THE NEW YORK ACADEMY OF SCIENCES

(Lycrum or Natural History, 1817-1876)

OFFICERS, 1911

President—FRanz Boas, Columbia University

Vice-Presidents--GrorcE F. Kunz, FrepErIc A. Lucas,

WILLIAM CAMPBELL, R. 8. WoopworTH

Recording Secretary—Evmunp Otis Hovey, American Museum

Corresponding Secretary—Hrrmon C, Bumpus, American Museum

Treasurer—EMERSON McMituin, 40 Wall Street

TAbrarian—Ratpeu W. Tower, American Museum

SECTION OF GEOLOGY AND MINERALOGY

Chairman—GeorGE F. Kunz, 401 Fifth Avenue

Secretary—CuHAk Les P. Berkey, Columbia University

SECTION OF BIOLOGY

Chairman—Freveric A. Lucas, American Museum

Secretary—L. Hussaxor, American Museum

SECTION OF ASTRONOMY, PHYSICS AND CHEMISTRY

Chairman—WILLIAM CAMPBELL, Columbia University

Secretary—EpwarpD J. THaTcHER, Teachers College

SECTION OF ANTHROPOLOGY AND PSYCHOLOGY

Chairman—R. 8. WoopwortH, Columbia University

Secretary—F. LyMAN WELLS, Columbia University

CONTENTS OF VOLUME XXI

Page

PEL Oy A Ces icevats hate etavelc suclevelovey sicher cratetasiolat setae) crevesehaleveieilel aisle Srerercieersioa dle oe vals i

(Qi Wsaa oo ocodon DUB OOODOUCCOOOSOGUGUDUEOOOOD gi ttr tsetse ects t eee ee ee ii

COTTE 5h 5 5.0 db OOO CeO Goal bocuE CDADeOODCE.CCG GUGOOO OL OC OnRCr DME EmCaor iii

WMAteShOre EU pDiCaAtOns OtcAULNOLS s SCPALAGCCS «le cielsretelololelelslelsiciicls|s/ cies s\cleleyaic’s iii

HOTS O te USE ALIONS 2 2121. c/s) -lovcic! oreo SOPaCADAOOOGOIS BOO DOD OC DOOD RCOOGOOOOT iv

The Ravenswood Granodiorite. By Victor ZirEGLER. (Plates 1-II)....... il

Geology of the Cortlandt Series and Its Emery Deposits. By G. SHER-

BOURNE PLOGERS a) (E1ACES DET Vill) Ss. cere «c1s<.5 5 .ciale ote ci creer ohenete le efeveleuaiere 11

The Influence of Heredity and of Environment in Determining the Coat

Colorssin) Mices By, 22 EH Morgan. (Plates) VII—WX5) <n os eerie 87

On Some New Genera and Species of Pennsylvanian Fossils from the

Wewoka Formation of Oklahoma. By Grorce H. Girty........... 119

A List of the Type Species of the Genera and Sub-genera of Formicide.

ES Van EAI VL ORTON WEUBEIMNR 5 Se. js/eccvoise carci © 0.65 chee cic o.h a cer ermremiele 157

The History of the American Race: By FRANZ BOAS............cceecce«s ECE

Records of Meetings, 1911. -By EDMUND OTIS HOVEY..........-ccccceccee 185

PHN Orxani ZA] ONO fet yA GAGE ciscrs oles oles e) core co Sie aei cole: che sulci e oilers PPL

new Orieinell a © WaArberc ays, se syscfeisre, craiel erelers sialecere wicks eid. wiecesa/a nie tehermre oma PPL

CO RGLEEFO fe OO Us bee reread oi eteavols vel che oy oh eae saved e,) ot on siiel eae Gud Siac aletbinin igre ahelse ao

Mier Am end@edeG Marveleee ers ere synch sors he Societe iano i erere) ae. 8." aeselloyéieis Aion ect otereee 230

OOMSITTETLED Me 5 Oso oo RAC CCID eG COE. CIO DIS OLIGO CEC nee aaa aN EAD 233

LEWIN hetecees olen och CRANE ESCO COR CRRCIS OSE CEN EC Fra ee eer a 234

MEemMPErESHIP MISTS ole CCEMD eT al Oil rare stars accls, © citcvcisicie o's: « o o\aterore cielcrelecerceie 241

ire Cl oxcretemereteronsretevene cratorctere lace terciokenetcns¥ caer ecko seis ene aval clei svar sie le Slaven o's Guseatoreiaie w wate 253

DATES OF PUBLICATIONS OF AUTHORS’ SEPARATES

Edition.

Pp. 1-10, 10 May, 1911. 325 copies.

Pp. 11-86, 15 May, 1911. 275 copies.

Pp. 87-117, 5 July, 1911. 200 copies.

Pp. 119-156, 26 August, 1911. 50 copies.

Pp. 157-175, 17 October, 1911. 250 copies.

Pp.177-183, 20 March, 1912. 150 copies.

Pp. 227-252, 25 March, 1912. 50 copies.

iii

LIST OF ILLUSTRATIONS.

Plates.

I.—Photomicrographs of Ravenswood Granodiorite.

II.—Distribution of the Ravenswood Granodiorite.

II1I.—Pyroxenite and Diorite.

IV.—Gneissoid Structure.

V.—McCoy Mine and Schist.

VI.—Geological Map of the Cortlandt Series.

VII.—Changes in Coat Colors of Mice.

VIII.—Changes in Coat Colors of Mice.

IX.—Changes in Coat Colors of Mice.

Text Figures.

Poikilitic Relation of Biotite and Olivine in Peridotite................... Tae

Diagram of the Relations of the more important Types of the Cortlandt

S12) 4 (ne en cae nmr ne Ana Ae oad acai dic boo oN Go 58

Variation Diagram) of the Cortlandté ‘Seriesiecicc civ sceeeeireieteiaie tele ieee 59

Relations of Spinel, Corundum and Magnetite in Hmery.................. 68

Sillimanite: Sehist... Vs nc5 4g wleseieionst eves nie aie aut ele alee sie etesiete ace nite arene etatetenrte 71

ANNALS OF THE NEW YORK ACADEMY OF SCIENCES

Vol. XXI, pp. 1-10, pll. -Il

Editor, EpMunp Otis Hovey

~THE RAVENSWOOD GRANODIORITE

Victor ZIEGLER

NEW YORK

PUBLISHED BY THE ACADEMY

10 May, 1911

THE NEW YORK ACADEMY OF SCIENCES

(Lyceum or Natural History, 1817-1876)

OFFICERS, 1911

President—FRranz Boas, Columbia University

Vice-Presidents—Guorce F. Kunz, Freprric A. Lucas,

R. 8S. WoopwortH, WILLIAM CAMPBELL

Recording Secretary—Epmunp Otis Hovey, American Museum

. Corresponding Secretary—HeEnry- E. Crampton, American Museum

Treasurer—EMERSON McMitt1in, 40 Wall Street

Librarian—RatrH W. Tower, American Museum

SECTION OF GHOLOGY AND MINERALOGY

Chairman—GeorceE F. Kunz, 401 Fifth Avenue

Secretary—Cuar.is P. Berkey, Columbia University

SECTION OF BIOLOGY

Chairman—Freperic A. Lucas, Brooklyn Museum

Secretary—L. Hussakor, American Museum

SECTION OF ASTRONOMY, PHYSICS AND CHEMISTRY

Chairman—WItLiAM CAMPBELL, Columbia University

Secretary—EDWarD J. THATCHER, Teachers’ College

SECTION OF ANTHROPOLOGY .AND PSYCHOLOGY

Chairman—R. 8S. WoopwortH, Columbia University

Secretary—F reprric LyMan WELLS, Columbia University

The sessions of the Academy are held on Monday evenings at 8:15

o’clock from October to May, inclusive, at the American Museum of

Natural History, 77th Street and Central Park, West

[ANNALS N. Y. Acap. Sci., Vol. XXI, pp. 1-10, Pll. I, II. 10 May, 1911]

THE RAVENSWOOD GRANODIORITE

By Victor ZIEGLER

(Presented before the Academy, 6 March, 1911)

CONTENTS

Page

Geoiogical Section of New York City.......... Stet Saar a nieve s: aha, cites Sey eke a!

