VOL. XLVII, NO, 553 JANUARY, 1913

THE

AMERICAN

NATURALIST

A MONTHLY JOURNAL

Devoted to the Advancement of the Biological Sciences with

Special Reference to the Factors of Evolution

CONTENTS

Page

Factors and Unit Characters in Mendelian Heredity. Professor T. H. MORGAN 6&

m

Vertical Distribution of the Chætognatha of the San Diego Region in Relation

to the Question of Isolation vs. Coincidence. ELLIS L, MICHAEL ~ -1

H

Ill. A Family of Spotted Negroes. Q.I. Simpson and Professor W. E. CASTLE ~ 50

The Effect of Fertilizers on Variation in Corn and Beans, Professor J. K. SHaw 57

d

THE SCIENCE PRESS A

LANCASTER, PA. GARRISON, N. ¥.

NEW YORE: SUB-STATION 84

The American

Naturalist

intended for publication and books, etc., intended for review should be

MSS.

sent to the sated of THE AMERICAN NATURALIST,

Garrison-on-Hudson, New York.

containing research work bearing on the problems Borba, organic evolu-

tion are eepecially welcome, and will

e given preference in publica

ndrea ha sige of contributions are supplied to authors free of charge.

One hu

Further reprints will be s

THE SCIENCE PRESS

NEW YORK: Sub-Station 84

Entered as second-class matter, April 2, 1908, at the Post Office at Lancaster, Pa., under the Act ot

ongress of March 3, 1879.

Lancaster, Pa.

Garrison, N. Y.

FOR SALE

ARCTIC, ae and GREENLAND

D8’ SKIN

Well Prepared pa Prices

articulars of

G. DIN NESEN. Bird Collecto

Husavik, North idaland: Via Leidle, England

TENTH EDITION.

THE MICROSCO PE,

1 to Microscopic Methods and to Histology,

by SIMON poprad GAGE, of Cornell University. Over

la and above fig in this new

and fully ion. Price $2.00, postpaid.

COMSTOCK PUBLISHING CO., Ithaca, N. Y.

Reptiles, Amphibians

and Fishes

W. F. H. ROSENBERG,

57, Haverstock Hill, London, N. W.,

England

begs to announce the publication of a new Price List

(No. 15) of Reptiles, Batrachians and Fishes contain-

ing over 400 species from all parts of the World. This

Catalogue is conveniently arranged for reference,

with authors’ names and indications of localities, It

will be sent post free on application as will the

following lists :

No. 11, Birds’ Skins, (5,000 species); No. 12, Le-

ppidotera, (5,000 species) ; No. 13, Coleoptera ; No.

14, Mammals; No. 15, Birds’ Eggs. Also list of Cabi-

nets, Collecting Apparatus, etc.

stock in the world of specimens in all

branches of Zoology.

Please state which lists are required.

University of Colorado

Mountain Laboratory

Courses in Field Biology (animal and

plant), Ecology, Systematic Botany, Bi-

ology of Ponds and Streams. Suitable

work for undergraduates and graduates.

Daily Field excursions.

The laboratory is at Tolland, Colo., in

the Rocky Mountains at an altitude of

8,889 feet. Climate cool and healthful.

Fees and expenses moderate.

For pamphlet address the professor in

charge.

Dr. Francis Ramaley

University of Colorado

BOULDER COLORADO

THE AMERICAN NATURALIST

THE i

AMERICAN NATURALIST

A MONTHLY JOURNAL

DEVOTED TO THE ADVANCEMENT OF THE BIOLOGICAL SCIENCES

WITH SPECIAL REFERENCE TO THE FACTORS OF EVOLUTION

VOLUME XLVII

NEW YORK

THE SCIENCE PRESS

19t3

THE

AMERICAN NATURALIST

VoL. XLVII January, 1913 No. 553

FACTORS AND UNIT CHARACTERS IN MEN-

DELIAN HEREDITY

PROFESSOR T. H. MORGAN

COLUMBIA UNIVERSITY

Tae factorial hypothesis has played an important rôle

in Mendelian heredity, and while students of Mendel’s

principles have had on the whole a pretty clear idea of

the sense or senses in which they have made use of factors

or symbols, yet those not engaged in the immediate work

itself have, I believe, often been misled in regard to the

meaning attached to the term factor, and by the assumed

relation between a factor and a unit character. The con-

fusion is due to a tendency, sometimes unintentional, to

speak of a unit character as the product of a particular

unit factor acting alone, but this identification has no real

basis. It has, in fact, more than once been repudiated,

yet the confusion has been so persistent that I venture to

try to make clear my own position at least—it is one I

think with which in the main many students of heredity

will agree—in regard to the relation between unit-factors

and unit-characters. I shall do this by means of several

examples taken from my breeding experiments with the

fly, Drosophila ampelophila.

The eye of tbis fly is red. A mutant arose with a ver-

milion eye. Crossed to the wild or red-eyed fly, the new

color proved to be a Mendelian recessive.

5

cc

THE AMERICAN NATURALIST [Vou. XLVII

According to the scheme that Mendel followed, red, R,

and vermilion, V, are symbolized as complete and con-

trasting characters carried by the germ-plasm of the

hybrid. They are assumed to separate in the germ-cells,

and as a consequence two kinds of these cells are pro-

duced.

According to a more modern interpretation, known as

the presence and absence theory, vermilion is supposed

to arise through the loss of something from the germ-

plasm of the wild fly. This something is not supposed to

be the factor for vermilion, but another factor. On this

scheme the red eye would be represented by the letters

RV, and the vermilion eye by rV ; as though the vermilion

color arose through the loss of a red factor.

The relative advantages of these two modes of repre-

sentation become apparent when two pairs of factors are

involved. For instance, a new eye color—pink—ap-

peared as a mutant. It, also, was recessive to red.

Mendel’s scheme would make the pink character the mate

of the red character, just as vermilion had been before.

But if pink and vermilion were mated to each other, it is

not clear whether vermilion and pink should be treated

as contrasted characters, or whether each should still be

treated as allelomorphic to red. If either of these alter-

natives is adopted, the scheme fails to account for what

actually happens. Mendel did not meet with such a situa-

tion, for none of his paired characters involved two

changes in kind in the same organ, and consequently the

problem did not exist for him.

Bateson did meet with just this situation in the case of

the comb of fowls and the coat color of mice. His’scheme,

if applied to the present case of the eye colors in Droso-

phila, would be to represent red by RV, vermilion by rV,

and pink by Rv. This scheme illustrates first why when

vermilion is bred to pink a red-eyed fly, rVRv, should

result; second, why in the second generation the propor-

tion 9:3:3:1 should appear;! and third why in the eye

* Except in so far as modified by sex-linkage.

No. 553] MENDELIAN HEREDITY $

color series a new color is expected in the F, generation,

represented here by rv. This new color I called orange,

and since rv only meant two absences, I followed the con-

ventional method and added the symbol O to stand for

orange. The completed formulæ were :

RVO red

rVO vermilion

RvO pink

rvO orange

This is identical with the scheme that Bateson adopted

for the mouse color series, viz:

GBCh gray

gBCh black

GbCh cinnamon

gbCh chocolate

In a later paper (1912) I used the symbol P instead of

R, so that the series stood:

PVO red

pVO vermilion

PvO pink

pvO orange

Let us now examine some of the possible interpreta-

tions of these symbols to see in what sense the letters

were used for factors.

It is undoubtedly implied, on the presence and absence

scheme, that something is lost from the original germ-

plasm PVO when the vermilion pVO arises. The ver-

milion color is supposed to be the product of what is left

when this something (called P) is lost. It is not sup-

posed on this hypothesis that the vermilion factor alone

is responsible for the vermilion color, for it is hypothet-

ically only a part of what is left when something (P) is

lost. Yet it is the identification of the vermilion factor

with the vermilion eye-color that the opponents of Men-

delism seem anxious to impute to the Mendelians.

8 THE AMERICAN NATURALIST. [Vou. XLVII

Again, when the pink eye mutant appeared, it would

have been assumed, on the presence and absence theory,

that something was lost, so that the formula is Pv0O.

Here again the pink color is the result of all that is left

when something (V) is lost. Pink is not assumed to be

produced by a factor P, but by what is left when a factor

V is lost. An egg is supposed to have lost something and

vermilion developed, another egg is assumed to have lost

something else and pink developed. It was the loss of the

vermilion factor that allowed pink color to develop, and

the loss of the pink factor that allowed vermilion color to

develop. When pink and vermilion are mated together,

the original color—red—is restored, because on this

_ scheme what each has lost is made good by what is found

in the other.

To my series of eye color factors the letter O was

added to indicate the nature of the color produced when

two factors, P and V, were assumed to be absent. The

symbol O at that time did not seem to stand in the formu-

læ on the same footing as P and V, because it stood for a

color, and not for a factor that had been lost from the

germ-cells of the wild fly. But since on the presence and

absence scheme O stood for the residuum after P and V

were lost it stood for the same sort of thing as did P and

V, for P and V also stood for residua, when they were not

used as symbols for factors. This will be made clearer

later.

When the experiments had progressed to this stage, a

new eye color appeared that was called eosin. Mated to

orange it gave red; therefore, it seemed that this mutant

must have contained P and V, and I inferred that it owed

its color to the loss of an imaginary O factor. Eosin was

represented, therefore, by PVo. But a moment’s thought

will show that on this scheme, as long as P and V are

present, any loss from the germ-plasm (giving a new eye

color) added to orange should give red, because orange

would contain what the new mutant had lost.

The history of this case will show how, with the best of

No. 553] MENDELIAN HEREDITY 9

intentions, one may be led into a paradoxical position in

regard to the use of factors. Even admiting that the rep-

resentation is purely symbolic, the letters used may unin-

tentionally come to stand for different things. Thus in

` the case first cited, the letter P in the formula vP stood

for a residuum that gave pink, but in the formula Vp, the

letter p stood for the loss of a P-factor, yet p is the allelo-

morph of P, which latter, as stated, meant the residuum

when V was lost. In other words, a double meaning was

attached to P, for it stood both for the P-factor, which was

only a part of the residuum, and also for the residuum as

a whole. It is this doubleness of meaning that gives the

opponents of Mendelian inheritance an occasion to im-

pose upon the factorial hypothesis a meaning that is

really foreign to it. Admitting that the Mendelians them-

selves have not always taken the pains to state explicitly

that the symbols represent both a factor and a residuum,

there is still little or no justification in imputing to the

presence and absence theory the view that a given char-

acter, pink color, for instance, is the product of a pink

factor alone. The attempt to impute to the factorial

hypothesis the same interpretation that Weismann made

use of in his theory of determinants rests largely upon an

erroneous understanding of the symbolism employed.

Weismann identifies each character of the organism as

the product of a special determinant. The factorial

hypothesis assumes only that the cell in one case is differ-

ent from the cell in the other, the difference relating, it is

true, to some part, but the character produced may be the

result of the whole or of much of the cell, and not of one

part alone.

There is a further difference between these two points

of view. A change in a factor may have far-reaching

consequences. Every part of the organization capable of

reacting to the new change is affected. Though we seize

upon the most conspicuous difference between the old

type and its mutant, and make use of this alone, every

student of heredity is familiar with cases where more

10 THE AMERICAN NATURALIST [Vou. XLVII

than the part taken as the index is affected. Weismann’s

theory, on the other hand, seems as a rule to identify each

character with a special determinant for that character,

and his meaning is clear when it is remembered that the

process of development on Weismann’s view is a process ;

of sorting out of the determiners of the germ-plasm into

different regions of the body. The factorial hypothesis

makes no such assumption, but refers differentiation to

the interaction of the parts on each other—every cell

retaining the full complex of the original germ-plasm.

Hence the possibility of the far-reaching effects of any

change in the germ-plasm!

II

The presence and absence system of nomenclature

(aside from its implications as to what is meant by pres-

ence and absence) has till the present time justified itself,

when properly interpreted, by its usefulness. It seems

to me that as a system of nomenclature it may be used, if

one so desires, quite apart from the idea, that a loss in

a character involves necessarily a loss in the germ-plasm.

I can bring forward one clear case at least that seems to

me difficult to explain if absence is taken literally to mean

the loss of a factor from the germ complex. I refer to a

mutation ‘‘backwards,’’ which in the older terminology

meant reversion, or atavism.? In my pure cultures (at

rare intervals) individuals have appeared like the orig-

inal progenitors of the stock. I have not serupled to put

aside this evidence, because contamination, even with

extreme care, will occasionally occur; and even if a rever-

sion had occurred there would be no way of proving that

it was such and not contamination. In fact, eosin first

appeared in white-eyed stock and seemed to arise through

reversion, but at the same time it seemed so improbable

that this could happen that I tried to account for its ap-

pearance in a roundabout way. Now I should say that

the factor w reverted to W.

It is needless to add, perhaps, that atavism by recombination is not

here for a moment brought into question.

No. 553] MENDELIAN HEREDITY 11

But the clear case referred to above is the following:

Quite recently there appeared in a culture bottle that had

been producing for more than four months (probably for

twelve generations) only wingless flies, an individual with

one ‘‘wingless’’ wing and one normal wing on the other

side. Here the evidence is conclusive that reversion had

occurred. The wingless stock in which the asymmetrical

form arose had purple eyes and the same eye color was

present in the new type. As the eye color was relatively

new at the time the chance that contamination had oc-

curred was rendered very unlikely. Had contamination

by a red-eyed fly occurred, making the new type a hetero-

zygote, the eye color would have been the dominant red.

When the asymmetrical fly (g) was bred to wingless

females only wingless flies appeared, for three or more

generations. The reversion, therefore, was somatic and

did not involve the germ-plasm, yet this fact does not

invalidate the question here raised.

In the light of this evidence, as well as the evidence

from ever-sporting varieties (that may also be consid-

ered, I think, as mutatingand reverting as regular proc-

esses), I believe it unwise to commit ourselves any longer

to a view that a recessive character is necessarily the

result of a loss from the germ-cell. We need only assume

that some readjustment occurs, and as the result a new

factor is produced. A simile may make this clearer, if

not taken too literally. If we suppose that a factor is a

labile aggregate, and that a rearrangement in it occurs,

then the new aggregate in connection with the other parts

of the cell produces a character that differs from the old

one. Here there need be no loss, but only a change in

configuration with a corresponding change in the end

product in which the changed part plays a role, along

with the other parts of the cell. A factor, in this sense,

may exist in two or more forms according to the state

of equilibrium; one of its states is dominant-producing,

and the other is recessive-producing. Such a view may

make it easier for us to appreciate that a mutation need

12 THE AMERICAN NATURALIST [Vou. XLVII

not be a loss, and that a recessive may revert in the sense

that it may mutate. In chemical terms, the process is

reversible.

III

As I have pointed out, the presence and absence nomen-

clature, if properly understood, offers no practical diffi-

culties so long as only two changes in the same organ are

involved, but in experiments with Drosophila we have

passed beyond this stage and must have at command a

system by means of which more than two factors may be

easily and conveniently represented. How impossible it

becomes to use the presence and absence nomenclature

when new characters are appearing may be shown by the

following illustrations.

As already stated, Mendel’s method of representing

the allelomorphic pairs sufficed so long as one new char-

acter is contrasted with the original one. In this sense

the relation of a vermilion-eyed mutant to the red-eyed

fly could be fully represented by treating red (R) and

vermilion (V) as allelomorphs. But when another muta-

tion in eye color appeared the scheme was no longer fea-

sible. Now, in the same sense in which it became neces-

sary to supplant Mendel’s scheme by another one, it be-

comes necessary to change the presence and absence

scheme when a third mutation appears in the same organ;

for, the presence and absence scheme is not sufficiently

elastic to allow the introduction of a new term in the

series, unless a complete revision of the method is made

each time that a new mutation in kind occurs.

For example, when it becomes desirable to.compare the

eosin eye with the vermilion-pink (or orange eye already

known) it becomes puzzling to know what symbols to

adopt. If, as I assumed, the symbol O in VPO is changed

to small o, then the formula for eosin becomes VPo. But

this is inconsistent with the scheme already adopted be-

cause the small letter o stands for a character called

eosin. If to avoid this ambiguity a letter E (or e) is in-

troduced for eosin the situation is even more puzzling.

No. 553] MENDELIAN HEREDITY 13

The only logical method that could be followed, if an

attempt is made to apply consistently the current scheme

of presence and absence? would be the following:

When it becomes necessary to construct a series, let us

say one involving three characters (PVE), the three

double recessives (Pve, pVe, pvE) must be made up and

suitable names given to them, the initial letters of these

names then become the factors sought. Such a procedure

not only involves holding in suspense the naming of the

factors until all the double recessives have been obtained,

but involves renaming all the factors, each time a new

series is made up.t This method is not likely to recom-

mend itself if a simpler one can be employed. The plan

here advocated avoids such difficulties.

The first letter (or the first and second or some other

significant letter) of thename of the new character stands,

as heretofore, as its symbol; thus P stands for the pink

factor and small p stands for the correlative factor of the

pink-eyed fly. Whether small p represents the loss of the

P factor, or a change in that factor when the pink eye

appears, is immaterial. The large letter represents the

dominant character in conformity with the current

scheme.” The eye color series will then be:

Red PVE

Vermilion PvE

Pink pVE

Vermilion-pink pvE

Eosin PVe

Eosin-vermilion Pve

EKosin-pink pVe

Kosin-pink-vermilion pve

3It is the nomenclature that is here brought into question and not, for

the moment, the underlying conception of presence and absence, for even in

my scheme this conception might still be held if it seemed desirable to do so.

‘When, as in the case of the mouse colors, all the members of the series

are known, there is no difficulty in finding suitable symbols, for the current

names of the characters give the letters for the symbols.

5 When a new dominant character appears it is represented by the capital

letter and its allelomorph in the original form by a small letter.

14 THE AMERICAN NATURALIST [Vou. XLVII

The same scheme might be followed by using the small

letters for the factors in the original red eye: thus red =

pve; and the capital letter for the corresponding factor

in the mutant ; thus, vermilion — pVe,ete. A disadvantage

of this scheme is that the large letter now stands for a

recessive condition and the small letter (its allelomorph)

for a dominant condition. Usage has, however, made us

accustomed to interpreting a large letter as a dominant,

its corresponding recessive (its allelomorph) by a small

letter and therefore the plan first suggested seems more

desirable. It is with much reluctance that I suggest this

change in our present nomenclature. It has become nec-

essary, however, in the case of the Drosophila to find

some way to represent consistently those cases in which

three or more factors are involved in the same organ.

The change is not one of any theoretical importance, but

a practical necessity for all cases of this kind. When one

new character is contrasted with the original one, Men-

del’s way may still be the simplest and easiest way of

formulating the results, and will, no doubt, be followed.

When two new characters are involved the formula of

presence and absence is a sufficient way of representing

the symbols. But when new mutations are appearing

some other plan must be adopted. The one here sug-

gested has at least two merits: it is as easy to use as

either of the foregoing for one and for two characters,

and can also be utilized when any number of further

mutations appear in the same organ.

The scheme applied to body colors is as follows: Two

mutants arose, yellow and black, and by recombination, a

‘*hrown’’ or yellow-black fly was obtained. The symbols

would be wild fly—=YYBB, yellow—yyBB, blackYYbb,

and yellow-black yybb. Two other mutations in body

color have appeared, both dark, one is called ebony, eb

and the other sable, s. When brought in connection with

the preceding mutation the gametie symbols would be:

Wild fly YBE.S

Yellow yBES

No. 553] MENDELIAN HEREDITY 15

Black YbEW

Yellow-black ybE.S

“Ebony YBe.S

Sable YBE.LWS

Ete.

Another combination is represented by certain wing mu-

tations. A mutant called miniature appeared and may be

represented by m; another mutant appeared, called rudi-

mentary, and may be represented by r; and a third form,

produced by recombination, was called miniature-rudi- —

mentary, mr. The symbols for this series would be:

Wild fly MR

Miniature mR

Rudimentary Mr

Rud.-min. mr

Later several other mutations in wings also appeared.

Six of these may be selected for illustration, viz: Vestigial,®

va; Bifid, bi; Are, ar; Curved, cv; Jaunty, 7; Balloon, ba.

If these are brought into connection with the foregoing

the symbols for the wild fly in terms of these factors

would be:

Wild Ny ==, R, Vo, Bi, Ár, O, J. Be:

In order to study the relation of these characters to

each other it has become necessary to combine many of

them and in order to represent the results some system

of symbols must be adopted. Obviously, it would be

highly undesirable to be obliged to revise the system each

time that any of these new mutations are brought into

relation with those that have already been compared.

It may be asked, why may not the current scheme be

retained, since in most cases only two characters are

likely to be involved and characters can always be con-

trasted in pairs on this scheme? The answer is that more

€ This is the wingless fly of former papers.

$

16 THE AMERICAN NATURALIST [Vou XLVII

characters in the same organ have already been obtained,

and it is at least as important to have a scheme by which

they can be represented as it is to have a scheme where

two characters only are studied. Moreover, the more

characters that are obtained that show association in

inheritance the further we may hope to go in our analysis

of the consitution of the germ-plasm, which is admittedly

the fundamental problem in the study of heredity. We

must have some convenient way of representing the

symbols in order to carry out this analysis, and on the

grounds of convenience alone some scheme other than the

current one must be found, at least for such a case as this

of Drosophila. Another scheme has, in fact, been adopted

by Baur and Hagedoorn. The letters that stand for the

factors bear no relation to the name of the characters in-

volved. This scheme allows the addition of any number

of new factors to a series under consideration. In prac-

tise, however, this plan makes it extremely difficult to

~~ understand what any formula means without continual

reference to the key of symbols used. We have found in

practise that the scheme is so puzzling when several

factors are under consideration that we have been led to

follow the current method of representing each factor by

the initial letter (or other suggestive letters) of the char-

acter that it stands for. Except in this regard the method

of formulation here suggested is similar in principle to

the A.B.C. scheme of Baur.

VERTICAL DISTRIBUTION OF THE CHATOG-

NATHA OF THE SAN DIEGO REGION IN

RELATION TO THE QUESTION OF

ISOLATION VS. COINCIDENCE

ELLIS L. MICHAEL

Scripps INSTITUTION FOR BIOLOGICAL RESEARCH OF THE UNIVERSITY

oF CALIFORNIA, LA JOLLA, CAL.

INTRODUCTION

Ever since Jordan (’05) called attention to the almost

universal neglect of Moritz Wagner’s contention that

geographical isolation is an important factor in the for-

mation of species, ‘‘Jordan’s Law’’ (’05, p. 547) that

‘‘oiven any species in any region, the nearest related

species is not likely to be found in the same region nor

in a remote region, but in a neighboring district sepa-

rated from the first by a barrier of some sort,’’ has been

subject to much controversy and diversity of opinion.

The conclusions of those biologists dealing with land

fauna have, as a rule, emphasized the fact of isolation,

whereas those of the marine biologist have tended to

emphasize the fact of coincidence, or at least to doubt

the truth of isolation. It is therefore part of the busi-

ness of the marine biologist, through whose investiga-

tions new and important data have been accumulated, to

throw as much light as possible upon the problems of

isolation and coincidence. This is particularly true with

regard to data concerning the Chetognatha because, as

pointed out by Kofoid (’07), the group is exclusively

marine and pelagic, and so completely circumscribed as

to make it probable that their entire evolution has taken

place within the confines of the open sea.

At the outset, the fundamental differences in the prob-

lem with reference to land and marine fauna must be

emphasized. Kofoid (’07, p. 241) has pointed out that

17

18 THE AMERICAN NATURALIST [Vou. XLVII

‘barriers are far less in evidence in the environment of

the pelagic fauna than in that of the shore or of the

land,” and that, while there do exist ‘‘limited regions

along the margins of great ocean currents’’ which might

afford means of hydrographic isolation, changes in

hydrographic conditions such as temperature, density,

substances in solution, illumination, etc., are so grad-

ual that stratified areas do not exist to any large extent.

Furthermore, land faunas are segregated into neigh-

boring or remote areas almost entirely with reference to

latitude and longitude. With pelagic faunas this is not

necessarily, perhaps not usually, the case, for a third

dimension—depth—is involved. It therefore follows

that closely related pelagic organisms may be coinci-

dently distributed as regards latitude and longitude, and

still be completely isolated in their vertical distribution.

This fact signifies that data respecting the isolation of

a pelagic fauna will be wholly inadequate unless the ver-

tical distribution of the particular species or group

under consideration be capable of determination and

analysis. In his discussion of ‘‘the coincident distribu-

tion of related species of pelagic organisms as illustrated

by the Chetognatha’’ Kofoid (’07), while recognizing

the force of this point, has utilized data pertaining al-

most exclusively to latitude and longitude. This was

consequent upon no lack of appreciation on his part of

the real problem involved, but solely to the fact that the

necessary data were missing. Moreover, what little has

been previously discovered relative to the vertical dis-

tribution of this group was based upon observations

scattered over such large areas as to make any approach

to critical analysis of the problem of isolation almost

impossible. However, through the efforts of the San

Diego Marine Biological Station, a mass of data has been

collected which enables an entirely new light to be

thrown upon this problem.

Since 1904 this station has centered its collecting upon

an irregular area of about 30 square miles lying be-

No. 553] DISTRIBUTION OF THE CHASTOGNATHA 19

tween 32° 20’ and 33° 30’ N., and between the coast and

119° W. From this small area 68,962 specimens com-

prising ten species of Ohetognatha have been collected,

and, as all depths between the surface and 350 fathoms

have been examined with horizontal closing nets, the

depth from which each specimen was obtained is known

with the nearest approach to certainty permitted by any

known method of collecting. As a critical analysis of

this data has been published elsewhere [see Michael

(711)] reference must be made to that paper for the

methods, problems and details involved in determining

the vertical distribution of each species, so that, in the

following pages, only the fruits of that research bear-

ing directly upon the present subject-matter will be dis-

cussed. It will be shown (1) that of the most closely

related ‘‘couplets’’ of species only one has been taken

from the San Diego region, (2) that, of those species oc-

curring in this region, each has its own definite and

specific manner of vertical distribution, (3) that the most

diverse species (morphologically) have the most coinci-

dent vertical distribution, and (4) that, while several

species have sometimes been taken in the same haul,

rarely more than one was represented by sexually ma-

ture individuals.

RELATIONSHIPS BETWEEN THE SPECIES OF CH#TOGNATHA

Adopting Ritter-Zahony’s (’11b) careful revision of

the Chetognatha as our starting point, the group be-

comes separable into six genera, Sagitta, Pterosagitta,

Spadella, Eukrohnia, Heterokrohnia and Krohnitta.

Sagitta is represented by eighteen valid and four rather

doubtful species, Pterosagitta is represented by one,

Spadella by one, Eukrohnia by two, Heterokrohnia by

one and Krohnitta by two.

Now the eighteen valid species of Sagitta fall into two

sharply contrasted groups by virtue of the presence or

absence of a collarette which is a conspicuous thickening

of the epidermis posterior to the head. Ten species are

20 THE AMERICAN NATURALIST [Vou. XLVII

provided with this structure, while in eight it is entirely

missing. The species comprising each group are listed

below: 7

Species with Collarette Species without Collarette

S. bipunctata S. enflata

S. decipiens S. hexaptera

S. neglecta S. lyra

S. regularis S. gazelle

S. ferox S. serratodentata

S. planktonis ; S. bedoti

S. hispida (robusta Doncaster) S. elegans

S. tenuis S. macrocephala

S. pulchra

S. siboge

Those species having the collarette may be further

separated into diverse groups by means of the following

main characteristics: (1) Those in which the body is

transparent as contrasted with those in which it is

opaque, (2) those in which the collarette extends to the

ventral ganglion as contrasted with those in which it never

extends more than half way to the ganglion, (3) those

in which the anterior fin extends to the ventral gang-

lion as contrasted to those in which it does not, (4) those

having more than 50 per cent. of the posterior fin in

front of the tail-septum as contrasted with those having

more than 50 per cent. of the fin behind the tail-septum,

and (5) those in which the anterior fin is shorter than

the posterior fin as contrasted with those in which the pos-

terior fin is the shorter. Let it not be thought that these

are the only characteristics used to differentiate the

various species of Sagitta provided with the collarette.

Far from it! Many others of great specific importance

are made use of, but, if classified according to those just

enumerated, the most closely related ‘‘couplets’’ remain

inseparable while those species not so closely related are

readily separated from each other. This may be graph-

ically represented by arranging these five pairs of con-

trasted characteristics into a series of ‘‘rows’’ and

No. 553]

DISTRIBUTION OF THE CHÆTOGNATHA

21

‘‘columns’’ and then writing the names of the species,

having those in question in each square made by the

Such an arrangement is given

intersecting ‘‘arrays.’’

below:

TABLE I.

More than 50 Per | Less than 50 Per | Anterior Fin | Anterior Fin

Cent. of Posterior | Cent. of Posterior | Shorter than Longer than

Fin in Front of Fin in Front 0 Posterior Posterior

Tail-septu m Tail-septum | Fin Fin

|

| |

Body trans- S. bipunctata | S. tenuis | S. bipunctata | S. pulchra

parent. S. decipiens | S. decipiens

S. pulchra | | S. tenuis

Body opaque. S. planktonis | S. neglecta | 8. neglecta | S. planktonis

S. siboge S. regularis | S. regularis |S. ferox

| S. ferox | S. hispida | S. siboge

| S. hispida | |

Collarette ex- S. planktonis | S. neglecta | s. neglecta |S. ferox

tending to ven- S. regularis | S. regularis | S. planktonis

tral ganglion S. ferox | |

| l

oo eet ex-| S. bipunctata | S. hispida | S. bipunctata | S. pulchra

tendin S. decipiens S. tenuis | S. decipiens | S. siboge

alf tind ton ven-| 8S. hra | S. hispida |

tral ganglion. _tral ganglion. | S. siboge | S. tenuis |

| L

T fin ex- = S. planktonis | S. neglecta | S. neglecta |8. ferox

ding to ven-; 5S. pulchra S. regularis | S. regularis | S. i

ser ganglion S. siboge S. ferox 8. tenuis | S. 8

S. tenuis |S. od 5c

Anterior fin not) „S. bipunctata | S. hispida S. bipunctata |

extending to S. decipiens | S. decipiens |

ganglion, | S. hispida |

An examination of this table shows that in every

square where S. bipunctata occurs there also S, decipiens

is found, and the same relation holds between S. ne-

glecta and S. regularis. These four species, then, are to

be regarded as constituting two very closely related

‘‘couplets.’?? A third ‘‘couplet,’? whose constituent

Species are somewhat less closely related, is that of S.

ferox and 8. planktonis, which only differ in the propor-

tional extent of the posterior fin in front of the tail-

septum. The remaining species are clearly distinct.

Turning attention to those Sagitta devoid of the col-

larette, S. enflata, S. hexaptera, S. lyra, S. gazelle and

S. bedoti are exceedingly transparent, while S. serrato-

22 THE AMERICAN NATURALIST [Vou XLVII

dentata, S. elegans and S. macrocephala are opaque.

While the three opaque species are unmistakably dis-

tinct, we find that, in the transparent group, S. enflata

and S. hexaptera form one closely related ‘‘couplet’’

while S. lyra and S. gazelle make another. This is shown

more clearly below:

TABLE II.

Anterior and

Posterior Fin Ex- Posterior Fin Anterior Fin | Posterior Fins

tends Caudally to | Never Extends to | Confluent with pahi y Sepa

Seminal Vesicles | Seminal Vesicles Posterior Fin rated by an

Interval

More than 50 per! S. lyra S. enflata S. lyra S. enflata

cent. of pos- S. gazelle S. hexaptera S. gazelle S. hexaptera

terior fin in

front of tail-

septum.

S. bedoti S. bedoti

Less than 50 per

os-

teri n

front of tail-

septum.

All told, then, we have in the genus Sagitta five closely

related ‘‘couplets’’ of species. It is not to be presumed

that every ‘‘couplet’’ expresses the same degree of close-

ness between its two members, for such is not the ease.

Unquestionably the two species most closely related are

S. neglecta and S. regularis, and the two least so—S.

ferox and S. planktonis. Now if we list these ‘‘couplets’’

in one column and the species so far taken from the San

Diego region in another, the interesting fact is evident

that, except in one case, the San Diego Sagitta contain

only one species of each ‘‘couplet.’’ Such lists are given

below.

San Diego Sagitta “Couplets”

S. neglecta S. neglecta-S. regularis

S. bipunctata S. bipunctata-S. decipiens

S. lyra S. gazelle-S. lyra

S. enflata S. enflata-S. hexaptera

S. hexaptera

S. planktonis S. planktonis-S. ferox

S. serratodentata

No. 553] DISTRIBUTION OF THE CHÆTOGNATHA 23

Looking to the other genera we find Eukrohnia com-

posed of two species (E. hamata and E. fowleri), Kroh-

nitta of two (K. subtilis and K. pacifica), and Hetero-

krohnia, Pterosagitta and Spadella of one each. Now

E. hamata and E. fowleri form an exceedingly closely

related ‘‘couplet,’’ but only the former is known to occur

in the San Diego region. Again, K. subtilis and K.

pacifica are so nearly alike that it is very difficult to de-

scribe their differences although they are probably valid

species. Yet, only the first has been found in California

waters. Of the three remaining genera Heterokrohnia

and Spadella are not represented in our collections, and

Pterosagitta by only one individual of its single species

P. draco.

In so far, therefore, as the relationships among the

Chetognatha have been correctly interpreted, it is evi-

dent that, except for the occurrence of both S. enflata

and S. hexaptera, there is no instance of two of the most

closely related species having been taken from the San

Diego region.

GENERAL DISTRIBUTION OF THE ‘‘COUPLETS’’

Having pointed out that only one of a ‘‘couplet’’ of

the most closely related species occurs in the San Diego

region, it will be interesting to ascertain to what extent

the same relation holds in other parts of the world.

Furthermore, wherever both members of a ‘‘couplet’’

are recorded from the same vicinity it will be to the

point to determine, if possible, to what extent their dis-

tribution within the area is coincident or isolated.

S. NEGLECTA AND S. REGULARIS

The members of this ‘‘couplet’’ may be designated as

warm water, epiplanktonic species whose northern and

southern limits of distribution are 35° N. and 9° S. The

highest surface temperature recorded in connection with

their capture is 29° C. and the lowest 15°.5 C. They were

both originally described by Aida (’97) from Misaki

24 THE AMERICAN NATURALIST [Vou. XLVII

Harbor, where, so far as known, they are coincidently

distributed.

In the Siboga area Fowler (’06) records 45 surface

hauls containing either one or the other species. Of

these, 35 contained S. neglecta but not S. regularis, and

6 contained S. regularis but not S. neglecta. In only 4

hauls were both species obtained. When we remember

the large area covered by this expedition these results

point toward contiguous and slightly overlapping, rather

than coincident distribution.

The two expeditions of the Pola to the Red Sea ob-

tained both S. neglecta and S. regularis. The collections

of the first expedition (1895/96) were made in an area

limited by 21° 27’ and 29° 45’ N., and 32° 30’ and 38° 30’

E., while those of the second (1897/98) were made some-

what further south and east within the limits of 15° 1’

and 28° 42’ N., and 32° 56’ and 42° 31’ E. Ritter-Zahony

(709) records 32 surface hauls made during the first ex-

pedition that contained one or other of the two species.

Of these, 25 contained only S. regularis, 6 only S.

neglecta, while in but one haul were both species taken.

During the second expedition 27 surface hauls were made

of which 12 contained S. neglecta only, 10 S. regularis

only, and 5 contained both. These data strongly suggest

that the two species are distributed in contiguous re-

gions which overlap considerably along the edges.

Of the other expeditions, the Biscayan, Plankton and

National failed to catch either species. The Gauss ob-

tained both in the region of Port Natal, but never in the

same hauls. Doncaster (’02) records both under the

names of S. septata and S. bedfordii from the Maldive

and Laccadive Archipelagoes, but nothing is stated as to

whether they were obtained in the same hauls or not.

Ritter-Zahony (’10a) records S. regularis from Sharks

Bay, Australia, but failed to find S. neglecta.

No. 553] DISTRIBUTION OF THE CHASTOGNATHA 25

S. BIPUNCTATA AND N. DECIPIENS

S. bipunctata is a eurythermal, euryhyaline and cos-

mopolitan species recorded from the epiplankton of the

arctic, sub-arctic, north temperate, tropical and south

temperate Atlantic Ocean, the south temperate and trop-

ical Indo-Australian Ocean, and the north temperate

Pacific Ocean, as well as from the mesoplankton of the

north temperate and tropical Atlantic. Its northern

limit is 74° N. and its southern 28° S. The highest tem-

perature recorded in connection with its capture is

33°.6 C., while the lowest is 0°.2 C. S. decipiens, on the

other hand, is mesoplanktoniec. Both were taken from

the Bay of Biscay, S. decipiens from between 100 and

200 fathoms, and S. bipunctata only in open vertical

hauls made between 50 and 200 fathoms to the surface,

the total yield being only 7 specimens. Ritter-Zahony

(710 b, p. 4) records both from the Irish Sea, but, in re-

gard to S. decipiens, says: ‘‘S. decipiens is purely meso-

planktonic and in the Irish area was only found at depths

varying from 164 to 1,150 fathoms.’’ Concerning S.

bipunctata he says that it is ‘‘confined to the epiplankton.

... The quantity of S. bipunctata in the upper epiplank-

ton is larger than in the lower.’’ Finally, both species

have been taken in the Atlantic Ocean between 60° N.

and 8° S. but, while S. decipiens occurred only in open

nets from below 100 fathoms and in closing nets from

between 100 and 600 fathoms, S. bipunctata occurred

only in the epiplankton. From this evidence it seems

that wherever the two species occur in the same region

they are isolated by their manner of vertical distribution.

S. LYRA AND S. GAZELLE

S. lyra is a cold water, nearly eurythermal species

ranging from 73° N. to 7° 33’ S., the highest temperature

recorded in connection with its capture being 18°.6 C.

and the lowest 1°.1 ©. It has been found in the epiplank-

ton of the arctic, sub-arctic and north temperate Atlantic,

and sub-antarctic Pacific Oceans, as well as in the meso-

26 THE AMERICAN NATURALIST [Vou. XLVI

plankton of the sub-arctic, north temperate and tropical

Atlantic, the tropical Indo-Australian and the north

temperate Pacific oceans. S. gazelle, on the other hand,

isarare form. <A few specimens were first taken during

the Gazelle expedition from the Indian Ocean (43° S.)

from a depth of 75 and 1,300 fathoms. It is also recorded

from the Atlantic Ocean (35°.5 S.), where it was taken in

a single haul from about 1,400 fathoms, and from the

Antarctic Ocean between 60° and 66° S., where it was

taken from 10, 25 and 50 fathoms. These data indicate

that S. gazelle is confined to the southern hemisphere

and tends to be distributed circumpolarly. Records of

the Gauss expedition show that out of 88 hauls contain-

ing either S. lyra or S. gazelle, 39 contained only the

former, 42 only the latter, and 7 both species.

S. ENFLATA AND S. HEXAPTERA

S. enflata is a warm water purely epiplanktonic

species whose northern and southern limits of distribu-

tion are 40° 24’ N. and 34° 52’ S. The highest tempera-

ture recorded in connection with its capture is 32° C. and

the lowest 15°.5 C. It has been taken from the north

temperate, tropical and south temperate Atlantic, the

south temperate and tropical Indo-Australian and the

north temperate Pacific oceans. S. hexaptera, on the

other hand, is a eurythermal, nearly cosmopolitan species

found in the lower epiplankton or mesoplankton of the

arctic, sub-arctic, north temperate, tropical and south

temperate Atlantic, the south temperate and tropical

Indo-Australian and the north temperate and sub-ant-

arctic Pacific oceans. Its northern and southern limits

of distribution are 74° N. and 28° S., while the extremes

of temperature recorded in connection with its capture

are 29° C. and 6° C

Both species have been taken together from the same

areas during a number of expeditions. In the Siboga

area Fowler (’06) records 58 surface hauls containing

one or other of the species, of which 31 contained both,

No. 553] DISTRIBUTION OF THE CHAITOGNATHA 27

while 26 contained S. enflata but not S. hexaptera, and

only one contained S. hexaptera alone. During the first

expedition of the Pola to the Red Sea 32 surface hauls

were made which contained both species, 21 which con-

tained S. enflata but not S. hexaptera, and only one

which contained S. hexaptera alone. During the second

expedition not a single S. hexaptera was taken in surface

hauls that was not accompanied by S. enflata, there being

13 hauls containing both and 29 containing S. enflata

alone. Finally, during several expeditions covering

parts of the Adriatic, Ionian and Ægean seas, Ritter-

Zahony (’08) records 45 surface hauls containing S. en-

flata, of which 6 also contained S. hexaptera, and only 4

hauls in which the latter species was taken without the

former.

