sitjyks

TRANSACTIONS

OF THE

WISCONSIN ACADEMY

OF

SCIENCES, ARTS, AND LETTERS

VOL. XV, PART 1

1904

WITH EIGHT PLATES

EDITED BY THE SECRETARY

Published by Authority of Law

MADISON, WIS.

Democrat Printing Co., State Printer £([& b /

1905

,

TRANSACTIONS

OF THE

WISCONSIN ACADEMY

OF

SCIENCES, ARTS, AND LETTERS

VOL. XV, PART I

1904

WITH EIGHT PLATES

EDITED BY THE SECRETARY

Published by Authority of Law

MADISON, WIS,

Democrat Printing Co,, State Printer

1905

OFFICERS.

President,

JOHN J. DAVIS, Racine. '

Vice-Presidents,

CHARLES H. CHANDLER, Ripon.

HENRY E. LEGLER, Madison.

DEXTER P. NICHOLSON, Appleton.

Secretary,

ERNEST B. SKINNER, Madison.

Treasurer,

ROLLIN H. DENNISTON, Madison.

Librarian,

WALTER M. SMITH, Madison.

Curator,

CHARLES E. BROWN, Milwaukee.

TABLE OF CONTENTS

PAGE

The reproductive organs of the female maia moth,

Hemileuca maia (Drury), (with two plates)

Wm. S. Marshall, 1

The correlation of fracture systems and the evidences of

planetary dislocations within the earth’s crust (with,

one plate) . . William Herbert Hobbs, 15

The nature and origin of the binucleated cells in some

Basidiomycetes (with three plates)

Susis Per civ al Nichols , 30

The Russulas of Madison and vicinity

Rollin H. Demniston , 71

The relation of the Andrenine bees to the entomophilous

flora of Milwaukee county . 8. Graenicher, 89

Observations on the wintering of grain rus+tt

A. H. Christman j 98

Habits and anatomy of the larva of the caddis fly*

Platyphylax designatus (Walker), (with two plates)

Charles T. Vorhies , 108

Notes on the occurrence of Oscillatoria prolifica (Gre-

ville) Gomont in the ice of Pine Lake, Waukesha

county, Wisconsin . . . Edgar W. Olive, 124

Infection experiments with Erysiphe graminis, D. C.

George M. Reed, 135

VI

Table of Contents.

The state administration of taxation in Wisconsin

James D. Barnett , 163

The parts of speech in the child’s linguistic development

M. V . O'Shea, 178

On the nature of the process of osmosis and osmotic pres¬

sure with observations concerning dialysis

Louis Kohlenberg , 209

LIST OF PLATES

Plates To face pages

TTL Marshall on Hemileuca mam . 12, 14

III. Hobbs on Correlation of Fracture Sys¬

tems ..... 18

IV- VI. Nichols on Binueleated Cells in Some

Basidiomycetes . . 66, 68, 70

VII-VIII. Vorhies on Larva of Platyphylax

designatus (Walker) . 120, 122

THE REPRODUCTIVE ORGANS OF THE FEMALE MAIA

MOTH, Hemikuca Maia (Drury).

BY” WM. S. MARSHALL.

Hemileuca maia is found late in the summer near Madison

occurring very abundantly in the marsh land at the margin of

one of the lakes adjacent to the city. Here for a few days both

the males and females fly low over the marsh or settle on the

grass and small twigs, the latter to lay their eggs. One year a

number of the moths were collected and saved for study; some

were prepared by having the body cut open before throwing them

into alcohol, while from others the reproductive organs were re¬

moved and hardened in corrosive sublimate or Flemming's

solution.

The reproducitve organs of the female moth are, in general,

similar to those which have been described for other Lepidoptera.

Each ovary consists of four long ovarian tubules all of which

are bent and coiled forming a large irregular mass within the

abdomen. When the tubules are separated from each other

each one is seen to be a long tube having the same diameter

throughout, except for a short distance, at the distal end, where

it is narrower (Fig. 4.) Those moths from which the eggs have

been expelled show the ovarian tubules shorter and much nar¬

rower. The eggs within the tubules, which, before egg-laying

generally exceed forty in number, give to each tubule the appear¬

ance of a string of beads. After the expulsion of part of the

eggs there are here and there in the tubule considerable distances

between neighboring eggs, no regularity, however, being apparent

2 Wisconsin Academy of Sciences, Arts, and Letters.

in this respect; such spaces in the tubules being found inde¬

pendently in each one, and all showing slight differences. The

four ovarian tubules of each ovary meet to form a short oviduct

and the two oviducts join near the median line of the body to

form a wide oviductus communis. This passes towards the

posterior end of the body soon, however, enlarging to form the

vagina into which open the bursa copulatrix, dorsally, and the

receptaculum seminis on its ventral surface. The vagina has

dorsally at its distal end an enlarged saccular evagination or

pouch into the apex of which the tube from the receptaculum

seminis opens. The bursa copulatrix, a pyriform sac, is situ¬

ated on the left side of the body. It communicates with the

exterior through the ostium bursae, ventrally on the eighth ab¬

dominal segment, and with the vagina through a narrow tube,

the doctus seminalis, which in Tlemileuca is without the vesi¬

cular swelling present in so many Lepidoptera. The recepta-

eulum seminis, generally on the left side of the body, is double,

consisting of a large rounded part and of a second smaller more

tubular piece which, near its distal end, bears a narrow, tubular

appendix. The cement or sebaceous glands, glandulae sebaceae,

two long narrow tubes, lie in the right side of the body where

they form a bent and coiled mass. The proximal part of each

gland enlarges to form a reservoir, these two parts then uniting

and emptying by a common duct into the vagina. As already

mentioned, the glands lie in the right side of the body, but the

enlarged portions are dorsal in position, just above the vagina.

The intestine is entirely functionless, the moths taking no food;

it is a long narrow tube which, near its posterior end, becomes

much enlarged to form the rectum on the distal end of which

is a large saccular outgrowth.

A ventral view of the eighth abdominal segment discloses a

large genital plate the shape of which is shown in (Fig. 3).

This plate extends from near the middle of this segment to its

anterior margin, where, in the median line, is the ostium bursae.

The same figure shows the opening of the vagina on the ventral

surface of the ninth segment.

Ovary. Each of the two ovaries is made up of four ovarian

Marshall — Reproductive Organs of the Female Moth . 3

tubules, these are not bound together in any way, but each one

throughout its entire length is separated from the others. Be¬

fore the expulsion of the eggs each tubule is from 70mm. to

80inm. in length. The number of eggs each contains is not

constant, thirty-eight being the average in those counted. The

tour tubules on each side unite to form an oviduct, 1.5 mm. m

length, and the two oviducts join in the median line to form an

oviductus communis 2 mm. long.

Throughout the entire length, from the distal end of the

tubule to that point where the oviductus communis passes into

the vagina, the wall is very similar in structure. On the inner

surface there is a folded chitinous layer, somewhat thinner near

the distal end. The next layer, the epithelial, has, as such,

nearly disappeared, and in its place is an empty space contain¬

ing a few scattered nuclei each of which shows a few irregular

chromatin granules. The cytoplasm, which we can assume was

present when the cells were active, has entirely disappeared.

Externally each tubule is lined with two muscular layers, an

inner circular, and an outer longitudinal layer. Throughout

the entire length of the tubule there is only a very slight, if

any, difference in the comparative thickness of these two layers

(Figs. 5 and 6). Sections cut through the distal end and the

middle of the tubule will show this.

The oviductus communis shows a slight change from what we

have just described for the ovarian tubule. The remains of the

epithelial cells are much more marked, the nuclei appearing at

fairly regular intervals, and the cell boundaries, while not com¬

plete, are present in such a condition as would allow us to limit

the boundaries of the cells which were earlier present in this

layer. Both muscular layers show a greater development (Fig.

7), and while the circular longitudinal layers bear to each other

the same relative thickness that they did in the tubule, both are

here very much thicker. In all specimens of Hemileuca exam¬

ined the eggs were fully developed, and in most of the moths

they had been in part or entirely expelled from the body. The

egg-laying goes on rapidly, and but a short time is needed for

the expulsion of all the eggs from the body. The only use the

4 Wisconsin Academy of Sciences, Arts, and Letters.

ovarian tubules and oviducts could at this stage possess would

be to aid in expelling the eggs, and we find the parts which

are useful in doing this, the muscular layers, still normal, while

the other part of the wall, epithelial cells, shows a marked de¬

generation.

Beceptaculum seminis. — The seminal receptacle is double con¬

sisting of a larger bladder-like, and a smaller tubular piece, the

latter of which is pointed and near its distal end bears a thin

coiled appendix (Fig. 8). The two parts have a common duct

which opens into the apex of a saccular outgrowth near the dis¬

tal end of the vagina. The relative size and shape of the two

parts can best be understood by the figure from which there

are, however, many variations, either in a general increase or

diminution in size of the receptacle, or a relative change in one

part to the other. From an external view the duct appears to

come equally from each part, a section shows at first a similar

condition ; the cells lining the duct are more like those in the

smaller than in the larger part, being in fact the same cylin¬

drical cells, shortened, but not, however, flattened as in the

larger part of the receptacle, flflie appendix, near the distal end

of the smaller part, shows a considerable variation in length.

In the specimens of Eemileuca examined both parts contained

spermatozoa.

An examination of sections shows the wall in the two parts

to be different, similar, however, in being lined over the entire

internal surface with chi tin. In the smaller part of the recep¬

tacle (glandular part), the greater portion of the wall is a layer

of long cylindrical epithelial cells (Fig. 9) which show a de¬

cided striation at both ends but more marked in the free than

in the basal end. The large ovoid nucleus in each cell contains

a number of chromatin granules connected by a linin network.

On the outer surface of the wall are a number of circular mus¬

cles which, three or four layers thick near the proximal end,

gradually diminish until but a single row is left at the apex.

Sections through the tubular appendage of this part (Fig. 10)

show differences from what we have just described. The cell

boundaries were not visible, the nuclei have the same structure

Marshall — Reproductive Organs of the Female Moth. 5

as the others, but are much wider and situated in the basal part

of each cell. These cells do not show any longitudinal striation,

but large vacuoles are present in the cytoplasm. The inner

chitinous layer is much reduced in thickness. Near the free ends

of these cells there is a row of bodies which are apparently

nuclei. They have the same general structure as the large nuclei,

but differ from them in shape and size. These nuclei, if they

are such, lie free in the cytoplasm, that wdiieh surrounds them

being darker than in other parts of the cells, without any- bound¬

aries separating them from each other or cutting them off from

other parts of the cells.

Where the large and small parts pass into each other the cells

of the wall change very gradually (Fig. 12), the cylindrical

cells of the smaller part gradually getting shorter and shorter

until they become flattened as we find them in the larger part of

the receptacle. Here (Fig. 11) the wrall is much thinner, due al¬

most entirely to a flattening of the epithelial cells, the chitinous

lining and layer of circular muscles both being nearly as thick

as in the small part. The epithelial cells in the smaller part have

an active glandular appearance, but here in the larger part they

are very much reduced. Their nuclei have chromatin granules

which are gathered in an irregular mass near the center, and

have the appearance of being functionless. Even the circular

muscles (we take those bodies we have drawn on the outer sur¬

face of the wall to be such) do not have the same appearance

as those we find in the small part of the receptacle, and we

judge that this wall is entirely inactive.

Bursa copulatrix. — From the opening in the gential plate,

ostium bursae, a short tube .1 mm. in length leads into the bursa

copulatrix. This is an elongated sac pyriform in shape, some

5mm. in length, 1.25mm. wide at ■ narrowest, and 2inm. at the

widest part. The wail is thin, having apparently no function

other than that of an enclosing sac; in section it appears to be

without any definite structure, but composed of a fibrous-like

mass in which nuclei lie irregularly scattered. These nuclei

show no structure other than that each contains a number of

what appear to be small chromatin granules. From the pros-

6 Wisconsin Academy of Sciences, Arts, and Letters.

imal part of the copulatory pouch a narrow tube 1 mm. in length

leads into the vagina, opening opposite the duct from the re-

ceptaculum seminis.

Cement glands. — The cement glands are two long gradually

tapering, tubular parts, very thin distally, but soon beginning

to increase in diameter; the proximal part is much larger, this

enlarged portion, the reservoir, being about one-quarter of the

entire length. Each gland is 30mm. in length, the two uniting,

as is common in Lepidoptera, in a common duct 2mm. long, which

empties dorsal ly into the vagina 1mm. anterior to its proximal

end and almost opposite, but a little posterior to, the opening

of the receptaculum seminis. The entire thinner glandular por¬

tions lie in a twisted mass at the right side in the posterior part

of the abdomen. The larger reservoir is generally dorsal to

the vagina.

Sections through the gland at any place in the distal third

show the wall to consist of a layer of epithelial cells, each ceil

long and narrow, containing an elongated nucleus in the basal

half and a number of small vacuoles scattered throughout the

cytoplasm. Over the free ends of these celis is a loose chitinous

layer which in the sections appears wavy. Scattered just under¬

neath this chitinous layer were a number of small nuclear- like

bodies, many of which were elongated and nearly tubular (Fig.

14). Seen in a surface view the chitin appears marked off into

small irregular spaces each one of which contains one of these

nuclear-like bodies (Fig. 15). The chitinous layer covering this

part of the cement gland is a continuation of the same layer

which is found to cover the inner surface of the gland through¬

out its entire length. The nuclear-like bodies appear detached

from the epithelial cells, and attached to the chitinous. layer,

but this might easily be due to poor preservation of the tissue

and not normal.

Fig. 16 shows a section cut nearer the proximal end, although

still in the narrow part of the gland, and we notice that the

very narrow cells just described are replaced by wider ones

which are somewhat pointed at their free ends and may be

separated from each other either throughout their entire length

Marshall — Reproductive Organs of the Female Moth. 7

or, oftener, for only a part thereof. Near the free end of each

cell is a peculiar small ovoid body which does not so closely re¬

semble a nucleus as the bodies just described as present under

the chitinous layer (Fig. 14). They are quite homogeneous,

stain darker than the surrounding cytoplasm, but are not seen

distinctly in all cells of this region. Towards the proximal

end of the thin glandular part just before it widens to form

the reservoir, a distinct change is noted in the structure of the

wall. The epithelial cells of which it consists are here of the

same length as those just described, but are much wider. The

nucleus is also different here, and, instead of lying with its long

axis parallel to the long axis of the cell, lies transversely across

it ; it is also bent with the convex surface towards the basal end

of the cell. Along its other surface which is concave, lies a

peculiar body, rounded in outline, of a dark homogeneous appear¬

ance, with a yet darker small central portion from which radiate

a number of dark lines none of which reach the periphery (Fig.

17). Similar cells have been described in many glandular tis¬

sues of insects by Dierckx (3), Glison (5), and in the reproduc¬

tive organs of the Orthoptera by Fenard (4). From the dark

central portion of this body a duct passes into the lumen of

the gland penetrating the chitinous layer which is here present

as in other parts of the gland.

The wide proximal part of the gland has become a reservoir

to hold the secretion of the other parts. Internally there is a

chitinous layer much wider than in the glandular part, and very

much folded. The epithelial layer has undoubtedly become

functionless, and is represented by a thin layer of cytoplasm,

without cell boundaries, in which are scattered a number of

ovoid nuclei here much smaller than in any other portion of

the gland. There is an outer layer of longitudinal muscles

which is not present in other parts of the gland.

Vagina. — Tbe vagina is a continuation of the oviductus com¬

munis from which it differs, externally, only in its slightly

greater width. Near its distal margin, along the dorsal wail,

it bears a saccular evagination which receives the duct from the

receptaculum seininis; nearly opposite to this the ductus sem-

inalis opens.

8 Wisconsin Academy of Sciences, Arts, and Letters.

The wall of the vagina shows in section a greater develop¬

ment of the muscular layers than any other part of the repro¬

ductive organs. A similar enlarged muscular layer was found

by Verson and Bisson (12) for Bombys. The thin internal chit-

inous layer is folded and in the epithelial layer the cells show

a partial degeneration; the nuclei appear normal but the cyto¬

plasm has in part disappeared from many of the cells. The

layer of circular muscles is wide and followed externally by a

narrower longitudinal layer.

Zoological Laboratory,

University of Wisconsin,

February, 1905.

Marshall — Reproductive Organs of the Female Moth. 9

BIBLIOGRAPHY.

1. Burgess, E. Contributions to the anatomy of the milk-weed

butterfly {Dana, is A.r chippies) . Anniv. Mem. Bost. Soc.

Nat. Hist. 1880.

2. Cholodowsky, N. Ueber den geschlechtsapparat von

Nematois metallicus. Zeit. f. Wiss. Zool. Bd. XL’II.

1885.

3. Dierckx, F. Etude comparee des glandes pygidiennes chez

les Carabides et les Dyiiscides. La Cellule. Tome XVL

1899.

4. Fenard, A. Recherches sur les organes complementaires

internes de F appareil genital des Orthopteres. These.

1896.

5. Gilson, G. Les glandes odoriferes du Blaps mortisaga et

de quelques autres espeees. La Cellule. Tome Y.

1889.

6. Griffiths, A. B. On the reproductive organs of Noctua

pronuba. Proc. Roy. Soc. Edinburgh. Vol. XX.

7. Herold. Entwicklungsgeschichte der Schmetterlinge anat-

omiscb und physiologisch bearbeitet. Cassel und Mar¬

burg. 1815.

8. Packard, A. A text book of entomology. New York, 1898.

9. Petersen, W.Beitrage zur Morphologie der Lepidopteren.

Mem. de F Acad, St. Petersbg, 8th Ser. Vol. IX. 1900.

10. Petersen, W. Zur Morphogenese der doppelten Bursa

copulatrix bei Schmetter] ingen, Allg. Zeitschr. f. Ento-

mol. Bd. YI. 1901.

11. Stitz, II. Der genitalapparat der Microlepidopteren. Zool.

Jahrb. Bd. XIY. 1901.

12. Verson, E. and E. Bisson. Die postembronale entwicklung

der ausfiihrungs gange und der Nebendriisen beim weib-

lichen geschlechtsapparat von Bombyx mori, Zeit. f.

Wiss. Zool. Bd. LXI. 1896.

10 Wisconsin Academy of Sciences, Arts, and Letters.

EXPLANATION OF PLATES.

All figures except 1. 2. 3, 4, and 8 drawn with camera lucida.

Marshall — Reproductive Organs of the Female Moth. 11

PLATE I.

12

Wisconsin Academy of Sciences , Arts, and Letters.

EXPLANATION OF PLATE I.

Fig. 1. Reproductive organs, the eight avarian tubules have been re¬

moved. x 3.

Fig. 2. More enlarged view of part of same. The different parts are

not in their normal positions, the receptaculum seminis is

thrown over to the right; the vagina is shown in lateral

and the oviducts in dorsal view.

Fig. 3. Ventral view of eighth and ninth abdominal segments show¬

ing the opening o. b. into the bursa copulatrix on the ante¬

rior margin of the eighth, and the opening of the vagina

on the ninth segment. Enlarged.

Fig. 4. A single ovarian tubule.

Fig. 5. Longitudinal section through the wall of distal part of an

ovarian tubule, x 320.

F.g. 6. Transverse section of wall of same, near the middle, x 250.

Fig. 8. Receptaculum seminis. Enlarged.

Trans. Wis. Acad., Vol. XV.

Plate I.

W. S. M., del.

Marshall — Reproductive Organs of the Female Moth.

PLATE II.

13

14

Wisconsin Academy of Sciences, Arts, and Letters,

Trans. Wis. Acad., Vol. XX.

Plate II.

W. S. M., del.

THE CORRELATION OF FRACTURE SYSTEMS AND

THE EVIDENCES OF PLANETARY DISLOCATIONS

WITHIN THE EARTH’S CRUST.

BY WILLIAM HERBERT HOBBS.

In a recent paper tlie author lias attempted to correlate within

a complex and varied geological province the somewhat

scanty observations which deal with the orientation of fracture

systems.3 4 The results of this correlation possess considerable

significance inasmuch as there is a clear indication that over

quite an appreciable fraction of the earth’s surface the main

lines of fracture betray evidences of a common origin. Since the

publication of the paper so many verifications of this conclusion

have come to the writer from geologists whose work was unpub¬

lished, or which apply to the extension of the province treated,

that it seems desirable to extend both the area considered and the

general topic in order to include the new material.

That the fracture systems observed in a local district betray

close relationship to those of neighboring districts, was early

shown by Phillips2 and Ilaughton,3 and later by Kinahan1 for

1 Lineaments of the Atlantic Border Region. Bull. Geol. Soc. Am.,

Vol. 15, Nov. 1904, pp. 483-506. Pis. 45-47. Read hy title before the

International Geographical Congress at St. Louis, Sept., 1904.

2 Phillips, John. Illustrations of the Geology of Yorkshire, Part 2.

The Mountain Limestone District. London, 1836, pp. 90-98. Also,

Manual of Geology, London, 1885 \(Etheridge and Seeley Edition), pp.

33-34.

sHaughton, Samuel. “On the physical structure of the old red sand¬

stone of the County of Waterford, considered with relation to cleav¬

age, joint surfaces and faults. Trans. Roy. Soc. Lond., Vol. 148, 1858,

pp. 133-348. Also, On the joint systems of Ireland and Cornwall and

their mechanical origin. Ibid, Vol. 154, 1864, pp. 383-411.

4 Kinahan, Gerald Henry. Valleys, their relations to fissures, fract¬

ures and faults. London, 1875, pp. xiv and 240.

16 Wisconsin Academy of Sciences, Arts, and Letters.

portions of the British Isles, and also by Kjerulf1 for Norway.

Haughton, and more recently Brogger,2 have conclusively proven

. that in the districts which they studied the normal faults con¬

form in direction to the general system of joints; but the import¬

ance of this result, though reinforced and emphasized by

Daubree,3 lias never been fully appreciated.

For the region of the eastern United States it was developed

in the author’s paper referred to, that the joints and normal

faults of greatest prominence follow in any particular part of

the province one or more of three or four general directions.

These directions approximate to the meridian and the equator,

and to diagonal intermediate bearings. There seems good rea¬

son to believe that as regards the first two of the directions men¬

tioned the approximations to meridian and equator are fairly

close — generally within 5 degrees. As regards the intermediate

directions this is far from being the case, and there are in general

not two but several intermediate directions; yet their general

tendency to occupy rather distinctly intermediate positions be¬

tween the meridian and the equator is sufficiently manifest.

Wherever close observations have been made it has been found

that not four but a considerably larger number of directions may

be made out, as will be clearly indicated in the following tables.

These observations receive, moreover, strong support from the

studies of Green, Prinz, and others dealing with the orientation

of the broader earth features upon the entire planet.

North Carolina . — Since the author’s paper4 wms published it

was learned that a careful measurement of joint and dike direc¬

tions had been made by Mr. F. B. Laney within the Newark area

of North Carolina, and as a result of his studies Mr. Laney

stated that joints and dikes alike were oriented mainly north and

south, east and west, northwest and southeast, and northeast and

southwest. Mr. Laney has kindly turned over to the author his

i Kjerulf, Theodor. Die Geologie des siidlichen und mittleren Nor-

wegen. Authorized German edition by Gurlt. Bonn, 1880, pp. 1-350.

2Brogger, W. C. Spaltenverwerfungen in the Gegend Langesund-

skien, Nyt Magazin for Naturvidenskaberne. Yol. 28, 1884, pp. 253-

419, with map.

s Daubree, Geologie experimentale. Vol. I, pp. 289-385.

4}. c.

Holts — The Correlation of Fracture Systems. 17

detailed observations, from which it is clearly seen that the in¬

termediate directions which have greatest importance, while prop¬

erly enough described in general terms as northeast and north¬

west, fall principally in six and not two series. The comparison

of the general results determined for the lineaments of the At¬

lantic border region with the general directions observed in in¬

dividual districts examined by the writer expresses a like result.

The actual observations made by Mr. Laney have been tabulated

by the author with the following result.

In the North Carolina Newark both joints and dikes stand ap¬

proximately vertical, and it appears that the dominant direction

Is N 15° E. The remaining joint series are so oriented as to fall

into three conjugate sets.. Moreover, the joints observed at any

locality were found generally to be in pairs corresponding to one

or the other of the sets.

The Finger Lakes district of West-Central New York. Mr.

Charles G. Brown of Ithaca, N. Y., has measured the directions

of more than 1000 joint planes which occur in the vicinity of

Ithaca in the basins of the near-lying lakes. In order of rela¬

tive numerical importance the joint directions of the district

are found to be i1

N. 20° W. (288), N. 10° W. (139), N. 70°— 75° E. (131),

N. 15° — 16° W. (88), N. 60° W. (86), N. 30° E. (68), N. 80° E.

(62), N. 40° W. (51), N. 4°— 6" W. (43), N. 30°-— 34° W. (35),

N. 85° E. (35), and N. — S. (15). These twelve directions in¬

clude 941 of the 1,004 measurements and with the exception of

i Jour. Geol., vol. 13, 1905, pp. 367-374.

18 'Wisconsin Academy of Sciences, Arts, and Letters.

the direction N. 12° W., whose 14 observations should perhaps be

added to the 139 directed N. 10 '' W., no other direction is repre¬

sented by more than six measurements.

The fracture system of which these are the more prominent

series has controlled in an important way the topography of the

region, as has been described in the paper referred to.

The larger basin of the near lying Lake Ontario has been

shown by Wilson1 to have its lines of drainage determined by

fracture planes, though no indication is afforded of the cardinal

directions characteristic of the system. This author says:2

“The direction of the master joint fractures is intimately asso¬

ciated with the trend of all the master valleys, though the

modern post-Glacial channels are independent of them. . .

These master joints with other associated valleys and the parallel

system of valleys of similar trend on the adjacent Archean areas

are probably associated with an extensive system of faults of

pre-Ordovician date.”

This province like that surrounding the Ausable chasm near

Lake Champlain should afford a most promising field for the

observation of cardinal directions within fracture systems.

The French River District of Ontario. — Bell has called atten¬

tion to the marked orientation of the inlets and streams within

the French River district of Ontario,3 and has ascribed this ori¬

entation to the control by fissures, joints, and dislocations. The

directions are strongly marked in the district and trend E-W or

very nearly so, and approximately NW-SE, NE-SW. Bell

says :

“The effects of cleavage and bedding, fissures and joints, lock-

crushing, dislocations, intrusive dikes, etc., on the production of

geographical features are here so well marked as to make it

worth calling attention to some points in connection with this

subject. In any part of the district we may select, it will be

found that the joints, fissures, and dislocations, generally run in

two sets intersecting each other at large angles, but those of

1 Wilson, A. W. G., Trent River system and St. Lawrence outlet.

Bull. Geol. Soc. Am., Vol. 15, pp. 211-242. Pis. 5-10.

2 Bell, Robert. Report on the geology of the French River district,

Ontario. Geol. Surv. Can., Ann. Rept, N. S., Yol ix, 1896, pp. 20-1 to

21-1.

Trans. Wis. Acad., Vol. XV.

Plate III.

Hobbs — The Correlation of Fracture Systems. 19

either set are parallel to each other. Usually one set is more

strongly marked than the other and exercises an important influ¬

ence in the decay and disintegration of the rocks, and this in its

turn affects the contours of hill and valley and determines the

positions of streams, inland lakes and of the inlets, etc., of Geor¬

gian Bay.

‘ ‘ The dikes which traverse both the Laurentian and Huronian

rocks of the district and the fissures and lines of crushing wTiich

occur more particularly in the former, have given birth to some

of the more striking features of the map. . .”

In an earlier paper Bell1 had already shown that many of the

long and straight valleys within the Archean area of Canada

now occupied by rivers, lakes, or by inlets of the larger lakes,

have been formed as a result either of the relatively rapid de¬

cay on dikes or on lines of close joints. The valleys now occu¬

pied by water are further extended by valleys filled in with

drift. Examples are Onaping Lake, 30 miles long; Long Lake,

52 miles in length ; and Sepiwesk Lake, which with Nelson River

forms a trench 96 miles in length. The Mattagami River is

thus guided for 160 miles, and Lake Temiscaming, 35 miles in

length, with Deep River, forms a rectilinear trench in places

more than 2000 feet in depth. Soundings show that the rec¬

tilinear inlets of Georgian Bay are extended lakeward by

straight channels at their fronts. The prevailing joints of the

region correspond perfectly in direction with these trenches.2

The map just published3 by the Department of Crown Lands

and covering the area about Lake Temiscaming reveals this

orientation in great perfection (see Plate III). Dr. C. K. Leith,

who has done much recent geological work in the district con¬

firms the general correctness of this map and the correspondence

of drainage lines with the direction of the prevailing joints.

As the map indicates, the prevailing joint direction is about N.

40° W., with N. 50° E., and N-S the directions next in import¬

ance.

iBell, Robert. Pre-Paleozoic decay of crystalline rocks north of Lake

Huron. Bull. Geol. Soc. Am., Vol. 5, 1894, pp. 357-366, Pis, 15, 16.

2 Personal communication from Dr. Bell.

s Map of part of the District of Nipissing, showing agricultural lands

surveyed on Lake Temiscaming, Ontario, 1905.

20 Wisconsin Academy of Sciences, Arts, and Letters.

Southwestern Wisconsin and Northern Illinois. — As early as

1866 Whitney3 showed that the Galena limestone formation is in¬

tersected by two sets of vertical fissures whose directions are ap¬

proximately east and west, and north and south.

“All through the mining district, indeed, in Wisconsin and

Iowa as w'ell as in Illinois, the heaviest diggings will usually be

found on crevices varying but little from east to west in their

general direction. . . The norths and souths on the other

hand or those crevices which have a course approximating to the

meridian, are much less important, although these in some in¬

stances are the most productive ones of portions of the lead re¬

gion.” Buckley as a result of his measurement of joint direc¬

tions within the quarries of the State of Wisconsin makes the

following general statement.

“As wdll be seen in the accompanying map the joints of the

sedimentary rocks strike in four main directions. The prevail¬

ing general direction of the joints is northeast and southwest.

The other directions are northwest and southeast, east and west,

and north and south.’’

An examination of Buckley’s map will show that the interme¬

diate directions vary in many cases widely from the forty-five

degree positions. Under the writer’s direction Mr. E. C. Harder2

has made a careful study of the joint systems which are devel¬

oped in the rocks of southwestern Wisconsin. For this province

as a w-hole he finds that in order of numerical superiority the

dominant joint directions of the district are, N. 35° E., N. 75° E.,

N. 35° W., N. 55° W., N. 45° E., N. 85-90° E., N. 25° W., N. 45°

W., N. 15° W., N. 25° E., N. 65° E., N. 75° W., N. 65° W., and

N. 55° E. The tendency of the five degree interval to appear

is noticeable and indicates that here as elsewhere the observa¬

tions of slightly curving planes become adjusted to the larger

unit of the compass. Thus the equatorial direction, which is the

dominant one, should probably include the measurements rang¬

ing between N. 85° W. and N. 85° E., or 25 in all. Only less

noticeable aggregations appear elsewhere in the table, where

i Whitney, J. D. Geology of the Lead Region. Geol. Surv. Ills., Yol.

1, 1866, p. 194.

2 Jour. Geol. Yol. 13, 1905, pp. 363-366.

Hobbs — The Correlation of Fracture Systems . 21

probably if the joints could be measured with greater accuracy

they would fall between two five degree limits on the side of the

one having the numerical superiority.

Iowa. — McGee1 has called attention to the conjugate set of

joints which penetrate the quarry rocks of the State of Iowa.

Of these joints there are two classes recognized, which are desig¬

nated by the quarrymen as “clay seams” and “dry seams.”

The chief set consists of two series of “clay seams” crossing

each other approximately at right angles, with two series of “dry

seams” also crossing about at right angles and approximately

bisecting the angles formed by the principal series. He further

adds :

“The phases presented are, however, variable. Either of the

two classes or either system of either class may be absent, or

additional and generally iess conspicuous systems . . . some¬

times so blended with the predominant class as to separate the

strata into either irregular or tolerably regular polygonal blocks

. . . may be introduced ; and the two classes pass impercep¬

tibly into each other.”

In a personal letter Dr. McGee adds:

“Unhappily my memory is not as clear as I could wish; so far

as I am able to recall the less distinct double-sets are orthogonal,

departing a few degrees from N-S and E-W, though the direc¬

tion of the departure from the meridian and parallel escapes me.

The more prevalent and conspicuous do ubi e-sets are also orthog¬

onal or nearly so and approximately NW-SE, and NE-SW. So

nearly as I can recollect the minor and major double-sets do not

exactly divide the quadrangle. The difference is somewhat more

(or less) than 45° ; . . .

“As I dwell on the matter in this writing the impression re¬

vives that the dominant single-set trends about N. 55° W., and

S. 55° E., with the correlative orthogonal single-set about N. 35°

E., and S. 35° W. ; and that the stronger joints of the minor

double-set trends about N. 5° W. to S. 5° E., and the weaker one

about N. 85° E. to S. 85° W. In any event such a scheme will

i McGee, W. J., Note on jointed structure. Am. Jour. Sci. 3rd Ser.

Vol. 25, 1883, pp. 152-155.

22 Wisconsin Academy of Sciences, Arts, and Letters.

illustrate approximately the relation of the joints worked out in

my building-stone researches in 1880. My best examples were

found in the central and southern sections of the state.”

Arkansas. — Professor Branner1 2 has been good enough to call

the author’s attention to similar results obtained by Professor

Newsom and himself when studying the eastern portion of the

Boston Mountains in Arkansas, in which investigation the pe¬

culiarities of drainage, the parallelism of streams, the similarity

of their elbows, and the relation of these streams to structural

features in the adjoining regions were put upon record. The

dominant directions of drainage are NE-SW, and NW-SE.

With greater definiteness these directions are given as N. 60°-67°

E., and N. 51°-65° W. The NE-SW series seems to be con¬

trolled by the monoclinal folds. Attention is called, however, to

the relationship existing between these stream directions and a

series of dislocations in the province, and it is further stated

that ‘ ‘ there are two other systems of joints in the area here

especially considered, one running N. and S., the other E. and

W.”

The Great Basin of the Western United States. — Since the

publication of the early reports upon the geology of the Great

Basin of the western United States it has been generally recog¬

nized that the dominant faults trend near the meridian. Gilbert

has also called attention to the fact that formations rather gen¬

erally end abruptly on east and west lines. Spurr1 in his gen¬

eral conclusions upon the origin of the basin ranges of Nevada

and California says:

“The faulting in general seems to be about as frequent as in

other regions which show' the same amount of folding. The

chief faults belong to the north-and-south and east- and-west

systems. There are also diagonal ones running northeast and

northwest and in each of the systems they , may have a very great

displacement. ’ ’

The most important directions described by Spurr are in per-

1 Newsom, J. F., and Branner, J. C. The Red River and Clinton

Monoclines. Am. Geol., Vol. 20, pp. 1-13.

2 Spurr, J. E. Origin and Structure of the Basin Ranges. Bull Geol.

Soc. Am., Vol. 12, 1901, pp. 217-270. Pis. 20-25.

Hobbs — The Correlation of Fracture Systems. 23

feet accord with those described by Gilbert as occurring in the

clays of Lake Bonneville and controlling the local drainage di¬

rections. The data were secured by Mr. Israel C. Russell, then

his assistant, of whom Gilbert2 says:

“He found that the details of drainage were controlled by a

compound and extended system of joints. The principal series

trend almost precisely" north and south and the subordinate

series east and west. They are all vertical and straight and

(within each series) closely parallel. They are readily traced

from top to bottom of the walls of the lateral ravines and not in¬

frequently a wall exposes a broad flat sheet face caused by the

removal of the clay from one side of the plane of jointing. Else¬

where the faces of the bluffs are buttressed by square pilasters or

ornamented by outstanding rectangular columns, the forms of

which have been determined byr the two systems of joints . . .

“The point of especial interest is that these joints have been

developed in post-glacial time within the series of strata not per¬

ceptibly indurated and which repose undisturbed in the place

where they were deposited. The strata are nearly horizontal

and their inclination of less than V20 northward is presumably

the pitch slope of the bottom upon which they were thrown

down. ’ ’

Similar joints were observed by Russell1 in the clays of Lake

Lahontan. He says:

“The marly clays forming the upper and lower members of

the Lahontan series usually break into prismatic and cubical

blocks on weathering,- the vertical faces of the blocks are deter¬

mined by joint planes and the horizontal by planes of lamina¬

tion. In many localities a more pronounced jointing occurs,

forming two approximately" vertical systems that are nearly at

right angles to each other. Judging from the number of in-

1 Gilbert, G. K. Post-Glacial Joints. Am. Journ. Sci., 3rd Ser., Vol.

23, 1882, pp. 25-27. See also same author, Lake Bonneville. Mono¬

graph I, U. S. Geo-1. Surv., 1890, pp. 211-213.

2 Russell, I. C. Geological History of Lake Lahontan. Monograph

XI, U. S. Geol. Surv., 1885, pp. 162-163.

ilddings, Jos. P. A Fracture Valley System. Journal of Geology,

Vol. 12, 1904, pp. 94-105. PI.

24 Wisconsin Academy of Sciences , Arts, and Letters.

stances observed at widely separated localities, the joints in ques¬

tion may be traced through the entire series of Lacustrian beds. ? 1

Yellowstone National Park. — Quite recently Iddings2 has

shown that upon the Livingstone quadrangle, which is included

in the area of the Yellowstone National Park, a net-work of

faults has determined a fracture valley system in which the

dominant directions are NE-SW, NW-SE, N-S, and E-W.

More definitely the diagonal directions are given as N. 30° E.,

and N. 60° W. He says:

“A study of the topographical map reveals the angular char¬

acter of much of the drainage system, and the prevalence of cer¬

tain parallel and sub-parallel lines which appear in various

streams and occur in quite diverse portions of their channels.

Along parallel lines different streams may be flowing in oppo¬

site directions: . . . The persistency of these lines becomes

more striking when the geological structure of the region is taken

into account and it is observed that certain drainage lines tra¬

verse rocks of such diverse nature as gneiss, schist, volcanic tuff,

breccia, solid lava, limestone, sandstone, and shales.

“The relation of some of these directions of drainage to

known fracture planes will be pointed out. The dominant

drainage lines in the southern tliree-fourths of the quadrangle

trend about NE-SW, and NW-SE, more nearly N. 30° E. and

W. 30° N., the angle between them being approximately 90 de¬

grees. There are other systems of almost rectangular lines

somewhat differently oriented, namely, N-S and E-W.”

Other Districts . — Only less important in a study of the orien¬

tation of fracture systems are other studies which will not be

here specifically referred to except as they have been included in

the following table. To facilitate the study of directions there

are given in separate columns : 1st. Those bearings of fracture

systems which approach the direction of the meridian; 2nd.

Those which approach the direction of the equator ; 3rd. Those

which occupy intermediate positions within the quadrants NE

and NW ; 4th. Those which occupy intermediate positions in

the quadrants NW and SE.

Hobbs — The Correlation of Fracture Systems.

25

Tabulation of Fracture Systems for the United States.

(Partial.)

Bearing of Fracture Series.

District and Au¬

thority.

Great Basin of Western

United States.

General direction for Gt.

Basin. — Giijert.

Basin Ranges in Cali-

forni? and Nevada.—

Spurr.

Upper Sevier Basin.— Gil¬

bert.

Utah and Arizona. — Gil¬

bert.

General Direction in Colo¬

rado plateau —Gilbert.

Lake Bonneville (Old

River Valley.)— Gilbert.

Lake Lahontan. — Rus¬

sell.

Lake Lahontan tN. of 37°

lat.)— Russell.

Telluride quadrangle in

Cole. — Purington.

Rico Mts. in Colo. — Ran-

some.

Globe Quadrangle in Ari¬

zona — Ransome.

Bisbee Quadrangle in Ari¬

zona. — Ransome.

Albuquerque and Magda¬

lena Mts. in New Mex¬

ico.— D. W. Johnson.

Same. — Keyes.

Death Valley in So. Cal.

— Campbell.

Boise Quadrangle, Idaho

— Lindgren

Livingston Quadrangle,

Montana. — Iddings.

Butte Min. District.— Em¬

mons & Tower.

Merid¬

ional or

near¬

ly so.

N— S

N-S

Equator¬

ial or

near¬

ly so.

E-W

E-W?

In quad¬

rants NE

&SW.

NE-SW

N 10°-

30° E

N NE-

s sw

N 38Q E

N 53° —

63° E

N 25° —

65° E

NE-SW

In quad¬

rants

NW &

SE.

NW-SE

Remarks.

Formations also end

in E — W cliffs.

In clay.

Also a perpendicular

vertical set in clay.

N 21° — Directions N 53° — 63®

51° W

N 87° W

N-N 45°

W

NW-SE

to WNW

-ENE

E and N 21°— 51° W

recognized as each

containing several

series.

Also some others

common.

N-S

NE— SW

These are the direc¬

tions of major

faults in modem

mts.

E-W

E NE—

w SW

Rarely NE-SW.

E-W

N 30p E

N 60° W

Mainly

near E—

W.

In quadrant between

NE and SE but

mainly nearE — W.

‘ ‘Secondary frac¬

tures” with dips

from 45° — 65° at

right angles to

mineral veins.

26 Wisconsin Academy of Sciences , Arts, and Letters .

Tabulation of Fracture Systems for the United States— Continued.

District and Au¬

thority.

High Sierra of California.

—Becker.

. Upper Kern Val¬

ley.— Lawson.

Jb&wer Mississippi Val¬

ley.

Boston Mts. in Arkansas.

— Branner-Ne wsom.

Ozark Mts. in Ark. —

Powell.

Brunswick Dist. in Indian

Territory . — Eld ridge.

Balcones Fault Zone in

Texas.- Hill &Vaughan.

TDValde Quadrangle in

Texas.— Vaughan.

Upper Mississippi Val¬

ley.

Satire State of Wiscon¬

sin, — Buckley.

So. Wisconsin— Percival.

$W Wisconsin. — Harder.

Entire State of Iowa.—

McGee.

Eastern Iowa.— Bain and

Calvin.

Northern Illinois. (Also

Southwestern Wiscon¬

sin.'— Whitney.

Kosioiare Dist. in So. Ill.

— Emmons.

Minnesota. — Upham.

Eastern United States.

So Conn. Valley in Conn.

— Davis.

Conn. Valley in Mass. —

Emerson.

Merid¬

ional or

near¬

ly so.

N-S

N-S?

“ssrsr* ««*a

near¬

ly so.

rants NE

&SW.

ENE— W

SW

E-W

NE-SW

N— S

E-W

N 85° E

N 85° W

N 5‘

E i E— W

' N 85u W

N 85° E

N5»W

circa.

N-S

(nearly.)

N-S

N 85° E

circa

Noarly

E-W

E— W

N of E—

S of W

NE-SW

In quad¬

rants

NW &

SE.

N NW— S

SE

Remarks.

NW-SE

N 20° W

N 50° W

N 35° E

N 45° E

N 50° E

N 55° E

N 65° E

N 75° E

N 35° E

circa

NW-SE

N 70° W

N 80° W

N 25° W

N 35° W

N 45° W

N 55° W

circa

N 25® E

N 32° E

N 50° —

57° E

N 60° E

N 30° —

45° E

N 10®-

15° E

N 20° E

N 20° —

30° E

N 45 *V

N 20° —

30° W

N 50° W

Two diagonal sets

sometimes found

striking between

N-S and NW-

SE, and dipping

45® either way.

Also rarely hori¬

zontal sets.

Other directions of

minor importance.

Also minor faults.

Main direction NE-

SW.

Form two rectangu¬

lar sets.

Main direction E — W

A third series is

mentioned at 45° to

others.

Not sufficient num¬

ber o 1 planes ob¬

served to clearly in¬

dicate these as

series.

Scattered observa¬

tions.

Directions taken

from map.

Values are approxi¬

mate only and

taken from map.

Hobbs — The Correlation of Fracture Systems,

27

Tabulation of Fracture Systems for the United States— Continued.

28 Wisconsin Academy of Sciences, Arts, and Letters .

In the above table have been included all reports which have

come under the eye of the writer, or have been communicated to

him personally, in which the orientation of the fracture systems

has been indicated, however crudely; and this entirely without

reference to whether the results favor one theory more than an¬

other. One condition only has been imposed, namely, that the

fracture systems described shall be made up of individual sur¬

faces that in steepness approach the vertical — are in general

steeper than 70° — thus insuring that such tilting as may have

occurred since their formation has not materially affected the

plan of their arrangement.

There are undoubtedly many districts in which no such regu¬

larity of arrangement of fracture series can be discovered, but it

will generally be found that the planes of jointing or of veins or

dikes, are many of them inclined at comparatively low angles to

the horizontal. Even where this is not the case, the number of

directions of joint planes may be so many that no law of ar¬

rangement is discernable ; as might well be true when the belt

of rocks has been subjected to successive deformation within the

zone of fracture either from a single or from several directions.

If we are to discover any laws governing the orientation of frac¬

ture systems, it will be by proceeding from the simple to the

more complex areas, and there is ample ground for assuming that

where belts of flat-lying, homogeneous rocks without pre-existing

fracture planes are deformed within the zone of fracture, there

is normally produced a vertical prismatic system composed of

intersecting parallel series. Moreover, observations would ap¬

pear to show that even where rocks are far from homogeneous

and lie in other than horizontal positions, an approximation to

this result still obtains. While it has not as yet been demon¬

strated by experimentation, it is difficult to avoid the conclusion

based upon field observation, that a second deformation of rocks

which are already possessed of a simple prismatic system of

joints through renewal of compression from the original direc¬

tion, in the main merely increases the number of series within the

vertical joint system.

It is certainly of much significance that the systems of frac¬

tures which are developed throughout the area of the United

Hobbs — The Correlation of Fracture Systems.

29

States should so clearly correspond with the orientation of the

grander features of the planet as they have been worked out in

the studies of Prinz1 and others. Prinz has found that the

major features of the earth are arranged in two nearly recti¬

linear series running northwest and southeast, and northeast and

southwest, with an intermediate series directed nearly along the

meridian. It is evident that the present study merely blazes the

way for a more thorough and careful correlation to be made

when the available data are more nearly adequate for the solu¬

tion of the problem. It is therefore hoped that geologists will

co-operate to the extent of measuring and recording the direc¬

tion of joint and fault planes within the districts which they

individually have opportunity to examine.

University of Wisconsin,

Madison, Wisconsin,

May 20, 1905.

i Prinz, W. Sur les similitudes que presentent les cartes terrestre

et planetaires (Torsion apparent des planetes.) Ann. de robservatoire

royale de Bruxelles, 58th year, 1891, pp. 304-337.

THE NATURE AND ORIGIN OF THE BINUCLEATED

CELLS IN SOME BASIDIOMYCETES.

SUSIE PERCIVAL NICHOLS.

INTRODUCTION.

Rees (20) was among the first to attempt a careful study of

the mycelium of the Basidiomycetes with reference to the ques¬

tion of the origin of the carpophore. By making artificial cult¬

ures of Coprinus stercomrius in dung decoction on slides he was

able to observe the formation of erect short hyphae on which

he believed sexual cells, spermatia and carpogonia were borne.

He also believed that he found a spermatium fused with a car-

pogonium. After fertilization, branches arise from the base of

the carpogonium which developed into the carpophore.

Yan Tieghem (25) also germinated the spores of Coprinus

stercorarius and radiatus in dung decoction and studied the de¬

velopment of the carpophore. In his first paper he agreed with

Rees. He found the swollen end cells on the lateral branches

of the mycelium. These “carpogones” usually terminated in a

papilla with which the spermatia fused. The carpogone then

divided into three cells, the two lower developing a system of

lateral branches which curve around and enclose the terminal

cell. Their further development in slide cultures was prevented

by lack of nutriment. But by observations made on larger cult¬

ures they were seen to be the beginnings of carpophores. Later

Yan Tieghem reversed this opinion. *

Brefeld (3, 4) grew mycelium of Coprinus stercorarius from

single spores in dung decoction and figures a series of stages in

Nichols — Binucleated Cells in Some Basidiomyceles. 31

the development of the young carpophore. From the older por¬

tion of the mycelium a perpendicular hypha very rich in proto¬

plasm arises. This hypha branches profusely forming a dense

snarl from the center of which a bundle of parallel hyphae de¬

velop forming the first indication of the stipe. Lateral branches

are formed increasing the size of the mass and at the same time

the stipe grows rapidly in length. The pileus and gills are

differentiated very early in the development of the fruit body.

Brefeld found no evidence of the existence of sexual organs at

the formation of the carpophore.

With the study of the nuclear phenomena new stand-points

arose. The work of Rosen (21), Rosen vinge (22), Wager (26,

27, 28), and Dangeard (5) has established the fact that the

cells of the carpophore are frequently multinucleated while the

basidia are at first binucleated. In typical basidia the two

nuclei fuse and the fusion nucleus divides into the four spore

nuclei.

Maire (15) found that the cells of the young carpophore are

binucleated and that the cells of the hymenial layer never have

more than two nuclei, but that the cells of the stipe and pileus

may become multinucleated through the amitotic fragmentation

of the two nuclei originally present. The young basidiimi when

it is formed from the hymenial ceiis receives two and only two

nuclei which unite to form the large fusion nucleus of the basid-

ium. He further states that the nuclei in the series of binue-

leated cells in the young carpophore divide by conjugate divi¬

sions so that the two nuclei which fuse in the basidium are of

widely different origin. But his evidence is not conclusive on

this point.

Maire describes the division of the nucleus in the basidium in

detail. The nuclear membrane disappears and the spindle ap¬

pears at about the same time. The chromatin filaments break

up into irregular granules or protochromosomes which are placed

on the spindle without any definite order. At the end of the

prophase these protochromosomes unite into two definite chro¬

mosomes. That the formation of only two chromosomes is not

universal among the Basidiomycetes as Maire assumes has been

shown by Wager, Ruhiaud and others. Maire states further

32 Wisconsin Academy of Sciences, Arts, and Letters.

that the chromosomes after a longitudinal splitting are pulled

apart at the center and move to the poles. The second division

is similar to the first. The four centrosomes remain at the

summit of the basidium while the nuclei move to the center or

base of the basidium. boon a sterigma is formed above each

eentrosome, and fibres appear extending from the centrosome

to the nuclei which now move to the summit, probably through

the influence of these fibres.

Two notes of Maire’s (16 & 17) in the Comptes Eendus report

that the last two or three cells of the ascogenous hyphae of

Pustularia vesiculosa, Galactinia succosa and Acetabula acetab-

ulum are binucleated. The ascus like the basidium is the last

of a series of binucleated cells. In order that such a comparison

should have any value we must know how the ascogenous hyphae

originate.

In Hypochnus Harper (12) was able to trace a series of bi¬

nucleated cells from the hymenium to the mycelium in the sub¬

stratum. The mycelium did not form dense wefts or strands but

spread through the decaying wrood where it could be readily

studied. The cells were regularly binucleated. The stages of

nuclear fusion and division were similar to those described by

Wager. At the equatorial plate stage the chromosomes were

distinct and at least six or eight in number.

The origin of the binucleated cells was not determined by

these observers. Maire states that the two nuclei in the spore

of Coprinus radiatus pass into the germ tube and a cross wall

may or may not be formed between them. The mycelium is

then of two types, the one apoeytic, the other cellular. The

cells of the latter are uninucleated. He did not observe the

transition from these stages to the binucleated condition found

in the young carpophore.

In a preliminary notice Blackman (2) has given a brief ac¬

count of the life history of two of the Uredineae. He finds that

the spermatia do not have the structure of conidia but of male

cells; a thin wall, no reserve material, a very large nucleus with

no nucleolus and cytoplasm greatly reduced in amount. He

also studied in detail the development of the aecidium of Phrag -

midium violaccum. The aecidium arises as a layer of uninu-

Nichols — Binuclealcd Cells in Some Basidiomycetes. 33

cleated cells just beneath the epidermis of the leaf. Each of

these cells divides into a sterile cell above and a fertile cell below.

The fertile cell becomes binucleated not by the division of its

original single nucleus but by the migration through the wall

of the nucleus of a neighboring vegetative cell of the mycelium.

He sajte, “In the presence of the spermatia with their special

cytological characters, etc., the only view that seems capable of

explaining the facts is that the fertile cell was formerly fer¬

tilized by the spermatia, but that now the process has become

reduced, fertilization by means of spermatia having been re¬

placed by the more certain method ol! fertilization by the nucleus

of a neighboring vegetative cell. ” In view of these facts ne

holds it to be evident ‘ 4 that the Uredineae present an alterna¬

tion of generations which is as sharply marked as that of the

higher plants.”

Since the young carpophores invariably have binucleated cells

these must originate either at the first formation of the car¬

pophore or sometime during the growth of the mycelium. The

latter hypothesis is suggested by Harper’s observations on Hy¬

po chnus. In order to obtain some evidence on this point the

study of the nuclei in the mycelium of some of the Agarics was

begun at the suggestion of Professor R. A. Harper under whose

direction the work was carried on.

METHODS.

Spores were collected from a large number of Agarics in the

following manner. Zinc racks were washed in alcohol and passed

through a flame and then placed in plates which had been washed

in 95 per cent, alcohol. The racks were covered with bell jars

which were also washed in 95 per cent alcohol. Two racks were

placed side by side under each jar. Slides were washed m

alcohol and passed through a flame and then placed on the lower

bars of the racks, each rack holding four. The pileus from a

mature fruit body was carefully removed from the stipe and

placed on the upper bars of the rack. When the basidia dis¬

charge their spores they fall on the slides below thus lessening

3

34 Wisconsin Academy of Sciences, Arts, and Letters.

the danger of infection from the gills. After the spores were

discharged the pileus was carefully removed and the bell jar

replaced the slides remaining on the racks until needed. Spores

preserved in this manner remained pure for a year.

A decoction of string beans, (about 392 grams to a liter of

water) proved to be the best nutrient although a decoction of

Coprinus and of dung was used for some forms. For the early

stages in the development of the mycelium small cultures were

made in dishes holding 10 c. c. of the nutrient solutions. Large

quantities of spores were sown and at the end of 12, 24, and 48

hours, the nutrient solution was removed by a pipette until only

2 c. c. remained, the dish was then filled with fixing solution

which would thus be reduced about one-fifth in strength. After

fixing 24 hours the spores were stippled on the slide by the

method described by Harper (11) in his paper on the nuclear

phenomena in the smuts.

Spores were also sown in thin films of agar-agar on sterilized

slides. When the mycelium had attained the desired growth

the entire slide was immersed in fixing solution. If the film

loosened from the slide it was easily fastened again by a film of

albumen.

To obtain mycelia, cultures were made similar to those de¬

scribed by Falk (8). Rye bread cut in slices two or three inches

thick were moistened in bean decoction and fitted into battery

jars five inches deep by four wide. For covers petri dishes four

and a half inches in diameter were used. A thin layer of cotton

was placed between the cover and the dish to allow free circula¬

tion of air. Agar-agar plates were also used.

The spores germinate in from six to eight hours and at the

end of two or three days form a growth of mycelium which ap¬

pears as a white mat about a quarter of inch in diameter on

the surface of the bread. The mat increases in size rapidly until

it is two or three inches in diameter. At the same time there

appear all over the culture small white dot-like masses of mycel¬

ium. These are new growths from oidia scattered from the

first myc.elium. Falk has also described and illustrated such

oidial colonies. These small secondary growths were removed

Nichols — Binuclcated Cells' in Some Basidiomycei.es. 35

whole with about a quarter of an inch of the substratum.

Larger myeelia were cut into small pieces for fixation. In order

to force the fluid through the thick felt which the mycelium

forms the material was placed in a small bottle and well cov¬

ered with the fixing fluid. The bottle was then fitted into the

end of a rubber tube which was connected with the air pump.

The air was pumped out from the closely matted hyphae, after

which the fixing fluid was renewed.

The material was fixed in Flemming’s solutions both the

stronger and the weaker. Merkel’s and Herman’s solutions

were also used. The best results were obtained from Flemming’s

weaker solution. Both Flemming’s triple stain and Heiden-

hain’s iron haematoxylin gave satisfactory results.

Ilypholoma 'perplexum, Pk.

The spores of Ilypholoma perplexum were collected in great

abundance. The carpophores appeared in great profusion on

decaying stumps and logs of oak throughout September and Oc¬

tober. The spores were collected and stored after the manner

already described. A large per cent of the spores obtained from

mature pilei germinated in a shorter time than those obtained

from younger ones. The spores were sown in the hanging drop

agar cultures made with bean decoction. The cultures wrere kept

in a dark box at a temperature of 20° c. As all the spores do

not germinate at the same time but vary from four to forty-eight

hours in the time of the appearance of the germ tube a large

number of stages are obtained from the same slide.

The mature spores were studied in the drop cultures and also

in the cross sections of the gills where they were still in connec¬

tion with the sterigmata.

The mature spore of ilypholoma perplexum has a dark brown

opaque wTall which before germinating swells to two or three

times its original thickness becoming much lighter in color and

transparent. At this time it is easy to distinguish two nuclei

lying close together near the center of the spore. They show

the Usual structure of the resting nucleus and are small spherical

bodies with a w7ell defined membrane, a large distinct nucleole

36 Wisconsin Academy of Sciences , Arts, and Letters.

and a fine granular chromatin which fills the remaining space

(Fig. 1). The chromatin stains a light blue with the triple

stain while the nucleolus stains with safranin. Previous to

germination there is no indication of a germ pore but the germ

tube always appears at the same place on the spore. On germ¬

inating a tube in the form of a large round bulb of protoplasm

is protruded from the end opposite the point of original attach¬

ment to the sterigma. The germ tube increases rapidly in size

retaining its spherical form as the cytoplasm passes into it from

the spore. One of the nuclei passes from the spore into the

tube as soon as there is room (Fig. 2 & 4) the other may pass

out at once and it will then be seen near the first (Fig. 3) or it

may remain in the spore for some time. Strands of cytoplasm

caused by streaming from the spore into the tube are clearly

shown in many of the preparations. As the nucleus which has

remained in the spore is carried along by the streams of cyto¬

plasm it becomes drawn out into a blunt point on the side to¬

wards the germ tube. (Fig. 2). As the nucleus approaches the

narrow passage from the spore into the germ tube it is very

nearly cone shaped. The vacuoles are very small at this stage

with the exception of a single large vacuole which forms in

the spore back of the second nucleus.

The tube begins to lengthen immediately in the direction of

the long axis of the spore. The bulb has not formed a firm wall

at this time and as the tube lengthens it becomes drawn out

in the general direction of growth until its diameter is only

slightly greater than that of the spore. The two nuclei do not

remain together, one moving towards the apex of the tube -while

the other remains near the base. One of the preparations at this

stage showed late division figures (Fig. 5). The chromosomes

have already passed to the poles so that it is impossible to de¬

termine their number. The spindle fibres still connect the two

masses. The two nuclei lie at some distance from each other.