Surface distribution of the Ravenswood granodiorite................... 2

PereOeLanine GeSCEItIOns ANG ANALYSES <2)... 60% 2 <0 sis wre oie socials vie o's euera bleleiwe 2

Mineral occurrences in the granoGiorite...............--2002 Pieler ae 3

Granite phaseé.........0.4+. yeigisniselseeeiessicces fh tae witieinve ainiat aMgldbel ommetioute 4

IDIGe JUPR OS 6 S53 ee he B06 DEOMI OE rer ODO LS BE Grerrer bir st - coc 6

Probable extent of the rock as disclosed by borings................-ee00% 7

Robi iaety fai OM OTT GLENS) aR yarere ais, « cldte Tale’ a orsicie.e lets Slee ig ele 6 «sis, o% ole arertie ele. sion £¢

Comparison with the Harrison granodiorite..... ween eee e eee eee e eens 9

GEOLOGICAL SECTION oF NEw YorK CITY

In the New York City folio of the United States Geological Survey,

the following succession of formations is recognized: Fordham gneiss,

Inwood limestone, called Stockbridge dolomite, and Manhattan schist,

called Hudson schist. In addition, there are several minor intrusive

masses, of which two, the Harrison diorite and the Yonkers gneiss, are

large. The former is located along the eastern side of Westchester County,

and the latter, immediately east of the city of Yonkers. To these two

large intrusives, we may now add a third, exposed along the western shore

of Long Island and first recognized by James F. Kemp,” who, in a paper

describing the section of the rocks revealed by a tunnel beneath the Kast

River, applied the name “Ravenswood granodiorite” to it. Subsequent

field work and the exploratory borings undertaken by various engineering

enterprises have proved the importance of this intrusive.

1The writer takes pleasure in acknowledging his obligation to Professor James F.

Kemp, and especially to Professor C. P. Berkey, for aid in securing data and for many

‘kind and helpful suggestions.

2Trans. N. Y. Acad. Sci., vol. 14, p. 273.

9 ANNALS NEW YORK ACADEMY OF SOIENSES

SURFACE DISTRIBUTION OF THE RAVENSWOOD GRANODIORITE

The western part of Long Island is covered by drift, so that exposures

of rock are rare, except along the shore of the East River and about the

Blackwell’s Island bridge. Here is found a series of scattering outcrops,

beginning near the corner of Boulevard and Potter Avenues and recurring

at intervals as far south as Fifth Street. North and south of the above

areas, drift masks everything, and the only clue to the underlying forma-

tions is afforded by a series of bore-holes which will be described later. In

the New York City folio, most of the exposures of this rock are mapped

as separate acidic intrusives in the Fordham, but the idea that these repre-

sented one large igneous body was not expressed. Since there are no

traces of Fordham gneiss found between the above exposures, and since in

a number of cases granodiorite was found between some of the outcrops in

the excavations for the foundations of buildings, it seems to the writer

that we may safely map this whole area as granodiorite. Thus we cut off

at once an area of at least a square mile from the old Fordham formation

and add it to the series of igneous intrusives mentioned above.

PETROGRAPHIC DESCRIPTION

The Ravenswood granodiorite is a light to dark gray rock, varying from

granitoid to gneissic in texture and ranging from a hornblende-biotite-

granite to a diorite in composition. Garnet is present in nearly all speci-

mens and in some cases is very abundant. For the purpose of establishing

a type composition for the rock, the following table has been prepared to

show the minerals present in a number of slides. A system of notation

devised by C. P. Berkey, in which X denotes an essential, Y an accessory

and Z a secondary mineral, is used.

ZIEGLER, THE RAVENSWOOD GRANODIORITE

Mineral Occurrences in the Granodiorite

E Pe eta |

es Semesters) ct | Oe | ses O8 lacs: el cee alley ol ee | et | os

Bee eS Oe | 4 ao ee Ss | eee VS:

iL alte allie 4 eer XE + | Veal Nea Ve EN iN Y

mee xe ee ‘cel Ae Nal A: ail avail celle sia] sc

3 Mee yl teal XO OX. SW | | Weave ave ease Y

4 5 ISG geld ecient 4 Valls ealnx y|y

5 RGRAY CVS Int [eh eH | ea Won ¥ Y

6 ar eee | Ne Me XY. ey Wwe |X Y

7 pKa RemeVen|eXG Cea seve WY lay. | YY Y

BRN fe: Y Nan en ee nee UID |) Me | oe | Y

ell ee | OE. EU Biull ail ev desl hs gd Y

10 pKa PONE lyse |) WAH EX: an oy tly yay Y

u SN ees tz SSE 3/8 x ye || ye | AY

12 SS Saal Wee Ns Alexa sad line be Y

Ieee eee ONAT Val ervanexeeie Sux: | | oy | OY. | x

14 CNS ohare calm 4 Vee | Ob aWa ie |W.

Nome re eae ese Vas ave eX |4aeel oxo ae | Wea) We |

16 Ru GY | Rh aR | IK SPH YG EY ea CY.

owe oll O° Ee 1a Mlle Nit cui al, aa al aa Ug Y |

18 > pal a Bd (dp. BEN eve Vly. Y |

19 Gu [eal i ail ip aleve Ny, «ONE || NG

ERCP aXe Ba eee Ye | Ys | Y

NNN SN ‘See. Carbonates.