These data certainly indicate a high degree of coinci-

dence. However, the fact that S. enflata is rarely re-

ported other than from the upper epiplankton and that

S. hexaptera is more typical of the lower epiplankton and

mesoplankton, suggests isolation with respect to sex-

ual maturity. Concerning this Ritter-Zahony (’10)),

who has been very careful to distinguish immature from

mature specimens, says: ‘‘Like S. serratodentata, S.

hexaptera is a species which can not endure low tempera-

tures until it has reached the adult stage. ... We do

not, as a rule, find large specimens until we come to the

lower epiplankton.’’ Until more is known regarding the

stages of growth of these specimens taken on the surface

together with S. enflata we can not regard the cases of

coincidence revealed above as anything more than nega-

tive evidence of isolation.

S. PLANKTONIS AND S. FEROX

S. planktonis is a eurythermal species recorded from

both epi- and meso-plankton of the north temperate At-

lantic and Pacific oceans. It has not been reported north

of 32° 45’ N., nor south of 8° 30’ S., except for a few

small specimens from the Antarctic between 65° and

28 THE AMERICAN NATURALIST [Vou. XLVII

66° S. Its temperature range is from 27° C. to 4°.7 C.

S. ferox, on the other hand, is a warm-water species con-

fined, so far as known, to the epiplankton of the tropical

Indo-Australian region. Both species were taken during

the Siboga expedition, but, while S. ferox was taken in

abundance from the surface, S. planktonis was taken only

from the mesoplankton. There is no record of both hav-

ing been taken in the same hauls except in those made

with open vertical nets.

E. HAMATA AND E. FOWLERI

It is still an open question in my mind whether Æ.

fowleri is a valid species or merely a synonym for £E.

hamata. Ritter-Zahony (’11 b) describes certain differ-

ences, but the characters used appear indicative of varia-

tion within the species rather than of constant specific

} ters. If they should prove synonymous, then

Eukrohnia would be represented by only one species.

However, assuming their validity, then E. hamata would

be distributed in the mesoplankton of the Indian, Altantic

and Antarctic oceans, while E. fowleri is rarer and per-

haps more cosmopolitan, occurring in the Irish sea

between 200 and 1,100 fathoms, in the Bay of Biscay be-

low 325 fathoms, in the Malay Archipelago below 460

fathoms, and rarely in the open Atlantic below 500

fathoms. It might be added that E. hamata also occurs

in the epiplankton of ‘the Arctic and Antarctic regions,

while E. fowleri always remains confined to the meso-

plankton. During the Plankton expedition the species

were taken together in only one closing-net haul made

between 500 and 600 fathoms, and only twice out of 18

open vertical hauls from a variety of depths.

K. SUBTILIS AND K. PACIFICA

K. subtilis is regarded as a eurythermal cosmopolitan

species ranging from 60° 12’ N. to 29°30’ S. The tem-

perature corresponding to its capture varies from 30°.8

No. 553] DISTRIBUTION OF THE CHAITOGNATHA 29

C. to 5°.3 C. It is reported from both epi- and meso-

plankton of the north temperate and tropical Atlantic and

tropical Indo-Australian oceans, as well as from the epi-

plankton of the south temperate Atlantic and south tem-

perate Indo-Australian oceans and from the mesoplank-

ton of the north temperate Pacific. K. pacifica, on the

other hand, is a warm-water epiplanktonic species from

the tropical Atlantic and Indo-Australian, and the north

temperate Pacific oceans. Its northern limit is 35° N.

and its southern 7° 30’ S. During the Siboga expedition

both species were taken together in but one haul, and that

one made by an open vertical net from 1,000 fathoms.

This is the only instance, so far as I can ascertain, where

both species have been obtained from the same area.

In summing up we find that the members of each ‘‘coup-

let’’ tend to be isolated in one way or another. S. neglecta,

for instance, maintains a distribution which, while over-

lapping more or less, is contiguous rather than coin-

cident with that of S. regularis. In the case of S. bipunc-

tata and S. decipiens the data show that wherever both

are taken within the same area the former is confined to

the epiplankton, while the latter occurs only in the meso-

plankton. With S. lyra and S. gazelle the distribution is

never coincident, but, in some instances, contiguous and

overlapping. S. enflata and S. hexaptera present the

most striking evidence in favor of coincidence but, even

here, the chances are that only the immature of S.

hexaptera occur in the upper epiplankton, so that an

effective physiological isolation is probably maintained.

S. planktonis and S. ferox, while they do occur together

in the Siboga area, are isolated by their manner of ver-

tical distribution, S. ferox being epiplanktonie and S.

planktonis mesoplanktonic. The members of the doubtful

“couplet” comprising E. hamata and E. fowleri are only

rarely taken in the same net hauls, which indicates con-

tiguous rather than coincident distribution, although this

appearance may be due entirely to the fact that E. fowleri

is not abundant anywhere. Finally, K. subtilis and K.

30 THE AMERICAN NATURALIST [Vou. XLVII

pacifica have never been taken from the same region,

excepting in the case of the Siboga expedition when they

were obtained together in only one haul from 1,000

fathoms to the surface.

VERTICAL DISTRIBUTION OF THE CHH®TOGNATHA OF THE

San Dreco Recron

The 68,962 specimens of Chætognatha obtained from

the San Diego region were distributed among the various

species as follows:

8. Pee A T Ok ea E r N eile 51,670

Di PERO E RE EE P T E A OIE TEs ORES 10,127

8. Cana. aS care CON See Path ake ace Ae pee 6,575

Be ee hoe ee ale ce ee eee a ee

PSs MO COCD OE Se pas eae ie 8 aos aes is SRC eA EVE EROS 127

R? honat Oe OOS SS Oi ee oe ae 72

Ke MOr st oie a Cee ee ew ree 50

st MONGOO ce Se a ee es ae ata en 41

Bs BOLUDVONE ore Sew oy vas Vee Corus Oey wae coe y 28

Page a Babies g E er REMY Footy ier eta arre hes ee pee remit a 1

Turning attention first to those species that must be

regarded as visitants rather than residents of this region

we find that S. enflata, S. neglecta, and the single

specimen of P. draco were all obtained from the upper

epiplankton mainly during February, 1905, when the sur-

face temperature was 15°.5 C. One surface haul made on

the morning of February 25. obtained 3,500 S. enflata,

many of which were sexually mature, 9 immature S.

neglecta, and the single very immature specimen of P.

draco. A second surface haul, made the same morning,

contained 3,100 S. enflata (most of them sexually mature)

and 75 immature S. neglecta. Six more S. enflata and 38

immature S. neglecta were obtained in a surface haul

made on April 29, 1905, and a seventh S. enflata in a sur-

face haul made on June 11, 1908. Of the remaining 5S.

enflata, 3,507 were obtained in open vertical hauls (from

ten fathoms or less) during the fall of 1904, 3,500 having

been taken in one haul. The five remaining S. neglecta

were also obtained in the same haul, and the 13 S. enflata,

No. 553] DISTRIBUTION OF THE CHASTOGNATHA 31

still unaccounted for, were all obtained in open vertical

hauls from 45, 75, 110 and 290 fathoms.

These data indicate that these three species can not be

regarded as typical of the San Diego region, and since

they occur abundantly in the surface waters of more

tropical seas where the temperature reaches 34° C., it

seems likely that they have been carried here by currents

from the warmer regions, although no such currents are

known with certainty. The probability of this supposi-

tion is somewhat increased because of their reoccurrence

here during the past winter after an absence of over two

years.

Of the remaining species, S. bipunctata, S. serrato-

dentata, and S. lyra are the most typical of the San Diego

region. The number of each species obtained from the

various depths with horizontal nets is shown in the fol-

lowing table:

TABLE III

TOTAL NUMBER OF SPECIMENS OBTAINED WITH THE HORIZONTAL NETS

Number

Depth in Fathoms S. bipunctata S. serratodentata S. lyra of Hours of

Hauling

0-25 30,733 93 5 108.1

25-75 275 106 20 11.0

75-150 10 106 20 6.5

150-250 0 174 | 17 3.1

250-350 0 43 54 5.4 ¢

This table reveals the fact that S. bipunctata was ob-

tained in by far the greatest numbers between the surface

and 25 fathoms, and that it was not taken at all below 150

fathoms. S. serratodentata, on the other hand, appeared

in greatest abundance between 150 and 250 fathoms, and

S. lyra between 250 and 350 fathoms. However, the mere

tabulation of the number of specimens taken from the

various depths does not reveal the true significance of the

data, for it is obvious, from the last column, that the

amount of hauling, and consequently the amount of water

filtered, has varied with the depth so that the relative

density or abundance in the various depths is not repre-

32. THE AMERICAN NATURALIST [Vou. XLVII

sented by the total number of specimens obtained. A

more accurate and justifiable presentation is to express

the total number of specimens obtained from each of the

above depths in terms of the average number per unit of

time consumed in hauling. The following table reveals

this relative abundance of the three species as thus deter-

mined:

TABLE IV

RELATIVE ABUNDANCE OR AVERAGE NUMBER OF SPECIMENS OBTAINED PER 20

Hours oF HAULING

Depth in Fathoms S. bipunctata S. serratodentata S. lyra

0-25 5,685 17 i a i

25-75 420 193 36

75-150 31 326 61

150-250 0 1,123 110

250-350 0 159 200

It is evident that this table brings into still more strik-

ing relief the fact that S. bipunctata is most abundant

between the surface and 25 fathoms, from where it de-

creases in abundance as the depth increases, while S.

serratodentata increases from a minimum between the

surface and 25 fathoms to a maximum between 150 and |

250 fathoms, and S. lyra increases from a minimum near

the surface to a maximum in the deepest water (250 to

350 fathoms). While it is very improbable, owing to

variations in many environmental conditions affecting the

abundance of the three species in the various depths, that

subsequent collecting would ever result in exactly the

same averages as given above—it is just as improbable

that, if the hauls were distributed in approximately the

same manner, with regard to such environmental condi-

tions, as those from which the above data were derived,

we should find the relative abundance much altered. Con-

sequently it is no exaggeration to say that each of these

three species has its own definite and specific manner of

vertical distribution just as truly as each has its own

1 As relative abundance is independent of the particular unit of time

selected for standardizing the data, a unit of 20 hours has been used instead

of the more obvious 1 hour in order to eliminate fractions in the case of

8. lyra.

No. 553] DISTRIBUTION OF THE CHA'TOGNATHA 33

specific morphological characteristics, and it would be

quite as easy to identify the species from an analysis of

data regarding its vertical distribution within an area

analogous to the San Diego region as it would from the

usual taxonomic descriptions.

Detection of specific differences in the vertical distribu-

tion of the remaining species is rendered more uncertain

because so few specimens have been obtained. However,

by taking the species one at a time, it will be seen that

tendencies, at least, toward specification are revealed.

SAGITTA PLANKTONIS

Eliminating those catches made with open vertical nets

as of little or no value in determining the depths from

which specimens were obtained, we find that seven S.

planktonis were taken between the surface and 150

fathoms, two between 150 and 200 fathoms, six between

200 and 250 fathoms, and eleven between 250 and 300

fathoms. If we separate those obtained with horizontal

from those obtained with vertical closing nets the relative

abundance of the species in these various depths may be

expressed as in the following table:

TABLE V

RELATIVE ABUNDANCE OF Sagitta planktonis

Horizontal Closing-Net Vertical Closing-Net

Catches Showing Number Catches Showing Number

of PT per 20 Hours of Specimens per 500 Fath-

ng

Depth in Fathoms auling oms of Hauli

0-150 1 none

150-200 5 1

200-250 8 8

250-350 19 -I8

This table shows that this species increases in abun-

dance as the depth increases and reaches its maximum in

the neighborhood of 300 fathoms. When we realize that

approximately the same relative abundance is obtained

from independent considerations of data supplied by

horizontal and vertical closing nets, this conclusion is

placed upon solid ground, in spite of the few specimens

dealt with.

34 THE AMERICAN NATURALIST [Vou. XLVII

SAGITTA HEXAPTERA

Some indication of the relative abundance of this

remaining species of Sagitta may be gleaned from the

following table:

TABLE VI

RELATIVE ABUNDANCE OF Sagitta hexaptera

Horizontal Closing-Net Vertical Closing-Net

Catches Showing Number Catches Showing Number

of Specimens per 100 Hours of Specimens per 1,000 Fath-

Depth in Fathoms of Hauling oms of Hauling

0-50 9 none

50-100 88 4

100-150 none 2

150-350 none none

When to the evidence contained in this table we add

that the species was not obtained in hauls made with open

vertical nets from above 45 fathoms, the facts suggest

that S. hexaptera maintains its maximum abundance

between 50 and 100 fathoms. The number of specimens,

however, is too small to afford basis for any more positive

conclusion.

KROHNITTA SUBTILIS

Regarding this species, we find that the horizontal clos-

ing nets obtained four specimens from 200 fathoms, but

none from above or below this depth. The vertical clos-

ing nets, on the other hand, obtained twelve from between

50 and 200 fathoms, 25 from between 200 and 250 fathoms,

and five from between 250 and 300 fathoms. Only three

were obtained by the open vertical nets and those in one

haul made from 250 fathoms to the surface. The follow-

ing table gives a more accurate idea of the relative abun-

dance of this species:

TABLE VII

RELATIVE ABUNDANCE OF Krohnitta subtilis BASED ON VERTICAL CLOSING

Net CATCHES

Average Number ef Specimens

Depth in Fathoms per 1,000 Fathom Haul

0-50 none

50-200 18

200-250 63

250-300 20

No. 553] DISTRIBUTION OF THE CHASTOGNATHA 35

The table indicates that K. subtilis maintains its maxi-

mum abundance between 200 and 250 fathoms, and all

the data agree that it does not occur above 50 fathoms.

EUKROHNIA HAMATA

Two specimens of this species were taken with hori-

zontal closing nets from 110 fathoms, two from 300

fathoms, and two from 350 fathoms. The vertical closing

net obtained nine from between 25 and 50 fathoms, one

from between 150 and 200 fathoms, six from between 200

and 250 fathoms, and one from between 250 and 300

fathoms. None were obtained in open vertical hauls

made from above 250 fathoms. These data show that E.

hamata is typical of the mesoplankton, and suggest that

the region of maximum abundance i is in the neighborhood

of 250 fathoms.

The essential facts presented in this brief discussion of

vertical distribution may best be summed up by classi-

fying so far as possible the various species within the San

Diego region on the basis of similarities and differences

in their manner of distribution. When this attempt is

made we find that a key somewhat as follows may be

built up:

KEY TO THE SPECIES OF CHÆTOGNATHA OF THE SAN DiEGO REGION BASED

ENTIRELY UPON Facts OF DISTRIBUTION

A. Species conspicuously epiplanktonic, very rarely extending to

$ 150 TACBOINS sa so ee es Ca ek ees opt eee es ee B.

AA. Species conspicuously mesoplanktonic, very rarely occurring

DOTO TOU TOLNOMB 2. a pn 0 ee ace cen ee eee E

B. Species confined to the upper 10 fathoms ...................... D

BB. Species veg depth of maximum abundance is below 10

FRENO aare veins o A a e ly

C. Species sain in large numbers and distributed from the

surface to 75 fathoms, but occurring in much the greatest

abundance between the surface and 25 fathoms. ..S. bipunctata.

CC. Species not occurring in large numbers, the region of greatest

dance being at least below 40 fathoms....... S. hexaptera.

D. Species occurring rarely, but in large numbers (1,000 or more

per haul not being unusual) «2. 6.66. beet ee es . enflata.

DD

)

: sana ee ae rarely and in very small numbers (more than

O per haul being wnweeil) 2056 iis. ee eel S. neglecta.

36 THE AMERICAN NATURALIST [Vou. XLVII

E. pa increasing in relative p as the depth increases,

aching a maximum at a depth of 250 fathoms or more..... fs

EE. ae RTR in relative PEATA as the depth increases,

ut reaching a maximum between 150 and 250 fathoms...... G.

F. Species ot alae common occurrence above 150 fathoms. .8S. lyra.

FF. Species whose occurrence above 150 fathoms is exceedingly

PONG © ev pe ee hats wee Sewer oe ey OC ee Ce ers 8. planktonis.

G. Species never occurring above 50 fathoms.............. ili

GG. Species never occurring above 25 fathoms .............. E. hamata.

GGG. Species occurring at irregular times above 25 fathoms, and s

times even on the surface ........--++-+s:: 8. ES

It is unnecessary to state that this key is not published

for the purpose of furnishing a ready means of identi-

fying the various species of Chetognatha. Perhaps,

when all the species from the four quarters of the globe

have been studied as critically in regard to their behavior

and ecological relations as they have in regard to their

morphology, it will be possible to construct a ready means

of identification on such a basis, but at present we can do

no more than point out that the key does work for the San

Diego region and ascertain what this fact signifies.

Its primary significance is that species are quite as dis-

tinguishable from their manner of distribution as from

their morphological characteristics. In other words, each

species has its own definite and distinctive mode of

behavior and each adapts itself to the hydrographic and

other elements of its environment in quite as definite a

way as any of the other species.

This being true, the question at once arises: To ink

extent are morphological differences between the species

proportional to, or correlatable with, their distributional

differences. Ritter (’09) has pointed out that, if

‘‘change of environment and of environed organism are

wholly and inseparably linked together,’’ one ought to be

able to measure and correlate the differentials between

organisms with the differentials between their environ-

ments. However, in attempting to find such a ‘‘necessary

correlation’’ in the case of Halocynthia johnsoni, native

to the San Diego region, and H. hauster, native to the

Washington coast, the results were negative. It is un-

No. 553] DISTRIBUTION OF THE CHÆTOGNATHA 37

necessary to point out that this is an exceedingly impor-

tant line of investigation, for, if change of environment

and of environed organism are not inseparably linked

together, the hypothesis of ‘‘natural selection,’’ with its

attendant hypotheses of ‘‘survival of the fittest,’’

‘‘struggle for existence,’’ etc., are at stake. Ask yourself

if it is not a priori impossible for any of these hypothet-

ical factors to operate in the formation of species except

on the basis of variations in structure which are more or

less adapted to the conditions of existence in which an

organism finds itself? Again, does not logic demand that,

if isolation be a necessary cause of species formation, two

similar species must occupy similar but not identical or

vastly different environmental complexes, because both

could not be equally adapted to the same conditions by

virtue of their organic difference nor to radically differ-

ent conditions by virtue of their organic similarity?

Such questions sufficiently indicate the importance of

our inquiry regarding the relation between the morpho-

logical and distributional characteristics of species and

in this connection the key reveals the fact that those

species having the most coincident vertical distribution

are those having the greatest morphological difference.

In other words, when the Chetognatha of this region are

classified in the usual taxonomic fashion, five groups can

be distinguished, of which each group contains species _

having the same fundamental morphological character-

istics; but, when classified according to similarities and

differences in vertical distribution, the species consti-

tuting any one of the five groups are those differing from

each other in fundamental distributional characteristics.

We have, then, two methods of classification, one of which

results in groups of similar morphological but dissimilar

distributional species, while the other results in groups of

Similar distributional but dissimilar morphological

species. To illustrate concretely, the groups resulting

from each method of classification are tabulated below:

38 THE AMERICAN NATURALIST [Vou. XLVII

Groups of Similar Distributional Species Groups of Similar Morphological Species

S. enflata S. enflata

1

ee S. neglecta Group 1 JS. hexaptera

S. bipunctata S. lyra

2

arn p hexaptera S. bipunctata

8. lyra Group 2 45. neglecta

G 3 :

E F, planktonis S. planktonis

S. serratodentata Group 3 {8. serratodentata

Group 42K. subtilis Group 4 {K. subtilis

E. hamata Group 5 {E. hamata

By referring to the key (p. 35) it will be seen that

S. enflata is separable from S. neglecta only by the fact

that the former occurs in large numbers while the latter

occurs in small numbers. These two species then consti-

tute Group 1 of similar distributional species, but, while

S. enflata falls in Group 1 of similar morphological

species, S. neglecta is found in Group 2. It will therefore

be worth while to see just how extensively the one species

is morphologically differentiated from the other. To this

end I have arranged, in the following table, some of the

most striking differences between the two species.

TABLE VIII

STRUCTURAL DIFFERENCES BETWEEN Sagitta enflata AND Sagitta neglecta

Structures Sagitta enflata Sagitta neglecta

Collarette. Entirely wanting. Extending nearly to the

ventral ganglion.

Anterior fin. Separated from ventral Extends to the ventral

lion.

ganglion by an interval ganglion

f

total length of animal.

Length of anterior fin. 7.4-15.9 per cent of total 18-23 per cent. of total

length of animal. length of animal.

Length of posterior 12-18 per cent. of total 21-26 per cent. of total

fin. length of animal. length of animal.

Extent of posterior Not more than half way To seminal vesicles,

fin.

Per cent. of posterior More than 50. Less than 50.

fin in front of tail-

septum.

Appearance of body. Very transparent. aqu

Width of body. 7-12 per cent. of total 4.2-6.4 per cent. of total

length of animal. length of animal.

No. 553] DISTRIBUTION OF THE CHASTOGNATHA 39

Muscles. Weak and thin. ve ~ thick.

Lateral fields. Very large

Length of tail, 16-24 per cent. ot total 280 per ae of total

length of ani ength of animal.

External process of At least 10 ea ‘aes a over 4 times longer

vestibular ridge. than broad. than broad.

Doubtless the number of differences could be increased

were we to search details, but the twelve set forth in the

above table sufficiently emphasize the fact that the two

species are fundamentally distinct from a morphological

point of view. It would, in fact, be difficult to find any

species within the genus more differentiated from S.

enflata than is S. neglecta.

Looking to Group 2 of similar distributional species

and referring to the key (p. 35) we find that S. bipunctata

is separable from S. hexaptera only by virtue of occurring

in large numbers and maintaining its maximum abun-

dance above 25 fathoms, whereas S. hexaptera occurs in

small numbers and maintains its maximum abundance

below 40 fathoms. In contrast to this we find that,

while S. hexaptera occurs in Group 1 of similar morpho-

logical species, S. bipunctata occurs in Group 2. The fol-

lowing table, therefore, reveals their most fundamental

structural differences.

TABLE IX

STRUCTURAL DIFFERENCES BETWEEN Sagitta bipunctata AND Sagitta

hexaptera

Structures Sagitta bipunctata Sagitta hexaptera

Collarette. Conspicuous but not ex- Entirely wanting.

tensive.

24-55 mm. when ma

7.3-11 per cent. of ak

Length of body. 12-17 mm. when mature.

Width of body.

Length of anterior fin.

Interval between an-

n and ven-

tral ganglio

Extent of postarloe

fin.

Vestibular ridge.

Anterior teeth.

5-7 per cent. of length.

15.9-23.7 per cent. of 8.6-11.8 per cent. of total

total length of animal. length of animal.

5-9 rarely 10 per cent. of 11.5-18.5 per cent. of to-

total length of animal. tal length of animal.

To seminal vesicles. Never to seminal vesicles,

Provided with the usual Without the usual skele-

skeletal parts. tal parts,

P

5-7 in number. 2-3 in number.

40 THE AMERICAN NATURALIST [Vou XLVII

2—4 in number

Provided with short mas-

sive crest.

12-14 in number.

Without crest.

Posterior ten

Seizing jaw

This table shows similar and just as fundamental

morphological distinctions as those found between S.

enflata and S. neglecta.

Group 3 of similar distributional species is composed

of S. lyra and S. planktonis, and on referring to the key

(p. 36) we see that the two species are distinguishable

only by the fact that S. lyra is of relatively common

occurrence above 150 fathoms. However, we find that

S. lyra is placed in Group 1 of similar morphological

species, while S. planktonis is placed in Group 2. The

following table reveals the main morphological differ-

ences between the two species.

TABLE X

STRUCTURAL DIFFERENCES BETWEEN Sagitta lyra AND Sagitta planktonis

Struct

Collarette.

Body.

Muscles.

Length of anterior fin.

Relation of anterior

fin to posterior fin.

Extent of posterior

n.

Length of tail.

Lateral fields.

Vestibular ridge.

Posterior teeth.

gitta lyre

Entirely wanting.

Translucent, nearly trans-

arent.

Tumid, but not retaining

its form we

Weak and thin.

31.4-44.5 per cent. of

total length of animal.

Confluent.

To seminal vesicles.

15.6-24.8 per cent. of to-

tal length of animal.

Large.

Skeletal parts missing.

3-9, rarely 10 in number.

Sagitta planktonis

Massive, extending to

race ‘on and

exceptionally opaque.

Firm and rigid, retain-

ing fe almost

perfectly.

Strong and thick.

18.8—27 per cent. of total

length of animal.

Separated by an interval

of 8-11 per cent. of

total length of animal.

Never to seminal vesicles.

24-38 per cent. of total

length of animal.

Very small.

Skeletal parts well de-

veloped.

11-15 in number.

Here again we find that there is no question concerning

the great morphological difference between these two

similar distributional species.

No. 553] DISTRIBUTION OF THE CHATOGNATHA 41

Finally we find that Group 4 of similar distributional

species consists of S. serratodentata, K. subtilis, and E.

hamata, and by referring to the key (p. 36) we see that

they are separable only by the fact that S. serratodentata

occurs to some extent above 25 fathoms, while E. hamata

never occurs above this depth, and K. subtilis never

occurs above 50 fathoms. Yet, we have as members of

this group three species belonging to three genera, so

that there can be no question regarding their funda-

mental morphological difference.

In what way then do these facts answer our question :

‘To what extent are morphological differences between

species proportional to, or correlatable with, their distri-

butional differences?’’ It is obvious that the only reply

permitted by our data is that there is a very definite cor-

relation, but one that is the exact reverse of what would

a priort be expected on the basis of the Darwinian theory

of ‘‘natural selection’’; namely, that the morphological

difference between two species is inversely proportional

‘to their distributional difference, or, to state it otherwise,

the coefficient of correlation between morphological and

distributional differences among species approximates

closely to —1.

RELATION BETWEEN SPECIES OBTAINED IN THE SAME HAULS

WITH Respect TO SEXUAL MATURITY

Under this head it is proposed to briefly consider the

evidence of physiological isolation or coincidence between

species relative to their maturity in those cases where

two or more were obtained in a single haul. It is obvious

that open vertical and vertical closing net hauls do not

yield data relevant to this question, for the reason that the

vertical distance covered is so great (25 fathoms or more)

that it is impossible to tell whether the specimens of two

or more species were taken from the same depth or not.

Concerning the horizontal hauls, however, this objection

can not be made, and when they are examined we find that

only 14 out of 148 surface hauls and 23 out of 108 closing-

net hauls obtained more than one species.

42 THE AMERICAN NATURALIST [Vou. XLVII

TABLE XI

SURFACE HAULS THAT OBTAINED MORE THAN ONE SPECIES

No. of

Haul Species Obtained Specimens Stage of Maturity

N tained

216 |S. bipunctata....... 200 Over 50 fully mature

S. hexaptera....... 2 Both small and very immature.

411.8. entaetea 2 ks 3,500 (Over half fully mature

SB. ROQUHE oe owe « One nearing, but none fully —

S. bipunctata....... 75 All small and very imm

ei te genta 1 (Very immature, ovary suns visible.

TORRES fan 1 Very immature, ovary not visible.

419 S. ata ioan 3,100 (Many fully mature

ls hexaptera....... 4 vee as and very i im

S. serratodentata 1 mature, phen "barely visible.

bipunctata....... 64 All's slap and imm

De noaea. OE 75 (None even E maturity.

ATS S: eniaint oe. 6 |One Pape kes rest nearly so.

B naate t io 38 |All imma

S. bipunctata....... 6 (One ein ake the others clearly im-

mature.

1,416 |S.:enflata. 2.0... .. Nearly but not quite mature.

S. bipunctata....... 1,620 |All stages, many fully mature.

1,422 |S. bipunctata....... 9 Several stages, one fully mature.

S. planktonis....... 1 |Very immature, ovaries invisible.

1,426 S. serratodentata 1 All small and immatur

S. bipunctata....... 1,250 (|All prs many fully wk

1,582 i AE 7 |All small and immature.

EEN a 600 (All stages, 25 mature or nearly so.

1,591 5S. a 5 |All small and immature.

Bear 200 (All stages, but mostly immature.

1,605 S. oka ress Small and very immature.

ire oes 35 (Mostly immature

1,686 pi hexaptera........ 1 (Small and very imma

. bipunctata....... 1,600 (All stages, over 200 fal. mature.

1,716 A esi oie ore re 1 mall and very immature

ma bipunctata....... 50 Mostly immature, one fully mature.

1,738 = ne stoasvot gga sarge 14 (All small and immature

S. bipunctata....... 105 All prs some fully mature.

1,772 S. pec 1 Small and immatur

bipunctata....... 9 |All small, none fully. mature.

Concerning the 14 surface hauls, the following table

reveals the fact that in only one haul (1,416) were

representatives of two species taken which were nearly

mature, and in this case the one specimen of S. enflata

No. 553] DISTRIBUTION OF THE CHASTOGNATHA 43

did not appear to be fully mature. In-every other haul

only one species was represented by sexually mature

individuals.

The following table, which contains data relative to

hauls made with horizontal closing nets, shows no in-

stance of two species having been taken in the same haul

both of which were represented by sexually mature indi-

viduals.

The facts revealed in Tables XI and XII, when taken

together with the foregoing discussion of vertical distri-

bution, suggest that the various species reach maturity

for the most part during different seasons, and that fer-

tilization probably takes place in different strata of

water according to the species. In the case of S. enjflata,

for instance, fertilization unquestionably takes place be-

tween the surface and ten fathoms and then only during

the winter, if at all, in the San Diego region. With S.

bipunctata, on the other hand, evidence is at hand [see

Michael (’11)], which space forbids presenting here,

showing that the species maintains a ‘‘center of migra-

tion’’ between 15 and 20 fathoms, from which center the

Species moves up and down in response to variations in

light, temperature, salinity and other factors of its en-

vironment, which facts indicate that fertilization is

mainly, if not exclusively, confined to this depth of 15 to

20 fathoms. In the case of S. hexaptera only the imma-

ture have been taken above 50 fathoms, which shows that

fertilization must take place below this depth. Again,

only the very immature of S. serratodentata have been

taken above 100 fathoms, except at night when the larger

specimens ascend to 50 fathoms. Similarly with S. lyra,

the larger more nearly mature specimens do not occur

above 200 fathoms to any extent, and so on with the other

species,

It is quite true that much more knowledge is needed

concerning the vertical distribution of most of the species

before positive conclusions relative to the depth at which

fertilization occurs can be advanced. Were the deeper

water (below 350 fathoms) thoroughly investigated, dif-

44

HORIZONTAL CLOSING NeT HAULS THAT OBTAINED MORE THAN ONE SPECIES

THE AMERICAN NATURALIST

TABLE XII

[Vou. XLVII

No.

Depth Bpel:

Haul linFath-| Species Obtained mens Stage of Maturity

No. oms Ob-

tained

1,873 BSG. cee Fes 1 (Small and very immature.

5. e ik PE A AE 5 |3 fully mature, 2 immature.

1,877 15 18. sorore: es 4 |All small and immature.

S: bipunctata....... 6 |All small and immature.

1,748 S5 Sae 2 One ee. 5 |All very small and immature

S. serratodentata.... 6 One any mature, the rest very im-

matur

S. bipunctata. ...... 4 Small a remote from maturity.

1,761 25 S a cae ea fea a 1 |Very small and imm

S. Saan PA ae 1 la but not fully nant

1,851 25 |S. rite Eg ee 7 |All small and imma

S. bipunctata....... 2 micas but not ‘ally cB

1,858 35 S. serratodentata... 2 |Small and immature

S. bipunctata....... 1 [Nearly but not fully mature.

1,476 | 50 apma aa 1 |Remote from maturity.

S. bipunctata....... 300 |All stages, several mature.

nktonie.....<> 1 [Remote nea maturity.

1,575 75 |S. serratodentata 76 |Mostly large and nearly mature.

S. bipunctata....... 65 ? Fog ? ?

1,688.) 100 (Solara ee er ee 4 |Very small and im ure,

S. serratodentata 4 |One nearly cues ye rest remote '

from maturity.

ETIS 200 S ia.. 1 (Small and immature

S. serra 9 [One sopini mature the rest small and

immat

1,767) 100 S tre. 22 eee 1 [Large and nearly mature.

S. ieai EE, 2 ? ? ? T ?

1,813 100 iS. tyra. 6c. fo a 2 |Large but not m

S. rA 12 |All large but ene tel mature.

1,979 | 100 iS. tree evince 1 (Large but not mature.

S. serratodentata 30 |All stages, none fully mature.

RST (425. i5. Bee. eS 11 (All stages, ve nearly ee ay

S. serratodentata 21 |All eer ne fully m

S. BONIS. cs 1 Small see vies ‘a de eer

Torso i 100 IS BG. 6 oi 1 (Small and ve mature

S. serratodentata 16 |All large and danrin mature.

1,926; 200. iS. yra: S 13 |All stages, none fully mature.

S. serratodentata 46 |All , Some nearly mature. į

S. planktonis....... 1 (Large bat very immature.

ee SURG ok 3 ‘Very immature.

No. 553] DISTRIBUTION OF THE CHA'TOGNATHA 45

No. of

Haul Depth Speci-

No. inFath-| Species Obtained mens Stage of Maturity

o. oms

tained

TaS 200 IS. Brans as 2 (Large but not matu

S. serratodentata....| 13 (Large, some nearly ig

ESI 20018; Ge. e n 1 (Large but not mature.

S. seih R A ...| 29 Large and nearly mature.

PART T T POA E A 1 (Very immature.

23732) | 220" iS. UN Cees: 1 (Large but imma

S. serratodentata....| 13 (Large and HO Aa ctr.

ODT- BOO Wy APO: os ale ee 50 |All stages, several nearly if not fully

mature.

S. serratodentata....| 24 |Large but immature.

E720 250 SR: es 2 |Small and immature.

S. serratodentata 17 (Large, some nearly mature.

1,550 + 350 S.. Um.: osii 2 |Large but not mature.

S. Siati bea 3 |Large and nearly mature.

1,567 | 350 |S. serratodentata 2 mall and immature.

B: amen eo iG 2 [Large and nearly mature.

ferentials would undoubtedly be established between S.

lyra and S. planktonis with reference to their depths of

maximum abundance. Again, more extensive data rela-

tive to all depths, by increasing the number of specimens

and hauls with which to deal, would enable us to split the

region of 250 to 350 fathoms, for instance, and thus dis-

tinguish strata and establish specific differences within

much smaller limits. However, on the basis of data al-

ready accumulated, all the evidence points toward the

probability that the fertilization of each species nor-

mally takes place within the limits of its depth of maxi-

mum abundance.

SUMMARY AND GENERAL SIGNIFICANCE OF THE’ DATA-

In the foregoing pages the following facts relative to

the question of isolation and coincidence have been re-

vealed:

1. Of the most closely related ‘‘couplets’’ of species

only one occurs in the San Diego region, except in the

case of S. enflata and S. hexaptera, the former of which

can not be regarded as resident in this vicinity.

46 THE AMERICAN NATURALIST [Vou. XLVII

2. The general distribution of each member of a

‘‘eouplet’’ is never entirely coincident with that of the

other, but varies from a contiguous and overlapping

to a radically isolated distribution, according to the

‘“eouplet.”’

3. Each species occurring in the San Diego region has

its own definite and specific manner of vertical distribu-

tion just as truly as it has its own specific morphological

characteristics.

4. Those San Diego species having the most similar

vertical distribution are those possessing the most dis-

tinctive morphological characteristics or, to state it

otherwise, the morphological difference between species

is inversely proportional to their distributional dif-

ference.

5. Whenever two or more species have been obtained

in the same haul, never more than one was representd by

sexually mature individuals.

6. With one or two possible exceptions, the mature

specimens of each species occur in different strata of

water.

On the basis of these facts we are forced to conclude

(a) that the more closely related species of Chetognatha

are isolated from each other either horizontally, ver-

tically or by virtue of physiological differences causing

fertilization to take place in different strata of water,

and (b) that ‘‘Jordan’s Law”’ is only partly true, when

tested by vertical distribution, for, while the more closely

related species do not inhabit the same environment,

they do inhabit the most remote environments.

Aside from these obvious conclusions the primary

significance of this paper is that of emphasizing the need

of more exhaustive and quantitative data relative to or-

ganisms, on the one hand, and their environments, on

the other, before any solid basis can be had upon which

to build theories regarding the operation of isolation,

adaptation, natural selection, mutation and other factors

supposedly concerned in the evolution of species. The

No. 553] DISTRIBUTION OF THE CHÆTOGNATHA 47

existence of this need relative to pelagic organisms and

their conditions of vertical distribution is readily recog-

nized and our first impression may be that the extent of

this particular need is exceptional, but an acquaintance

with the literature on evolution shows plainly that the

need is very general. Indeed, this literature is fairly

bulging with evidences of mimicry, protective coloration,

natural selection, etc., based upon an abundance of data

concerning many organisms as well as their environ-

ments, which data, while supporting the hypotheses,

rarely include any facts relative to the essentially quan-

titative nature of either the organisms or the environ-

ments investigated. The mere fact that this sort of data

supports a hypothesis and that the logic is sound is not

adequate scientific proof that the hypothesis is true, for,

as Pearl (’11) and others have demonstrated, logic may

carry conviction, be supported by numerous data, and

still prove erroneous when the quantitative relations of

the facts included in such data are considered. Therein

lies the mischief of much of our a priori reasoning rela-

tive to evolution, namely, that it causes us to depend so

largely upon logic that we overlook or neglect as insig-

nificant the quantitative nature of organisms and par-

ticularly of environments. Our most urgent present

need, therefore, is not so much the accumulation of addi-

tional qualitative data as it is an exhaustive and quanti-.

tative treatment of those facts now at hand.

While the biometrician and, to some extent, other stu-

dents of evolution are treating their data quantitatively,

the ease with which large numbers of individuals of

pelagic species may be obtained without apparently di-

minishing the supply, gives an unusual opportunity to

the marine biologist for applying quantitative methods

on an extensive scale to many of the important problems

of evolution. If all the planktological expeditions would

join hands by publishing all the data relative to every

haul (those that did not as well as those that did contain

the species or group under consideration) and by record-

ing the approximate, if not the exact, number of speci-

48 THE AMERICAN NATURALIST [Vou XLV11

mens of each species obtained, instead of publishing only

data relative to successful hauls and recording the

species as abundant or rare, many problems now so

largely discussed from hypothetical points of view could

be analyzed entirely on a factual basis without involving

committal to any hypothesis whatsoever. For pa anay

by such means it would be possible:

1. To measure the degree of variation in the habite of

distribution of species.

2. To measure the extent of correlation between varia-

tion in the vertical distribution of species and variation

in their horizontal distribution.

3. To measure the degree of correlation between

morphological and ecological characteristics of species,

and so arrive at an accurate analysis of the causes of

adaptation.

4. To measure the range of adaptation accompanying

the same structures.

5. To measure the range of variation in structure

adapted to the same environmental conditions.