One spindle which is near the tip is nearly parallel with the

walls of the germ tube. The spindle of the other, near the spore,

is placed obliquely. Near one of the nuclei are two small deep

staining bodies which may be remnants of the old nucleoles.

(Fig. 5a). In an older germ tube the next division was ob-

Nichols — Binuclcalcd Cells in Some Basidiomyceles. 37

served. In this hypha three nuclei were in the resting condi¬

tion. One is somewhat separated from the others towards the

tip of the hypha and is a densely staining homogeneous body.

Two other nuclei side by side near the center are elongated with

the nucleole at one end. The fourth nucleus is dividing and

is in the equatorial plate stage. The separate chromosomes

could not be distinguished. The spindle fibres show an even

distribution at the center and are collected into a definite pole.

The presence of a central body could not be distinguished

definitely (Fig. 6).

As the germ tube elongates the cytoplasm forms a much thin¬

ner layer at the periphery of the cell and a number of large

vacuoles appear. The number and size of the vacuoles increase

until they are only separated by fine lamellae of granular cyto¬

plasm. The nuclei multiply rapidly and are frequently in pairs

for some distance but this arrangement is not at all uniform.

The nuclei may be separated from each other by long distances

or they may lie in groups of three or four. Cross partitions

are not formed in the young germ tube. It frequently branches

two or three times before the first cross wall is formed. The

cells which are then formed are very variable in length. The

number of nuclei which they contain is from one to four (Fig.

7). The mycelium does not consist in these earlier stages of

regularly uninucleated or binucleated cells. The mycelium

grows rapidly and branches freely and the branches interlace

forming a loose net-work. The branches do not show any reg¬

ularity in their origin. A few are formed near a cross wall but

the majority bud out somewhere near the center of the cell.

The majority do not show a cross partition at their base. The

first cross wall appears at varying distances up the branch.

Usually a nucleus is found at the base of each newly forming

branch but this rule is not constant for in a few cases a fairly

long branch was found without any nucleus near. The lateral

branches are narrower than the hyphae from which they arise.

The first few cells of the branch contain as a rule fewer nuclei

than the cells in the main hypha but the cells nearer the tip

may have a large number. The number of nuclei at this stage

varies from one to eight or nine; the same branch frequently

38 Wisconsin Academy of Sciences , Arts , and Letters.

showing both extremes in number. The nuclei are slightly

smaller than those in the young germ tube but show the same

general structure, a definite membrane and a nucleole which ap¬

pears separated from the finely granular chromatin by a small

clear space. In some of the preparations the nuclei were drawn

out into long rather slender bodies with larger and very irreg¬

ular chromatin granules loosely scattered throughout their

length.

Special lateral branches may also be formed which have reg¬

ularly uninucleated cells. These branches vary considerably in

length and general shape. Some are long and nearly straight,

or only slightly coiled, with long slender cells. The nucleus

lies near the center of the cell and has the structure described

above. Towards the ends of these branches very short cells are

formed, only about twice as long as broad. These cells separate

readily thus forming oidia. The character of the cytoplasm in

these oidial cells is the same as that of the branch from which

they originate. It is sometimes dense with very small vacuoles

similar to that found in the spore. In other cases it has much

larger vacuoles and the cytoplasmic granules as much larger

and more irregular. In the majority of cases the nucleus is

relatively large occupying nearly the entire diameter of the

cell. In some instances the nuclear membrane and nucleole

could not be distinguished and the chromatin was collected in

two or four deeply staining masses usually oblong in shape.

These nuclei may have been in some stage of division.

In one instance a slide culture that was placed in a very warm

moist chamber (22° c. ) showed a pronounced modification in

the usual habit of growth. In many cases the germ tube was

divided at once into uninucleated cells that bore numerous short

series of oidia. The branches were coiled about the tube mak¬

ing it impossible to determine their number or structure. In

other cases the hyphae were much longer and very slender with

two or three branches on whose ends were large tangled masses

of oidia. The oidia were rather slender and long but showed

the usual structure.

Carpophores were not formed in any of the cultures but my

studies on Ilypholoma perplexum were continued on material

Nichols— BinucUated Cells in Some Basidiomycetes. 39

which had developed spontaneously in nature. During the win¬

ter specimens of Hypholoma perplexum appeared on an oak Jog

in the green house. The rotten pieces of wood were broken away

exposing an expanded sheet-like rhizomorph or perhaps sclerot-

ium, with some young fruit-bodies just forming. A sharp dis¬

tinction between rhizomorphs and sclerotia probably cannot- be

made. Between the round tuber-like sclerotium of Coprinus

ephemerus and the long branched mycelial strands of Armillaria

mellea are many intermediate forms, among which are the sheet-

like mycelial masses of Hypholoma perplexum. These have an

outer layer of hard brown cells and a central mass of thin wailed

hyphae bearing, perhaps, a closer resemblance to the structure

of the sclerotium than to that of rhizomorphs. But the sheet-like

masses do not have any definite shape and, as far as this ma¬

terial showed, may be unlimited in their growth — characteristics

not usually associated with a sclerotium. The central portion

of this hyphal mass is formed of parallel hyphae which do not

show any protoplasmic contents. Their walls have become gel¬

atinous and in the triple stain become light blue. Near the sur¬

face on each side there is a layer of thin walled hyphae with ir¬

regular cells. The cells are very closely packed together form¬

ing a pseudoparenchyma. On the outer surface there is a layer

formed by small closely packed cells which are filled with a hard

brown substance. When a carpophore is to be formed some of

the thin walled hyphae force their way through the hard outer

layer where they unite to form the carpophore. Occasionally

the carpophores are formed singly but they are usually in

clusters.

Closer examination shows that the thin walled hyphae just

beneath the surface of hardened cells are formed of short irreg¬

ular cells that contain very little granular cytoplasm. They are

regularly binucleated. The nuclei are large, with a distinct

nuclear membrane and a small nucleole. The chromatin does

not appear as granular as in the majority of nuclei but takes a

uniform light blue stain. The two nuclei are usually pressed close

together near the center of the cell. In forming a carpophore

a number of the thin walled hyphae force their my through the

outer crust, branch profusely at the surface and spread out

40 Wisconsin Academy of Sciences, Arts, and Letters.

slightly to form abroad base for the carpophore. The branches

are short and branched in turn to form a knot of hyphae which is

somewhat similar to the tangled mass formed at the base of the

carpophore of Coprinus. But the cells are shorter and straightcr

so that the hyphae are not so matted and difficult to trace. Above

the tangled hyphae at the base, the branches gradually assume

a more vertical direction. They are still interlaced like the

meshes of a net which has been pulled out lengthwise. Near

the top of the button the number of branches increase and in¬

stead of continuing in a vertical direction they spread out radi¬

ally. This is the first indication of a pileus. The outer branches

of the central system of hyphae are not so closely interlaced but

form a loose open network covering the pileus and stipe still

further to the exterior. The surface of the button is formed

by a layer of nearly straight hyphae. They arise as lateral

branches near the base of the button and covering the entire

surface, disappear among the hyphae at the apex of the pileus

where the loose network of branches is thicker than at the sides.

These hyphae are formed of long slender cells with very little

protoplasmic contents. The shape of the cells and the direction

of growth of the hyphae differentiate this outer layer very

sharply from the tissues in the interior of the button. All of

the cells are binucleated. The nuclei are large and have a

nuclear membrane, a small nueleole and very finely granular

chromatin.

At this time the young carpophore is an oval body and as

indicated its tissues can be separated roughly into three distinct

portions. The central portion formed of closely interlacing

hyphae, which may be still further divided into the base

of closely packed hyphae, the more vertical intervening hyphae

from which the main portion of the stipe is formed, and

the rather loosely spreading mass at the top from which

the pileus develops. Second and outside of this central portion

there is a poorly defined layer of loosely tangled hyphae which

will untimately form the veil above and the outer covering of

the stipe below. This layer is not sharpelv separated from the

central portion and is formed by lateral branches from that

region. The third and outer portion is a layer of parallel hyphae

Nichols — Binuclcaied Cells m Some Basidiomycetes. 41

which form the external covering’. This covering is doubtless

a simple type of volva.

In a carpophore which is a little further developed the cells

above the basal tangle have enlarged and elongated. By the

increase in size of the cells the hyphae are pressed close together

and with the elongation of the stipe are forced into very nearly

vertical series. At the same time the cells in the pileus increase

in size but remain more or less intertwined. There is no differ¬

entiation of hyphae for the formation of gills as yet. As the

stipe elongates the outer covering of parallel hyphae is irreg¬

ularly broken in several places, and now appears as small par¬

ticles or scales which, soon disappear. The protoplasm in the

eells, which are rapidly increasing in length, forms a much thin¬

ner peripheral layer, the central vacuole having increased greatly

in size. A longitudinal section through a young carpophore just

before the breaking of the veil shows the gills already formed.

Sections which pass through the center of a gill show the struct¬

ure very clearly. The hyphae that spread out to form the pileus

are very irregularly twisted about each other. Near the lower

side of the cap there is a narrower layer of much straighter

hyphae which extends from the stipe to the margin of the pileus.

From the lower half of the pileus but especially from this

layer of straighter hyphae, branches are formed that grow down¬

wards in parallel series forming the trama. The hyphae of the

trama branch freely, the branches forming a layer near the sur¬

face of the gills, the subhymenium. The hymenium is formed

of closely packed basidia which are terminal cells of the hyphae

in the subhymenial layer. Fig. 24 is from a section showing

the connection of a basidium with the subhymenial hyphae. The

large basidium has two nuclei lying near together at the center

of the cell which is separated from the stalk cell by a cross wall.

A second basidium is just forming from the basal cell but does

not as yet contain any nuclei.

The hyphae at the margin of the pileus are continuous with

the outer covering of the hyphae on the stipe and form the veil.

The original external covering of parallel hyphae has disap¬

peared with the exception of a few small fragments near the

base of the stipe. All of the cells are regularly binucleated.

42 Wisconsin Academy of Sciences , Arts, and Letters.

When the basidia are formed there is an uninterrupted series

of binueleated cells from the binueleated cells of the selerotium

through the stipe, pileus, trama and subhymenium to the basidia.

Later the cells of the stipe and pileus contain from six to eight

nuclei.

The stages of nuclear fusion and subsequent division in the

basidium have been fully described by Wager, ITarper and

others. In nearly all the essential points Hypholoma perplexum

agrees with the forms studied by these authors. The nuclei are

in the spirem stage at the time of fusion. The fusion nucleus

moves to the summit of the basidium when it divides. The pro¬

cess of the formation of the spindle was not studied. The

spindle is always at a right angle to the long axis of the basidia.

The chromosomes in the equatorial plate show a tendency to

aggregate into a number of larger masses probably as a result

of fixation. In every case there are eight or more. Certainly

there are many more than two as claimed by Maire.

The second division follows the first very closely. The chro¬

mosomes in this case are not scattered on the spindle but defin¬

itely arranged in an equatorial plate. The four nuclei move to

the center of the basidium. At this time or a little later fibres

extending from, the sterigmata to the nuclei at the center were

observed in a large number of basidia, but their origin was not

worked out. As to these fibres Maire says ‘ ‘ Apres la formation

des noyaux-fils definitifs la baside, ceux-si se massent le plus

souvent a la base ou au milieu de la, cellule, tandis que les cent-

rosomes restent au sommet si les mitoses etaient apicales, s’y

rendent dans le cas contraire. En face de chaque centrosome

apparait rebouche se produit une differenciation kinoplasmique

qui s;e propage jusque’ aux novaux at meme quelquefois plus

loin vers la base de la cellule, orientant ■ tant le cytoplasma de

la baside par rapport aux sterigmates. ’ ’

Each spore receives a single nucleus which immediately divides.

In a number of cases the spindle is at right angles to the long

axis of the spore. The chromosomes both at the equatorial plate

and where they are drawn back to the poles appear to be four

or more.

Nichols— B inu cheated Cells in Some Basidiomycetes.

43

Coprinus.

The dung inhabiting C opr ini are so easily obtained from dung

cultures that the spores were not collected and stored as for the

other species studied. The spores were transferred by means

of a sterilized needle directly from the gills to the culture med¬

ium. For the study of the germinating spore and the early

mj^celium the agar hanging drop cultures are the most satis¬

factory. But for the older mycelium and for the formation of

the carpophore agar plates in Petri dishes give the best results.

The spores of Coprinus ephemeras ( ?) germinate in from four

to forty-eight hours, l'n the majority of cases of spores ob¬

tained from mature carpophores germination occurred in six

hours. The spore wall does not swell and become transparent

in germination as in Hypholoma perplexum and it is impossible

to observe the position of the nuclei in the spore. At one end

of the spore there is a thinner place in the wall — the germ pore.

The germ tube appears as a small globular protrusion. One

of the nuclei passes out from the spore at once into the germ tube.

The nuclei divide very rapidly so that the young germ tube

soon contains from three to six nuclei. At this time a nucleus

can frequently be seen in the spore either near its center or

just passing through the pore into the germ tube. At the end

of the spore opposite the germ tube there is a dark hemispherical

body, which perhaps marks the position of a second spore that

has been closed by a cellulose thickening (Figs. 8 & 9).

The tube grows very rapidly forming a main hypha which is

a direct extension from the spore and also a lateral branch which

is nearly at right angles to it. This is well illustrated in Fig.

10 which is drawn from a germinating spore at the end of six¬

teen hours. The main hypha is slightly constricted at irregular

intervals and in nearly every instance it is possible to determine

the presence of a cross wail at the constrictions. The number

and size of the nuclei vary in the different cells. The first and

fourth cells counting from the spore have a very large nucleus

which is more than twice the size of the nuclei in the other cells.

The nucleole is very large and the chromatin is very much denser

than in the nuclei usually found in this species. The other cells

44 Wisconsin Academy of Sciences, Arts, and Letters.

of the main hypha have from one to three small nuclei. The

lateral branch does not show any of the constrictions found in

the main hypha. It is very difficult to locate the cross walls

since when newly formed they are extremely delicate and easily

confused with a strand of cytoplasm. In one of the cells there

are fourteen very small nuclei many of which are only slightly

larger than the nucleole of the large nuclei in the main hypha.

They have a nuclear membrane, a minute nucleole and a few

scattered granules of chromatin. The other cells of the branch

have one or two large nuclei.

At the end of twenty-four hours a well developed mycelium is

formed. Fig. 11 is drawn from a mycelium that was formed in

a hanging drop culture in twenty-four hours. In addition to

the main hypha and the lateral branch mentioned in the pre¬

vious description a second smaller branch is formed close to the

spore. These hyphae branch repeatedly, without any regularity.

The original hyphae are larger and straighter than any of the

lateral branches and for convenience in distinguishing them I

shall call them the primary hyphae. In them the cells are very

unequal in length but the majority are from seven to eleven times

as long as they are broad. The number of nuclei varies also.

The cell next to the spore contains a very large nucleus in the

center and a small one lying at one side near the junction of one

of the branches. The next few (five or six) cells of each of the

primary hyphae have three or four nuclei which are slightly

smaller than the one found in the first cell. The nuclei are usu¬

ally near together at the center of the cell but in one or two in¬

stances one of them is slightly separated from the others. The

next cells for some distance (eight or nine cells) have two nuclei

which may be close together or separated. In the remainder of

the hyphae the nuclei are separated at some distance from each

other but in many cases it is impossible to distinguish a cross

wall between them.

The lateral branches are much smaller and more irregular

than the primary hyphae and the cells are shorter, three or five

times as long as they are wide, and usually contain a single

nucleus although cells containing four or five are frequently

found. In a few instances where the branches extend to the

Nichols-— Binucleated Cells in Some Basidiomyceles. 45

edge of the agar the terminal cells become smaller and contain a

large number of nuclei closely packed together.

Arising from some of the cells near the base of the primary

hyphae are short thick eonidiophores. Their cells are nearly as

broad as they are long. The end cells bear one or two very short

branches which produce clusters of very slender hyphae formed

of cells two or three times as long as broad. The cells separate

readily into rod shaped oidia (Fig. 12). These hyphae do not

appear above the surface of the substratum.

Near the center of the oidiurn there is a small well defined

nucleus in which the nueleole and granular chromatin can be dis¬

tinguished. The remainder of the ceil is filled with finely vacu¬

olated cytoplasm. This is in marked contrast with the form

found by Maire on C opr inns radiaius. lie says ‘ ‘ Chacpie oidie

renferme un seul noyau reduit d ’ordinaire a une petite masse

homogene chromatique et un grande vacuole.” On account of

the number of oidia I was unable to see the nuclei in the cells of

the supporting branch.

Occasionally the cells of the main hypha are binucleated. In

this case the first few cells near the spore have from one to five

nuclei. The cells further from the spore are regularly binucle¬

ated. At the end of five or six days branches appear on some of

the older cells near the spore that are much larger than any pre¬

viously formed and are frequently club-shaped. The cells of

these branches are binucleated, the nuclei lying near together,

sometimes side by side, near the center of the cell. The branches

with binucleated cells may be formed from any cell of the myce¬

lium near the spore. Fig. 13 shows one of the branches with

binucleated cells formed on a hypha with uninucleated cells.

The branch is formed very near the center of the cell instead of

at one end as is usual with the mycelial branches. In my opin¬

ion the newly formed branch probably received a single nucleus

which immediately divided. Fig. 14 is drawn from a hypha

with multinucleated cells showing one of the larger branches just

forming. Two nuclei of the multinucleated cell are just at the

base of the newly forming branch.

Fig. 15 shows a group of these branches closely crowded to¬

gether — the first indication of a fruit body. The hyphae have

46 Wisconsin Academy of Sciences, Arts, and Letters.

many branches spreading outwards. In the center the hyphae

are so densely packed that it is impossible to tell whether the

branches originate from a single hyplia or from several. As a

number of hyphae from different directions in the substratum

lie near together at this point it is probable that they arise from

several.

The cells are regularly binucleated. The cytoplasmic strands

are very delicate and form a loose network at the periphery of

the cell. Very small convex plates are present at the cross wails

marking the regions of protoplasmic connection between the cells.

This method of formation of the carpophore was repeatedly ob¬

served in the hanging drop cultures. It is probably an adapta¬

tion to the culture conditions, especially the small supply of

nutrient. In larger cultures of agar-agar in Petri dishes I find

quite another method for the formation of the carpophore.

In an agar culture in a Petri dish the main hyphae with their

system of branches can frequently be traced for one or two

inches. Pig. 16 was made from the end of one of these systems.

Squares of the agar a little smaller than a % inch cover glass

can be fixed in Flemming’s weaker solution and then fastened

to a slide with a film of albumin. Preparations are best stained

with iron haematoxylin. The stain washes from the agar in the

iron solution leaving the mycelial hyphae sharply defined so that

they can easily be studied.

The main hyphae are formed of very long cells which have

from five to eleven nuclei that are smaller than in Hypholoma,

but show the same general structure, they are elliptical with a

nucleole at one end, finely granular chromatin, and a small body

which is usually opposite the nucleole. This small body is flat¬

tened against the nuclear membrane and takes a very deep blue

stain. The great regularity of its occurrence and its staining

qualities have led me to consider this as the same structure as

the central body which Harper described in the ascospores of

Erysiphe. The primary branches which arise from the main

hyphae show great regularity of arrangement. They are formed

in pairs at one end of every cell or every other cell. They are

never exactly opposite, one always appearing slightly below the

other. These branches are approximately of the same size as

Nichols- —Binucl cal ed Cells in Home Basidiomy cedes. 47

the main hypha and their base is enlarged in the same fashion as

is the base of the germ tube. The young branch receives a single

nucleus from the parent cell. A cross wall is then formed sepa¬

rating the branch from the main hypha while it is still very

small. The nucleus divides very rapidly so that the young

branch soon contains two or three nuclei. These continue to

multiply until the mature cells contain as many or nearly as

many nuclei as the cells of the main hyphae. The clamp con¬

nections between adjacent cells are quite abundant although they

are not formed at every cross wall. They are cut off from one

of the cells by a cross wall which always shows the deeply stain¬

ing convex plates originally described by Hoffman (13) and

since by Strasburger (24) and other authors. In every instance

they show the same cytoplasmic structure as the cells which they

connect. I have found one or two cases in Coprinus in which a

branch has arisen from one of these clamp connections.

The secondary branches which arise from the primary hyphae

do not show the same regularity m their formation. Only one

is formed from a cell. They are very long and slender and run

in every direction through the substratum. They anastomose

with other branches frequently thus increasing the complexity of

the hyphal system. The nuclei are separated by long distances,

they are elliptical with a slightly larger nucleole than is found in

the nuclei of the main hyphae. The cross Avails are difficult to

locate AAdien, as frequently happens, the convex plates at the

cross walls are absent.

The mycelium does not appear at the surface of the culture

until a fruit body is to be formed. At this stage the character

of the branching of the mycelium changes. The cells are shorter

and branch three or four times, each successive cell branching in

the same way. The cells are also curved backward towards the

main hypha, thus a tangled mass is formed (Fig. 17). Other

hyphae from various directions pass into this snarl and branch

in the same way increasing its size and complexity. Fig. 17a is

from one of these tangled masses and shows the Y shaped cells

Avhich are frequently found near the center. The cell walls are

thick and tinged with yellow. The clamp cells are found at

every cross wall connecting the adjacent cells.

48 Wisconsin Academy of Sciences, Arts, and Letters.

All of the cells contain two nuclei which may lie close together

but are more frequently separated by the entire length of the

cell. In the Y shaped cells I have generally found one nucleus

in each of the arms.

In my cultures the hyphal masses developed into carpophores

only when they were formed near the surface or some break in

the agar. A large number of small tangled masses were formed

throughout the cultures but they did not develop into carpo¬

phores. Fig. 18 is from a section through a young carpophore

which is developing from one of these tangles. The mass* of

hyphae is much larger than in the earlier stages but it shows the

same structure. A large number of hyphae are found converg¬

ing from different directions in the substratum to form by the

interlacing of their branches near the surface of the agar this

large tangled mass. The hyphae are so closely packed and so

crooked that it is impossible to trace the cells their entire length

but as far as I can determine they are in all cases binucleated.

The hyphae outside of the tangled mass that are growing towards

it and whose branches form the tangle invariably have binu¬

cleated cells. In the younger tangles the cells are binucleated

and in this large hyphal mass the nuclei are usually found in

pairs.

A more complete series of young carpophores of Coprimes

ephemerus were obtained from dung cultures in the green house.

By fixing the young carpophores when they first appear as

small white dots on the substratum the main stages in the early

development of the fruit body are very easily obtained. They

are very similar to the corresponding stages of Coprinus ster-

corarius as described by Brefeld.

The very young carpophore is an oval body formed by a cen¬

tral mass of closely tangled hyphae and an outer covering of

very loosely interlaced hyphae. The hyphae near the center are

copiously branched. The branches branch in turn and gradually

force their way upward and outward, the density of the tangle

becoming less as they grow farther from the center until they

are only loosely woven together at the surface forming the outer

covering. The cells of all the hyphae are very short and are

densely filled with a finely vacuolated cytoplasm. They are reg-

Nichols — Binucleated Cells in Some Basidiomycetes. 49

ularly binucleated, the nuclei showing the same structure as

those found in the mycelium.

A section through a carpophore which is more mature shows

the hyphae in the central portion arranged in parallel series ver¬

tical to the base. At the center there are a number of slender

hyphae. Near the top these slender hyphae spread out and

slightly downward to form the young pileus. The outer network

of loose hyphae still encloses the growing cap and stipe. It is

scarcely attached to the cap except near the apex where a few

slender hyphae from the center of the pileus pass out into it.

Below the cap it is continuous with the hyphae of the stipe. At

this time previous to the formation of the gills the cells of the

stipe have from four to eight nuclei. The cytoplasm is filled

with large vacuoles.

Maire has described the formation and structure of the gills

of Clitocyhe aurantiaco Mycena galericulata, S Iropharia semi-

globata, ITypholoma appendiculata, and Psathyrella disseminata.

He finds that when the basidia are formed the cells of the trama

are binucleated but in the stipe and pileus the cells are multinu-

cleated. Harper has found this* true also for Coprinus ephe¬

meras.

I have also studied the young carpophore of other forms with¬

out growing them from spores to determine the arrangement of

the nuclei in the young carpophore and in the mycelium in the

substratum.

A species of Crepidoius was found growing on a rotten log in

the green-house during the winter. The specimens were fixed

with a portion of the substratum. For the most part the hyphae

fill the vascular elements of the wood densely and are difficult

to distinguish. But there are also single hyphae which grow in

some of the large ducts that are favorable for study. The cells

are very poor in protoplasmic contents. Two nuclei lie close to¬

gether near the center of the cell surrounded by a little granular

cytoplasm. At the surface the mycelial hyphae come together

and twine around each other forming a short tangled mass which

is the base of the carpophore. The hyphae branch profusely

and curve upward growing in a vertical direction to form the

4

50 Wisconsin Academy of Sciences, Arts, and Letters.

stipe. At the top of the stipe they spread out in every direction

to form the p ileus and gills. The cells are regularly binucleated

and are filled with very dense cytoplasm.

Certicium lilacino-fuscum was also studied. It appeared in

great abundance on a damp log in the green-house. Portions a

quarter of an inch square were removed with a thick layer of the

substratum and fixed in Flemming’s weaker solution.

The surface layer is formed of densely packed short celled

hyphae which grow perpendicularly to the substratum. The

hyphae branch repeatedly, the branches crowding in between the

hyphae. The cells are long and slender and so closely crowded

together that I could make out the contents in only a few cases.

In favorable preparations in a few instances cells were seen with

two nuclei near the center and the remainder of the cell was

filled with very granular cytoplasm. The clamp connections be¬

tween adjacent cells are nearly always present. On the lower

surface there are a large number of thick walled hyphae, desti¬

tute of cytoplasm, which form a loose felted layer and then

penetrate the wood for some distance. They probably serve as

a protective layer and also hold the mat close to the substratum.

Among these basal thick walled hyphae there are others with thin

walls and protoplasmic contents which also penetrate the wood

in every direction and are usually single but occasionally are

found in masses filling a vascular element of the wood. They

are formed of long slender ceils with two nuclei.

Rhizomorphs.

In many Agarics the hyphae form mycelial strands which

often grovr to great length. These strands may branch fre¬

quently but without regularity and resemble roots. In some

species the branches anastromose frequently forming a loose net¬

work; a character which is especially pronounced in DictyopJiora

duplicata. The structure of these so-called rhizomorphs is best

known through the descriptions of Armillaria mellea by De

Bary, Goff art and others. The forms wdiich I have studied show

a much simpler structure but they apparently have the same

origin, the same general character of growth and doubtless serve

the same general function. They provide a storage for reserve

Nichols — Binuclealcd Cells in Some Basidiomycetes. 51

material, a means of spreading the species in the substratum and

are capable of retaining their vitality during a long dormant

period.

Poria. (Fig. 19.)

The rhizomorphs of a species of Poria were collected in April

from the underside of a board sidewalk where they had hiber¬

nated. Thus they were in a resting rather than the active

growing condition of the other species described later. The

main strands were about an eighth of an inch in diameter.

These branch dichotomously a few times and from them a num¬

ber of very slender branches are formed. These branches ex¬

tend in every direction over the substratum; they subdivide

freely, the branches frequently anastomosing and flattening out

to form web-like expansions. The younger portions of the

strand are a dull yellow while the older portions are coated with

a yellowish brown layer.

Their structure is very simple. The entire strand is formed

of slender hyphae which are very straight, only twining about

each other slightly. At the center a large number of the hyphae

have lost their contents and the walls are very thick. But

among them are other hyphae with thin walls and rich proto¬

plasmic contents. The cytoplasm is filled with small deeply

staining bodies but there are no crystals. Towards the surface

these thin walled hyphae become fewer and finally disappear

leaving the outer layer formed wholly of thick walled hyphae

(Fig. 19a). In a number of the thin walled cells two nuclei

were found. They were very small and dense lying near to¬

gether at the center of the cell. In some cases the deeply stain¬

ing granules were so abundant and so closely packed that it was

impossible to distinguish the nuclei. Among the hyphae there

were numerous openings or pockets filled with octahedral crys¬

tals which dissolve slowly in hydrochloric acid indicating that

they are calcium oxalate (Fig. 19c). Just beneath the surface

layer of thick walled hyphae there is a layer of similar crystals

very closely packed together (Fig. 19b).

52 Wisconsin Academy of Sciences , Arts , and Letters.

Pholiota praecox. Pers. (Fig. 20.)

The mycelial strands of Pholiota praecox were collected from

a mass of decaying leaves. A large number of mature caps were

removed with all the connecting mycelial strands. The rotten

leaves wrere easily pulled away leaving the strands free from all

foreign matter. The strands were cut into pieces one fourth

inch in length from the tip and from the older parts which show

the branching.

The outside of the strand is covered by a thin layer of loosely

tangled thick walled hyphae which doubtless serves as a protec¬

tive layer. Beneath this is the main portion of the strand which

has the same structure throughout with no differentiation into

medulla and cortex. It is formed principally of large thin

walled hyphae which show some slight regularity of' * arrangement

although the longitudinal rows of cells so pronounced in Armiil-

aria were not found. In some parts of the strands there are

a few central hyphae that run very nearly straight. On both

sides of this central strand the hyphae grow obliquely towards

the surface. In other parts bundles of thirty or more hyphae,

slightly entwined with one another coil around similar bundles

or separate hyphae. Frequently it was impossible to make out

any regular arrangement. The great majority of hyphae are

formed of large cells but arising as secondary branches from

them are a few narrow celled hyphae which run in the same di¬

rection as the larger hyphae until they reach the surface. At

the surface instead of turning back towards the center as do the

larger hyphae their walls become thickened and they form the

felted layer. Thus the outer felt consists of the thickened term¬

inal portions of the slender hyphae of the interior.

It is only by very careful staining with the triple stain that I

was able to make out the contents of the cells. The walls of the

large cells are partly gelatinized and have lost their smooth out¬

line becoming wrinkled and creased. There is a very thin layer

of cytoplasm with small light staining granules. The nuclei are

flattened against the walls by the central vacuole. They are

large and distinct with a very small nucleole on the side towards

one end of the cell and a deeply staining body, probably a cen-

Nichols — Binuclealed Cells in Boyne Basidiomyceles. 53

tral body, on the opposite side. The remainder of the nucleus is

filled with a finely granular* chromatin which takes a stain that

is only slightly darker than that of the granules in the surround¬

ing cytoplasm from which it is sharply separated by the nuclear

membrane. All of the cells contain two and only two nuclei

which may lie near together or separated by the entire length of

the cell. Occasionally the cells had one or two small crystals.

The slender hyphae are formed of very long narrow cells

which are filled with very dense cytoplasm. The nuclei are large,

occupying nearly the entire diameter of the cell and lie close

together at the center. When these hyphae reach the surface of

the strand their walls become thickened leaving only a very small

central strand of cytoplasm.

All of the cells have deeply staining convex plates on opposite

sides of their cross walls. These convex plates always appear in

pairs with their concave surfaces towards each other.

Lepiota naucina. (Fig. 21.)

The mycelial strands of Lepiota naucina are very similar in

appearance to those of Pholioia. They were found growing in

the newly deposited mulch around the base of some young trees.

It was impossible to obtain the mature carpophore attached to

the strands but by loosening the substratum carefully I could

obtain long strands still attached to the young buttons.

The structure of the strands differs in a few respects from

those just described. In the interior of the strand the hyphae

do not showT any regularity of arrangement. The large celled

hyphae are separated by a number of fine branches and are much

fewer than in Pholioia. They are formed of large cells, with

slightly rounded ends where the cross walls are formed. The

slender hyphae arise as lateral branches of the larger ones and

are very abundant. They wind among the larger hyphae form¬

ing a dense weft. Near the surface there is a layer of several

cells thick entirely composed of slender hyphae. The cells are

much shorter than in the center of the strand and each cell

branches so that a close net -work is formed. Outside of this

there is a felted layer of thick walled hyphae. The thick w'alled

54 Wisconsin Academy of Sciences, Arts, and Letters.

hyphae were in this case so closely packed that it was impossible

to trace their origin.

The contents of the cells are very similar to those of Pholiota.

The large cells have a very thin peripheral layer of cytoplasm

which contains a large number of small very deeply staining

granules which may be very minute crystals. Larger crystals

are very abundant, many of the cells containing twenty-five or

thirty. They vary from the smallest size determinable to those

which are one fourth of the diameter of the cell. They take a

bright red color in the triple stain and are octahedrons in the

majority of cases but a few appear to have more faces. Because

of the numerous dark staining bodies, it was frequently impossi¬

ble to differentiate the nuclei but in the cells where there were

very few if any crystals or granules two nuclei were found. The

large cells have wide pits in their walls through which it is easy

to trace a protoplasmic strand. In a few cases the pore is so

large that it is possible to trace a connection not only between

the peripheral layers of cytoplasm but also between the central

vacuoles. The protoplasmic granules frequently accumulate

around these pores in large numbers but the convex plates de¬

scribed above are not formed. The slender hyphae are filled

with a finely granular cytoplasm without any crystals or large

granules. The cells are regularly binucleated. The convex

plates are always present at their cross walls.

Dictyophora duplicata. Ed. Fisch. (Fig. 22).

The mycelial strands of Dictyophora duplicata were collected

from the beds of mulch around the base of young trees on the

University drive. By loosening the soil thoroughly large masses

of the strands were obtained still in connection with young car¬

pophores. The strands are fine and delicate with numerous ir¬

regular swellings. They branch frequently and the branches

sometimes anastomose forming a very loose network.

The structure of the strands is decidedly different from the

types already described. The very center of the strand is occu¬

pied by a few large hyphae which twine about each other

slightly. They are formed of very long cells which have lost all

of their contents and their walls are slightly gelatinized.

Nichols — Bmucleatcd Cells in Some Basidiomycetes. 55

Towards the surface these large hyphae become much fewer and

are separated by other hyphae wThich are only slightly smaller

but very rich in protoplasmic contents. This second form of

hyphae are the lateral branches of the first and correspond to

the slender hyphae of the other species but are very different in

appearance. They are loosely twined together towards the in¬

terior, forming an open mesh work through which a few of the

larger hyphae run and also leaving numerous air-spaces.

Towards the outside they become more and more closely woven

together until they form a layer of densely packed hyphae just

beneath the surface. Their cells are not so long as in the large

hyphae but are nearly as wide in many places. They are irreg¬

ular in width, especially towards the surface and contain large

numbers of deeply staining bodies probably of a proteid char¬

acter. These bodies take a deep blue color with the triple stain

which requires a long exposure to Orange 0. to remove. On ac¬

count of the number of these dark bodies it was possible to locate

the nuclei in only a few of the cells. In favorably stained cells

there were two large nuclei with the usual structure, a small

nucleole, and very finely granular chromatin, surrounded by a

nuclear membrane with a deeply staining body at one side.

The hyphae. do not have their walls thickened when they reach

the surface as in all of the preceding species but their cells be¬

come shorter and very vesicular. The hyphae are very loosely

wroven together forming a very thick outer layer. The cells do

not contain any of the dark bodies so abundant in the layer just

beneath, but have a single large nucleus, which is very irregular

in shape, near the center of the cell. This large nucleus is im¬

bedded in a central mass of cytoplasm from which strands radi¬

ate in every direction towards the walls, as is the case in

Spirogyra cells. I did not find any crystals in any of the cells

or any convex plates at any of the cross walls.

Lycoperdon pyriforme. Schaeff. (Fig. 23.)

The mycelial strands of Lycoperdon pyriforme were found

growing in a rotten stump. By breaking off pieces of the wood

large quantities of the strands were obtained free from foreign

56 Wisconsin Academy of Sciences , Arts, and Letters.

€

matter. The strands are fine and very tough, in marked con¬

trast to the two preceding species.

In the center of the strand the hypliae run very nearly

straight only slightly twining around each other. The large

celled hyphae are formed of very long cells so that it was only

occasionally that I found a cross wall. The walls are very deli¬

cate and do not show either a pore or the lens shaped plates.

The cytoplasm forms a peripheral network of deeply staining

strands which are very irregular and contain a number of large

round granules. Only one or two octahedral crystals were found

in any of the cells. The cells contain two large nuclei which

take a very light stain as they have apparently only a small

amount of chromatin. The nuclear membrane is very distinct

and clearly defines the nucleus. The nucleole is very small.

The slender hyphae are very abundant and form the larger

portion of the strand. They arise as lateral branches of the

larger ones and twining among each other in a slightly oblique

direction gradually work their way to the surface. At the sur¬

face they form a layer of densely packed hyphae by the inter-

Jacing of their branches. On the outside of this layer the hyphae

are more loosely wmven together and their walls become gradu¬

ally thickened to form. a. protective covering. The cells of the

fine hyphae are long and slender in the central portion of the

strand, becoming much shorter as they approach the surface.

At every cross wall very large distinct convex plates are present.

The cytoplasm forms a finely granular layer at the periphery.

There are always two large distinct nuclei near the center of each

cell.

Section which pass through the main strand and the base of a

lateral branch showr that the hyphae which form the branch do

not arise in any very definite fashion. Hyphae from various

parts of the central and outer layers curve in the same direction

towards one side and unite at the surface to form a branch. In

one case I was able to trace a hyplia which was growing on the

other side of the strand, until it was nearly oposite the branch,

where it curved sharply, crossed the main strand and entered the

branch on the other side.

Nichols — Binucleated Cells in Some Basidiomycetes.

57

CONCLUSIONS.

The results obtained from the above studies show that the bi-

nueleated cells do not originate through the formation of any

special reproductive apparatus. Neither is there any structure

in the formation of the hyphae or in the cells of the hyphae that

indicates in any way where the binucleated cells will first ap¬

pear. That they arise previous to the formation of a carpophore

in a large number of forms is very evident. The long series of

binucleated cells found in the mycelial strands or rhizomorphs of

Poria, Pholiota, Lepiota, Dictyopliora and Lycoperdon show that

in these forms the origin of the binucleated cells is only distantly

associated with the formation of a carpophore and it is probable

that many strands never develop carpophores. In Hypholoma

verplexum there is an expanded mass of mycelium with binu¬

cleated cells formed previous to the appearance of a carpophore.

In many forms in which there is not any specialized structure

between the filamentous mycelium and the carpophore, as in

Crepidotus, Corticium, and in Coprinus ephemerus it was found

that the cells of the mycelium in the substratum in the region of

the carpophore were regularly binucleated. Binucleated cells

were also observed in the cells of the mycelium of Ilypochnus by

Harper. As was mentioned previously, he w'as able to trace the

series of binucleated cells from the hymenium into the mycelium

in the substratum. The above cases show' that the formation of

binucleated cells is not necessarily followed immediately by the

formation of a carpophore.

Further in the fairly complete series of stages from the germi¬

nation of the spore to the mature carpophore vdiich wrere ob¬

tained for Hypholoma perplexum and Coprinus ephemerus the

same fact appears. A summary of the development of these

species is as follows:

The spores of both species contain two sister nuclei which at

germination pass from the spore into the germ tube. In Coprinus

the nuclei divide very rapidly, forming a multinucleated germ

tube which may contain twenty or thirty' nuclei before the first

wall is formed. The nuclei do not divide as rapidly' in the germ

tube of Hypholoma so that it seldom contains more than six or

58 Wisconsin Academy of Sciences, Arts, and Letters.

eight when the first wall is formed. In the germ tube of

Eypholoma the nuclei divide independently as was shown by the

appearance of a single nucleus dividing in one of the germ tubes.

A septate mycelium is formed in Ilypholoma, the cells of which

contain from one to nine or ten nuclei. The mycelial cells of

Coprinus frequently contain as many as fifteen nuclei. Lateral

branches are formed in both Eypholoma and Coprinus, the cells

of which have a single nucleus. The cells are short and separate

from each other readily. In Eypholoma these cells germinate

forming a normal mycelium, thus proving conclusively that they

are oidia. In Coprinus the cells did not germinate but their

similar origin and structure indicate that they also are oidia.

Apparently after a varying period of growth hyphae are

formed which have regularly two and only two nuclei in their

cells. In the forms studied this association in pairs does not

• , # r ^

arise at any otherwise differentiated time or in any special

structure.

We must remember as described above in Coprinus that binu-

cleated cells were sometimes formed very near the spore, the

main hyphae at a distance of two or three cells from the source

of its origin having regular binucleated cells. In other cases

special branches having binucleated cells were formed. When

these branches arise from the uninucleated cells they probably

receive a single nucleus which immediately divides. This divi¬

sion is not followed by a cross wall separating the two nuclei

and thus the first binucleated cell is formed. If the branch hav¬

ing binucleated cells originates from a multinucleated cell it may

receive two nuclei from the parent cell or it may receive a single

nucleus that divides to form the second nucleus.

The hyphae with binucleated cells branch frequently and may

develop immediately into a carpophore. This method of carpo¬

phore formation has been described for Coprinus ephemerus .

The cells of the young carpophore are always binucleated and a

series of binucleated cells extends to the formation of the

hymenium. Thus the basidium with its two nuclei is the last

of a long series of binucleated cells. That these two nuclei have

maintained distinct lines of descent from the first cells with

two nuclei occurring in the mycelium is not certain. I have not

Nichols — Binucleated Cells in Some Basidiomyceles. 59

found evidence of conjugate division in tlie hyphae but it seems

possible that it exists and that the two nuclei of the young

basidia have remained distinct throughout the series of binu¬

cleated cells.

The evidence seems satisfactory that in the rusts the series of

nuclei in the binucleated cells are formed by the association of

two nuclei in so called conjugate division. A wall is formed be¬

tween the two pairs of daughter nuclei. Thus the two nuclei in

a cell are not sister nuclei. The nuclei which fuse in the teleuto-

spore have thus remained distinct through generations of binu¬

cleated cells. The wide occurrence of a series of binucleated

cells in the young carpophore suggests a method of division sim¬

ilar to that in the rusts. Maire believes that the series of nuclei

in the binucleated cells of the forms studied by him are formed

by conjugate division but his figures do not seem conclusive. In

Hypholoma and Coprmus each of the four spores borne on the

basidium receives a single nucleus which immediately divides

forming a binucleated spore. Maire states that a similar process

occurs in Clavaria vermicularis, Clitocyhe aurantium, Mycena

galericulata, and Amanita pantherina. This resemblance be¬

tween the nuclear history of the Uredineae and of the Basidiomy -

cetes has been discussed fully by Maire, Harper and others.

In view of the fact that the hyphae with binucleated cells are

found originating from uninueleated cells in Coprinus, it is very

possible that further study may reveal some forms in which the

mycelium will show regularly uninueleated cells and that this

uninueleated series will extend to the first formation of a carpo¬

phore.

At present there is no evidence that the binucleated cells of

the Basidiomyceles ever originate by a fusion of two adjacent

cells such as Blackman (2) finds at the base of the aecidium in

Phragmidium violaceum and Gymnosporangium clavariaeforme.

The occurrence of regular binucleated cells through a large

part of the life history of the Basidiomyceles leading to the

formation of basidia while binucleated cells are unknown in the

life history of the Ascomy cetes makes it difficult to consider the

two groups as phylogenetieally related. Maire (16) states that

binucleated cells occur in the ascogenous hyphae in Pustularm

60 Wisconsin Academy of Sciences, Arts, and Letters.

vesiculosa, Galactinia succosa and Ace tabula acetabulum. But

this hardly seems adequate proof for the conclusion he suggests.

The origin of the ascocarp in a sexual process is in marked con¬

trast to the origin of the carpophore as described above. It

seems difficult to imagine that they are homologous structures.

Certainly neither asci nor basidia can be considered as oogonia

as Dangeard proposes.

That true cell fusion does not occur in Coprinus is very evi¬

dent. It is doubtful husv far this fact should influence our in¬

terpretation of the fusion of nuclei in the basidium. Two nuclei

of more or less widely separated origin fuse and this is at least

a common characteristic of sexual fertilization. Raciborski (19)

believes that the binucleated cells of the rusts represent a pro¬

longed vegetative stage intercalated between the cell fusion and

the nuclear fusion just as in the zygospore of- Basidiobolus a

period may intervene between cell and nuclear fusion. He pro¬

poses the terra zeugite for all cells in which occurs a fusion of

nuclei belonging to the same cytoplasmic mass. But the fact

that the binucleated cells of the Basidiomycetes are not the result

of actual cell fusion makes it difficult to compare directly the

delayed fusion in the zygospore of Basidiobolus and the nuclear

fusion in the basidium.

Maire proposes to distinguish two types of fusion: “La

sexualite avec fecundation ’ ’ and ‘ ‘ la sexualite avec mixie. ’ ’ The

first type is found in the higher plants and animals, where the

nucleus resulting from the fusion of two sexual nuclei contains

twice the number of chromosomes contained in either of the

fusing nuclei. The second type he believes is found in the lower

plants and in the basidium. In this type the nucleus result¬

ing from the fusion of two nuclei has the same number of chro¬

mosomes as was contained in each of the fusing nuclei.

Maire describes the nuclear fusion and division in the basidia

in Mycena galeriadata, Psatkyrella disseminata and a large num¬

ber of other Basidiomycetes. In all of these species he describes

the two conjugate nuclei as each having two chromosomes and

states that the fusion nucleus of the basidium in the first divi¬

sion has also only two chromosomes. But his figures are some¬

what diagrammatic and decidedly inconclusive. On this evi-

Nichols - Binucleated Cells in Some Basidiomycetes. 61

dence Maire has formulated a theory of the alternation of gener¬

ation in the Basidiomycetes. He believes that the origin of

the binucleated cell is comparable to the fertilization of the

higher plants. The carpophore having cells with paired nuclei

corresponds to the .sporophyte. The fusion of nuclei in the basid-

ium is then not a fertilization but a reduction-process. Black¬

man’s discovery that in the rusts the binucleated cells arise by

an actual cell fusion is strong evidence in favor of this view

and Blackman develops the conception into very satisfactory

form so far as the rusts are concerned. It is possible that the

origin of binucleated cells in the ordinary course of mycelial

growth as described above may have come to serve as a sub¬

stitute for cell fusions and that thus the stages of development

with binucleated cells in the Basidiomycetes may be considered

the equivalent of a sporophyte. Since, however, the point or

points at which binucleated cells arise is very variable such a

sporophyte can by no means be considered as representing a so

definitely differentiated stage of development as does the sporo¬

phyte of the moss or fern.

62 Wisconsin Academy of Sciences, Arts, and Letters,

BIBLIOGRAPHY.

1. Bambeke, Ch. Van. L ’Evolution Nucleaire et la Sporula-

tion chez Hydmangium Carenum. Bull. L’Acad. Roy.

d.Belgique 6: pp. 515-528. 1903.

2. Blackman, Y. H. On the Fertilization, Alternation of

Generations and General Cytology of the Uredineae, New

Phytologist 3: pp. 23-27. 1904.

3. Brefeld, 0. Botanische Untersuchungen liber Schimmel-

pilze. III. Heft. Basidiomycetes. Rev. in Bot. Jahresber

5 : pp. 131-135. 1877.

4. Brefeld, 0. Die Entwickelungsgeschichte der Basidiomy-

ceten. Bot. Zeit. 34: pp. 49-62. 1876.

5. Dangeard, P. A. Memoire sur la Reproduction SexueUe

des Basidiomycetes . Le Bot. 4 : pp. 119-181. 1894-5.

6. Dangeard, P. A. La Reproduction Sexuelle des Cham¬

pignons. Etude Critique. Le Bot. 7 : pp. 89-130. 1900.

7. De Bary, A. Comparative Morphology and Biology of the

Fungi, Mycetozoa and Bacteria, pp. 22-59, Oxford. 18S7.

8. Falk, Richard. Die Cultur der Oidien und ihre Riickfiihr-

ung in die Hohere Fruchtform bei den Basidiomyceten.

Beitr. z. Biol. d. Pfl. 8 : pp. 307-342. 1902.

9. Ferguson, Margaret C. Germination of the Spore of

Agaricus campestris and other Basidiomycetous Fungi.

U. S. Dept. Agric. Bureau of Plant Industry, Bull. No.

16. 1902.

10. Harper, R. A. Kerntheilung und freie Zellbildung im

Ascus Jahrb. f. Wis. Bot. 30 : pp. 249-282.

11. Harper, R. A. Nuclear Phenomena in Certain Stages in

the Development of the Smuts. Trans. Wis. Acad. Sci.

12: pp. 475-498. 1900.

Nichols — Binuclealed Cells in Some Basidiomy cotes. 63

12. Harper, R. A. Binucleate cells in certain Ilymenomycetes.

Bot. Gaz. 33 : pp. 1-23. 1902.

13. Hoffman, H. Die Pollinarin und Spermatien von Agar-

icns. Bot. Zeit. 14: pp. 137-148. 1856.

14. Holden, R. J. & Harper, R. A. Nuclear Division and Nu¬

clear Fusion in Coleosporium Sonchi-arvensis. Trans.

Wis. Acad. Sci. 14: pp. 63-77. 1903.

15. Maire, Rene. Recherches Cytologiques et Taxonomiques

sur les Basidiomycetes. Theses Presentees a la Faculte

des Science de Paris. Serie A., No. 429.

16. Maire, Rene. Recherches cytologiques sur le Galactinia

succosa.

Compt. Rend. 137 : pp. 769-771. 1903.

17. Maire, Rene. La formation des asques chez les Pezizes et

P evolution nucleaire des Ascomycetes. Conipt rend,

d. I. Soc. d. Biol. Rev. in Bot. Centbl. 95: pp. 149.

1904.

18. Petri, L. La formazioni della Spore nell’ Hydnangium

Carneum b. Nuovo Giornale Bot. Italione, 499 p. 1902.

19. Raciborski, M. Ueber den Einfluss ausserer Bedingungen

auf die Wachthumsweise des Basidiobolus ranarum.

Flora. 1896.

20. Rees, M. fiber den Befruchtungsvorgang bei den Basidi-

omyceten. Rev. Bot. Jalirb. 3 : pp. 209-216. 1875.

21. Rosen, F. Studien liber Kern und Membranbildung bei

Myxomyceten und Pilze. Beitr. z. Biol. d. Pfl. 6 : pp.

237-264. 1893.

22. Rosenvinge, M. L. Iv. Les Noyaux des Ilymenomycetes.

Ann. d. Sci. Nat. Bot. VII, 3 : pp. 75-91. 1886.

23. Ruhland, W. Zur Kenntniss der intracellularen Karyo-

gamie bei den Basidiomyceten. Bot. Zeit. 59 : pp. 187-

204. 1901.

24. Strasbltrger, E. Bot. Pralctikum. 325 p.

64 Wisconsin Academy of Sciences, Arts , and Letters.

25. Van Tteghem. Sur le Developpenient du Fruit des

Coprins el la Pretendue Sexualite des Basidiomyceies.

Comp. Kend. 80 : pp. 373-379. 1875.

26. Wager, H. On Nuclear Division in the Hymenomycetes .

Ann. Bot. 7 : pp. 489-515. 1893.

27. Wager, H. On the Presence of Centrospheres in Fungi.

Ann. Bot. 8 : 322 p. 1894.

28. Wager, H. The Sexuality of the Fungi. Ann Bot. 13:

pp. 575-597. 1899.

Nichols — Binudeated Cells in Some Basidiomycetes.

PLATE IV.

66 Wisconsin Academy of Sciences, Arts, and Letters.

EXPLANATION OF PLATE IV.

All figures were drawn with the aid of the Abbe camera lucida.

Fig. 1.

Fig. 2.

Fig. 3.

Fig. 4.

Fig. 5.

I

Fig. 6.

Fig. 8.

Fig. 9.

Fig. 10.

The spore of Hypholoma perplexum previous to germination.

X 2700.

Spore of Hypholoma perplexum with a short germ tube which

has received a single nucleus from the spore. X 2700.

Spore of Hypholoma perplexum with a short germ tube which

has received both nuclei from the spore. X 2700.

Same as Fig. 2.

The nuclei :a germ tube of Hypholoma perplexum show late

division figu/es. X 2700.

One nucleus, in a germ tube of Hypholoma perplexum , is in

the equatorial plate stage of division. X 2700.

Spore of Caprinus ephemerus with a short germ tube contain¬

ing several nuclei. X 2700.

Same as Fig. 8.

A germinating spore of Caprinus ephemerus at the end of six¬

teen hours. X 2700.

Trans. Wis. Acad., Vol. XV.

Plate IV.

S. P. Nichols, del.

Nichols-

-Binuciealcd Cells in Some Basidiomyceles. 67

PLATE V.

68 Wisconsin Academy of Sciences, Arts, and Letters .

Fig. 7.

Fig. 11.

Pig. 12.

Fig. 13.

Fig. 14.

Fig. 16.

Fig. 18.

Fig. 24.

EXPLANATION OF PLATE V.

The mycelium of Hypholoma perplexum. X 1200.

Mycelium of Caprinus ephemeras. X 850.

Oidial branches of Caprinus ephemerus. X 1500.

A branch with binucleated cells formed on a hyp ha with uni-

nucleated cells. X 1500.

A large branch forming on a hypha with multinucleated cells

X 1500.

Mycelium of Caprinus ephemerus showing the arrangement of

nuclei in the cells. X 1200.

Section through the base of a young carpophore of Coprinns

ephemerus. X 850.

A section of Hypholoma perplexum showing the connection

of a basidium with the subhymeniun. X 1500.

Trans. Wis. Acad., Vol. XV.

Plate V.

S. P. Nichols del.

Nichols-

-Binuclcated Cells in Some Basidiomyceles.

70 Wisconsin Academy of Sciences , Arts, and Letters .

EXPLANATION OF PLATE VI.

Fig. 15. The first indication of a fruit body of Coprinus ephemerus .

X 1500.

Fig. 17. Same as Fig. 15. a, Y-shaped cell. X 1500.

Fig. 19. Longitudinal section of the rhizomorph of Poria. a, Outer layer

of thick-walled cells; c. Pocket containing crystals; b, Layer

of closely packed crystals. X 1200.

Fig. 20. Longitudinal section of the rhizomorph of Pholiota praecox .

X 1200.

Fig. 21. Longitudinal section of the rhizomorph of Lepiota naucina.

X 1200.

Fig. 22. Longitudinal section of the rhizomorph of Dietyophora dupli-

cata. X 1200.

Fig. 23. Longitudinal section of the rhizomorph of Ly coper don pyri-

forme. X 1200.

Trans. Wis. Acad., Vol. XV.

Plate VI.

S. P. Nichols del.

THE RUSSULAS OF MADISON AND VICINITY.

H. R. DENNISTON.

The genus Kussula is well represented in Wisconsin, several

species being extremely abundant. They grow on the ground

and prefer a rich hilly woods where there is considerable moist¬

ure, but certain species are also found in low woods or open

meadows.

An unusually rich harvest of Russulas was gathered in the

summer of 1903, when there was an abundant rainfall through¬

out the season. It was noted that in such a favorable season the

bulk of the Russulas came in the latter part of July and the first

part of August ; after that, although large numbers of fungi were

found, the Russulas were comparatively scarce, and other genera,

especially the Cortinarii and Tricholomas, became relatively

more abundant.

The Russulas are easily distinguishable by their bright colors,

brittle, white or cream colored gills and fleshy stem, but within

the genus, a number of species are particularly difficult to sepa¬

rate.

This difficulty is due largely to their extreme variability. If

our common R. Integra is taken as an example, it is found that

the color of the pileus may be purple, livid, greenish, brown or

tawny. All of these colors may be present in the same indi¬

vidual and are often found in specimens growing in the same

neighborhood. This variability of color is a feature of a num¬

ber of the commoner species of Russula, but not of all of them.

The shape of the pileus usually changes as the plant grows

older. On this account it is quite necessary to note the age of

the specimen at which it has a certain form. In young plants

72 Wisconsin Academy of Sciences , Arts, and Letters.

the pileus is commonly convex, later explanate and finally de¬

pressed.

Massee separates the Russulas into two groups according to

their acrid or mild taste, and makes this feature the principal

basis for classification.

I found that certain specimens of B. alutacea were intensely

peppery the instant they were tasted, others of the same species,

were more slowly peppery. In some cases, the pileus of B. al-

utacea is quite mild, but the gills of the same plant are peppery.

For these reasons it does not appear that the taste is a suitable

cnaracteristic upon which to divide the genus. The word acrid

is used by Massee synonomously with hot or peppery. This use

of terms is misleading for, to many people, the words peppery

and acrid convey different meanings.

While the taste characteristic is hardly sufficiently constant to

be used as the chief basis of classification, it is a point which no

collector should fail to test on the fresh specimens.

The flesh is white in most cases and spongy, but may become

discolored by being bruised or upon drying. When there is a

separable cuticle, a little of it should be removed, for oftentimes

the flesh beneath has a characteristic color.

A number of the Russulas have characteristic odors. Our Wis¬

consin species, B. foe-lens, derives its name from this feature.

The odor is pungent and nauseating and is present in the fungus

at all stages of its development. Curiously enough, I find the

odor of the dried specimen is very pleasant. B. atropurpurea

develops an exceedingly rank odor while drying.

The gills are usually inter veined, a point which should be

noted in the fresh specimens, as later it is difficult to observe.

'The attachment of the gills to the pileus is an important charac¬

teristic and on this point it is advisable to examine as many

specimens and at as many different ages as possible, for fre¬

quently gills which appear in young specimens to be free, upon

the expansion of the pileus become apparently decurrent or ad-

nate.

It is of considerable value to compare the color of the fresh

gills with a color chart, for there is a color change so delicate

from white to straw, from straw to cream and from cream to

Denniston — Russulas of Madison and Vicinity. 73

ocher, that the unaided eye will scarcely distinguish it. In not¬

ing the condition of the gills, it is well where possible, to examine

a portion under the microscope, for it is frequently found that

the spores are tardy in developing, and until they are fully de¬

veloped, the gills remain perfectly white. In some species the

spores cause the gills to have a powdered or dusty appearance.

Probably most of our commoner Russulas are edible, although

this is a point which needs further investigation. They are

looked upon with suspicion by many people and this is no doubt

because of the bright colors and the peppery taste of a number

of species. According to Mcllvane, all of the Russulas are edi¬

ble, at least none of them are poisonous. He admits, however,

That a number of them are extremely unpleasant in odor and

taste.

The species in which the question of edibility is most debated

is probably R. emptied , a small form with a red cap, white gills

and a peppery taste. Mcllvane claims to have eaten it in quan¬

tities with no unpleasant results.