NNN N

NNN NN N

A summary of these occurrences for each individual constituent is as

follows, F denoting the absence of the mineral from the slide:

a(cageteLeuehel se ivlleel ails eke © is)w alse). ¢ 0 \6..¢ = 10)4),8 6, a \¢ s.e:\0

Sie ele) 6, 8) y ae \e (oA tse eiieli= eh e).e)0 0) ¢ 0 ls «le se a alee

Pee SSeS Be ee eels a6 2S Se eee eee

Hornblende

Ow) af'v) 6) Nie elev ala fel ve: ui) 6 0.\0,'e) @,s 0c 6, 8 6) 8) 6 0 e610

Mi wiwEe ela) \s\ «leis (6 1s)\s)\s'/\ 0) wisi) «| 6 ¢ 6) ¢.c) oe) © 6) 0) 06 (610 eve © e's

«) wiella 69,0) iellsive) valle lele) es 66 © 6, 00 le)(e, m6 © 0/8.) 8 6 608

PGY:

Ss) ad

a) Ye

8 Y

OWE

0 Z

13 F

18 F

4 ANNALS NEW YORK ACADEMY OF SCIENSES

GreATT CLE aes oR ee este ele aio eern as 34 ON 15) 4 OZ 2F

ANU IG) bio ooo So AG e oa On nae Ob SOC ORenr 0x 20 Y¥ 0Z OF

FATE COON Ae 5 SAR As, SEP ee a nner 0X 20°¥ 0Z OF

JX ORUNIES Cot Glen bobbie S.cOGHee OO cao Me Ae ric OIe 0X 20-Y OZ OF

Ie Adt a ae ae en see Uc etvato sxc as ote siaeia eae 0x ons 0Z 13 F

Pye AOUIGS ee Breet civic slacole Fieie's"olayc's Oke mime 0x 5 Y 0Z 15 F

NIA OME Me Rete ines os olers a s&s. she cere evel siptas 0X ae 0Z ILS

(CanlOI Lester Bictar (eis oss okie 2th =e oes eves secs 0x Om UAE 8 F

TRSYOININ, 2 6.4 So Hons Rp OOO OR nee e ao nh orca .c 0X Ons 14Z 6F

From this summary, we see that quartz, orthoclase, oligoclase, horn-

blende and biotite are the essential minerals, and that garnet, titanite,

zircon and apatite are the common accessories, while microcline, anortho-

clase, augite, pyrite, pyrrhotite and magnetite are rarer. The acid min-

erals are dominant.

Granite Phase

The hand specimen shows a medium dark gray rock, evenly and finely

granitoid in texture and essentially composed of quartz, feldspar, horn-

blende and mica. Pink garnets of small size are present throughout the

specimen and give it a mottled appearance. The rock is slightly gneissoid.

The feldspars are fresh and semi-transparent and show polysynthetic

twinning. The hornblende is a dark green variety, sometimes almost.

black.

Under the microscope, great differences in the size of grains are no-

ticed, and there is also a tendency for the smaller grains of quartz and

feldspar to surround the larger ones. The small individuals seem to be

the result of the crushing to which the rock was subjected. Fractures are

common in the rock and are coated with limonite and in some cases with

quartz. The following minerals have been identified in the slide: quartz,

orthoclase, anorthoclase and hornblende as essential minerals, also micro-

cline, oligoclase, garnet, titanite, apatite, zircon, augite, magnetite and

pyrite. Secondary alteration is slight, although limonite and kaolin are

noticeable.

The quartz occurs allotriomorphically and also as small elongated

prisms included in the feldspars. The latter case show between crossed

nicols a fine rim of interference colors. Included in the quartz are zircon,

apatite, biotite and fine trails of dust. Some of the quartzes have deep

embayments and irregular outlines due to resorption. Nearly all show

wavy extinction.

The predominant feldspar is orthoclase. This is, as a rule, allotrio-

morphic. Microperthitic structure is faintly developed. Carlsbad twins

are common. Cleavage parallel to (010) is well marked, not so strong

ZIEGLER, THE RAVENSWOOD GRANODIORITE 5

parallel to (001). Several grains of microcline are present usually show-

ing, between crossed nicols, the characteristic “Scotch-plaid” structure.

The plagioclase is of the oligoclase type and is twinned according to the

albite law, although some patches are twinned under the pericline law in

addition. The bands due to the latter law are commonly the wider. In-

clusions of all the other minerals occur.

The hornblende is a dark green variety, in irregular shreds and grains.

Good cleavage is shown parallel to (110) and strong absorption along this.

The pleochroism is intense, ranging from dark olive green to light green

to yellowish brown. The extinction angle is 21 degrees.

Light green augite is present. It has weak pleochroism and an extinc-

tion angle of 36 degrees. Uralitization has developed, and the greater

part of the augite has gone over into green hornblende. This uralite

closely resembles the other hornblende in the rock in appearance and in

optical properties. Both augite and hornblende seem to be primary min-

erals, and the alteration of the augite seems to be the result of the pressure

which is also suggested by the wavy extinction and the crushed rims.

Brown biotite is present in small amounts, some of it as an alteration

product of the hornblende. It then spreads along the cracks towards the

center of the grains of hornblende. It is also present in small shreds and

in lath-shaped crystals and often occurs in beautiful aggregates.

The garnet in this type of the rock is light red in color, pale pink in

thin sections. It is associated with the darker minerals.

Magnetite and pyrite are usually present in small amounts. Titanite,

zircon and apatite are found in all slides.

Chemical analysis:

Per cent

IOS. pabidtcU deen dood dc OC UOT aS COO In OS renee enact 75.61

INUGH deabt tet pddacodd6c bo COD OC OC: Rene aera ices PER

HO ramyafet thterctarelater ete) ates to) cia Slee eraile esale 6. 6 eyelv'a7a\e aches whaleiere Pape

INZO) Gady wb bos 60H ODOR ECONO OD OURLOO Oe OCC eerie 86

WIEO) Seda AbGR OOS GUS OO DO EIEIO CIEE TCI aS 20

CRYO): Sr SPA b 6 OOS GOTO CHEE LET CNC MERC RO NERC IO a Re sees)

140) 54 A oid OO HIELO OI OC IO ECO DEIO OCI CISC IGOR Rao rae 3.18

INO Beererei evel totoneict ciererectetsversislcharste lens) shel seus erecta oie oS ielaverallerausls aya

IBLO) So Gada Uo OGOG Ec cOC 6 DOGO TIBCTEC DeICIC apn orion 6 09

101.25

Recasting into the component minerals we obtain

Per cent

(QUEIIA opts Ob COG obs DO RROOE DEO a ROA ODODE RO OU CO SOO 31.62

ORUMO CLASS eters rererai lore hiss 'exerelissole: orev stolavetet tev cues ober enctsteraterebere 16.68

6 ANNALS NEW YORK ACADEMY OF SCIENSES

Per cent

JNU OEIC HRCI AACS GOODE aD Oooo do DOA COU Bae .83

GIOSSUIATITC (oo ois oe Sesis arse ats cies 6 ate store ansttersiotelsteweletoceleters< 1.35

PAUTTIVATIGI UC ere roc. cire,s tavere. 8 sa)6 lo oi re.0. 8 ole we emecotoh Meno CeToE Peters 1.49

18 Ko) gc 6) (2000 (A Re AA RIA ACLONGTY cir, cow 0 oc. 8 ODI 2.83

SI OELES Bebb ate wi aval Sate ss ostere soa jo chek a elaievalel alevate enReeeaneter ates ke IU 7 Ee

13 Ginko eee eee nciccc moe asc bind bom odo ae 1.28

101.27

Diorite Phase

From this highly acidic type, the Ravenswood rock varies through inter-

mediate stages to a quartz-diorite. It is then darker gray in color,

medium in grain and slightly gneissic in structure. The mass of the rock

is made up of oligoclase, green hornblende, brown biotite and garnet.

Under the microscope, orthoclase, augite, zircon, titanite, apatite, pyrrho-

tite, pyrite, magnetite and kaolin were also identified. Here, as in the

more acid type, the rock is granitoid with the component grains irregular

in outline and interlocking. In size, they average about .8 to 1 mm. in

diameter.