6. To analyze the natural behavior of a species with-

out involving the necessity of first placing collected indi-

viduals under the artificial conditions of the laboratory

and then reading the results, arrived at by experiment,

back into their natural environment. I do not wish to

minimize in the least the immense value of laboratory

experiments on behavior, but, no matter how great the

achievement, such experiments can not afford a reliable

basis for interpreting the natural behavior of a species

until it becomes possible to re-create nature in miniature.

While these are but a few of the problems that are

urgently calling for solution, I can not help but feel that,

in the foregoing pages, we have touched the fringe of a

line of quantitative investigation destined to yield much

of importance to the student of evolution.

BIBLIOGRAPHY

Aida, T.

’97. The Chætognatha of Misaki Harbor. Annot. Zool. Japon., Vol. 1,

pp. 79-81, pl. 4.

No. 553] DISTRIBUTION OF THE CHATOGNATHA 49

Doncaster, L.

702. nenas hens a Note on the Variation and Distribution of the

oup. Fauna and Geogr. Maldive-Laccadive Archip., Vol. 1, pp.

sie: —218, pl. Pi figs. 39-40 in text.

Fowler, G. H.

705. peda a Collected During a Cruise of H.M.S. Research,

9 . The Chetognatha. Trans. Linn. Soc. London,

Ser. 2, va 10, pp. 55-87, pls. 4-7.

706. The Chetognatha of the Siboga Expedition, with a Discussion of

the Synonymy and Distribution of the Group. Siboga Exped.

onogr. No. 21, 86 pp., 3 pls., 6 maps.

Jordan, D. 8.

705. The Origin of Species through Isolation. Sci., N. S., Vol. 22, pp.

545-562.

Kofoid, C. A.

707. The Coincident Distribution of Related Species of Pelagic Organ-

isms as Illustrated ed the Chetognatha. AMER. Nat., Vol. 41,

pp. 241-251

Michael, E. L.

’11. Classification and Vertical Distribution of the Chætognatha of the

San Diego Region, Including Redescriptions of Some Doubtful

Species of the Group. Univ. Calif. Publ. Zool., Vol. 8, pp. 21-178,

pls. 1-8, 1 fig. in text.

R.

’11. Data on the Relative Conspicuousness of Barred and Self-colored

Fowls. AmER. Nat., Vol. 45, pp. 107-117, figs. 1-4 in te

Ritter, W. E;

’09. Halocynthia johnsoni n. sp. A Comprehensive Inquiry as to the

Extent of Law and Order that Prevails in a Single Animal

Species. Univ. Calif. Publ. Zool., Vol. 6, pp. 65-114, pls. 7-14.

Von Ritter-Záhony, R

%08. Chätognathen. Ber. Comm. Erforsch. östl. Mittelmeer, Zool.

Ergebn. z 18 pp, 1 pl.

09. Expeditionen S.M. Schiff Pola in das Rote Meer nördliche und

siidliche Sap 1895-96, 1897-98. Ber. Comm. ozeanogr. Forsch.,

Zool. Er 2 pp., 4 fig.

10a. Die Fauna Südwest- PEE DE Lief. 3—Chetognatha. Ergebn.

Hamburg siidw-austral. Forsch., Bd. 3, pp. 125-126.

106. Chetognatha from the Coasts of Ireland. Fisheries Ireland.

Sci. Invest., 1910, No. 4. 7 pp.

"lla. Die Chiitognathen der Plankton-Expedition. Ergebn. d. Plank-

ton-Exped. d. Humboldstiftung, Bd. 2, 33 pp., 11 figg. im text.

*11b. Revision der Chiitognathen. Deutsche Siidpolar-Exped., Bd. 13.

Zool. 5, 71 pp., 50 figg. im tex

Wagner, M.

68. Ueber die Darwinische Theorie in Bezug auf die — Al

breitung der Organismen. Sitzber. d. bayer. Akad.

München, Bd. 1, pp. 359-395.

—

A FAMILY OF SPOTTED NEGROES

Q. I. SIMPSON AND W. E. CASTLE’

Ir is the purpose of this note to put on record an inter-

esting variation in human skin color which made its ap-

pearance as a mutation or sport in a negro family of the

southern United States some sixty years ago and has

shown itself fully hereditary through two generations of

offspring. The nature of the variation is shown in Figs.

1-4. It consists of a ‘‘piebald’’ condition of the skin.

which is spotted with white in a fairly definite pattern,

not? unlike that of certain domesticated animals. A more-

or-less continuous white area begins on the top of the

head, which has a crest of white hair, extends down over

the face (where, however, it may be interrupted) and

broadens out on the chest, which is either entirely white

or finely mottled. Im the whitest individuals the chest

area extends around the sides of the body on to the back

(see Fig. 4), but fails to reach the mid-dorsal line. It

also extends on to the arms in like proportion to its ex-

tension elsewhere on the body, but the lower forearm and

hands, like the feet, are in all observed cases dark. The

ventral white area continues downward from the waist

line, and in at least one case (Fig. 4) covers the legs,

which are nearly free from black spots down to the knees.

There larger and more numerous specks of black begin,

which become continuous above the ankles.

If we should describe the pattern in terms of its black

1 The material on which this paper is based was collected by the senior

author; the junior author has merely assisted in preparing the material

for publication

2 A photograph in our possession of the same four individuals shown in

Fig. 1 together with the father of the three children, taken when the children

were small, but now too faded for successful _Teproduetion, makes it clear

vening period. As in other piebald mammals the pigmented areas have

definite boundaries fixed at birth and not subsequently changeable.

No. 553] A FAMILY OF SPOTTED NEGROES 51

Fic. 1. Mrs. Eliza D., her sons Jim and Robert (the taller one) and

daughter, Lillie. Photographed 1910.

areas, we should mention as its most prominent feature

the back-stripe (Fig. 4) which begins on the head and

extends the entire length of the trunk, narrowing below

and ending on the buttocks. In the taller son, Robert,

52 THE AMERICAN NATURALIST

Fic. 2. Back view of Mrs. Eliza D.,

seen in front view in)Fig. 1

and states positively that

there were no spotted

negroes previously in that

region. The colored skin

of Mrs. S. A. is ‘‘medium

dark’’ as is that of her

husband, who is entirely

normal in appearance,

being free from spots.

The pair were married

in 1868 and have had fif-

teen children, all of whom

are living, a fact which

indicates a healthy vigor-

ous stock. Of the cebi-

dren eight are spotted like

the mother, the remaining

seven being normal, with-

out spots, but varying in

[Vou. XLVII

Fig. 1, the back-stripe is

so wide that it covers the

sides of the body also.

The original mutant,

founder of this line of

spotted negroes, Mrs. S.

A., is still living. She

was born in 1853 in Loui-

siana, both her parents

being normally colored

negroes, the father

‘‘dark,’’ according to

the statement of her

husband who grew up in

the same neighborhood

1

Fic. 3. Lillie, daughter of Mrs. Eliza D.

Compare Fig 1

No. 553] A FAMILY OF SPOTTED NEGROES 53

depth of pigmentation, as is usual in mulatto families.

The pigmentation of spotted children and grandchildren

Fig. 4. Back view of Jim, seen in front

view, at the left of Fig. 1.

likewise varies in intensity from light mulatto to coal

black. The white spots are however in all cases entirely

devoid of pigment.

54 THE AMERICAN NATURALIST [Vou XLVII

Six of the fifteen children of Mr. and Mrs. S. A., three

normal and three spotted, married normal negro mates

and have had from two to four children each. The nor-

mals have had only normal children, in all seven. The

spotted ones have had nine spotted and two normal

children.

The normal children of Mr. and Mrs. S. A. who mar-

ried consisted of two daughters and one son; the spotted

ones consisted of two sons and one daughter. There is

evidently no sex-limitation in the transmission of the

spotted pattern, which behaves consistently as a simple

Mendelian dominant character, the only peculiarity of

the case being the excess of spotted grandchildren over

the expected one half. But this quite probably isa chance -

deviation due to the small numbers under consideration,

or to failure to secure as complete a report of the un-

spotted as of the spotted grandchildren.

The descendants of Mr. and Mrs. S. A. are now widely

seattered through the United States and Europe, certain

of the spotted ones being connected with ‘‘museums.”’

Their peculiarity is therefore an economic asset and not

likely to interfere with their racial increase. The indi-

viduals thus far produced are clearly from their parent-

age all heterozygous for the spotted character, which

they transmit in half only of their germ-cells. If in the

course of time two spotted individuals of this race, not

closely related, should marry each other, we might on

Mendelian principles expect the production of a new type

of individual, one homozygous in spotting, which would

transmit the character in all its germ-cells. What the

somatic character of such an individual would be we can

at present only conjecture. Our experience with the

domesticated animals leads us to think that it certainly

would not be an albino with pink eyes and unpigmented

or faintly pigmented skin, since true albinism is genet-

ically entirely distinct from spotting with white and is

recessive in heredity whereas this character is dominant.

More likely it would resemble ‘‘black-eyed whites’’ such

No. 553] A FAMILY OF SPOTTED NEGROES 55

as occur among mice, rabbits, guinea-pigs, cats, dogs,

eattle and horses. Our experience with these animals

would lead us to expect that the homozygote in this strain

of spotted negroes would be either wholly white, that is,

with snow-white skin and hair but with colored eyes, or

spotted but with pigmented areas still further reduced in

extent than in the heterozygote. Some student of genetics

generations hence may be able to answer the question.

To this end we shall deposit with the Eugenics Record

Office at Cold Spring Harbor, N. Y., our original data

including the correct names and present whereabouts of

these people.

Three of the spotted children of this family, of whom

we have been unable to secure pictures, are undoubtedly

identical with ‘‘The Three Striped Graces” figured

(Plate VV) and described (p. 248) by Pearson, Nettle-

ship and Usher in ‘‘A Monograph of Albinism in Man,”

London, 1911, after Hutchinson, British Medical Journal,

June, 1910, p. 1480. The names given by Pearson, et al.,

for the three individuals are ‘‘Mary, Rose and Fanny,”

which agree sufficiently well with individuals VII, VIII

and X, of our table. Our own information obtained from

members of the family indicates that at present VII is in.

America, while VIII, X and XIV together with the

grandchild, Beatrice, are in Europe.

TABLE

DESCENDANTS OF Mr. AND Mrs. S. A., THE FORMER A NORMAL,

THE LATTER A SPOTTED NEGRO

Children Grandchildren

I. Mrs. Eliza D., spotted, 1. Spotted son (dead) ;

Figs. 1 and 2; 2. Spotted, Jim (pigment dark),

Mate, mulatto. Figs. 1 and 4;

3. Spotted, Robert (pigment

light), taller son, Fig. 1;

4. Spotted, Lillie (pigment me-

dium dark), Figs. 1, right,

- and 3.

II. Mrs. Eugenia —, normal; Two normal.

Mate, colored.

THE AMERICAN NATURALIST [Vou. XLVII

Mr. Horace A., spotted; 1. Normal, light brown;

First. mate, light. mulatto; 2. Spotted, brown;

Second mate, black. 1. Spotted, black

Mr. Jake A., normal; Three normal.

Mate, colored.

r. John A., spotted; 1, Spotted, Beatrice;

Mate, dark. 2. Normal;

3. Spotted.

Mrs. Jane —, normal; Two normal.

Hattie, normal.

THE EFFECT OF FERTILIZERS ON VARIATION IN

CORN AND BEANS

J. K. SHAW

MASSACHUSETTS EXPERIMENT STATION

THE data here reported were secured in the summer of 1909

from a field of sweet corn and beans which were fertilized with

nitrogen phosphorus and potash separately and in combination

after the manner described later. The original purpose of the

investigation was to determine if the differences caused by fer-

tilization were in any degree transmitted to succeeding genera-

tions. Owing to development of other work it has been impos-

sible to carry this on as planned. It is thought that the data se-

cured may have sufficient interest and value to warrant their Ere

sentation.

The plot of land used appeared much exhausted of both plant

food and humus. It had previously been used as a raspberry

patch. It lay on a gentle southeastern slope sheltered on the

opposite side by a belt of woods which, however, was far enough

distant to prevent injury from shade or root trespass.

The field was rectangular in shape, 300 feet long and 60 feet

wide. It was divided crosswise into twelve plots, each 25 x 60

feet. The fertilizers and their amounts were as follows:

Plot Lbs. Plot Lbs,

1... Nitrate of Soda ......... 12 10. Check

2. Check -of Soda 12

3. P itrato of Soda ......... i IL [asa r Phosphate ......... 30

Acid Phosphate ......... 30 SR of Potash ...... 8

£ Acid Phosphate ....:.... 30 E .. 500

jeu FP HOSUBRGE 4 ea ea os 30 12 bias of Potash ...... 10

` ĮSulphate of Potash ...... 8 1 t Acid Phosphate ......... 40

6. Check Nitrate of Soda ......... 16

7. Sulphate of Potash ...... 8

8 Nitrate of Soda ......... 12

*)Sulphate of Potash ...... 8

Nitrate of Soda ......... 8

9.4 Acid Phosphate ......... 20

Sulphate of Potash ...... 6

THE AMERICAN NATURALIST

Corn

of Soda -

2 |lbs

30’

Cheich

v

Nitrate £

Acid Phos

Soda 12 lbs

phate 301lbs.

Acid Phosphate 30 ibs

Q

Acid Phos

Sulphate

phate 30 Ibs.

Potash 8 lbs

Gheqh

Sulphate

Dat asch Bibs

œ

Nitrate of

Sulphate

Soda 12 lbs

Potash 81bs.

Nitrate of

Sulphate

Soda 8&lbs

ate 20/lbs.

Potash 6ibs

Gheich

—

a

Nitrate of

Acid Phosphate 30

Sul phate

Soda 121lbs

l bz.

Potash 3lbs.

Manure

[Vou. XLVII

It is seen that plots 1, 3, 4, 5, 7, 8

and 11 received the several fertilizing

materials singly and in combination

in what may be called normal

amounts of 12 pounds nitrate of

soda, 30 pounds acid phosphate and

8 pounds sulphate of potash; plot 9

received all these in considerably

smaller amounts; while in plot 12

an excess of the chemicals was applied

together with a liberal quantity of

ordinary barnyard manure, the idea

being to supply here a maximum of

the stimulus that ordinary fertiliz-

ing materials may be expected to af-

ford. The lower half of the entire

plot was planted to corn and the

upper half to beans. These facts

are graphically shown in Fig. 1.

One half of the nitrate and the entire

amounts of the other materials were

applied broadeast on June 8 and har-

rowed in. The corn was of the Red

Cory variety and was planted June

14, and the Davis white wax beans

the next day in drills running

lengthwise of the area. Both were

thinned where necessary to give the

corn space of four inches and the

beans three inches. The remainder

of the nitrate was applied about three

weeks after planting, but the exact

date can not be given. The plants

were cared for in the usual way and

the measurements taken from Sep-

tember 14 to 16, or possibly a few

days later. With the corn a record

was made of the following: distance

from ground to uppermost ear, dis-

tance from ground to lowest branch

of tassel, and the number of ears

formed. Where ear-bearing suckers

No. 553] VARIATION IN CORN AND BEANS 59

occurred, the same records were made in such a manner as to

indicate their parent stalks. The plants on space of two feet on

each end of the plots were omitted in the measurements.

TABLE I

EFFECT OF. FERTILIZER ON YIELD OF CORN

Barren 2-Eared Per Cent,

Plot Total Stalks, Stalks, | Ear-bearin

Stalks Per Cent. | PerCent. | Suckers

i: ee OU S00R ee 241 2.90 | 1.66 3.32

Si Cheek i oe ae tg 248 43.15 40 40

Nit of soda

3. E Aa E pE TA oy 281 5.34 4.27 7.47

4. Acid TRT rarer a aagenee © 244 44.26 -60 Al

Acid phosp

{ oe of Sotak } EE aes a S5 | a Br

6. DOC TEA E PAE E e EE 251 29.87 .00 40

fe arse Se Of potash. 223) 3 314 28.03 -00 .64

Nitrate of soda

See a ee ee 269 4.83 3.35 7.06

itrate of soda

9.5 Acid phosphate F.. oa. 255 9.80 1.57 5.88

Sulphate of potash

($ norm

OO or eee ae ee re 225 28.44 44 .88

Nitrate of soda

11.4 Acid phosphate |+........... 241 7.05 3.73 10.37

lphate of potash

(no amounts)

Nitrate of a

Acid pope

12. ap te of potash } ........... 248 4.43 | 10.09 | 39.92

(in kolas

an

Aves Checking o n hee es 33.82 28 56

Table I gives figures bearing on the productiveness of the

corn, and indicates that the deficient element was nitrogen.*

There is no indication that the addition of potash or phosphorus

decreased the number of barren stalks at all, nor did either

alone increase the number of two-eared stalks of ear-bearing

suckers, though there appears to be considerable benefit from

each when applied with nitrogen, and still more when all three

are supplied. The addition of manure results beneficially in all

ways, possibly on account of its physical action as well as by the

direct addition of plant food.

* While the discussion following is in terms of the elements of fertility,

it is of course possible that other carriers of the same elements might have

given different results. has seemed simpler to express the matter in

terms of the elements and with a full reading of the text no misunderstand-

ing on this point is possible.

[Vor. XLVII

THE AMERICAN NATURALIST

00

96'FI $99 98 FF 126% 67'e Ce Tt eo le ee ee ee ee ee e S00 © 6 +612 ao yoy 4 ‘AV

ainuvyy

(s890xe UT)

88+ OSZI GG * GS L Te" = 99°99 OV T+ 98°SE CT’ L6'P E EENE A See as 2 eg ee ysvjod jo oyeydyng } “zy

ayeydsoyd poy

BPOSs JO IJBIMN

(syunoure ewou)

; ; ; 4 5 5 ; i ; : ‘ ihe ds Gis eek as Tii ysvjod jo ayeydyng | 'II

PP = 6S'ST 6G F 869 Of SEIS SO'T = 62°08 el + SOF Sie £6 Ct ayeydsoyd poy

Bpos JO Oo] BIJIN,

OS" = TEST 1° F SP'9 62° = TZP TI SL'TE GI 82'S OU ESO ee E ey ee eee 8

a ean p

. . , . . é . y ; á . A E r Se 8 ysvyzod jo oyeydng

EF FOP IF GOL O&' = 6F 67 ELT POZE el’ =OP 61° ZOST ayeydsoyd poy [6

BpOs JO 94BIJIN

. . 3 ` : A : ý ; ` A e Ls eee s auo uss yseyjod jo eyeydng ) ,

OF + 99°EI correo | 8Z'=9F09 | 96 #9708 | GI+STh | SI ESET wpos jo ovemtny | 8

82°ST 61° O12 L6°* ZO LV 66° = Z067 Gl’ = L9°E LV 19'GT ee ee a ysvyod jo oyeyding '4

Sv FLP 02° + FL'9 66° = IL’SP LO°'T* 99°23 GIF GSE Mt tt ee neg eels oe es bata ‘9

. ‘ ` ‘ ‘ : : ‘ : ‘ ` ER E A R a Ses os & yseyod jo oyeyding ) ,

€S° = €€°9T GS EOL TS" = LO’EP IGT L'8S El FOEL 61° + 69°IT ayeydsoyd poy G

SF == 99ST og: = LOL 62° = OL FF FL = CZ'SZ FL = 6£'E 0z = 00°ZT Wee) ot ks ee a ae ee ee Oe E R » T poy ‘$

° $ J i ; ; : ‘ ; ‘ ; ee E E U ck tobe as see eyeydsoyd poy

oe" + SUSI ST'=PT9 | Seesros | LOTFZeee | Freeh | LV *Slel gee jo oveayty f È

OF = PS'FL 02° = ZL‘9 62° = 92°SF EAT = Z987 PT + 98° 6° ES 8Z'II sr 4 0 86 O58 64 08) FERS 6 6 6 Det tod 0 to 4 eS yoo z

OF = GST ST’ + 08'S GSF 89°9F 94° T+ 8ST'9E El’ + 02°F Gre 4t Ue ce eae ee ante Bpos JO IBIN ‘I

AQypiquise uone q AYT[IQBIIBA nonerag

jo anio go prepuvS uH Jo UATE) prepunys TOR ii

1439H AVIS 1439H 184

SINVIG NYOD JO AAALVIQ NI NOLLVIUV A

It Wiavi

VARIATION IN CORN AND BEANS 61

No. 553]

‘HIOD PUL suvəg UT UOMBVELA ‘Z ‘DI

H = suvaqd

745174

4) $|

216

YIPIS - Y409

G| Og] a

spod 7219

eer

24 | oir 6 Ç

oye

|

N

N

291?

SHSHSH SHG

RPessaa ss”

oS

“Ss

=

N =

We)

SN

SH oH SF

< 46 a Typ 3

fe BE

te

“«

62 THE AMERICAN NATURALIST [Vou. XLVII

Table II shows the variation in stature of the corn plants.

With respect to ear height it is seen that the addition of the

mineral elements, especially if nitrogen was also supplied, ap-

parently raised the mean. It will be remembered that ear height

is taken to the uppermost ear in two-eared stalks, and a com-

parison with Table I indicates that the ear height is significantly

greater only where there is a considerable percentage of two-

eared stalks, making it very probable that none of the fertilizers

influenced in any marked degree the height of the ear on single-

eared stalks, or of the lower ear where two were present.

The differences in variability of ear height so far as they are

significant are probably to be referred to the same cause, an in-

creased number of two-eared stalks resulting in greater varia-

bility, as would be expected.

TABLE III

VARIATION IN YIELD OF BEANS

Pods per Vine

Plot Coefficient of

Mewes Damia Variability

1 Witsate of sodas. <4. 0 8.63+.14 | 3.49+.10 | 40.441+1.36

E ak o a 4.81=.14 | 3.18+.10 | 66.11 +3.21

3. { PEE E \ Se ges 9.62+.23 | 5.25+.16 | 54.42 +2.43

4. Acid phosphate................ 7.61+.18 | 3.94+.13 | 51.77 +1.60

gir

5. ra eov sono. 10.71+.25 | 5.05.17 | 47.15 +1.93

6 Chok oe ee ao 9.13.22 | 4.34.15 | 47.53 +2.02

t. Sulphate of potash... 2. 65... 52. 10.09 +.18 3.76+.13 | 37.26 +1.41

Nitra

Slasti a 9.82+.16 | 3.59+.11 | 36.56 +1.30

Nitrate of so

a

9. a or ot ki erete 12.26+.21 | 4.39+.15 | 35.79 +1.36

{ ( s mal |

10, Cheek a a 6.75+.16 | 2.96+.11 | 43.85 +1.93

| Mie of ie |

id p j

11. Sulphat Bee oa h Ba. 13.20+.21 | 5.01+.15 | 37.95 =1.30

no amounts) |

Nitrate of soda

Acid phosphate |

12. { Sulphate of potash } ........... 15.34 +.30 6.09+.22 38.38 +1.54

(in exe |

Manure

POTE OE oe 6.89 3.49 52.49

Fj

There is little evidence that either of the mineral elements

alone or both together increased the stature of the whole plant,

No. 553] VARIATION IN CORN AND BEANS 63

though nitrogen did have this effect. Nitrogen together with

either phosphorus or potash had a more pronounced effect, and

when all three were applied together in plot 11 the stature was

still greater. Increasing the amounts of the commercial fertil-

izers and adding manure in plot 12 gave still taller corn.

The standard deviation is apparently increased by the mineral

elements either alone or together, while nitrogen operates to

lessen this measure of variability. When all are used together

both influences seem to operate and the standard deviation is

increased, but not so much as with the mineral elements alone.

The increase of the standard deviation where considerable

amounts of complete fertilizer is added is not in proportion to

the increased mean, and the coefficient of variability is lessened.

It was not found possible to make any measurements bearing

on the vegetative vigor of the bean plants; the only figures avail-

able are those of yield as measured by the total number of pods

on each plant. Table III indicates that potash was most effec-

tive in increasing the mean number of pods per vine, though it

will be remembered that this element was of the least avail with

corn. Nitrogen seems next, and there is a possible beneficial

effect from phosphorus. When all are applied the average is

higher even on the plot receiving only two thirds the normal

amount. Increasing the quantity of chemicals and adding ma-

nure increase still further the yield. The results are not as

consistent as in the ease of stalk height of corn. This variability

in yield finds further expression in the larger variation within

each plot as expressed by the coefficients of variability which

are nearly three times as great as those for stalk height of corn.

The effect of the single elements on the standard deviation is

not very clear, but it seems that two or more together increase

it, but if nitrogen is present this increase is not in proportion to

the increased mean and the coefficient of variability is lessened.

There is some indication that potash has a similar effect. Owing

to the uncertainty of pod setting a great many data are neces-

sary in order to give definite information on these points.

In Fig. 2 is shown graphically the difference in the effect of

the fertilizer on yield of beans and stalk height of corn. This

shows that fertilizers greatly extended the range of variation of

total pods per vine; the minimum is only slightly raised, but

there are many more plants with a considerable number of pods

and the maximum number of pods per vine too is markedly in-

64 THE AMERICAN NATURALIST [Vou. XLVI1

creased. There is a greater ‘‘scatter’’ to the distribution. With

the stalk height of corn the ‘‘seatter’’ is not much increased, but

the whole polygon is moved bodily to a higher position. The fer-

tilizer had a direct effect on every stalk of corn, increasing its

stature, but it had little effect on the productiveness of a few of

the bean plants, though increasing that of most of them.

J. K. SHaw

MASSACHUSETTS EXPERIMENT STATION

THE SNAKES OF SOUTH AFRICA

Their Venom,

and The Treatment of Snake Bite

By F. W. FITZSIMONS, F.Z.S., F.R.MS.

Director, Port Elizabeth Museum, Author

of ‘‘ The Monkeyfolk of South

Africa.”

New Edition. Enlarged and Thoroughly Revised.

With 193 Illustrations. 8vo. $4.50 net. (Postage 23 cents. )

4th Avenue

& 30th Street LONGMANS, GREEN, & CO, NEW

YORK

The American Journal of Science

Established by Benjamin Silliman in 1818

The Leading Scientific Journal in the United States

Devoted to the Physical and Natural Sciences, with special reference to

Physics, and Chemistry on the one hand, and to Geology and Mineralogy

on the other.

Aa tide Editor: EDWARD S. DANA.

tors: Professor GEORGE L. GOODALE, JOHN TROWBRIDGE, W. G. FARLOW and WM. M. DAVIS

of Cambridge; Professors A. E. VERRILL, HENRY S. WILLIAMS and L. V. PIRSSON, of New

Haven; Professor JOSEPH s. AIMES, of Baltimore ; MR. J. S. DILLER, of Washington.

Two volumes annually, in monthly numbers of about 80 pages each.

Shiny x pee ended its first series of 50 volumes as a quarterly in 1845; its second

wake volumes as a two-monthly in 1870; its third series as a monthly ended Dec-

T, 1895. A Fourth Series commenced in 1896.

s on Pete Price, $6 per year or 50 cents a number, postage prepaid in the United States

nada ; $6,40 to Countries in the Postal Union. Back numbers at p

se-Ten-Volume Indexes, Vols. I-X XI-XX, fourth series, price one dollar. Address

The American Journal of Science

New Haven, Conn.

The American Naturalist

A Monthly Journal, established in 1867, e to the Advancement of the Biological Sciences

with Special Reference to the

tors of Organic Evolution and Heredity

CONTENTS OF JULY NUMBER

Genetical peg on Oenothera, IIL Dr. Bradley

E Š

Evidence of A Int the F Genera-

tion from son prise Bos indicus on Bos taurus.

Shorter Articles ae Diseu a On the Inheritance

the Tricolor Coat in Guinea-pigs, and Its Rela-

tion to Galton’s Law of Ancestral Heredity. Pro-

fessor W. E. Castle.

CONTENTS oF AUGUST NUMB

mage pore: in Successive Alfalfa Generations. L. R.

Some Tentiro of gen een in Fresh-water

Fishes. Henry W. Fow

The Formation of pe Pali Correlation Tables

Lae: the Number of Combinations is large. Dr.

ussion: The Production

nisms from Het-

x a

c

Analogies of B jologieal 3 A Dr. C F.Cook.

a Fairness ksd Accuracy in Scientific Review-

ng, Dr. J Arthur Harris.

Notes and Literature: Distribution and Origin of

ein America, T. Barbo

CONTENTS OF SEPTEMBER NUMBER

On Diferential tesa a Respect p Seed

Weight gouna a s Field a of Phaseolus

A Case of Polymorphism in Asplanchna simulating

Mutation, Prcfessor J. H. Powe

Shorter Articles and Discussion ee al Symbio-

tic Relation between a Waterbug aad a Crayfish.

Professor James F. Abbott. Double Eggs. Pro-

fessor Charles W. Hargitt.

Notes and re American Permian Verte-

brates, Dr. O. P. Hay. Federley’s Breeding Ex-

pomen ts with the Mot oth Pygaera, A. H. Stur-

vant

CONTENTS OF OCTOBER NUMBER

I. Gametic riae as a Cause of Correlations.

G. N. Collins

IL Mice; Their Breeding and Rearing for Scientific

Pur Dr, J. Frank Daniel

III. The agag of — arenicolor Cope, with

Notes on vd ec ts and Variation. C. H,

on

IV, Aleurone Colors in F, if a Cross between Non-

colored Varieties of Maize. Professor R. A.

W: eee en of Zea mays L. Dr. Walter

VI. Notes and pene re : Patten on the Origin of

Vertebrates and t the General Value of Specu-

lations on the TAPEET of Organic Beings.

Professor Wm. E. Ritt

CONTENTS OF NOVEMBER NUMBER

e Mendelian Notation as a Description of

h ysiological Facts. Professor E. M. East.

Pang gnc = ane ecarri of the Starvation :

O cendants u Characteristics o;

the Descendants oo Dr. J. Arthur Harris.

oy

:

H

Sara Relations in Xenoparasitism, Dr.

Shorter rensa a Discussion: On Tricolor

ars t in Dogs and uinea-pigs. Arend. L.

.

4

VY. Notes and Literature: The Classification o:

Liverworts: Professor Douglas Hou a

Campbell. Invertebrates, Professor opt ton

wood Kefoid.

CONTENTS OF DECEMBER NUMBER

I. The Mendelian ace of Sete prvi od in the

Domestic Fowl. Dr. Raym

IL Reflections on the Autonomy ‘ot iai Sci-

ence. Professo

II. The Spawning Habitsof the Sea Lamprey, Petro-

myzon marinus. Dr. L. Hussakot.

IV. gge Articles and Daik, A Simple a

the Goodness of Fit of Mendelian Ratios.

Dr. J. ur

V. Notes — Literature: Notes on Ichthyology.

d Starr Jordan.

<

H

mäe a Volume XLVI

Single Number 35 Cents

peun a

Yearly Subscription, $4.00

The NATURALIST will be sent to new subscribers for four months for One Dollar

THE SCIENCE PRESS

Garrison, N. Y.

Sub-Station 84: NEW YORK

Lancaster, Pa.

VOL. XLVII, NO. 554

|

.

A

FEBRUARY, 1913

THE

AMERICAN

NATURALIST

A MONTHLY JOURNAL

Devoted to the Advancement of the Biological Sciences with

Special Reference to the Factors of Evolution

CONTENTS

Page

Spee othe Natural Selection and TE Professor MAY-

RD M. MET

ESE: in ae ‘avin and Non-iving Professor Bunrox EDWARD

LIVINGSTON

Adaptation in Animal Reactions. Professor G. H. PARKER

eee

Adaptation from the Point of View of the ier Ben Professor ALBERT P.

MATHEWS

The Fitness of the Environment, an Inquiry into the Biological wees

of the Properties of Matter. Professor LAWRENCE J. HENDERSON - -

Shorter Articles and Discussion: Mutations in Oenothera een L?: Professor

BRADLEY M. Davıs. A Convenient Microscope LBERT

Case essor À M.

REESE. A Literary Note on the Law of Goriaitai Continuity: Dr. W. W.

STOCKBERG

- 116

THE SCIENCE PRESS

LANOASTER, PA. GARRISON, N. Y.

NEW YORK: SUB-STATION 84

The American

Naturalist

intended for AAST and books, etc., intended for review should be

MSS.

sent to the Editor of THE AMER

A

tion are especially w

ICAN a Garrison- -on-Hudson,

rticles rab research work bearin

elcome, and will be given froter in publica

New Yor

n the problems A Se evolu:

hundrea reprints of a amg are supplied to authors tres of charge.

One

Further reprints will be supplied a

iptions and Spn EAA should be sent to the publishers.

subesriptien. pao is four dollars a

Canadian postage twenty-five cents additional.

The

Foreign postage is fifty cents and

The charge for single copies is

ge.

forty cents. The advertising rates are Four Dollars for a pa

THE SCIENCE PRESS

NEW YORK: Sub-Station 84

Entered as second-class matter, April 2, 1908, at the Post Office at Lancaster, Pa., under the Act ot

Congress of March 8, 1879.

Lancaster, Pa.

Garrison, N. Y.

FOR SALE

ARCTIC, ICELAND and GREENLAND

BIRDS? SKIN

Well Sees ig Prices

articulars of

G. DINESEN. Bird Collect

Husavik, North Iceland, Via Leidiė i Sribiand

TENTH EDITION.

THE MICROSCOPE,

an introduction to Microscopic Methods and to Histology,

By SIMON Hener GAGE, of Cornell University. Over

350 la A a tf gen res in this new

and fully revi ition. Price $2.00, postpaid.

COMSTOCK PUBLISHING cO., Ithaca, N. Y.

To Ornithologists

and Museums

W. F. H. ROSENBERG

Importer of Exotic Zeological Specimens

57, HAVERSTOCK HILL

London N. W., ENGLAND

Begs to announce the publication of a new Price List

(No. 17) of BIRD SKINS. This catalogue con-

tains nearly 6,000 species, and is the largest and

most complete price list of birds ever published. It

is arranged in systematic order with authors’ names,

indications of localities, and an index to Families.

It will be sent gratis and post free on application, as

will the following lists:

No. 14, Mammals; No. 15, Bird’s Eggs; No. 16

Reptiles, Amphibia and Fishes ; No. 12, Lepidoptera ;

No. 13, Coleoptera,

Largest stock in the world of specimens in all

branches of Zoology. Specimens sent on approval,

Please state which lists are required.

University of Colorado

Mountain Laboratory

Courses in Field Biology (animal and

plant), Ecology, Systematic Botany, Bi-

ology of Ponds and Streams. Suitable

work for undergraduates and graduates.

Daily Field excursions.

The laboratory is at Tolland, Colo., in

the Rocky Mountains at an altitude of

8,889 feet. Climate cool and healthful.

Fees and expenses moderate.

For pamphlet address the professor in

charge.

Dr. Francis Ramaley

University of Colorado

BOULDER COLORADO —

THE

AMERICAN NATURALIST

VoL. XLVII February, 1913 No. 554

ADAPTATION THROUGH NATURAL SELECTION

AND ORTHOGENESIS!

PROFESSOR MAYNARD M. METCALF

OBERLIN COLLEGE

Mr. President, Members of the Society, and Friends:

When the president of our society, Professor Conklin,

asked me to open this discussion, he suggested that I

speak in advocacy of natural selection as the dominant

factor in adaptation, saying that several speakers would

follow who would attack this conception. As I under-

stand it, a paper in advocacy of natural selection is

desired not so much that it may be a target at which fol-

lowing speakers may direct their shafts, as that this

conception may not stand without defenders.

It involves some temerity to speak to an audience of

American naturalists upon natural selection, for the con-

ception is more than a decade old. It dates even from

the dark ages of the middle of the last century, and we

naturalists, like the Athenians to whom Paul spoke, are

continually seeking some new thing. I can not even, as

DeVries and Bateson in the case of Mendel’s principles,

claim the honor of a re-discoverer, for through all the

years since its first promulgation the conception of nat-

ural selection has been constantly to the fore and has been

discussed and re-discussed until many, I fear, have

wearied of it, and for this very reason may be ready to

give too undiscriminating welcome to other conceptions

that seek to supplant the old idea.

* Read at the Symposium on Adaptation at the meeting of the American

Society of Naturalists, Cleveland, January 2, 1913.

65

66 THE AMERICAN NATURALIST [Vou. XLVII

There are two sets of factors in producing and deter-

mining the direction of evolution, those within the organ-

ism, and those external to it. In earlier studies of evo-

lution the external factors received by far the most atten-

tion, and only recently have the internal factors begun to

receive the thought they deserve. Of the external factors,

natural selection is the one for which the largest claims

have been made and the one whose claims have been most

cordially allowed. I think, however, there are few, if any,

of us who do not feel that some of the advocates of nat-

ural selection have gone too far. Wallace, for example,

would not allow that there are any characters of animals

or plants which are not useful to their possessors and have

not been perpetuated and emphasized because of this use-

fulness. Admitting, as we must, that we know but little

of the intimate life of organisms, and that the use of

many really most useful characters. may fail to impress

us only because of our ignorance of real conditions, still

I think most of us feel not only that the claim that all

structures and qualities of organisms are useful is an

exaggerated claim, but that very many characteristics

are either of indifferent quality, or are so slightly useful

as not to be of selection value, or even are slightly dis-

advantageous. The more closely one studies any organ-

ism the more will he become impressed with the number

of these non-useful or doubtfully useful qualities, and

close and careful students are likely to find their vision

of natural selection grow dim, like the pilgrim who, in the

midst of the woods, could not see the forest for the trees.

But as we view organic nature in its wider aspects is

there any other feature so prominent as the adaptation

of organisms to their environment and to the lives they

must live in the midst of this environment? However

many details of structure or behavior may fail to show

their utility, still it remains true that there is no phenom-

enon of organic nature more impressive than adaptation.

In our study of evolution we have, then, this thing to

explain—namely, the universal prevalence of a high

degree of adaptation of organisms in habit, function and

structure, to their environment, and yet the presence of

No. 554] ADAPTATION THROUGH SELECTION 67

many qualities, for the most part minor qualities, which

so far as we can judge, are non-adaptive. I would sug-

gest for your consideration the propositions that the

broader adaptive features are due to natural selection,

and that the non-adaptive minor characters are the result

partly of factors within the organism, and partly of ex-

ternal factors other than natural selection. This paper

will not refer further to these other external factors, but

will discuss one of the factors within the organism. +s

Evidence accumulated during the last decade seems to

indicate that fluctuating variations are not significant,

and that only mutations are stable and can serve as a

foundation for evolution. Perhaps we have accepted this

statement a little too hastily. The actual evidence in its

favor is not all that could be desired, but it seems to

be the general opinion to-day. If this opinion is well

founded, it is through the study of the origin and nature

of mutation that we may hope to gain most light upon

the problem of the origin of adaptation.

When one comes to think of it, we really know very

little of mutation. Few instances have been carefully

observed and recorded. But in the midst of this scant

knowledge of mutation one fact stands out clearly, that

mutation apparently is not indeterminate, occurring in

all directions equally, and changing from generation to

generation, as would be the case were it purely fortuitous.

Our best studied instances of mutation are still seen in

(Enothera lamarckiana, and in all the observations upon

the mutation of this species nothing is more patent than

that there are certain definite tendencies for particular

mutants of very complex and very clearly recognizable

types to appear generation after generation, in numbers

that are very much greater than could be accounted for

by any chance aggregation of unit characters to make

these few well-marked types.

In 1905 I wrote as follows: :

In the Amsterdam Garden the mutant albida appeared in four differ-

ent generations from lamarckiana parents, previous to 1902, 15 albida

appearing in one generation, 25 in another, 11 in another and 5 in

another. Nanella appeared 5 times in one generation, and in other

68 THE AMERICAN NATURALIST (Vou. XLVII

generations, respectively, 3, 60, 49, 9, 11 and 21 times. Lata, oblonga,

rubrinervis and scintillans appeared frequently.