Miller says it is much eaten in Indiana and Illinois. Steven¬

son, on the other hand, says that it acts as an emetic and Peck

gives it as deleterious.

It is possible that the fungus varies in its effect on the human

system but it is also possible that Mcllvane and others have mis¬

taken other closely allied species for R. emetica.

It is probable that our two commonest species R. alutacea and

re. integra are perfectly edible, since there is nothing unpleasant

in their flavor when raw, and no adverse reports against them.

The genus is a favorable one on which to experiment, for the

worst that can be said against any of them is that they act as

an emetic.

Special emphasis should be placed on the following features

to be recorded in field notes from the study of fresh specimens

of Russula.

General : All characteristics shall be given from fungi of

different ages, and under different conditions of growth.

74 Wisconsin Academy of Sciences, Arts, and Letters.

f Presence of cuticle.

| Color of flesh under cuticle.

JPil<: us. ■'j Taste and odor.

| Condition of margin, young and old (striate or even).

[ Color of flesh when bruised.

" Form.

Thickness.

Color, young and old.

Gill 3. Surface, dusted or smooth.

In ter venation.

Equal, heterophyllous or branched. (Relative num¬

ber. )

Stem.

-<

Shape.

Color, young and old.

Substance.

Changes in color.

A classification according to color of the Russulas in the vicin¬

ity of Madison may be of service to the amateur collector.

Red or blood colored: R. alutacea. R. emetica, R. roseipes, R.

veternosa.

Purple or livid: R. atropurpurea, R. integra, R. decolorans,

R. amoena, R. ochrophylla, R. ochrophylla var. albipes.

Brown or ferruginous : R. foetens, R. pectinata, R. ochracea «.

Yellow: R. lactea, R. lutea.

Green: R. viresccns, R. olivascens , R. furcata.

Black: R. adusta, R. nigricans.

White: R. delica.

R. adusta (Pers) Fr.

Pileus: convex, explanate or depressed, sometimes unequal;

margin indexed and even; color, fuliginous, smooth, viscid

when wet ; 6-12 cm. ; flesh, white.

Gills: adnate or decurrent, not broad, narrow* towrard margin;

white, changing to lead color, subcrowrded, heterophyllous,

forked occasionally.

Stem: stout, cylindrical, fleshy, 3-4 cm. long, 1.5-2 cm. thick,

smooth, white, soon becoming blackish. July and August.

Spores: globose, slightly echinulate 6-9 u.

Denniston — Russulas of Madison and Vicinity.

75

Characters f Pileus : fuliginous, solid, margin even. 6-9 cm.

on which j Gills: adnate, white to lead color,

identification ] Stem : stout, short, white becoming blackish,

is based. ^ Taste: mild.

Pileus : wrinkled, umbrinus to fuliginous,

cracking at margin.

Gills : lead color to blackish.

Stem : longitudinally wrinkled, umbrinus or

blackish.

Habitat. On ground under trees in grass.

Characters |

of dried -{

specimens. |

Locality. Blue Mounds, Madison.

Edibility. When well cooked it has a good flavor. (Mcl.)

R. alutacea Fr.

Pileus: convex, explanate or depressed, smooth, pink beneath

separable pellicle, rosy, bright red, or purple red, olivaceous,

often yellow at center or yellowish spots; margin striate,

thin ; 4-12 cm. ; flesh, white, slightly peppery, gills more so

than pileus.

Gills : free, equal, broad (% -Ty2 cm.), occasionally forked

(every 3-10th forked) and heterophyllous, white then ochra-

ceous, rounded anteriorly, not powdered by spores. Sub

crowded, interveined.

Stem : cylindrical or tapering downward, smooth, white,

tinged rosy, brown where rubbed, fleshy, 6-10 cm. long, 1-2

cm., thick, spongy. August, September, October.

Spores: globose or ellipsoidal, ochraceous, echinulate. 7-7 u.,

7-9 u., 9-12 u.

f Pileus : rosy red, often yellow at center.

Characters j Stem: stramineus, retaining shape, sometimes

on which J reddish.

identification ] Gills : light ochraceous, retaining color when

is based. | dried.

^ Taste: more or less peppery.

Characters |

of dried -J

specimens, j

i

Pileus: rosy or wine red, sometimes yellow at

center. Thin, margin faintly striate.

Gills: bright, ochraceous.

Stem : usually not shrunken, straw colored,

rarely brownish.

76 Wisconsin Academy of Sciences, Arts, and Letters.

Habitat. Largest specimen from ground in hilly oak woods.

Locality. Star Lake, Blue Mounds, Madison (Eagle Heights).

Edibility. When fresh it is very good. (McL)

B. amoena. Quel.

Pileus : convex then depressed ; 4-6 cm. ; smooth or pulverulent ;

atroviolaceus; margin not striate; taste mild, odor fruity;

flesh white then cream.

Gills : white then cream ; reddish at edge when dry ; subcrowded,

interveined.

Stem: rigid, tapering upward, 4-6 cm. long, 1-1% cm. thick;

powered, purple red. August.

Spores: subglobose, scarcely echinulate. 6x7 u.

Characters f Pileus : atroviolaceus, powdered, margin even,

on which ! Gills: subcrowded, white then yellowish; edge

identification , tinged with red.

is based. Stipe : larger toward top ; purple red.

f

Characters |

of dried -{

specimens. |

L

Pileus: atroviolaceus; depressed at center,

wrinkled.

Gills : adnate, ochery with reddish margin.

Stem : larger above, ochery with rosy tint.

Odor, pleasant.

Habitat. On ground in moist woods.

Locality. Blue Mounds.

Edibility. ?

B. alropurpurea Pk,

Pileus : convex, at first globose, finally slightly depressed,

smooth; dark purple, livid, or deep purple red, disk often

blackish; margin, even at first slightly striate when old.

6-12 cm. ; flesh, white, blackish where broken ; odor foetid

when drying. Taste, mild or slightly peppery.

Gills: free, subdistant, rounded anteriorly, interveined, rarely

forked, white to cream color, sometimes with rosy tint.

Stem: smooth, white, tapering upward slightly, 8-10 cm. long,

2 cm. thick, brown where rubbed, fleshy, white, sometimes

rosy. July, Aug., Sept., Oct.

Denniston — Russulas of Madison and Vicinity .

77

Spores: globose or ellipsoidal; stramineus or cream, echinulate,

8x12//, 7x9/x, 7x7 //.

Pileus: dark purple color. Flesh, grayish un¬

der pellicle. Odor of drying plant foetid.

Gills and stem brown where bruised.

Stem: sometimes rosy.

Spores with slight rosy tint. 8x12//,.

Characters

on which

identification |

is based.

Characters

of dried

specimens.

f Pileus : dark blackish purple, usually darker in

center. Margin even or striate.

^ Gills: isabellinus, or ochraleucus.

Stem: stramineus, shrinking but little, covered

with brown spots.

Habitat. Rich oak woods on ground.

Locality. Madison, Burlington, Blue Mounds.

Edibility. Should be perfectly fresh to have good flavor.

(Mcl.)

R. decolorans Fr.

Pileus: subglobate then convex, smooth, slightly viscid when

moist, livid, brownish red, lighter in center. Symmetrical,

faintly striate and tuberculate at margin. 6-8 cm.; flesh

white. Taste, slowly peppery.

Gills: almost free, subcrowded, equal, rounded in front, inter-

veined, white then yellowish.

Stem: 6 cm. long, 1.5 cm. thick, cylindrical, solid, minutely

striate, white then darker. July.

Spores: Globose, ellipsoidal, echinulate. 8x10//, 9x11//.

Characters

on which

identification

is made.

| Pileus: firm, 3^ellowish red, pale at center.

Gills: subfree, white then yellowfish.

! Stem: cylindrical, white with darker spots.

f Pileus : badius at margin, center ferrugineus, re-

of“ J taming its shape well.

0 rie I Gills: little changed in form, ochraceous.

Stem : stramineus, cylindrical.

specimens.

PI abitat. On ground in oak woods.

Locality. Madison.

Edibility. Esculent and of good quality. (Morgan).

78 Wisconsin Academy of Sciences , Arts, and Letters.

R. delica.

Pileus: infundibuliform, smooth then pruinose, rough, scaly

and cracked, white or tan. Margin incurved, entire. 7-12

cm.; flesh, white changing to yellowish, thin, dry, slightly

and slowly peppery.

Gills: decurrent, distant, thin, narrow, heterophyllous, inter-

veined, occasionally forked, white or cream.

Stem : stout, cylindrical, smooth and solid, or pruinose. White,

2.5-6 cm. long, 2- 2.5 cm. thick. July, August, in woods.

Spores: echinulate, subglobose, 8x9 p, 10x12^, 7x8/x.

Pileus : white, infundibuliform, roughened.

Margin : entire, incurved.

Stem: stout, short.

Gills: decurrent, narrow.

i© iixcu.ic. j rpas^e . slightly peppery.

f Pileus : stramineus or brownish, wrinkled, indi-

arac ers | funcpi)Iliii‘orm Margin, incurved.

o cne 1 Gills: yellowish or isabellinus.

specimens. ^ g^-em . shrunken, longitudinally wrinkled.

Habitat. On ground in woods.

Locality. Lake Waubesa, Blue Mounds, Madison.

Edibility. Edible and of fair quality. (Mcl.)

R. emetica Pr.

Pileus: convex or explanate. Pink or rosy, tawny when old,

sometimes white, smooth, margin striate. 4-10 cm. Flesh,

reddish under separable pellicle. Taste very peppery.

Gills : almost free, not crowded, heterophyllous, interveined, tri¬

angular, white.

Stem: 4-8 cm. long, 1-2 cm. thick, smooth, white or rosy,

spongy. Aug., Sept., Oct.

Spores: 8x9 jx, ellipsoidal, echinulate white.

Characters f Pileus : tawny or white color, when old.

on which I Stem : reddish or white.

identification ] Gills : always white.

is made. [ Taste : peppery.

Characters i

on which [

identification i

Denniston — Russulas of Madison and Vicinity. 79

f Pileus: red or white or tawny. Margin striate

Characters ! and cracked,

of dried j Q.jps: cream color, wrinkled,

specimens. ^ g^em . stramineus or rosy, longitudinally striate.

Habitat. On ground in hilly woods.

Locality. Madison, Star Lake, Florence.

Edibility. Claimed by some to be poisonous; by others to be

perfectly harmless.

R. fociens Fr.

Pileus: globose at first then fiat to concave or depressed at

center, sometimes gibbous, viscid when moist. Color, young,

wrhitish to pale brown ; older, badius or vinosus. Membran¬

ous pellicle, elastic and tough, marked with radiating brown

lines. Margin thin, striate or pectinate. Size 5-10 cm.

Taste mild, nauseous or acrid ; odor foetid. Becoming brown

when broken.

Gills: adnexed or sinuate, broad, not crowded, interveined,

rounded anteriorly, forked near stem, few heterophyllous;

white, spotted brownish when old.

Stem: 4-12 cm. firm, tapering upward, smooth, white to gray¬

ish, reddish spots and lines. 1.5-2 cm. thick. June, July,

Aug.

Spores: white, ellipsoidal, echinulate. 8x10/a.

f Pileus : brown, depressed with shining viscid

Characters | pellicle,

on which _j Margin : thin striate,

identification \ Stem : often brown spotted,

is made. j Gills: often brown spotted.

[ Odor : foetid, rank.

Characters

of dried

specimens.

' Pileus : reddish brown, much folded, wrinkled and

striate.

■{ Gills : yellowish or salmon.

| Stem : depressed in spots, yellowish, brown spotted.

L Odor, mealy.

Habitat. Moist ground in oak woods.

Locality. Madison, Blue Mounds, Minneapolis, Minn.

Edibility. Not poisonous, but unpleasant in flavor. (Mcl.)

80 Wisconsin Academy of Sciences, Arts, and Letters .

R. furcata Fr.

Pileus: explanate to depressed. Olivaceus to sordid, smooth

or slightly roughened. Margin, incurved, even. 6-8 cm.

Flesh white, pinkish under pellicle. Taste mild.

Gills: many, adnate to dec nr rent, strongly interveined and

forked ; white.

Stem : 6-8 cm. 1-1 f ^ cm. thick, cylindrical, slightly longi¬

tudinally wrinkled, white.

Spores: ellipsoidal, echinulate, 8x10//.

f Pileus: dark-greenish color.

Characters on which J Stem: cylindical, white.

identification is made. j Gills : adnate, forked, white.

Taste: mild.

Characters of dried

specimens.

f Pileus: margin even, shining.

Gills: thick, fumosus.

1^ Stem: white, wrinkled longitudinally.

Habitat. Ground in woods.

Locality. Parfrey’s Glen.

Edibility. Edible (Mcl).

R. integra Fr.

Pileus : convex, explanate or slightly depressed ; smooth,

slightly viscid when moist. Purple, livid, brownish, tawny,

greenish, darker or lighter at disk. Margin thin, even at

first, then pectinate or cracked when old. 6-8 cm. broad;

flesh white, dark where broken. Taste, mild.

Gills : free or adnexed, not crowded, 1% cm. broad, interveined,

few forked, heterophyllous, (few short) rounded anteriorly

and posteriorly; straw to yellowish. Pulverulent, when

mature.

Stem: 6-13 cm.; stout, cylindrical or large above and below,

2^2-3 cm. thick; smooth or slightly ridged; white, brown

where rubbed, spongy. July, August, September.

Spores : ellipsoidal, globose, echinulate, ochraceous. 7x9//, 7x7//,

8x10 [i.

Denniston — Russulas of Madison and Vicinity.

81

Characters

on which

identification

is made.

f Pileus.

| Stem:

^ rubbed.

Gills:

l Taste:

varicolored, dark purple prevailing,

never reddish. Becoming brown where

powdered when mature,

never peppery.

Characters

of dried

specimens.

f Pileus : usually showing different colors at margin

| and disk, dark purple, livid, greenish, yellowish

j or wine; cracking at margin.

] Gills: isabellinus, wrinkled and folded.

| Stem: plump or shrunken; straw colored with

[ brownish spots.

Habitat. Largest specimens grow on ground in nilly oak woods.

Locality. Madison (Eagle Heights), Devils Lake, Burlington,

Blue Mounds.

Edibility. Of good flavor when fresh. (Mcl.)

R. lactea Fr.

Pileus: convex, explanate, gibbous or depressed, rigid, surface

pruinose then cracked, no separable pellicle; color, stramin-

eus, yellow or pinkish yellow. Margin, even, rounded.

10-19 cm. ; flesh, white, bitter ; cheesy odor.

Gills: free, equal or slightly forked, broad, solid, subdistant,

interveined, rounded anteriorly, white to straw.

Stem: cylindrical or larger at top; solid, white, 2.5-6 cm. long;

1-2 cm. thick. August.

Spores: globose, minutely echinulate, 9/x.

Characters f Pileus: rigid, yellowish, incurved,

on which j Stem: solid,

identification j Gills : thick, straw color,

is based. [ Taste: somewhat bitter.

Characters

of dried

specimens.

T Pileus, yellowish or brownish, retaining shape,

j Margin, incurved.

| Gills : thick, narrow near stem, rounded anteriorly.

L Stem : longitudinally wrinkled, smaller at base.

Habitat. On ground, not common, in oak woods.

Locality. Blue Mounds, Madison.

Edibility. Edible and of good flavor. (Mcl.)

82 'Wisconsin Academy of Sciences , Arts , and Letters.

R. lutea Fr.

Pilens : conical, convex or depressed, smooth, viscid when moist,

pale or bright yellow, surface sometimes puberulent or

floccose. Margin even or later striate. 3-8 cm. ; flesh white ;

taste mild.

Gills: adnate to free, interveined, crowded; rounded anteriorly;

equal, white then yellowish.

Stem: tapers up, smooth, white, fleshy; 3-5 cm. long; 5-8 cm.

thick. July, August, September.

Spores: cream, ellipsoidal, echinulate. 8x10/*.

Characters f Pileus :

on which j Stem:

identification j Gills:

is based. [ Taste:

pale or bright yellow, small size,

short.

crowded, yellowish,

mild.

Characters ( Pileus: retaining yellow color,

of dried < Gills: yellowish or salmon,

specimens. ( Stem : retaining its shape, stramineus.

Habitat. On ground in woods.

Locality. Lake Waubesa, Blue Mounds, Madison.

Edibility. Edible and of delicate flavor. (Mcl.)

(a) R. ochrophylla Pk. var. albipes Pk.

Pileus: convex then flattened and depressed, slightly viscid

when moist, smooth at first, then broken up into patches;

margin at first incurved not striate, blackish or brownish

olive. 7-12 cm. ; flesh white, reddish when bruised.

Gills: adnexed, rounded near stem, heterophyllous, thick, sub-

distant, white, reddish where broken.

Stem: 4-7 cm. long; 21/2-3 cm. thick; cylindrical, dirty white,

then blackish, solid. July.

Spores: 10x11/*, 11x13/*, ellipsoidal, coarsely echinulate.

Characters

on which

identification

is made.

f Pileus : fleshy, blackish, margin even.

! Gills : thick, subdistant, white, red

| bruised.

I Stem : thick, cylindrical, blackish.

where

Denniston — Russulas of Madison and Vicinity.

83

f Pileus : umber or blackish, margin wavy. In-

Characters ! curved.

of dried ] Gills . darfc? reddish where bruised,

specimens, ^gtem: umber or blackish.

Habitat. On ground in woods.

Locality. Madison.

Edibility. Edible, but not equal to most Russulas. (Mcl.)

R. ochracea Fr.

Pileus : flat to convex, fulvus : darker at center. Margin, irreg¬

ular, thin, cracked, striate incurved; surface, rough, scaly.

Size, 4-9 cm. Taste, mild; odor, none.

Gills : free, subcrowded, broad, straight, equal, interveined,

white.

Stem: 2-2.5 cm. long, 1-2 cm. thick, tapering upwards or ven-

tricose, surface, smooth, stramineus, hollow. July.

Spores: 6x6^, 6x7. 5jjl, echinulate, elliptical or globose.

Characters f Pileus: fulvus, dark at center, striate, thin.

on which j Gills : free, subcrowded, white.

identification • Stem: short, thick.

is made. [ No odor or . taste.

Characters | GiUg;

ofdrled Stem:

specimens. odor;

fulvus, roughened and dirty, striate,

wrinkled and wavy, isabellinus.

short and hollow,

mealy.

Habitat. Hilly woods on ground.

Locality. Madison (Edgewood).

Edibility. Probably edible.

R. ochrophylla Pk.

Pileus : convex, explanate or depressed ; margin, even or

slightly striate, 6-10 cm. ; atropurpureus, reddish under the

separable pellicle. Flesh, white. Taste, mild.

Gills: equal, broad, adnat'e, subdistant interveined, yellowish

then bright ochraceous ; dusted by spores.

84 Wisconsin Academy of Sciences, Arts, and Letters .

■.’vyrnr-gft

Stem: Subcylindrical, smooth, whitish with rosy tint; firm or

spongy. 5-7 cm. long, 1-2 cm. thick. July, August.

Spores: globose, ellipsoidal, echinulate, 8x10 jn, 9xl2/u.

Characters

on which

f Pileus : dark purple color ; reddish beneath pel-

licle.

identification ; Gills :

is based. [ Stem :

broad, equal, dusted bright ochraceous.

rosy.

f Pileus : dark purple or livid, smooth ; margin

Characters I slightly striate.

0 f16 I Gills: bright ochraceous.

specimens. ^ gtem : longitudinally wrinkled, rosy.

Habitat. Open woods on ground.

Locality. Madison, Blue Mounds.

Edibility. Rather tough, but not disagreeable to the taste.

Milk in which it is stewed takes on pink color.

(a) B. ochrophylla Pkt var. albipes Pk.

Pileus : deep red ; stem white ; otherwise like the type. A num¬

ber of specimens answering to this description were found at

Blue Mounds in August, 1903.

B. olivascens Fr.

Pileus : explanate or concave, smooth slightly viscid when

moist, olivaceous; margin, even; 5-10 cm.; flesh, white.

Taste, mild.

Gills : adnexed or adnate, subcrowded, broad at margin, narrow

toward stem; interveined, few forked, white.

Stem: 4-7 cm. long, cylindrical or tapering upward slightly,

2 cm. thick, spongy, white, smooth. July, August.

Spores: scarcely echinulate, 8x 10/x, 8x9/*, ellipsoidal.

Characters f Pileus: smooth, shining, olivaceous,

on which j Margin : even, taste mild,

identification j Gills: broad at margin, white,

is based. [ Stem: white, smooth.

Denniston — Russulas of Madison and Vicinity .

85

Characters \ ^eus: smooth and shining, darker at center, dark

of dried J olive §Teen or hronze green. Margin entire.

• j Gills: stramineus.

^ * i Stem: longitudinally wrinkled, whitish.

Habitat. Rich oak woods on ground.

Locality. Blue Mounds.

Edibility. Taste mild.

R. pectinata Fr.

Pileus: convex, then flat or concave; smooth, ochraleucus to

f errugineus, darker at center ; margin, thin, striate tubercu-

late, at first infiexed. 6-8 cm.; taste, slightly pungent,

smell, soapy. No changes.

Gills: free, broad and rounded anteriorly, subcrowded, equal,

white.

Stem: 3-6 cm. long, l%-2% cm. thick, rigid, equal or ventri-

cose ; smooth or pulverulent, white ; spongy when old. July

and August.

Spores : ellipsoidal or globose, 7x7/*,, 7x9/*,.

Characters f Pileus: f errugineus, thin, pectinate.

on which j Stem : slender, white.

identification • Gills: rounded anteriorly.

is made. [ Odor not unpleasant.

Characters \ ^eus : latericius or yellowish, darker in center ;

n . • . j very thin, striate almost to center,

o ! : yellowish white.

[Stem: stramineus, slightly roughened.

Habitat. On ground in grass under oaks along road.

Locality. Madison, Blue Mounds.

Edibility. Unpleasant in flavor. (Mcl.)

86 Wisconsin Academy of Sciences , Arts , and Letters.

B. roseipes (Seer.) Bres.

Pilens : convex, plane or depressed, somewhat farinaceous ; rosy

red, ocher and whitish. Margin, thin, at first incurved to

stem; striate at edge. Cracking at edge and divided into

areas when old. 3-6 cm. broad. Flesh, white, pinkish un¬

der pellicle. Taste mild.

Gills: free or slightly adnexed, subcrowded, rounded poster¬

iorly, stramineus to light yellow.

Stem: 4-6 cm. long, l%-2 cm. thick, terete, tapering upward,

smooth; white, sometimes rosy, spongy.

Spores: globose or ellipsoidal 6x6/*, 7x9/*, light ocher yellow,

eehinulate.

Characters

on which

f Pileus:

Stem:

identification ■, Gills:

is based.

Taste:

rosy and ocher,

reddish,

ochery.

mild.

r Pileus : brittle and cracked at margin, retaining

Characters red color as when fresh,

of dried ■{ Gills: ocher, close and wrinkled,

specimens. I Stem : retaining its shape well ; rosy ; color persist-

l cnt. ; j i:f!|

Habitat. Open mixed woodfe on ground.

Locality. Madison (Picnic Point).

Edibility. Agreeable in flavor. (Peck. )

B. virescens Fr.

Pileus: convex, flat or depressed; thick at disk, dry, breaking

up into darker areas ; green or grayish. Margin, striate or

even. 5-15 cm.; flesh, white. Taste, mild.

Gills : free, appearing adnate by expansion of pileus, broad and

thick, subcrowded, interveined, few forked near stem, white

to cream.

Stem: 4-7 cm. long; stout, fleshy, cylindrical or small at base.

11/2-3 cm. thick, smooth, white. July and August.

Spores: sparingly eehinulate, subglobose, 6x6/*, 6x8/*.

Dennist on— Russulas of Madison and Vicinity.

87

Characters

on which

identification

is made.

f Pileus : dry, without pellicle, broken up green-

j ish.

■{ Stem: stout, smooth, fleshy.

| Gills: broad, subcrowded.

[ Taste, mild.

Characters

f Pileus: yellowish tawny with greenish areas.

of dried *|

Gills : regular, changing form but little in drying,

str a mineus.

specimens. I c^em; whitish, wrinkled

Habitat. Rich oak woods on ground.

Locality. Madison, Pdue Mounds.

Edibility. Good flavor. Can be eaten raw. (Mcl.)

R . veternosa Fr.

Pileus : convex then explanate ; smooth rosy, with paler yellow¬

ish disk. Margin even, 7-8 cm.; flesh, white.

Gills: adnate, subcrowded, broad, heterophyllous, few forked,

cream colored.

Stem: tapering upward, 4 cm. long; white; 2 cm. at base; soft,

fleshy, then hollow; smooth. June.

Spores : globose, ellipsoidal, echinulate 8x10^, white.

Characters f Pileus: rosy or pink, center yellowish,

on which ! Gills: subcrowded; few, forked,

identification 1 Stem: stout, white, becoming hollow,

is made. L Taste : stem and pileus, very peppery.

Characters ( Pileus : rosy, paler in center, changing but little

of dried < Gills : ochery.

specimens. (Stem: stout, fleshy, becoming hollow.

Habitat. Ground in woods.

Locality. Cottage City, Minn.

Edibility. ?

88 Wisconsin Academy of Sciences , Arts, and Letters.

Pileus: flat to depressed; umber at margin, darker at disk.

Margin coarsely striate, thin ; smooth or with few yellowish

fibrils. 5- 12 cm. Taste, mild; odor none.

Gills : free or slightly adnexed, not crowded, equal, thin, inter-

veined, white.

Stem: 4-5 cm. long, 1-2 cm. thick, mealy, cylindrical, white to

ocher; fleshy then hollow. July.

Spores: globose, spinulose, 9x9/*.

Characters

specimens.

f Gills : wrinkled, thin, drab and leather

of dried! colored.

I Stem : rough, ocher to brown.

[ Pileus : umber, dark at disk, striate, thin.

Habitat. Ground on lawns.

Locality. Madison, Burlington.

Edibility. ?

This Russula is close in its affinities to B. consobrina var. soro~

ria Fr. but has a mild taste. It is close also to B. ochracea Fr.

but has not the ocher gills.

THE RELATIONS OF THE ANDRENINE BEES TO THE

ENTOMOPHILOUS FLORA OF MILWAUKEE

COUNTY

S. GRAENICHER.

Owing to their large number of species, as also to the close

relations of many of the species to certain flowers, the bees of

the family Andreninae occupy a prominent position among the

flower-visiting insects of our neighborhood. This paper deals

with 47 species, representing about one-fifth of the entire bee-

fauna of our region. Several species make their appearance

quite early in the spring, and from this time on the family is

represented throughout the floral season, although not a single

one of the species extends its time of flight over two months and

a half. Some are of rather common occurrence, while others

are extremely rare, and are only occasionally met with. For

several years past I have given much attention to the bees of

this family, and have gradually come across 17 new species of

Andrena from Milwaukee county, which have been named and

described as follows :

Andrena subcommoda ^

Andrena sigmundi ■

Andrena multiplicata T. D. A. Cockerell, Canadian Entomologist,

Andrena radiatula j 1902, p. 45.

Andrena rufosignata j

Andrena clypeonitens J

Andrena graenicheri

Andrena parnassiae

Andrena peckhami

T. D. A. Cockerell, Annals and Magazine of

Natural History, 1902, p. 101.

90 Wisconsin Academy of Sciences, Arts, and Letters.

Andrena thaspii

Andrena cocker elli

Andrena milwaukeensis

Andrena viburnella

Andrena aWofoveata

S. Graenicher, Canadian Entomologist, 1903,

p. 162.

Andrena fragariana )

Andrena wheeleri L S. Graenicher, Entomological News, 1904, p. 64.

Andrena persimilis \

Poly tropic and oligotropic bees. Some bees are in the habit

of visiting a great variety of flowers, while others may be ob¬

served on the flowers of a single or those of a few species only.

Loew1 called the former polytropic and the latter oligotropic.

Robertson ° states that “in the economy of the host-bees (those

not inquiline) the most important flowers are those from which

the female gets the pollen upon which her brood is fed.” Ac¬

cordingly he proposes to use the term oligotropic for a species

of bee, of which the female obtains her pollen-supply from a

single species, or several closely related species of plants, i. e.,

plants belonging to the same genus, or the same natural family,

and on the other hand to call those bees polytropic that collect

pollen from plants of different families. In considering the

relations of our Andrena species to our flowers, I make use of

the terms oligotropic and polytropic in the same sense as Robert¬

son. In the following table the time of flight, or in the case of

insufficient observation the date of capture of each species is

given, as also the names of the plants visited for pollen. When

the bee has been recognized as oligotropic this is stated, other¬

wise the species is considered polytropic. Following Robertson’s

example, I consider in the case of an oligotropic bee the latter

adapted to the genus if found collecting pollen from more

than one species of that genus, and to the family if it obtains

pollen from plants belonging to different genera of that family.

Graenicher — Andrenine Bees and Entomophilous Flora. 91

Flowers Visited for Pollen.

Oligotropic, Salix.

Oligotropic , Salix.

Species of Salix and Thaspium.

Species of Salix, Primus, Clayto -

nia, Cornus , Crataegus, Vi¬

burnum, Angelica, Symphori-

carpos, etc.

Oligotropic, Salix.

Species of Salix, Claytonia and

Crataegus .

Species of Salix, Sanguinaria

Erythronium , Caltha , Ribes

Rosa, etc.

Species of Salix, Pru»us, Vibur

num, Angelica . Sniraea, Ce

lastrus, etc.

Species of Acer , Salix, Claytonia

Cornus , Vagnera, Hydrophyl

lum, Gera nium, etc.

This species is known in the male

sex only.

Oligotropic, Claytonia Virginica.

Species of Caltha and Lonicera.

Species of Salix.

Species of Claytonia, Ribes, Strep-

topus, Symp? wricarpos,1 Angeli¬

ca, Spiraea, Celastrus, Diervilla,

etc.

Oligotropic, Salix.

Species of Salix, Taraxacum and

Rubus.

Species of Taraxacum, Vagnera

Thaspium, Angelico, Viburnum

Symphoricarpos, etc.

Species of Ribes, Vagnera, Thas

pium, Heracleum, Cornus, Ru

bus, Geranium, etc.

Species of Salix, Taraxacum, Vi

burnum. Geranium, Crataegus

Cornv s, Thaspium, Sanicula, etc

Claytonia Virginica.

Oligotropic, Fragaria Virginiana.

Species of Salix, Thaspium, An¬

gelica, Cornus , Viburnum, Rhus ,

etc.

Oligotropic, Umbelliferae.

Oligotropic, Hydrophyllum .

Uvularia grandiflora .

Oligotropic, Geranium macula-

tum .

Species of Vagnera and Crataegus.

Species of Thaspium, Viburnum ,

Crataegus , Symphoricarpos,

Rosa, Spiraea, Rhus, etc.

Species of Viburnum, Sanicula ,

Angelica, Cornus, Rubus, Rhus,

Rosa, etc.

Species of Thaspium, Cornus,

Spiraea, Symphoricarpos,

Rhus, Ceanothus, Veronica, etc.

Oligoitropic, Umbelliferae.

92 Wisconsin Academy of Sciences , Arts, and Letters ,

The data relating to the time of flight, as contained in this

table, enable us to construct a flight-curve for the family Andre-

ninae as follows:

April May June July Aug. Sept. Oct

Graenicher — Andrenine Bees and Entomophilous Flora. 93

With us the blossoming of the first catkins of our earliest

species of willow, Salix discolor falls together with the appear¬

ance of our earliest species of Andrena, A. cockerelli, an oligo-

tropic visitor of the willows. In a certain locality in the Menom¬

onee valley numerous specimens of Salix discolor occur, and

among these a large specimen, bearing pistillate catkins opens

its blossoms regularly in advance of all the other specimens.

This particular plant was kept under observation in the early

days of spring for 2 successive seasons, with the result, that

on the first warm and bright day bringing out its blossoms the

presence of the bee Andrena cockerelli in both sexes was noticed.

This was the case on April 6th, .1902, and again on March 31st

in the exceptionally early spring of 1903. I have never suc¬

ceeded in coming across a species of Andrena before our willow-

blossoms appear, although two species of entomophilous plants,

Erigenia bulbosa and Hepatica acuta open up their flowers ear¬

lier than this willow. As the willow-blossoms become more

abundant additional species of Andrena arrive on the scene, so

that at the end of the third week in April 8 species are on the

wing, 4 of which are oligotropic, depending for pollen on the wil¬

lows exclusively, although they also visit other flowers for nectar.

These facts point to the importance of the willows in the econ¬

omy of our first arrivals among the Andrenine bees. As the

season advances several other flowers attractive to insects open

up, among them being Claytonia Virginica which usually ap¬

pears towards the end of April, and also has an oligotropic

visitor Andrena erigeniae. At the beginning of May 13 species

of Andrena are present, and this number is gradually increased

during the month until a maximum with 24 species is reached

in the latter part of May, and lasting throughout the first week

in June. This is the maximum of our spring-group of Andre-

ninae, and it corresponds with the blooming period of a great

variety of flowers, representing different families. Seven of

these bees are oligotropic, two of which collect pollen from the

late species of willows, one from Fragaria Virginiana, one from

Hydrophyllum Virginicum, one from Geranium maculatum, and

the two remaining ones from umbelliferous plants. Two species

of Umbelliferae Thaspium Irifolialum aureum and Taenidia in-

94 Wisconsin Academy of Sciences, Arts, and Letters.

tegerrima produce flowers in great abundance, and are very at¬

tractive to many species of Andre na besides the two oligotropic

species of the family flying during the maximum. In addition

to the plants mentioned in connection with the oligotropic vis¬

itors various species of Viburnum, Crataegus, Cornus, Kibes,

Rubus, etc., supply many of the species forming the spring-max¬

imum with pollen and nectar. From this maximum on there is

a gradual decline of the curve until a minimum is reached, ex¬

tending from about July 17 to July 20, and represented by a

single species. At the end of the third week in July a renewed

increase sets in, culminating in a maximum of the summer-

group of Andreninae, with 8 species in evidence at the end of

August and the beginning of September. This summer-group

comprises altogether 11 species, and these are with but one

exception oligotropic bees of the family Compositae. The plants

of this family with the numerous species of Solidago, Aster,

Helianthus, Kudbeckia, Eupatorium and many other genera are

dpminant factors in the make-up of the flora of the late sum¬

mer months. The earliest species of goldenrod begins its bloom¬

ing period about the middle of July, around the 20th of the

month the first aster appears, and mostly a trifle later the first

sunflower. Corresponding with the appearance of these com-

posite flowers the earliest Andrena of the summer-group A. peck-

hami begins to fly about July 21, and before the end of the month

3 additional oligotropic visitors are present on these flowers.

From the first week in September on there is a falling off in the

number of these bees, and around October 8th Andrena asteris ,

the last one of the A.ndrcninae disappears. Although any one

of these visitors of the Compositae may collect pollen from flow¬

ers belonging to diff erent genera of the family, . they still show

a decided preference for certain genera. A. peckhami, A. cly-

peonitens, A. aliciae and A. kelianthi favor the sunflowers', A.

nubecula, A. americana, A. solid aginis and A. persimilis the

goldenrods, A. asteris and A. graenicheri the asters.

The exceptional position held by Andrena parnassiae, the

oligotropic visitor of Parnassia Caroliniana has been referred to

above. This is so closely related to A. peckhami, and the two re¬

semble each other to such an extent as to leave no doubt regard-

Graenielter -■ And? cnine Bees and Entomophilous Flora. 95

ing their origin from a common ancestor. But while A. pech+

hami, a bee adapted to the Composilae appears together with the

early flowers of Helianlhus, A. parnassiae is adapted to a plant

belonging to quite a different family, and it flies considerably

later, from August 25th to September 2Gth, daring the blooming

period of Parnassia. The first specimens of this bee were taken

in a certain locality south of Whitefish Bay on the bluffs border¬

ing Lake Michigan, where the plant Parnassia Caroliniana oc¬

curs in large patches with ail abundance of flowers. The latter

are especially attractive to flies, 17 of the 25 recorded visitors

belonging to this order, but the bee Andrena parnassiae may be

observed regularly, season for season, although not a frequent

insect. It has up to the present time not been met with at any

other point within our County, nor has it been reported from

elsewhere.

Oligotropic species of Andrena. In the foregoing several of

our oligotropic species have been mentioned in connection with

the flowers visited. For the sake of completeness a list of all

of our oligotropic Andreninac is offered below.

Robertson2 has published a list of the bees of Carlinville,

Southern Illinois regarded by him as oligotropic, and 13 of the

20 species of oligotropic Andrenina-e of that locality occur also

in our region. They are as follows:

Name of species.

Andrena illinoensis, Rob.

Andrena mariae, Rob.

Andrena erythrogastra, Ashm.

Parandrena andrenoides , Cress.

Andrena erigeniae, Rob.

Andrena ziziae, Rob.

Andrena geranii, Rob.

Andrena geranii maculati, Rob.

Andrena aliciae, Rob.

Andrena nubecula, Sm.

Andrena helianthi, Rob.

Andrena solidaginis, Rob.

Andrena asteris, Rob.

Plants visited for pollen.

Salix.

Claytonia Virginica.

U mb elli ferae.

Hydrophyllum.

Geranium maculatum.

Compositae.

96 Wisconsin Academy of Sciences , Arts , and Letters.

To these I add 11 species recognized

surroundings :

Name of species.

Andrena cocker elli, Graen.

Andrena fragariana, Graen.

Andrena thaspii, Graen.

Andrena wheeleri , Graen.

Andrena albofoveata, Graen.

Andrena peckhami , Ckll.

Andrena clypeonitens, Ckll.

Andrena americana, D. T.

Andrena persimilis, Graen.

Andrena graenicheri, Ckll.

Andrena parnassiae, Ckll.

as oligotropic bees of our

Plants visited for pollen.

Salix.

Fragaria Yirginiana.

Umbelliferae.

Gompositae.

Compositae.

Gompositae.

Parnassia Garolimana.

According to this list 24 of the 47 species of Andreninae con¬

sidered in this paper, or fully one-half are oligotropic. The

Compositae supply 10 of these with pollen, and the willows come

next with 5 oligotropic bees. The importance of the TJmbelli-

ferae in this respect is also evident, 4 such visitors being adapted

to them. As regards Andrena gcranii this bee figures in Bob-

ertson’s list as an oligotropic species of Hydrophyllum appendi-

cidalum, but in our surroundings it collects pollen from Hydro -

phyllum Virginicum , the only representative of that genus in

our flora, and it therefore has to be considered an oligotropic

bee of the genus Hydrophyllum.

Seasonal forms. There are numerous instances recorded of

an insect-species appearing at one period of the season in a form

differing more or less from the form assumed at another period.

These are called seasonal forms. A few cases are mentioned in

the literature, all of them from the Eastern states, in which a

species of Andrena taken later in the season has been regarded

as identical with some species flying in the spring. Observa¬

tions carried on throughout a number of years warrant the

statement, that in our region no seasonal forms of Andrena oc¬

cur. Species after species makes its appearance in the order

indicated in the list at the beginning of this paper. As regards

their time of flight there is a great diversity among the different

species, some of them flying over 2 months, while others are

Graenicher — Andrenine Bees and Entomophilous Flora. 97

present during a few weeks only. In each species one genera¬

tion is produced annually, and the bee appears the following

season at the time of flight of the respective species.

It has been pointed out that all of our species of the summer-

group are oligotropic, and all but one adapted to the C oviposit ae,

and in this respect they differ essentially from the species of

the spring-group. In connection herewith it may be emphasized

that while the early Umbellif erae, represented especially by the

genera Thaspium, Taenidia, Heracleum, Angelica and Sanicula

are very attractive to members of the spring-group of Andren-

inae, the late Umbellif erae with the genera Cicuta, Slum, Oxy-

polis and Conioselinum have no relations whatever to the An-

dreninae of the summer-group. Comas stolon if era has its flow¬

ering season in the spring, but some specimens produce flowers

throughout the summer and as late as the middle of September.

It is significant that the flowers of this species are visited very

abundantly by many Andreninae of the spring-group, but that

after the middle of July a single species has been noticed on the

flowers, and this is A . multiplicata, the latest species of the

spring-group, which holds out until the end of July. All of

these considerations point to the fact that our Andreninae of

the summer-group, so far as their relations to flowers are con¬

cerned have nothing in common with those of the spring-group,

and they furthermore support the statement, that in our region

at least, no seasonal forms of Andreninae are produced.

January 12, 1905.

References.

1. E. Loew. Blumenbesuch von Insekten an Freilandpilan-

zen, Jahrbucb des bofanischen Gartens zu Berlin, III, 1884.

2. Chas. Robertson. Flowers and insects, XIX, Botanical

Gazette, XXVIII, p. 27.

7

OBSERVATIONS ON THE WINTERING OF GRAIN

RUSTS.

A. H. CHRISTMAN

The manner in which grain rusts pass the winter in northern

climates and in regions where the barberry is wanting, is still

considered an open question. Eriksson and Henning (1), while

they record the germination of uredospores collected during the

winter and early spring, conclude that these spores and the

mycelia producing them play no important part in perpetuat¬

ing the fungus. As is well known, they hold that rust may be

transmitted as a nonhyphal, so called, mycoplasm in the cells

of the host. It is quite possible that their adherence to the my¬

coplasm hypothesis on other grounds has biased their judgment

on the question of the ability of uredospores and mycelium to

pass the winter. Ward (2) rejects the mycoplasm hypothesis

entirely and probably most mycologists regard it with great

skepticism.

On account of the position taken by Eriksson and Henning,

it may be well to review some of the earlier observations re¬

corded as bearing on this question. As early as 1875 Kuhn (3)

found uredospores of P. graminis Pers. in the early spring near

Halle, in Germany, and in same year Nielson* 2 * 4 concluded that

in Denmark, P. rubigo-vera D. C., passes the winter as a myce-

lEricksson and Henning, — Die Getreideroste Stockholm, 1896.

2 Ward, H. M. On the Histol. of U. ddspersa, Erikss., and the “My¬

coplasm” Hypothesis. Phil. Trans. Hoy. Soc., Yol. 196, p. 29, ’03.

3Uber die nothwendigkeit eines Verbotes der Pflanzung und Anlage

der Berberitzenstrauches. Kwhn J. Landw. Jahrb. Bd. 4, 1875, p. 399.

4 p. Nielsen,— De for Landbruget farligeste Rustarter og Midlerne

mod dem. Ib., 1875, Bd. 1.

Christman — Wintering of Grain Rusts. 99

lium in the leaves of grain. He gives it as his opinion that

the rust retains its vitality so long as the leaves of the host

remain green. Blomeyer,1 in 1876, found uredospores of P.

graminis Pers. on grain near Leipzig in May and maintains

that these had matured too early to he attributed to infections

from an aecidium of that season. In England, Plowright2

found open pustules of uredospores of P. graminis on Triticum

repens on December 31, 1881, and on the same grass again in

March of the next year. Yon Thiimen,3 in 1886, also observed

that in Austria uredospores continue on certain grasses through¬

out the entire year.

The investigators mentioned above give us no data as to the

minimum temperatures in the seasons in which their observa¬

tions were made. It will be remembered, however, that the

minimum temperature for the winter in Denmark, Germany,

and Austria is about the same as that of Southern Illinois or

Kentucky, while that of England is about the same as that of

Tennessee.

Similar investigations were undertaken by H. L. Bolley4 at

Lafayette, Indiana. At various times during the winter of

1888-1889, he found healthy mycelium within the leaves of in¬

fected wheat plants. During the first warm days of March there

was a general outbreak of uredospores. Bolley concludes that

wheat rust passes the winter in Indiana as a mycelium within

the host.

Ericksson and Henning,,5 at Stockholm, Sweden, where the

winter temperatures are similar to those of Madison, Wisconsin,

record the finding of viable uredospores of Puccinia dispersa in

abundance on November 29th, 1891, and again on the same

plants, April 2nd, 1892. The postules then disappeared and

fresh ones were not again found until April 30th. Viable

1 Blomeyer, — Vom Versuchfelde des Landwirthschaftlichen Institutes

zu Leipzig. Landw. Bd., 25, 1876.

2 Plowright, — The connection of wheat mildew with the barberry

aecidium. Gardner’s chronicle. Series 2, Vol. 18, 1882.

3 Von Thiimen, — Die Bekampfung der Pilzkrankheiten unserer Kul-

turgewachse. Wien., 1886.

4H. L. Bolley, Wheat rust. Bull. Agr. Exp. Station of Indiana. No.

26, 1889.

5Loc. cit

100 Wisconsin Academy of Sciences, Arts, and Letters.

uredospores of P. graminis Pers., were found on Aira caespitosa

on April 1st, 1892, and of P. Plilei-pratensis Erik, and Henn.

December 29th, 1891, and again on. March 28th, 1892. In ex¬

perimenting with P. glumarum (Schm) Erik, and Henn., cer¬

tain leaves were marked in the fall of 1892 and the amount of

infection noted. After a very severe winter the leaves were

again examined April 27th, 1893, and three of ninety which

were marked were found to bear open pustules. P. coronata

Corda was also found in late fall and early spring. Viable

uredospores of this rust were found on Melica nutans, Novem¬

ber 31st, 1891, and in the same place April 5th, 1892. Besides

the above records, the abundance of uredospores of P. glumarum

on February 5th, 1894, is incidentally mentioned. They also

noted that a general outbreak of uredospores of the same rust

occurred on wheat within a week after the snow disappeared in

the spring of 1892.

In spite of these facts, it is Eriksson’s conclusion that, in a

climate like that of Sweden, the rusts do not winter to any ap¬

preciable extent either as uredospores or as a mycelium within

the host plant. His statements to this effect are most positive

except in the case of P. glumarum. Even in the case of that

rust, he does not believe that the number of uredospores appear¬

ing at the close of winter ^ three rusted leaves in ninety as

shown by his experiments) would be sufficient to insure the con¬

tinuation of the fungus. He holds for the wintering of rust,

as for its propagation through seed, to his well known myco-

plasm hypothesis. The evidence for the actual existence of

this non-filamentous mycoplasm imbedded in the protoplasm of

the host cells as presented by Eriksson and Henning, and later

by Eriksson and Tischler1 seems entirely inconclusive. The

structures described are doubtless artefacts or distorted ele¬

ments of the host cells themselves. The temperatures at Stock¬

holm for the winters of 1890 to 1894 are recorded and may be

tabulated as follows for comparison with the temperatures at

Madison, as given in a later table.

i Eriksson and Tischler, — K. Svenska Vetenskaps Akademiens Hand-

lingar. Bd. 37. No. 6, 1904.

Christman — Wintering of Grain Busts.

101

Temperature records at Stockholm.

In a recent paper1 Klebahn describes observations which pos¬

sibly suggest the existence of a my coplasm such as Eriksson and

Tischler2 have described in their latest communication on the

subject. Klebahn figures and describes minute granules within

the host cells which stain, as he says, like the nuclei of the rust

hyphae. These granules are very small and do not show the

membrane, chromatin, and nucleole, which I have always found

in the nuclei of rusts whether collected in winter or summer.

Klebahn himself is doubtful whether the structures observed

are not artefacts.

In the winter of 1 892-1898, Hitchcock and Carleton3 made

a series of observations on the wintering of P. rubigo-vera D. C.

at Manhattan, Kansas. They record the germination of uredos-

pores collected November 5th, January 9th, January 17th, Janu¬

ary 24th, January 25th, February 25th, and March 1st. They

also record the maximum and minimum temperatures of each

of the winter months.

Maximum. Minimum1.

December . 67° — 9°

January . 53° — 1°

February . 61° — 6°

Considerable snow covered the ground during the winter.

Carleton4 gives it as his opinion that in the latitude of Man-

1 Klebahn, — Einige Bemerkungen uber das mycel des Gelbrostes.

Bericbte der Bot. Gesell., XXII, 1904.

2Loc. cit.

3 Hitchcock and Carleton,— Preliminary report on the rusts of grains,

Kan. Agr. Exp. Station, Bull. 38, 1893.

^ Hitchcock and Carleton, — Rusts of Grains, Kan. Agr. Exp. Station,

Bull. 46, 1894.

102 Wisconsin Academy of Sciences , Arts, and Letters.

hattan the grain rusts winter as a mycelium within the host and

produce spores from time to time as the warmer periods occur.

In a still more recent paper, Carleton1 asserts that the uredo on

Poa pratensis winters alive as far north as Lincoln, Nebr.

In order to further test and determine in a latitude still

farther north, the ability of rusts to winter as mycelium and

uredospores, the writer undertook to follow the history of sev¬

eral of our common rusts through the winter of 1902-1903.

In the fall of 1902, volunteer grain was very abundant in the

vicinity of Madison and well rusted plants could be easily found

until late in November. Luring the winter and early spring

material was gathered from plots of Sclilansted rye and Bed

Clausen winter wheat on the University Experimental Farm.

The Poa and oat material was gathered from plots in the city.

The plots on the University Farm were situated on a piece of

ground sloping to the northwest. From these plots the snow

drifted leaving the ground bare during the greater part of the

winter. The oat and Poa plots were well covered with snow.

From time to time plants bearing uredospores were taken into

the laboratory and the spores were germinated in water. These

water cultures were made within a few hours of the time of

gathering the material. The dates of collecting and the pre¬

vailing temperatures may be taken from the following table.

Initial letters are used to indicate the particular rust collected.

There are also shown the maximum and minimum temperatures

for each day of the month, as indicated at the head of each

vertical column. Spores were germinated on every date noted

except in the case of P. poarum collected February 18th.

i Carleton, — Investigations of Rusts, Bull. 63, Bur. Plant. Ind., U. S.

Dept. Agric., July, 1904.

Prevailing temperatures and dates of collections of rusts, Madison , Wisconsin , 1902-1903 .

Christman — Wintering of Grain Rusts.

103

g. t. = Puccinia graminis tritici. r. s. = Puccinia rubigo-vera secalis. p. = Puccinia poarum,

c. = Puccinia coronata. r. t. = Puccinia rubigo-vera tritici. * — No germination.

104 Wisconsin Academy of Sciences , Arts, and Letters.

It is plain from the above table that in the latitude of Mad¬

ison and with a period of three months, during which the tem¬

perature scarcely rises above the freezing point viable uredo-

spores may be obtained at practically any time during the win¬

ter. As noted also, the spores were taken from very exposed

situations. In each case about 10 °/o of spores germinated. The

cultures made after January 1st did not vary greatly in this

respect excepting in the case of the oat rust spores collected

January 26th. Of these spores 60% germinated. In all cases

the germ tubes were healthy in appearance and reached a nor¬

mal development. The time required for germination varied

from three to six hours.

In order to make permanent preparations showing the per

cent of spores which germinate and the length and appearance

of the germ tube, the following method was devised. On one

side of a clean slide a thin him of albumen lixative, prepared by

mixing equal parts of egg albumen and glycerine, is applied.

On a slide so prepared is placed a drop of distilled water con¬

taining the spores. When germination has occurred, the water

is allowed to evaporate until the slide is nearly dry and the slide

is then immersed in a killing solution. I have found Flem¬

ming’s weak solution very satisfactory. Thirty minutes is

usually sufficient time for the exposure. After killing, wash

with water and harden by carrying the preparation through the

different grades of alcohol, as follows :

30% . 3 minutes.

50% . 5 minutes.

70% . 5 minutes.

80% . 5 minutes.

95% . . . 5 minutes.

100% . 1 minute.

The slides may then be bleached or stained at once. Short ex¬

posures to the stains are most satisfactory. I have used Flem¬

ming’s triple stain exposing to the Saffranine three-fourths of a

minute; Gentian violet', two minutes, and using the shortest

possible exposure to the saturated solution of Orange G., or

using that solution diluted with three times its volume of water.

Mayer’s haematoxylin brings out the nuclei perhaps even more

clearly. This stain was prepared by mixing a solution of 0.1

Christman — Wintering of Grain Busts. 105

gm. of Haematein dissolved in 5cc. of 90% alcohol with a solu¬

tion prepared by dissolving 5 gm. of alum in 300cc. of water.

Thirty minutes’ exposure to this stain followed by a very short

exposure to the dilute Orange G., gave the most satisfactory

stain tried. This method of fixation and staining has been em¬

ployed with good results in preparing slides of various kinds

of germinating spores and also of various kinds of germinating

pollen-grains. The albumen fixative seems to have no injurious

effect in any case which was tried.

It may be noted early in the winter that in the case of badly

rusted grain, many of the leaves become spotted with pale areas.

Sections showed that the tissue beneath these spots contained

mycelium and undeveloped spores. In order to determine

whether such spots developed further and became open pustules

during the winter and also to determine whether the oat plants

could be wintered over with some protection a clump of badly

spotted oat plants was covered in the following way. A bank

of earth six inches high, enclosing an area twenty by twenty-two

inches, was built around a patch of oat plants and the whole

covered with a pane of glass. Certain leaves were placed in

such a position that they could be easily seen. From time to

time during the winter, the snow was removed from the glass

and a little warm water applied when the oat plants could be

seen very clearly.

The oat plants were enclosed as above described on November

18th, and a number of the pale spots marked. On December

20th, it was observed that the epidermis was ruptured and open

pustules had been formed in certain of the marked spots. Many

of the white spots were still apparent. On January 17th the

snow was again removed. The pustules and spots were in much

the same condition as on December 20tli. The oat plants were

fresh and healthy except where in contact with the glass. On

January 23rd the conditions under the glass remained un¬

changed. After this date the experiment was interrupted and

no observations were made on these plants. It seems probable,

however, that uredospores and mycelium of P. coronata are at

least as resistant as the oat plants on which they grow. As is

shown in the table, spores from oats were germinated as late as

106 Wisconsin Academy of Sciences , Arts, and Letters.

January 26th, and it seems probable that the mycelium would

have withstood the remainder of the winter if the host could

have been kept alive.

Besides the data recorded in the tables, observations Avere

made on the spring development of the rust of wheat and rye.

On March 5th, a number of the winter leaves of rye, spotted in

the manner described above, were marked. Fifteen days later,

or March 20th, it was found that many of the spots had devel¬

oped to form open pustules. On this last date a general out¬

breaking of new pustules was noted, which reached its height

on about April 3rd, when fresh vigorous spores were abundant.

The winter leaves novr began to wither and disappear. After

April 8th, there Avas a period of about four weeks when it was

impossible to find a single spore. On May 6th, the new leaves

began to show a diseased appearance. The plot was visited

again May 13th, when open pustules were to be found in abund¬

ance.

It will be seen that there were in the spring in question twro

distinct outbreaks: The first occurred on the old wdnter leaves

in the two weeks following the first warm weather. Uredo-

spores did not appear again until sometime later when a second

outbreak occurred on the first of the spring leaves.

Eriksson and Henning also describe two distinct outbreaks

and use this fact as evidence for their mycoplasm hypothesis.

The first appearance of uredospores they admit is often caused

by mycelium that has survived the winter, but the later out¬

break comes from the mycoplasm. In their observations on P.

ddspersa, they found uredospores April 2, 1892, but think even

in this case that they have no reason for believing that these

spores were produced by mycelia living over from the previous

year. The explanation of the two separate outbreaks is prob¬

ably to be sought in another direction.

I have found by experiment that in the cooler weather of

spring the incubation period following inoculation with uredo¬

spores is usually lengthened to between three and four weeks

and this explains the existence of a period with no rust after the

first attack. The winter leaves die in early spring and with

them the winter mycelium, but not until it has produced uredo-

Christman — Wintering of Grain Busts. 107

spores which inoculate the new leaves. Then follows a period

of incubation which may be lengthened more or less according

to the temperature and other conditions in the spring.

The possibility that uredospores may be carried great dis¬

tances by the wind and in this way move northward to infect

the crops in successive regions is very interesting as bearing on

this question. Klebahn has recently attempted a quantitative

determination of the number of rust spores occurring in the air.

He collected spores on sheets of cotton exposed to the open air

during the summer of 1901. His results show that in the course

of the summer immense numbers of uredospores are to be found

floating in the air. 'Whether sufficient floating uredospores are

present at the first of the growing season to cause the abundant

outbreaks of rust frequently observed, requires further proof.

As noted above Carleton thinks that viable spores found dur¬

ing the winter are produced from time to time during the peri¬

ods of warmer weather. This, indeed, is very likely in a climate

like that of Kansas where the temperature rises to about 60° F.

us a maximum for each of the winter months. From the table

it will be seen that the uredospores of P. rubigo-vera triticu

which were gatheiod on February 6th, were either sixty-seven

days old or had matured at a temperature not higher than 6'

above the freezing point. It will also be seen that the tempera

ture did not rise above 42° F. for a period of ninety-three days.

I am inclined to think that at our winter temperature the uredo¬

spores may remain dormant for long periods without losing

their vitality. Very likely, however, these spores play little

part in producing infections in the spring, since with the first

warm days the mycelium produces new pustules with a fresh

crop of spores.

As the severity of the weather must affect the amount of

healthy host tissue that survives the winter, it must limit the

amount of mycelium and so the number of uredospores at lianc7

in the spring, and is, in all probaoility, one of the chief factors

in determining the violence of early outbreaks of rust.

Madiso7i, Wisconsin,

November i, 1901.

HABITS AND ANATOMY OF THE LARVA OF THE

CADDIS-FLY, PLATYPHYLAX DESIGNATES, WALKER.

C. T. VORHIES.

The larvae of the caddis-fly, Platyphylax dcsignatus Walk.,

are found in great numbers in a certain group of cold springs'

on the southern shore of Lake Wingra, near Madison, Wisconsin.

There are several other large springs about the shoves of the

same lake, but the larvae are not abundant in any of the others

and in some are not found at all. The conditions found in the

group inhabited by the larvae are as follows: cold water in

abundance throughout the year at a temperature of 8° C. never

freezing in the most severe winter weather: plenty of clean

rather fine sand, with numerous coarser particles; many larger

stones, under which the larvae lie hidden during the daytime;

Water-cress, Nasturtium Officinale , in great quantities, and some

water-milfoil, Myriophyllum, on which plants the larvae feed.

A few larvae may be found during the day under the denser

clusters of water-cress, where there is little light, but not much

evidence of feeding by day may be seen. As the loose stones-

under which the larvae are hidden are often at a distance of

five or six feet or even more from the food plant, and as the in-

testine is always found distended with food in these specimens,

the conclusion is at once forced upon us that they feed almost

entirely at night. The fact that during the day the larvae in

dishes in the laboratory cluster in the darkest shelter obtainable

lends support to this conclusion. When a loose stone is lifted

under which dozens of larvae are gathered, what at first appears

to be a mass of sand begins to heave and move and soon resolves

itself into a number of individual larval cases, each being labori-

VorJiies — Larva of Platyphylax Designatus. 109

■ously dragged away to a new retreat by a visible brown head

and six legs. Hundreds of these larvae may be seen in a few

minutes time in this one group of springs.