Oligoclase is the predominent feldspar. Polysynthetic twinning is

shown, sections parallel to which have extinction angles of about seven

degrees. Pericline twinning and Carlsbad twinning also occur, and in

several cases all three are in combination. Strong wavy extinction is

shown. Inclusions of hornblende, biotite, apatite, zircon and pyrrhotite

are common.

The hornblende and the augite are of the same type as described in the

granite phase and show the same alteration. The alteration of the augite

is especially well shown. This takes place from the edges of the grains

toward the center, but in some cases the change has taken place through-

out the entire grain, the surface of which is spotted and blotchy from the

formation of hornblende.

A chemical analysis of the typical diorite rock gave the following

result :

Per cent

SLO Svc ec Se Rei alae kena rete re elie otto faledotalcle Creda, sit eheeyst MeeeL teas 64.98

IRV Olg'oi5's SOS OMS haga leter ore ehatic caterers oi shetene, ofele nuaite @, 6: sistBtal eis ce MRT LO

GLO), | (a ste evens eee eter tetotensie le 2 8 aratherasoraie enais/ 910,019 @ieleter stores 50

OO) io cos eee he eee aed Masa ore kouete of ava: oiate Sie lee reiait ah 2:00

1 Cee 0 RE PEED, 33.0) cy in ERD POS LOLA ONG 0 Sidon ene ns CRONE core RC 93

CSO ivkicis oo cient ee lordiole ie oh Pore iwlate sets Rists| ote s/eie fee 4.23

1. OEE INT 56 Bbc kin Scki0Ao OO GO COORG CMe 1.84

INGO cia tla Sin Pee erectus etA RE ese oo wae ots oie §.22

NS 2 0 RRR AR Ae RPE Cro BIND CIE cl CCLRC ROR ERR RRC Be 11

ZIEGLER, THE RAVENSWOOD GRANODIORITE Y

Recasting into the component minerals we obtain

Per cent

QUAL ee sel sis cece ereleveras UP aNoLWeliot oi cPepopsicieuerce petri hayare estas 18.48

Orthoclase 2.3... PS IDE CUO IE OCICS OS HOSS Sar 8.34

INVONIGY Bo aoe ae a iiss wal sueirairene tate raat avonten sreicnste mete lake Ai aren nisia lay wi gkertons 44.02

PATOL OL oe oo: avenici'ar «. die7s crave ehererenaysenetare oft SOD Oro este LOO

GROSSUIAT ICS Bie os}. s,5 aia si 5}. alone aa Mo ato a pee sie nora ct eel alc 2.25

PAUIMANGIC Pe os <6 cis cm sos We OMe OM eeu SaaS Oa os 3.48

Hornblende and Augite....... oud) aheh sitet @taiat elslevouelete Saas. 5 5 UePatl

LOGIE Mars crc, sess ats sie asi wl'a''s chore ata my oua oes aay a itueLa ane al onatias 1.05

Corundumyes ss. Sraiaee gen Sereisvers onernie SSUES ASS as 5 Oks

MASI CLILAM ye Face Lklc cis ssn we dd dnenddas A Wale a eet aces 70

PROBABLE EXTENT OF THE RAVENSWOOD GRANODIORITE AS DISCLOSED

BY BoRINGS

Aside from the actual outcrops of the granodiorite so far discussed,

records of it are found in borings undertaken in the drift-covered areas.

Thus, north of the exposures and slightly southeast of Lawrence Point

the Consolidated Gas Company of New York while drilling for water put

down an eighteen-hundred foot borehole. The entire core is the typical

Ravenswood rock. Again, south of the actual exposures, a series of bor-

ings* is now available from the corner of Myrtle Avenue and Gold Street.

Brooklyn, across the East River into Manhattan, as far as Hester and

Allen Streets. Several other drill records are also known as depicted on

the map. In all of these, the core is a massive gray rock, in all respects

like the exposed Ravenswood. It has the same appearance, texture and

mineral composition, and it seems io be a part of the same igneous mass.

Whether the rock is actually continuous so as to include all these areas,

cannot be said with certainty, but the conclusion seems probable, and the

corresponding extent has been so indicated on the accompanying map.

ORIGIN OF THE GARNET

Some of the cores brought up on the above line and located near the

intersection of John and Bridge Streets show a highly garnetiferous rock,

reddish-brown in color. The rock is finely granitoid. Under the micro-

scope, the following minerals were found: garnet and quartz as essential,

also some feldspar, which varied extremely in different specimens, and

%The corundum molecule in this and the following analyses is probably to a great

extent contained in the biotite.

4The borings and cores here referred to, samples from which were made available for

this study, belong to the series of explorations made by the New York City Board of

Water Supply along the proposed line of the distribution conduit for Catskill water.

8 ANNALS NEW YORK ACADEMY OF SCIENSES

some biotite, hornblende, titanite, apatite, pyrite and zircon. The garnet

occurs in small grains closely bunched and cemented by quartz. They

appear to be the crushed fragments of originally large garnets.

An analysis gave

Per cent

SIO EE oc iegition Acc ok een ee Rhee eee Pere rene 52.82

LUO as aa REMC a Orme SO e yo Goaoh do Sab aco 22.75

INMOM Sagenactodd duencedad soWoos op oodb don oaddaos ones HESS)

1G Oe anne errr a contra odbonda aed 9.50

MEOMs cc cist da woke eee PEER ERO EE REE eC EERE ROE eC EEE alicy

(O}:10 ea Aer Pe Ia a RIN AMa nS Sa deuutEA Gd Gon Acco Kdo gon. a6 % 7.43

EO) Bavaieldlerorraiolele sissere wise Misvaehe orele Va eeom et Rr Re ea oRtenener one 86

INEM OM AA Gam OSU nS OOOO GOODGIC OOOO OOOOH ON CHObORO Soo 0,0 123

13 0 Wee eo ac UA tot et aie oer eeRMine Bele Oecdtic Bicier e aaa 14

98.87

Recasting into the component minerals we obtain

Per cent

OUCHEWAL po OUdo O.6co GOS MA COD AIOA MMOS aU mG Ooaecdacke 22.44

@MUEEHOCIAS@ Mee ere eieierele 6 ie aiave. clsieelsrisl one toils ovens totetofoletensheneroteeonatte 2.78

PAUP ILO a aetene racers. ctiaFe a te ole vaia a\istele lo: l0 sys /o oe leucvav ese ehcreiel reletome tere 10.48

ANITORUOILE A Sagic-d Aguado GOO On Mae OO SOUS TOGO Ano OC 8.34

CUVOSSUM ETUC prise eh rare coiers jo te-s sieyatate,cxecsveystencieaeretovel steteronebetenenets 13.50

PANTING tasereretapevare ale letevs (cielictensre elerenerelclonehodeyabareneveteleNerotenetat= 14.94

IRICEN NMOS. Sa caocanncdocma os OQuNbUMsouUC Uda acodaggonoC PPS

IDOI oon Dro OCOD OO DMOGDaOOs AOU emo Oo Oaoaca Gonos aNdC 1.53

(OD TAULTD CULT a eeve reeves ove rave areye Ge aie se ote tencicks Concreeteredel cictietere ores tober .82

IGGIICRO NM GGpe ays cisrs aves ctorere & cheers ciclo eiterioiS ecto eke ter 2.08

Glancing over the analysis, we see that this variety of the rock is very

low in silica, as compared with the other two phases. It is high in

alumina, lime and ferrous oxide. The latter three constituents are in the

garnet, thus making it a composite of the grossularite and almandite

molecules. There is also a decrease in the amount of soda and potash.