In the fourth generation, along with 14,000 lamarckiana plants, there

appeared 41 gigas, 15 albida, 176 oblonga, 8 rubrinervis, 60 nanella, 63

lata and 1 scintillans, all bred from lamarckiana seed. In the fift

generation, similarly bred from pure lamarckiana seed, among 8,000

lamarckiana plants were found 25 albida, 135 oblonga, 20 rubrinervis,

49 nanella, 142 lata and 6 scintillans. In the fourth generation one

plant in 80 was oblonga. In the fifth generation one plant in 60 was

oblonga. DeVries himself says: “ A species therefore, is not born only

a single time, but repeatedly, in a large number of individuals and dur-

ing a series of consecutive years.

Oblonga differs from the parent species lamarckiana not in a single

feature, but in an elaborate complex of characters. The other mutants

likewise are distinguished from lamarckiana z) a complex of characters

rather than by a single feature.

The mutation can hardly be entirely fortuitous if, for several genera-

tions, out of every thousand offspring of pure lamarckiana parents,

there appear more than ten plants marked by the particular complex

group of characters which designate oblonga. Were oblonga demar-

«cated from lamarckiana by but a single character, it would be remark-

‘able to find it appearing repeatedly and in such numbers. When we

remember that it is defined by an extensive series of characters differ-

entiating it from lamarckiana and from all other mutants observed, are

we not led to the conclusion that mutation in Ginothera lamarckiana

is not wholly fortuitous, but is to a degree predetermined; and that there

is some tendency to the production of the oblonga and other types in

numbers much greater than would be secured by purely fortuitous and

indeterminate mutation?

Mutation in our most carefully observed instance,

therefore, is clearly determinate. There is in Ginothera

lamarckiana a tendency to mutate in certain definite

directions generation after generation. This trend to

mutation in certain particular directions is an example

of a condition within the organism which might decidedly

affect the course of the future evolution of this Gno-

thera and its descendants.

But we can safely go further. Not only have we evi-

dence that there exist tendencies to produce certain mu-

tants repeatedly, paleontological records show, I believe,

that there have existed trends toward an increasing

emphasis upon certain characters and that these trends,

in some instances at least, lay along lines that produced

no more perfect adaptation of their possessors to their

environment.

No. 554] ADAPTATION THROUGH SELECTION 69

Neumayr’s well-remembered figures of fossil Palud-

inas from Slavonian lake deposits show an increasing

rugosity of shell and irregularity of aperture, but the

separate steps of this change could hardly have been

of selection value and have occurred under the con-

trol of natural selection. One is irresistibly drawn to

believe that there was in these Paludinas an internal

tendency to mutate from generation to generation toward

increasing rugosity of the shell and increasing irregu-

larity of its aperture.

Similar trends in the development of limb and tooth

structure of the horse were long ago emphasized by

American paleontologists. There is no conceivable utility

in the modified shell form.of the Paludina, and, in the

case of the horse, the wonder is that the race has not long

ago been exterminated. Among the Cephalopods are

some forms which show most complex patterns of the

partitions. There are complete intergrading stages

shown in the known species between simple suture lines

and the most fantastically fimbriated. There seems here,

again, to be evidence of a trend toward increasing convo-

lution of the sutures, rather than evidence of the appear-

ance of a useful character and its increase by successive

steps each of selection value. In a somewhat similar way,

over-ornamentation and bizarre character developed in

the Trilobites by steps which can hardly be imagined

useful. Many such examples might be cited.?

It may seem that I am arguing against the importance

of natural selection, but I think I am not, and for this

reason: the very trends, of whose existence I think we

have good evidence, are themselves subject to natural

selection, and if they are in hurtful directions, they will

ultimately cause the extermination of the species exhibit-

ing them. I very much doubt if such a monstrosity as

the horse could long persist, except for man’s aid, for it

* On the afternoon of the day when this paper was read Professor B. M.

Davis reported that among the offspring of his @nothera hybrids were some

mutants with flowers larger than either parent species, and that for two

generations (as far as his experiments went) the flowers increased in size.

is is soepi an example of just such a trend in mutation as I have

indicated above

70 THE AMERICAN NATURALIST (Vou. XLVII

is about the least adaptable beast we know. The Cepha-

lopods with complex partitions have all perished. The

bizarre Trilobites similarly persisted each for but a brief

period.

I would not claim that the over-ornamentation in these

cases was the cause of extermination. It seems more

probable that it is merely one noticeable indication of a

general unbalanced condition of the organism, a condition

affecting probably function as well as structure. I sus-

pect that more species have perished because of physio-

logical maladjustment than from any disadvantageous

structural qualities.

In quatenary times in numerous species, great empha-

sis seems to have been laid upon bulk and the very huge-

ness of some of these species probably aided in their ex-

termination.

Non-adaptive qualities, when first appearing, may

often be comparatively harmless, at least may not be

sufficiently hurtful to lead to the extermination of the

species in which they appear. But qualities compara-

tively innocent in their beginnings, when over-empha-

sized by such trends as we are discussing, may go beyond

the limit that even long suffering nature will allow and

extermination follow. The goblin of natural selection

will get him in the end if he doesn’t watch out. It is a

case of giving the species rope enough and letting it hang

itself. Instead, therefore, of supplanting natural selec-

tion, such orthogenesis as this really acts in the end to

aid in eliminating many species whose characters in their

beginnings were indifferent, natural selection finally dom-

inating and compelling adaptation.

Since mutation-trends in helpful directions will be

aided by selection, through the destruction of the rivals

of their possessors, while hurtful trends will cause ex-

termination, we see that natural selection has a determi-

native influence upon the direction of evolution, steering

the species along safe paths of progress.

Further advance in the study of the method of evolu-

tion may be expected from the study of trends in muta-

tion. Such study should be continued for many years,

No. 554] ADAPTATION THROUGH SELECTION 71

breeding in great numbers for many generations from

some highly mutating stock. Such work can not well be

undertaken by a single individual, for it should continue

through several human generations. It should be under-

taken by some institution which will ensure the endur-

ance of the investigation beyond the present generation.

Possibly the best object for such study would be the Syra-

cuse Trilliums—Trillium grandiflorum—whose remark-

able variation Britcher* has so well described. Variation

in these Syracuse Trilliums is more universal and more

extensive than in any other plants or animals I know. It

is so great that it can hardly be of the fluctuating type.

It is, in all probability, true mutation, and if so it offers

the best opportunity I know for the study of the phenom-

ena of mutation.

SUMMARY

Organisms show adaptation in their more important

characters, while many of their minor characters fail to

show their utility.

There are definite tendencies to mutation in particular

directions, and there is abundant paleontological evidence

of trends toward increasing modification in particular

directions.

Qualities so appearing may be indifferent in their be-

ginnings, but may through this orthogenesis become suffi-

ciently useful or hurtful to affect selection. Such trends,

when they affect physiological qualities, are likely to

bring about an unbalanced, distorted physiological con-

dition and be peculiarly hurtful. Probably this has been

one of the chief causes of the disappearance of species.

Orthogenesis as thus interpreted is but the handmaiden

of natural selection which, acting upon all qualities thus

developed to major proportions, will cause them to disap-

pear if ill-adapted. At the same time advantageous

trends will be encouraged in the struggle for existence

and the direction of evolution be turned toward further

adaptation.

Adaptation is the most salient result of evolution and

natural selection its great cause.

* H. W. Britcher.

ADAPTATION IN THE LIVING AND NON-LIVIN Gt

PROFESSOR BURTON EDWARD LIVINGSTON

Tur Jouns HOPKINS UNIVERSITY

Tue fundamental differences in concept and mode of

thought, which may be remarked between the sciences of

the living and those of the non-living, are perhaps no-

where better exemplified than in the interpretations and

in the degree of prominence which they respectively give

to the idea of adaptation. A general survey of the nat-

ural sciences results in the somewhat startling discovery

that biology is the only one of these which deals conspic-

uously with this idea. I have, therefore, been led to take,

for my present paper, the somewhat bizarre title wh ch

has been announced, and I shall here attempt partially

to set forth some characteristics and implications of the

biological concept of adaptation, and, in certain respects,

to compare these characteristics and implications with

those of similar concepts which have the place of adapta-

tion in the sciences of the non-living. The term adapta-

tion is used in a passive and in an active sense. I shall

consider the two sorts of adaptations in order.

Passive Adaptations.—Adaptations are ch teristics,

properties or qualities attributable to natural objects.

They imply, however, not only mere qualities, but also

the presence or absence, in the object considered, of po-

tentialities or capabilities to be or to do certain things

under certain conditions. The term always lays stress

on potentialities but it does not imply at all that these

are, or have been, realized. If they were actually real-

ized, it would amount to a redundancy to note the exist-

ence of the adaptations at all; an adaptation ‘‘caught in

the act,” an already realized potentiality, is so self-evi-

dent that we do not need to mention it as such. In such a

* Read at the Symposium on Adaptation at the meeting of the American

Society of Naturalists, Cleveland, January 2, 1913.

72

No. 554] ADAPTATION IN LIVING AND NON-LIVING 73

case, the statement of the realization implies the poten-

tiality, for an object is obviously adapted to doing and

being what it does and is.

However, doing and being are only relative, for any

object may change its state more or less effectively and

may possess attributes in different degrees, and our in-

terest in realized potentialities lies not in the fact, but in

the degree of adaptation. Furthermore, the degree of

adaptation depends clearly upon the extent to which the

object considered possesses those properties or qualities

about which our idea of adaptation centers, and so atten-

tion is at once turned to the properties or qualities as

such.

I may draw an illustration of actually observed adap-

tations from the science of geology, which perhaps in-

terests itself as much in the survival and distribution of

rock masses as does biology in the survival and distribu-

tion of living things. My example has to do with the nat-

ural selection and distribution of certain boulders and

pebbles in a deep Californian valley.

At the time of the filling of the Salton sink by the un-

ruly Colorado River, the only loose stones of the inun-

dated area that were able to keep themselves in contact

with the air were fragments of pumice. These’ were

adapted to float upon water, and they largely refused to

be submerged. With the rising waters they also rose, and

thus were able to take advantage of air movements to re-

distribute themselves. Had it not been for the floating

adaptation, these pumice pebbles would have suffered

temporary extinction in the form of submergence, and

they would not have been able to survive and to gain

dominance in the pebble population of certain Salton

beaches which they forthwith proceeded to invade. We

have reason to believe that this sort of spontaneous mi-

gration of pumice pebbles has taken place many times

before in the Salton valley, at periods and seasons when

edaphic and climatie conditions happened to favor such

readjustments of the tension lines, and that the present

`

74 THE AMERICAN NATURALIST [Vou. XLVII

distribution of these curiously endowed stones has been

largely brought about through their possession of the

floating adaptation.

= Probably a geologist would not mention adaptation in

this connection; he might succinctly state the apparent

specific gravity of pumice, and might then proceed to pre-

sent the case in terms of this internal character and in

terms of more or less quantitatively known features of

the surroundings; for the idea of adaptation here takes

account of the low specific density of the rock and the

ability to float upon water implied thereby, and to state

that pumice is adapted to float is redundant, after we

know its specific density. For aught I know also, there

may be different species of pumice, some of which might

be observed to float higher or a longer time upon water,

and in such ease, what we might term the varying de-

grees of adaptation in the different species should be

quantitatively brought out—and then dismissed—by an

adequate study of the internal qualities of the rocks.

But not nearly all of the potentialities discoverable in

natural objects are of this realized, and consequently di-

rectly observable sort. The future is no doubt pregnant

with hitherto untested adaptations and our imagination

frequently suggests these as problems. Of course mod-

ern natural science responds to the proposal of this sort

- of adaptation, try it and see if it is true, and many of us

are busy just now in doing this very thing.

If we find observational proof of the suggested prop-

erty, interest in its adaptational aspect fades, for the

case then passes over into my first category, of actually

observed adaptations. Also, the experimental test of a

proposed potentiality—as to whether it is attributable or

not to the object considered—is but a case of observation,

properly prepared for. Not readily finding the necessary

conditions and the object together in nature, we find

them separately (at different places or times) and pro-

ceed to bring them together. Thus my first experience of

the Salton valley was had at a time when it contained

No. 554] ADAPTATION IN LIVING AND NON-LIVING 75

little or no visible water. Let us suppose that interest

was at that time aroused in the distribution of pumice

pebbles upon certain areas of the dry valley floor, and let

us suppose that a previous migration of these, similar to

the one just described, was suggested as a possibility.

At that time the direct test by observation was not pos-

sible, but the whole question—as far as the floating adap-

tation is concerned—could have been settled readily

enough either by bringing water to the pebbles or by

taking some of the pebbles to water. But of course we

should have been dealing, in this instance again, with the

determination of the presence or absence of a certain

property in the pebbles, as related to a certain property

of water.

In the vast majority of the cases of this sort that at-

tract our attention at the present time, however, natural

science is unable to obtain direct observational tests, even .

of the experimental sort, and some indirect method of

comparison must be resorted to. Now, indirect methods

for determining the degree of a proposed adaptational

property consists in nothing more than the determina-

tion, by whatever means may be convenient, of the de-

gree to which this property exists in the object consid-

ered. Thus, without ever bringing water and pumice to-

gether, it is perfectly possible to establish the ability of

the latter to float, as by determining the comparative

weights of equal volumes of the two substances.

From what has preceded it is suggested that every

passive adaptation that we may consider resolves itself,

upon adequate analysis, into a problem of the measure-

ment and quantitative comparison of qualities or proper-

ties of objects. If neither the direct experimental test

nor the requisite measurements can be carried out, then

the suggestion of an adaptation is no different from the

Statement of any other problem for which no method of

attack has yet been devised. But it must be recognized

im this connection that the usual biological adaptation is

not always appreciated as a problem in comparative

76 ' THE AMERICAN NATURALIST [ Vou. XLVII

measurement, and that its proposal, especially if made

by one in an authoritative position, is far too apt to be

received as a declaration rather than as a question.

Thus, for example, our elementary texts may tell the

innocent beginner that brightly colored flowers are better

adapted to fertilization by insects than are their less

gaudy neighbors, and without critical analysis, a very

complex and exceedingly difficult problem is at once re-

garded as solved. As a matter of fact, this problem in-

volves comparative measurements for which methods

have not yet been devised, so that the cautious biologist

must regard the question of this proposed adaptation as

utterly beyond us for the present.

Apparently possible potentialities which have not been

actually observed in nature, or which have not a basis in

quantitative comparisons so as to be possible of definite

. establishment or refutation, have not played an impor-

tant rôle in the modern development of the sciences of

the non-living, and consequently the adaptational aspect

of the qualities of natural objects is seldom mentioned

in these sciences. The relative ease with which the quali-

ties of the non-living may now be analyzed into funda-

mental concepts renders the use of any other terms than

those of matter and energy quite out of place in their

serious discussions. On the other hand, biological in-

quiry has still much to do with theoretical attributes

which can not be put to any satisfactory test, and this

condition may be in part responsible for the prevalence

of the adaptational point of view in our science. It seems

to be partly because biological problems are too complex

for ready analysis at present, that the adaptational prop-

erties of living things are so often stated in terms other

than those of the fundamental concepts of matter and

energy.

_ In this connection it is, however, to be remembered that

ease of analysis depends as much upon the state of the

analyzing mind as upon the complexity of the analyzed

object. A mind is conceivable, I think, that would con-

No. 554] ADAPTATION IN LIVING AND NON-LIVING 77

sider the phenomenon of cell division as just as capable

of analysis as is that of a chemical reaction like flame;

and we are sure that there have been in the past (and are

indeed at present) minds to which flame would appear

quite as hopeless of analysis as does cell division to us.

I have said that the qualities of living things are too

complex for analysis at present, I might as well have

said that we are at present too ignorant and too feeble to

analyze such qualities. Our science is young yet, not in

years, perhaps, nor yet in absolute achievement, but in

the relation of its present phase of development to the

task which is set before it. It appears to be this youth-

ful quality in biology which may partially explain, as I

have said, the somewhat starting generalization with

which we began.

Active Adaptations.—Biological writing employs the

term adaptation in an active sense as well as in the pas-

sive one heretofore considered in this paper, and it re-

mains to give some attention to this usage, and to an ap-

parent confusion of cause and effect which is connected

therewith. To obtain a clearly legitimate case of active

adaptation we shall have to turn to human affairs, for

reasons which will soon be evident, and the familiar

adaptation of the watch will serve our need as well as

any other. The little machine is complex, too complex

for most of us to understand, and it seems to be much

better adapted than any living thing to long-continued,

uniform motion of a certain specified kind. If I make in-

quiry regarding the causes, or antecedent conditions, to

which this adaptation is due, I find that the various parts

have been made and assembled with reference to the very

adaptation which I am investigating. In my search after

causal relations I have been entrapped in a mesh of un-

investigated psychological phenomena and have discov-

ered the puzzling truth that the watch is what it is,

simply because it was to be what it is! In other words,

the cause of the effect which we are considering is re-

garded as some sort of disembodied spirit of the effect

78 THE AMERICAN NATURALIST (Von. XLVII

itself, and this effect, in order to be the cause of itself

must have existed before it came into being.

Of course we realize that we have thus come into con-

tact with the darkest problem with which biological sci-

ence has to deal, namely, the problem of human purpose-

ful action and of the human will. While I see no reason

for doubting that this problem may eventually yield to

analysis and comparative measurements, yet it must be

admitted that progress in this direction has only just

begun, so that anything but the most superficial consid-

erations in this connection is at present but waste of time

and trouble. We have here, for the present, to acknowl-

edge our fundamental ignorance, and to hold our minds

in that state of suspended judgment to which, in less

complex affairs, students of all the natural sciences have

become used.

Although we are as yet unable to analyze into simple

terms of matter and energy the antecedents which con-

ditioned the adaptation that is before us in the watch,

yet it seems that our analysis of the universe about us

has progressed far enough so that we may be justified in

frankly maintaining that the problem of purposeful cau-

sation has no place in any of our considerations, except-

ing solely those wherein human consciousness has been

involved among the causal conditions. To employ other

terms, I think we are bound to regard the nature of the

future outcome of all processes as totally non-existent,

and consequently absolutely without influence in the pres-

ent, excepting alone (and temporarily) those processes

for which the human will is accounted a necessary ante-

cedent.

We may now inquire as to the causes which have been

in operation to bring about the peculiarly low specific

gravity of the pumice pebbles in my case of the floating

adaptation of these bodies. We are assuredly unable to

state these causal conditions in anything approaching

completeness, but we are nevertheless sure that human

purposefulness has played absolutely no part in the

No. 554] ADAPTATION IN LIVING AND NON-LIVING 79

matter, so that we put the floating adaptation of these

pebbles entirely out of mind as soon as we begin our

search for the causes thereof.

Geology, it seems, does not find purposeful action nec-

essary in its explanations. Neither does any progress

come from a consideration of purposeful changes in the

readjustments of atoms with which chemistry deals, nor

in the phenomena of heat migration which constitute a por-

tion of the field of physics. Modern astronomy sees no-

purposeful activity in the motions of sidereal space, nor

does meteorology seek in the effects of a storm any sugges-

tion of the causes which brought it into being. Many

branches of biological science, however, although confess-

edly not dealing with human purposeful activity, seem

_ frequently to seek in the future the causes of the present.

Thus our science teems with purposeful reactions, and

this feature of the idea of adaptation adds its influence to-

the ones already mentioned, playing an important rôle in

keeping the sciences of the non-living essentially and fun-

damentally separate from those of the living.

The history of the idea of causation in the natural sci-

ences suggests much that may have a bearing on our

judgment as to whether this present distinction between

the two groups of sciences is necessary and permanent or

merely temporary and passing. To primitive man all

problems were too complex for adequate analysis, and

purposeful activities of many kinds were devised to ex-

plain not only the doings of his fellow men but also the

doings of all nature. The whole world was then a world

of teleological causation. The heavenly bodies moved

_and the chemical elements combined or separated accord-

ing to the capricious wills of innumerable deities and

demons. Men then heard in the howling of the storm

and in the rumble of the earthquake the terrible voices of

the spirits of the air and earth. All living things were en-

dowed with a man-like consciousness and power of will-

ing to do, and everything struggled with everything else:

in a never-ending conflict of capricious wills.

80 THE AMERICAN NATURALIST [Vou. XLVII

In the development of our race, however, an increasing

experience of deterministic causal relations has been ac-

companied by a progressive effort to expunge the idea of

purposefulness from our thinking. The numerous myth-

ological personages just called to mind have gradually

suffered a curtailing of their powers for good and evil, ’

and have, in general, by natural science at least, been

totally disearded. Many relics of the past of course re-

main in all our mental life; many of our words and not a

few of our instinctive modes of thought are survivals

from the teleological period of our development. Jupiter

and Venus still play their part in modern astronomy, and

Vulean’s name is still heard among geologists. But ob-

vious teleological expressions have been generally out-

grown and discarded by all of the sciences that deal with

the non-living. In biology alone they persist, mainly as

personifications of plants and animals, making our mod-

- ern writings a curious jumble of exactly stated observa-

tions and conclusions, together with many statements

that might have been taken bodily from primitive fairy-

tales. Foreseeing of the future and conscious purpose

are apparently attributed to living things in which we

have no evidence for the existence of consciousness. The

eye develops in the animal in order that it may see, the

leaves of the plant are for the purpose of obtaining car-

bon dioxid from the atmosphere. The list of such state-

ments might be made very long, but you are quite fa-

miliar with their nature.

Not only are the organisms with which we deal fre-

quently personified to the extent of attributing to them

the foresight and will needed to carry out complicated

plans, but they are also frequently supposed to be ca-

pable of making a mistaken judgment. I find in Gibson’s

translation of Jost’s ‘‘ Plant Physiology,” 1907 (page

389), an excellent example of this assumption, where it

appears that one lower organism may be clever enough

to outwit another. The statement in question reads,

‘‘The gall, for example, is of service only to the insect,

No. 554] -> ADAPTATION IN LIVING AND NON-LIVING 81

but is highly disadvantageous to the plant; we must as-

sume, indeed, by way of explanation, that the insect suc-

ceeded in deluding the plant, so that instead of treating

the insect as an enemy and an intruder it behaved towsres

it as if it were a bit of itself.”

I think it is perfectly clear that the non-biological sci-

ences have all passed through a much earlier stage in

which purposeful adaptations were seriously considered,

-and it seems quite as clear that such concepts are not any

longer accepted in the serious studies of these sciences.

There seems also to be no doubt that the biological sci-

ences, notably in their physiological aspects, are tending

at the present time more and more to adopt a non-teleo-

logical point of view. From these points I again draw

the conclusion that ours is a developmentally young sci-

ence, that it still retains features of its early youth, and

that the concept of purposeful adaptation is one of these

features, sooner or later to be totally abandoned, even as

the same concept has already been abandoned by the

other natural sciences.

If my conclusion should be wrong, then one of two

propositions must follow: either the sciences of the non-

living have fallen into error, ought to have retained the

concept of purpose in natural phenomena, and will sooner

or later return to this concept; or else there is a great

and fundamental difference between the living and the

non-living, and teleology has a logical place in considera-

tions of the former objects but not in those of the latter.

Although it is to be realized that the possibility of one or

the other of these propositions can not be rigidly denied

at the present time, yet the probability of either one is

definitely decreased by every analytical conquest of sci-

ence. The controversy here suggested—which seems in

our time as wastefully to absorb our energies as did the

discussion of special creation in the time of Charles Dar-

win—is characterized by this peculiar feature, that, while

all evidence presented for teleological causes is conspicu-

ously based upon our ignorance and present inability to

82 THE AMERICAN NATURALIST [Vou. XLVII

analyze our problems, the evidence offered in the opposite

direction is just as conspicuously positive and consists

of cases which have already been subjected to relatively

complete analysis. As Cowles! has pointed out, there has

never been any retrogression in these matters; all phe-

nomena now explained non-teléolgically were once ex-

plained teleologically, but no non-teleological explanation

once attained, has ever been replaced by one involving

purpose. Under these circumstances, a pragmatic judg-

ment must be rendered, at least tentatively, in favor of

the position here taken, that teleological thinking should

have, and will at length have, no place in our science at all.

Conclusion.—I think it is to be concluded from the con-

siderations here set forth that there is nothing known of

the nature of living things which should lead the bio-

logical sciences to base their inquiries on any other

methods or modes of thought than those employed in the

sciences of the non-living. In both its aspects, passive

and active, the dominance of the concept of adaptation,

which now distinguishes our science from the non-bio-

logical ones, is related to the comparatively youthful

stage of development so far attained by biology, and not

to any observed character of the living objects with which

we deal. It seems obvious that biology is advancing

slowly but steadily along the path already traversed by

the other natural sciences, and I think our present opera-

tions may best be guided by the hypothesis that all these

sciences will eventually come to deal with the same funda-

mental concepts and modes of thought. Should this con-

dition of affairs ever come to actual attainment, then the

discussions which now center about the idea of adapta-

tion might be expected to give place to other discussions

of causal relations between measured qualities and prop-

erties of the objects dealt with, such as are already begin-

ning to be common in many lines of biological study.

* Cowles, H. C., in Coulter, Barnes and Cowles, ‘‘Text-book of Botany,’’

2: 948. 1911.

cd

ADAPTATION IN ANIMAL REACTIONS!

PROFESSOR G. H. PARKER

Harvarp UNIVERSITY

In recent times no feature of animal or plant life has

been accorded greater emphasis than adaptation, and

this property has been repeatedly declared to be one of

the fundamental characteristics, if not the most funda-

mental, of living bodies. Nevertheless, the field of or- —

ganic adaptation has been scarcely more than superfi-

cially surveyed. With the naive instinct of the collector,

the biologist has been content to roam over this vast

territory and gather here and there what seemed to him

to be striking examples of adaptation till our texts are

rich storehouses of instances of nature’s apparent in-

genuity. How well founded even the more striking of

these examples are, no one really knows, though the ef-

fect of the whole collection on the naturalist is to en-

gender in him a feeling akin to awe for the adaptive

capacity of living beings. May it not be, however, that

we overestimate this aspect of organic nature and em-

phasize beyond reason its real value as a factor in the

organic world? It is my opinion, at least, that many ani-

mal reactions which we have been accustomed to call

adaptations should not be thus designated, and that the

difficulties that we often meet in attempting to account

for such reactions are due to our consideration of them

from the standpoint of adaptations when in reality they

are far from being such.

Adaptations are indissolubly connected with the ac-

tivities of animals and plants. We are only just begin-

ning to learn that an organism in its essentials is an ac-

tive, working system, that the moment we think of it as

*Read at the Symposium on Adaptation at the meeting of the American

Society of. Naturalists, Cleveland, January 2, 1913.

83

+

84 THE AMERICAN NATURALIST [ Vou. XLVII

a machine standing still, we divest it of precisely that

element which is most distinctive of it. The anatomical

conception of an organism as a mechanism at rest, useful

and important as it has been and is still, is thus funda-

mentally defective. The essential feature of every liv-

ing thing is incessant activity, and adaptations are part

and parcel of this activity. Even such apparently pas-

sive examples as the adaptation of a moth to the bark of

the tree on which it rests depends for success as much

on the position of the moth’s body, the pose of its wings,

etc., all features of muscular action, as upon the color

` pattern of the exposed parts. Thus adaptation, like

other truly organic phenomena, exhibits a fundamental

dependence on organic activity, a condition that favors

the conception of the organism as put forward long ago

by Lamarck.

But adaptation is not only essentially associated with

the activities of organisms; it is also conditioned by the

continuity of this activity. Every organic line of descent

has had its past history and can look forward to a pos-

sible future. The continuity thus indicated is an es-

sential part of organic nature. The inorganic may at

any moment be completely resolved into its elements and

recombined at the next moment into a new order without

violating the law of its existence. But an organism can

not undergo such a revolution without annihilation; an

animal or plant subjected to such a process would cease

to exist. Hence life is not only activity; it is activity so

directed as to be continuous, to be self-perpetuating.

Continuity of action is, therefore, an inherent part of

the make-up of living things, and adaptation is condi-

tioned by this continuity in that those reactions are

adaptive which make for a continuance of life.

Strictly speaking, adaptations exist no more in nature

than does a species; it is a word in the dictionary, a fig-

ment of the human brain. Just as the systematist finds

the individual animal or plant the real object of his in-

vestigation, so the student of adaptation finds individ-

w

No. 554] ADAPTATION IN ANIMAL REACTIONS 85

ual animal movements the material for his study.

Broadly speaking, these movements range from the hid-

den internal processes in the animal economy to the more

obvious external forms of behavior.

The very fact that adaptations have been classed

under one head to the exclusion of other forms of ani-

mal reactions has given them a certain undue promi-

nence, but this is not the only reason for their usurping

more than a fair share of attention. Many animal reac-

tions that are in no proper sense adaptations have been

brought under this head, and in certain quarters this

process of appropriation has gone to such an extent that

every animal reaction has been supposed to have some

adaptive significance. How far this is from the truth

can be made clear by a homely example. When a person

faints, his best position for recovery is a horizontal one,

and into approximately this position he is very likely to |

fall. Furthermore, he falls with limp muscles, a method

which under the circumstances is much safer than that

of falling with a tense body such as usually occurs in con-

sciousness. Thus the position into which a fainting per-

son falls and the method of arriving at it, have all the

appearances of adaptations. Yet, in my opinion, any

one who would interpret these movements as adapta-

tions would lay himself open to a charge of unreality.

. The new position, favorable as it is for recovery, is in

fact the mere consequence of the faint, and as such it

completely loses any claim as an adaptation. That it is

advantageous is purély incidental; it might equally well

have been disadvantageous. Thus responses, even when

of a favorable nature, are not necessarily adaptations,

though they may resemble them to a striking degree.

An adaptation is not only a favorable response, it is a

favorable response especially developed to meet a par-

ticular emergency.

If it is so easy to point out pseudo-adaptations among

our own activities, it is highly probable that they also

occur among the responses of the lower animals, And

86 THE AMERICAN NATURALIST (Vou. XLVII

such seems to be the case. That an isolated blastomere

representing a half or a quarter of the egg of Amphioxus

should be able under favorable conditions to develop

into a complete larva is at first sight a surprising fact

and seems to give evidence of a remarkable power of

adaptation. But such an interpretation is far from justi-

fiable. The growth of the isolated blastomere seems to

me much more like the falling of a fainting person than

like a process devised to meet a special emergency, and,

important and illuminating as this growth is from the

standpoint of our understanding of the mechanism of

the egg, it is, I believe, a good case of pseudo-adaptation

rather than of true adaptiveness. All eggs certainly do

not show this trait and to single out the egg of Amphi-

oxus and extol this reaction as an adaptation is to give

to it weight beyond its deserts. To call this an adapta-

tion is to read adaptation into it as only an overzealous

advocate could do. Dame Nature under the circum-

‘stances might well be likened to a certain English poet,

who, on visiting incognito an exposition of his own

verses, was amazed at the wonders they were said to

contain.

The majority of animal reactions are, in all probabil-

ity, neither conspicuously advantageous nor disadvanta-

geous to the life of the individual. They are dependent

chiefly on the material composition of the given organism,

and, so long as they are relatively indifferent to the con-

tinuance of life, they pass without special consequence.

Relatively speaking only now and then do we have condi-

tions where a vitally important form of response, an adap-

tive response, appears. On the whole the flow of action in

the daily life of many organisms requires little of such

special activity and proceeds at the level of mild indiffer-

ence. In other words, adaptive reactions as the controll-

ing factors in animal life are, I believe, by no means so

universal as some of their advocates would have us think.

The world at large affords an environment in which

each animal has a wide range for possible reactions and

No.554] ADAPTATION IN ANIMAL REACTIONS 87

of a number of responses that might be made to a given

set of conditions, one may be quite as appropriate for

the continuance of life as another. In other words ver-

satility seems to be a more truthful description of the

actual conditions in animal life than the rather rigid

state implied in the application of the idea of adaptive

responses. Animal reactions in most cases seem to be

more of the nature of fluctuations than of mutations, to

borrow a related phraseology; they are individual idio-

syneracies that are insignificant so far as the race is con-

cerned and are usually not interfered with because of the

generous latitude permitted by the environment. From

this standpoint animal reactions have a variety whose

explanation is to be sought for, as adaptations, but as an

expression of the momentary physical and chemical

make-up of the individual, a condition which does not

easily repeat itself and which therefore agrees with the

diversity of reactions exhibited by the organism.

Yet it is not for a moment to be assumed that adapta-

tions are not evident among animal reactions. When it

is remembered what enormous numbers of young are

lost in the process of producing one adult and that much

of this loss is due to misdirected animal reactions, it is

impossible to believe that adaptations, as roughed out by

a crude selective process, should not have become in-

grained in most animals. In fact any adequate survey of

the general field of animal reactions shows at once that

the main topographic features are adaptational and when

one reflects that this has probably been brought about in

large part by the elimination of myriads of individuals

mainly on the basis of some false step in their reactions,

one is compelled to admit that in our zeal for the study

of animal behavior, we may have missed the importance

of the lesson to be drawn from animal misbehavior. But

however this may be, I am convinced that, though the

main reaction systems of animals are essentially adap-

tive, the details of their ordinary flow of responses is

mostly free from adaptive influence and proceeds on

88 THE AMERICAN NATURALIST (Vou. XLVII

lines determined chiefly by the momentary state of the

individual concerned.

That animal reactions of an adaptive kind may possess

qualities that apparently exceed the possibility of origin

through selective operations has often been pointed out.

In fact it is from this standpoint that adaptation as a

sort of transcendental property of the organism. has

gained its most ardent votaries. And it must be admitted

that the illustrations given in support of this view are

most baffling and perplexing to the opponents. That a

dog which has had its diet changed from bread to meat,

should quickly exhibit a change in its pancreatic juice

from a type well adapted to bread and poorly adapted to

meat to another in which the reverse is true, is a fact of

adaptation the explanation of which seems beyond reach.

Here we are face to face with what appears to be a quick

adaptation of a thoroughly successful kind and without

the intervention of nervous activity. No wonder that in

face of facts, such as these, the more speculative members

of the biological camp seize their entelechies as the only

weapons with which they may hope to do battle. But

after all is the entelechy a reliable weapon. In all reac-

tions of the kind just mentioned, we are prone to say that

though there is not the least reason to suppose that intel-

ligence has really intervened, the whole affair passes off

as though directed by some such agent; hence we assume

some intelligence-like factor, some entelechy, to be active

for the time. But when we look at the matter deliber-

ately, we must admit that intelligence is only our own ex-

pression for that aggregate of nervous states and actions

which is our chief means of adaptation. To say then that

one category of adaptive acts, the adjustment of secre-

tions to particular kinds of food, has a fundamental re-

semblance to another category of adaptive acts, our in-

telligent performances, is not to offer an explanation but

to leave the matter where it was. When we know more

of the real nature of intelligence, we shall be in a better

position to attack the problem of adaptive reactions, and,

No.554] ADAPTATION IN ANIMAL REACTIONS 69

conversely, when we know more about adaptive reactions,

we shall be in a better position to attack the problem of

intelligence. Meanwhile, do not let us deceive ourselves

by confusing an argument in a circle with real progress.

In attempting a solution of the problem of adaptive reac-

tions, it is well to remember that entelechies and other

like notions do not really bring us forward, for they are

at most soporifics to the mind that would naturally be

excited to research by precisely those questions that they

tend to obscure. 7

In conclusion then I would maintain that the details

of animal reactions are in the main free from adaptive

restraint and that their diversity is dependent chiefly

upon the fluctuating momentary condition of the animal

body; further, that the main outlines of animal reactions

are adaptive, but that when we attempt to explain this

condition by assuming that it is dependent upon some-

thing like intelligence, we are arguing in a circle, for in-

telligence is merely our name for our own ch‘ef means of

adaptation.

ADAPTATION FROM THE POINT OF VIEW OF

THE PHYSIOLOGIST"

PROFESSOR ALBERT P. MATHEWS

THE UNIVERSITY OF CHICAGO

I FEEL much ashamed in having to expose my intellec-

tual nakedness before the members of this society. When

I came to this meeting I supposed that adaptation, or the

fitness of organisms to their environments, was a physio-

logical truism; that fishes were fitted by their structures

and functions to a life in the water; that frogs were so

constituted that they could live either on land or in water ;

and I was even so ignorant as to believe that many struc-

tures of a bird’s body adapted it to flight. But it appears

from the paper of one of my colleagues that in all of these

things I was most woefully mistaken.

I feel some hesitation, also, in appearing before a

society composed largely of American students of genet-

ics, for I have no new and confusing terminology to pro-

pose; and owing to my ignorance of the language they

speak and of the short-hand symbols sometimes employed,

I am, perforce, compelled to speak in ordinary English

which may be understood by any one; all of which, I fear,

must invest all I have to say with an air of superficiality,

or even of simplicity. I am besides a confirmed con-

servative in the matter of evolution, holding fast to the

explanation of adaptation given by Darwin of natural

selection of small variations; having little or no con-

fidence that genes, unit characters, mutations, saltations,

allelomorphs, determiners, inhibitors, dominants and re-

cessives, genotyes and phenotyes, are anything more than

ghosts, without substance; and looking always for simple

explanations of a physical and chemical kind, capable of

* Read at the Symposium on Adaptation at the meeting of the American

Society of Naturalists, Cleveland, January 2, 1913.

90

No.554] ADAPTATION AND THE PHYSIOLOGIST 91

expression in ordinary language, of the apparent com-

plexities of evolution. I avow myself as a physiologist to

be a follower of Darwin, admiring his methods of careful

experiment and observation, his long cogitations, and

with confidence in the soundness of his judgment. There

has been a tendency of recent years in certain quarters

to belittle his work, to make fun of his conclusions, to

deny that evolution has been a slow and steady continu-

ous process, as the rocks show, and to assert that it has

taken place by a hop, skip and a jump, and that it would

have taken place anyway without natural selection.

Physics and chemistry have attempted to express the

physical world in terms of matter and energy, and many

biologists are attempting to extend this method to the

living world. While this is a necessary and admirable

thing to do, it must not be forgotten that in doing so

they are neglecting the main fact of life, consciousness,

and that the phenomena of life can not be accounted for

if this is neglected. It is obvious, too, that the physicist,

with his present conception of matter and energy, is mak-

ing as great a mistake in neglecting the psychical side of

matter as the biologist would make if he neglected the

physical side. For the psychical, like the physical, must

be due to the properties of the atoms, or at least is asso-

ciated always with them. For'the atoms are the same in

living and lifeless, their properties are inherent in them

and can not be taken away and added to them as if they

were wagons, which changed horses, as Du Bois Ray-

mond has put it.

It is my opinion that physiology comes powerfully to

the support of Darwin’s conclusions; that it shows clearly

that there are no such things as independently variable,

unit characters; that a jump is a physiological impossi-

bility; and that most so-called mutations are in reality

reversions, as Darwin thought; and inthis position physi-

ology is, I believe, supported by paleontology.

But while accepting many of Darwin’s conclusions, we

must all admit that many phenomena are very hard

92 THE AMERICAN NATURALIST [ Vou. XLVIL

to understand on the basis of Darwin’s explanations.

Among these difficulties, most of which were recognized

by Darwin, there are the phenomena of parallel evolu-

tion among different species and genera, which, though

diverse, appear all to be moving forward in the same

direction; the phenomena of steady, limited progress in

one direction which point toward orthogenetic variation ;

the phenomena of the appearance of rudiments and their

development until useful. It is exactly these difficulties

upon which physiology throws some light; and it is of

them that I particularly wish to speak.

In the evolution of animals two movements may be

perceived: a spreading out and a progress; a diversifica-

tion and a movement forward. The first movement is

illustrated by the formation of many different species

in one genus; or of many genera of the same type of

animal; the second by the movement forward in the line

of nla oR of all these species. These two movements

were not sharply distinguished by Darwin, but they have

been more or less clearly recognized by several philoso-

phers. It is this double movement which has given the

animal kingdom the form of a branching tree instead of .

a single trunk. Darwin dealt mainly with the first of

these movements, which gives rise to genera, species and

varieties; which is shown by the diversification of ani-

mals sad plants in domestication by human selection ;

and he explained it by the progressively better adaptation

of forms to particular environments. He believed the

second movement, the movement upward, was due to the

same cause.