The case is very beautifully constructed of sand grains, and

is in the form of a slightly curved tube, (Fig. 21.) open at both

ends, though the posterior, narrower end, usually has the mar¬

gin turned in so as to partially close the orifice. (Fig. 22.) The

concavity of the case is ventral and a slight projection or hood

extends forward from the dorsal portion of the anterior mar¬

gin.

The eggs of this caddis-fly are deposited in large numbers in

April. They are attached to the lower surfaces of loose stones,

mostly at the edge of the water, in very moist situations. The

larvae hatch in a short time, probably in less than two weeks,

though the exact time has not yet been determined. They are

.about 1% mm. in length wdien first hatched, and their heads

are larger and legs longer relatively than the same parts of

older larvae. The interesting fact was noticed that these newly

hatched larvae are positively heliotropic to a marked degree

when on a dry surface, but at once become negatively helio¬

tropic when placed in a dish of water. The necessity of getting

out from beneath the stones where the eggs are placed in order

to find water, and of getting beneath stones for protection while

building a case, after reaching it, offers an explanation of this

peculiarity. The young larvae at once begin building cases

when placed in a dish of water with sand in it, and are capable

of fashioning a fairly good one in four or five hours. They

probably do not feed until safely housed in a case. Small larvae

a few millimeters in length are plentiful in the late summer and

early fall. From November to January" more and more larger

larvae are found and small individuals become few in number.

About the middle of February the majority of the cases are

found to have larger irregular stones attached to the anterior

ends, evidence of the approach of pupation, while some are found

fastened to the lower surfaces of the large rocks by a mass of

silk at the anterior end. Many of the latter are also closed

at the posterior end with larger stones of the same kind as

those already mentioned. If the closed cases be broken open

110 Wisconsin Academy of Sciences , Arts , and Letters.

at such a time very few pupae will be found, but numbers of

larvae in an inactive state may be seen. Several days seem to

be necessary for pupation after closing the case. The first adult

was reared about March 15. On April 18 many adults were

taken beneath the stones about the springs and at the same

time many eggs were found. Probably some were out as early

as the first of April.

Gross dissection was found practicable as a means for much

of the work done, the respiratory system being worked out en¬

tirely in this way. The full grown larvae, — in the last instar —

average about 16 mm. in length by 3% in breadth of abdomen.

The greatest hindrance to dissection wras encountered in the

numerous leaf -like fat bodies with which the organs are sur¬

rounded. In a few specimens dissected after being dead a few

hours the fat bodies could be washed out easily by means of a

pipette. For microscopic preparations the common fixing agents

and stains were used. Delafield's haematoxylin was found serv¬

iceable in staining the nuclei of the spinning glands, for whole

mounts of the same. These glands, after fixation, may be split

with a fine scalpel along one side and spread out. This shows^

the whole area of the surfaces of the cells and gets rid of the

secreted mass of silk within the lumen, wdiich, taking the stain,

would otherwise obscure the nuclear structures, A rather long

exposure to the haematoxylin is necessary to penetrate the nu¬

clei of such cells. The cytoplasm is thus stained a deep blue

which must be washed out with acid alcohol. The addition

of a counter stain, such as eosin, rather tends to obscure the nu¬

clei than to aid in differentiation, owing to the thick mass of

cytoplasm.

Respiratory System. — The internal respiratory system

consists of two large longitudinal tracheae extending throughout

the body and lying in a series of curves, each the length of a

segment. (Fig. 1.) Each trachea has connected with it a

series of smaller branches which pass to the external respiratory

filaments and to the various organs and muscles of the body.

In the prothorax each longitudinal trachea divides into two-

large branches which pass forward into the head, entering at

different levels. Before entering the head each of these again

V orhies — Larva of Phityphylax Designatus. Ill

divides, the four subdivisions supplying the head only. The

dorsal branch (A) will be described first. One of its two

branches lies ventral to the other for a short distance, then

curves outward and passes through the muscles directly to the

eye, giving off several small branches which at once break up

in the muscles. From a point near the eye this branch then

curves inward again and, after making a dip posteriorly, where

it gives off a small branch, it curves caudad and ventrad and

passes beneath the brain, giving off a small branch to that organ,

and then another small branch which anastomoses below the

brain and dorsal to the oesophagus with its fellow of the op¬

posite side. From here this trachea may be considered as a

branch of the ventral trachea (B) with which it is continuous.

The dorsal branch of A passes slightly inward so as to lie along¬

side the corresponding opposite tracheae, giving off immediately

after entering the head a branch on the outer side which curves

upward and breaks up in the muscles of the top of the head.

At a point posterior to the brain the main trachea forks, the

inner, larger branch, immediately anastomosing with its oppose

ite, while the outer, smaller one, curves outward and upward

and breaks up in the muscles. From the point of anastomosis

a median trachea passes forward dorsal to the brain and sup¬

plies the muscles of the labrum.

Trachea B divides into two nearly equal branches before en¬

tering the head. The outer one curves outward and passes in a

direction nearly identical with the outer branch of A but at a

lower level, giving off small branches to the muscles, and finally

ending in the mandible. The muscles supplied by this branch

seem to be mostly mandibular. The inner branch of B passes

cephalad and only slightly inward, the first important branch

being found posterior to the median anastomosis of A. This

branch soon divides, a small part passing forward to the mouth

and the main part passing upward to anastomose with the

branch described from A. Another important branch from the

inner fork of B passes outward and forward and supplies mus¬

cles which seem to be manibular. The remaining portion again

divides into two parts which pass forward and break up in the

muscles of the floor of the mouth. Between the first and second

112 Wisconsin Academy of Sciences, Arts, and Letters .

branches of the inner fork of B, but from the inner side, a small

branch passes off to the sub-oesophageal ganglion.

The dorsal branches are shown on the right in the figure (Fig.

1.) while those ventral to the intestine or which pass to external

respiratory filaments are shown on the left. A regular series of

large branches is seen _to supply the intestine with the excep¬

tion of the oesophagus proper. Each of these branches breaks

up in a complex fashion on its own side of the intestine. The

large intestine is particularly well supplied with numerous small

branches vrhich are not arranged on the regular plan of the

mid-intestinal supply. A smaller longitudinal trachea lying on

either side of the dorsal median line is formed by the union

of loops arising from the main trunk. Additional similar loops

in the meso- and metathorax are peculiar for the manner in

wdiich they dip down toward the leg joints, there giving off in

each case a branch which, aided by two others, supplies the two

pairs of limbs arising from these segments. There is a sug¬

gestion here that these loops may later become the source of sup¬

ply for the wings also. Of the three branches supplying the

meso-and metathoracic legs, the one sweeping in a curve toward

the median line before entering breaks up in the first segment

of the leg. The second fork of this same branch passes to the

second segment, while that from the loop continues to the ex¬

tremity, supplying the remaining segments. The branches sup¬

plying the thoracic ganglia all unite with their opposites by a

small anastomosing branch just anterior to the ganglia. The

spinning glands, notwithstanding their activity in the secretion

of silk, receive no regular supply of important tracheae, the

only noticeable branch being a small one in the prothorax. The

branches connected with respiratory filaments are marked with a

circle at the point of exit through the body wall. Some of the

external filaments consist of several branches, in which case

the tracheae break up accordingly.

Alimentary Tract. — The alimentary tract is straight, of

the same length as the body, and begins at the mouth as a

small thin-walled tube. It extends to about the end of the met¬

athorax with but little variation in size, narrowing slightly at

the junction of the head and thorax, thence gradually widen-

YorMes — Larva of Platyphylax Designatus. 113

ing to the beginning of the next division, the mid-intestine.

The posterior part of the fore-gut seems to function as a sort

of crop, as it is frequently seen somewhat swollen with food

when it assumes a rounded outline. (Fig. 1.)

The mid-intestine, as may be expected from the herbivorous

habit of the larva, is very large when normally tilled with food.

It begins with an abrupt enlargement of the alimentary tract

at the posterior border of the metathorax. The fore gut ap¬

parently is telescoped into this. The width of the mid-intes¬

tine at this point is more than one-half that of the average ab¬

dominal segment. From here to the posterior end, in the mid¬

dle of the sixth segment, it tapers gradually to less than one-

third the width of the abdomen. The end of the mid-intestine

is marked on its external surface by the attachment of the Mal¬

pighian tubules, which are six in number. These tubules ex¬

tend anteriorly to the first abdominal segment; they then turn

back on themselves, extending posteriorly to the eighth segment,

when they again turn forward and end in the sixth or seventh

segment after forming several loops in that region. They are

pigmented so as to appear reddish brown, the pigment granules

being generally grouped most thickly near the nuclei.

The hind-intestine may be divided macroscopically into a large

posterior portion beginning in the seventh segment, and a small

intestine only about the length of one segment, and rather nar¬

row. In the circular furrow formed by this narrowing of the

alimentary tract the distal folds of the Malpighian tubules are

very numerous. The anus is a vertical slit in the posterior end

of the ninth segment, between the projecting parts on which are

borne the prolegs.

When examined microscopically by means of sections, the

oesophagus is found to consist of an epithelium of thin flattened

cells, lining which is a chitinous layer bearing groups of chit-

inous spines, which point backward toward the mid-intestine.

From two to eight or nine spines constitute a group. (Fig. 2.)

The muscular coats consist of an inner circular and an outer

longitudinal layer, both being striated. Sections, both longi¬

tudinal and transverse, show clearly that the fore-intestine is

telescoped into the mid -intestine for a short distance. The

8

114 Wisconsin Academy of Sciences, Arts, and Letters.

portion thus extending into the mid-intestine as a double fold

has the same character of epithelium as the fore-gut, and further

proof of its origin is given by the layer of chitin extending to

the point where the folded portion meets the anterior end of

the mid-intestine. The length of this part is about one-half

the width of the anterior end of the intestine,, so that if pushed

forward by the food it might nearly, if not quite, close the open¬

ing. An oesophageal valve is thus formed. (Fig. 3.) As the

posterior portion of the oesophagus as well as the intestine is

usually distended with food, the chitinous hooks of the oesopha¬

gus probably aid in the function performed by the valve.

At the beginning of the mid-intestine there is a marked change

in the epithelial cells, which are here columnar and bear on

their inner surfaces a well-marked peritropliic membrane, the

thickness of which is about equal to the average width of the

cells. (Fig. 5.) Nests of regenerative cells, very similar in ap¬

pearance and staining reaction to those described by Needham

(7) for certain dragon-fly nymphs, are numerous, and placed

at regular intervals. A thick basement membrane and longitud¬

inal muscle fibers are present, the latter somew'hat scattered.

The circular fibers form a nearly continuous coat and are un-

striated.

This columnar epithelium extends to a point a little posterior

to the openings of the Malpighian tubules, but the peritrophic

membrane is supplanted by a chitinous layer just at the poster¬

ior border of the lumen of the tubules. That is, the beginning of

the hind-intestine is marked only by the change from peritrophic

membrane to chitin, and not by an immediate change in the

character of the epithelium, except that no nidi are present pos¬

terior to the tubules. (Fig. 4.) Within a short distance, how¬

ever, the columnar epithelium gives way suddenly to a layer

of flattened epithelial cells similar to those of the oesophagus.

The chitin covering this area of columnar epithelium of the

hind-intestine, is beset with numerous spines. At about the

middle of the small intestine another band of chitinous spines

is developed and a heavy band of circular muscles is present

just at this point, indicating that the structure probably acts

as a rectal valve. (Fig. 6.)

Vorhies — Larva ofPlatypliylax Designatus. 115

The larger posterior portion of the hind-intestine can be read¬

ily divided microscopically into a large intestine and a rectum.

In section the wall of the former is seen to be pouched or folded

forward over the posterior end of the small intestine. The ep¬

ithelium of this part consists of a single layer of large, flat cells

of considerable thickness, containing nuclei with finely granular

contents., (Fig. 7.) The inner surface of these cells has a per¬

ipheral membrane of about the thickness of the peritrophic

membrane of the mid-gut and of somewhat similar appearance,

being marked with striae perpendicular to the surface. (Fig.

8.) A constant layer of substance, — blue staining in haema-

toxylin and in triple stain, — lies within the lumen. This makes

it appear that the cells are secretory, but Yan Gehuchten, (8)

who figures similar cells for the larvae of the Dipteron, Ptyckop-

tera contaminata, takes the ground that they are exclusively ab¬

sorptive in function. Circular muscles are found within the

peritoneum, but no longitudinal fibers. In the furrows between

six irregular outfoldings or loose pouches of the large intes¬

tine lie six bands of longitudinal muscles, the latter being out¬

side the peritoneum.

At the beginning of the last segment the character of the

epithelium again changes abruptly and the lumen gradually be¬

comes vertical. The epithelium here also consists of a single

layer of flat cells, but their size and thickness is much less than

that of the cells of the large intestine. This portion is the

rectum proper, though no accumulation of waste material is

ever found within it, Within the anal aperture and extending

out on the external surface a short distance is an area bearing

strong, slightly curved chitinous teeth. Betten (1) apparently

has overlooked this last division of the intestine.

Spinning Glands. --The spinning glands of the larva of

P. designatus are very well developed and have been found very

interesting as regards the nuclear structures. The glands are

about one and one-haJf times as long ns the body and lie ventrad

and laterad to the intestine, each forming three principal folds,

with the distal end lying in smaller curves of varying shape.

(Fig. 12.) Each gland may be readily divided into conducting

and secreting portions, a slight enlargement with a constriction

116 Wisconsin Academy of Sciences , Arts, and Letters.

in its middle showing the exact location of the dividing lines.

In a microscopic preparation the character of the nuclei may be

seen to change abruptly at this constriction. The anterior,

conducting portion lies in the head and at the base of the labium

joins with its fellow in a common duct rather strongly chitinized

and leading to an opening at the tip of the labium. In this

chitinized portion is the ‘ ‘ press ’ ’ which has been fully described

by Gilson. (2, 3.)

The secreting portion is made up throughout its length of

cells containing the characteristic branched nuclei of spinning

glands of insects but in the small cells near the anterior end

the branching is not extensive. (Fig. 14.) Each cell, in sur¬

face view is typically the shape of a flattened hexagon, the

shorter axes lying in the direction of the length of the gland,

two such cells forming the entire circumference. Such a gland

when opened along one side and spread out gives the appearance

seen in the diagram. (Fig. 13.) The nuclei do not show dis¬

tinct centers of branching, as figured by Ilenneguy (6), nor

are they broken up into separate fragments, as described and

figured by Gilson. (3) They do, howrever, in some cases of

complex branching have this appearance, but a careful examin¬

ation with a high power shows a connection, in every case, of

such apparently detached pieces with the main body of the

nucleus. Whether or not the branches ever anastomose is dif¬

ficult to determine. Such anastomoses are shown by Helm (4)

.for the nuclei of Lepidopteran glands, but such cases are rare,

if they ever exist, in the corresponding structures of P. desig¬

nators. What at first appear to be such anastomoses are com¬

mon in the larger and more complex nuclei, but careful focus¬

sing and study of the outlines of the nuclear membrane, in a

majority of cases, proves beyond a doubt that it is only appar¬

ent and not real, the appearance being caused by overlapping

branches lying at different levels. (Figs. 15-20.)

Nervous Svstem. — The nervous system shows a small

degree of concentration in the larval stage. It consists of a

chain of thirteen double ganglia, (Fig. 9.) which Betten (1.)

says is the number given by Klapalek* for Trichoptera, though

*Klapalek’s work was not available.

Vorhies — Larva of Platyphylax Designatus. 117

he (Betten) noted ony six abdominal ganglia, or a total of

eleven, in the larva of Molanna cAnerea. The supra- and sub-

oesophageal ganglia show nothing worthy of particular note.

The three thoracic ganglia occupy the pro-, meso-, and meta¬

thorax, the third lying slightly anterior to the middle of the

metathorax. The first abdominal ganglion, smaller than the

third thoracic, lies quite close to it posteriorly, within the meta-

thorax. The second abdominal ganglion, slightly smaller and

more elongated than the first, lies near the middle of the first

abdominal segment. The third ganglion, about the size of the

first, lies at the juncture of the second and third segments, and

is the only ganglion found in these two segments. The fourth,

fifth, and sixth ganglia lie in their respective segments, but the

seventh lies just within the posterior border of the sixth seg¬

ment and the eighth, closely applied to it posteriorly, lies just

within the anterior border of the seventh segment. Although

the last two ganglia are closely applied to each other they are

entirely separate, as may be readily seen when stained and ex¬

amined microscopically. This arrangement of ganglia will be

seen at once to offer a possible explanation of Betten ’s failure

to find more than six abdominal ganglia, because, in serial sec¬

tions, only six abdominal segments would probably be noted to

contain ganglia. These would be the first, either the second

or third, but not both, and the fourth, fifth, sixth and seventh.

In the two cases where two ganglia are rather closely applied to

each other, very careful observation is probably necessary in

order to distinguish that two are present, though the point may

be easily seen by means of dissection. The first abdominal gang¬

lion would hardly be looked for in the thorax A

Eeproductive System. — The only traces of reproductive or¬

gans found in the larvae are ovaries and testes in the

early stages of development. These lie in the fourth and fifth

segments, and are found only in the large larvae which are pre¬

paring for pupation, which period is determined by the addition

ovaries, which are easily distinguishable, are elongate bodies,

* A dissection of a species of Molanna has since been made and the

ganglia found substantially the same as in P. designatus.

118 Wisconsin Academy of Sciences, Arts, and Letters.

of larger particles of stone to the anterior end of the case. The

lying deep down on either side of the mid-intestine. The testes

are similarly situated, but are not so readily observable, par¬

ticularly in their later stages of development, as they are then

closely enveloped with a layer of fat of the same color as the

surrounding fat bodies. The sperms become quite well de¬

veloped in the latest larval stages, but accessory organs of re¬

production do not appear.

Glands of Gilson. — In the prothorax of the larva care¬

ful dissection discloses a small, elongated structure, with irreg¬

ular, wavy outlines, lying beneath the oesophagus. (Fig. 9.)

Its anterior tapering end passes ventrad between the connect¬

ives uniting the sub-oesophageal and the first thoracic ganglia,

and may be traced to a connection with the base of a curved

chitinous spine, which, lying between the first pair of legs, curves

forward close to the head. A microscopic examination makes it

clear that this is a glandular structure, with its opening at the

tip of the spine. (Fig. 10.) The posterior end of the gland

is free in the cavity of the body and may lie either to the right

or to the left of the median line. This is the only representative

in P. designatus of the Glands of Gilson, so-called from the in¬

vestigator who first described them. The original paper was

not available, but M. Henseval (5) has given an interesting ac¬

count of the glands as studied by him in several Trichopterous

larvae. In the possession of only one of these structures, P.

designatus comes in the same group as Limnophilus fiavicornis,

L. rhombicus, and Anaoolia nervosa, as opposed to Phryganea

grandis and an undetermined Phryganid in which three glands

of a somewhat more complex structure are found, one in each

thoracic segment. Henseval (5) offers proof of an oily secre¬

tion and ascribes an excretory function to these glands.

Under direction of Prof. Wm. S. Marshall,

Zoological Laboratory, University of Wisconsin,

Madison, May, 1905.

Vorhies — Larva of Platyphylax Designatus.

119

PLATE VII.

120 Wisconsin Academy of Sciences, Arts, and Letters.

EXPLANATION OF PLATE VII,

All figures except dissections are drawn with a camera. Magnifica¬

tion in diameters given after explanation of each figure.

Fig. 1. Dorsal view showing tracheal system and alimentary tract.

Secondary longitudinal trachea, SI., thrown to the outside of the

main trunk. A., dorsal branch entering head. B., ventral branch

entering head. Pr., prothorax. Mo., mesothorax. Mt., metatho¬

rax. Circles on the ends of the tracheal branches indicate points

where same enter external respiratory filaments. Oe., oesophagus.

C., crop. Md., mid-intestine. Mp., Malpighian tubule. S., small

intestine. L., large intestine. R., rectum. X 8.

Fig. 2. Surface view of portion of chitinous lining of oesophagus, p.,

posterior border. X 300.

Fig. 3. Longitudinal section of oesophageal valve. An., anterior. Ch.,

chitin. Pm., peritrophic membrane. Cm., circular muscle. Lm.,

longitudinal muscle. X., point showing end of chitin and begin¬

ning of peritrophic membrane. X 110.

Fig. 4. Longitudinal section through junction of mid- and hind-intes¬

tine. Mp., Malpighian tubule. Pm., peritrophic membrane. X.,

beginning of chitinous layer, Ch., of hind-gut. Cm., circular

muscle. Lm., longitudinal muscle. Y., beginning of flattened epi¬

thelium of small intestine. Pg., pigment granules. X 110.

Fig. 5. Section through nidus of mid-intestine. Cm., circular muscle.

Lm., longitudinal muscle. Bm., basement membrane. Pm., peri¬

trophic membrane. X 285.

Fig. 6. Longitudinal section through rectal valve. An., anterior. Ch.,

chitin. E., epithelium. Cm., circular muscle. Lm., longitudinal

muscle. X 90.

Fig. 7. Surface view of epithelial cells of large intestine. X 190.

Fig. 8. Section of same. Cm., circular muscle. Pm., peripheral mem¬

brane. X 285.

Trans. Wis. Acad. Vol. XV.

Plate VII.

2.

C. T. Vorhies del.

Vorhies — Larva of Platyphylax Designatus.

122 Wisconsin Academy of Sciences , Arts, and Letters.

EXPLANATION OF PLATE VIII.

Fig. 9. Dissection of nervous system and Gland of Gilson, G. 1-9, ab¬

dominal segments. I-VIII, abdominal ganglia. X 4.

Fig. 10. Gland of Gilson, showing connection with external hollow

spine. X 43.

Fig. 11. Diagram showing number and position of the external res¬

piratory filaments on one side of body. V., ventral half of seg¬

ment. 1. and 2., first and last segments of abdomen.

Fig. 12. Dissection from dorsal side showing spinning glands. C.,

conducting portion. S., secreting portion. P., union of conduct¬

ing portions in press. X 4.

Fig. 13. Diagram showing shape and relation of cells of spinning

glands when split along one side and spread out.

Fig. 14. Two small cells, showing simple nuclei, from the small an¬

terior end of the secreting portion. X 160.

Figs. 15 and 16, More complex nuclei in larger cells taken anterior or

posterior to the widest part of the secreting portion. X 80.

Fig. 17. Still more complex nucleus. Note pieces X, Y, attached by

long slender threads to main nucleus. X 80.

Figs. 18 and 19. Complex nuclei filling larger proportion of cell than

the preceding. Some apparent anastomoses could not be satisfac¬

torily determined. X 80.

Fig. 20. Complex nucleus showing many pieces attached by slender

threads. Many of the apparent anastomoses in this nucleus are

undoubtedly not real. X 80.

Fig. 21. Larval case of P. designatus, approaching time of pupation,

shown by larger stones on anterior end. X 3.

Fig. 22. Posterior end of same, showing how the border of the case is

turned in. X 3.

Trans. Wis. Acad., Vol. XV.

Plate VIII.

C. T. Vorhies del.

Vorhies — Larva of Platyphylax Designatus.

123

BIBLIOGRAPHY.

1. Betten, Cornelius. The Larva of the Caddis-fly Molanna

cinerea. Journal of the New York Entomological So¬

ciety, Yol. X. Nr. 3. p. 147. 1901.

2. Gilson, G. Recherches sur les cellules secretantes. I. Lepi-

dopteres. La Cellule, T. VI. p. 119. 1890.

3. Gilson, G. Recherches sur les cellules secretantes. II. Tri-

chopteres. La Cellule, T. X. p. 37. 1894.

4. Helm, F. E. Ueber die Spinndrusen der Lepidopteren.

Zeitschrift fiir wissenschaftliche Zoologie, Bd. XXYI.

p. 434. 1876.

o. Henseval, M. Etude comparee de glandes de Gilson, organes

metameriques des larves insectes. La Cellule, T. XI.

p. 327. 1895.

6. Henneguy, L. Les insectes. p. 463. Paris, 1904.

7. Needham, J. G. The Digestive Epithelium of Dragon-fly

Nymphs. Zoological Bulletin, Yol. I. Nr. 2. p. 103-

113.

8. Yan Gehuchten, A. Recherches histologiques sur l’appareil

digestif de la larve de la Ptyckoptera contaminaia. La

. Cellule, T. YI. p. 185. 1890.

NOTES ON THE OCCURRENCE OF OSCILLATORIA PRO-

LIFICA (GREVIELE) GOMONT IN THE ICE OF PINE

LAKE, WAUKESHA COUNTY, WISCONSIN."

EDGAR W. OLIVE.

The presence of this minute alga in Pine Lake is of particular

interest since, so far as 1 am able to determine, it has been here¬

tofore reported from but one locality in America, viz., from

Jamaica Pond, a smail lake in Jamaica Plain, a suburb of Bos¬

ton. Mr. G. J. Hansen called the attention of Professor Birge

to the peculiar growth in Pine Lake, furnishing some notes on

its occurrence, and sent to him material which he had collected

on March 25, 1905, which was floating on the surface of the lake

where the ice had melted. Dr. Birge in turn gave the matter

into my hands for further examination.

Isabel P. Hyams and Ellen H. Richards have had Osciilatoria

prolifica under observation in Jamaica Pond since 1887, and

their three papers on the subject so far published embrace the

“Life History” (01); “Chemical Composition” (02); and

“Coloring Matters” (04). It will be of advantage to review in

considerable detail some of the long-continued studies of these

writers in order to supplement our meager observations in con¬

nection with the occurrence in Pine Lake and thus be able to

judge of the conditions which could bring about, in this instance,

such an unprecedentedly vigorous growth of the plant.

Osciilatoria prolifica was first noted in Jamaica Pond by these

two writers in 1887, and since that time has been so abundant

as almost to exclude other forms of plant life. In some years

it was much more abundant than in others, dependent upon the

*This work was done while the writer was serving as a research

assistant of the Carnegie Institution of Washington.

Olive — Occurrence of Oscillatoria Prolifica . 12*^

temperature and sunlight. During the year in which the plant

was most abundant, it did not entirely disappear during the

winter, but was found imbedded in the ice which was cut from

the water. During the period of most rapid growth — generally

in June — the water of Jamaica Pond frequently became turbid

and opaque. I have myself noted, some years ago, that from the

hills above, the water of this pond appeared reddish or of a

brownish chocolate color in bright sunlight. .The authors state

that after a hot, sunny day, the gas created by the vigorous

:growth of the plant often caused the alga to rise to the surface

:and there float as a reddish brown, frothy cream; on June 11

and 12, 1901, for example, the floating masses were of such

abundance that fully “one hundred barrels might easily have

been obtained” (01, p. 308). When this scum was driven by

the breezes on the shore, it decayed on the rocks, giving to them

a rich purple coating. The odor of decay was “intensely dis¬

agreeable, fetid rather than putrefactive” (01, p. 301).

The authors assert that, whenever the plant was found below

the surface, it was blue-green in color; the red pigment, which

they have tentatively named “rubescin” (04, p. 274), according

to their observations, appeared in the filaments only when a

luxuriant growth of the plant had taken place. They suggest

the theory that the red substance is probably “an important

factor, if not the chief one, in the vigorous growth of plant

life.” (04, p. 274). The authors, in reaching this conclusion,

apparently assume that this red pigment is derived from the

chlorophyll of the Oscillatoria. They further bring in for com¬

parison the association of the red coloring matter in the young

shoots of oak, maple, etc., with the vigorous growth of these

shoots ; and, from the a pparent close chemical relation of the red

color in plants to the haemoglobin of animal blood, they conclude

that “good red blood and a portion of red in the chlorophyll of

green plants wherever found seem to indicate robust life.” (04,

p. 274.)

According to their further investigations, the optimum tem¬

perature for the growth of the alga iies between 64° and 66° F.

After a season of such favorable conditions, a vigorous growth

usually culminated in one or two days, never more, when large

126 Wisconsin Academy of Sciences , Arts, and Letters.

quantities of the plant would come to the surface, buoyed up by

the gas bubbles, a phenomenon to which they refer as the

“blooming time” of the Oscillatoria. They say further in this

connection that this phenomenon never happened except when

an air temperature of 80° F. and over was accompanied by a

bright sun and quiet water. (04, p. 270). Only on two days in

thirteen years did this “blooming” result in a large amount of

scum; while on about five other days of this time; a slight scum

was formed.

The maximum growth, which took place generally during the

long days of June, was followed in July and August by a de¬

creased development, when the plant would be colored usually

a more or less brilliant blue-green. In September, the “spore¬

like” (04, p. 270) bodies were formed, always few- in number

in comparison with the amount of the plant in the water, which,,

in their opinion, apparently serve to carry the plant through ex¬

ceptionally severe conditions. Even in late October, however,

these authors have observed that a succession of warm days will

frequently permit of a rejuvenescence and a fairly vigorous

growth results along the edges where the water is warm. Dur¬

ing the winter, according to these investigators, the plant lies

dormant, either in the form of broken threads of various lengths

and of various stages of arrested development, or in the form of

spores. These fragments or spores rest either on the bottom in

shallow water or they appear to be held suspended in the denser

water near the bottom of the deeper portion. As the spring ad¬

vances and the surface of the water becomes warmed to a tem¬

perature of about 60° F., the authors assert that the plant be¬

gins to grow; and growth continues until the water reaches a

temperature of 72° F., when rapid breaking up of the filaments

occurs.

The alga in Jamaica Pond was at last, in September, 1903,

killed by means of an application of copper sulphate, and the

authors were able to find during the following spring only the

merest trace of its recurrence.

The scum which forms on the surface of many ponds and

lakes, following a hot summer season, has been much studied

and has been found to consist in most cases of various blue-green

Olive — Occurrence of Oscillatoria Prolifica. ±27

aigae. The phenomenon has received various names: in Eng¬

land it is called “breaking” ; in this locality, it is sometimes

called the “working” of the lakes. “Wasserbluthe,” “Flos

aquae,” “ waterbloom, ” are also variously applied to the scum.

In this country, Far low (77; 83, I; 83, II) has done more

work on the subject than any other investigator. According to

Trelease (89), Dr. Fallow .first observed the purplish color in

Jamaica Pond, in Massachusetts, in the spring of 1884, and he

gave to the alga the name Oscillatoria diffusa. This species was

subsequently found, however, to correspond to the earlier de¬

scribed 0. prolifica (Grevilie). In his earlier paper (77), Far-

Low discusses the odors caused by the deeayr of various organ¬

isms sometimes found in water supplies and speaks of Oscilla¬

toria and Lyngbya as causing “indescribably suffocating”

odors; of Beggiatoa, sulphurous odors; and of the Nostocs,

Plectonema, etc., “pig-pen” odors. In this same paper, he

ascribes the death of the algae forming the water-bloom as due

to the broiling hot rays of the sun.

Magnus (83) investigated an instance in which the ice cut

from a pond near Berlin contained a greenish growth, and, on

examination, he found it to contain a species of Aphanizomenon.

Trelease (89) studied the “Working” of the lakes at Madison

and gives, in connection with his paper, a long list of articles

relating to water-bloom. Chodat (96) has recently published

observations on Oscdlatoria rubescens D. C., which, together

with 0. prolifica (see Goraont, 93, p. 225), gives a reddish color

to the surface of Lake Moral, in Switzerland. This author ap¬

pears to agree with Klebahn (95), in that vacuoles of gas are

regarded as present in the cortical region of the cells of the

Oscillatoria; and he further believes that it is this gas which

causes the alga to rise to the surface and float.

Moore (01) also agrees with Klebahn ’s views concerning the

presence of gas-vacuoles in water-bloom, since, after his study

of Oscillatoria prolifica from J ainaica Pond, he came to the con¬

clusion that the buoyancy of the algal masses was due to the

presence of the vacuoles, which he thought contained nitrogen.

He was also lead to believe that the red color of the plant was

caused by the refraction due to the presence of large numbers

of these gas-vacuoles, as had been suggested by Klebahn.

128 Wisconsin Academy of Sciences , Arts, and Letters.

A most recent discussion of Wasserbliithe and of the green,

yellow or red colors given to bodies of water by various organ¬

isms has been written by Zaeharias (03).

Pine Lake is similar in one respect to Jamaica Pond, viz., in

that neither has any outlet to speak of. But, on the other hand,

Pine Lake is considerably larger than Jamaica Pond. The

former has an area of 1.2 sq. mi. and is about 2 miles long by 1

mile wide; whereas the latter has an area of only 65 V2 acres.

The smaller lake is between 50-60 feet in its deepest part;

while the greatest depth of Pine Lake is about 90 feet.

The growth of Oscillaioria prolifica in Pine Lake during the

summer and fall of 1904 must have surpassed in luxuriance even

the richest development of the plant in Jamaica Pond. For,

inquiries show that the ice around the shores of the whole lake

contained quantities of the alga, as evidenced by the fact that

ice harvested on all sides contained the red color imparted by

it. Mr. Hansen, as did the great majority of the residents

about the shores of the lake, threw away all of his colored ice

and replenished his supply from the neighboring Beaver Lake,

which did not show any of the reddish growth. Two of the

residents, however, retained some of their cut from Pine Lake,

and I have examined specimens from the ice-houses on the

estate of Messrs. Mayer, situated almost opposite and about one

mile from the estate of Mr. Hansen. I was told that this ice

was cut in January, 1905. In the most of the ice-cakes exam¬

ined, the reddish color, resembling the juice of crushed cherries,

appeared in small amount only, diffused about air-bubbles and

cracks in the ice. One of these colored areas was melted, and,

on microscopical examination, the water thus obtained was

found to contain the faintly reddish filaments of Oscillatoria

prolifica. These filaments appear rather rigid and refractive,

probably partly owing to the iarge amount of silica which, in

the investigations quoted above, was found to be present. The

diameter of a filament from the Pine Lake material measures

about 4/*-5y; while the component cells are likewise about 4/x-

5 n long.

I was told that some of the ice at Mr. Mayer's place was col¬

ored throughout with the ‘‘crushed cherry" color; whereas

Olive — Occurrence of Oscillatoria Prolific a, 123

other cakes showed ten inches or so of clear top, with the red¬

dish substance frozen into the ice only below the ten inches.

The great abundance of the alga is further proved by the

large masses which were left floating on the surface of the lake,

in March last, where the ice had melted. Mr. Hansen, who col¬

lected for the purposes of identification an abundance of the

alga on March 25, 1905, says that some of the floating masses

were about 12 inches in diameter, while others were small — ■ 1 the

greater part of them being about the size of an oak leaf” (from

letter of April 8th to Prof. Birge ; . Mr. Hansen and many

others mentioned the peculiar smell readily noticable at the

lake shore, which came from the decaying plant. One described

the odor as resembling that from decayed flesh; but Hyarns and

and Richards describe it rather as fetid, not putrefactive.

An old resident claimed that at intervals during the past

twenty years this red color had appeared in the ice taken from

Pine Lake. Another said that the ice at North Lake, which is

situated only a short distance north of Pine Lake, was colored

two or three years ago in this same way. Some ventured the

explanation that the fact that Pine Lake has practically no out¬

let except at high water, might explain the abundance of the

alga here, as wrell as its present confinement to this lake. For,

a visit to ice-houses at North and Okauchee Lakes failed to dis¬

cover any signs of the growth, and careful inquiries at Mouse

and Oconomowoc Lakes and Lac la Belle showed that none was

present, at least, in the ice harvested from these bodies of water

last winter. It is perhaps a significant fact that all of these

lakes mentioned have strongly flowing outlets and inlets, ex¬

cepting Pine Lake and the neighboring Beaver Lake, so that

it may well be that this lack may assist in explaining the abund¬

ance of the plant in Pine Lake during the past season.

I have had opportunity to examine the plankton of Pine Lake

collected on three days only — on Aug. 23, and Oct. 18, 1900, and

again on July 26, 1905. The first collections were taken from

waters 8 — 20 meters deep, but only a few blue-green forms were

here found, and among them no Oscillatoria.

The more recent material was taken from the surface of both

shallow and deep waters, and from various parts of the lake;

9

130 Wisconsin Academy of Sciences, Arts, and Letters.

the net was also lowered in various localities to a depth of about

25 feet and then hauled straight up. Not a particle of Oscil -

latoria prolifica was found in any instance, but in all of the

latter collections there was present in large quantities a species

of Gleotrichia, the little colonies of which could readily be seen

floating in the lake, with the naked eye, together with a small

amount of other common piankton forms.

We are thus struck at once with an important difference in

the midsummer conditions of Jamaica Pond and Pine Lake. In

the case of Jamaica Pond, other forms of plant life appear to

have been practically excluded for years by Oscillatoria prolif¬

ica; during the months of July and August, moreover, the

growth of the alga in this Pond was but somewhat decreased

from the earlier more luxuriant development. In Pine Lake,

on the other hand, we have in midsummer an abundance of an¬

other species of blue-green alga, and the seeming total disap¬

pearance of Oscillatoria prolifica. This last fact is to me inex¬

plicable, since one can hardly conceive of the killing off entirely

of the luxuriant growth of the past year by the severity of the

winter cold. I think that it is more than probable that repeated

observations will surely reveal this species of Oscillatoria again

during another season, if not later during this one. There is

thus presented by Pine Lake a most interesting problem involv¬

ing the seasonal variation and the varying predominance of dif¬

ferent plankton forms.

When we try to obtain insight into the conditions of the past

season which allowed of the production of Oscillatoria prolifica

in such phenomenal abundance, we at once note the unusually

favorable weather conditions of the latter part of last year.

According to data kindly furnished by Mr. J. L. Bartlett,

Weather Observer, while last October at Madison was about the

average in temperature, November, on the other hand, was 6°

warmer than usual. During the month of October 1904, which

had a mean temperature of 51° F., several warm spells were re¬

corded at Madison; on Oct. 1, 72°; Oct. 9, 77°; Oct. 17, 75°.

On Nov. 3, the temperature reached 68° ; on Nov. 19, 67° ; while

the mean for this month was about 40° F. But the most strik¬

ing weather conditions of this time were furnished by the long

Olive — Occurrence of Oscillaioria Prolifica. 131

drought which then prevailed; so that possibly this also assisted

the favorable conditions for algal growth. The large lakes at

Madison, Monona and Mendota, froze on Dec. 13 and 14, 1004,

respectively. I have no record concerning Pine Lake, but it is

quite probable that it was frozen over at about the same time,

notwithstanding the fact that it is considerably smaller than

the Madison lakes. The likelihood of the earlier freezing of the

smaller lake is somewhat counterbalanced, in this instance, by

the fact that the climate of the locality of the latter is to a cer¬

tain extent influenced by the proximity of Lake Michigan, since

it is only 20 miles awTay from the large lake, and over 50 miles

nearer than Madison.

Now, granting that the conditions for the growth of Oscilla-

toria were unusually favorable in Pine Lake particularly dur¬

ing last October, how can we account for the occurrence of

such vast quantities of the alga, frozen up in the ice? ITyams

and Richards speak of an occasional fall growth, a sort of reju¬

venescence due to a new warm season, which, resulted in one in¬

stance in some of the alga being found in the ice. Birge (98, p.

420) says that in the autumn there is normally “a period, be¬

ginning a little before the first of October and extending to the

freezing of the lake, when the algae are present in immense

quantities, and are distributed with approximate equality

through the whole mass of the water.” If those observers were

correct, who assert that the first ten inches of the ice from Pine

Lake was clear, and the alga appeared only in the lower strata,

then we must assume that, after considerable freezing had been

accomplished, the severe cold must somehow have killed the alga

and thus caused it to rise to the surface. I have not, however,

myself seen an instance in which one side cnly of the ice-cake

was colored, but those which I examined were instead reddish in

small areas, about cracks and air-bubbles. It has been suggested

that possibly those who made the observation recorded above

were mistaken and that it was the upper part of the ice which was

thus colored and not the lower; I have not had an opportunity,

however, of verifying this suggestion. But in the event of this

being the case, then we must suppose that the unusually mild and

long-prolonged growing season of last fall culminated in a

‘ ‘blooming time,” or “working,” of the alga. Should this prove

132 Wisconsin Academy of Sciences, Arts, and Letters.

true, then the floating scum would have been frozen directly into

the top ice. I have no means at present of determinating cer¬

tainly which of these two ideas is correct, but the weight of the

long-continued observations of the above quoted authors on the

seasonal habits of the plant inclines me to believe that the alga

did not form a scum, but was probably present, late in the sea¬

son, in vast quantities in the deeper waters of the lake and that

somehow the extreme cold of the month of J anuary caused it

to rise after the surface had become frozen.

Concerning the reddish color which appears in these plants,

particularly on their rising to the surface, or on their under¬

going decomposition, I wish to record here a suggestion at con¬

siderable variance from the theory held by Hyams and Rich¬

ards, who come to the conclusion that the reddish substance is

an important factor in the vigorous growth of plants.

These authors have themselves stated the fact that ' ‘ whenever

the plant is found below the surface it is blue-green in color ■ ’

(01, p. 310), and that the reddish pigment appeared “when

luxuriant, or whenever the growth is rapid’7 (04, p. 271).

They have further said that it is the mass of filaments near the

surface of the water, and the floating scum, and the decayed

alga on the rocks of the shore which display the reddish or vio¬

let tints.

While it may be correct, as do the two authors above cited,

to assume the probability of the great importance of these red¬

dish pigments and perhaps even their chemical combination

with the chlorophyll of these plants, it is hardly allowable, in

my opinion, to bring in, to assist in establishing their point of

the great importance of these substances, comparisons with the

other reddish coloring matters sometimes present in the cell-sap

of the young shoots and leaves of higher plants. For, I think

that it is not at all established that these reddish pigments of

the higher and lower plants are similar to each other, either

chemically or physiologically.

The appearance, in the case of Osciliatoria prolifica as well

as in other common species of Osciliatoria, of the reddish coloring

matters in the filaments after they have risen to the surface and

particularly on their undergoing evident decomposition, suggests

that such colors arise as decomposition products, rather than that

Olive — Occurrence of 0 scillatoria Prolifica. iZo

this reddish pigment is associated with a vigorous growth. The

intensification of the color, as decomposition proceeds, argues

also strongly for this conclusion.

The daily determinations, from March to October, of the

amount of carbon-dioxide dissolved in the water of Jamaica

Pond, as made by Hyains and Richards, may be regarded as fur¬

nishing a very important clew to the revealing of one cause, at

least, of the formation of a surface scum and the appearance of

the reddish color. These authors found that whenever the

Oscillatoria grew vigorously, the normal content of “ carbon di¬

oxide disappeared and the water became not only neutral but

alkaline . With the decay of the plant, the alkalinity disap¬

peared and carbon-dioxide again became normal and in one or

two instances appeared in excess.” (02, p. 310).

It seems to me readily conceivable, at any rate, that we may

have in the lack of this important food -substance, carbon-diox¬

ide, a condition perhaps brought about by its being used up by

the plant during the vigorous growth, the prime cause of the

beginnings of decomposition and the consequent rising of the

alga to the surface of the water and the appearance of the red

pigment.

Madison, Wisconsin, Aug. 3, 1905. \

BIBLIOGRAPHY.

Birge, E. A. 98. — Plankton Studies on Lake Mendota. II.

Trans. Wis. Acad. Sciences. 11 :274-448. 1898.

Chodat, R. 96. — Sur la structure et la biologie de deux algues

pelagiques. II. Oscillatoria rubescens DC. Journal de

Botanique. 10: 341—349. 405—409. 1896.

Farlow, W. G. 77. Remarks on some algae found in the water

supplies of the city of Boston. Bull, of the Bussey Inst. 2:

75—80. 1877.

- 83, I. Notes on fresh-water algae. Bot. Gaz. 8 : 224.

, 1883.

- 83, II. Relations of certain forms of algae to disagree¬

able tastes and odors. Proc. Amer. Assoc. Adv, Sci. 32:

306. 1883. Abstract.

Gomont, M. 93. Monographic des Oscillariees. Paris. 1893.

134 Wisconsin Academy of Sciences , Arts, and Letters .

Hyams, Isabel F., and Ellen II. Richards. 01. Notes on

Ociilaria prolifica (Greville). I. Life History. Tech.

Quarterly. 14. No. 4: 302 — 310. 8 text figs. 1901.

- 02. II. Chemical Composition. Ibid. 15. No. 3 :

308—315. 1902.

- 04. III. Coloring Matters. Ibid. 17. No. 3 : 270- -

276. 1 Plate, 1 text fig. 1904.

Klebahn, H. 95. Gasvacuolen, ein Bestandtheil der Zellen

der wasserbliithebildenden Phycochromaceen. Flora, 80:

241—282. Taf. IV. 1895.

Magnus, P. 83. Das AuOtreten von Aphanizomcnon fios aquae

(L.) Ralfs im Eise bei Berlin. Ber. d. D. bot. Ges. 1:

129- 132. 1883.

Moore, G. T. 01 . The cause of the red-brown color in certain

Cyanophyceae. The Soc. for Plant Morph, and Pliys.

Science. N. S. 13: 248. 1901. Abstract.

Trelease, W. 89. The “working” of the Madison lakes.

Trans. Wis. Acad. Sciences. 7: 121—129. PI. X. 1889.

Zacharias, O. 03. Geber Grim, Gelb- und Rothfarbung der

Gewasser durch die Anwesenheit mikroskopischer Organis-

men. Forsch. Ber. Biol. Stat. Ploen. Theil X: 296 — 303.

1903. Ref. in Bot, Centralbl. 96: 430. 1904.

INFECTION EXPERIMENTS WITH ERYSIPHE

GRAMINIS DC.

GEORGE M. REED.

Assistant in Botany, University of Wisconsin.

Neger is to be credited with the discovery that among the mil¬

dews, as in the rusts, our ordinary morphological species may

consist of a number of physiologically specialized forms which

are limited in their occurrence to a, single host plant or to a

group of closely related host plants. Neger ’s (8) first experi¬

ments were made during the summer and fall of 1901. His gen¬

eral conclusion was that, in the mildews he studied, specializa¬

tion has gone to such an extent that conidia from one species

will not infect a species of any other genus. In some cases the

specialization has gone still further so that conidia from one

species is incapable of infecting another species of the same

genus. His results may be summarized as follows:

1. Conidia of Erysiphe cichoracearum DC. from Artemisia

vulgaris will infect A. vulgaris but not A. Absinthium, Alchem-

illa vulgaris, Galium silvaticum, G. rotundifolia, Hicracium

murorum, Lactuca muralis, Lcontondon taraxacum, Lithosper-

mum arvense, Plantago lanceolata, Ranunculus repens, Senecio

vulgaris, nor Sonchus oleraceus.

Conidia from Lactuca muralis will infect L. muralis but not

Galium silvaticum, Hieraceum murorum., nor Pulmonaria offici¬

nalis.

Conidia from Hieracium murorum will infect IL murorum but

not Artemisia vulgaris, Galium silvaticum, Hypericum monia-

num, Lactuca muralis , Leoniodon taraxacum nor Sonchus olera¬

ceus.

Conidia from Senecio vulgaris will infect S. vulgaris but not

136 Wisconsin Academy of Sciences, Arts, and Letters.

Hieracium murorum, Lactnca muralis, Pulmonaria officinalis

nor Symphytum tuberosum.

Conidia from Plantago major will not infect the same host

P. major, Artemisia vulgaris, nor Hieracium murorum.

Conidia from Lappa major will not infect Artemisia vulgaris

nor Senecio vulgaris.

Conidia from V erbascum thapsiforme will not infect Arte¬

misia vulgaris.

Conidia from Pulmonaria officinalis will not infect Hieracium

murorum.

Conidia from Litliospermum arvense will infect L. arvense but

not Hieracium murorum, Pulmonaria officinalis, nor Symphy¬

tum tuberosum.

2. Conidia of E. polygoni DC. from Heracleum spondylium

will infect H. spondylium but not Aegopodium podagraria, An-

thriscus silvestris, nor Hypericum montanum.

Conidia from Galium silvalicum will infect G. silvaticum but

not Aegopodium podagraria, Ranunculus repens, Senecio vul¬

garis, nor Vicia sepium.

Conidia from Ranunculus repens will infect R. repens but

not Galium silvalicum.

Conidia from Trifolium incarnation will infect T. incarnatum

but not Galium silvaticum, Hypericum montanum, Trifolium

repens, nor Vicia sepium.

Conidia Prom Hypericum perforatum will infect H. perfora¬

tum but not H. montanum nor Galium silvalicum.

3. Conidia of E. galeopsidis DC. from Galeopsis tetrahit

will infect G. tetrahit but not Calamintha acinos, Glechoma

hederacea, nor St achy s rectea.

4. Conidia of Microsphaera astragali DC. (Nev.) from

Astragalus glycyphyllos will infect A. glycyphyllos and A. ciccr

but not Robinia pseudoacacia nor Vicia sepium.

5. Condia of Uncinula salicis DC. (Winter) from Salix

purpurea will infect S. purpurea and S. caprea.

6. Conida of Uncinula aceris DC. (Sacc.) from Acer pseu-

doplatanus will infect A. pseudoplatanus and A. campestre.

7. Conidia of Phyliaclinia corylea (Pers.) Karst, from Cory -

lus avellana will infect G. av ell ana.

Reed — Infection Experiments.

137

In his experiments, Neger found that the incubation period

was two to three days.

Neger raised the question whether this specialization which he

discovered in the conidia of the mildews also extends to the asco-

spores. He says he has never observed perithecia with the asco-

spores formed on Senecio vulgaris , Galium silvaticum, Cala-

mintha acinos, Symphytum tuberosum , Hieraceum murorum and

others. The question arises how these plants become infected

each spring. The conidia cannot live over the winter. Neger

found that conidia from Plantago, if kept dry, lost their germi¬

nating power in seven days and conidia from Artemisia in

twelve days. In order to explain the facts as observed by him,

he advances the theory that the ascospores have the capacity to

infect a wider range of host-plants than the conidia. To illus¬

trate this we might take the three host-plants, a, b, and c, the

conidia of each constituting a well-defined physiological species.

We may further assume that perithecia with ascospores may be

formed only on a. In the spring, however, these ascospores may

be capable of infecting all three host-plants, a, b, and c, and

thus spread the fungus.

If the ascospores have a wider range of host-plants than the

conidia of the same species, the way in which the conidia become

specialized each season has to be explained. Neger attempts to

explain this by citing the experiment of Brefeld who found that

the spores of certain smuts, if grown in a nutrient solution, lost

their ability to infect a living plant. Ward also found that

conidia of Botrytis cinera, if grown in an artificial nutrient sub¬

stance, could not infect turnips. It must be remembered, how¬

ever, that it is often a difficult thing to get spores to germinate

in a nutrient solution and the plants produced are frequently

not vigorous and so in turn do not produce a very high per cent

of viable spores.

Neger compares this assumed capacity of the ascospores to in¬

fect a wider range of host-plants than the conidia to a similar

capacity which he claims exists in the aecidiospores of the rusts.

On this point he cites the work of Eriksson (2) who found that

Avena sativa and Alopecurus pratensis were both infected with

aecidospores taken from the same aeeidum cup, the uredospores

of each constituting distinct forms. It must be stated, however.

138 Wisconsin Academy of Sciences, Arts, and Letters.

that Eriksson himself does not believe that the experiment was

carried out with sufficient accuracy to give absolutely certain re¬

sults.

Neger offers no evidence that the ascospores actually do have

a wider range of host-plants than the conidia of the same species.

Salmon (10) distinctly states that he has frequently observed

perithecia on Senecio vulgaris and Symphytum, while others

have recorded them on Calamintha and Hieraceum, all of which

are plants upon which Neger did not find perithecia. It is pos¬

sible that' with more careful observation, perithecia will be found

on all plants on which conidia are produced.

Marchal (7) , during the spring of 1902, attempted the infection

of several grasses with conidia of Erysiplie graminis. He gives

no detailed account of his work but states that conidia from

wheat would not infect rye, oats and barley; conidia from rye

would not infect wheat, oats and barley; conidia from oats

would not infect wheat, rye and barley ; and conidia from barley

would not infect the other cereals. As a result of his work he

makes seven physiological species which he names as follows:

1. E. graminis f. spec. Tritici upon T. vulgare, T. Spelta, T.

polonicum, T. turgidum , not on T. durum, T. monococcum, T.

dicoccum.

2. f. spec. Hordei upon II. hexastichon, H . vulgare, II. tri -

furcatum, II. nudum, II. jubatum, and II. murinum, not on H.

maritimum} H, secalinum, nor H. bulbosum.

3. f. spec. Secalis upon S. cereale and S. anatolicum.

4. f. spec. Avenae upon A. sativa, A. fatua, A. orientalis, and

Arrhenathcrum elatius.

5. f. spec. Poae upon P. annua, P. trivialis, P. pratensis, P.

caesia, P. mutalensis, P. memoralis, and P. serotina.

6. f. spec. Agropyri upon Agropyron.

7. f. spec. Bromi upon various Bromes.

It is impossible to tell how far Marchal ’s results are reliable

since he gives no data as to the number of his experiments or

the conditions under which they were performed.

There is, in fact, considerable evidence that there are more

than seven special forms of mildew which infect these grasses

mentioned as hosts.

Reed — Infection Experiments.

lod

Salmon (10) made several infection experiments with the mil¬

dew of the grasses, especially of the Brome grasses. He finds that,

instead of there being one special form for all the Brome grasses,

<as stated by Marchal, there are at least four and probably five

distinct biologic forms on the Bromes.

Salmon also obtained the following results with the mildew

from other plants. These results in general confirm the work

of Marchal.

1. Conidia of Erysiphe graminis DC. from wheat will infect

wheat and Triticum Spelia but not oats, barley, rye, nor Ag-

ropyron repens.

Conidia from oats will infect oats, Avena brevis, A. nuda,

A. sterilis, A. strigosa, and A. orientalis, but not wheat, barley,

rye, Festuca elatior, F. heterophylla, Poa annua, Dactylis glom-

erata, Arrhenatherium elatius, Trisetum pratense, Phleum pra-

tcnse, Alopecurus pralcnsis, nor Lolium Ilalicum.

Conidia from Avena nuda will infect oats, A. brevis, and A.

nuda.

2. Conidia of Erysiphe polygoni DC. from Trifolium pra¬

tense will infect T. pratense, but not T. agrarium, T. repens, T.

medium, T. montanum, T. incarnatum, T. hybridum, T. fill-

forme, Lotus corniculatus, Melilotus arvensis, Medicago sativa,

Lupinus luteus, nor Pisum sativum.

Conidia from Pisum sativum will infect P. sativum but not

Lupinus luteus, Colutca arborescens, Onobrychis saliva, nor

Trifolium pratense.

It is interesting to note that the mildew on the oats was able

to infect a number of other species of grasses belonging to the

same genus whereas the mildew on Trifolium pratense was un¬

able to pass over to any other species of the same genus.

Salmon (12) studied still further the infection power of the

conidia of several species of mildews. His results are as follows :

1. Conidia of Erysiphe graminis from Avena sterilis will in¬

fect A. sterilis, A. sativa, A. pratensis, but not Arrhenatherum

avenaceum, Lolium temulentum, Festuca elatior, Bromus unio-

loides, nor B. stei'ilis.

Conidia from Agropyron repens will infect A. repens, A.

tenerum, A. caninum, but not A. glaucum nor A. acutum.

140 Wisconsin Academy of Sciences, Arts, and Letters .

Conidia from Poa pratensis will infect P. pratensis, P. an¬

nua* and P. nemoralis* but not Festuca elatior var pratensis y

F. arundinacea, F. heterophylla, Lolium perenne, L. temulent-

um, Dactylis glomerata, Phleum pratense, Alopecurus pratensis r

Avena sativa, Hordeum vulgar e, Triticum vulgar e, Secale

cereale, nor Agropyron repens.

Conidia from Dactylis glomerata will infect D. glomerata but

not Avena sativa, Secale cereale, Triticum vulgare, Lolium

temulentum, Hordeum vulgare, nor Agropyron repens.

2. Conidia of Sphaerotheca humuli from Potentilla reptans

will infect P. reptans but not Alchemilla vulgaris, A. arvensis r

Fragaria sp. (cult), Spiraea TJlmaria, Agrimonia Eupatoria,

nor Potesium officinale.

3. Conidia of Sphaerotheca humuli var. fuliginea from Tar¬

axacum officinale will infect T. officinale but not Plantago

media, P. lanceolata, nor Fragaria sp. (cult.)

Conidia from Plan tag o lanceolata will infect P. lanceolata but

not P. major nor Taraxacum officinale.

4. Conidia of Erysiphe cichoracearum DC. from Plantago

major will infect P. major, P. media, but not P. lanceolata,

Galium Aparine nor Eupatorium cannabinum.

5. Conidia of Erysiphe galeopsiclis DC. from Ballota nigra

will infect B. nigra, but not Salvia verticillata nor Leonurus

Cardiaca.

Salmon (11) also made some infection experiments with asco-

spores from Hordeum vulgare during the spring of 1903. The

ascospores infected Hordeum vulgare, H. zeocriton, and H. trifur-

catum, but not Triticum vulgare, Avena saliva, Secale cereale,

H. maritimum, H. secalinum ■, H. jubatum, nor H. bulOosum.

The ascospores corresponded in their infecting powers to the

conidia from the same host except in the case of H. jubatum.

Marchal says that the conidia from H. vulgare will infect H.

jubatum. Salmon states, on the contrary, that he lias not been

able to infect H. jubatum with conidia from H. vulgare. How¬

ever, so far as the infecting power of the ascospores alone is

concerned, one experiment is not sufficient to determine.

As a result of the investigations summarized above, it may be

* “Sub-infection.”

Reed — Infection Experiments.

141

assumed as proven that there are physiological differences be¬

tween the mildews of one morphological species by which it is

split up into a number of forms, each limited to one or few host

plants. In what these differences consist is by no means yet de¬

termined. Salmon (13), during the summer of 1903, carried on a

series of experiments to determine the behavior of these special¬

ized forms under specially controlled conditions. He cut out of

the leaf to be inoculated some of the epidermal cells on one sur¬

face and most of the mesopliyll tissue. The epidermal cells on

the other surface were left intact. The spores were sown upon

these uninjured epidermal cells and the leaf was then placed

with the cut surface downwards in a petri dish. Salmon found

that when the leaves were injured in this way that they were

able to bridge the fungus over to other host piants although the

healthy leaves of these same plants are immune to the attacks

of the fungus. For example, under these conditions spores from

wheat will infect bar ley, oats and Ilordeum sylvaticum ; spores

from barley will infect wheat, oats, Ilordeum murinum and H.

sylvaticum; spores from Bromus commut'atus will infect B. race -

mosus, B. asper and Ilordeum sylvaticum ; spores from Bromus

secalinus will infect B. asper and barley; spores from Arena

stmgosa will infect barley and wheat. He further found that

spores from wheat will infect T. monococcum and T. dicoccum.