On the whole, the rock has the appearance of a contact zone developed

from an aluminous and ferruginous limestone, which furnished part of

the lime, alumina and iron necessary for the formation of the garnet.

Or else the aggregate may be due to the absorption of a mass of impure

limestone in the granodiorite. Since Professor Berkey has shown that

there are interbedded limestones practically everywhere in the Fordham

gneiss, this explanation seems to the writer to be the best. According to

Van Hise,® combined contact and mechanical action furnish the most

favorable conditions for the formation of garnets, which are usually the

5 Monograph on ‘“‘Metamorphism,” p. 300.

ZIEGLER, THE RAVENSWOOD GRANODIORITE 9

result of the rearrangement of two or more adjacent minerals. It is diffi-

cult to say just what minerals formed the garnet occurring so extensively

throughout the granodiorite, but the following reactions are offered.

Anorthite + wollastonite = grossularite + quartz

CaO, A1,0;,2SiO, + 2(CaO,SiO,) == 3Ca0,Al,0;,3Si0, + SiO,

Anorthite + hypersthene = garnet + quartz

CaO, Al,O3,2SiO, + 2( (MgFe)O,Si0O,) = Ca0,2(MgFe)O,A1,0;,3Si0, + SiO,

Pyroxene == anorthite = amphibole-

(MgFe)0, (AlFe),0;,Si0,,2(MgO,Si0,) + 2(CaO, Al,03,2Si0,) = 2(CaO) (MgFe)0O,

pyroxene molecule + garnet + quartz

(AlFe),0,,3Si0, -++ 2(Mg0O,Al,0;,Si0O,) + 3Si0,

Starting with orthoclase, the following reaction might take place.

Orthoclase + pyroxene =r wollastonite + hypersthene +

K,0,A1,03,6S8i0, + 2( (MgFe) O,(AlFe),0;,Si0,) + 2(Ca0,SiO,)+ (MgFe)0,SiO, +

water—garnet — biotite “=

H,O=2Ca0, (Mg¥e)0O, Al,O;,3Si0,4-2((HK)O, (AlFe),0;,2Si0,)2((MgFe)O,SiO,)+

quartz

38i0,

Yielding in this last case a complex garnet, biotite and quartz.

COMPARISON WITH THE HARRISON GRANODIORITE

A comparison of the Ravenswood granodiorite with the Harrison grano-

diorite as described by H. Ries® was made for the purpose of determining

any relationship which might exist between the two rocks. Professor

Ries says, regarding the Harrison diorite,

“Throughout its extent, the rock has a pronounced gneissic structure. Strong

effects of folding and crushing with the consequent formation of ‘augen’ of

quartz and feldspar are shown. The rock varies from a more or less massive

gneiss . . . to a mica-schist. The minerals forming the granite are quartz,

plagioclase, biotite, hornblende, orthoclase, and in lesser amounts garnet, tita-

nite, zircon, apatite, muscovite, and microline.

“Quartz composes from two fifths to one half of the rock. It occurs in

grains and in rounded masses, filling the spaces between the other minerals.

The grains are often cracked and show undulatory extinction and zonal struc-

ture. Dustlike inclusions are often present and are arranged in more or less

parallel rows, which often extend across the cracks from one grain to another.

Intergrowths with plagioclase are not uncommon, especially around the edges

of the feldspar augen.

“The plagioclase is rich in inclusions of biotite, apatite, and zircon, but

¢Trans. N. Y. Acad. Sci., vol. xiv, p. 80.

10 ANNALS NEW YORK ACADEMY OF SCIENSES

smaller undeterminable ones are present in countless numbers. Some of the

plagioclases show microperthitie structure.

“Orthoclase is less abundant than the quartz and the plagioclase. Microcline

is rare. Biotite composes about one half of the rock. MHornblende is less in

amount than the quartz, its pleochroism is green to brown.”

Considering this description, we may establish the following points:

1. The rocks have the same mineral composition.

2. Texturally, they agree closely. Both are gneissic and show augen of

quartz and feldspar, crushed rims, interpenetration of quartz and feldspar.

undulatory extinction, etc.

3. Micro-perthite is present in both.

4. The hornblende shows the same pleochroism.

These points seem to bring out a strong similarity in the two rocks.

We must remember, however, that this similarity may well exist in rocks

of the same group which may come from widely separated localities.

These points of semblance are offset by the following contrasts:

1. Quartz is very abundant in the Harrison granodiorite where it forms

as much as one half of the rock. In much of the Ravenswood rock, it is

present in small amounts and in some cases is absent entirely.

2. The gneissic structure is present throughout the Harrison diorite,

but it is lacking in much of the Ravenswood rock.

3. No muscovite was found in the latter.

4. Biotite is far more abundant in the former.

5. No augite was described from the Harrison occurrence, and no

uralite was noticed.

Considering both of the above groups of points, it seems to the writer

that the Ravenswood is not a continuation of the Harrison granodiorite.

He believes that there are two distinct igneous bodies, of which one, the

Harrison, has been the more thoroughly metamorphosed.

PETROGRAPHIC LABORATORY, COLUMBIA UNIVERSITY.

Fic.

Fie.

W1e. 3.

Fia.

Fig.

PLATH I

PHOTOMICROGRAPHS OF RAVENSWOOD GRANODIORITE

. Alteration of augite (A) into uralite (Hb). Biotite (B) and feldspar

(F). Field, 1.7 mm.

. Shows the relation of biotite (B), hornblende (Hb), garnet (G) and

feldspar (EF). Field, 1.7 mm.

The garnet phase. Garnet (G), biotite (B) and quartz (Q). Tield,

1.7 mm. -

. Nest of biotite laths (B), with feldspar (F), quartz (Q), apatite (A)

and zircon (Z). Field, 1.7 mm.

. Microperthitic feldspar (F) and quartz (Q) between crossed nicols.

Field, 1.7 mm.

. The diorite phase. Hornblende (Hb) and oligoclase (F). Field, 1.7

mm,

Se co eee 4

‘

ee dads fhe

Pete

. poor

yote e he petiat inti at yene 2 :

, bi ed

i Se ee

ee Ree

remand in both? aa

abe ne elie?

-aretaoidoK a0 usrmansr bi ae a ;

cole Sits similar | Sea

‘ in bo (Q atitoit a, osileur ee Ca gitar te |

B ‘ace (oy scr amy bunks a) » stile

eae

ee 4

aan rca ‘igeblod itive rata sHitoid 0 Jea%

- 1 seria as port (Xo oF DBS

rn d ; 2 .

ato mo Wy ay ba ae) taebit sisson a 5 ot

> pome

on Gs Fach

ANNALS N. Y. Acab. Scr VOLUME XXI, Puate I

RAVENSWOOD GRANODIORITE

o be Wid ae ae

ae ETS A IT a ss

PLATE II

DISTRIBUTION OF THE RAVENSWOOD GRANODIORITE

Map showing the outcrops of the Ravenswood granodiorite in solid black;

the boreholes reaching it, in black circular spots, and its probable extent, in

broken lines.