It is the second movement which has been so hard to

explain and which has particularly puzzled the paleon-

tologist; the successive series of dominating types on

the earth’s surface culminating in man; the progress

steadily toward the goal of consciousness and intelligence.

The question which I wish to raise is whether these two

movements, which are at right angles to each other, may

not be due to the natural selection of two different kinds

~-

No. 554] ADAPTATION AND THE PHYSIOLOGIST 93

of adaptations: first, adaptations of form and function to

different kinds of environments; and second, the natural

selection of the function of irritability, or, in other words,

to the selection of adaptability, or the adaptation to

changeableness of environment. Selection of the first

kind of adaptations may have given rise to varieties,

species, genera of the same type of animal, and have pro-

duced the spreading, or diversification; while selection of

the second kind of adaptation may have produced the

movement onward and upward of all animal forms.

These two kinds of adaptations do not always go

together and selection of the one may outweigh the other.

It is because selection to a specific environment some-

times is more important than selection of adaptations to

changeableness, that not all organisms have progressed

in the scale of evolution equally rapidly: but some have

persisted in special environments with slight changes of

structure for very long periods, or may even have retro-

gressed; while other forms, in which the second adapta-

tion has been rigorously selected, have moved rapidly on-

ward and upward, and show little adaptation to any

Special environment.

The question whether evolutionary progress is due to

the selection of this second adaptation, that of adapt-

ability, occurs very naturally to a physiologist, because,

in the first place, the evolutionary development of con-

sciousness and intelligence appears to him to be one of

the most important, if not the most characteristic move-

ment in evolution; and in the second place, his point of

View in considering evolution and adaptation is some-

what different from that of the zoologist or the paleontolo-

gist. To him the organism does not appear constituted

of bones, skin, horns, or other structures, but to be consti-

tuted essentially of a number of mechanisms in activity,

each mechanism having a definite function to perform.

Evolution, for the physiologist, is not evolution of struc-

ture primarily, but evolution of function; and he natu-

94 THE AMERICAN NATURALIST [ Vou. XLVII

rally expects to find that the adaptations of function have

been of great importance in determining survival.

Of all the physiological properties of the original proto-

plasm upon which natural selection might be supposed to

act, irritability, the most fundamental property of living

matter, would seem the most probable point of attack;

for irritability is that property of protoplasm in virtue

of which it adjusts itself to its environment. It is the

property of response; and since it is the environment

which is acting as the judge of the excellence of the

response and doing the selecting, it would seem that it

must be upon this property that all organisms must be

tested. It is, moreover, this property that Spencer has

very acutely selected as the most fundamental charac-

teristic of living organisms, namely, the power of continu-

ous adjustment of internal to external conditions. It

would seem probable that however well animals might be

adapted to special environments by the action of natural

selection, this particular property, or function, which has

to do with the continuous adjustment of internal to ex-

ternal relations must have been throughout the whole

course of evolution of predominant importance. And if

there has been any unity in the progress; if the course of

evolution has been at all in any single direction; and if

the natural selection theory is true; it must be in the

direction of the perfecting of this function.

I think this short statement will make it clear why the

physiologist turns naturally to this fundamental quality,

or property of living things, when he considers evolution

and adaptation; for however organisms may vary in

structure or other particulars, they all have irritability

in common. Moreover, I think most physiologists will

agree with me that this particular property has been too

often neglected by most students of evolution, among

whom physiologists have been unfortunately very rare.

Irritability shows itself in all cells by the power of

internal change in response to an external change. In

most cells of the body there is nothing especially adaptive

No.554] ADAPTATION AND THE PHYSIOLOGIST 95

in the nature of many of these responses; but it is quite

otherwise, if we consider the organisms as wholes. It

is clear that all organisms have not only the power of

reacting to an external change, but many of their reac-

tions are adaptive to a surprising degree. This is indeed

the very crux of the difference between living organisms

and lifeless things. A lifeless thing can not adjust its

internal to its external relations so that it can continue

to exist in a changed environment. A crystal in a solu-

tion of its kind must dissolve, if the concentration is kept

ever so little below saturation; a whole universe must

pass away, if anywhere within it there is a persistent

uncompensated difference of potential. With living things

it is quite otherwise. They have the power of interposing `

resistances to the potential difference. All living things

without exception have adaptive responses so that they

are able to continue in existence even though their sur-

roundings change in many different ways. They possess

adaptability. Their responses due to their irritability

are adaptive responses. The irritability of the organism

as a whole is, then, above everything else characterized

by power of adaptive response.

It is not difficult to imagine how this specialization of

the general property of irritability arose. Some of the

indefinite responses of the original organisms to environ-

mental change protected the organism against the change.

Organisms with such responses survived and their de-

scendants had the property of a limited adaptive response

to this particular change. From this crude beginning

further progress was easy. The changes in the environ-

ment, though many, are not indefinite in number, and

adaptations in the nature of direct responses easily arose

and were perfected.

Adaptability, then, appears to the physiologist as the

master word of evolution. And many facts also may be

urged as confirming this conclusion. For example, one

and all of the great physiological mechanisms of the body

have a single purpose: to secure adaptability. Not to

96 THE AMERICAN NATURALIST [Vou. XLVII

adapt an organism to one environment, but to all environ-

ments, and thus to make it superior to all environments.

Furthermore, the higher organisms are specially remark-

able for the development of that master tissue of the body

which is preeminently irritable and of which the main

function is the adjustment of internal to external rela-

tions, the nervous system; and finally that the inference

is sound may be concluded from the fact that it is by

adaptability and by no other quality whatever that organ-

isms may be arranged in the order of their evolutionary

progress.

It is not at all surprising that adaptability should be

the most important adaptation in nature, overpowering,

except in special cases, and dominating all others. For

there is but one certain thing in nature: namely: uncer-

tainty. The most constant feature of all environments,

but particularly of land environments, has been their in-

constancy. Changeableness is the chief characteristic of

all environments, whatever their special characters may

be. There are changes of light, temperature, climate,

oxygen and carbon dioxide, moisture; changes due to

the introduction of new species by migration upsetting

nature’s balance; changes in the food supply. Climates,

flora and fauna change; change alone persists. Change

is the essential thing. We may expect, therefore, if

Darwin be correct in his conclusion that variation and

natural selection account for evolution, that adaptation

to changeableness must be the chief adaptation in nature,

and more than all others, it must have determined the

general course of evolution. This is found to be the case

and the great physiological mechanisms of the body are

designed, as already stated, to subserve this fundamental

adaptation. Adaptability is that power which fits organ-

isms to withstand the unexpected: the vicissitudes of life;

special adaptations of form and color may contribute to

` the survival of animals; but the essential, or root, adapta-

tion is to changeableness. By adaptation to all environ-

ments they become finally superior to all environments.

No.554] ADAPTATION AND THE PHYSIOLOGIST 97

Superiority to environment, and not adaptation to it, is

secured through the irritability of the organism con-

sidered as a whole.

The great mechanisms of the body which have this

function are several. First, the heat-regulating mechan-

ism, for by means of this organisms are rendered inde-

pendent of the temperature of their environments. They

can exist in the tropics or in the arctics and withstand the

extremes of our own climate, while maintaining their

activities. This is a complex mechanism consisting of

insulating material in the skin; trophic nerves to the

internal organs; a closed vascular system; a power of

rapid oxidation; supra-renal capsules; pancreas; nervous

coordination; sweat glands; evaporation of water in the

lungs; temperature nerves. More than any other this

mechanism enabled the mammals to conquer the reptiles

and supplant them. The mammals became independent

of the temperature of their environments. A mechanism

not coming by jumps, but the rudiments found far down

in the fishes and slowly evolved.

A second fundamental mechanism of great importance

for the mammals in supplanting the reptiles and other

animals probably was that concerned in immunity. Most

of the toxins of poisonous reptiles are of a protein nature.

The mammals have developed a mechanism, the details

of which are still obscure, but which apparently consists

in the conversion of these protein toxins into bodies which

neutralize the toxins from which they are formed, that is,

into antitoxins, We find, as a matter of fact, that at least

many of the mammals are able apparently to make an

anti-toxin out of any kind of a foreign protein. Besides

this mechanism of defense, useful against bacteria, as

Well as against snakes, there is the primitive mode of

Phagocytosis and the chemical method of defense, which

consists either in the prevention of absorption, or in the

chemical neutralization of the poison by union with other

Substances. Thus the toxicity of phenols, benzoic acid

and many alkaloids are neutralized. By this mechanism

98 THE AMERICAN NATURALIST (Vou. XLVII

mammals are rendered superior to the attacks of many of

their enemies and to this extent rendered superior to their

environments.

Third, there is the mechanism for rendering mammals

tolerably independent of the moisture content of their

environment, a mechanism most highly developed in the

reptiles. A mechanism formed by the replacing of the

wet skin of the amphibian by a dry or scaly skin; the

perfecting of the kidneys to maintain osmotic pressure

of the blood; the control of the sweat glands and loss of

water by the intestines; the development of membranes

non-permeable to salts, so that animals may sit in fresh

water and not lose their salts. One of the most interest-

ing parts of this mechanism is shown in the reptiles and

birds, in the substitution of uric acid for urea in their

excretions. By this improvement reptiles have secured

almost complete independence of the water content of

their environments. They make enough water in their

own bodies to supply their small losses. This again is

a mechanism of which we can trace the steady growth

without a break from the invertebrates to man.

A fourth great mechanism makes mammals independent

of barometric fluctuations and less dependent on a fixed

atmosphere. By means of their blood loaded with hemo-

globin carried in corpuscles lacking all oxygen-consuming

power, they are able to live on lofty plateaus, or in deep

valleys; and in the presence of much or little carbon

dioxide.

The mechanisms having to do with reproduction and

the caring for the young afterward have this same

advantage of rendering the mammals independent of

environment.

A sixth mechanism is the alimentary mechanism, most

highly perfected in man. This has rendered him inde-

pendent of any particular kind of food. He can make

‘his body of any kind of plant or animal. He can make

carbohydrate out of protein and many other things. He

ean live in any climate largely because of this mechanism.

No. 554] ADAPTATION AND THE PHYSIOLOGIST 99

Again a complex mechanism, consisting of teeth, of diges-

tive glands tearing proteins and carbohydrates to pieces,

so that he can build up his own proteins from any other

kind, useless amino acids being converted into sugar and

urea.

The last and by far the most important of these great

mechanisms of adaptability is that which provides for

every contingency; for the unexpected. It seems that

nature, after elaborating these other mechanisms to meet

particular vicissitudes, has lumped all other vicissitudes

into one and made a means of meeting them all. One

can not but be pleased by the apparent ingenuity of this

solution. I refer to the nervous mechanism. It is obvious

how this mechanism, by substituting choice for blind in-

stinct, consciousness for unconsciousness, developing

memory, so that one can profit by experience, and intelli-

gent adaptation of means to ends, has provided finally for

all possible contingencies of the future. She has spoken

her last word. Adaptability, or superiority to environ-

ment, was the end so blindly sought ; memory, conscious-

ness, choice were the means, shall I say the means as

blindly adopted?

To the physiologist, then, adaptability appears to be

the touchstone with which nature has tested each kind of

organism evolved; it has been the yard stick, with which

She has measured each animal type; it has been the

counterweight against which she has balanced each of her

productions. However well adapted to a specific environ-

ment a type might be, did it lack ever so little of its possi-

bilities in this direction, it was sooner or later relegated

to the scrap heap. Some forms, to be sure, persisted in

Special environments, where they were protected from.

competition, as in Australia; or where the environment

was fairly constant, as in the sea; or in special environ-

ments for which they were highly suited; but the whole

trend of evolution, with these exceptions, may be summed

up by the statement: the general course of evolution has

been always from the beginning to the end, in the direc-

100 THE AMERICAN NATURALIST [Vou. XLVII

tion of increasing adaptability or increasing perfection

of irritability. This law may be put by the side of the

law for the evolution of universes: all spontaneous change

is in the direction of increasing entropy.

It is not by form, by color, by increasing complexity or

simplicity, that animals may be classified in the order of

their evolutionary appearance. It is by this property of

adaptability and this alone. At the summit is man; now

consciously attempting to carry on what nature has been

unconsciously attempting these millions of years, and to

secure mastery of his environment. Below him are the

other placental mammals of lower intelligence; beneath

them the marsupials, less adaptable than the mammals,

because of lower brain power; then the reptiles independ-

ent of water, but not of temperature; the amphibia, only

partially independent of water, but not of temperature;

the teleosts able to live in salt and fresh water; the

selachians, most without osmotic control and limited to

the sea; the arthropods living on land and sea, but depend-

ent on temperature, food and climate, cramped by an

external skeleton, and with the fatal defect of running

the alimentary canal through the nervous system, so that

for higher brain power, either a new nervous system or a

new alimentary canal would be needed; lower still the

molluses and annelids, closely limited to their environ-

ments; and last the echinoderms and protozoa. No adap-

tation or power of the body has been so consistently

attacked by natural selection as this; and it is this prop-

erty which seems to have been the determining factor in

the general course of evolution and to have determined

the steady development of the psychic powers.

I come now to the second part of my subject, namely,

correlation. By the first part I have attempted to show

that the selection of variations in adaptability is respon-

sible for at least a part of the steady progress in one

direction of many kinds of animals; and explains that

unity of progress which has been one of the main causes

for assuming orthogenesis. In this second part of the

No. 554] ADAPTATION AND THE PHYSIOLOGIST 101

paper, I hope to show that the development of our knowl-

edge of correlation removes some other difficulties which

Darwin had to meet, and probably explains some other

facts which have been urged as supporting orthogenesis.

Among the puzzles of evolution has been the steady

growth of rudimentary structures which have apparently

no function until they are considerably developed. I say

apparently no function, for the physiologist has learned

to be very cautious in saying that any part of the body

is without function or use. A few years ago it was quite

otherwise and it was supposed that various rudiments,

like the appendix, the hypophysis, the pineal gland, the

thymus and some other organs were without function;

the surgeons were busy explaining how much better we

were off without them; and the anti-Darwinian was fond

of presenting these things as not consonant with the view

of adaptation. At the present time the uselessness of

these rudimentary structures is no longer affirmed. We

must therefore be very cautious in supposing that any

structure we see, no matter how insignificant it may

appear, is without importance. Darwin himself felt the

great fact of correlation, and his pangene theory was

invented, in part, to account for these facts. He would

be both astonished and delighted could he know how com-

pletely physiology has vindicated his appeal to correla-

tion as the explanation of some difficulties.

Modern physiology has shown that the whole animal

organism is correlated by means of internal secretions;

that there is but one unit in the body, and that is the

whole organism. By the work of Knowlton and Starling

we now have the final proof of the correlation of the pan-

creas and muscles. The correlations between the hard

and soft parts of the body are of still greater importance

to the paleontologist, for it has been shown that the hard

parts are not independently variable, but that they are

dependent at every point upon the function of the soft,

internal organs. Who would have dreamed that the char-

acter of the skin, the hair, the shape of the skull, the in-

102 THE AMERICAN NATURALIST (Vou. XLVII

telligence itself, the length of the limbs, or the speed of

transformation of a tadpole into a frog would be depend-

ent on the thyroid or thymus gland? That the minute

parathyroids should be absolutely necessary to the life of

an organism thousands of times their weight? Or that

the development of the testes, the change of milk teeth

to the permanent dentition, the growth of the bones of

the extremities, should be dependent on the anterior lobe

of that apparently useless rudiment, the hypophysis? Or

that the secretion of milk and urine should depend on the

posterior lobe of the same organ? Who had the temerity

to suggest that the corpus luteum should be influencing

the development of the mammary glands? Do we not see,

indeed, that most of the characters of the body which

have steadily developed from the fishes to man are

secondary characters dependent on the anterior develop-

ment of these ductless glands? Is this fact without sig-

nificance to the paleontologist in helping him to under-

stand the apparent steady progress in one direction, the

appearance of orthogenesis? It will be asked, perhaps,

what has caused the steady development of these glands.

But the answer is not difficult. They are, in their turn,

parts of the mechanism of adaptability which has been

consistently selected in evolution. They are concerned

not only in the growth of bone, but in the growth of the

nervous system, the heat control of the body, the im-

munity mechanisms, the efficiency of muscles, and are in

the chain of reproduction itself. These facts largely

remove, in my opinion, the difficulties in understanding

how rudimentary organs could be useful.

But not only do these facts remove these difficulties in

the way of the selection theory, but they have a no less

important bearing on the problem of heredity. They

show that there can be no independently variable quali-

ties in the animal body. The body is a unit, and I, at

least, can imagine no part of it which can vary without

influencing other parts. Correlations are everywhere.

Pigment is often cited as a unit character, but how can it

No.554] ADAPTATION AND THE PHYSIOLOGIST 103

be so? Pigment is itself the result of a long and complex

series of changes. Ifa given cell produces no pigment it

is perfectly certain that its other chemical processes

are to some degree modified also, so that these other

things vary also. If this cell is changed so that it pro-

duces no pigment, then since it is the logical result of a

long series of changes in the developing organism, those

changes must have been different in animals producing

pigment and no pigment. But this means, since each

process in the early stage of development influences a

multitude of processes in the final change, that there must

be a host of differences correlated with the pigment

change. As a matter of fact, Darwin long ago pointed out

that pigment production was apparently correlated with

other factors; particularly with vital resistance, a fact

repeatedly mentioned to the writer, also by Whitman as

a result of his experiments in pigeon breeding. Darwin

cites the case of the Virginia pigs of which only the black

ones could eat a poisonous root without losing their hoofs;

and Whitman told me that always birds deficient in pig-

ment were also somewhat deficient in other characters

and were weaker. :

The essential unity of the organism is not only fatal to

the whole theory of unit characters, but it is an insuper-

able objection to the theory that evolution has been by

jumps. The organism is a finely adjusted mechanism of

a very complex kind; it seems impossible to a physiolo-

gist that one can cause a sudden large change in any part

of it and have it continue to function; it is as incredible

as if one should remove one of the wheels of a watch,

replace it by a larger one, and expect the watch to con-

tinue to run. Such a simple matter as the replacement

of urea by uric acid as an excretion, a change which the

reptiles introduced in their differentiation from the am-

phibia, a change which might conceivably be brought

about by the dropping out of a uricolytic enzyme, could

not take place suddenly. The kidneys and all other organs

of the body would need to be adjusted to this change.

104 , THE AMERICAN NATURALIST [ Von, XLVII

Finally correlation has greatly enhanced the value of

the old idea of checks in development and shows most

clearly that no organ of the body ever reaches its full

potentialities. What comes out of an egg is but one of

the infinite potentialities contained in it. Velocity of de-

velopment, like every other chemical reaction, is equal to

the affinity divided by the resistance. If resistances are

increased, or if vitality, in other words chemical affinity,

be reduced, the development must stop sooner than

normal; and we have the phenomena of reversion. If,

on the other hand, the reverse takes place, if vitality is

increased or resistance reduced, we have variation in the

direction of evolution. The development of nonviable

monsters is at one extreme of this process. Ontogeny is

like a runner, taking the first hurdles easily, but always

with increasing difficulty, sometimes trippling at one,

sometimes at another, but never reaching the end of his

race. :

In conclusion then: to the physiologist it appears that

the best explanation of adaptation is that given by Dar-

win of natural selection of small variations; that the

essential unity of the progress in evolution toward con-

sciousness and intelligence has been due to the natural

selection of the fundamental property of irritability, for

it is in virtue of this property that adaptability of organ-

isms has been increased. The recognition of this fact re-

moves one of the difficulties in the way of Darwin’s the-

ory. And,second, physiology by the establishment of the

` physiological correlation of all parts of the body, hard

and soft, interposes a final objection, in my opinion, to

the whole theory of unit characters, of independent vari-

ability of characters, and to the theory of evolution in

any other way than by a slow and gradual process, which

shall give time to the readjustments of every part of the

body necessitated by a change, however slight, in any

part of it.

THE FITNESS OF THE ENVIRONMENT, AN IN-

QUIRY INTO THE BIOLOGICAL SIGNIFI-

CANCE OF THE PROPERTIES

OF MATTER!

PROFESSOR LAWRENCE J. HENDERSON

HARVARD UNIVERSITY

Darwinian fitness is compounded of a mutual relation-

ship between the organism and the environment. Of this,

fitness of environment is quite as essential a component

as the fitness which arises in the process of organic evo-

lution; and in fundamental characteristics the actual

environment is the fittest possible abode of life. Such

is the thesis which I seek to establish. This is not a novel

hypothesis. In rudimentary form it has already a long

history behind it, and it was familiar doctrine in the early

nineteenth century. It presents itself anew as a result

of the recent growth of the science of physical chemistry.

In the study of fitness it has been the habit of biolo-

gists since Darwin to consider only the adaptations of the

living organism to the environment. For them in fact

the environment, in its past, present, and future, has been

an independent variable, and it has not entered into any

of the modern speculations to consider if by chance the

material universe also may be subjected to laws which are

in the largest sense important in organic evolution. Yet

fitness there must be, in environment as well as in the

organism. How, for example, could man adapt his civi-

lization to water power if no water power existed within

his reach?

At first sight it may well seem that inquiry into such

* Read at the Symposium on Adaptation at the meeting of the American

Society of Naturalists, Cleveland, January 2, 1913.

` This paper consists chiefly of excerpts from a book of the same title soon

to be published by the Macmillan Company.

105

106 THE AMERICAN NATURALIST [Vou. XLVII

a problem must end unsuccessfully in vague and unprofit-

able guesses. But the physico-chemical basis of life it at

length firmly established. On the whole, the composition

of living matter, its physical structure, the changes of

matter and energy which constitute the metabolic process,

together with the totality of such changes, which make up

the fundamental economic process of that largest com-

munity which consists of all living beings, are all clearly

defined.

THE CHARACTERISTICS OF LIFE

Under these circumstances it is certainly no rash enter-

prise to seek a definition of some of the essential char-

acteristics of life.. Although it is probably far beyond

our present power to make a complete study of the prob-

lem, I feel sure that a brief analysis will justify certain

very definite conclusions. Life as we know it is a physico-

chemical mechanism, and it is probably inconceivable

that it should be otherwise. As such it possesses, and,

we may well conclude, must ever possess, a high degree

of complexity—physically, chemically and physiologi-

cally, that is to say, structurally and functionally. We

can not imagine life which is no more complex than a

sphere, or salt, or the fall of rain, and, as we know it, it is

in fact a very great deal more complex than such simple

things. Next, living things, still more the community of

living things, are durable. But complexity and dur-

ability of mechanism are only possible if internal and

external conditions are stable. Hence automatic regula-

tions of the environment and the possibility of regulation

of conditions within the organism are essential to life.

It is not possible to specify a large number of conditions

which must be regulated, but certain it is from our present

experience that at least rough regulation of temperature,

pressure, and chemical constitution of environment and

organism are really essential to life, and that there is

great advantage in many other regulations and in finer

regulations. Finally a living being must be active, hence

its metabolism must be fed with matter and energy, and

No. 554] FITNESS OF ENVIRONMENT 107

accordingly there must always be exchange of matter and

energy with the environment.

Obviously these few conclusions can make no claim to

completeness. Fully to describe life, the discovery of

many other fundamental characteristics is necessary, in-

cluding such as are related to inheritance, variation, evo-

lution, consciousness and a host of other things. But in

the formation and logical development of such ideas

there is danger of fallacy at every step, and, since the

present list will suffice for the present purpose, further

considerations of this sort are best dispensed with. This

subject should not be put aside, however, without clear

emphasis that the postulates which have been adopted

above are extremely meager. The only motives for aban-

doning further search are the economy and the security

which are thereby insured and the very great difficulty of

extending the list.

THe ENVIRONMENT

Even at the earliest period in the evolution of a typical

Star there appears to be a progressive variation in the

chemical composition from center to periphery. Theo-

retically it seems inevitable that the heaviest elements

Should be concentrated in the interior and that those of

lowest atomic weight should be present in the greatest

amount near the surface. Actually, spectroscopic investi-

gation fully confirms this view. Thus the spectra of

typical hot stars show that hydrogen is an inevitable con-

Stituent of their superficial parts. Indeed the universal

Presence of hydrogen under such circumstances 18 un-

doubtedly one of the most clearly established facts of

stellar astronomy. As stars cool and become red the

Spectral changes quite as unmistakably point to the pres-

ence of carbon. Accordingly we possess the best of evi-

dence and the best of reasons for the belief that large

quantities of hydrogen and carbon must exist at or near

_ the surface when a crust forms upon a cooling star.

The nature of the chemical combinations into which

these elements at first enter is perhaps open to some

108 THE AMERICAN NATURALIST (Vou. XLVII

question. But as the temperature falls in the cooling of

a sun or planet the affinities of carbon and hydrogen for

oxygen increase so that carbonic acid and water must

normally result. For oxygen is almost certainly present

in the sun, it is found in meteorites, and the vast store of

it in the earth’s atmosphere and crust (roughly one half

of their total mass) justify the belief that it is every-

where one of the commonest of elements. Hence an

atmosphere containing water and carbonic acid appears

to be a normal envelope of a new crust upon a cooling

body. Even were not these substances at first present in

such an atmosphere, volcanos must soon belch them forth

in enormous quantities, to relieve the pressure which in-

evitable chemical processes set up.

In short just as living things permit themselves to be

simplified into mechanisms which are complex, regulated,

and provided with a metabolism the environment may be

reduced to water and carbonic acid. These are simplifica-

tions counselled solely by expediency. Neither logical

process is necessary, each involves a disregard for many

circumstances which might be of weight in the present

inquiry. But inthe end there stands outa perfectly simple

problem which is undoubtedly soluble. That problem ~

may be stated as follows: In what degree are the physical,

chemical, and general meteorological cl teristics of

water and carbon dioxide, the primary constituents of

the environment and of the compounds of carbon, hydro-

gen and oxygen favorable to a mechanism which must be

physically, chemically, and physiologically complex, which

must be itself well regulated in a stable environment,

. and which must carry on an active exchange of matter

and energy with that environment.

The first step in seeking a solution must be to review

the data of physics and chemistry which describe the

properties of water and carbonic acid, having due regard

to their meteorological significance. Such data of the

highest accuracy exist in great profusion, for almost

every conceivable property of these substances has been

No. 554] FITNESS OF ENVIRONMENT 109

studied with patient care. Next, the properties of the

compounds of carbon, hydrogen, and oxygen must be con-

sidered, and some of the characteristics of the chemical

reactions into which they enter must be discussed. For

this examination the unparalleled development of the

science of organic chemistry provides ample material.

All of these things must be scrutinized quantitatively as

well as qualitatively, and again there is no lack of neces-

sary formation.

Immediately one advantage of the method here pro-

posed becomes evident. We can deal with the familiar

abstractions of physical science—specific heat, coeff-

cient of expansion, solubility, heat of reaction, ete—and

thereby we shall gain all the advantages of the most exact

sciences. No qualifications, no doubtful or contentious

matter, no imperfect descriptions need enter.

In this manner it will be easy to estimate the absolute

biological fitness in certain respects of water and carbonic

acid, and at once a host of automatic results of their

properties will become evident. Many of these results,

such as the nearly constant temperature of the ocean, the

ample rainfall, the freezing of water upon the surface,

the great variety of carbon compounds, are familiar sub-

jects of speculation, though since Darwin little interest

has been manifested in them; others, only recently

brought to light by the growth of physical science, are.

nearly or quite unknown in this connection. All deserve

to receive more serious attention from biologists than is

at present vouchsafed them, for they constitute a part of

the very foundation of general biology, and they cause

many of the phenomena with which man is concerned in

his struggle for mastery of the environment. :

Yet the mere exposition of such facts and relationships.

can not suffice in a discussion of the fitness of the environ-

ment. In the first place these are in the main familiar

ideas, and if they were altogether conclusive to prove the

existence of really significant fitness, if they could be

regarded as alone adequate to establish the necessity of

fe

110 THE AMERICAN NATURALIST [Vou. XLVII

putting fitness by the side of adaptation as a coordinate

factor in causing the marvels of life, it is hard to believe

that they would have been so long neglected. In the sec-

ond place there is nothing comparative about such in-

formation. Water is indeed a wonderful substance which

fills its place in nature most satisfactorily, but would not

another substance do as well? Is not ammonia, for

example, a possible substitute? And are there not many

other chemical bodies which might, in a very different

world, serve equally useful purposes? Perhaps, too, the

great variety of carbon compunds which are known to

the chemist are known only because the vital processes

furnish an abundance of material with which to experi-

ment. Is it not possible, therefore, that another element,

perhaps for instance silicon, may enter into even greater

varieties of compounds? It is such questions, ever pres-

ent in the minds of men of science yet never yet care-

fully scrutinized to see if an answer be possible, which, I

suspect, have long deflected attention from this subject.

Clearly, therefore, it will be necessary to compare the

properties of water and carbonic acid and of the carbon

compounds with those of other substances. It will be

necessary to find out whether these substances are not

only fit but fittest—and this no doubt is a task of a very

different sort. It may even seem, at first sight, an im-

possible one, but I hope to show that this is not the case,

and that in spite of the incompleteness of our physical

and chemical knowledge, it may be pressed to a satis-

factory issue.

The very constant temperature of the ocean is a most

important factor in the economy of nature. It consti-

tutes, for example, a vital regulation of the environment

of a large proportion of all the living organisms of the

world, and it has many other important ‘‘functions.”’

This constancy of temperature is in large part due to the

magnitude of the specific heat of water. Other things

being equal the greater the specific heat of water the more

constant must be the temperature of the ocean. If then

No. 554] FITNESS OF ENVIRONMENT 111

the specific heat of water, as is actually the case, be nearly

or quite a maximum among all specific heats, it follows

that the fitness of water in this respect is nearly maximal.

Again the ocean contains an astonishing variety of

substances in solution, and they are present often in large

quantities. In this manner a very great supply of food

in very great variety is offered marine organisms. Of

course such richness of the environment is an exceedingly

favorable circumstance for the organism, and it is due

principally to the ability of water to dissolve a multitude

of things in large quantities. It is not to be supposed that

the substances present in sea water are all of use to every

organism. This need not be the case at all, but a variety

of supplies which may be adapted to special requirements

as they arise, here iodine, there copper, for instance, is

a very genuine advantage. Further the vast utility of

the solvent action of water in blood, lymph, and all the

body fluids is too patent to call for comment. If, now, it

can be shown that the solvent power of water is nearly or

quite a maximum, as it really is, among all known sol-

vents, then it must be evident that in another respect the

fitness of water is nearly or quite maximal.

Again the amount of energy that is required to tear

apart molecules of water and liberate hydrogen and oxy-

gen is very great indeed, and when hydrogen and oxygen

recombine to form water this energy must reappear,—

under ordinary circumstances as heat. This fact too is

very favorable for the organism, because almost all com-

pounds which contain hydrogen yield a great deal of

energy which can be tapped in the process of metabolism.

If therefore the heat of combustion of hydrogen be nearly

or quite a maximum, as it is, among all substances, it

is clear that water is again, in another respect, most

wonderfully fitted for life. Finally, if it be true, and

such is the case, that very few of the substances which

Share the fitness of water in one of these characteristics,

also share or approach its fitness in either of the others,

and that none possesses all these qualifications in a de-

gree that merits consideration, it must, I conceive, be ad-

112 THE AMERICAN NATURALIST [Vou. XLVII

mitted that so far as the investigation has proceeded,

water is the only possible fit substance.

A criticism may here be made, are there not other

substances which possess other groups of qualifications

which water lacks? And that is a difficulty which is

even harder to meet. But in the. first place it is evident

that there are not an infinity of important physical prop-

erties; in fact there are very few. And in the second

place it is evident, both from centuries of experience in

physical science and from the postulates above mentioned

regarding life, which undoubtedly do in the main describe

its physico-chemical characteristics, that very few prop-

erties indeed are of importance in the least comparable

with those which I have mentioned. Finally it is in the

highest degree probable that we are acquainted with most

of the truly essential physical properties, and know them

as biologically important, when they are so; and I believe

it has been possible to consider them all, and thus make

the argument complete.

Such is the nature of the argument; the facts, though

no less important than those above indicated, are far too

numerous to mention. They include the unique surface

tension of water and its very great ionizing power, the ab-

sorption coefficient and ionization constant of carbonic

acid, the extreme chemical activity of oxygen and hydro-

gen, the unique chemical combining power of carbon, the

number, complexity, variety and chemical activity of the

compounds and processes of organic chemistry, and the

vast complexity of the chemical system which inevitably

results from the reduction of a mixture of carbonic acid

and water. These properties result directly in a be-

wildering variety of conditions which in the most varied

ways promote complexity, durability and metabolism.

Analysis of all the facts justifies the following con-

clusions.

The physical and chemical properties which have been

taken into consideration include nearly all those which

are known to be of biological importance or which ap-

No. 554] FITNESS OF ENVIRONMENT 113

pear to be related to complexity, regulation and meta-

bolism. |

There are no other compounds which share more than

a small part of the qualities of fitness of water and car-

bonic acid, no other elements which share those of car

bon, hydrogen and oxygen.

None of the characteristics of these substances are

known to be unfit, or seriously inferior to the same char-

acteristics in any other substance.

Therefore the fitness of the environment is both real

and unique.

In drawing this final conclusion I mean to assert the

following propositions:

I. The fitness of the environment is one part of a re-

ciprocal relationship of which the fitness of the organ-

ism is the other. This relationship is completely and

perfectly reciprocal; the one fitness is not less important

than the other, nor less invariably a constituent of a

particular case of biological fitness; it is not less fre-

quently evident in the characteristics of water, carbonic

acid and the compounds of carbon, hydrogen and oxy-

gen than is fitness from adaptation in the characteristics

of the organism.

II. The fitness of the environment results from char-

acteristics which constitute a series of maxima—unique

or nearly unique properties of water, carbonic acid, the

compounds of carbon, hydrogen and oxygen and the

ocean—so numerous, so varied, so nearly complete

among all things which are concerned in the problem that

together they form certainly the greatest possible fit-

ness. No other environment consisting of primary Da

stituents made up of other known elements, or lacking

water and carbonic acid, could possess a like number of

fit characteristics or such highly fit characteristics, or in

any manner such great fitness to promote complexity,

durability and active metabolism in the organic mechan-

ism which we call life.

It must not be forgotten that the possibility of such

conclusions depends upon the universal character of

114 THE AMERICAN NATURALIST [ Vou. XLVII

physics and chemistry. Out of the properties of univer-

sal matter and the characteristics of universal energy

has arisen mechanism as the expression of physico-

chemical activity and the instrument of physico-chem-

ical performance. Given matter, energy and the result-

ing necessity that life shall be a mechanism, then the

conclusion follows that the atmosphere of solid astro-

nomical bodies does actually provide the best of all pos-

sible environments for life.

VITALISM

Modern vitalism consists in asserting the existence of

a directive tendency which manifests itself in or through

the organism alone and is peculiar to life.

In such speculations the properties of matter and the

processes of cosmic evolution have no place. Bergson

indeed very definitely, and it would seem gratuitously,

puts aside cosmic evolution, and also with slight reser-

vations the properties of matter, as of no essential con-

sequence in organic evolution.

Yet whoever is disposed to speculate about biological

fitness, and not even the incomparable finesse of M. Berg-

son’s dialectic can make fitness other than the most gen-

eral result of the process of organic evolution, must now

weigh well the cosmic processes. For, if allowance be

made for the results of natural selection, fitness of en-

vironment has the greater claim to be considered.

The two fitnesses are complementary; are they then

single or dual in origin? The simpler view would be to

imagine one common impetus operating upon all matter,

inorganic and organic, through all stages of its evolu-

tion, in all its states and forms and leading to worlds

like our own through paths apparently purposeful. Such

it seems to me is the natural hypothesis for the vitalist

to adopt. But then vitalism vanishes, only teleology re-

mains. Yet putting aside mechanistic differences is it

not now lost in any case? Has not modern vitalism in

accepting the limitation to entelechies or impetus de-

stroyed itself?

The situation, briefly, seems to be as follows: Two evo-

No. 554] FITNESS OF ENVIRONMENT 115

lutionary processes independently result in two comple-

mentary fitnesses, hence they are related. In the one

process the origin of fitness is in part explained by a

mechanistic hypothesis. Nevertheless many philos-

ophers, as is their right, declare that in this process a

further extra-physical influence is to be assumed. But

any one who makes such an assumption for the one proc-

ess must certainly make it for the other, thus he will be

led to see impetus or entelechies everywhere. Under

these circumstances it may be doubted if his acquaintance

with the nature of his impetus or entelechies is so inti-

mate that he will be able to distinguish the inorganic

from the organic, for he has surrendered all positive

physico-chemical differences between organic and inor-

ganic bodies and processes to the mechanist. Hence,

unless he is to make an arbitrary and unintelligible dis-

tinction, or to indulge in the spinning of cobwebs, his

vitalism has ceased to be exclusively organic, in short,

has ceased to be vitalism at all, and has become mere

universal teleology.

The whole process of cosmic evolution from its earli-

est. conceivable state to the present is, however, pure

mechanism, as the most perfect induction of physical

Science, based upon each and all of its manifold suc-

cesses in accounting for the phenomena of nature con-

clusively proves.

But if cosmic evolution be purely mechanistic and yet

issue in fitness why not organic evolution as well? Thus

once more we arrive, this time more completely, at the

negation of vitalism. Mechanism is enough in physical

Science, which no less than biological science appears to

manifest teleology; it must, therefore, suffice in biology.

We possess two arguments; the argument that, except

mechanistically, organic and inorganic phenomena are,

in such aspects as concern physical science, alike, and,

therefore, a specifically vital teleology is unnecessary,

and the argument that inorganic science unquestionably

has no need of non-mechanistic teleology. Hence we

are obliged to conclude that metaphysical teleology 1s to

be banished from the whole domain of natural science.

SHORTER ARTICLES AND DISCUSSION

MUTATIONS IN CNOTHERA BIENNIS L. ?

Ir is evident that the adherents of the mutation theory are

sensitive to the doubts freely expressed concerning the status

of G@nothera Lamarckiana, the behavior of which in throwing

off marked variants is cited as the most important evidence for

the origin of species by mutations. These doubts are in fact

criticisms of the assumption that Lamarckiana is representative

of a wild species and express the view that this plant is of

hybrid origin and that its behavior is of the sort to be expected

of a hybrid. Consequently, mutationists are likely to bring for-

ward as rapidly as possible any evidence that may seem to indi- —

cate the appearance of clear inheritable variations of a marked

character in forms of pure germinal constitution, i. e., in homozy-

gous material.

There are types of Ginothera that we have reason to believe

are now very pure and have been so for a great many years.

Such a form is the biennis of the sand dunes of Holland. This

species has apparently been established in its habitats in Holland

since pre-Linnean times. There has been little opportunity

for chance hybridization and its habits of close or self pollina-

tion in the bud are greatly in favor of the continuation of its

germ plasm in pure lines. Moreover, the type in experimental

cultures of De Vries and others has proved to be constant. I

then it could be shown that tested strains of this biennis are

able to produce new forms of specific rank or even marked varie-

ties the mutationists would have much stronger evidence in sup-

port of the mutation theory than that based on the behavior of

O. Lamarckiana, a form unknown as the component of any

native flora.

The title of a recent paper, ‘‘Mutation bei @nothera biennis

L.,’’ by T.. J. Stomps,' a former student of Professor De Vries,

naturally then arouses interest especially since he is working

with this same biennis of the sand dunes of Holland, a type well

known to a number of botanists who are conducting experimental

studies on cenotheras. A brief discussion of the claims indicated

T. J., ‘‘Mutation bei Oenothera biennis L.,’’ Biologischen

Centralblatt, XXXII, p. 521, 1912.

116

No. 554] SHORTER ARTICLES AND DISCUSSION 117

by the title of this paper, an analysis of the evidence presented

and its possible interpretation supplies the chief incentive for

this review.