All of these are cases in which no infection would occur on unin¬

jured leaves.

Salmon also sowed the spores produced on these injured leaves

on other injured leaves of the same plants and found that they

were able to produce infection. In this way the viability of the

spores formed on the injured leaves was proven. It was further

discovered that- spores produced on injured leaves of Ilordeum

sylvaticum by inoculating with conidia from wheat were in turn

able to infect healthy leaves of H. sylvaiicum.

These investigations indicate that although there are definite

physiological species, normally restricted to one or a few host

plants, yet injured host plants of one physiological species may

be infected by spores of another physiological species and, in

this way, the mildew may be bridged over from one host plant

to another.

142 Wisconsin Academy of Sciences, Arts, and Letters.

Various terms have been applied to these physiologically dif¬

ferent varieties, as “ special forms,” “biologic forms,” “adapted

races,” “biologic species,” “ physiological species.” Some re¬

gard such specialized forms as incipient morphological species.

It is possible that the specialized form of mildew adapted to

growth upon wheat, but not on other grasses, is in course of

becoming a species of Erysiphe which may eventually show

morphological differences in addition to the physiological ones

it already manifests.

Salmon believes that the physiological species are restricted to

certain host plants because of the presence of certain enzymes

or toxins, or perhaps both, in the cells of the parasite and anti¬

toxins, or similar substances, in the cells of the host plant.

Under normal conditions certain substances peculiar to each spe¬

cies of host plant are able to prevent the attacks of any mildew

except the specialized form adapted to withstand these sub¬

stances. 'When the host plant is injured these substances are

not formed by the cells in sufficient quantities to prevent the

attack of other specialized forms of the fungus and accordingly

infection results.

It is difficult to see how the spores from an injured leaf, which

has become infected by the special form from another host plant,

are able to infect uninjured leaves of the same plant. The cells

of the healthy plant, of course, form these substances which nor¬

mally inhibit the development of this special form of fungus.

In one generation, however, the fungus becomes able to resist

these substances and produce infection. The fungus must be

extremely variable and must readily adapt itself to the sub¬

stances formed by the cells of the plant infected. It is evident

also that the differences betvceen the various physiological spe¬

cies are very slight. They also probably differ among them¬

selves as to the degree of their specialization.

Just to what extent injured plants may serve to bridge these

specialized forms over from their own host plants to others must

be determined much more fully. W7e shall probably find that

the various specialized forms differ remarkably in this respect.

The farther apart the host plants are in their characters the

Reed — Infection Experiments.

143

more difficult will it be for the specialized form on the one to

pass over to the other.

Fifty-five species of grasses have been reported as infected by

this one mildew, Erysiphe graminis DC (14). In fact all grass

mildews are included in this one abundant and cosmopolitan spe¬

cies. Of this number only sixteen species have been reported in

this country as being infected with the mildew : Agropyron glau-

eum (3) (6), A. tenerum (4), Agrostis exarala (3), Arena sp.

indet. (3), Beckmannia erucaef ormis (3), Bromus unioloides

(4), B. breviaristatus (4), Elynius condensatus (3), Glyceria

nervata (3), Hordeum jubatum (3), Poa pratensis (3) (4) (9)

*(16), P. tenuifolia (3) (6), P. nemoralis (4), P. serotina (4),

Secale cereale, Triticum vulgar e (3).

This mildew occasionally causes serious damage to forage and

cereal crops, especially to wheat. Serious outbreaks of the mil¬

dew have occurred in Europe on winter sown cereal crops. An¬

derson (1), speaking o£ the attacks of the disease in Montana,

says: “It affects chiefly the Poas and is especially damaging

to P. tenuifolia, one of the most valued forage grasses.”

During the last year, I have carried on infection experiments

using the conidia of Erysiphe graminis from the host plants Poa

pratensis and Secale cereale. In the latter part of September,

1903, I brought into the greenhouse several sods of P. pratensis

infected by the mildew. The grass grew for most of the winter

but gradually died out. In order to keep a good supply of

conidia on hand, seed of P. pratensis was sown and the young

seedlings were placed beside the sods. They became infected

and served as a source for obtaining conidia for further infec¬

tions. In this way, the fungus has been kept growing in the

greenhouse from the last of September, 1903 to February 1st,

1905.

About the last of October, some rye which was growing in the

greenhouse was discovered to be infected with mildew. This

mildew was also kept growing in the greenhouse, by the methods

noted above, from the last of October to the first of July. Dur¬

ing the winter, the rye seedlings lived about six weeks and, in

order to have a supply of conidia, it was necessary to plant seed

every four or five weeks and inoculate the young seedlings.

144 Wisconsin Academy of Sciences, Arts, and Letters.

With the mildew on these two host plants, a series of infec¬

tion experiments was carried on to test their power of passing

over to and growing upon other grains and grasses. Most of

the experiments were performed by placing one and one-half

inch flower pots, containing the seedlings used, under a bell jar.

The bell jar was left over them from twenty-four to forty-eight

hours and, in many cases, during the course of the experiment.

All the seedlings in one half of the pot were inoculated, those

in the other half serving as controls. An ordinary wooden label

bearing the name of the plant, the date when the seed was

planted, and the date of the inoculation, was placed in the pot

in such a way as to separate the inoculated seedlings from the

controls. The cereals were grown in the pots used in the ex¬

periments. The other grasses were grown in larger pots and a

few seedlings transferred to the small ones for the experiments.

Four or five pots, containing usually as many different

grasses, were placed under the same bell jar. In every experi¬

ment, one pot contained uninfected seedlings of the host plant

used as a source of conidia for inoculation. In this way it was

always possible to tell whether good, viable spores had been used

in inoculating the seedlings for, if the conidia infected seedlings

of their own host plant, their viability was demonstrated.

For most of the experiments in which rye seedlings were in¬

oculated, the Sehlamslead variety of rye was used. In a few

experiments another variety, the Petkus, was taken. The mil¬

dew appeared to infect the two varieties with equal readiness.

The young seedlings of the cereals appeared in from four to

six days after sowing the seed. The oat seed was longer in ger¬

minating than the others. The grasses came up in from ten

days to two weeks after the seed was sown. The age of the

seedlings as stated in the tables is reckoned from the time of

sowing the seed. In most cases, the first leaf of the seedlings

of the grains was inoculated, although in older seedlings other

leaves were also inoculated.

In some cases, another method of experimentation, suggested

by Salmon (10) was employed. Green, vigorous leaves of grasses

or grains were placed on a moist filter paper in a Petri dish.

Some of the leaves were inoculated, the rest being kept as con-

Reed — Infection Experiments.

145

trols. Under these conditions the leaves retained their vitality

long enough to determine whether infection would take place.

The inoculations were made by removing with a scalpel a few

spores from an infected leaf and applying them to the leaf to be

inoculated. As the spores, under ordinary conditions, are light

and dry, it was difficult to get them to stick to the leaf to be in¬

oculated. This difficulty was overcome by placing infected

leaves on a moist filter paper in a Petri dish and leaving them a

few hours. In this way a very abundant supply of moist con¬

i' dia was secured which readily adhered to a leaf when applied.

Care had to be taken not to leave the spores too long before

using them as they readily germinate under such conditions.

The time elapsing between inoculation and the first appearance

of infection was fairly 'constant. Patches of mycelium ap¬

peared in from three to four days if infection occurred at all.

By the fifth day conidia were very abundant. The first obser¬

vation recorded was made when the conidia had become abund¬

ant, not at the first appearance of the mycelium.

Before taking up the detailed account of my experiments it

will be interesting to note some general differences in the be¬

haviour of the mildew on the rye and blue grass. On the young

rjTe seedlings the mildew spread very rapidly. When the

leaves of seedlings placed beside well infected plants showed

signs of infection, it was only two or three days until all of the

seedlings were nearly covered with patches of mycelium bearing

abundant conidia. In the case of Poa pratensis the conidia

were not produced nearly so abundantly, nor did the fungus

spread so rapidly. In one case a tuft of P. pratensis which was

growing about two feet from another that was infected re¬

mained free from the mildew for almost three weeks. Then at

first only a very few patches of mycelium appeared. It was

fully two weeks longer before the mildew had spread over the

entire tuft of the grass although the sod only covered an area

about three inches in diameter.

The mycelium and conidia also differed markedly in color on

the two host plants. The growths on the blue grass were pure

"white while those on the rye were quite pinkish in color.

When infected leaves were left in a Petri dish they gradually

10

146 Wisconsin Academy of Sciences x, Arts , and Letters.

died, turning yellow. The cells in the infected areas of the

rye leaf retained their green color for a longer time than the

others. The fungus seemed to act upon the infected cells in such

a way as to prolong their life. On the other hand the cells in

the infected region of the leaves of the blue grass died first, the

leaves turning yellow there, while the uninfected parts often

remained green for several days afterward.

My first experiments were aimed to show the possibility of

readily infecting leaves and seedlings of blue grass with conidia

from the same host. The first five or six experiments failed.

This was doubtless due to the fact that the spores used were

either weak or dead. They were taken from the plants which

had just been brought into the greenhouse and it was evident

that conidia were not produced at all abundantly until after the

plants had been m the greenhouse about three weeks. The

mildew seemed to be brought to a standstill by the change to the

greenhouse, not growing vigorously for some time. Later an

abundance of conidia wras produced on the infected grass and I

then had no more trouble in transferring the mildew. After

these first failures, forty-one additional experiments were per¬

formed in which the blue grass was inoculated with conidia

from the same host and only three were unsuccessful.

A series of observations was next made to determine to what

extent the mildew on the rye and blue grass had become special¬

ized. Spores from both of these host plants were sown on leaves

and seedlings of various grasses which are reported as hosts of

Erysiphe graminis.

I have used for the most part grasses that are commonly cul¬

tivated and must hence have been repeatedly exposed in nature

to infection from both rye and blue grass. Considerable diffi¬

culty was experienced in obtaining seed of many species of

grasses. Seed of Glyceria fluitans was obtained from Currie

Bros., Milwaukee, Wis. ; seed of Bromus mollis from Vaughn’s

Seed Store, Chicago, Ill.; seed of Lolium perenne, Festuca

elatior, Daciylis glomerata and Poa nemoralis from J. M. Thor-

bum & Co., New York; seed of Poa compressa , Phleum pratense

and Festuca heterophylla from Peter Henderson & Co., New

York; seed of Poa trivialis from the Bureau of Plant Industry,

Reed — Infection Experiments. 117

TJ. S. Department of Agriculture. The seed of the cereals was

obtained from Prof. R. A. Moore, Agronomist of the Wisconsin

Agricultural experiment station.

The seed obtained from these various sources, in some cases,

was not pure. Great care had to be taken to have only seedlings

of the desired species of grass for the experiments. The seed¬

lings of most of the grasses used have certain distinctive charac¬

ters by which they can readily be distinguished. The greatest

difficulty was with the various species of Poas as the young seed¬

lings are quite similar in general appearance.

To test the results reported by Marchal spores from rye were

sown upon wheat, oats, barley, E. jubatum, Bromus mollis , Poa

pratensis, P. irivialis, P. nemoralis and P. compressa. The re¬

sults of these experiments are given in the f olowing table :

Table I. — Conidia from Secale cereale.

The sign — indicates no infection ; — ? indicates that in one experiment there was, ap¬

parently, infection, an account of which is given below. The bell jar was left over the

seedlings either for 24 hours or 48 hours, or, in many cases, during the whole course of

the experiment. Three to five different kinds of grasses were placed under the same

bell jar, one of them always being rye; consequently in the table the record of the

observations on the rye is repeated in several cases. The actual number of experiments

with rye recorded in the table is fifty-two. Only two were unsuccessful.

These were all made by sowing spores on young seedlings and

placing them in a moist atmosphere under a bell jar, except four

of the experiments with wheat and two of those with blue grass

which were made by Salmon’s Petri dish method described

148 Wisconsin Academy of Sciences , Arts , and Letters.

above. Seedlings were left uninoculated in each experiment to

serve as controls.

With one exception none of the seedlings or leaves inoculated

became infected, although the experiments were continued from

eight to fifteen days. The controls also remained free from the

mildew. In one experiment, leaves of blue grass in a Petri dish

were inoculated, and when examined after five days were found

badly infected with the mildew. The pot containing the blue

grass from which the leaves were taken for the experiment was

then examined and it was found that the grass had, in some

way, become infected with the mildew. Consequently there was

no evidence in this experiment that the rye mildew could infect

the blue grass. In the other eleven experiments in which blue

grass was inoculated with spores from rye no infection occurred.

On the other hand the rye leaves and seedlings inoculated

with conidia from rye in connection with the above experiments

in every case except two had infected areas after about five

days. During the same interval of time, none of the control &

were infected. The seedlings were examined from day to day,

the controls especially being watched closely. Patches of myce¬

lium bearing conidia appeared on several of them eight or nine

days after the experiment was started. Evidently some of the

conidia that were first produced on the inoculated seedlings had

fallen upon the controls and then germinated, producing infec¬

tion. This infection occurred only after a period of time equal

to that which had elapsed between inoculation and the first ap¬

pearance of conidia on the inoculated leaves. The mildew was

also much more widely spread over the inoculated seedlings

when the controls became infected.

A further interesting experiment was made to show that the

spores from rye would not infect the other cereals. On March

25th, seeds of the four cereals, wheat, rye, oats and barley, were

sown in a large pot, about fifteen inches in diameter, one kind

of seed in each quadrant of the surface. Shortly after the

young seedlings came up they were all inoculated with conidia

from rye. The seedlings were left growing until about the

middle of July when they had matured, producing normal ker¬

nels. The rye seedlings became infected about four days after

Reed — Infection Experiments.

149

inoculation. The mildew was present on the rye during the

whole time, conidia being produced abundantly. On the other

hand, not one of the seedlings of wheat, oats or barley became

infected, although conidia from the rye must have fallen re¬

peatedly upon them.

Spores from blue grass were also sown upon several grasses

that had been experimented with by Marehal and with similar

results. Seedlings of rye, wheat, oats, barley, II. jubatum, and

B. mollis were used. The following table presents the results:

Table II.— Conidia from Boa pratensis.

The sign — indicates no infection ; — indicates that in one experiment there was, ap¬

parently, infection, an account of which is given below. The belljar was left over the

seedlings for 24 hours or 48 hours, or in many cases, during the whole course of the ex¬

periment. Three to five different kinds of grasses were placed under the same belljar,

one of them always being blue grass ; consequently in the table the record of the obser¬

vations on the blue grass is in several cases repeated. The actual number of experi¬

ments with blue grass recorded in the table is twenty-three. Only two were unsuccess¬

ful.

All of these experiments were made by the bell jar method

except four of those with rye which were made by placing rye

leaves in a Petri dish. Control seedlings were kept in each ex¬

periment. No infection occurred except in one instance. In

this case four rye leaves were placed in a Petri dish and two of

them inoculated with spores from the blue grass. In the same

dish two other rye leaves were inoculated with conidia from rye

and two were kept as controls. When examined a week after

inoculation all of the eight leaves had patches of mycelium with

conidia. This infection was doubtless due to allowing some of

the rye spores used for inoculating two of the leaves to fall on

the other six. In none of the other eleven experiments in which

rye was inoculated with conidia from blue grass did infection

occur.

150 Wisconsin Academy of Sciences , Arts , and Letters.

In connection with these same experiments seedlings of blue

grass were inoculated with spores from the blue grass. Alto¬

gether twenty-three experiments were made, several of the seed¬

lings in each case not being inoculated. Infection occurred on

the inoculated seedlings in all of the experiments except two.

None of the controls became infected.

My experiments with conidia from rye and blue grass thus

confirm Marchal’s results for the forms considered.

Marchal also states that the same special form of mildew oc¬

curs on all of a number of species of Poa. I have made a num¬

ber of inoculations with the mildewT from P. pratensis upon P.

nemoralis, P .trivialis and P. compressa and secured some in¬

teresting results which do not entirely confirm Marchal’s state¬

ment. My results are given in detail in the following table and

additional notes are given below.

Table III. — Conidia from Poa pratensis.

Reed — Infection Experiments . 151

Table III. — Conidia from Poa pratensis— Continued.

The sign 4~f indicates good infection ; + fair infection ; — no infection. || indicates

that several seedlings, usually eight to ten, were either inoculated or kept as controls.

In the table the record of the observations on the blue grass is repeated. The actual

number of experiments with the blue grass recorded in the table is twenty-seven. Two

were unsuccessful. The blue grass seedlings were from 43 to 96 days old.,

152 Wisconsin Academy of Sciences , Arts , and Letters.

Poa nemoralis.

The Wood meadow grass is now classed among our good

grasses for shaded pastures. It is particularly valuable for

shaded lawns.

The seedlings are very similar to those of blue grass. They

can generally be distinguished by their wider leaves and more

vivid green color.

Spores from rye produced no infection on this grass. The

experiments with conidia from blue grass were doubtful in their

results and quite interesting. As seen from the table, nineteen

experiments were made in which spores from blue grass were

sown upon the seedlings of P. nemoralis. The inoculated seed¬

lings in eleven of these experiments became infected with mil¬

dew; in the other cases there were no signs of infection. None

of the uninoculated seedlings in any experiment became in¬

fected. In six of the experiments in which infection occurred

there was a vigorous growth of the mildew, fully as good a

growth as was present upon the seedlings of blue grass which

were under the same bell jar. In the remaining five experi¬

ments in which infection occurred, there w^ere only a few small

infected areas, while the seedlings of blue grass in the same

experiments had numerous small infected areas producing

conidia very abundantly.

The seed was not pure as some of the grass was allowed to

grow until it flowered and it wTas found that P. pratensis was

also present. It is possible that some of the seedlings infected

were those of P. pratensis ; still I am sure that most of the seed¬

lings that were infected were seedlings of wood meadow^ grass.

Some of the infected seedlings of the wood meadow grass

were kept with a view of determining wThether the mildew would

pass from them to other grasses, especially blue grass. The

mildew, however, never developed on the wood meadow grass

with sufficient vigor to make it possible to get sufficient spores for

such experiments.

Reed — Infection Experiments.

153

Poa trivialis.

The Rough-stalked meadow grass, like Wood meadow grass,

is well adapted for shaded places. It is a spreading, thickly

matting, stoloniferous species. The young seedlings are nar¬

rower leaved than those of the other Poas considered and lighter

green in color. They resemble the Kentucky blue grass, how¬

ever, quite closely.

Eight experiments were made in wTiich spores from blue

grass were sown upon the seedlings of P. trivialis. A slight in¬

fection occurred in only two experiments. In each of the ex¬

periments 103 a and 107 a one patch of mycelium with conidia

was present on an inoculated seedling. These patches had dis¬

appeared when the seedlings were examined a second time a

few days later. In these two experiments the seedlings of blue

grass inoculated with spores from blue grass and placed under

the same bell jars with Poa trivialis, became only slightly in¬

fected.

I

Poa compressa.

Canada blue grass is not reported so far as 1 can find as a

host of E. graminis. As seen from the table eight experiments

were made in which seedlings of this grass were inoculated with

spores from Kentucky blue grass. It is a very hardy grass,

growing on poor and dry soils. It has creeping rootstalks,

forming a close and durable turf. It is distinguished from the

Kentucky Blue Grass by its flattened wiry stems and by its

decidedly bluer color.

None of the seedlings inoculated with spores from the blue

grass became infected except in two experiments. In one of

these, 107 b, a small patch of mycelium bearing conidia was

present on one inoculated leaf. In experiment 108 b there was

no visible infection when first examined seven days after inocu¬

lation. When examined a week later two small infected areas

were present on one inoculated seedling. In the first of these

two experiments, the seedlings of blue grass inoculated with

spores from the same host and placed under the bell jar with P.

154 Wisconsin Academy of Sciences , Arts , and Letters.

compressa bore only a few patches of mycelium; conidia were

not produced abundantly. In the other experiment, however,

there was a very good infection of blue grass.

As seen from these experiments the mildew on P. pratensis

will not readily infect the other species of Poa experimented

with. Still under certain conditions it seems able to pass over

to some extent to P. nemoralis, and, in a lesser degree, to P.

Irivialis and P. compressa. It is possible that there is a “sub-

infection”, as Salmon calls it, in these cases. The question of

the purity of the seed and the identification of the seedlings is

very important in connection with experiments on the various

species of Poa.

My observations indicate that the group is an interesting one

and further experiments may be expected to throw light on the

real nature of physiological species and their method of origin.

The facts thus far brought out lead us to expect that among

the mildews on the remaining grasses which have been reported

as hosts many will be found which have become more or less

physiologically specialized and limited in their power of pro¬

ducing infection. I have farther carried out a series of experi¬

ments with spores of rye and blue grass mildews to determine, if

possible, whether these forms can pass over to any other grasses

the seed of which I could obtain. All of these grasses are re¬

ported as hosts of Erysiphe graminis DC. The results of these

experiments are summarized in the following tables and further

notes on each grass are given below.

Reed — Infection Experiments. 155

Table YV. — Conidia from Secale cereale.

156 Wisconsin Academy of Sciences , Arts , and Letters.

Table IV. — Conidia from Secale cercale— Continued.

In the table the record of the observations on the rye is repeated. The actual num¬

ber of experiments with rye recorded in the table is twenty-six. All were successful.

The rye seedlings were from 6 to 19 days old.

The sign ++ indicates good infection, + fair infection, — no infection.

|| indicates that several seedlings, usually eight to ten, were either inoculated or kept

as controls.

Reed — Infection Experiments.

157

Table V. — Conidia from Poa pratensis.

158 Wisconsin Academy of Sciences , Arts , amd Letters.

Table V. — Conidia from Poa pratensis— Continued.

In the table the record of the observations on the blue grass is repeated. The actual

number of experiments with the blue grass recorded is twenty five. Only one was un¬

successful. The blue grass seedlings were from 43 to 96 days old.

Lolium perenne.

English Rye grass is a valuable pasture grass and is grown

abundantly associated with both the cereal grains and blue

grass. Like the latter il forms a perennial green sward. It is,

however, not closely related to either rye or blue grass and all

attempts to infect it with conidia from these failed. Mildew

has not been reported upon the English rye grass in this coun¬

try. The seed germinates quickly, the young seedlings coming

out of the soil in about ten days. It is favorable for experi¬

mentation for the seedlings grow rapidly. They are bright

green and very slender. The seed used seemed to be entirely

pure.

Reed — Infection Experiments.

159

Festuca elatior and F. hclerophylla.

Both of these grasses are natives of Europe and are grown as

hay and pasture grasses. The tall meadow Fescue is a coarse

growing grass. The young seedlings appeared above the soil in

about twelve days after the seed was sown. They grew vigor¬

ously, becoming rather broad leaved and coarse.

The seedlings of the various leaved Fescue were a little

darker green in color than those of F. elatior. They were also

a little longer in appearing above the soil, about two weeks

elapsing after the seed was sown before they appeared.

Spores from rye were sown on both of these grasses while

spores from blue grass were sown on F. elatior only. No in¬

fection occurred.

Dactylis giornerata.

Orchard grass is grown quite extensively for forage purposes.

It has the habit of growing in tufts. It is easy to work with as

the seedlings grow vigorously and rapidly. They appear in

two weeks after sowing. The mildew has not been reported

upon the grass in this country. Spores from neither rye nor

blue grass were able to infect it.

Phleum pratense.

Timothy is a native of Europe, but was naturalized in Amer¬

ica many years ago. It ranks in this country as by far the

most important of hay grasses and it is certainly commonly

associated with the grains. The seed germinates quickly, the

young seedlings appearing in about nine days. The seedlings

grow rapidly and are very hardy. No infection occurred from

inoculation with spores from either rye or blue grass.

Glycena fluHans.

Floating meadow grass is found in wet soils and marshes and

its damp surroundings should favor the mildew. The young

seedlings appear above the soil in ten days. They grow very

rapidly becoming quite tall and slender. Only spores from rye

were tried on the seedlings of this grass. No infection resulted.

160 Wisconsin Academy of Sciences , Arts , and Letters.

In connection with these experiments the seedlings of rye

that were inoculated with conidia from rye uniformly became

infected. The same was also true in the case of the experi¬

ments in which blue grass was inoculated with conidia from

blue grass, no infection occurring in only one case. The con¬

trols in every case were free from mildew when the inoculated

seedlings became infected except in experiments 87 and 89

where some of the controls near the inoculated leaves had small

infected areas on them. In the case of the rye many of the con¬

trols in several experiments when examined eight to ten days

after the inoculations were made had numerous small infected

areas on them. This was doubtless due to some of the spores

that were formed on the infected leaves falling upon the con¬

trols. The inoculated seedlings also had several additional in¬

fected areas on them when the infection of the controls was ob¬

served.

These experiments confirm the general conclusion that spores

of the mildew from one grass will not infect a grass belonging

to a different genus. Marclial, it. is true, states that the form

on oats is the same as that on Arrhenatherum elatius. Salmon,

however, was unable to verify this statement.

It is entirely possible that for E. graminis at least, there is

one, if not more, distinct physiological species for each genus of

grasses that contains species which are hosts for this mildew.

In many genera of grasses the specialization has undoubtedly

gone still further so that there may be a number of physiolog¬

ical forms upon the various species of the same genus. This is

Salmon’s conclusion in the case of the Brome grasses and my

work with the various Poas leads me to the conclusion that there

is more than one at least partially differentiated physiological

form for this genus also.

Further investigations will probably show that some physio¬

logical forms are much more fixed in their characters than oth¬

ers, just as is the case with morphological species. Such special

forms will doubtless be found to grade over into each other and

we may thus get interesting evidence as to their developmental

history.

R eed — Ii ifection Experiments.

161

This work has been done under the direction of Prof. It. A.

Harper and I am greatly indebted to him for his kindly criti¬

cisms and valuable suggestions.

Madison , Wisconsin ,

March, 1905.

BIBLIOGRAPHY.

1. Anderson, F. W. Brief Notes on a few common Fungi of

of Montana. Journ. of Mycology, 5:31. 1889.

2. Eriksson. Ueber die Specialisirung die Parasitismus bei

den Getreiderostpilzen. Ber. d. d. bot. Ges. 12. 189-1.

3. Farlow, W. G., and Seymour, A. B. Provisional Host-

index of the Fungi of the United States. 1890.

4. Griffiths, D. Some Northwestern Erysiphaceae. Bull.

Torr. Club, 26: 138-144. 1899.

5. Hitchcock, A. S. Partial List of Iowa Powdery Mildews.

Bull. la. Agric. Coll, for 1886, 64-66.

6. Kelsey, F. D. Study of Montana Erysiplieae. Bot. Gaz.

14: 285-288. 1889.

7. Marchal, E. De la specialisation du parasitisme chez

I’Erysiphe graminis. Comptes Rendus, 135: 210.

July 21, 1902.

8. Neger, F. W. Beitrage Zur Biologie der Erysipheen.

Flora, 90: 221-272. January, 1902.

9. Rose, J. N. Mildews of Indiana. Bot. Gaz. 11 : 60-63.

1886.

10. Salmon, E. S. On Specialization of Parasitism in the Ery¬

siphaceae. Beihefte zum Bot. Centralblatt. 14 :261.

1903.

11. - . Infection Powers of Ascospores in Erysiphaceae.

Journ. of Bot. 41 :159. 1903.

12. - . On Specialization of Parasitism in the Erysipha¬

ceae. New Phytologist 3 :109. May, 1904.

13. - . Cultural Experiments with ‘ * Biologic Forms”

of the Erysiphaceae. Phil. Trans, of the Royal Soci¬

ety of London. Series B. 197 : 107-122. 1904.

ii

162 Wisconsin Academy of Sciences } Arts , and Letters.

14. Salmon, E. S. A Monograph of the Erysiphaceae. Me¬

moirs of Torr. Bot. Club. Vol. 9. 1900.

15. Tracey, S. M., and Earle, F. S. Mississippi Fungi. Buli.

Miss. Agric. Exper. Sta. 34: 95-97. 1895.

16. Trelease, W. Preliminary List of Wisconsin Parasitic

Fungi. Trans. Wis. Acad. Sei. Arts and Letters, 6:

111-311. 1885.

THE STATE ADMINISTRATION OF TAXATION IN

WISCONSIN.

JAMBS D. BARNETT,

Assistant in Political Science , University of Wisconsin.

r-

I. The State Budget.

II. The Various Systems of Taxation.

1. The General Property Tax.

2. The Corporation Tax.

3. The Inheritance Tax.

4. The Suit Tax.

5. Miscellaneous License Taxes.

‘ The purpose of this paper is to describe the changes in the

form and in the powers of the central administration of tax¬

ation in Wisconsin since the organization of the territory, but

it is only indirectly concerned with the actual workings of the

system, and not at ail with the economic phases of the subject.

I. The State Budget.

It is only a little that the administration has to do with the

budget.

During the territorial period the congressional appropria¬

tions for “the contingent expenses” of the territory were made

upon estimates submitted by the secretary of the treasury

of the United States.1 In 1836 the president of the state

senate and the speaker of the house were direcetd to prepare

estimates as the basis of the secretary’s action,2 but these

1 0. L., s. 11.

2H. J. 1836, pp. 127, 141.

164 Wisconsin Academy of Sciences Arts , and Letters.

preliminary estimates were made by the governor apparently

during most of the period.1 No officer was charged with

preparing estimates for expenses to be paid out of the terri¬

torial treasury.

Since 1848 the secretary of state has been required to make

detailed estimates of the expenses to be defrayed from the state

treasury as a basis for the action of the legislature.2 The

secretary has based his estimate almost wholly upon the in¬

come received by the various departments during the preced¬

ing year, though recently the tax commission has recom¬

mended that all departments file preliminary estimates with

him.3

II. The Various Systems of Taxation.

Till recent years the general property tax has been the most

important source of the state revenue, and legislation has been

directed for the most part to this kind of taxation. But with

the growth of the corporation tax, the general property tax

has practically disappeared as a source of state revenue, and

henceforth the central administration will be interested in this

tax only in the control, which it has recently obtained, of the

administration of taxation by the local authorities. The chief

interest in state administration of taxes is now divided between

the problems of this central control and the problems of the

taxation of corporations. The minor sources of taxation have

some interest on account of the peculiarities of their adminis¬

tration.

1. The General Property Tax.

A. The Levy of State Taxes.

Of course the general property tax has been levied, in the

usual sense of that word, only by the legislature, till 1845 a

percentage on the gross amount of taxes assessed by each county,

and since that time a territorial or state tax levied in each

county with the county tax.4

iH. J. 1840-1, app., p. 85; C. J. 1842-3, app., pp. 49-53.

2L. June 1848, p. 115, s. 10 (2) : R. S. 1898, s. 144 (13) ; L. 1901, c. 368.

s Especially L. 1901, c. 97; Tax Comm. Rpt., 1903, pp. 243-4.

4 Const. Art. 8, s. 5; L. 1837-8, No. 93; E. g., L. 1845, p. 1, s. 2.

Barnett — Taxation in Wisconsin.

165

But since 1869, whenever, before the apportionment of the

state tax to the counties, it is evident that the appropriations ex¬

ceed the amount of state tax levied to meet the expenses of the

year for which the tax was levied, the secretary of state is

required to “levy” and apportion such an additional amount

as may be necessary to meet all authorized demands on the

state treasury up to the time when the next succeeding tax

will be due.1 For thirty years the validity of this measure

seems not to have been questioned, but in 1899 both the gov¬

ernor and the secretary of state were of the opinion that under

the constitution2 the legislature itself is the only tax-levy¬

ing authority of the state and that this power cannot be del¬

egated; and the secretary therefore refused to make the addi¬

tional levy called for by the statute under the circumstances.3

A somewhat analogous power was vested in a board consist¬

ing of the governor, secretary of state, and state treasurer by

a law of 1887 which directs that whenever in the opinion of

the board the public interest requires it, they may apply the

surplus in the treasury, or so much of it as they deem proper,

to reduce the state levy each year.4

B. The Assessment of Taxes.

Until 1852 the counties assessed the state taxes as they did

their own, without any further control by the state adminisra-

tion than a requirement of the filing of a duplicate of the county

tax and, a little later, also local valuation statistics with the

territorial or state authorities.5

The inequality of the burden of taxation under the old sys¬

tem had long been a subject of complaint,6 when the state

assumed a greater control in 1852 by establishing a state board

of equalization.

iL. 1869, c. 153, s. 1: R. S. 1898, s. 1071.

2 Const. Art. 8, s. 5.

sWis. (Weekly) State Journal, June 16, 1899; Wis. State Journal,

Sept. 28, Oct. 3, 12, 1899.

4L. 1887, c. 397: R. S. 1898, s. 1069a.

5L. 1837-8, No. 93, s. 2.

6 Especially A. J. 1852, app., pp. 4-5.

166 Wisconsin Academy of Sciences , Arts , and Letters.

a. The Development of the State Tax Commission. — The first

board of equalization consisted of the governor, secretary of

state, state treasurer, attorney general, and state superintend¬

ent,1 the lieutenant governor and bank comptroller being

added two years later.2 Beginning with 1858, for the next

fifteen years the board was composed of the state senate and

secretary of state.8 This change was induced partly because

of the dissatisfaction with the data available for the use of the

board in the returns from the counties, and partly by the ‘ 1 anti¬

republican” nature of the old board. The results of the change

seem generally to have been bad. Bings were formed in the

senate, and much logrolling took place to the great detriment of

some sections.4 Hence in 1873 a board composed of stale

officers was again established, the state board of assessment,3

consisting of the secretary of state, state treasurer, and attorney

general.6 This board was replaced by the present commissioners

of taxation in 1901.

The commission, the direct outgrowth of the tax commission

of 1897~8 (an investigating body), was established in 1899, “in

order to secure improved taxation within the state.”7 It

consists of the commissioner, first assistant commissioner, and

second assistant commissioner, all appointed by the governor

with the consent of the senate, all known to the governor “to

possess knowledge of the subject of taxation and skill in matters

pertaining thereto,” and serving ten years from 1899. Neither

the commissioner nor the assistants nor any clerk in the office

is permitted to “hold any other office or position of trust or

iL. 1852, c. 498, s. 1.

2L. 1854, c. 73, s. 1

3L. 1858, c. 115, s. 26.

4S. J. I860, pp. 817-8; Wis. State Journal, Mar. 12, 1860; April 7, 23,

May 5, 1868; Secy. State Rept. 1868, pp. 38-9; S. Proc. in Wis. State

Journal, Feb. 4, 1874; Wis. (Weekly) State Journal, Dec. 10, 1878.

5 The former board had been known as the Board of Assessors since

1870. (L. 1870, c. 144, s. 1.)

6L. 1873, c. 235.

7L. 1897, c. 340; Tax Comm. Rpt., 1897-8, p. 182; Wis. State Jour¬

nal, April 18, 1899.

Barnett — Taxation in Wisconsin.

167

profit, or possess any other business or avocation, or serve on

or under any committee of any political party.”1

In 1899 the commissioner was made a member of the state

board of assessment, presided at its meetings, and assisted the

board with his information.2 Finally in 1901 the old board

was abolished, and the commission of taxation became the state

board of assessment.3

While the board consisted of the senate and secretary of state

and met during the recess of the legislature, members received a

per diem and mileage the same as the members of the legisla¬

ture,4 but the members of the former boards (all ex-officio)

were paid no additional compensation for their- services. The

salary of the present commissioner is $5,000, and that of the as¬

sistants, $4,000 each.5 In 1899 the necessary traveling ex¬

penses of the commission were allowed, and the commissioner

was authorized to fix the number and compensation of any

clerks in the office; but upon the governor’s protest against plac¬

ing so much discretion in the commissioner’s hands, the maxi¬

mum amount for all the disbursements of the office was later

fixed by law.6 However when the commissioners took over

the taxation of railroads thej7 were given unlimited power in the

appointment of the necessary additional assistants etc., for

the purpose.7

b. The Assessment of Slate Taxes ; the State Supervision of

Local Taxation.- — By the law of 1852 the board was to meet an¬

nually to equalize the valuations made by the counties, “to pro¬

duce a just relation between the valuation of the taxable prop¬

erty in the state.”8 The action of the various boards has

always been annual with the exception of the years between

1859 and 1879, when it was biennial.9 Since 1870 the

iL. 1899, c. 206, ss. 1, 7.

2L. 1899, c. 206, s. 6.

3L. 1901, c. 237, ss. 1, 6; State Tax Comm. Rpt., 1900, p. 171; Gov.

M. 1901, pp. 9-12.

4L. 1859, c. 167, s. 28.

sL. 1899, c. 206, s. 7; c. 322.

eL. 1899, c. 206, ss. 7, 9; L. 1901, c. 220, s. 2; Gov. M. 1901, p. 8.

7L. 1903, c. 315, ,s. 27.

8L. 1852, c. 498, s. 2.

9L. 1859, c. 167, s. 29; L. 1879, c. 124.

168 Wisconsin Academy of Sciences , Arts, and Letters.

action has been recognized as assessment rather than equaliza¬

tion.1

The matter of getting correct returns from the localities has

been the subject of much legislation. Before any central equali¬

zation was attempted, beginning with 1841 reports from the

counties to the treasurer, auditor, or secretary were required,

showing the local valuation of property.2 After the crea¬

tion of the first board began a further series of laws to secure

proper returns of local valuation to the secretary of state as a

basis of state equalization,3 one of them authorizing the

secretary of state to send a special messenger for the required

statistics in case of the neglect of the county ^authorities.4

But the returns have never been satisfactory.

The board did not even attempt to make an equalization before

1854, and at that time the secretary of state declared the false

valuations received made any action on their basis “mere guess

work.”5 It wms claimed that the board of 1878 was the

first body which had before it a complete set of returns from

every county, and that theirs was “the first endeavor honestly

to live up to the law and equalize in fact as well as in name.”6

As early as 1861, in a complaint of the inequality of taxation

on account of the false returns of property, the secretary of

state declared it to be doubtful if a return of all property could

be secured unless through the appointment of assessors by the

governor or legislature, who by residence and tenure of office

would be removed from local influence.7 The state has not

gone to this extremity, but the powers of the present tax com¬

mission would seem to exhaust all remedies up to this point.

In 1899 the commission was given “general supervision of the

system of taxation throughout the state, ’ ’ but was really limited

to making investigations and reporting the results to the leg-

iL. 1870, c. 144, s. 1.

2L. 1840-41, No. 8, ,s. 6; L. 1843-4, p. 6, s. 6; R. S. 1849, c. 15, s. 41.

3E. g., L. 1854, c. 73, ,ss. 4, 6; L. 1881, c. 236, s. 4: R. S. 1898, ss. 1004

sq.; L. 1903, c. 315, s. 12.

4L. 1874, c. 43, s. 2: R. S. 1898, s. 1016.

5S. J. 1854, p. 510; Secy. State Rpt., 1854, pp. 43-4.

6Wis. (Weekly) State Journal, Dec. 10, 1878.

7 Secy. State Rpt., 1861, p. .222.

Barnett — Taxation in Wisconsin. 169

islature with recommendations. In making such investigations

the commission was empowered to require individuals and cor¬

porations to give information, to examine their records, to sum¬

mon witnesses, etc., and to direct the attorney general to proceed

against persons refusing their demands.1

In 1901. the powers of the commission were largely increased.

In addition to the assessment of state taxes in each county

made by the former boards, the commissioner has the following

powers :

1. To 4 £ exercise general supervision of the system of taxation

throughout the state;”

2. To exercise general supervision over the assessors, local

boards of review, and the assessment of property in the localities

by the county supervisors, “ so that equality of taxation shall

be secured according to law ; 5 ’

3. To advise and direct assessors, boards of review, and

county supervisors as to their duties under the statutes;

4. To direct that proceedings be instituted to enforce the

laws relating to the liabilities of officers, corporations, and in¬

dividuals for failure to comply with the tax laws; to cause com¬

plaint to be made to the proper circuit judge for their removal

from office for official misconduct or neglect of duty — in all these

cases to require the district attorney to assist in prosecution ;

5. To require local officers to report information as to the

assessment of property, collection of taxes, expenditure of pub¬

lic funds for all purposes, and any other information the com¬

mission may request;

6. To require individuals, corporations, etc., to furnish in¬

formation concerning their capital, “and all other information

called for;”

7. To summon witnesses to appear and testify in any mat¬

ter deemed material to the investigation of the system of taxa¬

tion and the expenditures of public funds for state and local

purposes. Both the commissioner and his assistants are author¬

ised to administer oaths to such witnesces. Refusals to testify

are reported to the attorney general who is required to proceed

against the offenders;

iL. 1899, c. 206, ss. 3, 4.

170 Wisconsin Academy of Sciences, Arts, and Letters.

8. To visit the counties for the investigation of the methods

of the local authorities in the administration of taxation, and to

examine into all cases where evasion of proper taxation is

charged, access to all documents of the state and localities being

allowed for such purposes; to ascertain wherever the existing

laws are defective or improperly administrated, and to investi¬

gate the tax systems of other states and countries;

9. To formulate and recommend such legislation as may be

found necessary to prevent the evasion of just and equal taxa¬

tion and for the improvement of the system of taxation;

10. To consult with the governor upon the subject of taxa¬

tion, and furnish him with such assistance and information as

he may require;

11. To transmit to the governor, before each regular session

of the legislature, a report showing the taxable property of the

state, with recommendations and such measures as may be for¬

mulated for the consideration of the legislature; and to hand

copies of this report to the members of the legislature.1

With the disappearance of the general property tax for state

purposes, these powders of the commission are reduced wholly to

the control of the local administration by the central adminis¬

tration.

c. The Apportionment of Taxes. — Under the system which

existed till 1845, the treasurer of the territory simply demanded

the amount of taxes due the territory according to the reports

from the counties.2 Under the present system the auditor

or the secretary of state has apportioned and certified the state

tax to the counties.3

d. The Collection of Taxes. — The secretary of state superin¬

tends the collection of all taxes as of all other moneys due the

state.4 The taxes have always been payable to the terri-

iL. 1901, c. 220; c. 237; A. J. 1901, pp. 24-7.

2L. 1837-8, No. 93, s. 3.

3L. 1845, p. 1, s. 6; R. S. 1849, c. 15, s. 42: R. S. 1898, s. 1070.

4L June 1848, p. 115, s. 10 (6): R. S. 1898, s. 144 (9). From 1859

to 1878 the secretary also, with the advice of the attorney general, was

directed to decide all questions as to the construction of the tax laws,

subject to an appeal to the supreme court. (L. 1859, c. 167, s. 50; R. S.

1878, s. 4978.)

Barnett — Taxation in Wisconsin.

171

torial or state treasurer by the county treasurer.1 Since

1849 the latter has been required to make with his payments to

the state treasurer, a statement of all state taxes as well as other

state moneys paid to him during the preceding year,2 and

since 1858, when he does not pay the full tax, to file with the

state treasurer as affidavit to the effect that he lias paid the whole

amount received by him.3

Until 1858 the counties were very delinquent in paying their

quotas of the state tax.4 From the beginning penalties

were enacted against the county treasurer for any neglect to

turn over the state taxes,5 but the delinquency of the smaller

localities in their payments to the county treasurer made these

penalties of no avail.6

For some years previous to 1858 it had been the practice of

the state treasurer to retain the school moneys apportioned to

the delinquent counties to balance their indebtedness to the state,

but that year the treasurer was satisfied that such a procedure

was not authorized by law,7 as was later decided by the su¬

preme court in the case of swamp land funds retained for the same

purpose.8 In 1858 a penalty was enacted against delin¬

quent counties, and the practice above mentioned was legalized,

no county being* permitted to draw any money from the state

treaasury as long as indebted to the state for taxes.9 The

operation of this law was later declared to have been 4 ‘most

happy,”10 but it was repealed the next year after its enact¬

ment, and even the penalties collected were returned,11 on

the ground that by reason of the delinquency of some of the

iL. 1837-8, No. 93, s. 3; L. 1845, p. 1, s. 4; R. S. 1849, c. 15, s. 85;

R. S. 1858, c. 18; L. ,1859, c. 14: R. S. 1898, s. 1121; Gov. M. 1859, p. 15.

2R. S. 1849, c. 10, s. Ill; R. S. 1898, s. 715 (5).

3L. 1858, c. 152, s. 3: R. S. 1898, s. 1122.

4E. g., H. J. 1838-9, pp. 317-8; C. J. 1839-40, app., pp. 249-50; S.

J., June 1848, app., pp. 22-4; Weekly Wis. Patriot, Nov. 27, 1858.

5L. 1837-8, No. 93, s. 4; R. S. 1898, s. 1123.

6 A. J. 1858, p. 1300; S. Proc. in Weekly Wis. Patriot, Feb. 26, 1859.

7 A. J. 1858, p. 1300.

8 State v. Hastings, 11 Wis. 448 (1860).

9L. 1858, c. 152, ss. 1, 2.

10 A. J. 1862, p. 641.

11 L. 1859, c. 29, s. 1; L. 1859, c. 67.

172 Wisconsin Academy of Sciences , Arts , and Letters.

towns of a county, the burden was thrown upon those which

had already paid.1

But the “old difficulty” returned,,2 and again in 1872 a

penalty was provided against delinquent counties, with the re¬

tention of all moneys due the county from the state except school

moneys.3 The “old difficulty” disappeared.4

2. The Corporation Tax.

The state taxation of corporations began in 1848 with tele¬

graph companies,5 and has since been extended to a large

number of corporations. The four general methods of the ad¬

ministration of the tax are seen in the “license” system, the

taxation of railroads and certain other carriers, the taxation of

street railways and electric light and power companies, and the

taxation of steam vessels.

A. The License Tax.6

This was the first form of the corporation tax used, and it is

still used in the case of many corporations. The method em¬

ployed is practically that of self-assessment. The tax is gener¬

ally estimated by the state treasurer upon the basis of reports of

the required data made by the corporations to him, and paid

directly to him. At present all insurance companies are li¬

censed by the commissioner of insurance, the company report¬

ing the required data to him, and sometimes also paying the

tax through the commissioner. The tax is enforced by money

forfeitures, forfeiture of license, or sale of the corporation's

property.7

iS. Proc. in Weekly Wis. Patriot, Feb. 26, 1859.

2E. g., A. J. 1862, p. 641; Secy. State Rpt., 1866, p. 37.

3L. 1872, c. 158: R. S. 1898, s. 1124.

4 A. J. 1873, app., p. 7.

5L. Feb. 1848, p. 257, s. 3: R. S. 1898, s. 1216.

6The validity of this “license” system has been upheld as constitu¬

tional (M. & M. R. R. Co. v. Board of Supervisors, 9 Wis. 410 (1855);

Kneeland v. City of Waukesha, 15 Wis. 454 (1862); Fire Dept, of Mil¬

waukee v. Helfenstein et al., 16 Wis. 136. But see Atty. Gen. v. W. L.

& F. R. P. R. Co., 11 Wis. 35).

7 A general example is L. 1891, c. 422: R. S. 1898, ss. 1222g-j.

Barnett — Taxation in 'Wisconsin.

173

B. The Taxation of Railroads, etc.

The license system of taxing railroads was established in 1854

and continued till 1903, when the present system was established.

At first the companies simply paid the tax to the treasurer esti¬

mated on the basis of reports made to him.1 Although

since 1874 the railroad commissioner had been required, to as¬

certain and report to the treasurer detailed information, which

might have been used as a check upon the reports made by the

railroads, the treasurer was not required to consider this infor¬

mation before issuing the license, the companies continuing to

be “their own assessors and own collectors,”2 After 1899

the approval of the report by the commissioner was required be¬

fore the license was issued.3 In case of failure to report

as required, the treasurer was to make the assessment without

the report and to sell the road for the tax, or the attorney gen¬

eral instituted proceedings for the forfeiture of the franchise.4

In 1903 the taxation of railroads was assimulated to the gen¬

eral property tax, and was turned over to the tax commission.5

Thirty years before the secretary of state had urged that the

roads should be taxed by the state board of assessment/

In performing this duty the commission is given access to all

records in the departments of the state and localities; is author¬

ized to require local officers to return information, to compel the

attendance of witnesses, and to administer oaths. All records

of the railroads are subject to examination by the commission.

Annual reports are required to be made to the commission by

all railroads, but if the report is not made the commission is di¬

rected to “'inform itself the best way it may” on the matters

necessary for valuation. The commission determines the aver-

iL. 1854, c. 74; L. 1860, c. 174.

2L. 1874, c. 273, s. 12: R. S. 1898, s. 1795; R. R. Comrnr. Rpt., 1883-4,

p. 13. In 1856, in case of the Wis. and Superior R. R. Co., the gover¬

nor was empowered, in order to ascertain the truth of the statement

of the earnings, to examine the books and papers of the company, and

to examine under oath the officers, etc. (L. 1856, c. 137, s. 23.)

sL. 1899, c. 308, s. 4.

4L. 1854, c. 74, s. 5; L. 1860, c. 174, s. 3; L. 1861, c. 68.

5L. 1903, c. 315.

e Secy. State Rpt., 1873, pp. 28-30; Cf. State Tax Comm. Rpt., 1903,

pp. 182-4.

174 Wisconsin Academy of Sciences , Arts , and Letters .

age rate of taxation on all the property in the state upon the

basis of the returns in the office of the secretary of state, and

applies this rate to the railroads.1

This system of taxation had already been applied in 1899 to

certain other carriers, formerly paying a license tax. Under the

earlier system, upon the payment to the state treasurer of an

amount computed by the railroad commissioner on the basis of

reports made to the latter, the treasurer issued the license 2

When the new system was adopted in 1899, the taxation was

accidentally put into the hands of the old state board of assess¬

ment instead of the tax commission, which took charge in 1903.

The procedure is practically the same as in the case of railroads,

though the commission has not such large powers for this pur¬

pose.3

C. The Taxation of Street Railways, etc.

Since 1895 street railways, and since 1897, also electric light

and power companies, pay a license tax to the municipality. A

proportion of the tax is paid by the municipality to the county

treasurer as a state and county tax, and the county treasurer re¬

mits the state’s share to the state treasurer.4 Till 1899 the

state had no check whatever on the localities with respect to

these taxes, but since that time the assessors have reported the

names, etc., of railways in their districts to the railroad commis¬

sioner, and the railway companies have also made a report to

him.5

D. The Taxation of Steam Vessels.

Just the reverse of the above method is pursued in the taxa¬

tion of steam vessels of a certain class by the system adopted

in 1901. The tax is paid to the state treasurer on the basis of

a statement made to the secretary of state, and the state treas¬

urer pays to the county treasurer the county’s share of the tax.®

iL. 1903, c. 315.

2E. g., L. 1883, c. 353; L. 1899, c. 112, s. 7.

3E. g., L. 1899, c. 112; State Tax Comm. Rpt., 1903, p. 9.

*L. 1895, c. 363; L. 1897, c. 223; L. 1903, c 197.

5L . 1899, c. 308; L. 1899, c. 329; L. 1901, c. 417.

6L. 1901, c. 192.

Barnett — Taxation in Wisconsin .

175

3. The Inheritance Tax.1

The central authorities have a strong control of the counties

in securing the payment of the state’s share of the inheritance

tax established in 1903, and to some extent even in the collec¬

tion of the tax in the first instance.

The tax is paid to the county treasurer,2 who reports to

the secretary of state the amount of tax received, and pays over

the state’s share to the state treasurer. The receipt given by

the county treasurer on the payment of the tax must be coun¬

tersigned by the secretary of state to be valid in the final ac¬

counting of the estate, and holders of securities belonging to the

decedent are prohibited from delivering them to the executors

without prior notice to the secretary. The county judge reports

to the secretary the name of every decedent whose estate is liable

for such a tax and upon whose estate an application has been

made for letters of administration, and also the valuation of the

legacy, etc., and the secretary may apply to the county court for

an appraisal of the estate. Composition of expectant estates

may be effected under certain circumstances by the county

treasurer, but only with the consent of the secretary of state and

attorney general. As the last instance of central control, the

commissioner of insurance, upon the application of the county

court, determines the value of future and contingent estates.3

4. The Suit Tax.

The tax on suits in the circuit court was created by the con¬

stitution in 1848. 4 Its most noteworthy feature is the dif -

ficulty with which it has been collected.

At first the tax was paid directly by the clerk of the circuit

court to the state treasurer,5 but a law of the next year re¬

quired the clerk to report to the secretary of state the amount of

1 The inheritance tax law of 1899, L. 1899, c. 355, was declared un¬

constitutional in Black v. State, 113 Wis. 205 (1902).

2 In certain cases it may be paid either to the county treasurer or to

the secretary of state.

3L. 1903, c. 44.

4 Const. Art. 7, s. 18.

5L. June 1848, p. 19, s. 17.

176 Wisconsin Academy of Sciences , Arts , and Letters.

the tax received by him, and to pay the same to the judge of

the circuit court, filing the latter’s receipt with the secretary,

who deducted the amount from the judge’s next quarter’s sal¬

ary.1 A change was again made by the law of 1855, which

directs the clerk to pay the tax to the county treasurer, who

remits it to the state treasurer, and to report the amount to the

secretary of state. The secretary is to notify the judge of any

failure of the clerk to report, and the latter is liable to removal

by the judge for such neglect.2

The law has never been well obeyed,3 but the state author¬

ities do not seem to have interested themselves in the matter at

all for a great many years.

5. Miscellaneous License Taxes.

Beginning with 1852 “ hawkers and peddlers” have been re¬

quired to pay license fees to the state treasurer directly, or

through the treasury agent, receiving a license from the secre¬

tary of state. The law has been extended from time to time to

include other such transients, and various special provisions

have exempted certain classes of persons from such payments.4

Though the validity of the tax had previously been upheld

by the supreme court, in 1904 it was declared unconstitutional

on grounds other than those advanced in the earlier cases.5

In 1866 the secretary of state reported that the law requiring

the payment of these fees was not generally obeyed and that the

state was overrun by these non-resident dealers.6 Accord¬

ingly the office of treasury agent was established in 1867 to en¬

force the law. The agent is appointed by the governor and

holds office during the pleasure of the governor, and his bond

of $5,000 is subject to the governor’s approval. As compensa-

iR. S. 1849, c. 10, s. 61.

2L. 1855, c. 56; R. S. 1898, sec. 743-4.

3E. g., A. J. 1851, app., p. 826; S. J. 1861, pp. 362-3; State Treas. Rpt.,

1868, p. 11.

4L 1852, c. 386, and many amendments: R. S. 1898, ss. 1570 sq; L.

1901, c. 341; Lt. 1903, c. 393.

sMorrill v. State, 38 Wis. 428 (1875); Van Buren v. Downing, 41

Wis. 122 (1876); State V. Whitcom, 99 N. W. 468 (1904).

e Secy. State Rpt., 1866, pp. 38-9.

Barnett — Taxation in Wisconsin.

177

tion he receives the penalty assessed for neglect to pay the li¬

cense fees and a certain percentage of his collections. The

agent may appoint an assistant agent and assign his duties.1

For some years the agent approved his assistant’s bond but later

it was made subject to the governor’s approval.2 Either

the agent or the assistant may appoint special agents to aid them

in some of their duties.3

The treasury agent is directed to superintend and enforce, if

necessary, the collection of the license fees, and both he and the

assistant and special agents have large powers for this pur¬

pose.4 Since 1878 the secretary of state has been expressly

authorized to direct the agent in enforcing the license laws.5

Recently the attorney general has been required to advise the

agent as to the discharge of his duties, and whenever the agent

deems it necessary, the attorney general must assist in actions

brought for the collection of forfeitures,6 The report of the

treasury agent, required since 1889, is made to the governor.7

Of course since the decision of the court in 1904 against the

validity of the license, the agent’s usefulness has disappeared.

A few other license taxes must be mentioned. Sellers of bank¬

rupt' stocks, etc., have been required, since 1891, to pay fees to

the state treasurer, who issue licenses to them,8 and since

1901 private employment agencies have been licensed in the

same way by the secretary of state.9 For a few years, upon

the payment of the required fees to the secretary of state, the

secretary issued to non-residents hunting licenses, countersigned

by the state fish and game warden, but since 1901 the license has

been issued by the state fish and game warden, countersigned by

the secretary of state.10

Madison, Wisconsin, Feb. 1, 1905.

iL. 1867, c. 176: R. S. 1898, ss. 1578-1582.

2L. 1872, c. 177, s. 3; R. S. 1878, s. 1579: R. S. 1898, s. 1579.

3L. 1870, c. 72, s. 15: R. S. 1898, s. 1580.

4L. 1867, c. 176, s. 1; L. 1870, c. 72, s. 15: R. S. 1898, ss. 1579-80.

5R. S. 1878, s. 1579: R. S. 1898, s. 1579.

6R. S. 1898, s. 161 (5).

7L. 1889, c. 172: R. S. 1898, s. 1579.

sL. 1891, c. 443, ss. 1-5: R. S. 1898, ss. 1584d-g.

QL. 1901, c. 420, s. 10; L. 1903, c. 434, s. 9.

io L. 1897, c. 221, s. 2; L. 1901, c. 358, s. 3.

12

THE PARTS OF SPEECH IN THE CHILD’S LINGUISTIC

DEVELOPMENT.1

M. V. O’SHEA,

Professor of Education, University of Wisconsin.

1. SENTENCE WORDS.

A number of students of infant linguistics, as Holden,2

Humphreys,3 Tracy,4 et al, have endeavored to determine the

relative frequency of the several parts of speech in the child’s

language during successive periods in his early linguistic de¬

velopment. They have made lists of all the words spoken by

a number of children between the ages of fifteen months and

three years approximately, classifying them according to the

standard grammatical categories. Following this method of

1 It should be explained that this essay comprises a short chapter ab¬

stracted from a larger study on the Psychology of Linguistic Develop¬

ment, which it is hoped may some time later be brought to the light.

It will be presented in two parts: 1. The Non-reflective Processes

in Linguistic Development; II. The Reflective Process in Lin¬

guistic Development. The present essay is Chapter III of Part I of this

study; and the chapters which precede and follow it should, if space

permitted, be given in order that it might have the proper setting. As

it is, however, I must be content with saying simply that the first two

chapters treat of Pre-Linguistic Activity, and the Earliest Reactions

upon Conventional Language. The chapters which follow the present

one in Part I treat of the Development of 'Inflection in Early Language ;

the Order of Words in Sentential Construction; and the Development

of Verbal Signification. Part II treats of the psychology of the various

phases of linguistic instruction in the mother tongue and in foreign

language; and it discusses current theory and practice at home and

abroad in respect of language training.