Pe ry oe

i- = 3

ANNALS OF THE NEW YORK ACADEMY OF SCIENCES

Vol. XXI, pp. 11-86, pll. II-VI

Editor, EpMunp Otis Hovey

GEOLOGY OF THE CORTLANDT SERIES

AND ITS EMERY DEPOSITS

G. SHERBURNE ROGERS

NEW YORK

PUBLISHED BY THE ACADEMY

15 May, 1911

THE NEW YORK ACADEMY OF SCIENCES

(Lyceum or NaTuRAL History, 1817-1876)

OFFIcErs, 1911

President—FRanz Boas, Columbia University

Vice-Presidents—GnoRGE F. Kunz, Freperic A. Lucas,

R. S. WoopwortH, WILLIAM CAMPBELL

Recording Secretary—Epmunp Otis Hovey, American Museum

Corresponding Secretary—Hunry EH. Crampton, American Museum

Treasurer—EMERSON McMIittin, 40 Wall Street

TAbrarian—Ratrxu W. Tower, American Museum

SECTION OF GHOLOGY AND MINERALOGY

Chatrman—GerorceE F. Kunz, 401 Fifth Avenue

Secretary—CuHakLES P. Berkey, Columbia University

SECTION OF BIOLOGY

Chairman—Freveric A. Lucas, Brooklyn Museum

Secretary—L. Hussaxor, American Museum

SECTION OF ASTRONOMY, PHYSICS AND CHEMISTRY

Chairman—WIt1L1AM CAMPBELL, Columbia University

Secretary—EpwarpD J. THatcoHer, Teachers’ College

SECTION OF ANTHROPOLOGY AND PSYCHOLOGY

Chairman—R. 8. WoopwortH, Columbia University

Secretary—F rEDERIC Lyman WELLS, Columbia University

The sessions of the Academy are held on Monday evenings at 8:15

o'clock from October to May, inclusive, at the American Museum of

Natural History, 77th Street and Central Park, West

[ANNALS N. Y. Acap. Scr., Vol. XXI, pp. 11-86, Pll. III-VI. 15 May, 1911]

GEOLOGY OF THE CORTLANDT SERIES AND ITS EMERY

DEPOSITS?

By G. SHERBURNE ROGERS

(Presented in abstract before the Academy, 6 March, 1911)

CONTENTS

Page

GINS CaP MEN IG LOLA ete ora ati toe ole) s ‘aval Gials dro. oe Slat wore dela wile: eleveideeetaldaverkie 4 eletersiaeaee 11

MAES eterna all MEMENTO RT Pal eats ey Yo tel cts) & jo: 1s aitol'a,e' Guede ws ow wile, dude duel ayaraya'd anwel #dlens ausveveee 13

CORE IGM: Sato s deh A Ry 2 ae ee erie ee YW 19

. PURDIOENS \ clad aeons Ge Bt ae Gre ee ee ee aa an a OR 20

LAO MOCKS MERE ee Reger ace aerials alislevaliels a) e' bie ay due ia celeleis ace io elk meee eee 20

PPP URC BTS() CAGE MEEREY Ourers LTA EH SMe re anata ck add felts axe dele iae alehel 6 <va.ele eilanero sale dverehe ere 47

ANS OTEYORPITRIZT NS COOTOT REE BE OCIS 5 Oy Og a og 50

TV CUITS TOM SIM eRe ii che kere iio! 6 Aidlanisih Joeu eddlabecs slopate'elete cet atte: ole eta: eores 52

PSL tee OO LMS aN Aas ai real eho ndrore: chk uiletecai ysis oletns eisiah Gkyeld lo atlorauaere 54

Fe AMON SH Otst he LY PES, WAL MGaNAlVSES... cs sj./s4 2 bac ae %sle ce es eeene sila atc 57

EBDAG ESS: OI SELGY GIGI, OS CaO 6 Cad OS COA, SEA PIA re een nearer. 66

Petrography of the emery and associated rocksS.............0.00eeee 68

EV MESmrO ben OCIEMIEREN Cll celeRynternon ais <6 cic tavs dislvs. chavelaudte Viet a shevele alale otis waters 72

TAAL eRe SUNSET cio saree. Siche i clo stelalenacy a clordererels eleeve Sisch(ethawe at 77

STORERTITEIAS Woe Ble 6 Gadhe Chow OED aeRO oe aE Ree PE to et Ry 85

INTRODUCTION

The Cortlandt Series is a small but remarkably complete igneous

complex, lying to the south and east of Peekskill, N. Y., or about 35

miles north of New York City. It was so named by James D. Dana, the

first geologist to bestow upon it any particular attention, because of the

approximate coincidence of its boundaries with those of the township of

Cortlandt, the most northwesterly in Westchester County. More careful

work upon its extent emphasizes the roughness of its conformation with

1The writer desires to acknowledge his obligation to Prof. J. F. Kemp, who suggested

the subject of this investigation and under whose general direction it was carried out.

To Prof. C. P. Berkey, the thanks of the writer are due for valuable advice and many

suggestions generously imparted in the course of the work; and to Dr. M. A. Lamme for

assistance and advice in connection with some of the chemical problems encountered.

(11)

12 ANNALS NEW YORK ACADEMY OF SCIENCES

the limits of the township, but the work of Dana and the later and more

detailed investigations of George H. Williams have served to give the

series a rather wide repute under this name. This area is between 25 and

30 square miles in extent, as will be seen from the map. The same rocks

are found also in a small patch, less than a quarter of a square mile, on

Stony Point, on the west side ef the Hudson opposite the main develop-

ment; these have been rather carefully mapped by Dana.? James F.

Kemp has also described an extension of the series at Rosetown, N. J.,

where several of the types have been found; and Wm. H. Hobbs has dis-

covered two large and well developed patches in western Connecticut.

The present paper will be confined to the type development of the series

around Peekskill, which exhibits the greatest complexity of differentiation.

The Cortlandt Series has hitherto, according to Williams’s description,

been thought of as a confused aggregrate of basic rocks, in which the

norite type predominated. As a matter of fact, although the norites are

the most prominent type, examples of nearly every group from pegmatite

to peridotite have been found, with local developments of very peculiar

and abnormal rocks. It is thus much more complete than was formerly

supposed and would merit a thorough study from the standpoint of mag-

matic differentiation. The rocks never show a distinct gneissoid struc-

ture; they are massive and unsheared, although there are several excep-

tions to this which will be considered later. The series is surrounded on

every side by the Manhattan schist, inclusiornf§ of which are sometimes

found in it but which are easily distinguishable on account of their schis-

tose structure. Inwood limestone, underlying the schist, outcrops along

the river bank from Verplanck Point north to Lent’s Cove, and the inclu-

sions of limestone are also found in the igneous rock, but the main areas

are usually separated from them by the schist. Still other basic gneissoid

inclusions are found which resemble the gneisses of the Highlands to the

north. The whole Cortlandt Series, therefore, is a very complex and

intricate mass and presents a petrographical problem of rare interest.

There is no lack of outcrops in the district, but it is seldom that the

rocks can be distinguished in the field ; all of them except the granites are

dark pink or gray, and while the writer after some practice became able to

identify typical specimens, the microscope could alone be relied on. It

was deemed wise, therefore, to conduct the field work in a somewhat

unusually careful manner; from 25 to 30 specimens to the square mile

were collected in the areal work alone, and ten of these on the average

were sectioned. Notwithstanding this care, there is, of course, consid-

7Am. Jour. Sci., (3), XXII, 112. 1881.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES ie

erable room for error; and it is to be understood that the various bound-

aries are probably not always exactly where they are drawn on the map,

although never far away.