The greater part of this paper consists of a discussion of cer-

tain criticisms directed against the mutation theory by those

who believe that O. Lamarckiana is of hybrid origin. Certain

objections of Stomps appear to the writer well founded,

but we shall not take the space to consider this portion of the

paper since the greater interest attaches to the value of the direct

evidence offered by him in support of the mutation theory.

en we come to the short account of the experimental work

of Stomps we find that the so-called ‘‘mutants’’ were not de-

rived from the pure Dutch biennis of the sand dunes but from

a cross between this race and a type designated O. biennis cruci-

ata. This fact seems to the writer of fundamental importance

in judging the conclusions of Stomps. It should be made clear

that the form ‘‘O. biennis cruciata’’ is recognized in the more

recent taxonomic treatments as a true species sharply distin-

guished from types of biennis by its floral characters. What-

ever may have been the origin of O. cruciata or its possible

relationship to O. biennis, a cross between these types must cer-

tainly be regarded as a cross between two very distinct evolu-

tionary lines and its product a hybrid in which marked modifica-

tions of germinal constitution are to be expected.

(nothera cruciata differs from O. biennis most conspicuously

in having very narrow linear petals, from 1-3 mm. wide, in sharp

contrast to the broad heart-shaped petals characteristic of

biennis. O. cruciata is found wild in certain regions of New

England and New York and is consequently a native American

Species. Stomps assumes that the cruciata in Holland is a mu-

tant from the Dutch biennis, but his belief rests upon no direct

evidence. Cruciata has never appeared in the extensive cultures

of the Dutch biennis grown by De Vries and Stomps. Neither

have we any direct evidence that the American cruciata has

come from any form of biennis. It is true that the species

cruciata and biennis appear to be closely related, but it is

equally clear that they constitute very distinct lines each with a

long period of evolutionary independence. I can not see the

Justification for Stomps’s attitude when he treats a cross between

the biennis and cruciata of the sand dunes of Holland as though

t were the combination of forms within the same species which

have similar germinal constitutions. ?

118 THE AMERICAN NATURALIST [Vou. XLVII

Stomps lays emphasis on the purity of his material of biennis

and cruciata which had been carried along for several years in

pure lines from original wild plants of the sand dunes. He

states that the crossing of these two forms is concerned alone

with the floral peculiarities of cruciata, since in all other char-

acters the two types are the same. It seems to the writer hardly

possible that lines so well established as biennis and cruciata

can be absolutely the same in all respects except that of flower

form, although this is obviously the most important point of dif-

ference. The American forms of cruciata are exhibiting among

themselves remarkable differences of germinal constitution.

The observations of Stomps are of interest. He obtained in

the second generations from crosses between biennis and cruci-

ata two marked variants. These are called biennis nanella and

biennis semi-gigas because of the similarity to somewhat cor-

responding variants from Lamarckiana. We are not informed

as to the proportions in which these new forms arose, a point of

importance since we should like to know whether they are very

rare, as say 1: 10,000, or more common.

The first variant, biennis nanella, appeared in the second gen-

eration of the cross biennis X cruciata. This cross gave an F,

hybrid with heart-shaped petals as in the mother plant; no

statements are made as to the size relations. In the F, genera-

tion there was a splitting into forms of biennis and cruciata;

we are told nothing of the proportions of these individuals in

the cultures or of their range of variation. One of the biennis-

like forms presented a dwarf habit which distinguished it from

the biennis parent in much the same way that nanella is dis-

tinguished from Lamarckiana. This plant, biennis nanella,

differed from Lamarckiana nanella by the same characters that

distinguish biennis from Lamarckiana. An important point of

resemblance to Lamarckiana nanella lay in its sensitiveness t0

bacteria which Zeylstra discovered within the tissues of this

type and showed to be responsible for certain abnormal char-

acters. De Vries has shown that Lamarckiana nanella grown in

a soil treated with calcium phosphate became healthy and Stomps

found that his biennis nanella responded in a similar way to this

treatment.

The second variant, biennis semi-gigas, appeared in a second

generation from the reciprocal cross cruciata X biennis. This

cross also gave an F, hybrid with flowers of the biennis type

No. 554] SHORTER ARTICLES AND DISCUSSION 119

and in the F, generation there was likewise a splitting of the

culture into forms of biennis and cruciata. One of the biennis-

like forms presented a more vigorous habit and a larger size

of buds, flowers and leaves, suggesting the differences between

Lamarckiana and its derivative gigas. The style was longer

than in biennis and self-pollination, characteristic of biennis,

was impossible. This plant proved to be almost sterile.

A count of the chromosomes as shown by mitotic figures in

meristematic tissue of young buds determined them to be 21 in

number. This important fact placed the plant in that group

intermediate between the usual types of @nothera with 14

chromosomes and that very rare variant from Lamarckiana,

called gigas, which has 28 chromosomes. It has recently been

shown that certain plants that have been mistaken for gigas have

21 chromosomes and for these the name semi-gigas has been pro-

posed. Consequently Stomps calls the plant from the cross

cruciata X biennis with 21 chromosomes biennis semi-gigas.

The observation of this remarkable plant and the determina-

tion of its chromosome count is a matter of great interest. The

fact that the number of the chromosomes (21) is not twice the

number of the parent types (14) shows that the germinal varia-

tion did not take place after a normal fertilization, for a doub-

ling of the number of chromosomes in the fertilized egg or

embryo would give a plant with 28 chromosomes. It indicates

that a gamete produced by one of the plants in the F, generation

had 14 chromosomes and that this element combining with a

normal gamete (7 chromosomes) produced this exceptional plant

with 21 chromosomes.

I have suggested? a way in which gametes of an Œnothera

might be formed with 14 chromosomes in place of the normal

number. The presence of 28 chromosomes instead of the normal

number 14, during a heterotypic mitosis in an @nothera might

come about from a somewhat earlier appearance of that prema-

ture division of the chromosomes which normally takes place as

early as anaphase of this mitosis. Thus a pushing forward of this

premature fission of the chromosomes from the anaphase to e

metaphase of the heterotypie mitosis would result in the distribu-

tion of 14 chromosomes to each pole of the spindle. Another fis-

sion introduced before the metaphase of the homotypic mitosis

would make possible a group of 4 nuclei at the end of the reduc-

* Annals of Botany, Vol. XXV, p. 959, 1911.

120 * THE AMERICAN NATURALIST (Vou. XLVII

tion divisions each with 14 chromosomes. From such nuclei

gametes would be formed with 14 chromosomes,

The position of Stomps is clear. He believes that the Dutch

biennis and cruciata have identical germinal constitutions except

for the factors that determine floral structure and therefore

with respect to other characters can be crossed as though they

were homozygous. Since the cross gave two marked variants

which differed from the parents in other respects than those of

floral structure these two plants are mutants. These conclusions

are then applied by Stomps to the problem of the status of

Lamarckiana in the following line of reasoning. Since biennis

mutates and since biennis is probably an older species than

Lamarckiana it follows that mutations among the cenotheras are

older than Lamarckiana and that consequently the mutations

of this species can not be the result of hybridization.

The line of argument rests primarily on the assumption that

biennis and cruciata have exactly the same germinal constitu-

tion except for floral characters. This I can not believe possible

considering the long evolution back of the two lines. Why did

Stomps find it necessary to cross biennis with cruciata to obtain

his ‘‘mutants’’? If homozygous in all respects except for

flower structure why should not biennis alone or cruciata alone

give the same mutants? From my point of view Stomps really

made a cross between two rather closely related species and

obtained first the segregation of flower types to be expected in

the F, generation among which from my experience I should

expect a wide range of variation, and second Stomps obtained

two marked variants due to some germinal modification as the

result of the cross.

This sort of phenomenon I am obtaining frequently in crosses

of my races of American biennis and grandiflora. The nanella

condition of dwarfed growth is very common in F, generations.

And, last summer in an F, generation a large plant appeared

with leaves so thick and stems, buds and flowers so stocky that

I have hardly a doubt but that the cytological examination will

show an increase in the number of chromosomes.

In so far as the observations of Stomps bear upon the problem

of mutation my interpretation would be exactly the reverse of

his. To me they further illustrate the same phenomenon that I

am obtaining through my hybrids of biennis and grandiflora,

namely, that behavior by which these hybrids in the F, generation

No.554] SHORTER ARTICLES AND DISCUSSION 121

throw off variants that in taxonomic practise would be consid-

ered new species readily distinguished from the parents of the

cross and from the F, hybrid. I have this past summer found

that F, hybrids similar in character to the F, will in the F,

generation repeat the performance of the F, and throw off again

some of the same marked variants.

It is a satisfaction to know that De Vries and Stomps stand

firmly by the original definition of a mutation as a germinal

variation (and this means inheritable) from a pure stock, i. e.,

from homozygous material. This is a valuable concept whether

or not mutation proves to be a rare phenomenon. Furthermore,

one of the most important lines of experimental study is that

which will endeavor to determine with precision the conditions

under which true mutations may arise. There has been a loose

usage of the term mutation which should it become prevalent will

take from the word the significance described above, and reduce it

to a meaning no more precise than that of a marked germinal

variation from any source. If the word mutation is to be kept in

the sense of De Vries it must be reserved for germinal variations

from homozygous stock.

BRADLEY Moore Davis

A CONVENIENT MICROSCOPE CASE

A VERY convenient case for holding microscopes, especially for

large, beginning courses where two or more students in different

Sections use the same instrument, is shown in the accompanying

photograph.

The case here shown was built to stand in a shallow offset in

the laboratory near a door, and fills a small space that would

otherwise be wasted. As is seen from the numbers below the sec-

tions, it will hold fifty standard microscopes. Each instrument

has a number on the base to correspond to the number on its re-

Spective section. Across the floor of each section, at the back, 18

nailed a 2 in. X 2 in. strip of wood to stop the base of the micro-

Scope and to serve as a shelf for the extra oculars. Holes of the

Proper diameter in the shelf would hold these oculars more

safely. The doors slide easily on a metal track with ball-bearing

Wheels and have brass pushes set flush with the surface of the

y ood. They may be fastened with a catch or with a lock and

ey,

122 THE AMERICAN NATURALIST (Von: XLVII

Below the case proper is a shelf to hold laboratory books. An

improvement over the case here shown would be to have two

shallow shelves, in place of one, divided into sections for the

further alphabetical distribution of the books. The case here de-

scribed is 84 inches high, 51 inches wide, and 13 inches deep,

outside measurements.

There is wasted, of course, a vertical strip about four inches

wide in the center of the case where the doors overlap, but it is

always hidden, whether the doors be open or shut.

Such a case, if well made, is practically dust-proof, and is eco-

nomical not only of space but of money as well, since the cost of

the individual microscope boxes may be saved in buying new in-

struments. <A case similar to this has been used by the writer for

several years and has proved entirely satisfactory.

ALBERT M. REESE

No. 554] SHORTER ARTICLES AND DISCUSSION 123

A LITERARY NOTE ON THE LAW OF GERMINAL

CONTINUITY

THE distinctive theory of germinal continuity or continuity

of the germ-plasm is, historically speaking, of much more recent

origin that the broader doctrine of genetie continuity from

which it was derived and with which, in the usage of some

writers, it is made synonymous. Genetie continuity in its

widest sense embodies the proposition that “living matter

always arises by the agency of preexisting living matter,’

and in a more restricted sense means that all living cells must

be derived by continuous lineage from the cells of preexisting

generations. The theory of germinal continuity, in its most

highly developed form, conceives the germinal protoplasm as

dividing into two portions, from one of which the somatic or

body cells of the offspring are developed while the other portion

is reserved unchanged for the formation of the reproductive

material of the adult individual. The general doctrine of con-

tinuity is fundamentally essential to both these theories, but

germinal continuity, at least in any Weismannian sense, always

involves the further assumption of a transmission from genera-

tion to generation of an unmodified residue of the specially

organized ‘germinal substance, the germ-plasm, through a defi-

nite series of cells, but this concept does not imply that there is

necessarily a direct connection between the germ-cells of con-

secutive generations.

To Richard Owen the credit is usually given of being the

first to recognize the distinction between body-cells and germ-

cells and thus to foreshadow the idea of germinal continuity.

Writing in 1849, he said:

Not all the progeny of the primary impregnated germ-cell are re-

quired for the formation of the body in all animals: certain of the

derivative germ-cells may remain unchanged and become included in

that body which has been composed of their metamorphosed and

diversely combined or confluent brethren: so included, any derivative

germ-cell or the nucleus of such may commence and repeat the same

Processes of growth by imbibition, and of propagation by spontaneous

fission, as those to which itself owed its origin; followed by meta-

morphoses and combinations of the germ-masses so produced, which

concur to the development of another individual; and this may be, or

“Huxley, T. H., ‘‘Lay Sermons, Addresses and Reviews,’’ New York,

1870, p. 350.

124 THE AMERICAN NATURALIST (Vou. XLVII

a! not be, like that individual in which the secondary gorm: -cell or

rm-mass was included.

When the primary division of the impregnated germ-cell takes place,

it must divide its properties with its matter between the two cells result-

ing from the spontaneous fission of its nucleus: and this result must

follow every subsequent division. It is scarcely figurative therefore to

say that the primary or parent germ-cell has equally divided its

spermatic virtue amongst its countless progeny.

Owen’s suggestions apparently received no consideration and

were later disregarded by the author himself. Somewhat similar

ideas were expressed by Haeckel? in some of his earlier specu-

lations. Galton, says Weismann,‘ was the first to express ideas

resembling the theory of germinal continuity, but these ideas

were later considerably modified.®

A clear expression of the conception of germinal continuity is

found in the writings of Jager, but his ideas made little impres-

sion, and inaccurate citation of his work has sometimes caused

his disparagement. In 1877, restating previously expressed

propositions, he said :°

The basis of heredity consists in this, that throughout whole series of

generations the germ-protoplasm of animals retains unchanged its

specific quality in spite of all external influences. In the actual ontogeny

the available germ-protoplasm may divide into two groups, the onto-

genetic, from which the existing individual is formed, and the phylo-

Seis. Richard, ‘‘On ianea London, 1849, pp. 5-6, 63-64.

* Haeckel, E., ‘‘ Generelle Morphologie,’ ? 1866, pp. 287-

‘Weismann: AS ‘*The Germ-plasm, A Theory of Heredity,’? New York,

1902, p. 198.

*Galton’s early ideas were expressed as follows:

‘‘From the well-known circumstance that an individual may transmit

to his descendants ancestral qualities which he does not himself possess, we

are assured that they could not have been altogether destroyed in him, but

ust have maintained their existence in a latent form. Therefore each

individual may properly be conceived as consisting of two parts, one of

which is latent and only known to us by its effects on his posterity, while

the other is patent, and constitutes the person manifest to our senses.

‘‘The adjacent and, in a broad sense, separate lines of growth in which

the patent and latent elements are situated, diverge from a common group

and converge to a common contribution, because they were both evolved out

of elements contained in a structureless ovum, and they, jointly, contribute

the assat which form the structureless ova of their offspring.’ ’—@Galton,

. ‘On Blood-relationship,’’ Proceedings of the Royal Society of London,

Vol. 20, 1872, p. 394.

* Jiger, G.,

‘*Physiologische Briefe,’’ Kosmos, Jahrg. I, Bd. I, 1877,

>. r:

No.554] SHORTER ARTICLES AND DISCUSSION ` 125

genetic, which is reserved until the time of puberty for the formation of

the reproductive material. This reservation of the phylogenetic material

I designated as the Continuity of the Germ-protoplasm.’

This clear expression of the doctrine of germinal continuity

apparently does not appear in Jager’s later work,® to which

reference is usually made.

Weismann in his essay on the ‘‘Continuity of the Germ-

plasm,’’® assumed that he was the first to give expression to this

conception but in a later work’® made acknowledgments to other

authors who had anticipated his theory. “With respect to Jäger,

however, he said :12

The praiseworthy attempt to do justice to my predecessors in this

particular subject has perhaps been carried too far. In Geddes and

Thompson’s “ Evolution of Sex” (p. 93), for instance, a quotation is

given from Jiiger which seems to prove that he anticipated me with

regard to the theory under consideration. The quotatién in which this

idea is expressed is, however, not taken from the book published in

1878 but from an essay written ten years later, and it concludes with

the following words: “ This reservation of the phylogenetic material I

described as the continuity of the germ-plasm.” But no mention is made

by — of the continuity of the gene in his book which ap-

m e früheren Auseinandersetzungen gingen dahin: Das Fundamen

der ace besteht darin, dass durch grosse Reihen von cease

hindurch das Keim Protoplasma eines Thieres eine sich stets gleichbleibende

Tro

Der Continuitat des Keimplasmas,’’ Jena, 1885; Essay IV in author-

ized er 2d ed., 1891, p. 163.

***The Germ-Plasm, A Theory of Heredity,” New York, 1902, pp.

Pay

"L. ¢., p. 200, footnote.

126 THE AMERICAN NATURALIST [Vou. XLVIL

peared in 1878, in which a connection between the germ-cells of

different generations is supposed to exist :—and this is not the ease. The

entirely new statement of his ideas has been influenced by those con-

tained in my essays which had appeared in the meanwhile.

As a matter of fact the quotation from Jäger which Weis-

mann repudiates actually appeared more than eight years before

the publication of Weismann’s essays, as the quotation from

Kosmos, given above, clearly shows. Although Jager coined the

expression ‘‘Continuity of the Germ-plasm,’’ the idea involved

seems to have attracted no attention until after the essays of

Weismann had aroused general scientific interest.

Jäger also assumed a material connection between the germ-

cells of different generations,’* and, in what Weismann char-

acterizes as ‘‘a few casual remarks,’’ Rauber*® expressed a con-

ception which some authors interpret as the same idea.

In the account of his exhaustive researches on the differentia-

tion of the reproductive cells Nussbaum gave clear expression to

the doctrine of the continuity of the germ-cells, which in a strict

sense means that germ-cells arise directly from one another.

The views held by Nussbaum?‘ were in part set forth in the

following words:

* See preceding quotations from Weismann.

* Rauber, A., ‘‘Formbildung und Formstérung in der Entwicklung von

Wirbelthieren, ws pet Jahrbuch, Vol. 6, 1880, p. 4. The ‘‘ casual

marks’’ are as follow

‘‘Die beiden Oa deren Verbindung das neue Wesen bewirkt,

sind bei den höheren Thierformen enthalten in besonderen Organen, den

Keimdriisen. Da aber die Keimdriisen die folgende Generation beherbergen,

so erscheint ein Individuum als der Träger zweier Generationen, seiner

theil der dualistischen Anlage. Die Triiger der Zukiinftigen Generation, die

Keimdriisen, stellen dagegen den Germinaltheil der dualistischen Anlage dar.

‘‘ Personal- und Germinaltheil gehen aber von einem befruchteten Ei

aus, eîn solches Ei enthält den Stoff mit dem Kriifteplan zu der genannten

dualistischen Anlage. Man kann darum << von einem Personaltheil und

—— des hetroskiatos Eies reden

* Nussbaum, M., ‘‘ Zur FOP EOL SiR des Geschlechts im Thierreich,’’

Archiv fiir mikroskopische AN Bd. 18, 1880, p. 112. The text of

the original is worded thus

‘*Es theilt sich sage das gefurchte Ei in das Zellenmaterial des

i len fiir In bei

Indivi und in die Zel die Erhaltung der Art. iden

Theilen geht die Zellenvermehrung continuirlich weiter; nur tritt im Leibe

Individuums die itstheilung hinzu, während in seinen Geschlechts-

zellen sich eine einfache additionelle Theilung vollzieht. Die beiden Zellen-

No. 554] SHORTER ARTICLES AND DISCUSSION 127

The segmented ovum divides into the cell-material of the individual

and into the cells for the preservation of the species. In both divisions

the cell-multiplication proceeds continuously, but in the body of the

individual division of labor occurs, while in the reproductive cells

simple division only takes place. Both groups of cells and their off-

spring are propagated quite independently of each other, so that the

reproductive cells have no share in the development of the tissues of the

individual, and no seminal or ovicular cell arises from the cell-material

of the individual. After the segregation of the reproductive cells the

history of the individual and that of the species are entirely distinct,

and because of this relation the “constancy” of the species is more

easily understood; that is, the sharp persistence of the phenomenon of

atavism by means of which ancestral traits are transmitted. For sperm

and ovum are not derived from the cell-material of the parent organism,

but have a common origin with it. However, since they are preserved

within it, they are subject to the conditions which modify the parent

organism; therefore the transmission of “ acquired” characteristies is

not excluded.

Nussbaum is said by some to be the first to suggest the idea of

the cellular continuity of successive generations, but this con-

ception is clearly implied in Virchow’s aphorism’ ‘‘omnis

cellula a cellula,’’ and was fully stated in 1858 in the Law of

Genetie Cellular Continuity first clearly formulated by Vir-

chow?! as follows:

Just as an animal can originate only from an animal and a plant -

only from a plant, so every cell must arise from a preexisting cell.

Although there are individual cases in which strict proof is still want-

ing, yet the principle is firmly established that for all living beings,

whether they be entire plants or animal organisms or integrant parts of

the same, there exists an eternal law of continuous development.

gruppen und ihre Abkémmlinge vermehren sich aber durchaus unabhängig

von einander, so dass die Geschlechtszellen an dem Aufbau der Gewebe des

Individuums keinen Antheil haben, und aus dem Zellenmaterial des Indi-

viduums keine einzige Samen- oder Eizelle hervorgeht. Nach der Abspaltung

der Geschlechtszellen sind die Conti des Individuums und der Art völlig

getrennt, und wir glauben aus diesem Verhalten die ‘Constanz’ der Art,

d.h. die in der Erscheinung des Atavismus gipfelnde Zähigkeit, mit der sich

die Eigenthümlichkeiten der Vorfahren vererben, begreiflicher zu finden.

Denn Samen und Ei stammen nicht von dem Zellenmaterial des elterlichen

Organismus ab, sondern haben mit ihm gleichen Ursprung; da sie aber in

ihm aufbewahrt werden, so sind sie auch den Bedingungen unterworfen,

welche auf den elterlichen Organismus modificirend einwirken, weshalb die

Vererbung der ‘erworbenen’ Eigenschaften nicht ausgeschlossen ist.’’

* Archiv für Pathologische Anatomie, Bd. 8, 1855, p. 23.

“Die Cellularpathologie im ihrer Begründung auf physiologische und

pathologische Gewebelehre,’’ Berlin, 1858, p. 25.

128 THE AMERICAN NATURALIST [Vou. XLVII

On the other hand, to Nussbaum is sometimes credited the

theory of germinal continuity, but in such cases authors appar-

ently do not sharply distinguish continuity of the germ-plasm

from continuity of the germ-cells. Thus Minot" says:

We owe to Moritz Nussbaum the theory of germinal continuity—the

only theory of heredity which seems tenable at the present time.

Aceording to this theory, the germ-cells are set aside during the seg-

mentation of the ovum and preserve the essentially undifferentiated

qualities of the protoplasm and nucleus of the ovum, from the division

of which they arise.

However, irrespective of the conclusions that may be reached

as to whom priority in the statement of the theory of germinal

continuity belongs, it is to Weismann that credit must be given

for the development of this doctrine into an important theory

of heredity.

There would seem to be a gain in precision and clearness of

expression in discussions involving the idea of continuity in

development if a distinction were always made between (1)

genetic continuity, or biogenesis, (2) genetic cellular contin-

uity, (3) continuity of the germ-cell and (4) germinal con- .

tinuity. Thus restricted the term germinal continuity expresses

more closely the conception held by the greatest exponent of this

theory. Since Jäger first used the phrase ‘‘Continuity of the

-Germ-plasm’”’ I suggest that his name be linked with that of

Weismann in referring to this principle, which may well be

called the Jiger-Weismann Law of Germinal Continuity, the

esential doctrine of which is thus expressed :18

In each ontogeny, a part of the specifie germ-plasm contained in the

parent egg-cell is not used up in the construction of the body of the

offspring, but is reserved unchanged for the formation of the germ-

cells of the following generation.

However, the real significance of Weismann’s theory of ger-

minal continuity and its bearing on theories of heredity can not

be fully appreciated without at least a general acquaintance

with the somewhat voluminous literature of this subject.

W. STOCKBERGER

BUREAU OF PLANT INDUSTRY,

WASHINGTON, DO

" Minot, C. S., ‘‘Laboratory Text Book of Embryology,’’ Philadelphia,

1910, p. 28.

* Weismann, A., ‘‘Essays upon Heredity and Kindred Biological Prob-

lems,’’ authorized translation, 2d sess E Oxford, 1891, p. 170.

SECOND EDITION

AMERICAN MEN OF SCIENCE

A BIOGRAPHICAL DIRECTORY

BDITED BY J. McKEEN CATTELL

A so rig directory re 8 revision if it is to maintain its use usefulness. Nearly a th

ly a third of the names in the

Present edition are new, and the s had which appeared in the first d ivision have in nearly every revised. The

amount of ore required to prepare the revision has been as great as that given to the first edition. There has been no

ge stric ob ing ù

change in the al f the work. Greater iness Ds in confining its scope to the natural and

exact and for this reason a few names included in the first edition have been Efforts have been exerted

make the book as co lete and accurate as ible. are of course omissions se some men

DPN even to repeated requests Jor the in tion e md leading men of sci ve been again selected

by the methods thai were used before, and stars kave been to the subjects of research in the case of 269 new men who

ve obtained pl on the list. the as in selecting this group of scientific men to make a study of

conditions on which scientific research depends and conditios

so far as may be to im these

an Te the two statistical studies that have been te ape it he Dk the Second Edition

i econd edition of the Directory extends to more than 600 pages and contains more than 5500 sketches

is well printed on all rag paper and bound in buckram with epin label. Aeae the work has been

noreased in size by more than 50 per oent., it is sold at the same price as the first editio

Price: Five Dollars, net, Postage paid

THE SCIENCE PRESS

GARRISON, N. Y LANCASTER, PA.

SUB-STATION 84, NEW YORK CITY.

The American Journal of Science

Established by Benjamin Silliman in 1818

The Leading Scientific Journal in the United States

Devoted to the Physical and Natural Sciences, with special reference to

Physics, and Chemistry on the one hand, and to Geology and Mineralogy

on the other.

Editor: EDWARD S. DANA.

Agso ig

Clate Editors: Professor GEORGE L. GOODALE, JOHN TROWBRIDGE, W. G. FARLOW and WM. M. DAVIS on

of Cambridge; Professors A. . VERRILL, HENRY 8. WILLIAMS and L, V. PIRSSON, of New

Two Haven ; wae JOSEPH S$. AIMES, of Baltimore ; MR. J. S. DILLER, of Washington.

ine annually, in monthly numbers of about 80 pages each.

. U8 Journal ended its first series of 50 volumes as a quarterly in 1845; its second

se Dec-

ries of 50 volumes as a two-monthly in 1870; its third series as a monthly ended

em

ber, 1895. A Fourth Series commenced i in 1896.

Subscription d in the United States

$6.25 Price, $6 per year or 50 cents a number, postage prepai e

to Cana nada ; $6,40 to Countries in the Postal Union. Back numbers at reduced prices.

*-Ten-Volume Indexes, Vols. I-X XI-XX, fourth series, price one dollar. Address

The American Journal of Science |

New Haven, Conn.

TA ip Seen igh tease ne ARPE RON Ree RE Te

renee

The American Naturalist

A Monthly Journal, established in

with Special Reference to the

1867, Devoted to the Advancement of the ae Sciences

F.

of Organic Evolution and Her

CONTENTS OF AUGUST NUMB

Hardiness in Successiv eta Generations. L. R.

ea

gogy tt ae oe pag cee SOE in Fresh-water

hes.

The Formation ap} Con n Correlation Tables

when e Seire Number of Combinations is large. Dr.

u

Shorter Articles an wate ras The Production

of Pure Homoz Organisms a ed

erozygotes by “Bel fertilization, Profes

8S. J nnings. and Pg uti Fact tore

Mice not ate rm "C. Little. Physical

Fags pe Reg ol soo Processes, Dr. C. F. Cook.

acy in Scientific Review-

3 ur

Notes and Literature: r E and Origin of

Life in America, T. Barbour,

CONTENTS OF SEPTEMBER NUMBER

Asymmetric Color Resemblance in the Guines Pig.

Professor Joseph H. Kastle and G. D. Buckner.

On ae eanen Mortality with Respect to Seed

Wei ET we ugh tht gobi Cultures of Phaseolus

15. r

wales

A Case of portent in Asplanchna simulating

Mutation. Prcfessor J. H. Pow

Shorter Articles and Discussion : An inat Symbio-

tic Relation between a Waterbug and a Crayfish,

Professor James F. ara Double Eggs. Pro-

fessor Charles W. Hargitt

r merica: n Permian Verte-

brate pact a. O. P. Hay. M Federleys Breeding EX-

periments with the Moth Pygaera. A. H. Stur-

Rae ee ee Re ra D EA ee ae ay ee E RETA

n en EN EA EENEI EE VPE EEE AE E NETE EET R E FE E SEE E ae

CONTENTS OF OCTOBER NUMBER

I. Gametic aoe as a Cause of Correlations.

G. N, Collins

II. Mice; Their Breeding and Dran ngio Scientific

Pu r. J. Frank Daniel.

III. The Aan of Hyla arenicolor Cope, with

Notes on its Habits and Vanation. C. H.

Richardson, Jr.

IV. nee sor Colors in F, if a Cross between Non-

Varieties of Maize. Professor R. A.

eiai

We KETE —— ot Zea mays L. Dr. Walter

VI. Notes and riers + Patten on the Origin of

Vertebrates = y the General Value of Specu-

ations on the Phylogen Beings.

Professor Wm. E. Ritter.

CONTENTS OF NOVEMBER NUMBER

The Mendelian Notation as a Description of

Physiological Facts. Profiles r E. M. East.

bat

.

I. A FirstStudy on the Influence of the Starvation

of the Ascendants aR the Characteristics of

the Descendants.—. J. Arthur Harris.

IH. Relations in Batka ei

tru

W. A. Cannon,

Iv. —— a and Diseussion: On Tricolor

n Dogs and Guinea-pigs. Arend. L.

kA.

V. Notes ma Literature: The Classification of

Liverworts: Professor chp cor

Compeatle "Invertebrates; ee Honen ion

Atwood Kofoid.

CONTENTS OF DECEMBER NUMBER

i a pew ag Inheritance of Fecundity in

estic Fowl. Dr. Raymond P “ry

IL, Reflections ie aeons the Autonomy of Biological Sei-

Ti he! TR brener = theses Lamprey, Petro-

a Hussakof,

< orter sel Daka A Simple Test

of the Good Goodness of Fit of tendelinn tation

V. Notes and Literature: Notes Ichth:

Dr. David Starr Jordan. oe

VI. eN ume XLVI,

CONTENTS OF JANUARY NUMBER

I. Factors and Unit aaay in Mendelian

Heredity. Professor T. H. Morgan

Vertical Distribution of the Chætognat

Sa San Diego ea in Eohfions a ‘guar

oa of Isolati . Coinciden

A Family 4 ‘of Spotted Negroes, I. Simpso®

and Professor W. E. Castle.

The Psi of Fertilizers on Variation in Corn

and Beans, Professor J. K. Sha

pi

yi

bi ë be

R

wrll ha ss

Yearly ae $4.00

Dollar

Jonas

Soie Number 35 Cents

The NATURALIST wi

HE SCIENCE PRESS

— 84: NEW YORK

Garrison, N. Y

four months f os

ae

VOL. XLVII, NO. 555 ~ MARCH, 1913

THE

AMERICAN

NATURALIST

A MONTHLY JOURNAL

Devoted to the Advancement of the Biological Sciences with

Special Reference to the Factors of Evolution

CONTENTS

Page

I. Distribution = Copeman. of e cis Professor VERNON

YMAN KEL i - - =i

II. Castration in Relation to the Secondary Sexual Characters of Brown Leg-

oms. +H. D. Goonies =. ee

II. Simplification of Mendelian Formule. Professor W. E. CASTLE - - -170

IV. Notes and Literature : Some Recent Advances in ET a Pro-

fessor Roy L. MOODIE - 183

THE SCIENCE PRESS

LANCASTER, PA. GARRISON, N. Y.

NEW YORE: SUB-STATION 84

The American Naturalist

MSS. intended for ee and books, etc., intended for review should be

sent to the Editor of THE AMERICAN NATURALIST, Garrison-on-Hudson, New York.

Articles containing research work „bearing on the problems = are evolu-

tion are Senethilly w welcome, and wiil be given preference in publica

One hundred reprints of ae Arba are supplied to authors tres of charge.

Further reprints will be supplied at c

Su scriptions and Saar a should be sent to the publishers. The

subscription price is four dollars a year. Foreign postage is fifty cents and

Canadian postage twenty-five cents additional. The ee for single copies is

forty cents. The advertising rates are Four Dollars for a pa

THE SCIENCE PRESS

Lancaster, Pa. Garrison, N. Y.

NEW YORK: Sub-Station 84

Entered as second-class matter, we 2, 1908, at the Post Office at Lancaster, Pa., under the Act of

ongress of March 3, 1879.

FOR SALE TENTH EDITION.

ARCTIC, ICELAND and GREENLAND | THE MICROSCOPE;

BIRDS’ SKINS, an introductio n to Microscopic Methods and to Histo(eey

Well ee Low Prices +A Siaa HENRY GAGE, of Cornell Uni vas

articulars of and aboye rg ain Apii:

i ition. Price

Ó DINESEN Kra Cclcces and fally orined edition d ee:

Husavik, North Iceland, Via Leidie, England | COMSTOCK PUBLISHING CO., Ithaca, N. Y.

E

e °

To Ornithologists FINE LAND SHELLS

I have in stock and for sale nearly ten thousand species

and Mus = land shells, also > yery good stock of marine. Anea

eums rately labeled. Iam prepared to make up collec $ all

Tustrating or phase = the sip es +o Soe aon i

countries and islands. ne illustra

W. F.H. ROSENBERG angaa E shells p pe olas ae an ta hae v

se cco ased

Importer of Exotic Zoological Specimens cash. Correspondence solicit ted with travelers and Fs WEBB,

Only the finest m al handled W. TE 4

57, HAVERSTOCK HILL 202 Westminister Fonts Rochester, N. Y., U.S. A

London N. W., ENGLAND

Begs to announce the publication of a new Price List

(No. 17) of BIRD SKINS. This catalogue con-| | Lhe smerny of Chicago

tains nearly 6,000 species, and is the largest and ca ee

most complete price list of birds ever published. It one ie year. eee

is arranged in systematic order with authors’ names, bhe reap

indications of localities, and an index to Families.

It will be sent gratis and post free on application, as

will the following lists : h

University staff which is et al

No. 14, Mammals; No. 15, Bird’s Eggs; No. 16, e T iomractars: fr

Reptiles, Amphibia and Fishes ; No. 12, Lepidoptera .

No. 13, Coleoptera.

Largest stock in the world of specimens in all

branches of Zoology. Specimens gent on approval.

Please state which lists are required.

Detail

sent upon application.

The University of Chicago

Chicago, Illinois

THE

AMERICAN NATURALIST

VoL. XLVII March, 1913 No. 555

DISTRIBUTION AND SPECIES-FORMING OF

ECTO-PARASITES!

PROFESSOR VERNON LYMAN KELLOGG

STANFORD UNIVERSITY, CALIFORNIA

; I

Four years ago I completed, and published in Wyts-

man’s ‘‘Genera Insectorum,” a catalogue of the Mallo-

phaga (biting bird lice) of the world. This list gives the

host distribution of each of the known species and the

place or country of capture of the actual host individuals

for all cases in which these localities had been recorded.

It is, then, a classified list of the known Mallophagan spe-

cies with their known geographic and host distribution.

I have now in hand, nearly completed, as complement

to this already published catalogue, a classified list of all

the hosts, bird and mammal, from which Mallophaga

have been taken, together with the names of all the para-

Sites recorded from each species, and not only again the

actual geographical records of capture but also a state-

ment of the general geographical distribution of each

host Species. I have determined, for the birds at least,

which form the great bulk of the hosts, the synonymy of

the various names used for them by the various collectors

and reduced them all to the basis of the British Museum

Catalogue of the Birds of the World.

By these two lists, namely, the catalogue of parasite

Anil of this paper were read before a general meeting at the Second

tonal Congress of Entomology, Oxford, August, 1912.

129 .

130 THE AMERICAN NATURALIST [Vou. XLVII

species with hosts, and the catalogue of host species with

parasites, this group of ectoparasitic insects should be

accessible to a wider attention and a more thoroughgoing

study from entomologists than it has hitherto had. In-

deed the group is deserving of some special attention

from general naturalists because of the interesting sig-

nificance of the conditions of distribution and species

forming which obtain in it. It is to the setting out of a

few facts and some of the significance concerning distri-

bution and species forming among these insects that the

present paper is devoted.

II

The Mallophaga compose a small fairly homogeneous

group of about fifteen hundred insect species (as so far

known), which pass their whole lives on the bodies of birds

and mammals. Those species found on birds are never

found on mammals, and those on mammals never occur

on birds. Indeed, with exceptions in only two genera,

the Mallophaga of birds are of distinct families from

those of mammals, the two groups being distinguished

structurally by the loss in the mammal-infesting kinds

of one of the two tarsal claws, an adaptation connected

with the difference between the feathers and hair of the

hosts as habitat. Of the 1,500 known Mallophagan spe-

cies less than one hundred are mammal-infesting, the

others all occurring exclusively on birds.

As to the relations of the Mallophaga to other insects

I am convinced that they should be treated as a distinct

order, finding their nearest affinities in the wingless

Psocide, such as Atropos, the familiar book louse. This

relationship is shown not only by similarities in external

structure, some of which, however, are only examples of

parallelism or adaptive convergence, but by certain more

important common characters of internal anatomy, not-

ably a curious pharyngeal sclerite (perhaps the greatly

modified hypopharynx), found in both these groups and

nowhere else among insects. The habitat and food

No. 555] SPECIES-FORMING OF ECTO-PARASITES 131

habits, also, of the two groups add to the likelihood of

the derivation of the Mallophaga if not directly from the

Atropidæ, at least from an ancestor that might well have

been the conmon progenitor of both types.

The Mallophaga are all of small size, ranging from 1.5

to 10 mm. in length, with the great majority of them not

more than 4 or 5 mm. long and from 1 to 2 mm. in width.

The body is much flattened dorso-ventrally, offering an

interesting adaptive contrast, in this respect, to the later-

ally flattened fleas of similar habitat. The body is always

wholly wingless, usually strongly chitinized and smooth.

The mouth parts are of true biting type, the food being

bits of hair or feather or dermal scales, and dried or fresh

blood only incidentally and not by a direct breaking of the

skin of the host. There are no compound eyes, but a

single pair of ocelli placed on the lateral margins of the

much flattened head, and the antennæ are from three- to

five-segmented, and in about half the genera are capable

of being concealed in a small protecting groove on the

ventral face of the cheeks. The legs are of normal num-

ber, rather long and strong, flattened and fitted for run-

mng and clinging. The body is whitish, pale brownish

or even dark-brown, often well patterned by blotches or

bands of light to black-brown, on the paler ground.

These markings almost always indicate specially heavily

chitinized portions of the body wall.

The eggs are fastened to the hairs or feathers of the

host, and the hatching young are much like their parents

m appearance except for their smaller size and paler,

unpatterned softer skin. They acquire maturity without

any marked metamorphosis. They run freely about

from birth, and feed on the hairs or feathers just as their

parents do. Neither they nor the adults leave the body

of the host except under unusual circumstances. In hen-

houses the parasites of the fowls have been occasionally

found on the perches or in the nests, but I have often

Sought carefully, without finding a single parasite, on

Seaside rocks from which I had just frightened scores of

132 THE AMERICAN NATURALIST (Vou. XLVII

closely massed resting cormorant, pelicans and gulls, all

of them bird kinds whose individuals are practically

always well parasitized. When a bird dies or is killed

some of its Mallophagous parasites will sometimes wan-

der off of the cold body, but usually they remain on the

body and die there within a few days. Before dying they

frequently grasp a feather barbule with their strong

mandibles and are found there firmly fixed after death.