2 On the Vocabularies of Children Under Two Years of Age; Trans.

Am. Phil. Assn. 1877, pp. 58 et seq.

3 A contribution to Infantile Linguistics; Trans. Am. Phil. Assn. 1880,

pp. 5 et seq.

4 Psychology of Childhood, Chap. V.

O'Shea — The Child's Linguistic Development. 179

treatment Tracy lias calculated that of five thousand four hun¬

dred words employed by twelve children from nineteen to thirty

months of age, and reported by several investigators, 60 per cent

are nouns, 20 per cent are verbs, 9. per cent are adjectives, 5 per

cent are adverbs, 2 per cent are prepositions, 1.7 per cent are

interjections, and 0.3 per cent are conjunctions.

Nov/, it will be apparent upon a little reflection that this

method of treating the child's vocabulary is quite external and

artificial. The classification is based upon what may be called

the structure of words viewed ah extra , rather than upon their

function in the child's expression. Tracy, and all who use his

method, take a formal or logical or grammatical, not a pyscho-

logical point of view. To illustrate the principle in question,

when K, at eleven months, says ha (hat), she always sees the

object and thrusts her arms toward it, indicating plainly enough

that she wishes to reach it. The word, if her mutilated copy

can be dignified by such a term, is uttered in an impulsive or

perhaps inter jectional way; and all her expressions show that

she has active desires with reference to the thing designated.

She is not simply naming it in any formal, logical, or purely

denotative manner. Looked at from this standpoint the word

is seen to be more than a mere noun ; it does duty for an entire

sentence in a highly generalized form.1 It is the : * undifferen¬

tiated linguistic protoplasm” out of which in due course various

sentential organs and members will make their appearance, ac¬

cording to some such general method of differentiation, possibly,

as a complex animal organism like the chick, for instance, grows

out of the undifferentiated germ cell contained in the egg. So

far as I can make out, K employs her word ha (and I speak of

this as typical of all the words she uses at eleven months), to

convey the notion, “I want that hat;” or “Take me to the

i Compare, among others. Sully: Studies of Childhood, p. 171; Lu-

kens: Preliminary Report on the Learning of Language, Ped. Sem.,

Vol. Ill, p. 453-455; Dewey: The Psychology of Infant Language, Psy¬

chol. Rev., Vol. I, pp. 63-66; Egger: Observations et re-flexions sur le

development de Vintelligence et du language chez. les enfants, Paris

1877; H. Ament: Die Entwiclcelung von Sprechen und DenTcen beim

Einde (Leipzig, 1899), p. 163; Meumann: Die Enstehung der ersten

Wortbedeut ungen beim Einde (Leipzig, 1902), p. 31.

180 Wisconsin Academy of Sciences , Arts , and Letters.

hat ;” or “I want to put that hat on. ? ’ I may add that I think

her attitude is not expressed by “See that hat” merely, for

she is exceedingly dynamic with reference to it. She is not con¬

tent to look at it simply or to induce me to look at it ; she must

do something with it, and her modes of expression are calculated

to affect me so that I will aid her in attaining her ends. It

seems to me, again, she does not have the attitude indicated by

“May I have the hat?” or “I wish I could have the hat,” for

she does not yet recognize clearly any power or authority to

which she must thus appeal in gaining her desires. She is not

pleading; she is demanding or commanding. Her attitude is

rightly expressed, I think, by the sentence, “I want that hat,

and you take me over there to it.” But the special point I

wish to impress is that her word ha denotes more than a mere

substantitive relation with the object; it denotes, in a general

way of course, all that can be indicated, though in a more par¬

ticular and definite manner, perhaps, by the grammatical ele¬

ments which in adult analy tic speech we designate as noun, verb,

pronoun, adjective, and preposition.

Sometimes the adult reverts to the infantile method of lingu¬

istic expression, and makes single words do for whole sentences.

For instance, he says simply * ‘ hat ? ” to the waiter in the restau¬

rant, at the same time looking up at the object which hangs

where he can not get it, and intoning in a characteristic manner.

This single word, used in this special situation, and supple¬

mented by gesture and characteristic vocal modulation per¬

forms the offices of an entire sentence. The psychology of the

process is clear enough; the waiter has learned from previous

experience that such a tone of voice and such a pose

always denotes a need, and the one word localizes the need,

so to speak. The notion expressed in conventional language by

“I want my” may be and in this instance is indicated plainly by

characteristic motor attitudes; indeed these attitudes could in

this case express the entire thought without the use of any word.

If the situations we encountered in life were never more complex

than in this instance, it seems hardly necessary to say that man

would not have invented parts of speech. Primitive races, as

O'Shea — The Child's Linguistic Development. 181

Romanes,1 2 3 Whitney/ Sayce,® Muller,4 Powell,5 Brinton,6 Bosan-

quet,7 and other students of primitive languages have pointed

out, get along with single-word sentences. It seems to be well

established that linguistic evolution on the phylogenetic side has

proceeded by continual differentiation of the primitive sen¬

tence this differentiation resulting in our parts of speech

and in their varied infected forms. So the infant’s expres¬

sion, on the verbal side, is a highly undifferentiated one; and

the process of development consists, for one thing, in con¬

tinuous differentiation with specialization of function, — just such

a process in principle as we see illustrated in the evolution of

language in the race.8 This method of development — continual

differentiation with specialization of function — has universal

validity in mental ontogeny, holding as well for linguistic as for

other activities.

It is apparent why, classifying the child’s vocabulary ah extra,

we find that three-fifths of his words are nouns, the names of

things, as Mrs. Moore,9 Mrs. Hall,10 Kirkpatrick11 and others

maintain. It is easy to overlook the pronominal, verbal, adjec¬

tival, adverbial, prepositional, and conjunctional function of the

first words ; as I have intimated, we unconsciously infer this func¬

tion from the child’s attitudes, gestures, facial expressions, into-

i Mental Evolution in Man, p. 294.

2 See The Encyclopaedia Brittanica, 9th edition, Vol. XVIII, pp.

766-722, article on philology.

3 Ibid., Vol. XI, pp. 37-43, article on Grammar.

4 See his Science of Thought.

6 See, among others of his writings, his essay on the Evolution of

Language; Trans, of the Anthropological Soc. of Washington, 1880, pp.

35-54.

6 Essays of an Americanist, pp. 403, et seq.

7 Essentials of Logic, pp. 82-86.

sLeFevre (See his Race and Language, p. 42) has attempted to show

that in phylogenesis all the grammatical categories have developed from

the primitive cry. The cry of animals, even, (contains the roots of

human speech. There is the cry of need which gives rise in time to

our interjection, and later to the elements of the sentence. The warn¬

ing or summoning cry in turn gives rise to our demonstrative roots, and

is the origin of the names of numbers, sex, and distance.

9 The Mental Development of a Child; Psych. Rev. 1896 (Mono. Supp.

No. 3).

io First 500 days of a child’s life; Child Study Mo., Vol. II, p. 607

(March, 1897).

41 Fundamentals of Child Study, p. 236.

182 Wisconsin Academy of Sciences , Arts , and Letters.

nation, and so on, and we disregard the part the interpreter plays

in reacting upon infant speech. But viewed from the stand¬

point of the child’s use of his words in his adjustments, it is evi¬

dent that they are never at the outset merely nominal in func¬

tion.1 Mrs. Hall thinks objects are' at first apprehended as

wholes, without regard to their qualities or their action, but this

seems extremely doubtful, to say the least. It appears rather

that the qualities of an object, as food, for instance, will be up¬

permost in the child’s consciousence in his reactions upon it; and

in naming it at any time he will really, so far as his own mental

content is concerned, be designating these qualities of the thing

and not the thing in itself, whatever this may be. To illustrate,

S. at twelve months liked buttered zwieback, and whenever he

saw any on the table he would call out bock , bock , though he did

not care for the plain variety. Surely his reaction — or in other

words — his expression, must have been incited by and had refer¬

ence to the peculiar gustatory quality of this special article. In¬

deed, the child’s mental states must usually if not always be con¬

cerned primarily with the sensory effects of objects, which would

occasion a predominant adjectival attitude toward them. In the

course of development one’s experiences of an adjectival charac¬

ter with any object will slowly become generalized into what we

mean by the term object ; and then when we refer to it we have

in mind first this generalized something which we may simply

designate, and then go on to specify certain particular experi¬

ences we have had or should like to have with it.2

But the young child’s attitude must always be special and

qualitative, not general and nominative. And at the outset the

actional is really but a phase of this general qualitative attitude.

When II. sees the kitten running after the ball, or her father tak¬

ing gymnastic exercise, or any thing else in movement, she indi¬

cates plainly that it is the actional characteristic of the thing

i Compare with this statement Dewey’s view, Psychol, of Infant Lang.,

Psych. Rev., I, pp. 63-66.

1 1 do not mean that we can form a notion of a thing apart from any

of its qualities, states, or actions, but nevertheless with repeated ex¬

perience with an object we seem to gain a kind of sense of its existence

independent of any particular quality, state, or action. Doubtless this

sense is 'for the most part verbal merely.

O'Shea — The Child's Linguistic Development. 183

which attracts her. She does not of course abstract the action

from the object and regard it as a thing apart ; but she is affected

differently by the object when it is at rest from what she is when

it is in motion. The conception of action as such arises only very

gradually as a generalization upon a body of experiences, where¬

in particular objects are seen to be capable of a variety of

actions. This results in establishing the feeling that there is a

something constituting an object which is not displayed in its

particular activities. In some such way object, action, quality

are differentiated; and our analytic language aids in the differ¬

entiation, and tends to make it permanent. The effort to employ

differentiated speech imitatively is a great stimulus to the defini¬

tion of elements in one’s original undifferentiated ideas.

2. NOMINAL AND VERBAL FUNCTION IN EARLY SPEECH.

It is probable, as I have intimated, that the child’s early in¬

terests center entirely in things as qualitative and dynamic ; and,

confining our attention here to the development of nominal and

verbal function, we see that it is only upon a multiplicity of ex¬

periences that the child can conceive of any object as distinct

from its various dynamic conditions. So that in the young

child’s consciousness noun and verb, viewing the matter func¬

tionally, cannot exist independently; the use of substantive

terms, speaking grammatically, always implies predicative char¬

acteristics. When the child makes his own terms they always

denote objects acting; just as do individual terms in primitive

languages. Only in our anayltic adult language, which has

been developed to express intricate and highly differentiated in¬

tellectual content — only in this Language are substantive and

predicate function more or less completely differentiated. Now,

when the child copies the forms of this highly differentiated lan¬

guage, we may argue that he must have back of them the same

differentiated thought as the adult has, but in this assumption

we are quite likely to go wide of the mark.

In illustration of this point, take a case like the following. I

give K. the term 4 ‘ runs ’ ’ for her brother cutting across the lawr

I repeat it on several occasions, and I find that soon she will

184 Wisconsin Academy of Sciences , Arts ■, and Letters.

point to the brother running and exclaim, unsl What is the

mental content back of such an expression? Manifestly her at¬

tention is engaged with this object in certain continually chang¬

ing attitudes ; she can not be concerned with the action as an in¬

dependent thing. However, as her experiences with her brothei

running and other objects running increase she will gradually

generalize this activity as a thing or quality characteristic of

these objects. But these same objects present themselves from

time to time under other and different conditions, each of which

will in due course be generalized as special characteristics; and

on the linguistic side, if she would express any particular charac¬

teristic of the objects, she finds that she must have some means

of designating them, without reference to any special attitude

or quality, and then she must have some means of designating

the special characteristic in question. If these objects always

appeared in the same role she would not need to have one term

for substantive and another for predicate in describing her ex¬

perience with them.

Some one has said that, viewed ab intra, the child’s nearest ap¬

proach to the use of a noun pure and simple is found in those

expressions, which from one point of view may be regarded as

exclamations or even interjections. To illustrate, S. hears a bark¬

ing dog at a distance, and he exclaims, bu ! bu ! (dog). lie makes

no effort to get the object, or to get away from it. His eyes, his

intonations, his bodily attitudes all show surprise and wonder,

however, but with no tendency to definite action. Now% in this

expression is he simply naming an object — either the dog, or

the barking as an independent auditory thing ? The strict nom¬

inal attitude, it will be agreed, is a purely intellectual one; but

in this case the child experiences lively emotion, though for the

moment it does not issue in adaptive action. Reaction is held

in check for the time being ; but nevertheless the individual is in

a dynamic attitude toward the object. He is on the qui vive to

detect what should be done in the premises. If one should at¬

tempt to express his attitude in a sentence, it would probably be

something like — “I wonder what that noise means?” or 4 ‘That’s

the dog ; what ’s he going to do ? ”, or “I can tell that ’s the dog

making that racket ; will he be likely to do me any harm ? ’ ’

O'Shea — The Child's Linguistic Development. 185

It should be added that as development proceeds the indi¬

vidual acquires a more and more impersonal relation toward

many objects and so in his speech he may reach the point where

he can simply designate them or name them ; that is, he can em¬

ploy the substantive in its grammatical function strictly. Again,

a child early finds pleasure in the ability to recognize objects, as

Groos has pointed out, and he always wishes to have others share

his achievements with him, so he may and probably does often

employ his words for the purpose of winning the applause of the

alter, and not for the purpose of imparting an idea by naming

an object, or predicating anything about it. That is, he uses his

terms in a simple denotative way, without attempting to express

his experience with the objects denoted.

Before the completion of the second year usually, and in some

cases as early as the eighteenth month, the child begins to ex¬

press himself in eliptical sentences, as, giving two of H’s ex¬

pressions, “ doggie-high ” (the dog is jumping over a high fence) ;

“Nann-come” (I want Anna to come and help me). Now, are the

differentiation of thought which we are making the basis of dif¬

ferentiation in speech? Viewed from without they appear to

be ; but in reality they are probably often mere mechanical imi¬

tations, with no subjective differentiation to correspond to the

external differentiated form. I have often said “doggie-high”

to TL, and she may be and probably at the outset is just copying

my words. In her own mental processes there may be but little

more differentiation with respect to this particular situation than

when she employed the single wrord “doggie” in reaction there¬

upon. Children from a year and a half on, constantly illustrate

this principle in their speech for a number of months. They

learn, as a matter of mechanical imitation, an expression like

“birdies fly” and they use it not only when they see a bird fly¬

ing, but also when the see it sitting on a limb or picking

worms from the ground. That is to say, the term fly may not

denote a clear and definite particularization in the child’s

thought though he uses it freely enough. It will not carry true

verbal function until he employs it for the purpose of indicating

a particular aspect or attitude of birds and other objects, and

186 Wisconsin Academy of Sciences , Arts, and Letters.

which he can and does distinguish from all other attitudes or

aspects.

Then again, such an expression as 4 ‘birdie fly” in the child’s

speech may be regarded as a single term describing a bird in

certain attitudes. The child is not aware that he is using the sub¬

stantive and a predicate; he imitates as a unity1 an expression

for a general situation, which expression in adult speech denotes

differentiation into object and action. The only way we can tell

for sure when substantive and predicate have become differen¬

tiated in the child’s speech is when he uses them appropriately

in situations where he could not have imitated them just as he

employs them; as when, dropping some bits of paper over the

hot-air register he sees them sail upward and exclaims, “paper

fly.” Here action is apprehended apart from the special thing

with which it was originally connected, and a beginning is made

in regarding it as a characteristic that may be possessed by many

different things. In due course “flying” or “to fly” will de¬

note a certain kind of thing conceived of as having existence

more or less apart from objects as such. The same may be said

of “running,” “jumping,” “shouting” and so on ad libitum.

It should be added that only very slowly and as a result of a

great variety of significant experiences does the conception of

action as such become differentiated from the conception of

things as such; and the differentiation always comes about

through a series of objects performing in the same general sort

of way as, for instance, boys, dogs, horses, etc., running. This.

‘ ‘ general sort of way ” is of course always very close in conscious¬

ness to objects; but yet, and especially when a particular verbal

symbol is used to designate it, it may be felt as having a certain

degree of individual existence. Doubtless, though, persons dif¬

fer greatly in respect of the extent to which action and quality

become dissevered, as it were, from things, and attain to a meas¬

ure of independence in the mental processes. It may be added

that there can never be complete severance of relationships and

iPreyer observed his son Axel at twenty-seven months saying mage-

nicht ( mag es nicht ) and tannenicht (kann es nicht). Any observer

may notice the same phenomenon and often quite late in linguistic de¬

velopment, after the child has been in school for several years.

O'Shea — The Child's Linguistic Development. 187

dependency in things of the mental commonwealth any more

than there can be in things of the social commonwealth.

The prominence which some grammarians ascribe to the verb

in linguistic expression may warrant its receiving a little special

attention at this point. Before the twenty-fourth month, as a

rule, the child uses sentences of two or three words, but the verb

is quite often omitted, and from my observations I should say

that the novice can get along very handily without it. To illus¬

trate, M. at twenty-live months will say, “Mamma — milk,”

(Mamma, I want my milk; or Mamma, have my milk brought

in). Taking my glasses in her hand she will say, “Baby — nose”

(I want to put them on baby's [my] nose). Watchng her nurse

prepare her bath, she will repeat many times, “Baby — bath.”

One may count instances of this kind literally by the hundred

every day in the life of an active child from his second to his

third or fourth birthday. The copula is „quite generally omitted

in the beginning. A three-year old will say “My — (or me or

perhaps I) hammering; (I am hammering); “Me — running;”

“Me — playing horse” and so on ad libitum. So he will ask,

‘ ‘ Where- —papa going ? ”, ‘ ‘ Where — papa been ? ”, “ Where — my

book?”, “My dog — running?”, and so on. Helen Keller says

in her Autobiography that when she was seven she used such ex¬

pressions as these: “Eyes — shut; sleep — no,” (Their eyes are

shut” — speaking of puppies — “but they are not asleep”.)

4 4 Strawberries — very good, 5 ’ and so on.

It is not difficult to understand why the child should thus do

violence to the logic of our sentences. He can convey his limited

range of thought adequately without the copula ; being an adept

at gesture and intonation he can make these latter discharge the

office of the former. Of course his expressions always relate to

very definite concrete experiences, so that he can make himself

intelligible even with an imperfect handling of our linguistic

tools; but when he comes to deal with more abstract situations,

where every detail of the thought to be conveyed must be sug¬

gested by his own expressions, receiving no aid from immediate

conditions and occurences, as the child’s expressions come grad¬

ually to be concerned with complex experiences remote in time

and place, then he will feel the need of having command of a

188 Wisconsin Academy of Sciences , Arts , and Letters.

larger and larger assortment of linguistic symbols, and of em¬

ploying them in the precise conventional manner, else he can not

make himself understood. Here again the child and primitive

man are on a level, as Powell, Romanes and others have pointed

out.

I think that so far as actual need is concerned, the child could

go on for a long distance, say up to the sixth or seventh year

ordinarily, neglecting the verb and particularly the copula in his

sentences; but with the logical forms of the adult constantly

ringing in his ears he comes to adopt them as a matter of con¬

vention at the outset, and not because he feels they are of any

special service to him. The parent and governess and teacher

are incessantly putting the standard forms before the novice,

and forcing them upon his attention, and as a consequence he

abandons his own original, abbreviated, gesture-symbol forms,

and takes up with the conventional models. Just observe a child

saying, for instance, 4 ‘Doggie — high’’ (The dog jumps, or is

jumping, over a high fence) and notice the parent repeating

after him “Doggie jumps high,” and asking the child to follow

suit. This is going on incessantly in the first years of language

learning; if the parent is not dictating conventional forms, then

the brothers and sisters and playmates are. Of course, the con¬

ventional forms are sailing about the child all the time, even

though the speakers are unconscious of his presence, and it is

inevitable that he should in time come to imitate these forms in a

more or less sub-conscious, mechanical way. So the child is not

let alone to do as he chooses linguistically; the social milieu re¬

sorts to various devices to get him to abandon his primitive lin¬

guistic forms before he feels the need of it. Not only are the

standard usages constantly thrust into his ear by all charged

with his care and culture ; but the people around him make gen¬

erous use of ridicule to hasten his progress in adopting the

standard modes. Observe an eight-year old boy making fun of

his three-year old brother for some of his childish phrases, and

the importance of this force in urging the child to abandon his

original expressions, though they serve him well enough, will be

appreciated.

O’Shea — The Child’s Linguistic Development . 189

3. INTERACTIONAL i FUNCTION IN EARLY SPEECH.

Thus far I have spoken only of nominal and verbal function in

the child's earliest sentence-making. Perhaps I should have be¬

gun with the interjection, since this, viewed from one standpoint,

is the first part of speech to appear. It may be observed, how¬

ever, that Meiklejohn* 2 and others say the interjection is no real

part of language, since it does not enter into the organism of the

sentence. But the observers of infantile linguistics have re¬

tained the interjection as a part of speech; and, according to

Tracy’s Summary, the vocabulary of the average child of about

two years contains less than two per cent of interjections.

Salisbury3 maintains that in the vocabulary of his child at

thirty-two months there were only five interjections out of a total

of six hundred forty-two vmrds. At five and one-half years

there was but one interjection out of a total of fifteen hundred

twenty-eight words. The table given by Kirpatrick4 shows

about the same prenomenon as Salisbury’s. Now, these classifica¬

tions are made strictly ah extra, following the formal grammati¬

cal categories. But, regarding the matter psychologically, there

is an interjection element in most of the child’s early words, as

Mrs. Hall5 appears to have observed. She maintains that the

language of her child from the two hundred thirty-third to the

three hundred fourteenth day was an “ inter jectional onomato-

poetic race-language. ’ ’

1 may illustrate the point in question by citing B’s use of

Kee (kitty). Whenever he uttered it, in the beginning, there

was always something of the “Oh!” quality about it. The kit¬

ten was for some weeks a fresh surprise every time he beheld it,

and he used his word with much feeling. He might with pro-

il do not here distinguish between inter jectional and exclamatory

function, though in strict grammatical treatment this should doubtless

be done. Professor Owen makes this distinction. The' interjection is

a sentence element, though it is not strictly a part of the sentence.

The exclamation may be expressionally self sufficient.

2 See his English Language, Part I, p. 60.

s Educational Review, Vol. VII, pp. 287-290.

4 See his The Fundamentals of Child Study, p. 236.

5 Op. cit. p. 601.

190 Wisconsin Academy of Sciences , Arts , and Letters.

priety have used “Oh!” in place of kee. One who follows a

child about as he learns new objects cannot escape the conviction

that his expressions all have for a time at least an inter jectional

quality, as the grammarians use the term. It is interesting to

note in this connection that anthropologists, as Aston,1 e. g.,

maintain that human speech originated in certain natural cries,

— hisses, shouts, grunts, — and these in time became interjections.

Interjections were in the beginning then the only parts of

speech ; all others were included in them. Whether this position

can be defended or not, it is at least evident to me that inter-

jectional speech comes very easy to the young, and it is promi¬

nent up until the adolescent period. One may hear children

(boys especially, I think), from five to twelve incessantly using

expletives such as Gee Whiz! Giminy Crickets! and so on

through a long list. They are all employed, it seems, in expres¬

sion of strong feeling, or to emphasize a thought put forward in

conventional fashion; and the child’s attitude is in some measure

interjectional, even if he does not use the particular forms recog¬

nized by formal grammar. He can use “horse” with interjec¬

tional content and function about as readily and effectively as

“Oh!” or “Whew!”, or “Hurrah!” In the course of develop¬

ment this exclamatory or interpectional coloring of the child’s

language gradually declines after the age of seven or eight, say,

so far as ordinary speech is concerned, though throughout the

whole period of childhood, and to a less extent during youth, in¬

terjectional function is much in evidence. One result of devel¬

opment is to gradually confine interjectional function to the con¬

ventional terms, whereas in the beginning, as I have said above,

every term may have a greater or less degree of interjectional

quality.

4. ADJECTIVAL AND ADVERBIAL FUNCTION IN EARLY SPEECH.

Let us glance for a moment now at the place of qualifying and

particularizing terms in early linguistic activity. The term

modifier suggests differentiation in mental content, and we should

not expect to find limiting terms and phrases employed, intelli-

i See the London Journ. Anthr. Inst., Vol. XXIII, pp. 332-362.

O’Shea — The Child’s Linguistic Development. 191

*

gently at any rate, until the child’s thought had attained a con¬

siderable degree of complexity, so that he might feel the need of

some particularization in his expression. Of course, the child’s

appreciation of particular properties of objects is implicit in his

reactions upon them, — in his attitude toward his kitten for in¬

stance, — long before he employs qualifying terms. He shows

that he regards his kitten as 4 4 nice,” for illustration; but still the

notion of niceness as a general attribute is not dissected out, so

to speak, from the mass of impressions in which it is imbedded.

Then the right use of a modifying term requires the generaliza¬

tion of numerous experiences, ail of the same general effect. So

two processes must go on pari passu in order that the child

should feel the need of verbal instruments to function as modi¬

fiers. In the first place, there must take place continual differ¬

entiation in the body of general impressions which together con¬

stitute any particular object; and in the second place, there must

be constant generalizing of similar experiences with objects, giv¬

ing certain types of experience which are designated by modifiers

and attached to objects according to the type of experience which

they yield. Of course, qualifying and particularizing terms

may very early be used which have the outward aspect of modifi¬

ers, but inwardly they do not function as such. They are the

resultants of mere mechanical imitation. Take, for example, H.

who at two and one-half years would say, when running to greet

her father returning to the house, ‘;nice papa.” She had been

taught this formula, and it probably was the expression of no

different mental content from when she said simply 4 4 papa. ’ ’ So

she would ask for a “nice story;” but what she wanted w*as a

story, not some special kind of a story, — a nice as distinguished

from some other sort of a story that she had heard. She did not

employ nice as a particularizing term. Again, I say in the pres¬

ence of S. at nineteen months, 4 4 nice mamma,” at the same time

patting her head. He imitates my action and words, but mani¬

festly he uses both words as a single term. Possibly the patting

suggests to him some of the mother’s special qualities denoted by

nice, but even so his conception must be extremely dim and unde¬

fined. Now, if a friend should happen in while he was caressing

his mother and using his new* term, it would be thought that he

192 Wisconsin Academy of Sciences , Arts, and Letters.

♦

was employing a modifier in an intelligent manner simply because

in grammatical form it resembled such. It should be noted, how¬

ever, that the use of a modifying term even mechanically does

tend to draw attention upon certain characteristic qualities of

the object in question ; so that mere formal imitation often ends

by hastening the intelligent use of terms, a point to be worked

out in detail later.

The principle is that at the outset the child views the kitten,

to keep to our illustration, in a certain very concrete, totalized

way, without differentiating the notions of niceness, of gentle¬

ness, of playfulness, and so on. But as experiences with the

kitten and other domestic animals (including human beings pos¬

sibly) increase, these ideas gradually gain a certain degree of in¬

dependence or individuality. The attribute denoted by nice for

instance, is, of course, always experienced in connection with

some definite thing ; but as the number and variety of such things

are augmented, the characteristic of affording pleasure of a

special sort, to which is attached the conventional symbol, nice,

being common to all, it acquires a kind of existence apart from

the particular things which occasioned it. When this stage is

reached the child can use the modifier in an intelligent manner.

He can say, “I have a nice doggie, ’ ’ and the adjective indicates

that a particular characteristic of his dog has come to clear con¬

sciousness in his reactions. He appreciates the quality as such,

too, for he can employ the term appropriately in reference to

other objects where he could not have mechanically imitated its

use. Nice then has become a true particularizing term; and the

principle is universal in its application to the natural history of

all modifiers. Of course, a term like the one in question is in¬

cessantly changing in its content and in the range of its applica¬

tion. As development proceeds extensions are made here, and

excisions there. Experience is all the time at work remodelling

it; and just what is accomplished depends upon the peculiar

character of the experiences. Here is a home where the children

hear the term applied frequently under a variety of circum¬

stances; both physical and spiritual characteristics are desig¬

nated by it. But here again is a home where the term is used

infrequently; the members of the family rarely take the atti-

O’Shea — The Child’s Linguistic Development. 193

tudes toward things denoted by this term. The children

from these two homes will have quite different bases for the em¬

ployment of this adjective; and the principle applies to the de¬

velopmental history of all qualifying terms.1

It may sound like a commonplace to say that the adjectives

which are earliest used relate to the impressive characteristics

(depending upon the child's peculiar experience), of the objects

with which he has direct, vital relations. The peculiar charac¬

teristics of different articles of food are among the very first to

become differentiated, and designated by separate terms,2 so that

the adjectives appearing first in the vocabulary are such as

4 ‘nice,” “sweet,” “bad,” “hot,” “good,” “cold,” and the like.

Some of the terms descriptive of the child ’s experience with food

apply also to experience with other objects, and it happens that

these special terms become more prominent than any others.

Large things early impress the child deeply, and his social en¬

vironment intensifies his natural tendency in this respect by lay¬

ing special emphasis upon big things in stories, and in all repre¬

sentations and descriptions of the child’s surroundings. So

“big,” “great,” “awful,” and the like, early acquire promi¬

nence in his vocabulary, as do “little,” “small,” “tiny,” and

similar terms. So if one should go through with all the types of

experiences of the child at different stages of his evolution he

would find that intelligent adjectival function depends directly

upon the degree to which particular attributes of objects become

differentiated from their general characteristics because of the

new relations which the individual, as an inevitable consequence

of his development, comes constantly to assume toward them.

Terms denoting abstract moral qualities in things appear in the

vocabulary last of all, unless such terms are imitated in a me¬

chanical way.3 Of course, it is utterly impossible to tell just

il discuss this matter in detail in the chapter on The Development

of Verbal Signification.

2 It will be appreciated, of course, that long before the cEild uses

conventional terms to denote the qualities of his food, for example, he

indicates his appreciation by gesture and facial expression with char¬

acteristic interjectional expression of rich variety and complexity.

3 All observers of child linguistics give instances in illustration of

this principle; but see Chamberlain (Ped. Sem., 1904, Vol. XI, p. 278).

13

194 Wisconsin Academy of Sciences , Arts , and Letters.

what is the extent and content of a term as the child employs it

at different stages in its developmental history ; but still the evi¬

dence all indicates that abstract moral qualities are not appre¬

ciated until relatively very late, so that the terms designating

them are not intelligently handled until the later stages of de¬

velopment. This is not to say that adjectives denoting moral

qualities in adult speech are never used early ; indeed such terms

as “good,” “bad,” “horrid,” “ugly,” “mean,” “nice,”

“naughty” and the like are applied to persons as early as the

third year; but they are always used in a very concrete, even

physical way. The young child has had some unhappy physical

experience with his playmate, and he calls him “bad,” or

‘ ‘ ugly, ” or “ mean, ” or “ horrid ; ’ ’ but as he develops he will be

apt to use these terms to denote more and more general attitudes

of persons; to designate “qualities of heart,” as well as, or per¬

haps rather than, mere muscular traits.

In her eighth year, H., who had been read to a great deal, and

who had herself at that time read twenty-five books of classic

fairy-tales and fables and myths, and nature stories, and even a

few novels which her parents were reading, — with this linguistic

experience H. occasionally used in her conversation such a term

as “excellent” or “genuine.” She would say, speaking of a

character in one of her books, he was an “amiable” or “genial”

or “excellent” person. Now, when I would test her understand¬

ing of one of these terms I would find that she had in mind some

definite act described in one of her books, and she had remem¬

bered this term as applied to the particular character in ques¬

tion, and had seized upon it without any adequate idea of its

significance. When I ask her to apply one of these terms to her

playmates she shows that she has but a slight and very hazy no¬

tion of its precise connotation. Of course, she has only a very

general and quite incomplete idea of the qualities denoted by

“virtuous,” say. It will require the experiences of many more

years before she can use this term with a true sense of its signifi¬

cance, as this has been determined by racial usage. In the mak-

His child, in her third year, used the word “sinecure” without the

slightest idea of what it meant. See also Hall: The Contents of Chil¬

dren’s Minds on Entering School. (Heath & Co., Boston.)

O'Shea — The Child's Linguistic Development. 195

ing of the term in phylogenesis some such stages have been

passed through, in growing from concrete and physical to more

and more general and abstract reference, as the child passes

through in his acquiring the ability to employ the term correctly

and efficiently to connote moral quality.

What has been said of the development of adjectival function

applies practically without modification to the development of

adverbial function. The only word needing to be added here

is that the adverb appears considerably later than the adjective,

and even when learned it is used much less frequently, as all

vocabularies indicate. According to my observations, the term

earliest used adverbially is one denoting place, — here 1 in ‘ 1 here I

am.” There and where are used early also. Mrs. Moore1 thinks

these antedate ail parts of speech except interjections and nouns.

But unless under exceptional conditions of training, it is prob¬

able that with the possible exception of “here,” “up,” “down,”

“there,” and “when,” they appear later than the more con¬

crete adjectives relating to quality of food and prominent char¬

acteristics of dogs, playthings,2 etc. As we might expect, ad¬

verbial function at the outset is confined to the immediate, con¬

crete, physical needs of the child, and relate to time and place

principally. S., in his fifteenth month, being on the second floor

and in his father’s arms, points to the stairs, at the same time

urging his body in that direction, and says, “dow” (down). So

he will point to objects and say “da” (there), “uh” (up), and

“he” (here), “more,” “out,” “now,” “where,” “away,” and

possibly two or three other adverbs are found in the vocabularies

of children before the close of the second year, though they are

not always used with precision, according to the traditional

standards.

5. PREPOSITIONAL AND CONJUNCTIONAL FUNCTION IN EARLY SPEECH.

From what has been said in previous sections, it must be ap¬

parent that the part played by connective terms in adult speech

is carried to a large extent by gesture in child linguistics. Of

1 Op. cit., p. 129.

2Cf., Hall, op. cit., pp. 604-606.

196 Wisconsin Academy of Sciences , Arts , and Letters .

course, connective function is almost wholly lacking in the in¬

fant’s expression, since his mental content is not sufficiently

differentiated to require connective terms; or at least he can get

on very well without them. When a boy of seventeen months

says “My — go — snow” (I want to go out into the snow) and a

little later Avhen he comes in, exclaiming “My — come — snow” (I

am coming in from the snow), — here the child is not in any ex¬

plicit sense aware of the difference between going to the snow

and coming from it. In his own thought snow occupies the all

important place. His attention is filled with his experience in

the snow. In the first instance he longs to have these experiences

repeated, and his sentence, “my — go — snow*,” will reveal his de¬

sires completely and definitely to his care-takers. In the second

instance his “my — come — snow*” also meets the needs of definite

expression; here his impulses concern the imparting of his ex¬

periences to his care-taker ; and these experiences do not include

to any appreciable extent the relation expressed by the preposi¬

tion in adult speech. The verbs — “go” and “come” — used in

this special connection, include the idea of prepositional rela¬

tion, so to say, — a principle exhibited in all primitive language,

according to Muller, Sayce, Powell, Romanes, and others. With

the child’s relatively undifferentiated experience, and with his

facility in gesture, as I have suggested, he may readily convey

his notions without such relational terms, and this is why he

never employs them at the outset of sentence making. As ex¬

periences multiply and become ever more complex, and there

arises an urgent need to express precisely experiences remote in

time and place, gesture is found sooner or later not to be definite

enough, and then prepositions will begin to find their way into

the child ’s vocabulary. It may be added by way of qualification

that the imitative tendencies of the child lead him often to

adopt connective terms before he has real need for them, but it

is probable that such terms are not imitated as readily as those

expressive of concrete elements of experience. However, me¬

chanical imitation must be reckoned with, and if it were not for

this the average child would, I think, leave relational terms

locked up in the other parts of speech somewhat longer than he

usually does.

O’Shea — The Child’s Linguistic Development. 197

One cannot easily detect tlie emergence of prepositional ele¬

ments in early speech. Their individuality is at first not at all

marked or distinct. It is as though they were still a part of the

organism in which they were originally imbedded. H. at nine¬

teen months says, as a typical expression, “Papa — go — u — Uni¬

versity,” the u here being evidently a mutilated form of the

preposition “ to At the outset it was lacking altogether ; but by

the twenty-sixth month it had become differentiated completely

from the verb. We catch it here in the nineteenth month in its

embryonic form; and so far as I have observed, all prepositions

have a similar history, which seems to be much the same in prin¬

ciple in phylogenesis as in ontogenesis. Powell,1 commenting on

prepositional function in the Indian language, maintains that

prepositions are often intransitive verbs. When an Indian says,

i ‘ That hat table on, ’ ’ we are to consider the “ on ” as an intransi¬

tive verb which may be conjugated. “Prepositions may often

be found as particles incorporated in verbs; and still further,

verbs may contain within themselves prepositional meanings

without ever being able to trace such meanings to any definite

particles within the verb . . . Prepositions may be pre¬

fixed, infixed, or suffixed to nouns, i. e., they may be particles

incorporated in nouns.”

In some children’s vocabularies “up” and “down” are given

as about the first prepositions to appear, and they are said to

be used properly by the eighteenth month, or so. The child

says, “up-stairs” (I want to go upstairs) and “down-stairs.”

Now, as I have observed the early use of these words, I should

say they were not employed with exclusive prepositional mean¬

ing at all. In the beginning the child says simply “up,” and

makes this expression definite by extending his arms upward,

by straining upward with his body, by looking upward, and by

so employing his voice as to leave no doubt respecting his de¬

sires. I should say his word really denoted the place he wished

to reach, and the method o£ reaching it, although neither of these

elements would be focal in consciousness in the sense in which

we can imagine they might be in the case of an adult who sat

1 Op. cit ., p. 46.

198 Wisconsin Academy of Sciences , Arts, and Letters .

down and reflected upon getting up stairs. The child’s con¬

sciousness might be said to be motor rather than ideal when he

is expressing himself in this wray. The word is just one phase

of a general motor excitement, and it is impossible that it should

be used with precise prepositional value. Before this word can

be employed as a preposition merely, a number of other words

will need to be used with it in the sentence, each to carry phases

of the meaning which it now carries alone.

In her twelfth month, K. w'ould throw objects from her high-

chair to the floor, and would exclaim “down!” To my mind,

this term denoted mainly the racket made by the objects when

they struck the floor. Prepositional relation was surely not a

prominent element in the child’s consciousness on such occa¬

sions. A little later she would take an object in her hand, and

at the moment of releasing it she would exclaim “down!” and

blink, evidently anticipating the noise to follow, wdiich was the

thing prominent in consciousness. Later on she would use the

term when she wished to get out of her high-chair, but here,

also, it had much more than prepositional meaning. Her con¬

sciousness could be expressed in adult language by the following,

perhaps: “Unfasten me so I can get onto the floor and play.”

It is improbable that the child uses such words as “up” or

‘ ‘ down ’ ’ with strict prepositional meaning, or adverbial meaning

either, before his third year at the earliest, and I should prefer

to make it a year later. In the fourth year one may hear ex¬

pressions like the following: “I am going down the street;”

“I climbed up the stairs,” etc., in which we doubtless have ex¬

amples of genuine prepositional function. The original terms

“up” and “down” have persisted, but much of their early

meaning has been drawn off from them, and loaded on to other

terms in the sentence; and such is the history of other preposi¬

tions, as on, in, etc.

It should be pointed out that there are prepositions which are

never used except with prepositional meaning pure and simple.

They describe relations which the child does not apprehend until

he has made good headway in differentiating the parts of speech,

and constructing the sentence. Before such terms as toward,

among, against, notwithstanding, and the like are employed he

O’Shea — The Child’s Linguistic Development. 199

has abandoned his primordial sentence-words, and in their place

he nses sentences with substantive, predicate, and modifiers, so

that any one word now carries special, differentiated meaning.

When S. at four says: “I throw it toward the house,” he shows

that he has reached the prepositional plane, so to speak, in

linguistic development. It may be added that the principle here

in question applies to development in respect of all the parts

of speech. To illustrate with our adjective nice , already often

cited, this is first used as a sentence-word ; but a term like virtu¬

ous is never employed until the sentence- word period is out¬

grown, and the word carries adjectival meaning alone.

It will not be necessary to dwell long upon the proposition

that conjunctions as such appear relatively late in the child's

language. It is probable that the primitive sentence-word car¬

ries conjunctional function, but to a very limited extent, I should

say. It is questionable whether the thought of a child of two

is integrated to the degree required for the intelligent use of

the conjunction. The central processes relating to any situation

are, relatively speaking, fragmentary, disjointed; or better still,

unconnected, or non-integrated. Now, development results in

the gradual integration of elementary processes, and this makes

necessary the use of the conjunction in expression. Probably

the earliest sort of integration has reference to objects acting

simultaneously in the same way. In the beginning the child

will say, “ Baby-go-stairs” (Baby is going upstairs) ; “ papa-

go-stairs” (papa is going upstairs). But before the completion

of the second year, one may hear this expression, “Baby a papa

going upstairs.” This example is typical of much that may

be heard as early as the twenty- fourth month. Judging from

my own observations I should say that objects acting simultane¬

ously and congruently are coordinated considerably earlier than

are the acts they perform, or the qualities predicated of them.

One may hear children after the third year say, “My run and

fall and get up again,” and “Mamma nice and good,” and the

like; but such expressions appear later than the first type men¬

tioned. It is probable that two objects acting in the same way

fuse in the child’s thought more readily than succeeding actions

or co-existent qualities of the same object. Baby and papa, go-

200 Wisconsin Academy of Sciences , Arts, and Letters.

ing upstairs together, are apprehended in a single act of atten¬

tion so they tend to stick together in representation, and in ex¬

pression they require to be named together. In atten¬

tion they gradually fuse into a unity which ultimately will be

expressed by we; but in the child’s speech the old habit of nam¬

ing each object separately persists, and so he must use “baby”

and ‘ ‘ Mamma ’ ’ The use of the and shows the growth in coordin¬

ation or integration of the two objects. Now, there is not quite

the same necessity for coordination in successive actions per¬

formed by the same object, though, of course, with development

they tend constantly toward integration, and by the fourth year

it is plain that coordination has been achieved. The child

then joins with the conjunction two or three of his own acts,

as well as those of his parents, his brothers and sisters, his dog,

and so on. And what has been said of the coordination of ac¬

tions applies also without modification, I think, to the coordina¬

tion of attributes.

The first conjunction appearing in the child’s speech is un¬

questionably and. As for the order of the appearance of the

other conjunctions one can not speak with certainty. Probably

or is the second to be used with strictly conjunctional meaning.

The child says, “Baby have apple or peach?” This expres¬

sion was forced upon H. early because of her being required

to choose between eatables, the parent saying: “Take this or

that.” The child early hears or used a great deal, — “Hurry

or I will go;” “look out or the baby will fall;” and so on ad

libitum. Of course, the central processes required for its in¬

telligent employment are quite a bit more complex than in the

case of and; and I think it is apt to be employed as a result

of mere imitation at the outset.

The general principle holds, that the appearance of any con¬

junction depends primarily upon the complexity of the thought

which it is employed to express, though imitation is always a

disturbing factor, leading to the mechanical use early of a term

much heard from the lips of parents and others. Because is

snch a term, I think. Quite early one may hear the child say¬

ing, ‘ ‘ ’cause I do, ” “ ’cause I must, ” “ ’cause I want to, ’ ’ and

O'Shea — The Child's Linguistic Development. 201

so on ; and it is probable that his thought is not complex enough

to really demand these expressions.

I have endeavored to determine just when such words as ex¬

cept, although, unless, lest, in order that, nor , whether — or,

and so on, appeared in the vocabularies of my children, but I

find I cannot speak with certainty about the matter. Of this

I am confident, however, that none of these terms is employed

with precision before the fifth year. V. at six and a half does

not. use one of them intelligently, so far as I can detect

But H. at nine uses them all fluently. She has read much, and

has been much read to ; and it is probable that these terms have

forced themselves into her vocabulary mainly because of their

prominence in her reading. She has heard them in the speech

of the people about her, and she has been correctly interpreting

them for years; but they have all played a minor part in her

consciousness of spoken language. So far as auditory language

is concerned, relatively unimportant elements are swallowed up

in wholes of greater prominence; but they are likely to gain

some measure of individuality when reading is begun, though

they are at the same time likely to lose it again as the reader

gains in facility in grasping and interpreting larger and larger

language units. It may be added that a child of five seems to be

able to express himself definitely and fully enough without re¬

sorting to any of these conjunctive aids that imply quite com¬

plex ideational integration. If he did not find these terms ready

to hand, and if they were not continually impressed upon him,

I think he would not miss them, at least not until he should

be placed in situations where he would be required to express

involved thought very connectedly and precisely.

6. PRONOMINAL FUNCTION IN EARLY SPEECH.

The absence from early speech of anything which could be

called a pronoun has attracted the attention of all students of

linguistics, and of psychologists and philosophers as well.

Philosophical literature is full of speculation concerning the de¬

velopment of self-consciousness in a child, indicated by his use

of the personal pronoun. The philosophers, many of them, have

202 Wisconsin Academy of Sciences, Arts, and Letters .

said that the child does not distinguish self from others, the

ego from the alter, until the terms “I,” “my,” “mine,” “you,”

“yours,” “he,” “him,” “his,” begin to appear in his vocab¬

ulary, which most observers have found to be somewhere about

the twenty-fifth month, though a few have not noticed it until

the beginning of the third year. Ament detected it in the

twenty-first month, Schultze in the nineteenth, and Mrs. Hall

even in the seventeenth. It is suggestive in this connection to

note that primitive languages show great confusion in the use

of the pronoun. Brinton1 maintains that in aboriginal American

languages there is no distinction between persons in the pronoun *

“I,” “thou,” and “he” are not discriminated, a single mono¬

syllable serving for all persons, and also for both singular and

plural numbers. In some American languages, however, there

is a great variety of pronouns, used to denote not only person

and number, but various conditions and aspects of the person

or persons designated, as that they are standing, sitting, or lying*

alone or with others, moving or stationary,2 3 and so on. Accord¬

ing to Powell,3 4 4 The Indian of today is more accustomed to say

this person or thing, that person or thing, than he, she or it.

Among the free personal pronouns the student may find an

equivalent of the pronoun 4 1,’ another signifying 4 1 and you/

perhaps another signifying 4 1 and he’ and one signifying 4 we/

more than two, including the speaker and those present, and

another including the speaker and those absent. He will also

find personal pronouns in the second and third person, perhaps

with singular and dual forms.” The pronouns are not in all

cases completely differentiated in these languages, but are in¬

corporated in the verb as prefixes or infixes or suffixes, and as

such they designate the person, number, and gender of both sub¬

ject and object, and in the conjugation of the verb they play

an important part.

How is it now with the child ? Is pronominal function in his

iSee his Essays of an Americanist (Philadelphia, 1890), p. 396.

2 Powell says that in Indian languages genders are, not confined to

sex, but are methods of classification primarily into animate and in¬

animate, which are again classified according to striking characteris¬

tics or attitudes or supposed constitution.

3 Op. cit., p. 43.

O'Shea — The Child's Linguistic Development. 203

case discharged by the verb or some other part of speech ? Does

he employ a single term for all persons and things ? What need

gives rise to the differentiation of special words to carry pro¬

nominal function? In discussing these questions it should be

said at the outset that from the very beginning the child in his

reactions distinguishes himself from others and from things.

Of course, he does not make this discrimination reflectively; but

nevertheless he does not confuse himself with foreign objects

when he is in need of food, say ; though, as President Hall1 has

shown, he may not recognize his fingers and toes as his own.

But when he is hungry he does not give his food to another,

thinking that the other is himself. As early as the sixth month

he exhibits in his reactions a certain realization of the opposi¬

tion between ego and alter, for he will squall if another takes

his bottle, or even if the mother shows, overt partiality for some

other child. This appreciation is very keen at a year and a half ;

though the child does not yet use terms that denote distinctions

in persons. I mention these obvious facts merely to suggest how

far astray some persons have gone in assigning the birth of the

ego to the period when the personal pronoun first appears.

Their treatment of the matter has been purely a priori and meta¬

physical. They have reasoned that because “I” denoted self,

the self must be in consciousness as a distinct object when tbe

term is used, and if it can not be used the self can not be con¬

ceived of as a distinct entity; but they have overlooked the fact

that for months before the child uses the pronoun he has been

using other modes of expression wThich show clearly that his

personal self exists in consciousness as a thing apart from all

other things.

What then are these modes? First of all, gesture, grimace,

pantomime, intonation. When a vigorous year-old child wishes

to be taken in your arms, no one who sees and hears him can

doubt that his discrimination between ego and alter is veiy

clear. All that can be denoted by “ I ’ ’ is exhibited by the chii i,

though in a generalized, consolidated, impulsive, instinctive, non-

reflective fashion. Again, when you see a child of this age

!Se,e his Some Aspects of the Early Sense of Self. Amer. Jour, of

Psych., April, 1898. Yol. IX, pp. 321-395.

204 Wisconsin Academy of Sciences , Arts , and Letters.

scream and strike at his brother who appropriates his food or

playthings, you cannot doubt that he possesses a rudimentary,

undifferentiated sense of ‘ ‘ mine. ’ ’ When, again, this same child

offers his father a taste of his sugar-lump, and exclaims

“Ndobbin,” 'ndobbin” with interrogative intonations, he is

certainly acting out the question “Will you have some sugar?”

The you as contrasted with 1 is involved in the child’s action,

though he can utter not a syllable to denote the distinction.

Further, when the child’s brother performs tricks for the babe,

and the latter turns to the father or other person, and pointing

at the brother laughs at him and gabbles about him, — in a reac¬

tion of this sort the idea of he, or possibly it, is clearly involved*

There is a third person in the case, who is not now in vital

relations with the speaker and listener. He is being talked at) out,

not to. In this latter situation the child shows in his reaction,

not reflectively, an appreciation of all three persons in their

grammatical relations to him, so to speak.

We have seen how, in the course of expressive development,

verbal symbols come gradually to take up the function which

had been originally discharged by gesture and pantomime; and

the principle obtains in respect of pronominal as of all other

forms of linguistic function. In the beginning the child desig¬

nates persons and things by gesture, and pronominal function

in this stage might be said, perhaps, to be demonstrative. Even

when he wishes attention turned upon himself he indicates it

by characteristic bodily attitudes and contortions and vocal

demonstrations, saying, in effect, “this person needs your help.”

But as development proceeds,, demonstrativeness in linguistic

function declines, and pure symbolization increases; and in the

matter of pronouns it results that terms introduced which

not so much point out or demonstrate as name or denote

merely. This is true, of course, of racial as of individual

evolution; to the primitive mind things must be made very ob¬

jective, concrete, explicit, but with mental development simple

suggestion becomes continuously more efficient. In other words,

language becomes ever more abstract, which means relieved of

direct, concrete reference.

In his pronominal evolution the child passes first from the

O'Shea — The Child's Linguistic Development. 205

pantomine to the nominative stage, he gives its name to every¬

thing to which he alludes, including himself. If his elders ad¬

dress him as ‘'baby” then he always uses this term when refer¬

ring to himself in any way ; or if his proper name is used, then

he employs this on all occasions. So he says, — a phenomenon

observed by every student of the matter, I think, — ‘ ‘ Baby wants

baby milk;” or “Baby hurt baby hand,” and so on ad libitum.

In the same way he says, addressing his father, “Papa take

baby,” usually, I may add, in the imperative mode. Similarly,

when speaking of his brother, he will say, “Stanley is putting

on Stanley coat.” V. continued in this nominative stage until

he was past five years ol‘ age; then with great swiftness he went

over into the pronominative stage. Within his linguistic range,

he used pronouns with considerable freedom, accuracy, and

efficiency by the time he was six and a half, though he still got

the cases of his personal pronouns mixed at times, and he could

not use the relatives according to the prevailing standard; his

whats and his thats, for instance, gave him trouble. H. and S.

were well into the pronominative stage by the time they were

three and a half; and by six they had overcome all their diffi¬

culties.

Why does the child pass through the nominative on his way

to the pronominative stage in linguistic function? To begin

with, the name of a person is far more definite and uniform than

his pro -name, and so all persons in speaking to the child use the

former and avoid the latter, as Preyer1 has already pointed

out. To illustrate : In addressing my child I say, ((Papa wants

this or that;” or “ Papa will do or did this or that,” and so on

ad libitum. The mother speaking of the father in the presence

of the babe says, “Papa loves baby;” or “Papa has come home,”

and so on. Now, everyone who mentions the father when the

babe is concerned uses this term invariably; and the same is

true in principle of the baby himself, and the mother and broth¬

ers and sisters, and every object mentioned. If the pronoun were

used, see what confusion (from the learner’s standpoint) would

result. When I referred to myself I would designate myself

iThe Development of the Intellect. (Trans, by Brown), p. 202.

206 Wisconsin Academy of Sciences , Arts } and Letters.

by “I” or “my” or “me;” when the mother addressed me di¬

rectly she w'ould designate me by “yon” or “your’s;” when

she spoke to the babe about me she wonld use “he” or “him”

or “his.” Here are eight symbols for the same object, looked at

from the child’s standpoint, and it would be a long story to tell

how he could orient himself with reference to each and all of

these terms for the same individual. To present the matter in

a sentence here, — the individual must reach a stage of develop¬

ment where he can organize a variety of experiences around a

common center before he can comprehend or use intelligently our

system of personal pronouns. As you watch him pushing for¬

ward in integrating ability, you see him adopting first one form

and then other forms of the pronouns. At the outset he makes

his one form do duty in all cases. “Him is a nice boy;” “Me

wants to go to him ’s (or perhaps he’s) house,” are illustrations.

We shall go into this in greater detail in the chapter on Inflec¬

tion; but it may be noticed here that the young child cannot

readily accommodate himself to the notion of having different

forms of his words apply to the same unchanging thing, un¬

changing so far as he can see. This leads parents, more or less

intuitively, to avoid the pronouns in speaking to young chil¬

dren, and this has the effect to retard the appearance in the

vocabulary of pronominal forms.

Then the pronoun, as employed in conversing with a child,

lacks individually, warmth, color. Try talking to your year-

old child in pronominative terms, and see how much weaker is

your speech in personal suggestiveness. On the other hand, to

continue the nominative stage too long is equally objectionable;

it seems silly, babyish, ineffective. The opening mind needs to

be assissted in its grasp of things by all possible concrete aids;

but once it has got a hold it knocks out the ladder by which it

lias ascended. This, I take it, is a principle of universal validity

in mental development, and is one of the forces incessantly at

work transforming the individual’s interests and abilities.

This will be the best place, perhaps, in which to glance at the

forms of the pronouns which are used most frequently at first.

I said above that one form of the personal pronoun is often

made to do duty for all cases; but wrhat is this form? Mrs.

O'Shea — The Child's Linguistic Development. 207

Hall’s1 boy used his first; Rzesnitzek2 says that the possessive

form mine is first used, while von Pfeil3 thinks that the pronouns

denoting second person are first mastered; then those denoting

third person, and last of all come those denoting the first person.

In Chamberlain’s4 account of the linguistic development of his

child, “I” and “ my ” appear very frequently after the begin¬

ning of the third year, but the other forms are not in evidence.

Preyer5 observed that his son, Axel, in his thirty-second month

used “I” meaning by it “you.” In the thirty-third month

came such expressions as “das will ich!” “das mocht ich.”

However, before this, in the twenty-ninth month, the objective

form of the third person was used, “gib mir,” and “biite heb

mich herauf .” The boy often used the third person, though,

in designating himself, as when the father would ask “Wo ist

Axel?” the latter would respond “da ist er wieder .”

These citations will perhaps suffice to indicate that there is no

certain and invariable order followed by all children in the em¬

ployment of the personal pronouns. In my own observations

my has been the first form to be adopted. In every case it came

before I. It was used in such relations as the following: My

want to do this or that ; my feel bad ; that is my pencil or apple,

or what not; take my to bed or out of doors. The form mine

came considerably later than my, and I still later. To my mind

the situations involving the use of my aremiore concrete, more

obvious than those involving 7, and it seems reasonable that it

should first appear, and once it gets started it will serve for me,

mine, and 1 for a time. The use from the beginning of all forms

of the pronouns, as given in some of the vocabularies, appears to

me very remarkable, and quite in contrast to the child’s usual

method of procedure in similar situations.

Why does the child not settle upon one form permanently?

For the very effective reason that his social environment will not

1Op. cit., p. 606.

2 Zur Frage der Psychischen EntwicTcelung der Eindersprache (Bres¬

lau, 1899), p. 35.

3 Wie lernt Mann eine Sprache , p. 5.

4 Studies of a Child; Ped Sem ., Vol. XI, pp. 264-291.

& Op. cit., p. 202.

208 Wisconsin Academy of Sciences , Arts , and Letters .

I>ermit him so to do. His parents, once he gets to using pro¬

nouns at all, keep putting the conventional forms before him

whenever he uses a form incorrectly ; his brothers and sisters and

playmates make fun of him for his lack of conformity to environ¬

mental standards, and the teacher tries to habituate him in the

use of the standard forms, and gives him rules for his guidance.

These are all powerful corrective forces, and no child can long

resist them, except in respect of the least important matters.

Then simple imitation, where the child more or less unconsciously

copies the models in his environment, is of immense importance

in leading him to appropriate the various forms employed about

him. It is suggestive to note in this connection that when an

adult tries to write or speak a foreign language with which he

is not very familiar he experiences much trouble in mastering

the cases of his pronouns. If he is just a novice one form will

answer for all cases, and one of the difficult tasks in using a for¬

eign tongue with ease is to gain facility in employing the right

form of the pronoun in different situations; and of course, this

principle applies to other parts of speech than pronouns.

ON THE NATURE OF THE PROCESS OF OSMOSIS AND

OSMOTIC PRESSURE WITH OBSERVATIONS

CONCERNING DIALYSIS.*

LOUIS KAHLENBERG.

INTRODUCTION.

A brief but excellent outline of the history of the develop¬

ment of our knowledge of osmosis up to 1877 is given by Pfeffer

in his well-known monograph, “Osmotische Untersuchungen. ’ ’

The great importance of osmotic phenomena in physiological

processes was clearly recognized as early as 1826 by Dutrochet,

and for half a century later osmotic investigations were con¬

ducted very largely, though not exclusively, in the interests of

physiology. Precipitated membranes were first used by the

botanist, Moritz Traube,* 1 in 1865, and these were employed by

Pfeifer in his researches above mentioned. Special interest in

osmosis has developed since 1887, when by using Pfeifer’s data

of osmotic pressure measurements van’t Holf sought to show

that the simple gas laws hold for dilute solutions. Since the

latter date, so-called semipermeable membranes have been used

almost exclusively in osmotic investigations. These membranes

usually consisted of ferrocyanide precipitates of some heavy

metal, copper ferrocyanide being the favorite precipitate for

osmotic work, though the ferrocyanides of zinc and nickel w'ere

occasionally employed, as were also a few other precipitates.