The massive character of the rocks precludes any topographic structures

of particular interest. The weathering of the rocks is like that in all

other such regions ; in the case of the olivine rocks alone is there anything

worthy of mention, and this will be taken up under their description

(Plate III, Fig. 1). The most striking valley in the district is that in

which lies the long curved tongue of schist, projecting north into the

series ; it is due, however, to a line of weakness along the contact and not

to the igneous rocks themselves. In like manner, Lake Meahach, a stag-

nant inlet, seems to occupy a similar, though much less pronounced de-

pression. Aside from these instances, the region is a succession of appar-

ently irregular hills and valleys, becoming to the east rather wild and

beautiful in a small way.

HistoricaL REVIEW

Considerable work has been done on this small district, and the evolu-

tion of our present ideas concerning it is interesting. Each of the earlier

~-workers expresses a different view, which have been regarded since Wil-

liams’s work as antiquated; yet it seems that some of these theories are

not entirely to be cast aside, after all.

It is a very curious and interesting fact that the emery in this district

was first seen and recognized by the first white man to ascend the river.

The following passage is taken from Henry Hudson’s report:* “The

Mountaynes looke as if some Metall or Minerall were in them. For the

Trees that grow on them were all blasted, and some of them barren with

few or no Trees on them. The people [Indians] brought a stone aboord

like to Emery (a stone vsed by Glasiers to cut Glasse) it would cut Iron

or Steele: Yet being bruised small, and water put to it, it made a colour

like blacke Lead glistering: It is also good for Painters Colours. At

three of the clocke they departed, and we rode still all night.” This trans-

action took place on his voyage down the river; and we read further that

he descended eighteen leagues below this and was driven by the Indians to

the shelter of a “Cliffe, that looked of the colour of a white greene.” This

was opposite Manhattan Island about two leagues below what is probably

‘Fort Washington Point; it was probably, therefore, the Castle Point

(Hoboken) serpentine. This would show that the emery was brought

3“The third voyage of Master Henry Hudson,’ New York Historical Society Collec-

tions, (1), I, 143. 1809.

14 ANNALS NEW YORK ACADEMY OF SCIENCES.

aboard about 54 miles (by the river route) north of this, or possibly a few

miles north of Peekskill.

W. W. Mather,* in 1848, described the rocks not as a series but merely

under the lithologic types of Westchester and Putnam Counties. Under

the caption “Sienite,” he states that this rock forms the east bank of the

Hudson River below Peekskill and is found in many of the hills southeast

of Peekskill. It is valuable and a durable building stone, although rather

hard to dress. It changes east of Verplanck to a black hornblende rock,

which may contain iron. Pits have been sunk for coal in this rock but of

course without success. Under “hornblende rocks,” he merely mentions

the dikes and stringers in the limestone at Verplanck. He evidently con-

siders the rocks adventitious occurrences of a syenitic type, making no

mention of the hypersthene, olivine, etc. and no suggestions as to their

origin. ;

Herman Credner,® in 1865, made the earliest identification of the

hypersthene which exists in many of these rocks. After describing it

briefly, he states that there is a gradual transition of the hypersthene rock

into syenite and hornblende schist. His work, however, was hasty and

cursory and unsupplemented by the microscope; and, since syenite is rare

in the district and hornblende schist unknown, his conclusions may be >

disregarded.

James D. Dana, in 1880, described these rocks in connection with his

work on the limestone belts of Westchester County, being the first geolo-

gist to recognize them as genetically related and to study them as a whole.

In his first paper,® he describes the rocks, classifying them as chrysolitic

and non-chrysolitic, and subdividing further into hornblendic, hypers-

thenic, augitic and biotitic rocks. He then discusses their origin. Since

they occur on Verplanck Point as dikes and stringers in the limestone and

schist, showing contact effects, he is led to believe that at one time they

had been molten, or at least plastic. From the considerable number of

schist and limestone inclusions in the western part of the district, which

generally lie approximately conformable to the strike of the main sur-

rounding areas of these rocks, he concludes that the Cortlandt Series is a

mass of old sediment, worked over by pressure and great heat ; these inclu-

sions would then be merely cases of unobliterated bedding. He finally

discusses the very sudden minor variations in the igneous rocks (which

are most frequent on Montrose Point) and believes that these also demand

* Geology of New York, pt. I, Geology of the First District, p. 528. 1843.

5“Geognostische Skizze der Umgegend von New York,”’ Zeit. der deut. geol. Gesell-

schaft, XVII, 390. 1865.

6 Amer. Jour. Sci., (3), XX, 194. 1880.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 15

this explanation, arguing that sediment may change suddenly in composi-

tion but that igneous flows are constant over large areas. For the source

of the heat, he postulates extreme local metamorphism of some kind.

In his second paper,’ he reviews the above arguments and proceeds to

the question of the origin and nature of the sediments. The rocks are so

very different in composition from the surrounding sedimentary rocks

that some peculiar local development must be postulated. There are

three hypotheses: (1) the material may have been contributed by the

Archean Highlands as detritus, or (2) as detritus supplemented by in-

gredients from the ocean, or (3) it may have been originally volcanic

ashes. While expressing no definite opinion, Dana appears to favor the

last theory, since successive outbursts might be very different in composi-

tion and yet come to rest in close juxtaposition. He suggests that the

voleanic source lay in the present bed of the Hudson, between Verplanck

and Montrose Points.

In his third paper,® he describes the phenomena exposed by the (then)

new railroad cut through Stony Point, which he considers entirely indica-

tive of the true igneous nature of the more basic members of the series at

least. He abandons, of course, all of the above theories; and this affords

an excellent example of the frankness, honesty and sincerity which char-

acterizes his hfe and work.

George H. Williams, at about this time, started a minute petrographic

study of the various types in the district, Dana having called his attention

to the wealth of variation exhibited. He did not, however, publish his

first paper until 1886.9 After reviewing Dana’s work, he announces his

intention of describing the types petrographically, beginning with the

most basic; then the contact metamorphism in the schist around the bor-

ders ; and finally of giving his theoretical conclusions and generalizations.

Starting with peridotite, he describes hornblende peridotite, devoting con-

siderable time to the schiller structure so often found in all the rocks of

the series, and then augite peridotite (picrite), giving an analysis of the

latter. He takes up each mineral, discussing its general habit and its

special peculiarities in these rocks, and often drawing valuable and schol-

arly comparisons with its occurrences abroad.

His second paper?® is on the norites, which, he says, are by far the most

abundant type in the district and cover the whole township east of the

™*Origin of the Rocks of the Cortlandt Series,’ Amer. Jour. Sci., (3), XXII, 103.

1881.

8 “Note on the Cortlandt and Stony Point Hornblendie and Augitic Rock,” Amer. Jour.

Sei., (3), XXVIII, 384. 1884.

9 ““Peridotites of the Cortlandt Series,’ Amer. Jour. Sci., (8), XXXI, 26. 1886.

10 “‘Norites of the Cortlandt Series,” Amer. Jour. Sci., (3), XXXIII, 135, 191. 1887.