That they can not live, except unusually, for more than

a few days off the body of the warm-blooded host I have

proved by keeping them on feathers in ovens at the tem-

perature of the bird body or on feathers in vials next my

own body. They die in from two or three to six or seven

days off the host body, or on it after death of the bird. It

may be that the normal life of the adult Mallophagan is

but a week long, but the fact that the young die off the

host, or on it after its death, just as soon as the adults,

indicates that the parasites are simply unable to live

apart from their live hosts, for the only two apparent

conditions peculiar to life on a bird’s body, namely, a cer-

tain constant temperature and feathers for food, are cer-

tainly not difficult to reproduce off the body.

At times of actual contact of the bodies of their hosts

they undoubtedly migrate from bird to bird. Thus they

can move from male to female, or vice versa at mating,

from mother and father to fledglings in the nest, andfrom

individual to individual of the same species in the case

of gregarious birds perching together.

The Mallophaga are, then, true, permanent ecto-par-

asites, of simple development, some degradation and

adaptive modification of body, which occur in the num-

ber of one to several species on probably all kinds of

birds and mammals, and on practically all bird and mam-

mal individuals, but which, except in special and rather

uncommon cases of unusual abundance, cause little seri-

ous injury to their hosts. The familiar constant search-

ing of the plumage with the bill by birds is almost always

due to the slightly irritating influence of these small para-

No. 555] SPECIES-FORMING OF ECTO-PARASITES 133

sitic insects, but only rarely can this annoyance reach

such a pitch as to interfere much with the bird’s feed-

ing or resting or sleeping. And the loss of minute bits

of feather can certainly be of no consequence at all. The

Mallophaga then may be called benignant parasites, as

contrasted with the malevolent, blood-sucking fleas and

true lice affecting the same hosts.

Tit

With this fleeting acquaintanceship with the various

principal structural and physiological characters of our

group of insects, we may attend now to some of the spe-

cial facts of their distribution and their inter-ordinal re-

lationships, and to the problems which these facts pose

o us.

First, with regard to the taxonomic conditions within

the group. I have divided the order into two sub-orders,

Sharply distinguished by certain structural differences

whose physiological or ethologie significance, however,

is not at all plainly apparent. The most convenient rec-

ognition character for this sub-ordinal separation is the

condition of the antenne, short, broad, capitate, three-

Segmented and concealed in special antennary fosse in

one group; longer, slender, five-segmented, projecting

and without receiving fosse, in the other.

Each suborder contains two families, one of which, with

two-clawed members, occurs exclusively (with the ex-

ception of one or two species in each of four genera in

one case) on birds, and the other, with one-clawed mem-

bers, exclusively on mammals. Both of the mammal-

infesting families include but one genus each; while the

two bird-infesting families include fifteen and ten genera,

respectively,

The whole order then, with its one thousand five hun-

dred species, comprises but twenty-seven genera,

Srouped into four families, arranged in two suborders.

The disposition of the species among the genera has a

rather extraordinary aspect. Nine, or one third, of all

134 THE AMERICAN NATURALIST [Vou. XLVII

the genera, are represented by but one species each;

seven are represented by from two to ten species; three

are represented by from twenty to thirty species, three by

from forty to sixty-five species, one by one hundred and

fifty species, one by two hundred, and three by from two

hundred and twenty-five to two hundred and fifty species.

Thus while sixteen genera contain less than ten species

each, four genera contain more than two hundred species

each.

This extraordinary condition of the species in their re-

lation to each other presents an attractive problem.

What is its significance? What are indeed the special

conditions influencing species-forming within the group?

From many years’ work with these insects, including

the description of several hundred new species, and the

examination of long series of individuals of species in

several different genera, I can say with confidence that

the evolutionary factor of isolation plays a conspicuous

part in Mallophagan species-forming. One soon comes

to the acceptance of a very flexible species description for

any given Mallophagan kind. While the score of indi-

viduals of one kind that one may collect from a single

host individual will agree well with each other as to

details of structure and pattern, the specimens of the

same kind from another host individual of the same host

species collected either in the same locality or a distant

one, and the specimens from a third host individual and

from a fourth and fifth, and so on, will all show many

obvious, if mostly small, variations from the specimens

taken from the first or any other host individual.

That is, each host individual is, in a way, a small island,

biologically considered, with its inhabitants more or less

nearly completely isolated from the inhabitants of other

host islands. So that each species is made up of many

dislocated small groups which may have, as when they

are on birds of solitary habit, but little opportunity for

mixing and cross-breeding with the members of the spe-

cies-body as a whole. The group on one host bird may

No. 555] SPECIES-FORMING OF ECTO-PARASITES 135

meet the group on the mate of this host, and these two

mingling groups may send their repr tatives or their

offspring to the young of the two mated hosts. But that

is about the extent of their participation in the life and

character of their species as a whole, and it is an extent

which plainly must result in the establishment of an

hereditary strain characterized by the special slight

structural idiosyncrasies peculiar to the few ancestors

from which the strain takes its origin. In the case of the

parasites of more gregarious bird kinds, as the seabirds

that mass for rest or brooding on ocean rocks or shore

cliffs, or the swallows and swifts that live in colonies in

caves and chimneys, or those gallinaceous birds like the

partridges of California that gather in close bands of

two or three score individuals, or others of any kind of

similar habit which may give chance for repeated actual

personal contact of body with body sufficient to permit of

migration of the wingless but active parasites from bird

to bird, this element of isolation is less accented. But

it still plays an important figure. For both the wingless-

ness and the manifest stay-at-home habits of the para-

sites make their movement from host to host at best a

desultory and almost accidental one.

This combination of conditions, then, may serve to ex-

plain partly both why each species must be given a very

flexible description and why one might describe and name,

if he liked, many varieties of each species; and it ex-

Plains, in some measure, why there are a good many

Species in the order, and why there are many in each of

a few genera, although it does not explain, perhaps, why

there are some genera with very few, and indeed even, so

far as yet known, single species.

The explanation of the actually small number of genera

and families depends, I think, upon one of the conditions

in the life of the Mallophaga which is directly opposed,

In its influence, in a way, to the isolation condition making

for a variation that results in numerous varieties and

Species. It is this. Although the different host species

136 THE AMERICAN NATURALIST [Vou XLVII

may differ much among themselves as to habitat, habits,

plumage markings, ete., yet as places of residence and

providers of food for their external parasites they are all

much alike. The temperature is the same, the feathers

as food are about the same. Although the parasite’s

host may live in the water, the parasite itself, safely

tucked away next the skin or among the feathers, lives on

dry land in free air, for the water, even where it con-

tinually covers part of the plumage, as in swimmers, or

occasionally all of the plumage, as in divers, only touches

the plumage surface. Beneath this surface it is always

dry and there is always free air.

Thus despite an isolated life for the inhabitants of each

host island, and the great variety of these islands as re-

gards name and relation to phyletic mainlands, the actual

life conditions are monotonously alike on all these

islands. So that there is, for the Mallophaga, no such

variety of conditions of habitat and food and food-getting

and mate-seeking and egg-concealing and young-rearing

as would tend sharply to select and promote variations,

with a result of genus and family making. There is no

external influence at work promoting wide divergence.

The generic and family distinctions tend to be few; the

varietal and specific tend to be many.

IV

- As a direct outcome of these conditions of life of the

Mallophaga there arises an extremely interesting state

of affairs concerned with their host and geographic dis-

tribution, a state of affairs which reveals, I think, a prin-

ciple or fundamental consideration concerning the dis-

tribution of wingless ecto-parasites in general. This

special subject may be introduced by a swift résumé

of our present knowledge of the facts of the distribution

of the Mallophaga. In this résumé I include some par-

ticular illustrations, by examples, of certain special dis-

tributional ondion,

As Mallophaga have been taken so far from but a

No. 555] SPECIES-FORMING OF ECTO-PARASITES 137

hundred species of mammals, representing 48 genera, 24

families and 5 orders, any special scrutiny of the con-

ditions of their distribution among mammalian hosts

would hardly be worth while. But such serutiny can cer-

tainly now be advisedly undertaken as regards the dis-

tribution among birds. For the Mallophagan host list

includes already more than 1,100 bird species, represent-

ing 33 of the recognized 35 orders of living birds. The

known living bird species number, according to the Brit-

ish Museum Catalogue, about 18,500. This catalogue, I

should note, elevates to the position of species, or at least

to the seeming of species, by cataloguing them binomi-

ally, the so-called varieties or trinomially dubbed sub-

Species of the continental and North American ornithol-

ogists; and I have followed this custom in my list—al-

though against my belief in its taxonomic implication—

for the sake of having a common and universally access-

ible basis for the host names.

Thus one out of every seventeen known living bird spe-

cies is now included in the Mallophagan host list, as are

625 out of the 2,700 recognized living bird genera, and ~

120 out of the 160 living bird families. As comparatively

few bird kinds are still unknown, and as on the other

hand only a good beginning has been made in finding

and describing the Mallophagan kinds, it is certain that

the list of hosts of these parasites will increase rapidly in

proportion to the total number of bird species. From the

proportion of the number of different bird hosts par-

asitized by each Mallophagan species and the propor-

tion of bird families and genera already in the host

list, I estimate, roughly, the total number of living Mallo-

phagan species to be about 5,000.

From the three Acarinate or Ratitian bird orders,

namely, the Rheiformes, or South American rheas, the

Casuariiformes or Australian cassowaries, and the

Struthioniformes or African ostriches, only five species

of Mallophaga have so far been recorded. On the rheas

Occur three species of Lipeurus, one being found on each

138 THE AMERICAN NATURALIST [Vou. XLVII

of two host species and the other two on a third. On one

species of Australian cassowary are found two Mallo-

phagan kinds, one of which is the same species as that

found on two of the South American rheas, while from

the African ostrich, Struthio camelus, are recorded two

parasite species, one of which is the same as that found

on the third rhea. Here, at the very outset, is a remark-

able case of distribution. Identical parasitic species on

hosts as widely separated, geographically, as Australia

and South America and Africa, but hosts all of a certain

degree of genealogic affinity.

The order Tinamiformes, the tinamous of South Amer-

ica, curious birds, rather pheasant-like but presumably

not really pheasants nor true Galliformes of any kind, is

represented by eleven species in the Mallophagan host

list. Most of these tinamous are well parasitized, a spe-

cies of Nothura having four parasite species, one of

Crypturus five, one of Tinamus six, one of Rhynchotus

eight and another of Tinamus even nine parasite species

representing five genera, of which two are peculiar to the

` group. Of the other Mallophagan genera found on the

tinamous two that specially characterize the pheasants

and other gallinaceous birds are, by odds, the most com-

monly represented. And this condition suggests another

interesting problem. Is it going to be possible to get

suggestions regarding the phyletic affinities of hosts from

the character of their parasitic fauna? Take, for ex-

ample, an order of birds troublesome to the ornithological

taxonomists. Will the evidence of the presence on mem-

bers of this order of certain parasitic genera character-

istic of another order indicate their affinities to this

second order? It does indeed seem, in the case of the

Tinamiformes and Galliformes, as if the evidence from

the Mallophagan distribution was in conformity with that

suggested by certain structural similarities in the two

groups.

The great order Galliformes, including the pheasants,

partridges, quail, etc., is represented in the host list by

No. 555] SPECIES-FORMING OF ECTO-PARASITES 139

seventy-eight species, from which are recorded about 150

Mallophagan species representing six genera, two of

which, Goniodes and Goniocotes, are the most abundantly

represented and occur much more commonly on birds of

this order than those of any other unless it be the Tinami-

formes, just spoken of. The other Mallophagan spe-

cies recorded from the Galliformes belong to the large

genera Lipeurus, Colpocephalum and Menopon which in-

clude species from most bird orders. The Mallophagan

genus Docophorus, the second largest of all in the matter

of number of species, and abundantly represented on al-

most all other bird groups, is totally unrepresented on

the Galliformes. The Gallinaceous birds are, as a rule,

strongly parasitized both as regards number of Mallo-

phagan species and number of individuals. One of the

brush-turkeys, Megapodius, has ten parasitic species, the

painted Chinese pheasant has nine, and the Texan quail,

eight. The domestic fowl has twelve Mallophagan spe-

cies and its reputed ancestor, the wild Indian jungle fowl,

Gallus bankiva, four, all of which occur on its domesti-

cated descendant. The Mallophagan species Lipeurus

variabilis is common to nine different hosts of the family

Phasianide.

The small order of so-called pigeon grouse, the Ptero-

clidiformes, has two species in the host list, each parasi-

tized by the single Mallophagan species, Nirmus alchate,

not found on other birds. The two hosts species have

overlapping geographic ranges.

The Columbiformes, or doves and pigeons, are repre-

sented in the host list by 40 species. The Mallophagan

genera Goniodes and Goniocotes, so common also on the

Pheasants and tinamous, are very well represented

among the pigeon parasites. The single Mallophagan

Species Lipeurus baculus is recorded from nineteen of

the forty pigeon host species, whose geographic distri-

bution includes Europe, Asia, Africa, North America,

Malaysia, Australia, Madagascar and the Galapagos

Islands. The European-Asiatic rock dove, Columba

140 THE AMERICAN NATURALIST [Vou. XLVII

livia, immediate ancestor of the domestic pigeon, has two

Mallophagan parasites of which one, the wide-spread

Lipeurus baculus, is found on the domestic pigeon. The

other species, Goniocotes compar, is common to several

other wild doves, but, curiously enough, it has not been

recorded from the domestic pigeon. The isolated Gala-

pagos Island dove, Nesopelia galapagoensis, peculiar,

both in genus and species, to these islands, is parasitized

by Lipeurus baculus, and by four other Mallophagan

species not found on any other pigeons.

From the hoazin, strange aberrant bird of the Amazon

forests, and single representative of the order Opistho-

comiformes, I have recorded three Mallophagan para-

sites, two of them new species, and one a member of the

genus Goniocotes, a genus, rather cl teristic of the

pheasants and pigeons. It is exactly to the pheasant-

like birds that ornithologists seem at present inclined to

associate this lonely South American bird.

The Ralliformes, or rails, gallinules and coots, are

represented in the Mallophagan host list by twenty-three

species. One small genus of parasites, Oncophorus, 1s

almost limited to the order. The old world coot, or

mudhen, Fulica atra, has seven Mallophagan species rep-

resenting six genera. Its congeneric sister species of the

new world, Fulica americana, has twelve Mallophagan

species, of which five are identical with those found on

the old world coot. The parasite species Oncophorus

bisetosus occurs on six different rails, three of them

North and Central American and three of them Malay-

sian and Australian.

The Podicipedidiformes, or grebes, are represented in

the host list by six out of the 25 known species of the

order, from which are recorded eight Mallophagan kinds.

On five of the six grebe species occurs the Mallophagan

species Menopon tridens, found elsewhere also on certain

loons, auks and ducks. The six grebes are geographic-

ally distributed as follows: two new world, three old

world, and one cireumpolar.

No. 555] SPECIES-FORMING OF ECTO-PARASITES 141

Seven species of Mallophaga have been taken from

four species out of the known five of the order Colymbi-

formes or loons, one of which is limited to the old world

with three circumpolar in range. On three of these loons

occurs the Mallophagan species Docophorus colymbinus,

and Nirmus frontalis is common to two, and Menopon tri-

dens to two. But the continuity of geographical range

among the loons does not seem to have produced any

special effect of commonness of parasites to different

host species. In the preceding order, for example, that

of the grebes, there is more commonness of Mallophagan

species, although at the same time more isolation of the

hosts geographically.

From two penguins representing the order Sphenisci-

formes, three. Mallophagan species have been recorded.

Two of these belong to the genus Goniodes, a genus best

represented among the pheasants and pigeons. The

third is type-species of a genus so far not elsewhere

recorded.

The Procellariiformes, or petrels and albatrosses, of

which about a hundred living species are known, are rep-

resented in the host list by thirty-two species, and give

evidence of being a strongly parasitized group of birds.

Ten Mallophagan kinds have been taken from one species

of Puffinus, nine from another, eight from another and

Seven from a fourth. Besides these, four other species

of Puffinus are in the host list. On the four Puffinus

Species most infested there is one parasite kind common

to all, and four parasite species common to three of

them. Six species of albatrosses, genus Diomedea, are

included among the Procellariiform hosts. On five of

them occurs the giant Mallophagan species Lipeurus

f eros, 10 mm. long, and on five also the large, broad

Species Hurymetopus taurus. Nine species of Mallo-

Phaga have been recorded from the single albatross spe-

cies Diomedea albatrus, of the North Pacifice Ocean.

F our Mallophagan genera, each of them containing but

a single species, are peculiar to the order. : The birds of

142 THE AMERICAN NATURALIST [ Vou. XLVII

this order range the great oceans in overlapping zones

and reaches.

The order Alciformes, including 29 known living spe-

cies of auks, murres and puffins, is represented in the

host list by sixteen species which are parasitized by six-

teen Mallophagan species belonging to but two genera,

Docophorus and Nirmus, with the exception of one spe-

cies of Menopon. These two genera are, however, not

at all limited to the Alciformes, but are two of the largest

and most widely distributed of the Mallophagan genera.

Nirmus citrinus occurs on four Alciform hosts, and Nir-

mus maritimus, Docophorus celedoxus and Docophorus

montereyi on three each. Nirmus pacificus and Doco-

phorus atricolor occur on two each.

The homogeneous order Lariformes, or gulls and

terns, including 122 known living species, is represented

by fifty species in the host list, of which two dozen belong

to the gull genus Larus, and one dozen to the tern genus

Sterna. Gulls and terns are strongly parasitized. Thir-

teen species of Mallophaga, representing four genera,

have been recorded from Sterna fuliginosa, and ten spe-

cies from the tropical noddy, Anous stolidus. The gull

and tern parasites are mostly of the genus Docophorus

and Nirmus with some Menopon and Colpocephalum and

a few Lipeuri. Docophorus lari occurs on nineteen spe-

cies of Larus, and Nirmus bilineolatus on eleven. Doco-

phorus melanocephalus occurs on four species of Sterna.

Many of the members of this bird order range widely,

but some are limited to new world or old world shores.

The large order of waders and shore-birds, the Chara-

driiformes, is represented in the host-list by sixty-three

species. The Mallophagan genus Nirmus is especially

commonly met with on these birds, and has many species

characteristic of them. .From the cosmopolitan sander-

ling, Arenaria interpres, with its individuals from old

and new world meeting in high latitude breeding

grounds, fourteen Mallophagan species have been re-

corded, of which six are Nirmi. Ten Mallophagan kinds

No. 555] SPECIES-FORMING OF ECTO-PARASITES 143

occur on the European curlew, Numenius arquata, of

which one occurs also on the new world curlew, Numen-

ius longirostris. It is the only Mallophagan so far re-

corded from this host. Hematopus galapagoensis, limited

to the Galapagos Islands, has three Mallophagan species,

of which one is peculiar to it, one is a duck-infesting spe-

cies, probably a normal straggler under conditions which

I shall explain later, and one is a form found also on

Hematopus ostralegus, the common oyster-catcher of

Europe, Central Asia and Africa. The old world avocet,

Recurvirostra avocetta, has six Mallophagan species,

while the new world avocet, Recurvirostra americana, has

four, of which two, Nirmus pileus and Nirmus signatus,

are common to both hosts. The other two are new. Two

species of the curious aberrant Charadriiform family

Parride occur in the host-list, each having but a single

Mallophagan species, and that the same for both hosts.

One of the host species is limited to Australia, while the

other ranges from India to the Malay Peninsula.

The Gruiformes, or cranes, thirty-four living species,

are represented in the host-list by twelve species, para-

Sitized by twenty Mallophagan kinds. The herons and

egrets, order Ardeiformes, are represented by forty-five

Species. Lipeurus leucopygus occurs on both the old

world bittern, Botaurus stellaris, and the new world one,

Botaurus lentiginosus. It occurs also on two other

herons, both old world species. From Butorides sunde-

valli, peculiar to the Galapagos Islands, I have had four

Species, all previously described by me from various mar-

itime birds of the Pacific. This is a case of straggling,

but as I shall point out later in connection with the condi-

tions shown by certain other Galapagos Island hosts,

a case of what may be called normal straggling, unusual

on the whole, but possible and especially common in the

case of Galapagos, and perhaps other, island hosts.

The Palamedeiformes, or South American screamers,

are represented in the host-list by two species, parasi-

_ tized by three Mallophagan kinds. One of these is com-

144 THE AMERICAN NATURALIST [ Vou. XLVII

mon to both hosts. The flamingoes, constituting the

order Pheenicopteriformes, are parasitized by four spe-

cies of Mallophaga of four different genera.

The Anseriformes, swans, geese and ducks, are repre-

sented in the host-list by sixty-four species, seven being

- swans, nine geese and the rest ducks. The swans have

a Mallophagan genus, Ornithobius, peculiar to them,

which occurs on four out of the seven species. The spe-

cies bucephalus of this genus occurs on two old world

and one new world species. Six of the seven swan kinds

belong to the genus Cygnus. Three of these are old

world, two new world and one circum-polar in range.

Trinoton conspurcatum has been recorded from the three

old world and the single cireumpolar species, and Doco-

phorus cygni from two old world and one new world spe-

cies. Cygnus cygnus of Europe has six Mallophagan spe-

cies. The curious Australian swan Chenopsis atrata has

two parasite species neither of which occurs on any other

swan. Among the geese are three species of Anser, two

of them old world and one new world, with the parasite

Lipeurus jejunus common to them all, and four species

of Branta, one of which also carries Lipeurus jejunus.

This Mallophagan species also occurs on the domestic

goose. The forty-four species of ducks of the list are

parasitized by forty-two species of Mallophaga. Of these,

Docophorus icterodes is recorded from eleven duck spe-

cies, Lipeurus squalidus from fifteen and Trinoton luri-

dum from nineteen, these duck kinds including African,

Asiatic, European, North American, South American and

cosmopolitan species. A duck kind from Australia and

Malaysia has four parasitic species, all peculiar to it.

Three species of ducks have the three most familiar duck-

infesting Mallophaga, mentioned above, common to them

all. Anas boschas, the ancestor of the domestic duck, has

these three and just one more. But so far only one of

the three has been recorded from the domestic duck.

While many species of ducks, and most individuals of the

species, are parasitisized, it is rare that more than two

No. 555] SPECIES-FORMING OF ECTO-PARASITES 145

or three or four—the maximum is actually six—Mallo-

phagan species are found on a single host species. The

great bulk of the parasitization comes from compara-

tively few Mallophagan species, notably the three spe-

cies already named.

The Pelecaniformes, 75 living species, including the

pelicans, cormorants, boobies, man-o’-war and tropic

birds, are represented in the host list by thirty-three spe-

cies. Thirteen of these are cormorants of the genus

Phalacrocorax, and nine are boobies of the genus Sula.

Lipeurus setosus occurs on two African and two Aus-

tralian and Malaysian species of Phalacrocorax. Lipeu-

rus toxoceros occurs on one cosmopolitan cormorant and

on two others, one of South and Central America and

one of North America. Lipeurus faralloni is recorded

from three North American west coast species. Five

Species of Pelecanus, three new world and two old world,

are included in the list. Menopon titan occurs on two

of the new world and one of the old world species, Lipeu-

rus bifasciatus on one new world and both of the old

world forms, and Lipeurus forficulatus on two new world

and one old world species. Eleven species of Mallophaga

are recorded from the cosmopolitan man-o’-war bird,

Fregata aquila, and eight from the beautiful white tropic

bird, Phaeton ethereus.

The Cathartidiformes, or new world vultures, are rep-

resented in the host list by four species. There are but

nine species in the order. The four species in my list

are the great condor of the Andes, ranging from Pata-

gonia to Keuador; the great king vulture of the northern

Andes, Central America and Mexico; the rare Cali-

fornian condor of northern Mexico, Baja California, and

California north to its middle region; and finally, the

ubiquitous turkey vulture, smaller and far more abun-

dant than any of the others, that ranges over all of North

America and, in winter, gets into northern South Amer-

ica. Thus the ranges of the four species combined ex-

tend the whole length of the western coast of the new

146 THE AMERICAN NATURALIST [ Von. XLVII

world. And although each bird has its own stretch of

coast mountains, the range of each overlaps that of some

other. The individuals of all these bird species, except

the last named, are few and solitary in habit, resting and

nesting in inaccessible mountain places, but meeting a

few of their kind occasionally at common table around

some dead or dying animal. Turning now to the para-

sites of these lonely birds, we find one species, the well-

marked, rather large Lipeurus assessor common to all

four vulture species. But assessor has been taken from

the wide-spread turkey vulture only in Panama, îi. e.,

within the range of the king vulture. Laemobothrium

delogramma is also found on both the king and the turkey

vultures, but has been taken from the latter host again

only from Panama specimens. The king vulture and

Californian condor, whose ranges overlap in Mexico, have

one parasite species, Menopon fasciatum, common to

both. These are the only cases of commonness of Mallo-

phagan species to two or more of these great vulture

kinds. And all are pretty well parasitized, seven Mallo-

phagan species being recorded from the king vulture, five

from the South American condor, five from the North

American turkey vulture, and two from the Californian

condor. It is well to keep in mind, in noting this rather

abundant parasitization, that the feeding habits of the

birds give some opportunity for the straggling of para-

sites from other bird or mammal kinds, serving, in the

persons of moribund or just dead individuals, as prey-

It is therefore indeed important to note that no mammal-

infesting Mallophaga have been taken from any vulture,

despite the excellent chances for such straggling. Per-

haps the difference between a mammal and a bird host is

too great to permit a parasite adaptively specialized for

life on one to persist successfully on the other. Or per-

haps there is a physiological antipathy, a negative chemo-

tropism, too strong to permit the straggling. Yet I re-

call that in my days as a bird-collector and maker of

skins, I repeatedly had the annoyance of discovering

No. 555] SPECIES-FORMING OF ECTO-PARASITES 147

that I was a temporary host for individuals of parasites

more normal to a duck or a barn owl than to man! But

these wanderers seemed as anxious to leave their chance-

found new host as was the host to be relieved of them; a

few moments was the usual extent of their stay.

This point of the reluctance of Mallophaga to migrate,

even with good opportunity, from normal or characteris-

tic host, to another, is emphasized when we come to ex-

amine the parasitic conditions of the next bird order, the

Accipitriformes, or faleons, hawks and eagles, almost all

species of which capture living mammals of one kind or

another. In the parasitic records of the seventy species

of this order included in my host list, there is not a single

record of a Mallophagan species of either of the strictly

mammal-infesting genera, and there are but three or four

records of bird-infesting species that are plainly strag-

glers from prey, as is, for example, a typical duck para-

site recorded from an American hawk, and a pigeon para-

site from a European falcon. In fact, the Mallophagan

Species taken from the birds of prey are about as charac-

teristic of their host-group as are those of any other

Sroup, although there are, indeed, no parasitic genera

Wholly peculiar to the birds of prey. The Mallophagan

Species Colpocephalum flavescens is found on twenty-one

Accipitrine species and Nirmus fuscus on eighteen, the

hosts representing species from all parts of the world,

Including, for Nirmus fuscus, at least, Australia. They

represent, too, most of the principal families and sub-

families of the order. Colpocephalum flavescens is found

on Thrysaëtos, a new world eagle genus, on Gypaétus, an

old world eagle, and on one cosmopolitan species of Aquila

and one old world Aquila. The eagles, like the great vul-

tures, are characteristically solitary birds, only the mem-

bers of each household, that is, male, female and young,

coming into contact with each other. They are typical

host islands, Some of the birds of prey are strongly

parasitized, as the golden eagle, Aquila chrysaétus, with

148 THE AMERICAN NATURALIST [Vou. XLVII

nine Mallophagan species, and the South American cara-

cara, Polyborus tharos, with eight.

In the Strigiformes, or owls, represented in our host

list by nineteen host species, we have added confirmation

of the Mallophagan hesitancy to straggle even with good

opportunity. There is no record of a mammal parasite

on any owl, nor but two or three of Mallophagan species

characteristic of birds of other orders. Of the charac-

teristic owl parasites, Docophorus cursor occurs on all

three species of Asio included in the host list. One of

these Asios is restricted to the Galapagos Islands, one

ranges Mexico and temperate North America, and one

occurs in both old and new worlds. The genus Strix,

barn owls, is represented by two species, one the old

world barn owl and the other the barn owl of temperate

North America. The Mallophagan species Docophorus

rostratus occurs on both of them. Asio accipitrinus, the

cosmopolitan hawk owl, carries seven Mallophagan

species.

The Psittaciformes, parrots and cockatoos, are repre-

sented in the host list by twenty-eight species, infested

usually by only one to two or three parasite kinds, al-

though five have been recorded from a Senegambiaa

Psittacus. It is pleasant to note that the cruel New Zea-

land Keas, which have adopted the extraordinary habit of

alighting on the backs of living sheep and tearing their

flesh, even through to the vitals, have at least three Mal-

lophagan parasite species to make life a little uncomfor-

table for them. One of these species, Lipeurus circum-

fasciatus, is recorded from three other parrots of Aus-

tralia and Malaysia.

The catch-all order Coraciiformes, including the

rollers, kingfishers, hoopoes, mot-mots, poor-wills, swifts

and humming-birds, is represented in the list by forty-

five species, of which six are kingfishers, six are hummers

and five are swifts. Of the five swifts three are of the

new world and two of the old. The Mallophagan genus

Nitzschia is peculiar to the swifts and is found, repre-

No. 555] SPECIES-FORMING OF ECTO-PARASITES 149

sented by four species, on all five in my list. Nitzschia

pulicaris is found on one old world and two new world spe-

cies. The humming-birds are not badly parasitized,

although three Mallophagan species have been recorded

from a single one of these tiny host kinds. They are

especially infested by the Mallophagan genus Physosto-

mum, although species of this genus occur on several

other passerine bird hosts.

The trogons, order Trogones, are represented in the

host list by two species, infected by two Mallophagan

species, both of the genus Nirmus. The Coceyzes, or

cuckoos, represented by twenty-three species, have usu-

ally but one, although sometimes two or three Mallo-

phagan kinds to a host species. And this condition of

slight parasitization is also true of the five species of

toucans and barbets, order Scansores, included in the list.

Docophorus latifrons is recorded both from the common

European cuckoo and, in varietal form, from the common

American cuckoo.

The order Piciformes, the woodpeckers, is represented

in the list by twenty-six species, each carrying but one to

two or three Mallophagan kinds to its discredit. The

woodpecker genus Dendrocopus is represented by six

_ Species of which five belong to the new world and one to

the old world. Docophorus superciliosus, described from

the old world species of Dendrocopus, occurs also on one

of the new world species and also on another old world

woodpecker of different genus. One of my Western

American Mallophaga species occurs on three wood-

pecker kinds in California, one in Baja California and

three in Costa Rica. It seems to be a pervasive parasite

of American West Coast woodpeckers. Of the seven

host species from which it is recorded three belong to the

one genus Melanerpes.

The order Eurylæmiformes, or Malayan broadbills, is

represented in the host list by one species parasitized by

two species of Mallophaga, while the order Menuri-

formes, the beautiful lyre-birds of Australia, three living

150 THE AMERICAN NATURALIST [Vou. XLVII

species, is also represented by one species, parasitized,

however, by six Mallophagan species all peculiar to it.

This brings us to the last bird order of our list, the

great group of Passeriformes, the perching and singing

birds, with its various familiar families of flycatchers,

swallows, wrens, thrushes, titmice, warblers, larks,

finches and sparrows, tanagers, blackbirds, crows and

jays, et al. It contains 5,000 known kinds, but is repre-

sented in our host list by but three hundred and eight

species, divided among more than a score of families.

Practically no Mallophagan species found on members

of this order occur on birds of any other order. Two

Mallophagan kinds have a wide host distribution within

the order. Docophorus communis has a host list of one

hundred and thirty Passeriform species, of which

thirty-eight are members of the family Fringillide,

this being more than half of all the Fringilline birds

from which Mallophaga have so far been recorded.

These one hundred and thirty hosts of Docophorus com-

munis represent most of the families of the Passeri-

formes and, in their geographic distribution, all of the

principal regions of the world. A score of varieties have

been named within the species, and a score more might

be. But this would be to say no more than that there is

a wide variation among the members of the species, and

to attempt to make categories of this variation is really

labor lost. This large variability of Docophorus com-

munis is simply the most conspicuous example within the

order of that condition of persistent variation, due chiefly

to isolation, that I spoke of at the beginning of this paper

as a condition occurring in almost all the Mallophagan

species; a variation fostered by isolation, unrestrained

by cross-breeding, but not specially emphasized by adapt-

ive modification, nor sharply selected for life and death

value. ;

Another Mallophagan species widely spread among

Passeriform hosts is Nirmus vulgatus, a species described

by me several years ago and which I have so far taken

No. 555] SPECIES-FORMING OF ECTO-PARASITES 151

from forty new world host kinds, including several

genera and species peculiar to the Galapagos Islands.

All the specimens from Galapagos Island hosts show a

number of small but obvious distinguishing characters,

and I have given them the varietal name galapagoensis.

This constant distinction would indicate that the Gala-

pagos individuals, though now infesting several different

host kinds, are all descended from a single original intro-

duction of the species; or that there is some external

modifying condition of life on Galapagos Island birds

that would produce a convergence among the descendants

of ancestors representing several introductions, a sup-

position hardly tenable, especially in the light of the pe-

culiar life conditions of the Mallophaga.

The Passeriform family Tyrannide, the new world

flycatchers, is represented in the list by eighteen species

parasitized by two Mallophagan species, of which but

two, one being Docophorus communis, are recorded from

old world Passeriform hosts. Although this family has

a continuous geographical distribution over North, Cen-

tral and South America, there is no unusual commonness

of parasitic distribution in it. No one Mallophagan kind

occurs on more than two flycatcher species, except in the

Galapagos islands, where Nirmus vulgatus var. gala-

pagoensis is found on all three of the flycatcher kinds

occurring there. All the other Mallophagan species on

these Galapagos flycatchers are forms restricted to the

islands, although not to the flycatchers.

The swallows, Hirundinidæ, are represented by ten

Species of which two are old world, five new world, two

both old and new world, and one from the Galapagos

Islands. From them are recorded twenty Mallophagan

Species, of which M enopon rusticum is found on one old

world, one new world, one old and new world, and one

Galapagos Island host. Docophorus excisus is found on

one old world and two new world hosts, and Nirmus

longus on two new world and the two old and new world

hosts. Seven Mallophagan species have been taken from

152 THE AMERICAN NATURALIST [Vou. XLVII

the Galapagos Island swallow, of which one is found on

two new world swallows and one is the Menopon rusticum

already mentioned as common to one old world, one new

world and one old and new world host.

Mallophaga have been taken from three species of wrens,

Troglodytide, of three different genera. Two species

of parasites have been taken from each host species, and

no one of these Mallophagan species occurs on more

than one host kind.

The Cinclide, or dippers, are represented in the list,

by two species of Cinclus, one from the old world and one

. belonging to the new. One Mallophagan species, a Men-

opon, is common to both hosts, and each host species has,

in addition, another Mallophagan species, a Docophorus

in one case and a Nirmus in the other.

The Mimide, or mocking-birds, are represented by

seven species, of which five belong to the genus Nesomi-

mus peculiar to the Galapagos islands. The other two

are North American. There is one Mallophagan species

on each of the American mockers, and from one to as

many as eight on the various Galapagos Island hosts.

But the Mallophaga of the Galapagos Island mockers

are mostly species common to numerous other birds of the

islands, some of these birds being widely separated phy-

letically from the Mimide. For example, the charac-

teristic Lipeurus baculus of the pigeons has been taken

from one of the mockers. This apparently abnormal

condition has a normal explanation that we shall take up

in connection with certain still more conspicuous ex-

amples of the anomalies among the Galapagos records.

The Turdide, or thrushes, are represented in the list by

twenty-three species. Docophorus communis occurs on

nine of them. Menopon thoracicum is recorded from the

old world Turdus viscivorus and from Merula grayi and

Catharus gracilirostris from Costa Rica. The Sylviide,

or old world warblers, are represented by five species, on

four of which Docophorus communis is the only parasite.

The Ampelide, or waxwings, are represented by four

No. 555] SPECIES-FORMING OF ECTO-PARASITES 153

. species, two of which are North American, one Central

American, and one common to both old and new worlds.

Docophorous communis and Nirmus brachythorax oceur

on three of the four species.

Mallophaga have been recorded from eight species of

Laniide, or shrikes, of which three are old world species,

three new world and two are Australian. Docophorus

communis occurs on five of the old and new world spe-

cies, but not on either of the Australian hosts. Four of

the eight species belong to the genus Lanius, three being

new world and the other an old world species. Doco-

phorus communis occurs on them all.

The Paride, or titmice and chickadees, are represented

in the host list by eleven species, four old world and

seven new world. Docophorus pari is recorded from

three old world species of three different genera. Doco-

phorus rutteri occurs on two new world species of the

Same genus and the same geographical region. Mallo-

phaga have been taken from three kinglets, family Regu-

lide, two of them old world species and one new world.

Physostomum frenatum is recorded from one old world

and one new world host species.

The rather large family of Mniotiltide, or wood-war-

blers, is represented by twelve species, three of them

from the Galapagos Islands, and the others from North,

Central and South America. Docophorus communis is the

only old world Mallophagan species occurring in the

group, being found on three of the American host species.

The Drepanide, or honey creepers of the Hawaiian

Islands, are represented in the list by three species of

three different genera. Seven Mallophagan species, all

except one peculiar to the host group, have been taken

from these three hosts. One of the Mallophagan species

1S Common to all three.

Mallophaga have been taken from five species of Alau-

did, or larks, three of them old world, one new world,

and one inhabiting both old and new worlds. Doco-

Phorus communis occurs on four of the lark species.

154 THE AMERICAN NATURALIST [Vou. XLVIL

Only three other Mallophagan kinds have been recorded

from the family.

The great family of sparrows and finches, the Fringil-

lide, including nearly 1,200 bird kinds, is represented in

the host-list by seventy-six species. The Mallophagan

species Docophorus communis has been recorded from

thirty-eight of these from the old and new worlds, and

Nirmus vulgatus from eighteen, all from the new world.

There have been almost no Mallophagan records made

from old world hosts since I described vulgatus, which is

possibly the explanation of the lack of any old world

records for it, although it may really be that the species

does not occur in? Europe. Among the Fringilline hosts

of the Mallophaga there are nine species representing

two genera, Camarhynchus and Geospiza, peculiar to the

Galapagos Islands. Most of these nine host species are

pretty strongly parasitized, one of them, indeed, Geopiza

fuliginosus, having had nineteen Mallophagan species

taken from it, the record as regards number of parasite

species from a single host form. Of these nineteen Mal-

lophagan species, four belong to the genus Docophorus,

five to Nirmus, five to Lipeurus, two to Colpocephalum,

two to Menopon, and one to Goniocotes. Among them

are included all of the seven species that have been re-

corded from Camarhynchus pallidus, the most parasitized

species of this sister sparrow genus peculiar to the

islands. But also there are included several Mallopha-

gan species found on various other host birds widely sep-

arated phyletically from Geopiza and Camarhynchus.

In fact, in studying the parasitization of the Galapagos

Island birds—and I have had in hand two very full col-

lections from them—one is struck by the breakdown of

the general rule that the Mallophaga of one host group,

as a genus or family or order, shall be more or less nearly

exclusively confined to members of the group, and hence

to be characteristic of it. For example, one does not

expect to find the abundant duck parasite, Trinoton luri-

*I have recently recorded Nirmus vulgatus from a starling from

Egyptian Sudan.

No. 555] SPECIES-FORMING OF ECTO--PARASITES 155

dum on pigeons, nor the familiar pigeon parasite, Lipeu-

rus baculus on ducks, and one does not so find them.