*This paper will be reprinted in the Journal of Physical Chemistry

for March, 1906.

i Centralblatt f. medic. Wissenschaften, 1865.

14

210 Wisconsin Academy of Sciences , Arts and Letters.

In this later work effort in two directions is clearly discern¬

ible. (1) Those who have been favorable to van’t Hoff’s hy¬

pothesis of solutions (based on the analogy between gases and

solutions) according to which the osmotic pressure, so-called, is

due to the bombardment of the semipermeable membrane, by

the dissolved molecules, have sought either to compare osmotic

pressures of aqueous solutions with each other, or to measure

directly the osmotic pressure of certain solutions in the hope of

securing data to uphold the theory. It is true, however, that

considering the vast importance of direct measurements of os¬

motic pressure for the van't Hoff theory of solutions, but little

effort has been made to measure osmotic pressures directly.

This has come about very largely because of the attitude taken

in the matter by the main adherents of the van’t Hoff theory,

who voiced and continually supported the dogma that the os¬

motic pressure is necessarily independent of the nature of the

membrane if it be semipermeable ; and that since it is very diffi¬

cult to measure osmotic pressures directly, it is better to con¬

tent one’s self with the so-called “indirect” measurements of

osmotic pressure, namely, with a computation of the latter from

vapor tension, freezing-point or boiling-point observations on

solutions, which, be it remembered, involves the assumption that

the gas laws hold for solutions. And so we have the rather re¬

markable situation that direct measurements of osmotic pres¬

sure, and indeed the general investigation of osmosis, has not

only been neglected by the chief advocates of the gas theory of

solutions, but they have in addition through the attitude they

have taken actually discouraged work in this direction. They

have even claimed to have proven by thermodynamics that the

osmotic pressure must be independent of the nature of the

membrane provided the latter is semipermeable. The assump¬

tions made in such “proofs,” and the fact that there is in real¬

ity no such thing as a semipermeable membrane in the strict

sense of the word, have been passed over lightly. (2) Quite a

different direction in the investigation of osmotic phenomena

has been taken by those who have held van’t Hoff’s conception

of the nature of osmotic pressure to be untenable. These men

have continually brought forward experiments, of a qualitative

Kahlenberg — Osmosis and Osmotic Pressure. 211

nature to be sure, showing that the hypothesis of van! Hoff

can not be held and that no special stress is to be laid upon di¬

rect measurements of osmotic pressure, which they have conse¬

quently not attempted to make.

In his efforts to measure osmotic pressures directly, Tam-

mann1 came to the conclusion that it is not possible to obtain

reliable, concordant results by means of the method adopted by

Pfeffer,2 which he consequently abandoned entirely and turned

his attention to comparing the osmotic activity of various solu¬

tions with one another. Attempts at direct measurements of

osmotic pressure have again been taken up recently by II. N.

Morse3 and his coworkers. They have measured the osmotic

pressures that are developed when aqueous sugar solutions are

separated from water by means of precipitated membranes of

copper ferroeyanide. The method they employed is essentially

that of Pfeffer, with the exception that they prepared the mem¬

branes with the aid of electrolysis. Enough can hardly be said

in praise of the care and perseverance exercised by Morse and

his assistants in this work, and yet they have neglected a very

essential point in their determinations as will appear from con¬

siderations given below, and consequently their experiments

are not conclusive in establishing, as they suppose, that the gas

laws hold fairly well for the osmotic pressures of aqueous sugar

solutions,, using copper ferroeyanide membranes. Furthermore,

attempts to generalize from the data collected by Morse and

Frazer on aqueous sugar solutions, as to the behavior of all

solutions taking no consideration of the membranes employed,

are quite unwarranted. Moreover, while according to the work

of Morse and Frazer and also according to Flusin4 the aqueous

sugar solutions show osmotic pressures in approximate conform¬

ity of the gas laws, the Earl of Berkeley and E. G. J. Hartley5

have found materially higher pressures than those deduced

iWied, Ann. 34, 299 (1888). See later attempts by new method Zeit,

Phys. Chem. 9, 97 (1892).

2 Pfeffer, be it remembered, worked solely in the interests of physi¬

ology and for his special purpose, his experiments were quite sufficient.

sAmer. Chem Jour. 34, 1 (1905).

4Compt. Rendus 132, 1110 (1901).

s Proceedings Royal Society (.London), 73, 436 (1904).

212 Wisconsin Academy of Sciences , Arts and Letters.

from the gas law, even though their membranes, which were

also prepared with the aid of electrolysis, were admittedly not

perfectly tight.

Among the many opponents of the van’t Hoff theory of os¬

mosis may be mentioned Lothar Meyer, Raoult, Fitzgerald,

Pickering, Quincke, Dieterici, J. Traube, Battelli and Stephan-

ini. The opponents of the van’t Hoff idea have generally held

that the so-called osmotic pressure is an ordinary hydrostatic

pressure brought about by entrance of liquid into the osmotic

cell. Concerning the reason for this entrance of additional

liquid into the osmotic cell there has, however, been difference

of opinion ; some holding that it is due to attraction that is es¬

sentially chemical in character ; others that it is due to capillar¬

ity; and still others that it is caused by surface tension. The

latter view has lately been prominently brought forth by

Traube,1 whose claim is that ‘ ‘ The difference in surface tensions

determines the direction and velocity of the osmosis.’’ His idea

is that the main direction of osmosis is always toward the liquid

having the greater surface tension. Again, Battelli and Ste-

phanini2 also express the opinion that difference in surface ten¬

sion is the cause of osmosis. But on the basis of their experi¬

mental work, they modify Traube ’s contention to the effect that

the main osmotic current is not always toward the liquid of

higher surface tension, but that the process always proceeds in

such a direction as to tend to equalize the surface tensions of

the liquids on the opposite sides of the septum. Very recently

Barlow3 has also brought forward cases which are not in har¬

mony with the theory of Traube.

In the present investigation the main purpose has been to

inquire into the nature of osmosis and osmotic pressure, and to

test whether the latter really follows the gas laws. To this end

a considerable number of osmotic experiments were performed

in 'which liquids of similar and also of very different character

were separated from each other by different membranes. In

this work observations were taken as to the direction of osmosis

with change of membrane, with change of solvent and also with

l Phil. Mag. (6) 8, 704 (1904).

2Atti della Reale Accademia dei Lincei 14, 3 (1905).

3 Phil. Mag. (6) 10, (1905).

Kohlenberg — Osmosis and Osmotic Pressure. 213

change of solute. The effect of temperature was considered.

The degree of permeability of the membranes for various sub¬

stances was noted, with the result that membranes as semiper-

meable as any known hitherto were found, which are yet not

precipitated membranes. Furthermore it was ascertained why

the membranes were semipermeable in some cases and not in

others. It was consequently possible to foretell for which sub¬

stances the membranes were permeable and for which sub¬

stances not permeable. Further, this work has cast light on the

process of dialysis. It has been possible to separate different

dissolved crystalline substances from each other by dialysis, and

also to separate dissolved crystalline from non-crystalline

bodies by having the non-crystalline substances pass through

the septum and the crystalline ones remain behind in solution

in the dialyzer. The direct measurements made with the semi¬

permeable membranes employed, moreover, showed that the gas

laws do not hold at all in these cases. It was also found that in

making direct measurements of osmotic pressures it is necessary

to stir the liquids separated by the membrane, a very important

fact which has been entirely overlooked in all osmotic pressure

determinations hitherto made. The necessity of such stirring

was really accidentally discovered, as will appear from details

given below, after having worked in vain for over a year try¬

ing to get reliable, concordant results without stirring. The

discovery that agitation of the liquids is essential in osmotic

pressure measurements is of paramount importance in deciding

as to the nature of the osmotic process.

In the presentation of the experimental work which now fol¬

lows, the experiments will not be described in the order in

which they were actually performed for the reason that greater

clearness will be obtained by detailing the results in the light

of the theory which was gradually evolved in the course of the

investigation.

A TYPICAL CASE OP OSMOSIS.

If chloroform A, Fig. 1, be placed in a glass tube and a layer

of water B be poured upon it, and again a layer of ether C be

carefully poured upon the water, and the whole be allowed to

214 Wisconsin Academy of Sciences , Arts and Letters.

stand, there will eventually be but two layers A' and B', Fig. 2.

An examination of A' shows that it consists of chloroform and

B

Fig. 1.

ether saturated with water, whereas the layer B' consists of

water saturated with ether and chloroform. In the change

which has taken place, one layer, that of the ether, has gradu¬

ally disappeared, and the lower layer has greatly increased in

bulk and lifted the aqueous layer to the top. The explanation

of this phenomenon is evident. Ether dissolves very readily in

chloroform, but in water it dissolves much less readily; again,

chloroform and water hardly dissolve each other at all. In the

arrangement we have in Fig. 1, the aqueous layer B dissolves

ether, and in turn the ethereal layer C takes up some water.

When the ether has gone into B until it touches the chloroform

layer A, the latter extracts ether from the aqueous layer B.

Thus the upper part of the chloroform layer A becomes enriched

with ether, whereas the lower part of the aqueous layer B be¬

comes depleted in ether. The latter depletion is made good by

a continuous supply of ether from the upper parts of B, which

are in turn supplied with ether from C. Again, the ether in

the upper layers of A gradually diffuses into the more distant

parts of A. This process of the transportation of the ether in

C through B into A proceeds, then, until the supply in C is ex-

Kohlenberg — Osmosis and Osmotic Pressure. 215

hausted. It is to be noted, however, that since chloroform is

more soluble in water containing ether than in pure water, the

aqueous layer B will take up more chloroform after becoming

charged with ether. Thus some of the chloroform is making

its way upward into the aqueous layer B, from which it also

passes in part into the ethereal layer C.

It is clear then that the ether of layer C is making its way

through the aqueous layer B into the chloroform layer A, and

that on the other hand the chloroform is passing into the

aqueous layer B charged with ether and further into C. Thus

we have currents of ether and chloroform going in opposite di¬

rections. The movement of the chloroform as to quantity is,

however, extremely slight as compared with that of the ether;

and so the movement of the latter practically predominates and

the ethereal layer is finally absorbed. It must be borne in

mind, however, that water, being somewhat more soluble in

chloroform charged with ether than in pure chloroform, will to

a slight extent go into the layer A as it becomes enriched with

ether; and, as has already been remarked, some water also

passes into the layer C, for water is somewhat soluble in ether.

When the change is complete, we have, as stated above, a lower

layer consisting of chloroform and ether saturated with water,

and an upper layer of water saturated with ether and chloro¬

form.

The rate at which this process goes on depends on the tem¬

perature and upon the pressure, but also upon the thickness of

the aqueous layer B and the area of the surface of contact with

the ethereal layer C and the chloroform layer A.i It is further

clear that the process would be hastened if each of the layers

A, B, and C were continually stirred, for by so doing the slow

processes of diffusion would be aided and the changes in con¬

centration which take place in the layers where they are in con¬

tact with one another would be lessened.

Summed up then, the observed change goes on because ether

is soluble in water, but much more readily soluble in chloro¬

form, so that latter is able to extract ether from the aqueous

solution of ether, B. The chloroform — ether solution in A

may be regarded as a solution of chloroform in ether, especially

216 Wisconsin Academy of Sciences , Arts and Letters.

after a considerable quantity of ether has accumulated in it;

and since but very little chloroform passes into the aqueous

layer B, the latter is practically permeable only for ether, and

therefore this aqueous layer B acts as a semipermeable septum.

It permits ether to pass from C to A, but allows very little,

practically no chloroform to pass into C. It is clear further

that as A becomes richer and richer in ether, the tendency for

more ether to enter A from B becomes less and less. Ether

dissolves more readily in chloroform than in water because

ether and chloroform have a greater mutual attraction for each

other than have ether and water; it is for this reason too that

chloroform is able to extract ether from an aqueous solution of

the latter. As ether accumulates in A, however, the power of

this layer to extract ether from the aqueous layer B diminishes,

because the attraction or affinity of chloroform and ether for

each other becomes more and more satisfied. Finally, suppose

we prevent the supply of ether in C from becoming exhausted

by adding some as may be required, a point will be reached

where compartment A has become so rich in ether that the at¬

traction or affinity of this chloroform — ether solution for addi¬

tional ether has decreased to such an extent that it can no

longer extract further ether from the aqueous layer B. In other

words, when the chloroform solution in ether, A, has become so

dilute that its attraction for additional ether just equals the at¬

traction of water for ether in the aqueous layer B, the process

is arrested. The point at which this occurs would clearly vary

with the temperature and also with the pressure.

If in Fig. 1 compartment A be filled with carbon disulphide

instead of chloroform, the process would go on as before in a

similar manner and for perfectly similar reasons. Indeed, any

liquid which in itself does not mix with water, practically

speaking, and yet has a greater attraction for ether than has

water, would serve in place of the chloroform. The rapidity

with which the process proceeds and the final point of equilib¬

rium reached would, however, also be a function of the nature

of the substance so employed. If in Fig. 1 the ether in C be

replaced by an oil, say olive oil or a hydrocarbon oil, retaining

Kohlenberg — Osmosis and Osmotic Pressure. 217

the water in B and chloroform in A, nothing will take place,1

for these oils are not appreciably soluble in water, that is to

say the attraction or affinity existing between them and water

is insufficient to overcome their cohesions to the extent neces¬

sary to cause a fusion, a blending, an interpenetration of their

masses. Again, if the water in B were replaced by a liquid in

which neither ether nor chloroform are appreciably soluble,

nothing would take place. Further if the chloroform A be re¬

placed by a liquid which does not dissolve water and has less

attraction for ether than has water, nothing will take place.

It is clear then that the process under consideration proceeds

because of the specific nature of the septum B and also that of

the two liquids that bathe it. It should also be emphasized in

this connection that while it is essential, as stated above, that

the layer B, the water, must be capable of dissolving C, the

ether, it is further necessary that this solubility be restricted

in character, as it is in fact, otherwise the layer B, would not

be distinct from C and would become so rich in ether that the

boundary lines between A and B would also disappear.

Returning now to the original experiment Fig. 1 in which

ether, water and chloroform are in A, B, and C respectively, let

us imagine the aqueous layer B as quasi solid and also immov¬

able, i. e. attached firmly to the sides of the glass tube, never¬

theless otherwise retaining its original properties. This would

make A a compartment whose volume remains fixed, and as the

ether enters it from B, for reasons already detailed, a hydro¬

static pressure would be produced upon the sides, top and bot¬

tom of A; and as this pressure develops, it would become more

and more difficult for additional ether to enter this compart¬

ment. Finally, if the walls of this compartment did not give

way, a point of maximum pressure would be reached. At this

point, at the temperature of the experiment, the affinity of the

i If olive oil be dissolved in ether and this solution placed in C,

whereas water and chloroform are retained in B and A respectively,

the ether would pass through the water into the chloroform leaving

the olive oil behind in C. We have here an illustration of what may

be called selective action on part of the membrane (the aqueous sep¬

tum B) in which property biologists are particularly interested. The

explanation is obvious. Water dissolves ether appreciably, but not

olive oil, so the latter is left behind.

218 Wisconsin Academy of Sciences , Arts and Letters.

ether — chloroform solution in A for additional ether is insuf¬

ficient to extract further ether from the aqueous layer B. In

other words, the affinity of water for ether aided by the hydro¬

static pressure developed in A (which militates against the

ether passing into A) just balances the affinity of the chloro¬

form — ether solution in A for additional ether. And yet the

ease is„ after, all not quite so simple, for it must, be remembered

that chloroform is somewhat soluble in water saturated with

ether at atmospheric pressure; now this solubility is increased

with increase of pressure, so that as the hydrostatic pressure

in compartment A increases due to the influx of ether, the out¬

flow of chloroform is continually slightly increasing, which

tends, of course, to relieve the pressure. The actual final max¬

imum pressure reached is therefore determined by the relative

influx of ether into compartment A and the outflow of chloro¬

form from that compartment. When this outflow is practically

nil, or at any rate very slight, we should be dealing with a so-

called semi-permeable membrane; when the outflow is not a neg¬

ligible quantity, as in the case of most septa, the final maximum

pressure attained is materially influenced thereby.

Now, it is easy enough to imagine the aqueous layer B firmly

held in place as we have done, but to realize this experimen¬

tally presents great difficulties. It may, however, readily be

demonstrated that such hydrostatic pressure is actually pro¬

duced, without to be sure making an attempt to furnish any¬

thing more than a qualitative proof that such pressure is really

formed. The apparatus used for this purpose is shown in Fig.

3. In this glass tube, D represents mercury, A, chloroform,

C, ether, and B is a slice of an excellent piece of cork which

had been kept under boiling water for some time so as to drive

the air out of it and thoroughly inject it with water; during

this process much soluble material was also extracted from the

cork. The cork thus thoroughly soaked with water was firmly

pushed into position. It is, of course, somewhat difficult to

avoid having air bubbles just above and below the cork, but

by careful manipulation it is possible to secure the arrangement

as shown in Fig. 3. The chloroform and ether used in this case

were each first saturated with water, S9 that they would not

Kohlenberg — Osmosis and Osmotic Pressure. 219

unduly rob the cork of its water content, which was necessarily

small enough to begin with. On allowing the apparatus to

stand, the mercury rises in the tube as indicated, showing that

pressure is produced on the walls of A. The mercury rose un¬

til the cork either broke or began to slip upward, the experi¬

ment being repeated three times. Using a cork not soaked in

water no pressure was obtained. It is quite probable that a

cork might be so fastened in place by the aid of mechanical

contrivances that an ordinary tube would give way before the

cork would move ; but since the maximum pressure could not

be measured in this manner, the qualitative demonstration of

the presence of the pressure was deemed sufficient. The ex¬

periment just described may be performed also in the form il¬

lustrated in Fig. 4, which is more like the usual arrangement

employed in osmotic experiments. In this figure, the liquid B

in the bottom of the bottle is ether; the end of the glass tube

is closed with a tight fitting slice of cork A saturated wtih water ;

in the tube above the cork is chloroform, C ; and above the ehlo-

220 Wisconsin Academy of Sciences, Arts and Letters .

reform is water, D. At E there is a cork collar which holds

the tube in place. This collar does not fit perfectly tight, yet

it minimizes the evaporation of the ether. The cork E is rather

loosely inserted; its purpose is simply to prevent undue evap¬

oration. This was also the purpose of the layer of water, D,

which to be sure also kept the chloroform saturated with water.

Both the ether and chloroform used were saturated with water

to begin with as in the previous case. With this arrangement

the layer C increased in bulk, the column rising until the layer

D touched the cork E. The experiment was continued for two

weeks, the cork at A being rather thick caused the change to

progress slowly.1

This case of ether, water, chloroform, Fig. 1, has been thus

described in detail because it illustrates all the essential points

to be taken into consideration in the study of the osmotic pro¬

cess. We have seen what conditions are necessary for the

process to proceed at all, what are the various factors that

modify the rate of the change, and what causes the so-called

osmotic pressure. It has further been shown under what con¬

ditions we get a so-called semi-permeable membrane, and it is

evident that an absolutely semi-permeable membrane exists only

in theory. On the basis of his elaborate experimental re¬

searches, Quincke2 has also arrived at this conclusion. He

stoutly contends that there really is no such a thing as a semi-

permeable membrane and that a theory which postulates such

can not be maintained. His words are, ‘ ‘ Ich bestreite, dass eine

halb durchlassige Miembran existirt, Hamit fallt aber auch der

osmotische Druck, dessen Theorie die Existenz einer halbdureh-

lassigen Membran voraussetzt. As stated above, there is in

general always an outflow from an osmotic cell as well as an

inflow, and when the former is extremely slight as compared

with the latter, the membrane is called semi-permeable ; but com¬

monly the outflow is quite sufficient to demand consideration.

At any rate, the osmotic pressure is always the resultant of in¬

flow and outflow caused by the attractions that come into play,

i Compare in this connection a similar experiment of Nernst, Zeit.

Phys. Chem. 6, 35 (1890).

2Drude’s Ann. 7, 682 (1902).

KaJilenberg — Osmosis and Osmotic Pressure. 221

which attractions are to the mind of the writer the same as

what is commonly called chemical affinity, and consequently

such so-called osmotic pressures may under suitable conditions

be very great indeed, while under other conditions they may

be quite small. They would, moreover, present considerable

variation according to the nature of the substances employed.

No originality is claimed for the ether, water, chloroform ex¬

periment, Fig. 1. It occurred to me after a goodly number

of the experiments described below had been performed; but in

looking over the earlier articles of the rather voluminous litera¬

ture on the subject of osmosis, I found that in 1854 L’Hermite

(Compt. rend. 39,1177) described the experiment in question.

His statements concerning its import and bearing are very

clear, though, of course, he does not speak of semi-permeable

membranes and osmotic pressures, for these concepts were at

that time quite unknown. A reference to the article of

L’Hermite is also made in the bibliography in Lehmann’s Mo-

lekularphysik.

However, after having L’Hermite ’s experiment clearly before

me, the principles it illustrates soon became the guide in future

experimentation, for they enabled me to foretell whether a

membrane would be permeable or impermeable for a certain sub¬

stance under given conditons; and if permeable for several

substances, which of these would go through most readily. 1

was thus enabled to forecast in which direction the main os¬

motic current would go. These matters were not only of con¬

sequence in direct measurements of osmotic pressures, detailed

below, but they were also of importance in dialysis as will ap¬

pear farther on. Again, by demonstrating the nature of the

process, it clearly appeared that in determining the maximum

osmotic pressure the contents of the osmotic cell must be con¬

tinually stirred. For, taking the arrangement as given in Fig.

1 and again imagining the aqueous layer B as quasi solid and

firmly fixed in place, it is evident that as ether is drawn into

compartment A, the liquid in that compartment becomes en¬

riched with ether just where it touches B, and that consequently

the osmotic pressure set up in A is not as great as it would be

if the liquid in A were continually stirred so as to rapidly dis-

222 Wisconsin Academy of Sciences , Arts and Letters.

tribute the incoming ether throughout the cell and present a

fresh surface to the layer B. When stirring is omitted, this

work of keeping the contents of B of uniform concentration has

to be performed by diffusion, a process which is very slow, and

consequently a lower pressure is obtained. It is clear that for

similar reasons the contents of B ought in general also to be

stirred in attempting to measure the maximum osmotic pres¬

sure, as should also the liquid C. But when B is quite thin,

as is the case with some membranes, stirring its contents is of

less consequence. Furthermore, when practically very little of

A enters C (i. e. when B is a so-called semi-permeable septum)

it is obviously not so necessary to stir the contents of C in

osmotic pressure measurements. But the liquid in the osmotic-

cell must always be stirred, though it is recognized that in

some forms of cells this is far more imperative than in others.

Attempts to make direct measurements of osmotic pressures-

without stirring the contents of the cell are comparable with

attempts to make a saturated solution of a salt by placing an

excess of it in a dish, pouring the solvent upon it and allowing

the whole to stand without agitation. The method of stirring

the contents of the osmotic cell will be described below in con¬

nection with the quantitative measurements of osmotic pres¬

sures.

QUALITATIVE EXPERIMENTS.

The following experiments which are largely of a qualitative

nature were performed in the course of the investigation in

order to determine the influence which the character of the

membrane has upon the permeability of the latter, and also

upon the main direction of the osmotic current. Unless other¬

wise stated, the osmometer used was a very simple one, con¬

sisting of an ordinary thistle tube, the mouth of which was

closed with the membrane employed. The latter was tied on

with thread, in which process great care was used to secure a

perfect contact between the glass and the membrane. To ac¬

complish this a thistle tube was selected with a flare of about

forty-five degrees at the mouth. On such a tube the mem-

Kohlenberg — Osmosis and Osmotic Pressure. 223

branes used could be perfectly securely tied. Very many

evenly applied turns of a stout, moderately fine, white cotton

thread were wound on after the membrane had been care¬

fully placed in position over the mouth of the tube. This part

of the work required much time, patience and perseverance,

as well as skill and practice. After thus securing the mem¬

brane over the mouth of the thistle tube, the latter was filled

with one of the liquids to a point slightly above the bell of the

tube, which was then immersed in a beaker containing the

other liquid. The level of the two liquids was, of course, the

same to begin with.

For convenience in future reference the experiments per¬

formed have been numbered. The liquid in the bell of the

osmometer will always be called the inner or inside liquid; that

in the beaker, the outer or outside liquid.

1 ) The membrane consisted of sheet rubber, vulcanized

caoutchouc, such as is used by dentists in making their “ rubber

dam. 77 This rubber was very elastic, being almost the pure

gum. It could readily be stretched over the osmometer and

tied on so as to form a perfect joint, for all folds could be

avoided. The inside liquid was a 20 per cent aqueous cane

sugar solution ; the outside liquid was pure water. The ex¬

periment ran 48 hours. No change took place. The liquids

remained at the same level and no sugar passed through the

membrane into the water on the outside.

2) A rubber membrane was used as before. The inside

liquid consisted of a 20 per cent aqueous NaCl solution, the

outside liquid of pure water. No change took place, the

liquids remaining on the same level and not a trace of salt

appearing in the water without. When the outside water was

replaced with toluene, the latter caused the rubber to swell,

but no salt or water passed into the toluene, neither did the

toluene ~ go into the brine. On afterward exposing the mem¬

brane to the air, it lost its toluene content and shrank to its

former size, remaining throughout quite elastic.

Experiments 1 and 2 show that it is possible to get the joint

between rubber and glass perfectly tight. It is perfectly clear

that no change took place when water and the aqueous solu-

224 Wisconsin Academy of Sciences , Aids and Letters.

tion touched the rubber, for the reason that there is no affinity

between these liquids and rubber. Thus .'the first condition

necessary for osmosis to take place was lacking, namely, that

the membrane must be able to dissolve (i. e. imbibe or take up)

one of the liquids that bathe it. When in the second part of

experiment 2 the water was replaced by toluene, the latter was

taken up by the membrane; but osmosis did not take place

because the liquid on the other side, the brine, having no af¬

finity for toluene to speak of was unable to extract toluene from

the rubber.

3) The inside liquid was water, the outside liquid 99.5 per

cent alcohol, the membrane rubber. The liquid rose in the

osmometer showing the main current to be from the alcohol

through the rubber to the water. This is what one would ex¬

pect for the alcohol is taken up by the rubber and then the

water by virtue of its affinity for alcohol extracts the latter

from the rubber.

4) This experiment was like No. 3 except that a 20 per cent

aqueous cane sugar solution was used as the inner liquid.

Again the liquid rose in the osmometer, alcohol passing through

the rubber into the sugar solution. No sugar, however, was

found in the alcohol on the outside. The rise in the osmometer

tube was slow, being about five centimeters in five days. The

inside diameter of the stem of the thistle tube was about 3.5mm.

The main direction of the current was, of course, such as

was to have been expected from what was said in connection

with the preceding experiment. No sugar was found in the

alcohol outside for sugar has so little affinity for alcohol, for

rubber and for rubber soaked with alcohol that it is not taken

up by any of these.

5) The inner liquid was pyridine, the outer liquid water

and the membrane parchment. The liquid rose in the osmo¬

meter showing the main current to be from the water through

the parchment to the pyridine. Some pyridine was also found

in the water.

6) The experiment was like No. 5, except that a rubber

membrane was used instead of parchment and the water was the

inner liquid and the pyridine the outer. The liquid again rose

Kohlenberg — Osmosis and Osmotic Pressure. 225

in the osmometer showing that the main current was in the

opposite direction as in the preceding experiment,, namely from

the pyridine through the rubber to the water. Some water also

passed into the pyridine, showing an appreciable minor current.

7 ) The inside liquid was pyridine, the outside liquid toluene

and the membrane parchment. No change took place.

8) This experiment was like No. 7 except that rubber was

substituted for parchment. The liquid at once rose in the os¬

mometer, showing the main current to be from the toluene

through the rubber to the pyridine. Some pyridine, however,

also passed into the toluene.

Rubber readily takes up pyridine, and imbibes toluene still

more readily; on the other hand parchment does not, showing

that it has but little affinity for these liquids. It is conse¬

quently easy to see why no change occurred in No. 7. When it

is further remembered that pyridine and water are consolute

liquids, as are also toluene and pyridine the observations in

Nos. 5, 6, and 8 are easily explained. In No. 5 the parchment

imbibes water which is then extracted by the pyridine; but

since pyridine is soluble in water soaked parchment, some pyri¬

dine also passes into the water outside. In No. 6 the rubber

imbibes the pyridine which is then extracted by the water; but

as water is somewhat soluble in pyridine soaked rubber, some

of it makes its way into the pyridine without. In No. 8 the

fact that toluene is imbibed more readily by rubber than is

pyridine again determines the direction of the main current,

though in this case, owing to the fact that pyridine has con¬

siderable affinity for rubber as well as for toluene, the minor

current is of considerable consequence. The cases just con¬

sidered well illustrate how the nature of the septum and of

the liquids that bathe it determines what will actually take

place.

In 1898, G. Flusin1 used carbon disulphide, chloroform, tol¬

uene, ether, benzene, xylene, petroleum ether, benzyl chloride,

turpentine, petroleum, nitrobenzene, methyl alcohol, ethyl al¬

cohol, and acetic acid, taking these liquids in all possible com-

iCompt. rend. 126, 1497 (1898); ibid. 131, 1308 (1900).

15

220 Wisconsin Academy of Sciences , Arts and LeUers.

binations in pairs and separating them from each other by

means of a membrane of vulcanized caoutchouc. He found the

main current to be from the liquid which is the more readily

imbibed by the rubber, through the septem to the less readily

imbibed liquid. Again in 1900 1 he used water, methyl alcohol,

amyl alcohol, amyl acetate, chloroform, benzene, ether, and

ethyl alcohol. He employed hog’s bladder as a septem placing

one side of it in contact with ethyl alcohol (the liquid which,

of those named, is according to him imbibed least readily) and

bathing the other side with each of the other liquids success¬

ively. He always found the main current to be in the direction

toward the ethyl alcohol and the rate of flow to vary with the

amounts of liquid imbibed by the membrane during the first five

minutes, which amounts were, of course, determined by independ¬

ent experiments. I have confirmed all of the results of Flusin

where he used rubber membranes. He says nothing, however,

about the minor current,, which I found to be present in all of

these cases to a greater or lesser extent. In other words, the rub¬

ber was traversed by both liquids of each pair, though the main

direction of the current was quite correctly determined. Flusin

shows that the affinity between membrane and liquid is to be

measured by the rate with which the latter is imbibed by the

former, and not by the total amount of liquid taken up by a

given quantity of membrane at the end of a long time, as

Tammann2 contends. Raoult3 separated methyl alcohol and

ether from each other by means of rubber. He always found

the direction of the main current to be from the ether

through the rubber to the methyl alcohol; and the direction

of the main current remained the same, even when the

ether was considerably diluted with methyl alcohol. When

he substituted a membrane of hog’s bladder for the rub¬

ber, the direction of the main current was reversed, it being

from the methyl alcohol through the septum to the ether. In

his article, Raoult has omitted to say anything about the fact

that in his experiments there is also a minor current in a di-

1 Compt. rend. 126, (1497) 1898; ibid. 131, 1308 (1900).

2 Zeit. Phys. Chem. 22, 491 (1897).

sCompt. Rend. 21, 187 (1895). Zeit. Phys. Chem. 17, 737 (1895).

Kahlenberg — Osmosis and Osmotic Pressure. 227

rection opposite to that of the major; in other words, that when

caoutchouc is the membrane and ether is going into the alcohol

there is also some alcohol passing into the ether; and when

the bladder is employed, there is some ether passing into the

alcohol, though the main current is from the alcohol to the

ether. Raoult was, however, perfectly clear in his own mind

with regard to this matter as will appear from a quotation from

a letter written by him printed below. |

It is hardly necessary to add that in the light of what has

already been said above, the results obtained by Raoult and

Flusin are exactly such as might have been foreseen.1

9) The inner liquid was glacial acetic acid, the outer liquid

distilled water and the membrane rubber. Within five minutes

blue litmus paper placed in the water turned red, showing that

the acid was passing through the rubber. This was the direc¬

tion of the main current, which might have been expected*,

since glacial acetic acid is more readily absorbed by rubber

than is water.

10) The inner liquid was a 10 per cent solution of acetic

acid in water, the outer liquid water, and the membrane rubber.

After 1.5 hours, the blue litmus in the water began to turn red,

indicating that in this case, too, the acid was passing through the

rubber into the water. The direction of the main current then

was not altered by diluting the acetic acid considerably with

water, which result is in line with what Raoult found in the

ether — methyl alcohol experiment referred to above.

11) The inner liquid was a strong solution of trichloracetic

acid in water, the outer liquid water and the septum rubber.

But very little acid passed through the rubber into the water.

Hardly any change took place in the 20 hours during which

the experiment was continued.

12) The inside liquid was a solution of trichloracetic acid

in benzene, the outside liquid water and the membrane rubber.

In this case acid went into the water in very considerable quan¬

tity. This was the direction of the main current. The ex¬

periment was run for 20 hours.

i Compare also the remarks by Tammann, Zeit. Phys. Chem. 22],

490 (1897).

228 Wisconsin Academy of Sciences , Arts and Letters.

13) The inner liquid was a solution of trichloracetic acid

in water to which considerable benzene had been added, the

outer liquid was water and the septum rubber. Very notable

quantities of acid passed through the membrane into the water

without. The experiment was continued for 20 hours.

The large increase in the amount of trichloracetic acid which

passed through the rubber in 12 as compared with 11 is easily

explained by the fact that benzene has considerable affinity for

trichloracetic acid and also for rubber. In imbibing the ben¬

zene, therefore, large amounts of trichloracetic acid are also ab¬

sorbed with that hydrocarbon by the rubber. When we dis¬

solve trichloracetic acid in benzene we really unite the acid

with the benzene.1 In this homogeneous liquid, the solution,

we have then the acid tied to the benzene; and because of the

great affinity of the latter for rubber, the benzene solution of

trichloracetic acid, the combination of the acid and the ben¬

zene, also has considerable affinity for rubber : and so the acid

is also drawn into the rubber because it is united with benzene.

To be sure, benzene and acid are not taken up by the rubber

in the same proportions in which they occur in the solution,

benzene being taken up in relatively larger amounts. This

means that the solution is to a certain extent decomposed, i. e.

altered in concentration, by the rubber. When the latter soaked

with benzene and the acid is in contact with water, the great

affinity between water and the acid again comes into play and

by virtue of it acid passes into the water, leaving the benzene

behind in the rubber almost completely. Similar consider¬

ations hold in No. 13 where a solution of trichloracetic acid,

water and benzene is separated from water by a rubber septum,

except that here the acid passes through in lesser quantity, and

minor amounts of water are undoubtedly also absorbed by the

rubber, since in the solution the water is tied on to the benzene

and the acid.

14) The inner liquid consisted of pure water, the outer of

0.1 normal solution of AgN03 in pyridine, and the membrane

of rubber. The liquid rose rapidly in the osmometer, reach-

i Compare, Kahlenberg, liber das Problem des Losungen, Chem. Zeit-

ung 29, No. 81, (1905).

Kohlenberg — Osmosis and Osmotic Pressure. 229

ing a height of 25 cm. in 4 hours. The experiment was then

stopped and the contents of the cell examined. It was found

that pyridine had passed into the water, but no appreciable

quantity of nitrate of silver. The entire content of the osmotic

cell was evaporated to dryness, the residue extracted with a

little water and a drop of nitric acid, and the filtered solution

tested with HC1 for silver, but none was found.

15) The inner liquid was a 0.1 normal AgNOs solution in

pyridine, the outer liquid pure water, and the membrane parch¬

ment. The main current was from the water, through the

parchment to the AgN03 solution in pyridine, for the level of

the liquid in the osmometer rose to a height of about 18 cm. in

20 hours. Water then goes into the cell, forming the main cur¬

rent, but both AgN03 and pyridine were also found in the

water outside, showing a considerable counter current toward

the outer liquid.

16) The inside liquid consisted of 0.1 normal AgN03 solu¬

tion in water, the outside liquid of 0.1 normal AgN03 solution

in pyridine, the membrane being rubber. The main current was

from the outer liquid to the inner one, the level rising in the

stem of the osmometer. During the night the liquid filled the

tube (which was about 32 cm. long) entirely and ran over.

17) The outside liquid consisted of toluene, the inside

liquid of 0.1 normal solution of AgN03 in pyridine, and the

membrane of rubber. The liquid rose 22 cm. in the osmometer

in 17 hours. The main current consists of the passage of tol¬

uene through the septum to the pyridine solution of the AgN03 ;

however, pyridine was also found in the toluene outside as

were mere traces of silver nitrate.

18) This experiment was identical with No. 17 except that

a parchment septum was employed. It was continued for 4

days and no change was observed. It seemed as though a slight

lowering of the level of the inner liquid might have taken place,

so the contents of the cell were evaporated to dryness, the res¬

idue taken up with water and a drop of nitric acid and tested

with HC1 for silver, after having been filtered. Only the faint¬

est indication of the presence of silver was thus obtained.

230 Wisconsin Academy of Sciences , Arts and Letters.

19) The inside liquid was 0.1 normal solution of AgN03

in pyridine, the outside liquid pyridine and the septum rubber.

The liquid rose slowly in the osmometer, indicating the direc¬

tion of the main current to be from the pyridine through the

rubber to the solution. An examination of the outer liquid

showed the presence of only a very small amount of nitrate of

silver. The experiment was repeated using a much heavier

piece or ordinary gray sheet india rubber as a septum, with the

same result. And again, it was repeated using the rubber dam

as a membrane once more, but supporting it by tying over the

outside of it a piece of muslin. With this arrangement the

liquid rose in the stem of the thistle tube to a height of 28.5

cm. in 18 days, remaining there constant for two days, and

then receding slightly. The temperature was very nearly 17

degrees throughout the test. The outer liquid was found to

contain appreciable amounts of AgN03, but hardly an estima¬

ble quantity. This shows that vulcanized caoutchouc is prac¬

tically impermeable for AgN03 under the conditions described;

in other words that it is a “semi-permeable” membrane.

20) The inner liquid was a 0.05 normal solution of AgN03

in pyridine, the outer pyridine, and the septum rubber. At

17° C. no change whatever was observed after 6 days. After

19 days a rise of about 0.5 cm. of the liquid in the stem of the

osmometer was observed. The outer liquid was then analyzed

for silver nitrate. Only traces were found, not an estimable

quantity. This shows that the membrane did not leak and

that the so-called osmotic pressure of a AgN03 solution in

pyridine which is 0.05 normal is practically nil at 17° C. when

vulcanized caoutchouc is used as the semi-permeable membrane.

21) The inner liquid was a normal solution of AgN03 in

pyridine, the outer liquid pyridine, and the membrane rubber.

The apparatus was kept at a temperature which varied grad¬

ually between — 16° and — 15° C. At the end of the second

day the liquid in the osmometer had risen to a height of 7.2

cm., the temperature being — 16.° At the end of the third

day the height of the column was 15.6 cm., the temperature

being _ 15.° The membrane was intact and but traces of

silver were present in the outer liquids.

Kahlenberg — Osmosis and Osmotic Pressure. 231

22) The inner liquid was normal solution of AgNOs in

pyridine, the outer liquid pyridine, and the membrane rubber.

In this ease a piece of common vulcanized rubber about 1 mm.

thick such as is used on footpower laboratory bellows was em¬

ployed. It was supported on each side by pieces of muslin and

perforated steel discs, and the whole was then securely screwed

to the lower end of an osmotic cell made of steel. The maxi¬

mum pressure which was read on a closed manometer, using

mercury between the air space and the inner liquid was 14.95

atmospheres at 20° C. The membrane did not “leak,” which

was evident from the fact that only mere traces of AgNOs

were found in the outer liquid,, though the experiment was run

for two weeks, the pressure remaining practically constant for

five days. This steel osmotic cell consumed considerable time

in its construction, and many difficulties had to be overcome in

perfecting it and attaching the manometer to it. However, it

is unnecessary to enter into a detailed description of the cell,

since no special significance will be attached to the single re¬

sult recorded here, it being given simply to show that a very

considerable pressure may be produced by a normal solution of

AgNOs in pyridine when it is separated from pure pyridine

by vulcanized caoutchouc at 20°, whereas at — 16° (No. 21)

the pressure formed is practically insignificant.

Silver nitrate is insoluble in hydrocarbons, which shows that

the affinity between that salt and hydrocarbons is slight. Now

as caoutchouc is a hydrocarbon substance, the affinity between

it and silver nitrate would be slight. Pyridine has consider¬

able affinity for silver nitrate. It dissolves the salt readily

with evolution of a considerable amount of heat. Pyridine is

soluble in all proportions in hydrocarbons — is consequently

readily imbibed by rubber. When a silver nitrate solution in

pyridine is placed in contact with rubber, the latter soaks up

pyridine, but also some silver nitrate with it, since the pyridine

and the salt are bound to each other by mutual attraction.

This accounts for the fact that traces of silver nitrate pass

into the pyridine when it is separated from a silver nitrate so¬

lution in pyridine by means of a rubber septum as in experi¬

ments 19, 20, 21 and 22, though the main current is that of the

232 Wisconsin Academy of Sciences , Arts and Letters.

passage of pyridine through the membrane to the solution.

Thus the reason why caoutchouc is a “semipermeable” mem¬

brane in these cases is given; and we should expect in all

cases in which the solute employed is insoluble in hydrocarbons,

like kerosene, benzene, etc., yet is soluble in pyridine, that vul¬

canized rubber will act as a “semipermeable” membrane when

it is employed in separating pyridine from the pyridine solu¬

tions of such solutes. This has been confirmed in the case of

cane sugar and lithium chloride which are soluble in pyridine

yet insoluble in hydrocarbons. The experiments are given be¬

low. Conversely, when a substance is soluble in hydrocarbons

as well as in pyridine, that substance will always pass through

vulcanized caoutchouc in notable quantities when its solution

in pyridine is separated from pure pyridine by means of the

caoutchouc septum. Examples of such cases will also be found

below. Though experiments 19 to 22 are only quasi quantita¬

tive in character they are already quite sufficient to show that

here the osmotic pressure does not follow the gas laws at all.

The change of the pressure with temperature is very much

greater than proportional to the absolute temperature; and

again the pressure varies much more rapidly with change of

concentration of the solute than is required by Boyle’s law.

Experiment 20 reveals the fact that at room temperature the

osmotic pressure of the 0.05 normal solution of AgNG3 in pyri¬

dine is practically nil under the conditions described, while on

the basis of the vant’ Hoff theory the osmotic pressure of this

solution ought to be over an atmosphere. We have in No. 20

the case where the solution has been diluted to such a point

that its affinity for additional pyridine is practically equal to

the affinity between pyridine and the rubber, so that the latter

can not be robbed of its pyridine content by the solution, and

consequently the liquid in the osmotic cell does not increase in

bulk.

The observations made in Nos. 14 and 16 are such as might

have been foreseen considering the fact that water has practi¬

cally no affinity for rubber; that the latter has considerable

affinity for pyridine ; that water and pyridine are consolute

liquids; and that silver nitrate, though soluble in water and

Kohlenberg — Osmosis and Osmotic Pressure. 233

pyridine,, is yet not soluble in hydrocarbons and consequently

has practically no affinity for rubber. Further, recalling that

water is readily taken up by parchment, and pyridine and

toluene not, and considering these facts in connection with

those already mentioned, the data obtained in Nos. 15, 17 and

18 are readily explained.

23) The inside liquid was a 7.06 per cent cane sugar solu¬

tion in pyridine, the outer liquid pyridine and the membrane

rubber, supported by muslin tied over it. The liquid rose in

the osmometer to a height of 22.7 cm. in 4 days, the height

after the second day being 21.8 cm. Sugar had not passed into

the outer liquid except in mere traces.

24) This experiment was like No. 23 except that xylene was

used as the outer liquid instead of pyridine. The liquid rose

to a height of 16.4 cm. in 4 days. Sugar was not present in

the outer liquid in appreciable quantities, but pyridine was.

25) The inside liquid was a 1.2 per cent solution of sugar

in pyridine, the outer liquid pyridine and the membrane rub¬

ber supported by muslin. The temperature was kept at 22.5°

C. A rise of the liquid in the osmometer tube was noted after

five minutes. After 3 hrs. the column measured 4 cm. ; after

12 hrs. about 5 cm. Sugar did not pass through the septum

in appreciable amounts. The experiment shows that the solu¬

tion used is able to produce but a very feeble osmotic pressure.

26) The inner liquid was a 0.125 normal solution of sugar

in pyridine, the outer liquid pyridine and the septum rubber

supported by muslin. The whole was kept at 0° C. After 3

days the liquid had risen only 0.5 cm. in the osmometer tube;

after 5 days, the rise was but 1.9 cm. The membrane was in¬

tact, and practically no sugar had passed into the pyridine

without.

27) This experiment was like No. 26 except that a 0.25 nor¬

mal solution of sugar was used as the inner liquid. The tem¬

perature was kept at very nearly — 16° C. During the first

day the liquid rose to 9.5 cm. in the osmometer; on the second

day the column measured 13.5 cm. : and on the fourth day 17.5

cm. The membrane was intact and only very slight amounts

of sugar were present in the outer liquid. The same experi-

234 Wisconsin Academy of Sciences, Arts and Letters.

ment performed at room temperatures showed a very rapid rise

of liquid in the osmometer tube, — see pressures measured in

the quantitative measurements described below.

28) The inner liquid was a saturated solution of Li Cl in

pyridine, the outer liquid pyridine and the septum rubber.

The liquid rose in the osmometer and Li Cl did not pass

through the septum into the outer liquid in appreciable quan¬

tity.

Experiments 23 to 27 show that vulcanized caoutchouc is a

‘ ‘ semipermeable ’ ’ membrane when it separates sugar solution

in pyridine from pure pyridine. The data lead one further to

the conclusion that the gas laws do not govern the phenomena,

which the measurements to be detailed later confirm. No. 28

shows that when Li Cl is used as solute in an otherwise similar

experiment, the rubber again acts as a semipermeable septum.

The reasons for this behavior have already been discussed in

connection with the AgN03 solutions.

29) The inside liquid was 0.1 normal AgN03 in pyridine,

the outside liquid a saturated solution of cane sugar in pyri¬

dine, the membrane being rubber. The level in the osmometer

fell, showing the current to be from the AgN03 solution to that

of the sugar. On further examination it was found that

AgN03 had also passed into the sugar solution but in small

amount.

30) This experiment was identical with No. 29 except that

parchment was used as the septum. The test was continued

for 3 days. No change was observed.

In the light of what has already been said, it is clear that

the results in Nos. 29 and 30 are such as might have been an¬

ticipated.

31) The inner liquid was a solution of FeCl3 in toluene, the

outer toluene, the membrane rubber. The main direction of

flow is toward the solution, but FeCl3 also passes through the

membrane in considerable quantity, which was to have been

expected since this salt is soluble in hydrocarbons. The FeCls

gradually disintegrates the septum.

32) This experiment was identical with that of No. 31 ex¬

cept that iodine was used as the solute instead of FeCl3.

KaMenherg — Osmosis and Osmotic Pressure. 235

The iodine also passed through the rubber, as was to have been

expected; and it disintegrated the septum more rapidly than

did the FeCl3. The liquid, however, first rose in the osmom¬

eter.

33) The inner liquid was a solution of copper oleate in

benzene, the outer benzene and the septum rubber. The liquid

in the osmometer rose to a height of 20 cm. showing the main

current to be toward the solution; however, large amounts of

copper oleate passed through the rubber into the outer ben¬

zene. This was to have been expected,, since copper oleate is

soluble in hydrocarbons.

34) This experiment was identical with No. 33 except that

parchment was employed as the septum in place of rubber.

No change whatever took place, which was to have been an¬

ticipated since none of the ingredients touching the membrane

have sufficient affinity for it.

35) This experiment was like No. 33, except that the copper

oleate was dissolved in pyridine, and pyridine was used as the

outer liquid. Again the main current was toward the solution,

but copper oleate passed into the outer pyridine in considerable

quantity, which is quite in harmony with the theory advanced.

When we think of a large molecule like that of copper oleate

readily travelling through vulcanized caoutchouc as in No. 35,

and that under like conditions cane sugar, AgNOs and LiCl do

not pass through that septum, it certainly must convince us

that the membrane does not act as a sieve. Again, No. 34 shows

that parchment is not a “porous” material as is so commonly

assumed from osmotic experiments with aqueous solutions in

which it is employed as septum.

36) The inner liquid was a strong solution of sodium oleate

in water, the outer liquid water, and the septum rubber. No

change occurred.

37) The experiment was like 36 except that parchment was

employed as the septum. In this case sodium oleate was found

in the outer water, though the inner liquid showed slight in¬

crease in bulk.

Sodium oleate though soluble in water is insoluble in hydro¬

carbons; bearing this fact in mind, the results in Nos. 36 and

37 are readily explained.

236 Wisconsin Academy of Sciences , Arts and Letters.

38) The inner liquid Consisted of a normal solution of

naphthalene in pyridine, the outer liquid was pyridine and the

membrane rubber. Practically no change in level occurred, but

large quantities of naphthalene passed into the pyridine with¬

out.

39) The inner liquid was a normal solution of camphor

in pyridine, the outer pyridine and the septum rubber. Prac¬

tically no change in level occurred, but considerable amounts

of camphor appeared in the outer liquid.

40) The inner liquid was a saturated solution of camphor

in 99.5 per cent alcohol, the outer liquid 99.5 per cent alcohol

and the septum rubber. After half an hour a slight lower¬

ing of the level of the liquid in the osmometer was noticed;

it continued to go down for 3 days, when the experiment was

stopped. Much camphor had passed into the alcohol without.

41) The inner liquid was a saturated solution of camphor

in toluene, the outer liquid toluene and the membrane rubber.

The liquid at once rose rapidly in the osmometer. In 9 hours

it reached a height of 32 cm. It continued to rise for three

days, the duration of the experiment. Large quantities of

camphor had passed through the septum into the outer toluene..

Camphor is very soluble in toluene.

That naphthalene and camphor should pass through rubber

was to have been expected, Nos. 38 to 41, since these substances

are very soluble in hydrocarbons. In No. 40 the main current

is from the solution of camphor in alcohol to the alcohol; while

in 41 the main current is from the toluene to the camphor

solution in toluene. This occasions no surprise when it is

borne in mind that toluene is imbibed much more rapidly and

more copiously by rubber than is alcohol,, and that an alcoholic

solution of camphor is imbibed by rubber more rapidly than

is alcohol. Again, remembering that it is the relative rate of

inflow and outflow which determines whether the bulk of the

liquid in the osmotic cell will change or not under given

conditions, the results in 38 and 39 are readily explained.

42) In the osmometer was placed a solid block of camphor

in form of a cube which weighed two grams; the outer liquid

was 99.5 per cent alcohol, and the membrane rubber. The os-

Kohlenberg — Osmosis omd Osmotic Pressure. 237

mometer was immersed in the liquid so that the membrane was

slightly below (about 0.5 cm.) the surface of the liquid. The

arrangement is shown in Fig. 5. After five minutes there

seemed the least evidence that the camphor was being attacked,

but even after 4 hours there was but little further change. Since

alcohol is not very readily imbibed by rubber this occasioned no

surprise.

43) This experiment was identical with No. 42, except that

toluene was used instead of alcohol. Plain evidence that the

camphor was dissolving appeared in three minutes. In forty

minutes the solid camphor had all disappeared and was found

in very large quantities in the outer liquid. The membrane be¬

came much distended, bulging downward. The level of the

liquid was alike in and outside after four hours, it having risen

slightly on the inside.

44) This experiment was like No. 42 except that xylene was

employed instead of alcohol. The observations were practically

identical with those of No. 43 where toluene was employed,

which might have been expected.

45) This experiment was also like No. 42 except that pyri¬

dine was employed instead of alcohol. After four minutes it

was evident that the camphor was being attacked. A very thin

238 Wisconsin Academy of Sciences , Arts and Letters .

layer of liquid, less than 1 mm. thick, appeared on the upper

side of the rubber. In four hours all the camphor had gone*

through the rubber into the pyridine, leaving a depth of solu¬

tion of only about 1 mm. on the upper side of the membrane.

During the process the membrane became somewhat distended

and bulged upward toward the camphor.

46) In this case the arrangement was again like in No.

42, except that CC14 was employed instead of alcohol. After

about four minutes it was evident that the block of camphor

was beginning to disappear. In forty-three minutes the whole

block was gone, having passed through the rubber into the CC14,

leaving on the upper side of the membrane a camphor solution

less than a millimeter deep. The membrane became distended

and bulged upward. After four hours, the liquid on the upper

side of the membrane was from 1 to 2 mm. deep.

47) This experiment was also like No. 42 except that CS2

was used as the liquid instead of alcohol. After four minutes

it was clearly evident that the block of camphor was being at¬

tacked. After 13 minutes a thin layer of liquid was visible

on the upper side of the rubber. After 45 minutes all solid

camphor had disappeared, having passed through the rubber

into the CS2, leaving on the upper side of the membrane a solu¬

tion about 3 to 4 mm. deep. After four hours the membrane

appeared very much distended, bulging downward, the liquids

in and outside being on a level.

48) In this case a cube of paraffine of rather high melting

point (70s) was separated from 99.5 per cent alcohol by means

of a rubber septum, the arrangement being as described in No.

42. After 24 hours the lower comers and edges of the par¬

affine cube appeared slightly rounded. No liquid was visible

on the upper side of the rubber, however. After three days

the large bulk of the paraffine was still intact, though it was

distinctly evident that the substance was slowly passing through

the rubber septum into the alcohol beneath, without any* liquid

layer appearing on the upper side of the membrane.

49) This experiment was identical with the preceding one,

No. 48, except that toluene was used instead of alcohol. In

this case solution gradually accumulated on the upper side of

Kohlenberg — Osmosis and Osmotic Pressure. 239

the rubber, though paraffine passed through into the toluene

on the outside. After 24 hours only about one-eighth of the

paraffine was left undissolved and the liquid on the inside was

practically on a level with that on the outside. Much paraffine

was found in the outer liquid.

50) In this case the cube of paraffine was separated from

pyridine by means of rubber, the arrangement being as de¬

scribed in connection with No. 42. The observations were

practically the same as when alcohol was employed, No. 48.

51) The experiment was identical with the preceding one,

No. 50, .except that CC14 was used instead of pyridine. After

24 hours about three-fourths of the paraffine was still left in

the solid state, it floated on the layer of liquid which had formed

on the upper side of the rubber. Considerable paraffine had

gone through the septum into the CC14 below. The membrane

was much enlarged and bulged downward. After two days,

all the paraffine had dissolved and the liquids in and outside

were nearly on a level.

52) About two grams of dry powdered A.gN03 was placed

on the upper side of a rubber membrane the lower side being

bathed by pyridine. The arrangement was as pictured in Fig.

5, the powdered AgN03, which was spread out over the sur¬

face of the membrane,, taking the pla.ce of the block represented

in the figure. The experiment then was the same as No. 45 ex¬

cept that the AgN03 was used in place of camphor. After 7

hours no liquid was visible on the upper side of the membrane.

Pyridine had, however, passed into the silver nitrate and formed

a solid addition product with it. No appreciable amount of

AgN03 had passed into the pyridine without. After 24 hours

all the AgNG3 had dissolved and the liquid in the osmometer

had risen to a height of about 1.5 cm. and continued to rise.

After four days the experiment was stopped and the outer

liquid examined for AgN03. It was, however, found to be

present only in mere traces. This shows that vulcanized caout¬

chouc is indeed a good “semi-permeable” membrane in this

case.

Considering experiments Nos. 42 to 52 in which solids were

separated from liquids by means of rubber septa, it is evident

240 Wisconsin Academy of Sciences , Arts and Letters .

that whether the solid will pass through the membrane or not

depends on the nature of the solid, the membrane and the

liquid employed. Furthermore, if the substance composing the

solid does make its way through the septum, the fact as to

whether the action is accompanied with an accumulation of

liquid on that side of the septum occupied by the solid or not

is clearly determined by the rate with which the solution formed

is absorbed by the membrane (which is determined by the

mutual attraction or affinity of the saturated solution and the

membrane for each other) and also by the rate with which the

pure solvent is imbibed from the other side.

Now it is well known that camphor dissolves very readily in

hydrocarbons and consequently has considerable affinity for

rubber, and we should, therefore, expect it to pass through the

latter when it is employed as an osmotic membrane. An alco¬

holic solution of camphor is more readily imbibed by rubber

than is pure alcohol, and so it occasions no surprise when in No.

42 we find the block of camphor slowly making its way through

the rubber septum into the alcohol, and without the appearance

of liquid on the side occupied by the camphor. The action is

slow because alcohol is not imbibed rapidly or copiously and

because the septum holds on to the strong camphor solution

very tenaciously, so that only a small portion of the camphor

thus saturating the rubber is washed out of the latter by the

alcohol on the other side.

When the same experiment is performed using a hydrocarbon

as the liquid instead of alcohol, Nos. 43 and 44, the action goes

on very much more rapidly on account of the great affinity be¬

tween rubber and the hydrocarbon or a camphor solution in a

hydrocarbon. The strong camphor solution is so greedily ab¬

sorbed by the rubber that but little liquid appears on the upper

side of the membrane while the solid camphor lasts, and it is

as though the solid camphor were passing through the rubber

septum by mere contact with it. The process reminds one

strikingly of the manner in which solid food placed in the ali¬

mentary canal is digested and absorbed. Here the presence of

the food in contact with the walls of the tract excites the flow

of the digestive juices toward the solids, the latter are acted

Kohlenberg — Osmosis and Osmotic Pressure. 241

upon and the resulting liquid is absorbed by the walls of the

canal. In our experiments the camphor in contact with the

rubber saturated with hydrocarbon excites a flowr of hydro¬

carbon towrard it. The hydrocarbon acts on the camphor

“dissolving” it, and the resulting solution is then so rapidly

absorbed by the septum as to leave no liquid on the side oc¬

cupied by the camphor.

When as in No. 45 pyridine is used instead of a hydrocarbon,

the camphor is still absorbed, but not as rapidly; for pyridine

is not taken up by rubber as readily or as copiously as are

hydrocarbons, neither has camphor so great an affinity for

pyridine as for hydrocarbons. On the other hand, camphor

very rapidly disappears when CC14 or CS2 (Nos. 46 and 47)

are employed, for these substances have great affinity for rub¬

ber and also for camphor.

When paraffine instead of camphor is separated from alcohol,

toluene, pyridine or carbon tetrachloride by means of rubber,

Nos. 48 to 51, the action is in all cases slower, which fact is

readily comprehended when it is borne in mind that camphor

dissolves more readily in the liquids named than does paraffine.