Nor to find on sparrows or hawks Mallophaga of the

genera Ancistrona, Eurymetopus and Philoceanus, char-

acteristic of maritime birds; nor on owls or cuckoos to

find certain maritime bird-infesting species of Lipeurus

and Docophorus, genera in themselves represented by

species on hosts of many orders. But in the two care-

fully made collections of Mallophaga from the Galapagos

Islands, representing in their host lists practically all the

bird species known to inhabit the islands, I have noted

the exception to be the rule. And from interviews with

the collectors—one of them was one of my own assist-

ants—I have determined the probable reason for this

unusual state of affairs. It is this: The birds of the

land, the birds of the shore, and the birds of the sea meet

and rest side by side on the shore rocks and sands. The

land birds live chiefly not in the dense, almost impene-

trable jungle of the interior of the islands, but in the

outer or shore fringe of it. Here they meet and mingle

with the hosts of sea birds that find resting and nesting

ground on these few small bits of solid earth set in the

midst of all the leagues of inhospitable moving waters

that constitute their range. This brings about the op-

portunity and the reality of an abnormal but natural

straggling, which results in an extraordinary and prob-

ably unique host distribution of the parasites. And thus

it is that the little Galapagos Island sparrow, Geospiza

fuliginosa, comes to be the bearer of more Mallophagan

Species than any other bird in the host list and has in-

cluded among its guests many that more rightfully be-

long to birds of ocean and shore sands.

The Tanagride, or tanagers, are represented in the

host list by ten species. Menopon thoracicum is com-

mon to four of them, all of Central American range. The

Icteridæ, including the blackbirds, grackles, meadow-

larks, and American orioles, are represented by twenty-

two host species, and the Sturnidæ, or starlings, by seven.

156 THE AMERICAN NATURALIST [Vou. XLVII

The Paradiseidex, the radiant birds of paradise, are rep-

resented in the list by eight species.

Finally, the Corvide, or crows and jays, are repre-

sented at present by thirty-two species, about one out of

ten of the known kinds, of which thirteen belong to the

genus Corvus. Both these numbers will be increased

when I am able to incorporate in the list the records,

already worked but not yet published, of a considerable

collection of Mallophaga from the crows and jays of

India, sent to me by Superintendant Annandale of the

Indian Museum at Calcutta. The Corvide are provided

with a number of parasite species characteristic of the

family, such as Docophorus atratus, Colpocephalum sub-

equale, Menopon mesoleucum, et al. This latter parasite

occurs on four species of Corvus, two of old world and

two of new world range. Colpocephalum subequale oc-

curs on two North American and one European species

of Corvus.

V

From these examples of Mallophagan distribution

illustrating various special conditions of host relations a

number of rather important points of significance appear.

Of two only, however, which indeed at bottom are really

one, shall I speak. First, there is apparent in Mallo-

phagan distribution a general faithfulness of parasite to

host kind or group of related host-kinds, and this without

much reference to geographical conditions. And second,

there appears a plain tendency for a single parasite spe-

cies to be common to two or more related host species,

even though these hosts be so widely separated geograph-

ically and so restricted to their separate geographic

ranges that all possible chance of contact between indi-

viduals of the different host species seems positively pre-

cluded. The American and the European avocets do not

meet; nor do the American and European coots. Yet

the two coots have five Mallophagan species in common,

and the two avocets two. The American and European

bitterns are both infested by a common parasite species.

No. 555] SPECIES-FORMING OF ECTO-PARASITES 157

The American and old world dippers or water ouzels are

in the same case. The same Mallophagan species occurs

on both an old world and a new world kinglet. One para-

site species is common to two old world and two new

world crows. Practically all of those isolated bird spe-

cies found only on the Galapagos Islands are infested by

Mallophaga that occur on related new world or even old

world hosts.

Now, removing all cases of even an imaginable rare

possible contact of bodies between these related but spe-

cifically distinct hosts, such as might occur in birds of

circum-polar range, or in gregarious maritime kinds

meeting on common mid-ocean islands, or in kinds occa-

sionally exported by man from their normal range, etc.,

and there are still left many cases of this commonness

of a parasite species to two or more usually rather closely

related host species of quite distinct geographic range.

How can this actual condition be explained?

I can see but one answer. That is, that the parasite

Species has been handed down practically unchanged to

the present specifically and even generically distinct sev-

eral bird species from their common ancestor of earlier

days. The parasite species dates from the days of this

ancestor. With the splitting up of the ancient host spe-

cies due to geographic wandering and isolation of groups

of its individuals, and their gradual divergence in plu-

mage, color and pattern, shape of bill or toes or wings,

caused partly by adaptation and partly by the simple per-

Sistence of chance variations fostered by the isolation

and inbreeding, there has been no equivalent evolution-

ary divergence of the isolated groups of individuals of

the parasite species. No adaptive changes have been

necessary for it. It has indeed been broken into isolated

Sroups of individuals, but no more than is normal to its

life under conditions less novel. I have already pointed

out the large variability that occurs within every Mal-

lophagan Species caused by the separation, more or less

complete and persisting, of its individuals into little

158 THE AMERICAN NATURALIST (Von. XLVII

groups and family strains each isolated on its host island

or succession of self-reproducing islands.

But the change of plumage markings, or of bill shape

or even of food and flight habit, of the separating host

kinds splitting off from a common ancestor need mean

nothing much for the parasite. So although in time we

come to have, derived from a common ancestor, an Amer-

ican avocet and a European one, an American coot and a

European one, we do not have an American avocet- or

coot-parasite and a European avocet- or coot-parasite.

But the parasite of the common avocet or coot ancestor

of the two present bird species remains unchanged and

is thus a single species common to the two geographically

separated, specifically distinct, never-meeting, host

species.

If this is a true explanation for the commonness of a

parasite of two separated host species, it is likely also

the explanation of the larger phenomenon of the general

faithfulness of certain parasite species or genera to cer-

tain bird groups—families, or even orders. I do indeed

believe that it is a commonness of the genealogy rather

than a commonness of adaptation that is the chief ex-

planation of this restriction of certain parasite groups to

certain host groups. It is in my eyes an unusually clear

example of the potency of heredity. There is more

nature than nurture in the upbringing of the Mallophaga.

Isolation and inheritance, then, are the two evolution

factors especially concerned in the species-forming and

the distribution of the Mallophaga. Adaptation seems

to play a very subordinate part. And this is a rather

unusual condition in insect biology. The plasticity of

insect nature combined with the stresses of insect life

and the necessary shifts for a living, have resulted in pro-

ducing among the insects some of the most striking ex-

amples of adaptive evolution to be found in the kingdom

of life. In the face of this fact, this little group of para-

sites may have by the very exceptionality of its evolu-

tionary behavior, an enhanced interest for us!

CASTRATION IN RELATION TO THE SECONDARY

SEXUAL CHARACTERS OF BROWN

LEGHORNS!

H. D. GOODALE

STATION FOR EXPERIMENTAL EVOLUTION

Introduction.—It has often been observed, in many

sexually dimorphic species of birds, that a female occa-

sionally occurs which exhibits many of the characters of

her mate. Frequently associated with this condition are

alterations, more or less pathological, in the ovary. It

has been said also that the male sometimes, and in par-

ticular the unsexed male of the domestic fowl, presents

characters resembling those of the female. However, it

has been pointed out that the supposed resemblance to the

female can also be interpreted as due to the failure of

the development of the normal adult male characters.

In other words, these modified males may be referred to

either a juvenile or to a female condition.

These considerations led, then, to the following ques-

tions. Will removal of the ovary from the young female

fowl cause her to assume the characters of the male?

Will removal of the testes from the young male cause him

to assume any female characters or will it cause him to

.Tetain his own juvenile characters or will it be without

effect? For answers the Brown Leghorns seemed to

furnish particularly favorable material, because: first,

the adult plumage, which is strongly sexually dimorphic,

is practically identical with that of the Jungle fowl;

Second, the comb of the female, while proportionally

smaller than that of the male, is larger than that of the

male of many other varieties; third, there are at least

*This paper with some changes was read before the American Society

of Naturalists at Princeton, N. J., December 28, 1911. Since then the

experimental results have been fully verified. Male characters have devel-

Oped on 25 females, following ovariotomy. A complete account of the

newer experiments will be published later.

159

160 THE AMERICAN NATURALIST [ Von. XLVII

three distinct stages in the development of the young

bird’s plumage before the adult color is reached. The

first two stages do not exhibit sexual dimorphism, nor are

they identical with that of the adult female. Sexual

dimorphism appears first in the third juvenile plumage.

At this stage, while the young female rather closely re-

sembles the adult female, the young males in their tout

ensemble are distinct from either the adult male or

female, though sometimes feathers like those of the fe-

male may be found. A study of the characters of the

Brown Leghorn capon or poullard, therefore, should show

whether their characters are juvenile, male or female.

For most characters, the results obtained from a pre-

liminary set of experiments are clear cut.

Description of the Secondary Sexual Characters.—In

Table I is given a comparative statement of the more

pronounced secondary sexual characters common to

nearly all races of domestic chickens. There are other

differences between the sexes, which are more subtle and

therefore not considered here. In Table II are given

other secondary sexual characters found in the Brown

Leghorns but not in all other races of chickens.

Comment on Tables—Such exceptions to the general

statements given in Table I as occur, are mainly negative,

that is, they are characters which for one reason or

another do not become patent, as for example, the failure

of the whole or a part of the color pattern to become:

visible, particularly in the case of uniformly colored

birds.

In considering plumage characters aside from shape of

feather there are three features to be taken into account.

First, the localization of color in certain definite regions

of the body, thus forming the body pattern; second, the

localization of color on definite regions of the feather,

forming the feather pattern; third, the pigments them-

selves, which are associated with each pattern in various

ways.

It is rather striking, that in spite of the numerous

No. 555]

CASTRATION IN BROWN LEGHORNS

TABLE I?

A COMPARATIVE STATEMENT OF THE CHIEF SECONDARY SEXUAL CHARACTERS

or DOMESTIC CHICKENS, COMMON TO NEARLY ALL VARIETIES

161

Size, includ-

Body color

patte:

mi.i —

vi

Feather

pattern.

Behavior...

es

Adult

Male

Female

Young of Both

Sexes

ieee large.

Well developed.

. Pie sat saddle ke

arrow, pointed a

Barbules absent distal

h

dle, and

w fond bom Gf.

Tail eoreti: including sick-

les, g, curved and

pointed

distinct color

. hackle,

baek, saddle, wing front,

bow, win ade Ser

bar

tail breast aera vitae

surface.

-Usually uniform’ or with

central stripe

Relatively erect carriage,

pugnacious, frequ uen

crows, rarely ‘‘sings,’’

appears to feck brooding

instinct.

Relatively small.

Absent.

Hackle feathers similar to

hose of male but rela

betes sho sgi Pagie ore

unded addle

feathers posse neart

shor . broadly

psc ip

Barbules absent teak distal |

third o e fe semen

verge A present U her

Ch Fig. 3, D.

Tail

short, broa

at tips.

coverts, relatively

d and rounded

Relatively red areas; viz. 3

mainder oi

were or penciled when

t like that of the male.

dgan to fe-

male.

Barbules

resent.

Similar to fe-

male.

Varies with

different

Relatively horizontal car-|Peeps for a

riage, not pugnacious, e, en

does not crow Meg t follows a

‘sings’ 5 great deal iod

when in laying Bo weg when voice

rooding instine

until

approach-

ing

ity.

varieties of domestic chickens, there is, with one excep-

tion, only one basic body pattern, that of the Jungle fowl.

(See table.) All other patterns, with the exception of the

white crest of certain Polish, must be considered as modi-

*These tables make no pretense of giving all the characters which differ

in the two sexes, I have also ignored minor details in some characterizations.

Un iform, except as cay under color can be distinguished from the sur-

face color.

See also

«

162 THE AMERICAN NATURALIST [ Vou. XLVII

fications of the Jungle, usually by the loss or non-devel-

opment of a part. I know of no other distinctive pattern

in chickens, such as, for example, that exhibited by the

Mallard in ducks.

TABLE II?

A COMPARATIVE STATEMENT OF THE CHIEF SECONDARY SEXUAL CHARACTERS

F BROWN LEGHORNS, OTHER THAN THOSE GIVEN IN TABLE I

Adult Juvenile (Third Stage)

Male Female Male Female

Comb..... eon Ose Lops. Blade Erect. Rela- Erect. Rela-

rtion- proportion- tively large tively small

st large. ately small. and of quick and of slower

growth. growth.

Spurs..... Present. Usually absent,

but frequently

present in var-

ious conditions

develop-

of

ment

Plumage

color....|/Hackle, reddish|Hackle, yelow,| Mostly black, Similar to adult

orange, with! with dark-| each feather’ but markings

black central) brown central! being splashed) coarser.

stripe ck ipe; breast, with various

: ail,| amounts of

ark crimson;| primaries and| red. Feathers

saddle, reddish} concealed por-| of back often

orange, often’ tionsofsecond-| like those of

lac ies, coe young fem

central stripe;| rest of Body pattern

wing bay, red; “oie = ight of adult more

dorsal white or less clearly

area at base of cael in| indie

tail; rest of pper and

body black. salt fashion.

The feather patterns fall into two groups, viz., those

associated with sexual dimorphism and those not so asso-

ciated. In the male there is no pattern of the first group.

But penciling of the type found in the Dark Brahma

female appears never to occur in the male and therefore

belongs in the second class. Stippling is also usually

distinctive of the adult female, but it occurs in the young

male and under certain conditions (heterozygosis) in the

adult male. In the Jungle female the stippling often

approaches very closely the form of narrow concentric

lines or vermiculations. Possibly the stippling of the

Brown Leghorn female is due to the breaking up of such

No. 555] CASTRATION IN BROWN LEGHORNS 163

vermiculations and so is not strictly comparable with

that due to heterozygosis. However this may be, stip-

pling is normally a female or a juvenile male character.

The males corresponding to these two types of females

usually have uniformly colored feathers ventrally. The

dorsal regions are less constant in type, being usually

uniformly colored or else striped. In other breeds the

feather pattern may be alike in the two sexes.

The question of behavior depends upon so many condi-

tions, that only a few of the numerous reactions have

been mentioned in the table.

The following comments are intended to apply only to

Brown Leghorns.

While the comb of the female usually lops to one side

or the other, it does not always do so. There are indi-

cations, moreover, that by proper matings this character

could be transferred to the male.

Spurs are usually absent from the female, yet they are

occasionally well developed. Their presence can not be

taken as a certain indication of the assumption of a male

character by the female, for there was an old Scottish

race in which both sexes were spurred. Nor need they

indicate the presence of an abnormal ovary, for in one

case in which they had become well developed at 9 months

of age, the hen proved to be a splendid layer. These

spurred hens are probably due to the presence or absence

in the germ plasm of some definite determiner, as certain

of my breeding experiments indicate.

The plumage color within certain limits is variable, so

that it is a little difficult to describe briefly. American

fanciers recognize only one type of plumage color for

each sex, that described in the ‘‘American Standard of

Perfection,” but beside the ‘‘Standard’’ female there is

one in which the salmon breast is replaced by one of

nearly the same color as the back. Beside the ‘‘Stand-

ard” male is one in which the striping of the hackle and

saddle feathers is lost, and third, one in which there is

some red in the under parts. It is evident, moreover,

164 THE AMERICAN NATURALIST [ Vou. XLVII

from experiments in cross breeding, that the body pattern

of the Brown Leghorn is not a unit, but is composed of

several parts.

These departures from the ideal and the probable vari-

ability in the gametie constitution of supposedly pure-

bred stock may be expected to influence the behavior of

the birds under castration. They indicate also that a

character which at one time may behave as a secondary

sexual character, at others forms a normal character of

both races.

Experimental.—Without taking time to describe the

details of the experiments, except to note that the gonads

were removed from the chicks when they were 21-28 days

old, that is, they were much younger than is the case in

commercial caponization, we may proceed directly to a

consideration of the results, taking up first the effects of

castration on the male.

Seven of the castrated males reached four months of

age, then three were accidentally killed. The autopsies

on these three showed an entire absence of testicular —

material in two cases. The third had a tiny nodule, pos-

sibly testicular, on the mesentary near the former site of

the testis. All four survivors were kept until they were

16 months of age. Two were then killed for study. They

developed the normal plumage of the male and were it

not for the small comb and wattles they would have had

nearly the same appearance as a normal cock (Fig. 1).

They are, however, somewhat fuller plumaged and rather

heavier looking birds than a normal cock. The spurs are

well developed but on the other hand these capons are

less active in their movements, are non-combatants and

show no sexual instincts. They have never been heard

crowing, though they may be made to squawk or even

cluck like a cock.

One of these capons was somewhat anomalous in that

he had much the same appearance as a normal cock. He

had a large comb but did not crow. At one time he

* Unfortunately this material was not sectioned.

No. 555] CASTRATION IN BROWN LEGHORNS 165

showed some inclination to pay attention to the hens, but

as far as I was able to observe, it never went as far as an

attempt at copulation. At the autopsy it was found that

there had been an autoplastic transplantation of a bit of

the testis.

Fic. 1, carey Leghorn Capon. The colors are exactly those of the male as

given in Table

It will have been observed from the foregoing descrip-

tion, that the small comb of these capons is the only char-

acter which might be considered female. In all other

points the characters of the capon are the characters of

the cock, sometimes exaggerated (feather length), some-

times infantile (crowing instinct). None are positively

female, except perhaps the comb, which is much too large

to be purely infantile, and too small and of the wrong

proportions to be that of a normal adult female. On the

whole it most closely resembles that of the mature young

female before she has actually commenced to lay, or of

the adult female when out of laying condition. Comb size

is affected by so many conditions that the question as to

whether or not the capon has a female type of comb is

not easily answered, and therefore must await further

studies,

We may turn now to the effects of castration on the

166 THE AMERICAN NATURALIST [ Vou. XLVII

female. Unfortunately, only one of the females from

which the ovary (naturally only the left, since it was

assumed that the right had completely degenerated) was

removed, reached maturity. There is no doubt as to the

effect of ovariotomy on this individual (Fig. 2), which

Female Brown Leghorn from which the ovary was removed. The

colors, with the exception of some feathers as noted in the text, are those of the

male as given in Table II. The long saddle feathers are hidden by the wing,

but see Fig. 3, 0.

commonly passed as a cock, with those unacquainted with

the bird’s history. Nevertheless, the assumption of male

characters has not been quite complete, as is shown by a

consideration of the following features. First, the pres-

ence of feathers on the back which are very much hen-

like (Fig. 3, B). Second, the wing bow is poorly de-

veloped. Third, the shank is too short for that of a cock.

In some points of behavior, such as lack of the crowing

instinct, non-combativeness, cackle and general indiffer-

ence to the hens, she is hen-like. On the other hand, her

carriage is cock-like and by suitable means she has been

made to cluck like a cock and even to pay a little atten-

tion to the hens, though this last reaction was produced

only once or twice. The peculiar shape and carriage of the

tail probably has nothing to do with the effects of castra-

tion, since this bird had only half a rump. Moreover, a

rumpless cock occurred in this same strain.

No. 555] CASTRATION IN BROWN LEGHORNS 167

Some additional light on the question is afforded by the

effect produced by castration on a second female, which

died at the age of four months. When the first feathers

of the third stage began to appear, they were like those of

the young male. The later feathers, however, were those

of the normal female. The explanation of this anomalous

behavior was found at the autopsy, for it showed the

———

Fic. 3. Feathers from the are region of the various types of birds

deserved A, normal male; B an d ©, two sorts of feathers from the bird shown

n Fig. 2; D, normal female; E, capon.

presence of an ovary about one half the size of that of a

normal chick of the same age. As long as the ovary or

its normal secretion was practically absent, the develop-

ment of this female was along lines similar to that of the

male, but with the regeneration of the ovary develop-

ment returned to its usual course.

These experiment s, then, indicate clearly that while the

emale may assume male characters following the re-

moval of the ovary, the male assumes no positive female

characters after removal of the testes.

168 THE AMERICAN NATURALIST [ Vou. XLVII

Theoretical_—Certain of my breeding experiments have

seemed to show that the Brown Leghorn plumage colors

and pattern followed a sex-limited mode of descent. Cer-

tain difficulties, however, have arisen. As stated above,

it has become evident that the coloration and pattern of

the Brown Leghorns is not due to a single inheritable

unit, but to a complex. This means that its more or less

independent parts must be discovered and the mode of

inheritance as well as the conditions under which each

part becomes visible in the soma, worked out. Until this

is done it will be impossible to say what part sex-limited

inheritance plays in determining sexual differences. The

following representation, then, is merely an attempt to

formulate a working hypothesis.

The male may be considered to be duplex for an internal

secretion S, produced by the testes, which is necessary

for the full development of the comb and less clearly for

the crowing instinct and sexual behavior, but not for

plumage and spurs. In the female this secretion is re-

placed by one S’ produced by the ovary. (S’ may, per-

haps, stand in some simple chemical relation to S.) To

its effect is to be referred the female’s form and color.

When the ovary is removed or becomes pathological so

that its normal secretion is no longer produced, then the

male characters develop to an extent which is determined

by factors at present unknown to us. Nor are we con-

cerned here as to what part of the ovary or testis pro-

duces this secretion (or secretions, for S and S’, respect-

ively may well represent a number of secretions which

behave in the same general way). In any case the in-

ternal secretion may be conceived to be associated with

sex in the same manner as any other sex-limited char-

acter. The formula for the male would then be Ig Sd,

that for the female SJS’?. S is of course to be consid-

ered dominant to S. If this scheme represents the actual

condition of affairs it may not be necessary to suppose

that the female Brown Leghorn is simplex for L where L

represents the whole or a part of the male Leghorn

No. 555] CASTRATION IN BROWN LEGHORNS 169

pattern complex, for S’ might be able to transform two

doses of Z into the condition observed in the female as

well as one dose. This is, however, a matter to be tested

_by experiment.

In insects, Meisenheimer and others have shown that

castration is without effect on the secondary sexual char-

acters of either sex. This lack of association between the

secondary sexual characters and the gonads may find its

explanation in the absence of a modifying secretion from

these forms, the secondary sexual characters being deter-

mined solely by differences in the gametic constitution of

the sexes, for example of the form AA in one sex and AB

in the other, B being dominant over A and linked with

the sex determiner, in the usual fashion for a sex-limited

factor. Secondary sexual characters of insects would

then belong to a different category from those of the

birds. There is some evidence, however, that certain

characters of the female fowl are not under the control

of the secretion of the ovary, for they do not become

male-like after the removal of the ovary and therefore

are like the insects in this respect. This point requires

further study.

There is one more point to be considered. I have shown

that in the Brown Leghorns as well as in Rouen ducks

that the male does not assume female characters as

a result of castration. Does the adult male of other

Sorts of birds ever exhibit female characters? An affir-

mative answer is possible only in those cases in which the

juvenile plumage of the male is like that of the female.

But if we consider only those cases in which the juvenile

male plumage is unlike that of the female, we shall find

that conclusive evidence on this point is wanting, at least

I have thus far been unable to secure such evidence.

5 There are, of course, other ways of representing this association.

SIMPLIFICATION OF MENDELIAN FORMULÆ

PROFESSOR W. E. CASTLE

Bussey Instrrution, HARVARD UNIVERSITY

Proressor Bessey in his recent presidential address*

expresses the opinion that Mendelian terminology is need-

lessly complicated. This opinion most biologists will

heartily endorse, and not a few Mendelians will be among

their number. For those who work most extensively with

Mendelian formule feel most keenly the need of simpli-

fication in these the tools of their investigations.

Professor Morgan, in the January Narurauist,? makes

a commendable effort to introduce reforms. I desire

heartily to endorse his effort, but would suggest certain

modifications in method. i

The Mendelian may say in justification of existing us-

age that it has arisen naturally step by step as knowledge

of Mendelian phenomena has advanced, but this is of

course no justification of its continued use, if it has be-

come a hindrance rather than a help in the further ad-

vance of knowledge.

Morgan clearly points out the two historical steps by

which present usage was reached. The first of these was

Mendel’s original recognition of segregating dominant

and recessive characters existing in contrasted pairs, and

his convenient designation of the former by capitals and

of the latter by small letters. ‘This usage answered per-

fectly so long as only a single modification of any char-

acter came under consideration, and indeed Mendel’s

observations did not go beyond this. But this system

broke down when characters more complex in nature came

under observation, as for example when Cuénot showed

that more than a single differential factor exists between

gray mice and albino mice. (2) The ingenious and useful

1 Science, January 3, 1913.

2 Vol. 47, pp. 5-16. :

170

No. 555] SIMPLIFICATION OF MENDELIAN FORMULZ 171

‘*presence and absence’’ hypothesis of Bateson was the

second step which led to our present usage. On this hy-

potheses gray in mice is not the allelomorph of white, but

of no-gray; while the allelomorph of white is color, or

more properly speaking white is equivalent to no-color

and this is the allelomorph of color.

Both of these steps have been amply justified by their

utility in making possible the prediction of the previously

unpredictable consequences of particular crosses.

It was natural that in applying the presence and ab-

sence hypothesis the usage of Mendel should have been

retained, in accordance with which capital letters were

used as the symbols of dominant characters and small

letters as the symbols of recessive characters. But this

retention has involved most unfortunate consequences and

is, I believe, the real seat of our present difficulty.

Mendel’s small letters stood for realities as truly as did

the capitals. His A was a round form of pea, his a was a

wrinkled form of pea; his B was a yellow-seeded, his b a

sreen-seeded pea. But the significance of these terms has

been changed under the presence and absence hypothesis.

A still means a round pea, but a is simply a not-round

pea; it may or may not be wrinkled. Likewise B is still

a yellow-seeded pea, but b is nothing but a not-yellow

pea; it may or may not be green under the presence and

absence hypothesis. For all that b signifies now, the pea

may be blue, violet, indigo or carmine.

It is most unfortunate, therefore, that the small letters,

having lost their original significance, were not discarded

altogether, for under the presence and absence hypothesis

they have done nothing but cause mischief.

The investigator who employs them starts out well in-

tentioned with a clear notion that the small letters stand

for negation only, that they are merely signboards to

Show what characters he is talking about, but presently,

unless he is unusually careful, we find him talking about

them as if they stood for something, instead of nothing;

he speaks of repulsions and couplings or associations

172 THE AMERICAN NATURALIST [Vou. XLVII

between a and B, or even between a and b. Think of it!

How can something be coupled with nothing? How can

nothing be inseparably bound up with nothing? It seems

to me the consequent effect on inheritance is absolutely

“nothing”!

Not only do the small letters thus lead to confusion of

thought, they also tend to make formulæ needlessly cum-

bersome, for they call for the use of two symbols for

every character difference dealt with. These two sym-

bols also are so much alike that both printer and reader

are in momentary danger of confusing them, with the

consequence that what is is not, and what is not is!

The small letters are not indispensable to accurate and

exhaustive analysis of Mendelian phenomena, or to lucid

exposition of them. See, for example, the fundamental

researches of Cuénot into the color inheritance of mice,

and his classic ‘‘notes’’ describing them. Like Cuénot, I

have not found the use of the small letters necessary ; but

among nearly all other Mendelians the double termin-

ology has become so nearly universal that a different

usage seems almost to demand an apology. Indeed Lang’

has suggested that such offenders against uniformity as

Cuénot and I should be haled before an International

Congress and be directed to conform; since which time

I had almost abandoned hope of ever seeing improve-

ment in the current confusing system, but Morgan’s pro-

test and proposal gives me new courage.

What we need first of all to symplify our present usage

is to abandon the dual terminology. Where we are deal-

ing with a single set of variations, let a single set of sym-

bols suffice. Let us give up either the small letters or the

large ones, it matters not which. If we retain A, then we

have no need of a, for it is not, as Morgan at one time

seems to assert and at another to deny, the ‘‘residuum”’

when A is lost; it means on the presence and absence

hypothesis nothing but this, that A is not present. The

rest of the organism is the ‘‘residuum.’’ Morgan points

° Zeitsch. f. ind. Abstammungs- und Vererbungslehre, 4, p. 40, 1910.

No.555] SIMPLIFICATION OF MENDELIAN FORMULZ 1738

out and his paper illustrates amply how under the dual

system ‘‘the letters used may unintentionally come to

stand for different things.” The obvious thing to do, if

we attempt reform, is to omit the superfluous symbol,

either the small letter or the large one.

Morgan, however, clings to the dual nomenclature, but

suggests a reversal of the usual significance. Thus the

factor for pink-eye, he assumes, is present only in animals

which are not pink-eyed, and the factor for black body

color, he suggests, is present in all sorts of flies except

those which are black bodied. This is confusion worse

confounded.

But, seriously, I do not see that it is possible to improve

the existing terminology, so long as we use two terms of

opposite significance with reference to a single germinal

variation. Certainly merely reversing the significance of

existing terms will not do it. What we need first of all is

one set of symbols, used in a single significance.

If this reduction is allowed, then I think that another

aspect of Morgan’s proposition might be extremely use-

ful, viz., that a mutation which behaves as a recessive in

crosses be designated by a small letter. This proposition

was put into effect more than three years ago in a paper

dealing with color inheritance in mice, though Morgan

does not seem to have observed it. See Castle and Little

(1909). In the paper cited, three recessive color factors

of mice were designated by small letters, viz., ‘‘d, the dilu-

tion factor’’; ‘‘s, the factor which causes spotting with

white;’’ and ‘‘p, the pink-eye (or paucity) factor.’’

In that same paper all dominant color factors of mice

were designated by capitals. This seems to me a very

necessary complement to the use of small letters to ex-

press recessive variations, and is in entire harmony with

Mendel’s original usage. But neither of these proposals

can help matters much, unless we discard the duplicate

set of symbols, which is the chief cause of present con-

fusion. Thus if we use s for spotting, then we have no

‘Science, N. S., Vol. 30, pp. 312314.

174 THE AMERICAN NATURALIST (Vou. XLVII

occasion to use S for no-spotting. We simply leave out

all reference to spotting, and we shall understand that

there is none, but that the normal condition prevails.

To be very explicit, my proposals for simplification of

Mendelian terminology are three:

1. To abolish the current dual terminology and use only

one symbol, where a single variation from the normal is

involved.

2. To use a small letter to designate the factor respon-

sible for a variation which is recessive in crosses with the

normal.

3. To use a capital letter to designate the factor respon-

sible for a variation which is dominant in crosses with the.

normal.

These proposals were made in substance in publica-

tions of the year 1909 and are here renewed under en-

couragement of Morgan’s suggestive paper. Let us see

how they would work if applied to the cases enumerated

by Morgan. The eye color series described by Morgan,

l. c., page 13, involving three recessive mutations, is as

follows:

Revised Terminology Morgan’s Terminology

CC ne bee E sas PVE

WOPMINON (2 0s 1.6 fat cies ee v PvE

DE pi Pt ee ie SE p pVE

Pok vermilion... -ou erhoa pv pvE

Pe ee os he bore. oe e FV e

VOO ea oe Se ve Pve

PIR OO ar a pe pVe

Pink-vermilion-eosin ........... pve pve

The revised terminology is obviously shorter and

simpler. It is obtained by merely omitting the capital

letters from Morgan’s terminology, letters which stand

only for negations.. The symbols used are suggestive of

the names employed for the various color categories of

eyes, whereas in Morgan’s terminology the most con-

spicuous symbols are suggestive only of other categories

than the true one.

The revised terminology is more convenient than

No. 555] SIMPLIFICATION OF MENDELIAN FORMULZ 175

Morgan’s in calculating the expected result of any ma-

ting, and it is equally reliable. The result of every pos-

sible mating within the series can be readily computed

without the confusing presence of the large letters.

To those who have grown accustomed to the presence

and absence terminology the objection will suggest itself

that in naming the recessive character and ignoring its

allelomorph, we are naming an absence or negative and

disregarding what is present and positive. But this does

not follow. Because a character is recessive it does not

follow that it is negative. I quite agree with Morgan that

the physiological condition which produces an eosin eye

is as real as that which produces a vermilion, a pink or

a red eye, and no mere negation; it is simply different. It

is quite impossible to decide, from its behavior as a domi-

nant or recessive in crosses, whether a character is posi-

tive or negative. This I have pointed out elsewhere

(1911) and the same view has been repeatedly expressed

by Shull. We have on record many instances in which

one and the same character may behave at one time as a

dominant, at another time as a recessive.

Our terminology may well recognize the dominant or

recessive behavior of a variation, without implying any-

thing as to its positive or negative nature, which must in

many cases be conjectural or possibly non-existent. Dif-

ferent gradations of color, such as we have in the eye-

Series of Drosophila described by Morgan, may result

merely from quantitative variations in cell constituents

and consequent activities, nothing being lost. This idea

concerning the possible nature of Mendelian factors in

general I have developed elsewhere, concluding that ‘‘it

1s the substantial integrity of a quantitative variation

from cell-generation to cell-generation that constitutes

the basis of Mendelism. All else is imaginary.’”

Morgan applies his altered system of nomenclature also

to the body-color series and wing mutation series which

he has discovered. This nomenclature we may simplify,

* AMERICAN Naruraist, Vol. 46, p. 358, June, 1912.

176 THE AMERICAN NATURALIST [Von. XLVII

as we did in the case of the eye-color series, without im-

pairing its utility.

Bopy-coLor SERIES

Revised aa Morgan’s Terminology

Wild MF ca enaa rasau mal YBES’

SAUN arrr S ae Yy yBES

CUOW“DIACE © 6 55 lh ws yb ybES

POR SEY SOTA OER Ss e YBeS

Bane ea eas fa 8 YBEs'

WING-MUTATION SERIES

Revised Terminology Morgan’s Terminology

RLY Sais pits SOG Liesl ¢ normal MR

SRA PP E E wine crammane m mR

SS ee a E r Mr

oer gears minature 2:05. -¢... mr mr

The taste of the reader will govern his choice between

these two systems. Doubtless either can be used success-

fully, though the revised terminology seems to me prefer-

able on the ground of simplicity and suggestiveness.

In the series with which Morgan has dealt, all the muta-

tions under consideration are recessive in character, so

that one can read the names of the varieties directly from

his formule, if one disregards altogether his large letters

and pays attention only to the small ones. To insure

this I have suggested omitting the large ones.

But if one were to extend Morgan’s terminology to a

series in which dominant mutations as well as recessive

ones occur, hopeless confusion would result. For here

some of the large letters would stand for mutations, while

others would stand for the negation of mutations, so that

without a key constantly at hand the formule would be

unusable.

If, however, we use the single system of symbols as I

have suggested, a series which includes both dominant

and recessive mutations may be handled without con-

fusion. In this case every symbol is significant, and its

° For simplicity I here use E instead of Morgan’s Eb.

* Morgan’s list here contains S, but this I suspect is a misprint for $;

if so, it is a living witness to the dangers of the dual system.

No. 555] SIMPLIFICATION OF MENDELIAN FORMULZ 177

dominant or recessive character is indicated by the sym-

bol, whether large or small. For example, consider the

mouse-color series as described by Castle and Little

(1909). In the paper cited, nine color factors were de-

scribed, three of which clearly recessive have already been

mentioned, viz., d, p and s. The remaining six were con-

sidered dominant factors. Mr. Little has since suggested,

and I think with good reason, that one of them had

better be omitted, since its existence has not been demon-

strated beyond question. The six as given were C, the

color factor; Y, the yellow factor; Br, the brown factor;

B, the black factor; R, the restriction factor (producing a

yellow coat); and A, the agouti or gray factor. _

Mr. Little would omit either C or Y, since it has not

been shown beyond question that the effects which had

previously been ascribed to these two are not due to one

and the same agency.

With the eight symbols which would remain, three being

small letters, the others being or beginning with capitals,

it is possible to write, without duplication of terms, for-

mule descriptive of the entire color series. But in so

doing it would be necessary to designate the original or

wild form in terms of factors supposed to be lost in its

derivatives, and which have only come to light through

such loss. This, as Morgan points out, involves rede-

scribing the wild form every time a mutation arises and

should be avoided if possible. I therefore favor Morgan’s

Suggestion that each mutation as it arises be given some

suitable descriptive name, the initial or other significant

letter of which shall be its symbol. If, as is commonly

true, the mutation is recessive in crosses with the wild or

original type, its symbol will be a small letter. But if the

mutation is dominant,’ its symbol should be a large letter.

The original or wild type need not be described in terms

of its mutations, as every duplicate system of termin-

ology, even Morgan’s, requires. The system would ac-

“I have met two dominant mutations in guinea-pigs, one in rabbits, and

one in mice, so that they can searcely be called rare.

178 THE AMERICAN NATURALIST (Vou. XLVII

cordingly be capable of indefinite expansion without

constant remodeling.

I favor Morgan’s further suggestion that as new forms

arise through recombination of simple ‘‘mutations”’ these

be described, so far as possible, in terms of the simple

mutations composing them. This principle is clearly

illustrated in the names chosen by Morgan for the eye

color series of Drosophila. It is surprising how little

change this system necessitates in the common names

with which we are already familiar, for example, in the

mouse-color series.

The color mutations? of mice with inal I am person-

ally familiar number seven. If all of these are independ-

ent, i. e., not ‘‘coupled’’ or ‘‘associated,’’ there should be

theoretically possible 127 different combinations involv-

ing one or more of them. A considerable proportion of

these combinations has been produced in my laboratory

in the course of the last twelve years, the earlier and

simpler ones by Dr. G. M. Allen or myself, the later and

more complex ones by Mr. Little, who has in press an

extensive paper dealing with his investigations. I shall

deal with the series as known up to 1909. The historical

order of appearance of the mutations is now unknown;

I shall place them in the alphabetical order of the sym-

bols used. It is also unknown whether each of them arose

directly from the wild type. More probably they did not,

but experiment shows that they might have done so, since

each behaves in crosses as if it had a distinct and inde-

pendent basis in the germ-plasm.

Witp TYPE AND iTS Seven MUTATIONS

1. Wild= gray.

2. a=albino (transmitting gray in crosses).

3. b= black.

4. c= cinnamon.

5. d— dilute.

° I use the term mutation in the sense of unit-factor variation, not in that

of DeVries.

No. 555] SIMPLIFICATION OF MENDELIAN FORMULH 179

6. p — pink-eyed.

7. s=spotted.

8. Y=yellow.

The a mutation, however combined, if present in a homo-

zygous condition, prevents the development of pigment

in the skin, hair or eyes. The d mutation, under like cir-

cumstances, makes the pigmentation of the coat dilute, or

pale; the p mutation reduces even more strongly the pig-

mentation of coat and eyes alike, but does it in a different

way ; the s mutation causes pigment to be altogether want-

ing in certain areas of the coat more or less definite in

position and extent, which areas accordingly appear as

white spots.

Combinations of these four mutations present no diffi-

culties of description or recognition, though breeding

tests alone suffice to differentiate the several sorts of

albinos, since all look alike. The nomenclature also is

perfectly simple. Thus,

ap—albino transmitting the pink-eye mutation in

crosses.

adp=albino transmitting both dilution and pink-eye

in crosses, ete.

Combinations of b, c, and Y, one with another, form the

fundamental and best known color varieties, which will

now be considered.

In the b mutation, the fur is black; in the c mutation,

it is brownish gray, called cinnamon. In the Y mutation,

the coat is yellow. Of the several mutations mentioned,

Y alone is dominant over the wild gray, but it occurs only

in a heterozygous state, and hence never breeds true.

The complete color series involving these three muta-

tions, but excluding all others, is as follows:

Wild —eray.

b= black.

c — cinnamon.

bc—black-cinnamon (chocolate).

= yellow (giving also gray offspring).

bY —black-yellow (giving also black offspring).

180 THE AMERICAN NATURALIST [Vou. XLVII

cY —cinnamon-yellow (giving also cinnamon offspring).

bey =black-cinnamon-yellow (giving also black-cin-

namon offspring).

To express the modification which this series under-

goes if the d mutation is added to it, we need only pre-

fix the symbol d to each of the formule given and omit

the term wild as no longer applicable. The series then

becomes

d—dilute gray.

db = dilute black.

de= dilute cinnamon.

etc.

Similarly an added p factor gives us the series

p= pink-eyed gray.