Finally in No. 52, where solid silver nitrate is separated from

pyridine by rubber, we have an illustration of a typical case

in which the liquid accumulates very greatly on the upper side

of the membrane. Here the solute has very little affinity for

the rubber, and so the solution of the salt practically does not

get into the septum on account of the fact that pure pyridine

is more readily imbibed and silver nitrate is difficultly soluble in

rubber soaked with pyridine, that is to say, in a hydrocarbon di¬

luted wTith pyridine. If a block of cane sugar or one of chloride

of lithium were separated from pyridine by means of a rubber

septum the action would be similar to that observed in the

case of silver nitrate, No. 52. The same would be true if any

solid which is soluble in pyridine yet not soluble in hydro¬

carbons were separated from pure pyridine by means of a

rubber membrane. It would be interesting to test this in the

case of more substances which are insoluble in hydrocarbons

and yet are soluble in pyridine, but the number of such sub¬

stances is rather limited, for it must be remembered that pyri-

16

242 Wisconsin Academy of Sciences , Arts and Letters.

dine is itself nearly a hydrocarbon in character and is eon-

solnte with hydrocarbons.

QUALITATIVE TESTS WITH ANIMAL AND VEGETABLE MEMBRANES.

A number of qualitative tests were made with membranes

of organic origin. It was thought best to place the results of

these experiments on record here, though it is contemplated

to take up the whole matter of the action of organic mem¬

branes separately at some later time.

Various animal membranes were tested as to their perme¬

ability for sodium chloride, urea, and cane sugar. The mem¬

branes used were stretched over square wooden frames quite

loosely so as to form bags or pockets into which one liquid was

placed; these bags were then suspended in the outer liquids, in

such a manner that only the lower sagging part, and not the

upper edges, came into contact with the outer liquids. As

membranes to be tested were selected the pericardium, dia¬

phragm, small intestine, large intestine, stomach, aorta, urinary

bladder and gall cyst of a young ox. The animal was perfectly

sound and normal and about three years old. The membranes

were used soon after the animal had been killed. On the up¬

per side of each membrane was placed an aqueous solution con¬

taining 23.4 grams of sodium chloride, 40 grams of urea and

342 grams of cane sugar in 2000 cc. ; while the lower side of

the membrane was immersed in pure water, the arrangement

being as already described. It was found that in case of all

of the membranes mentioned the sodium chloride passed

through more rapidly than urea and sugar, which fact was

established by examining the outer liquid from time to time

during the first four hours. After twenty-four hours besides

much salt very considerable quantities of urea and cane sugar

had gone through all of the membranes. A special test made

by separating water from an aqueous sodium sulphate solution

by means of the stomach membrane, showed that the latter is

permeated but slightly by sodium sulphate; for the amounts

of the latter salt in the outer liquid were but small.

KaMenberg — Osmosis and Osmotic Pressure. 243

The vegetable septa employed consisted of the rinds of Cali¬

fornia oranges and grape frnit and Florida grape fruit. The

grape fruit, also called shaddock, is the fruit of Citrus

decumana. In each case the fruit was carefully cut in two

transversely with a sharp knife, so that the halves after careful

removal of the pulp formed two cups. Each half rind was

filled to about half of its capacity with the liquid to be tested

and then suspended in pure water so that the latter was about

on a level with the liquid in the rind. The suspension of the

rind was accomplished by means of fine aluminum wire run

through small holes pricked through the upper edges of the

rind. Each experiment was continued for twenty-four hours,

unless otherwise stated. It was found to be impracticable to

continue the experiments much longer, for the rinds after be¬

ing immersed in water for a considerable time undergo alter¬

ation, becoming soft and loosing their waxy outer coating.

With the arrangement as just described and employing water

as the outer liquid and an aqueous solution as the inner liquid,

it was found that sodium chloride readily passes through

orange skins, but sodium sulphate very much less readily. In¬

deed, only traces of the latter salt were found in the outer

liquid even when strong solutions were employed. Again, urea

readily goes through the orange rinds, but sugar passes through

only in very small amounts. When an aqueous solution of

potassium alum wTas used it was found that a slight amount

of potassium sulphate appeared in the outer water, but no alu¬

minum sulphate, showing that the latter ingredient is left be¬

hind during the time of the experiment at least.

In the upper half of the rind of a Florida grape fruit was

placed 100 cc. of a solution containing 10 grams sodium chloride

plus 10 grams of sodium sulphate. The rind so charged was

suspended in distilled water as above described. The outer

water had a volume of 500 cc. Tests of the outer liquid showed

that the NaCl was passing through much faster than the

Na2S04. This experiment was run for seven days; the rind

did not seem to be altered much.

The lower half of the rind of a California grape fruit con¬

taining 70 cc. of an aqueous solution, which contained 10 grams

244 Wisconsin Academy of Sciences , Arts and Letters .

of NaCl plus 10 grams of sugar in lOOcc., was suspended in a

dish in 300 cc. of water. It was found that the sugar passed

through the rind much more slowly than the salt. The experi¬

ment was continued for seven days.

A similar experiment in which 80 cc. of an aqueous solution

containing 10 grams sugar plus 10 grams urea in 100 cc. was

placed in the upper half of the rind of a California grape

fruit suspended in a dish containing 400 cc. water, yielded the

result that both urea and sugar pass through the rind, but the

latter more slowly. After twenty-four hours sugar may be de¬

tected in the outer liquid with Fehling’s solution. The urea

acts on the rind thickening and hardening it. Much urea is

thus retained in the rind, also considerable amounts of sugar.

These facts were determined by an examination made after the

experiment had run for seven days.

To 70 cc. of a saturated solution of boric acid in water 10

grams of sugar were added. This solution was placed in the

upper half of the rind of a California grape fruit, the outside be¬

ing bathed by water. The experiment was continued for five

days, the rind remaining intact and practically unaltered during

this time. Only traces of boric acid passed through the rind.

Sugar passed through slowly ; but at the end of five days it was

found in the outer liquid in considerable quantity.

It was found further that in dilute aqueous solutions H2S04,

HC1, and HNOs readily pass through the skins of grape fruit;

but the rinds are much altered by the acids, appearing shrunken

and darkened in color. Citric acid passes through less rapidly

than the mineral acids mentioned. It is apparently retained to

a considerable extent in the rind.

The experiments made with the rinds of grape fruit were

also repeated with the skins of California oranges, with prac¬

tically the same results.

QUANTITATIVE MEASUREMENTS OF OSMOTIC PRESSURES.

The quantitative measurements of osmotic pressures were

made with so-called semi-permeable membranes, that is to say

with membranes through which the solvent* passes so much more

readily than the solute, that the amount of the latter which

Kohlenberg — Osmosis and Osmotic Pressure. 245

goes through the septum is practically a negligible quantity.

From what has been said above, it appears that the latter quan¬

tity is never absolutely nil, and that consequently there is really

no such a thing as a semi-permeable membrane, strictly speaking.

Now as has been intimated, the qualitative experiments above

detailed enable one to foretell when a membrane will permit

the solvent to go through so much more readily than the solute

that the amount of the latter which traverses the membrane

is so slight that the septum may be called semi-permeable.

From the ether, water,, chloroform experiment, already de¬

scribed as a typical case of osmosis, it appears that the ether

makes its way through the water into the chloroform because

(1) ether is soluble in water and (2) chloroform has much

more affinity for ether than has water, so that ether is extracted

from the water layer by the chloroform ; on the other hand, the

water does not permit chloroform to pass into it and into the

ether beyond to an appreciable extent, because chloroform is so

very slightly soluble in water even when the latter is impreg¬

nated with ether.

Holding these things in mind let us look for the proper solv¬

ent and solute to employ with rubber as the membrane so that

the latter shall be seminermeable. It must first be remembered

that rubber (vulcanized caoutchouc) is practically a hydro¬

carbon. The rubber employed was of excellent quality and was

almost the pure gum. On analysis it was found to contain 0.38

per cent ash, and a Carius determination yielded 0.30 per cent

chlorine and 0.95 per cent sulphur. Before the analysis was

made the sample was wiped superficially, washed with distilled

water, dried with filter paper and finally left in a desiccator

over strong sulphuric acid for twenty-four hours. It is evi¬

dent that in order to pass through a rubber membrane a sub¬

stance must be taken up by the rubber, the rubber must imbibe

the substance, in other words the substance must be soluble in

rubber.1 Again the liquid bathing the other side of the rubber

must be capable of extracting the imbibed substance from the

iThe act of such solution or imbibition is really mutual; i. e. the

rubber attracts the substance in question, and the latter in turn at¬

tracts the rubber.

240 Wisconsin Academy of Sciences, Arts and Letters.

rubber, thus completing the transference. On the other hand,

those substances which are not soluble in rubber, i. e. are not

taken up by the rubber, will obviously not pass through the

latter. And so what is required is a solvent which will read¬

ily be taken up by rubber, without, however, disintegrating the

same, and a solution of such a character that the solute shall

not be soluble in rubber. The less affinity the solute has for

rubber the better ; for then when the solution is brought into

contact with the rubber the latter will imbibe practically only

solvent to the exclusion of solute, thus leaving the solution

slightly more concentrated, that is to say making a partial sepa¬

ration of solvent and solute.

In casting about for a suitable solvent very many substances

were tried. It was of course desirable to secure a solvent that

was not too volatile at ordinary temperatures or too obnoxious.

In the course of this work it soon became evident that water,

alcohols and in general compounds containing considerable hy¬

droxyl, relatively speaking, are not suitable for they are not

taken up readily enough by rubber nor in sufficient quantity.

On the other hand many compounds, like hydrocarbons, their

halogen substitution products, carbon disulphide, ether, etc.,

though taken up readily by rubber and in considerable quantity

gradually act upon the latter to such an extent as to form with

it a very soft, sticky mass or even a liquid, a combination or

solution then of such mechanical properties as to be entirely

unsuitable for quantitative osmotic experiments. After trying

a large number of liquids, pyridine was finally taken as being

the most suitable for the purpose in hand. From its very na¬

ture pyridine is a substance which would perhaps be expected

to fulfill the requirements. Its high carbon and hydrogen con¬

tent make pyridine almost a hydrocarbon, indeed, it dissolves

in hydrocarbons in all proportions. Since rubber is a hydro¬

carbon, we should expect it to imbibe pyridine readily. It was

found that pyridine is imbibed by rubber, increasing the bulk

of the latter somewhat to be sure, but without otherwise mater¬

ially altering the mechanical properties of the rubber. Indeed,

I found that vulcanized caoutchouc may even be boiled in pyri¬

dine for hours, in which case there is a slight amount of ma-

Kahlenberg — Osmosis and Osmotic Pressure. 247

terial extracted from the rubber, giving the liquid a brownish

color; but the rubber is not disintegrated or affected materially

otherwise. The rubber used as membranes in the actual quanti¬

tative measurements was in fact thus extracted with boiling

hot pyridine. After such treatment rubber when dried has

practically all of its original properties; it is perhaps a little

easier to rupture it by stretching it hard. It might at first be

somewhat surprising that pyridine does not disintegrate rubber

more. However, while pyridine is closely akin to hydrocarbons

and consolute with them, it must be borne in mind that water,

which has but little affinity for hydrocarbons, also is consolute

with pyridine. Pyridine is then a rather unique substance, and

it is hardly surprising that it should be imbibed by rubber

sufficiently for the purpose in hand without unduly disintegrat¬

ing it. Direct experiment showed that at room temperature

(about 20° C.) 100 grams of the vulcanized caoutchouc used

imbibed 144.42 grams of pyridine in 24 hours, while the amount

imbibed in 17 days was 145.17 grams.

A suitable solute would be one that is soluble in pyridine yet

insoluble in hydrocarbons. For instance, a substance soluble in

petroleum or benzene would in general also be soluble in pyri¬

dine, but it would also be soluble in rubber (a hydrocarbon)

and hence would pass through the latter. Thus oleic acid, the

oleates of the heavy metals, ferric chloride, naphthalene, cam¬

phor are soluble in hydrocarbons, are consequently readily

taken up by rubber, and when dissolved in pyridine pass

through the rubber when the latter in an osmotic experiment

separates the solution from pure pyridine. Because of the pe¬

culiar nature of pyridine already alluded to above, this liquid

dissolves a goodly number of substances which are insoluble in

hydrocarbons. The solubility of such substances in pyridine is

to be sure rather limited as to quantity as a rule. So cane

sugar, silver nitrate, lithium chloride are insoluble in hydrocar¬

bons and yet reasonably soluble in pyridine. One would con¬

sequently expect that when solutions of either of these sub¬

stances in pyridine be separated from pure pyridine by means

of a rubber septum, practically none of the solute would pass

through the rubber; in other words, the latter would be prac-

248 Wisconsin Academy of Sciences , Arts and Letters.

tieally impermeable for sugar, silver nitrate and lithium chlor¬

ide. Now this is actually what was found in the qualitative

tests described above. These substances pass through rubber

in extremely slight quantities which are quite comparable with

the amounts of cane sugar that pass through the much studied

tests described abover These substances pass through rubber

an aqueous sugar solution. However, on account of the fact

that sugar has a high carbon and hydrogen content, one would

expect it to have more affinity for a hydrocarbon than either

silver nitrate or lithium chloride, and that consequently it

would pass through rubber a little more readily than these salts.

Experiment has also shown that this is actually the case;

though as stated above, the amount of sugar which passes

through the rubber membrane is quite small.

The quantitative measurements of osmotic pressures were

then made by using solutions of cane sugar, lithium chloride,

and silver nitrate in pyridine, these solutions being separated

in each case from pure pyridine by means of a membrane of

vulcanized caoutchouc previously treated with boiling hot pyri¬

dine so as to extract any soluble ingredients. It was not the

purpose of the quantitative measurements of osmotic pressures

to produce and measure enormous pressures; though as was

shown above in No. 22 a pressure of approximately fifteen at¬

mospheres was actually measured in the case of a normal solu¬

tion of silver nitrate in pyridine. The efforts were rather di¬

rected toward determining with a sufficient degree of accuracy

moderate pressures, using different concentrations of the solu¬

tions employed at several different temperatures.

The osmotic cells were made entirely of glass, excepting of

course the surface actually closed by the membrane itself. The

different parts of the cells were fused together so as to form

one piece, thus avoiding cemented joints of any kind. Figure

6 shows how these cells were made. To a stout, carefully made

thistle tube having a flare of about 45 degrees at E, a T was

attached, the tube being provided with a bulb and bent as shown

in the figure, C. To C was fused a manometer tube D having

a bore of about 0.5 mm. ; this tube was made as long as the ex¬

periment required. The small bulb and bent part of the tube

Kohlenberg — Osmosis and Osmotic Pressure.

249

C contained mercury. At R the tube was somewhat contracted

as shown, and after putting the required amount of pure clean

mercury into the apparatus and filling the rest of it with the

solution to be tested through the orifice F by means of a capil¬

lary funnel tube, the apparatus was carefully heated at B by

means of a small flame and finally drawn off leaving the whole

securely sealed. With practice this part of the operation may

be so performed that practically no air bubble remains in the

apparatus after the tube has been drawn off at B.

H

Fig. 6.

Before placing the mercury and the solution into the cell,

however, the membrane must be securely put into place. It is

first of all essential that the opening of the bell of the thistle

tube E be as nearly circular as possible, and that the points on

the outer edge of the orifice lie in very nearly the same plane.

The rubber membrane consisted of a high quality of sheet rub-

250 Wisconsin Academy of Sciences , Arts and Letters.

her as is used by dentists. Its ash, sulphur and chlorine con¬

tent have already been given above, and it has also been stated

that the material was extracted with hot pyridine before using

it. The rubber so prepared was carefully tied over the orifice

at E. In this process it was stretched only moderately so as

not to thin it unduly.1 In tying the membrane on, the rubber

was always stretched to such an extent, however, that no folds

whatever remained where it touched the glass. With proper

care it is possible to get a smooth surface of rubber to rest

snugly against the surface of the glass. The rubber was finally

securely fastened into place by carefully winding the whole

surface from H to E smoothly and closely with moderately fine

thread. When these precautions are observed one hardly ever

fails to get a perfectly tight joint between the glass and the

rubber, which is so all essential.

The next step consisted of supporting the membrane so that

it would withstand pressure. To accomplish this there was

first tied over the membrane a piece of smooth, soft yet strong

muslin. This cloth was drawn tightly over the membrane, and

securely tied on by winding moderately line yet strong thread

over it smoothly and evenly from H to E. It is of course im¬

possible to get rid of folds in the cloth where it is tied on the

surface H to E ; but it is quite possible to distribute these folds

fairly evenly around the circumference, in such a manner that

the cloth where it actually touches the rubber on the mouth of

the thistle tube and just at the edge at E, lies perfectly smoothly

and snugly against the surface of the membrane giving it

proper support. After this a circular perforated disc of stout

sheet steel made of proper size so as to just cover the lower end

of the thistle tube was placed on the muslin covering the rub¬

ber; and this disc was then securely held in place by tying

over it very firmly another piece of strong muslin by means

of stout thread evenly and tightly wound on as in the previous

cases. Thus there was the rubber tied on the glass, then the

muslin snugly covering and supporting the rubber, then the

steel disc pressing against and supporting the muslin, and fi-

iWhen finally in place, the thickness of the membrane was only a

small fraction of a millimeter.

Kohlenberg — Osmosis and Osmotic Pressure . 251

nally the outer layer of muslin firmly holding the disc in the

proper position. The perforations in the disc were about 1 mm.

in diameter and the disc itself was about 0.5 mm. thick. By

observing closely the precautions here laid down, cells were

practically always obtained without fail which were perfectly

tight, rigid and capable of withstanding pressures up to the

point of bursting the glass bell of the thistle tube. Further¬

more, on taking such an apparatus apart even after experi¬

ments that had continued for weeks, the membrane was found

to be intact and in perfect condition. Experiments demon¬

strating that such membranes do not leak have already been

given above, ISTos. 1 and 2. We have here then a simple, direct

and certain method of preparing so-called semipermeable mem¬

branes for osmotic pressure measurements, which are yet not

precipitated membranes: and furthermore we know why the

membranes are semipermeable for the solutions that come into

question.

After the membrane has been put into place as described the

apparatus is charged with the required amount of mercury and

then with the solution to be tested, the tube being drawn off at

B before the small flame of a blowpipe so as to leave no air in

the apparatus. The latter is then placed in a large vessel con¬

taining the pure solvent, and the rise of the liquid in the

manometer tube is observed from time to time, a pair of sensi¬

tive thermometers graduated to 0.1 degree being placed in the

solvent near the osmometer to indicate the temperature. The

apparatus was set up in a basement room, whose temperature

changed but slightly and only very gradually during the course

of the experiments. It is highly essential that the temperature

be kept as constant as possible during the experiments and that

sudden fluctuations be avoided. From what has been said it is

evident that the experiments were finally set up in much the

same way in which Pfeifer performed his tests:, only in this

case additional care was taken to cover up the dish so as to

minimize evaporation and to protect the surface of the pyri¬

dine from the moisture of the air. All measurements were

made with the open manometer and the height of the mercury

column was estimated to within 1 mm. in each case. It was

252 Wisconsin Academy of Sciences , Arts and Letters .

found to be quite needless to measure more closely than this,

for the results of separate duplicate experiments showed as a

rule greater variation.

I will first give the results of the experiments performed by

using the method just described, which method, though it is the

usual one, had to be modified later in one important particular

in order to secure reliable results. The individual experiments

are numbered consecutively with those preceding so as to

facilitate future reference to them.

53) The liquid in the osmometer was a solution of cane

sugar in pyridine containing one-fourth of a gram-molecule

per liter of the solution. The outer liquid was pure pyridine.

The experiment was run in triplicate, that is three separate in¬

dividual tests were made with the same solution. The temper¬

ature in each case was 17.5° C. a) In the first of these ex¬

periments the pressure came up slowly, remaining nearly con¬

stant after the second day. The experiment was nevertheless

allowed to go on for nine days when the pressure was finally

measured and found to be 186.2 cm. at 17.5° C. Before discon¬

tinuing the experiment, the whole of the outer vessel was

packed in melting ice for ten hours. The mercury column be¬

came constant at 125 cm. after three hours and remained there

for the remaining seven hours:. The temperature was then per¬

mitted to rise slowly and after three days the mercury column

was 159.5 cm. high, the tempreature being 16.4°. When the

whole had finally reached the temperature 17.5°, the mercury

column was again 186.0 cm. high. This then would seem to be

pretty good evidence that the maximum pressure had indeed

been reached. During the operation care was taken not to dis¬

turb the apparatus by jarring it in any way. The outer liquid

was found to contain traces of sugar, but the exact amount

was not determined, b) In the second independent yet per¬

fectly similar experiment, the pressure also rose gradually,

changing but slightly after the second day. The temperature

was kept at 17.5° C., and at the end of eleven days the column

of mercury measured 155.6 cm. This experiment was then al¬

lowed to run for four days longer. At 8 in the morning of the

fourteenth day the mercury column measured 166.4 cm. at 16.4°

Kohlenberg — Osmosis and Osmotic Pressure. 253

€. ; while at 5 in the afternoon of the fifteenth day it measured

166.7 cm. at 16.0° C. As in the previous case, care was taken

not to jar the apparatus during the experiment. In this case,

too, traces of sugar were found in the outer liquid, c) In a

third similar experiment the pressure also increased but little

after the second day. The experiment was stopped at the end

of four days in this case, when the mercury column was 107.4

cm. high, the temperature being 17.5° 0. Only traces of sugar

were found by testing the outer liquid.

It will be noted that the highest pressure was observed in (a).

In this experiment the pressure came up more rapidly than in

the other twro cases, accomplishing 115 cm. in the first twenty-

four hours. The discussion of the causes of the discrepancies

in the results of (a), (b) and (c) will be left until a little

later.

54) This experiment was performed in duplicate. The ar¬

rangement was exactly like that in No. 53 except that the

liquid in the osmometer consisted of a solution of cane sugar in

pyridine containing one-eighth of a gram-molecule per liter of

the solution, a) The pressure rose slowTy increasing but

slightly after forty-eight hours. On the sixth day the mercury

column measured 62.4 cm., the temperature being 17.5° C.

The whole was then surrounded with melting ice, and after two

hours the mercury column measured 0.2 cm. ; at the end of ten

hours the pressure was 0.6 cm. The temperature then gradu¬

ally rose during the night as the ice disappeared. At 14.5° C.

the column measured 48.8 cm. and at 17.5° C. it again came up

to 62.0 cm., nearly where it was before the chilling process.

In this case the experiment was left set up for five days longer,

when the column measured but 43.6 cm. at 18° C. Sugar was

found to be present in the outer liquid in small amount, b)

This was a duplicate of (a). The observations made were prac¬

tically the same as in (a) except that the pressures were differ¬

ent. The chilling process with ice was omitted in this case.

After two days the pressure increased but slightly. On the

third day it was measured carefully and found to be 52.8 cm.

at 17.5° C. During the next two clays the temperature fell

very' gradually to 14.5° C. and the column then measured 12.5

254 Wisconsin Academy of Sciences , Arts and Letters.

cm. The temperature then rose gradually to 18° C. during the

succeeding three days, when the pressure was 46.3 cm.

55) In this experiment the inner liquid was a solution of

cane sugar in pyridine containing 0.25 gram-molecule per liter

of solution, while the outer liquid was a solution of cane sugar

in pyridine containng 0J25 gram-molecule per liter of solu¬

tion. The pressure rose slowly, the mercury column reaching a

height of 137.5 cm. at 16.8° C. By far the most of this was

accomplished during the first two days. The final measure¬

ment just given was taken after thirty days. The experiment

was then left undisturbed for twenty-six days longer. The

pressure began to diminish gradually and finally measured only

89 cm., when the experiment was discontinued. The membrane

when examined at the end of this time was to all appearances

intact and but slightly changed, if at all.

56) Two experiments were made with solutions of cane

sugar containing 0.25 gram-molecule of cane sugar per liter of

solution in pyridine as the inner liquid, and pure pyridine as

the outer liquid, employing in this case, however,, a common

thick sheet rubber1 — such as is used on the ordinary foot-power

laboratory bellows — in place of the finer vulcanized caoutchouc

used in the other experiments. The experiments were con¬

ducted side by side as duplicates of each other. The pressure

rose gradually, the mercury column reaching a height of 43.0

cm. in one case and 39,5 cm. in the other at 22° C. The rubber

was found to be much softened when examined at the end of

ten days, the duration of the experiments. The maximum pres¬

sure was practically reached after fort}’- eight hours. The

pressure diminished gradually after seven days. Small amounts

of sugar were found in the outer liquid, though the exact

amount was not determined.

From the results just detailed it is evident that the methods

employed are not capable of yielding concordant values.

Though only attempts to measure osmotic pressures of sugar

solutions in pyridine have been described, similar experiments

were made using lithium chloride and silver nitrate solutions

in pyridine. In these cases the results were no more; concordant

than in those in which sugar was solute. At first it was thought

Kohlenberg — Osmosis and Osmotic Pressure. 255

that the difficulty lay in the membrane itself, the initial experi¬

ments having been performed by using the sheet rubber with¬

out further treatment; but actual tests showed that whether

the rubber was previously extracted with boiling hot pyridine

or not made no perceptible difference in the results. It will be

noted that one set of experiments was made using a common

thick sheet rubber (No. 56). In this set the results were very

different from those obtained with the thin rubber of high

grade (No. 53), and yet the duplicates did not differ from each

other more than when thin rubber was used. It was also de¬

termined by several trials that the non-concordance of the

results of duplicate experiments could not be laid to the fact

that in some cases the rubber was stretched rather more than

in others. It was not to be expected, of course, that the max¬

imum pressure would be reached in the same time in two ex¬

periments that were similar, for the areas of the surfaces of

the membranes and their thickness were not exactly the same.

It might further be possible that the differences in pressure

observed in the experiments that were duplicates of each other

occurred because slightly different amounts bf sugar passed

through the different membranes; in other words, that there

was more leakage of solute in one case than in the other. Such

leakage or lack of semi-permeability would operate to diminish

the osmotic pressure in two ways, (1) by directly letting ma¬

terial out of the cell as the pressure rises, and (2) by increas¬

ing the amount of solute in the outer liquid and so weakening

the cause which creates the pressure. It is quite true that in

all cases small quantities of sugar passed through the mem¬

branes, and that these were, perhaps, not always exactly the

same. The qualitative tests made, however, always showed that

the quantity wteich had passed through was very small and far

below the amount required to produce a noticeable osmotic pres¬

sure with the apparatus employed during the time of the ex¬

periments. It will be recalled that 1.2 per cent solution of

sugar in pyridine yields practically no osmotic pressure (No.

25) and that the same is true of a 0.05 normal solution of silver

nitrate in pyridine at room temperature (No. 20). Further¬

more, when two similarly charged osmometers were set in one

256 Wisconsin Academy of Sciences , Arts and Letters.

and the same outer vessel filled with pyridine, they neverthe¬

less in general failed to indicate identical pressures. All this

led me to the conclusion that the non-concordance of the pres¬

sures measured could not be due to leaks, defects or various

degrees of u semi-permeability ’ ’ of the membranes.1

I confess that I was about at my wits’ end to understand

why such duplicate experiments made as nearly as possible the

same in every way should yet yield results that were not more

concordant. In all, some seventy independent trials had been

made, in which great care had been used to get the experi¬

ments alike, but to no avail. Yet the membranes were not de¬

fective. One day after measuring carefully the height of the

mercury column in one of the experiments which had been

running for two weeks and in which the pressure had changed

inappreciably for several days, I happened to brush against the

apparatus in such a way as to thoroughly jar it without, how¬

ever, upsetting it or otherwise modifying it; in any way. Some

minutes later when I chanced to look at the apparatus, I

noticed that the level of the mercury, which had been prac¬

tically constant, had risen over two centimeters higher. The

apparatus was then shaken repeatedly from time to time, seiz¬

ing it with an insulating cloth so as not to alter the temper¬

ature, with the result that each time the pressure rose some-

i Measurements of the electrical resistance of the membranes, were

also made which further establish this fact. Placing on each side of

the membrane of an osmotic cell, such as was emjployed in the pres¬

sure measurements, a circular platinum electrode about 2.1> cm in

diameter, so that the planes of the electrodes were parallel to that

of the membrane, and that the metallic surfaces were close to the

membrane without actually touching it, the cell was filled with an

aqueous saturated solution of NaCl and the same solution was used as

the outer liquid. The electrodes were thus immersed in the saturated

NaCl solution and the membrane separated them from each other.

The electrical resistance between the plates was measured by means

of the Kohlrausch method; but it was found to be so great that it *

was not possible to estimate it accurately, being upwards of 70,000

ohms. Various rubber membranes tested in this manner gave results

of the samje order of magnitude. When the same experiment was

performed using a parchment membrane the edges of which protruded

so that they did not dip into the brine, the resistance measured was

less than an ohm. Using the same arrangement with a rubber mem¬

brane and a normal solution of silver nitrate in pyridine as the liquid

on each side, the resistance was over 90,000 ohms, and remained the

same for four days, when the experiment was discontinued. A small

hole pricked through the membrane with a needle in the latter case

caused the resistance to drop to 450 ohms.

Kohlenberg — Osmosis and Osmotic Pressure. 257

what higher. The explanation of this was, o?f course, per¬

fectly obvious at once. The pressure is produced by the en¬

trance of solvent into the omsotic cell, consequently right on

the inner side of the membrane the solution becomes more di¬

lute than at other points in the interior of the cell. If the cell

is left undisturbed the very slow processes of diffusion seek to

equalize the strength of the solution in the cell, but clearly

shaking the apparatus, or better yet, stirring the interior con¬

tent of the cell would at once accomplish what it would take

diffusion processes a very long time to do, even though they be

aided somewhat by gravity owing to the form of the osmotic

cell. And it was, moreover, perfectly {evident, too, that [to

shake the osmometer filled as it was with considerable pressure

on, would not thoroughly mix the contents of the osmotic cell.

It is clear that with a more dilute solution in immediate con¬

tact with the inner side of the membrane than is in the rest

of the osmotic cell the maximum osmotic pressure can not be

attained. Furthermore, as in cells slightly different in form

when left perfectly at rest the dilute layer in contact with the

inner side of the membrane would in general not disappear at

the same rate by diffusion and disturbances due to gravity and

difference of density of the layers, concordant results could not

be expected in duplicate experiments, however carefully per¬

formed, without stirring the contents of the osmotic cell. In

all direct measurements of osmotic pressures which have for

their aim the determination of the maximum pressure attainable

in a given case , it is consequently necessary to continually stir

the interior contents of the osmotic cell while the measurement

is being made, in order that the concentration of the contents

of the cell may remain uniform and a layer of maximum con¬

centration be kept in immediate contact with the inner side of

the septum. Again,, since there really is no such thing as a

semi-permeable membrane in the strict sense of the word — which

fact has already been emphasized above — it is evident that some

of the inner content of the cell, be it ever so slight, is making

its way through the membrane into the outer liquid. Thus right

next to the membrane on the side bathed by the outer liquid,

the septum is really not in contact with the pure solvent — the

17

258 Wisconsin Academy of Sciences , Arts and Letters.

outer liquid — but rather with a solution more or less dilute.

When the apparatus is at rest, the slow diffusion processes,

aided by gravity perchance, tend to keep the outer liquid of

uniform concentration; but these agencies clearly can not be

relied upon to keep pure solvent in contact with the outer sur¬

face of the membrane in osmotic pressure measurements. In¬

deed, when the surface of the membrane is not smooth and

presents considerable area to the outer liquid, a film of liquid

is apt to adhere tenaciously to the outer surface of the septum

in spite of the effects of diffusion. In the usual osmotic ex¬

periments using aqueous sugar solutions in a cell made by pre¬

cipitating copper ferrocyanide on the inner side of an unglazed

cup, the slight amount of sugar that passes through very likely

lingers very tenaciously in the pores of the cup just outside of

the actual membrane, forming there a film of solution of such

strength that its effect upon the osmotic pressure is not a neg¬

ligible quantity. At least its effect can not be assumed to be

negligible without further experimental work. Nor would

stirring the outer liquid in such a case as this be apt to remove

the difficulty. In the osmotic cells described above where thin

rubber membranes supported by cloth and perforated steel discs

were used, the effect in question is no doubt less than in the

walls of a porous cup, but it is by no means negligible. Here

the cloth and the disc hinder diffusion, and it is very necessary

to stir the outer liquid thoroughly and continuously in making

the osmotic pressure measurements.

1 would like to emphasize here once more then that it is very

essential to stir the contents of the osmotic cell and also the

outer liquid continuously in any attempt to measure directly

the maximum osmotic pressure that may he produced in a given

case ; and that since in all past experiments this has been en¬

tirely neglected , the results of such experiments can not he con¬

sidered as final and conclusive. In reality, as has been pointed

out above in connection with the ether, water, chloroform ex¬

periment of L’Hermite, the membrane itself ought also to be¬

stirred during the process. This is of course less necessary

when the septum is quite thin than when it is thick. It should

be borne in mind, however, that even stirring the contents of

Kohlenberg — Osmosis and Osmotic Pressure. 259

the cell and the outer liquid simultaneously can never entirely

keep the concentration of the liquid layers in immediate con¬

tact with the two sides of the septum exactly the same as the

concentration of the liquids further away from the membrane;

yet in many cases this may be accomplished with a fair degree

of approximation. Moreover, by using the stirring process, the

osmotic pressures may be measured fairly approximately even

when the amount of material passing out of the cell is not neg¬

ligible; in other words, when the membrane is not semiperme-

able.

The apparatus devised for stirring the contents of the cell

and also the outer liquid simultaneously during the measure¬

ment of osmotic pressures is pictured in Fig. 7 in diagramatic

form. In the figure, the beaker B of a capacity of 1000 cc. or

more, contains the outer liquid. The latter is stirred by means

of the stirrer F, which is moved up and down by the motion

of the crank C. In the experiments performed it was admis¬

sible to make this stirrer of iron, for this is not attacked by pyri¬

dine. F consisted then of a bright, stout iron wire bent in ring

form. Just above the cork closing the beaker, this stirrer F

was jointed so that it would not need so large an opening in

the cork in which to move up and down. A thermometer is

placed in the liquid to register the temperature. The whole

apparatus is set in a constant temperature room, or the beaker

B is immersed in a bath of constant temperature, not shown in

the figure. The osmotic cell S is made exactly as heretofore

described, except that before tying on the membrane, the stirrer

is inserted into the apparatus through the opening in the bell

of the thistle tube. This stirrer consists of a perforated disc

of light sheet iron fastened by riveting and a drop of solder, to

a sufficiently stout, yet flexible, iron wire the upper end of

which carries a lug of soft iron soldered on, as shown in the

figure. In order that the perforated disc at the low~er end of

this stirrer might not pound on the delicate membrane and in¬

jure it, prongs of wire wrere soldered on the main vertical wire

of the stirrer ; and after the latter had been introduced into the

osmotic cell, these prongs were bent outward in such a way that

they would strike the side of the bell of the thistle tube (see

Kohlenberg — Osmosis and Osmotic Pressure. 261

figure) as the stirrer moved downward, thus allowing the per¬

forated disc to come close to the membrane without actually

touching it. These prongs of course do not interfere with the

upward movement of the stirrer. The soft iron lug at the up¬

per end of the stirrer had a longitudinal groove (not shown in

the figure) filed into it to facilitate the introduction of the

liquid into the cell. After the glass part of the osmotic cell had

been blown in one piece, the iron stirrer was introduced and the

prongs bent into the required position; the membrane was then

tied on and supported with cloth and steel disc as heretofore

described. The proper quantity of mercury was then intro¬

duced and finally the cell was filled with the solution. A fine

long funnel tube was employed in introducing the mercury and

the solution into the osmotic cell. The upper end of the osnuv

tic cell, through which the cell had been filled, was then drawn

off before a blowpipe flame, as already described, so that prac¬

tically no air remained in the cell after it was sealed. The os¬

motic cell was mounted in the beaker as indicated in Fig. 7,

the cork closing the cell securely, leaving only a little play for

the stirrer F.

Over the sealed end of the osmotic cell is placed the solenoid

M; and by making and breaking the current in M, the stirrer

in the osmotic cell is moved up and down. By means of the

wires LL the small electric motor G is connected with a num¬

ber of storage batteries; the latter are not shown in the fig¬

ure. The motor G turns the wheel Yf, through the agency of

a belt, and thus operates the crank 0 which moves the stirrer

F. Now the wheel W is made of hard rubber and on it are

fastened two small brass plates PP which are connected with

each other by a wire. Mounted on separate supports, inde¬

pendent of the wheel W, are the two brass brushes DD, which

make contact with PP. These brushes DD are connected with

the source of current and the solenoid M by means of wires as

shown in the figure. As the wheel W turns and the brushes

DD come into contact with the brass plates PP on the wheel,

the current in the solenoid M is established, the soft iron lug is

attracted and the stirrer in the osmotic cell moved upward only

to-be dropped again as soon as W has turned far enough to de-

262 Wisconsin Academy of Sciences , Arts and Letters.

stray the electrical connection.1 Thus both the outer liquid

and that in the cell are stirred continuously. 11 is a resistance

to regulate the current in the solenoid. It was found that in

order to get good steady motion the hard rubber wheel W was

too light. This defect was remedied by boring holes A through

the wheel near its edge all around the circumference, and fill¬

ing these with lead. On the same axis with W was another

wheel (not represented in the figure) of the same size and

weight, which served to balance the apparatus and at the same

time to operate the stirrers of a second osmotic experiment like

that shown in the figure. Thus osmotic experiments could be

performed in duplicate, using but one motor. The arrange¬

ment for making and breaking the current in the solenoids of

course did not require duplication, since the same current could

be sent through both coils in series.

With this new apparatus the osmotic pressures of 0.125 nor¬

mal LiCl solution in pyridine and also of 0.125 solution of cane

sugar in pyridine were measured, using vulcanized caoutchouc

as the membrane and pure pyridine as the outer liquid. The

0.125 normal solutions were selected rather than 0.25 normal,

because the latter yielded a rather higher pressure than could

be conveniently estimated with an open manometer. The pre¬

liminary results (above recorded) obtained with 0.25 normal

sugar solutions without shaking do not indicate this, for the

highest pressure observed was below 200 cm. ; but it was found

that by stirring this pressure could be about doubled.

With the stirring apparatus described. Fig. 7, it, of course,

takes much more time to set up each individual experiment for

the arrangement is more complicated. The necessity of stirring

in osmotic experiments was found out after a long series of

preliminary tests made in the old-fashioned way without stir¬

ring, and so during the time left only a limited number of tests

could be carried out. Working with pyridine, moreover, has a

very depressing effect on the nervous system, it being impossible

iThe arrangement was such that the slight heating effect produced

by the current in passing through the solenoid was negligible, for the

readings of the manometer did not change perceptibly after the cur¬

rent had been turned off and sufficient time allowed for any difference

in tempearture to become equalized.

j Kohlenberg — Osmosis and Osmotic Pressure. 263

to keep1 the air in the laboratory entirely free from it. The

effect appears to be cumulative in character, for one’s system

seems to become more and moire sensitive to the substance.

For this reason I felt constrained not to prolong the work with

pyridine unduly. It will be remembered that I have carried

on experiments with pyridine continuously for about two years,

and off and on for a much longer time. During the investiga¬

tions I have at times found it necessary to counteract the de¬

pressing effects of pyridine by taking small doses of strychnine.

The results of the experiments performed with the stirring

apparatus will now be given. It need hardly be stated again

that the LdCl and the sugar were dry and of a high degree of

purity, as was also the pyridine.

57) The inner liquid was 0.125 normal solution of LiCl in

pyridine and the outer liquid pure pyridine. It required 46 cc.

to fill the cell. The volume of the outer liquid was 600 cc.

The temperature was 19.0,° and the experiment was run for

three days with constant stirring. The maximum height which

the mercury column reached was 51.2 cm. at 19.0° 0, corrected.

This height was practically attained during the first 24 hours.

At the end of the experiment, the whole of the outer liquid was

evaporated to dryness and the LiCl in the residue estimated.

It was found that the 600 cc. of outer liquid contained 0.0130

grams of LiCl, which shows definitely to what extent the solute

has actually passed into the outer liquid.

In the light of what has been said in the precediug pages, it

is evident that a determination of the exact amount of solute

present in the outer liquid at the time when the maximum pres¬

sure is attained is an essential part of all final osmotic pres¬

sure measurements,, yet such estimations have hitherto always

been omitted even in experiments in which great care has been

bestowed upon other details.

58) This experiment was exactly like the the preceding one

(No. 57) except that the outer liquid consisted of 1200 cc. of

pyridine instead of 600 cc. and the temperatures were as in¬

dicated below. The experiment was continued for 27 days, at

the end of which time an examination showed the membrane

264 Wisconsin Academy of Sciences, Arts and Letters.

to be perfectly intact. The entire outer liquid was finally evap¬

orated to dryness and the LiCl determined in the residue. It

was found that the 1200 cc. of the liquid contained but 0.0267

grams of LiCl. The pressure rose slowly, the mercury column

measuring over 50 cm. after the first day. On the sixth day

the mercury column measured 51.5 cm., the temperature being

24.0.° On the twelfth day the mercury column measured 53.2

cm., the temperature being 20.95;° on the fourteenth day, 54.9

cm., the temperature being 21.96;° on the fifteenth day, 54.2

cm., the temperature being 19.6.° The beater containing the

outer liquid was then packed in melting ice. The temperature

sank to 2.0° and was kept there for 12 hours After two hours

of thus cooling, the mercury column wTas only 9.0 cm. high and

remained there without change for the remaining ten hours.

The ice was allowed to melt gradually for the next twelve hours.

When the temperature had reached 21.4° C. the mercury meas¬

ured 56.1 cm., this was on the seventeenth day. During the

next three days the outer beaker was again kept surrounded

with melting ice. It w^as difficult to keep the temperature of

the apparatus perfectly constant during all this time. It grad¬

ually sank to 2.2° and finally to 1.6.° When the temperature

had reached 1.6° the pressure indicated by the manometer was

2.9 cm. and remained constant for six hours. The temperature

was then gradually raised, and at 20.5° the mercury column

reached a height of 53.0 cm. w-here it remained constant for

twelve hours. The outer beaker was then surrounded with a

bath of a temperature of 36° C., the change to lhat temperature

being made gradually; which, of course, was also done in the

preceding cases where the temperature of the bath was rad¬

ically altered. At 36° the height of the mercury column be¬

came constant at 101.0 cm. and remained so for about half a

day, when the temperature of the bath wras gradually raised

to 58.7° and kept there. At this temperature the mercury

column rose to a height of 128.3 cm., where it remained prac¬

tically constant for two hours. The temperature was then

gradually permitted to fall, and at 25.5° C. the column in the

manometer finally registered 51.9 cm. on the last day of the

Kahlenberg — Osmosis and Osmotic Pressure. 265

experiment. It was after all this that the entire outer liquid

was evaporated to dryness and the amount of LiCl in the resi¬

due determined with the result already recorded. The discus¬

sion of the results of this experiment will be deferred until after

the next two experiments have been described.

59) The inner liquid was a solution of sugar in pyridine

containing 0.125 gram-molecule per liter. The outer liquid

was pure pyridine and the septum vulcanized caoutchouc. The

experiment was run with constant stirring or three days, the

temperature being kept constant at 20° C. The maximum

pressure reached was 98.3 cm., nearly the whole of which was

attained during the first day. On concluding the experiment

the whole of the outer liquid consisting of 400 cc. was evapor¬

ated to dryness; the residue was taken up with water, ard the

sugar determined with Fehling’s solution, after inverting with

HC1. It was found that the 400 ce. contained 0.1149 grams of

sugar.

60) This experiment was as nearly as possible a duplicate

of the preceding one (59). On the fifth day the mercury in the

manometer reached a height of 91.8 cm., the temperature having

gradually gone up to 21° C. In this case, too, the pressure

increased but little after the second day. On the morning of

the seventh day the pressure was 94.0 cm., the temperature be¬

ing 21.95° O. ; in the evening of the same day 95.5 cun. at 22.15. °

On the eighth day the outer beaker was packed in melting ice,

the temperature in the beaker being 2.0° C. During this time

the mercury in the manometer registered practically no pres¬

sure, or at best only a few millimeters. The ice was then per¬

mitted to melt gradually, and on the ninth day the pressure

registered 92.0 cm. at 19.4° C. The temperature was then

slowly raised to 46° C., where it was, kept for twelve hours.

The pressure became constant after about three hours at 114.2

cm. and remained there for nine hours. The experiment was

then discontinued. The whole of the outer liquid consisting

of 1200 cc. was then evaporated to dryness, and the sugar was

taken up with water, inverted with HC1 and determined with

Fehling’s solution. The result was that in the 1200 cc. 0.2205

grams of cane sugar were found.

266 Wisconsin Academy of Sciences, Arts and Letters.

Taking into consideration the pressures developed while the

apparatus is kept nearly at a constant temperature, we find

that according to No. 57 the osmotic pressure of the 0.125 nor¬

mal solution of LiCl is 51.2 cm. at 19.0,° and according to No.

58 it is 54.2 cm. at 19.6.° Again, according to No. 59 the os¬

motic pressure of a 0.125 normal sugar solution is 9S.3 cm. at

20.0,° while according to No. 60 it is 91.8 cm. at 21.0.° Even

in these cases the agreement of the duplicates leaves much to be

desired, for the differences in pressure observed can hardly be

due to the slight differences in temperature. They are more

likely due to individual differences in the membranes used, and

to the fact that even with constant stirring, it was not quite

possible to keep the liquids in immediate contact with the mem¬

brane of perfectly uniform concentration on each side. These

difficulties have already been discussed at length above. There

can be no doubt, however, that the results obtained are approx¬

imately the osmotic pressures of the solutions in question under

the conditions of the experiments.

The effect of stirring is clearly shown by a comparison of ex¬

periment No. 54 where the highest osmotic pressure of a 0.125

normal sugar solution was found to be 6'2.4 cm. at 17.5° with¬

out stirring, and the result recorded in No. 59, where with stir¬

ring the osmotic pressure was found to be 98.3 cm. at 20.0.°

Taking the highest values found, namely 54.2 cm. for 0.125

normal LiCl solution at 19.6° and 98.3 cm. for 0.125 normal

sugar solution at 20.0,° we note (1) that the electrolyte gives by

far the lower pressure , and (2) that neither the LiCl nor the

sugar give anywhere near the pressure called for by the gas

laws, according to which a 0.125 normal sugar solution ought to

give a pressure of approximately three atmospheres (228 cm.

Hg) at 20. 5 Further, it will be recalled that a 1.2 per cent

solution of sugar yielded practically no osmotic pressure at 0,°

whereas according to the gas laws it ought to have produced

about 0.8 of an atmosphere. And again, 0.05 normal solution of

AgNOs yielded no osmotic pressure, whereas according to the

gas laws it ought to have produced over an atmosphere.

Further, the changes of pressure above recorded as caused by

changes of temperature are so very much greater than they

Kohlenberg — Osmosis and Osmotic Pressure. 267

would be if they were proportional to the absolute temperature

that additional comment seems quite superfluous. The osmotic

pressures here investigated, then, do not follow the gas laws at

all. Again, the electrolyte LiCl yields a much lower pressure

than the non-electrolyte, sugar, which is exactly the opposite

of what the theory of electrolytic dissociation requires.

OBSERVATIONS CONCERNING DIALYSIS.

When both cane sugar and copper oleate are dissolved to¬

gether in pyridine and the solution is separated from pure pyr¬

idine by means of a vulcanized caoutchouc membrane, the cop¬

per oleate passes through the septum and the sugar remains be¬

hind. This is what one would expect from what has been said

above. But here we have a case where a crystalline body, the

sugar, is separated from a non-crystalline substance, the cop¬

per oleate, by dialysis in which process the non-crystalline or

colloid body passes through the septum and the crystalloid re¬

mains behind. Again, when camphor and sugar are together

dissolved in pyridine and the solution is separated from the solv¬

ent by means of a rubber septum, the camphor passes through

and the sugar remains behind, which might have been antici¬

pated. Here we have a case in which crystalloids are separated

from each other by dialysis.

In fact it is clear that in general any substance which is sol¬

uble in both hydrocarbons and in pyridine may be separated

from sugar by dialysis, when the pyridine solution of it and

the sugar is separated from the pure solvent by a rubber sep¬

tum. In such cases the sugar will always remain behind in

the dialyser and the other substance will pass through whether

it be crystalline or non-crystalline. Any other substance sol¬

uble in pyridine but not soluble in hydrocarbons may, in gen¬

eral, be substituted for sugar. Thus the role of the nature

of the membrane in the process of dialysis is demonstrated.

The current view that crystalloids always pass through mem¬

branes more readily than colloids is evidently untenable, for

it has been shown that just the opposite may occur, and that

even crystalloids may be separated from each other by dialysis

268 Wisconsin Academy of Sciences , Arts and Letters.

when the proper septum is chosen. Whether substances can

be separated by dialysis or not does not depend at all upon

their crystalline or non-crystalline nature as is so commonly sup¬

posed l, but upon their affinity for the septum employed. After

what has been said above, it is evident that stirring the liquids

hastens dialysis.1

Several experiments in addition to those here described have

been performed, but as I hope to continue the experimental

work on dialysis their presentation will be reserved for a later

communication.

GENERAL REMARKS.

The conclusion reached above that the process of osmosis de¬

pends upon selective solubility (in other words, upon the af¬

finities involved) is in agreement with what Overton2 * * * * has found-

in his physiological studies. The importance of the recognition,

of the true nature of osmotic processes for progress in physi¬

ology can hardly be overestimated.

From the time of Dutrochet all investigators who worked on

osmosis experimentally with different membranes and various

liquids have been impressed with the important role of the na¬

ture of the membrane in determining whether osmosis would

occur at all or not, and if so, in what direction. The follow¬

ing quotation taken from the article of LTTermite, cited above,

shows how clearly he had grasped the situation. ‘‘Je pense

avoir demontre par la discussion des experiences de mes de¬

van ciers et des miennes, que Fendosmose n’est point le resultat

d’une force particuliere, mais de Faffmite elle-meme en etendant

1 'acceptation de ce mot a 1 ’attraction capillaire qui en est le

premier degre ’ ’. I should also like to quote the opinion of

Raoult, which he voiced in a letter written January 7, 1897,.

to Prof. W. D. Bancroft, and which, through the latter 7s kind-

i Since the above was written it came to my notice that in 1848

Jolly, (Lieb Ann. 68, 6) refers to a case where he shook the dialyzer

to hasten the process.

2Vierteljahrsschr. d. naturforsch. Ges. in Zurich, 44, 88 (1899). See

also the discussion in chap. 5 of Hober’s Physik. Chem. d. Zelle u.

Gewebe, Leipsig, (1902). Compare also Livingston, The Role of Dif¬

fusion and Osmotic Pressure in Plants, Chicago (1903).

Kahlenherg — Osmosis and Osmotic Pressure. 209

ness I am able to present here. This quotation, which requires

no further comment, gives everything contained in the letter

relative to the osmotic process.

“La question de 1 ’osmose, que beaucoup croient resolue par

des formules mathematiques fondees sur des hypotheses com¬

mence a peine a etre posee. Mes experiences auxquelles vous

faites allusion, prouvent qu ’ il y a certainement des diaphragmes

actifs. Existe-t-il egalement des diaphragmes passifs, comme

pooir les gaz? Pour le moment tout le mond parait le croire,

sur la foi de van’t Hoff; mais pour mon compte, je n’en ai

jamais pu trouver un seul.

“Si je n’ai pas continue mes publications sur ce suject,

c’est que j ’ai rencontre de tres grandes difficultes pour

mesurer les pressions osmotiques avec des diaphragmes mous.

Existe-t-il, d’ailleurs, une veritable pression osmotique? Je

n’en suis pas sur. II arrive certainement toujours une pres¬

sion pour laquelle le mouvement osmotique semble s’arreter;

mais j’ai lieu de croire qu’elle correspond au moment ou la

quantite de liquide qui passe dans un sens par osmose , est

egale a celle qui passe en sens oppose par filtration.

“La question est interessante et fort delicate. Aussi, ver-

rais-je avec un tres grand plaisir d’autres experimentateurs

s ’engager dans cette voie.”

Again, very recently Barlow (1. c.) working in the lab¬

oratory of J. J. Thomson has reemphasized the fact that the

osmotic current is caused by the mutual potential energy of

solution of the liquids and that the direction of osmosis is

conditioned by the character of the membrane.

Precipitated membranes, like those of copper ferrocyanide,

etc., are hydrates, that is they contain more or less water.

When such a membrane is used to separate water from an

aqueous solution, the latter abstracts water from the septum —

the hydrate — which again takes up water from the side bathed

by the pure solvent. The affinity of the solution for addi¬

tional solvent must be sufficient to extract some water from

the membrane in order that an osmotic pressure may be pro¬

duced. If, in addition, the solute has considerable affinity for

the septum, copper ferrocyanide for example, as many of the

salts of the alkali metals are known to have, for instance, the

270 Wisconsin Academy of Sciences , Arts a/nd Letters .

solute will pass through to a notable extent; if on the other

hand, the solute has very little affinity for the septum, as is

the case with cane sugar, the membrane will allow but very

small quantities to pass through, and hence will be “ semi-

permeable. ” It is perfectly clear, too, that, in general, dif¬

ferent precipitated membranes would lose part of their

wrater with different degrees of readiness, and hence in quan¬

titative measurements of osmotic pressures different results

would be obtained when different precipitated membranes are

used. This is shown to be true by the experiments of Pfeffer

(Osmotische Untersuchungen) who found that when copper

ferrocyanide membrane was used a one per cent sugar solu¬

tion yielded an osmotic pressure of 51.0 cm. at 14.2° ; when a

Prussian blue septum was employed the pressure was 38.7 cm.

at 13.9° ; and when a calcium phosphate membrane wa.s used

the pressure was 36.1 cm. at 15.2°. In all three cases the

sugar which passed through the septa was insignificant ac¬

cording to his determinations. While the results of Pfeffer

are not final as determinations of the maximum osmotic pres¬

sures, inasmuch as he did not stir the liquids while measur¬

ing the pressures, the experiments nevertheless serve to show

that the pressures developed with the different septa are by no

means the same. If one were to compute the molecular

weight of sugar from the osmotic pressures which a one per

cent aqueous solution of it develops when copper ferrocyan¬

ide, Prussian blue and calcium phosphate are used as mem¬

branes respectively, different values would obviously be ob¬

tained, that is the conclusion would be reached that the mo¬

lecular weight of sugar in one and the same solution varies in

different cases, which is obviously absurd. The experimental

facts are, of course, readily explained by what has been said

above.

Further, when precipitated membranes are used and the

osmotic process goes on very rapidly, it generally occurs that

the solution robs the precipitated hydrate of water faster than

the latter is taken up from the pure water bathing the other

side of the septum. In such cases the membrane develops

ruptures through which the solution oozes out of the cell,

hence the necessity of the usual practice of adding the “mem-

Kohlenberg — Osmosis and Osmotic Pressure. 271

brane formers” to the liquids on each side of the membrane

to repair such leaks. It is obvious that in any osmotic ex¬

periment the composition of the septum is always in a state

of change, though the extent of this may be slight in some

cases. Here lies one of the chief difficulties of measuring

osmotic pressures (which are equilibrium pressures) with

soft diaphragms to which Eaoult alludes.

It must be borne in mind that the application of the gas

laws, either in simple or modified form, to dilute solutions is

based upon the experiments which Pfeifer made with cop¬

per ferrocyanide membranes, and which Morse and Frazer

have recently sought to verify. But these experiments have

all been made without stirring and with but one membrane,

and hence are not final. Furthermore, the osmotic pressures

of sugar solutions in pyridine, using vulcanized caoutchouc

as the semi-permeable membrane, show definitely that the gas

laws do not obtain here at all. In the face of the experi¬

mental facts which we now have as showing the nature of

the osmotic process and the magnitude of the osmotic pres¬

sures under different conditions, the general, indiscriminate

application of the gas laws in their simple or somewhat mod¬

ified form to all dilute solutions, and even to some that are

not dilute, as now in vogue, can not be too greatly deplored.

To speak of the osmotic pressure of any isolated solution

without specifying what membrane separates it from what

other liquid is nonsense,1 in the light of the facts here pre¬

sented. And further, to assume that solutes are polymerized

or dissociated in dilute solutions because the osmotic pres¬

sures developed by the latter in given cases happen to devi¬

ate from values computed from the gas laws is evidently

equally unjustifiable practice.

SUMMARY.

In this paper it has been shown that whether osmosis will

take place in a given case or not depends upon the specific

nature of the septum and the liquids that bathe it; and if

1 Compare also views expressed by Van Laar, Cbemisch Weekblad, 2,

1—16. (1905.)

272 Wisconsin Academy of Sciences , Arts and Letters.

osmosis does occur, these factors also determine the direc¬

tion of the main current and the magnitude of the pressure

developed. The motive force in osmotic processes lies in the

specific attractions or affinities between the liquids used, and

also those between the latter and the septum employed.

These attractions or affinities have also at times been termed

the potential energy of solution, etc., they are to the mind of

the writer essentially the same as what is commonly termed

chemical affinity.

It has been emphasized that osmotic pressures , are equi¬

librium pressures, and that in osmotic processes there is al¬

ways a current in both directions; though the main current

may in specific cases be so much stronger than the minor that

the latter almost sinks into insignificance. In such cases the

septum is termed “semipermeable.”

Vulcanized caoutchouc has been found to be a “semiperme-

able” membrane when it separates pyridine solutions of sil¬

ver nitrate, lithium chloride, and cane sugar from the pure

solvent.

The necessity of stirring the contents of the osmotic cell

and also the outer liquid during osmotic pressure measure¬

ments has been pointed out, and a new apparatus for

measuring osmotic pressures accordingly nas been devised.

The results of the osmotic pressure measurements show

that the gas laws do not hold; and it has consequently been

pointed out that the latter can not serve as a basis for a sat¬

isfactory theory of solutions.

The advantage of stirring in processes of dialysis has been

indicated; and it has been shown that whether substances

pass through membranes or not does not depend upon their

colloidal or crystalloidal character, but solely upon their

affinity for the membrane employed and for the liquids that

bathe it.

In conclusion I desire to thank Messrs. F. L. Shinn, J. H.

Mathews, Wm. Marquette and IT. E. Eggers for assistance

which they have kindly rendered me from time to time in

the experimental part of the work.

Laboratory of Physical Chemistry ,

University of Wisconsin , Madison.

mmmma

ifHHf

M:

*

'"•i