16 ANNALS NEW YORK ACADEMY OF SCIENCES .

railroad. A glance at the map suffices to show the incompleteness of his

knowledge as regards the distribution. He discusses first norite proper ;

then hornblende norite, mica norite, augite norite and finally pyroxenite.

Under mica norite, he describes the “Butler Section,” a cliff in which

norite proper, augite norite and other rocks are arranged in layers. He

ascribes it to ordinary metamorphism. This and similar cases have been

described by the present writer’ and will be discussed in the paper here

contributed. Williams states also in this paper that the emery is a segre-

gation in the norite and discusses in this connection the composition of

the ore as compared with that of Ronsperg, Bohemia. In a later note,*

he gives an analysis of the orthoclase of these rocks, which removes the

suspicion that it may be merely unstriated plagioclase.

Williams’s third paper?’ is on the gabbros and diorites. He states that

the gabbro is a rather rare rock; but adds that it always shows metamor-

phism and always occurs adjoining the limestone, therefrom deducing the

fact that it is a hybrid formed by the action of the norite magma on this

latter rock. He divides the diorites into hornblende and mica diorite.

Under the caption hornblende diorite, he describes in turn brown horn-

blende diorite, hornblendite, green hornblende diorite and mica horn-

blende diorite. He subdivides mica diorite into mica diorite proper,

hornblende mica diorite, hypersthene mica diorite and quartz mica diorite.

In concluding, he states that the series is due to long-continued igneous

action, different types being successively produced which broke through

those already solidified, the more acid types apparently appearing last.

This concludes his petrographic study, and notwithstanding the incom-

pleteness of his knowledge of the distribution and extent of the types, his

work stands as a well-rounded and scholarly investigation. His nomen-

clature is particularly felicitous; in a district of such multifarious differ-

entiations, the free use of mineral prefixes is preferable to the adoption

of a number of new terms. It is greatly to be regretted that Professor

Williams’s untimely death prevented the further prosecution of his work

in the district; a complete study of the magmatic differentiation in the

Cortlandt Series would have been indeed valuable from a petrologist of

his experience and perspicacity.

Consonant with his original plan, having completed his study of the

igneous rocks themselves, he describes their effect on the surrounding

u “Original Gneissoid Structure in the Cortlandt Series,” Amer. Jour. Sci., (4), XXXI,

123: AOE

122 Op. cit., p. 243.

13 ““Gabbros and Diorites of the Cortlandt Series,” Amer. Jour. Sci., (3), XXXV, 438.

1888.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 17

rocks.1* He here states that the emery may be referred with certainty to

contact action on pre-existent material, apparently abandoning his former

view. In studying the other contact effects, he concentrated on several

linear sections, in the first two of which he describes the succession from

ordinary mica schist to the same rock carrying sillimanite, then to a rock

composed of sillimanite, mica and garnet, and finally to a garnetiferous

mica diorite. Analyses show that the change in the schist is in the direc-

tion of an increased alumina and iron content as the massive rocks are

approached. In the last section of the massive rocks on the limestone, he

finds that lime-bearing hornblende and pyroxene are formed along the

contact. He finally compares the effects here with those in several Euro-

pean localities, showing that a larger number of contact minerals is found

in the Cortlandt than anywhere else. The contact effects were thus shown

to be well developed, a point which will be reverted to in the discussion of

the genesis of the emery.

James F. Kemp,’® in 1888, described the Rosetown extension of the

series; he had also been over the main area under the guidance of Pro-

fessor Williams. This smaller body les about a mile west of Stony Point

and is itself about three quarters of a square mile in extent. It is sur-

rounded by gneiss and encloses a small patch of marble. The rocks them-

selves are all diorites, no hypersthene or olivine having been found in

them. Green augite occasionally assumes importance, but in the main it

is subordinate to the brown and green hornblende. Emery, similar to

that found in the main area, oecurs in these rocks. There is very notice-

able contact action along the borders, and numerous dikes radiate out

into the surrounding rocks. Six analyses are given in connection with

the petrographic description, three of the rocks and three of isolated

minerals.

William H. Hobbs'® has described the Connecticut extensions of the

series in an article which unfortunately is not particularly well known.

Two areas were found by him in the crystalline upland of western Con-

necticut, which resemble the original Cortlandt area both in the abun-

dance of norites and in the elaborate magmatic differentiation; these

constitute the northeasterly extension of the Cortlandt.

The Prospect Hill area, which is the larger and more important of the

two, covers about 40 square miles in the township of Litchfield, although

144“‘Contact Metamorphism produced in the adjoining Mica Schist and Limestone by

the Rocks of the Cortlandt Series,’ Amer. Jour. Sci., (3), XXXVI, 254. 1888.

15“QOn the Rosetown Extension of the Cortlandt Series,’ Amer. Jour. Sci., (3),

XXXVI, 247. 1888.

16 Festschrift zum siebzigsten Geburtstage von Harry Rosenbusch, p. 25, Stuttgart.

1906.

18 ANNALS NEW YORK ACADEMY OF SCIENCES

most of the types are found in a small patch about two miles square at

Prospect Hill itself. The igneous rocks are included in highly metamor-

phosed gneiss, quartzite and schist of pre-Cambrian and early Paleozoic

age, so that they are probably post-Ordovician. Along the border, there is

generally a mosaic of block faults. The rocks may carry considerable

percentages of pyrrhotite and chalcopyrite, which are exploited in many

prospect holes known locally as “nickel mines.” Hobbs describes the

rocks as varieties of the following types: gabbroitic. noritic, olivine-

hypersthene gabbroitic, dioritic, peridotitic, pyroxenitic and grano-dio-

ritic. Although his system of nomenclature differs somewhat from that

followed in the present paper, it seems that all of these types are counter-

parts of those found in thé main area. The most salient points of differ-

ence are the abundance of somewhat more basic plagioclase and of chal-

copyrite and pyrrhotite, with the absence of emery, in the Connecticut

areas. Five complete and valuable analyses are given, and the differentia-

tion is very suggestively summarized in a series of Brégger diagrams, In

the smaller, New Fairfield, area, the rocks are all of a dioritic nature,

resembling in this respect the Rosetown, N. J., extension.

From the description given, it is evident that the larger area at least

would warrant close study ; it is to be regretted that the boundaries of the

main districts, and the approximate distribution of the many interesting

types described, have not been more thoroughly worked out.

Charles P. Berkey,** in connection with an exhaustive study of the

geology of the Highlands and the region to’the south, has materially

added to our knowledge of the Cortlandt Series. He has been the first

geologist to map the boundaries; the writer has, of course, gone over the

same ground in connection with this present study, and the results are

entirely concordant. To Dr. Berkey also belongs the credit for the recog-

nition of the acid extreme of the series; the large granite area to the

northeast of the main body was overlooked by both Dana and Williams.

As will be shown later, there is every reason for considering this an inte-

gral part of the series; and if this be so, it completes the line of differen-

tiation from the most acid to the most basic, being thus an important

consideration from a theoretical standpoint. The writer is under great

obligation to Dr. Berkey for this as yet unpublished information.

This, then, constitutes the whole of the work done on the Cortlandt

Series. As will be seen, some of it has been superficial, some ill-directed

and some disproportionately minute. Kemp’s work on the Rosetown

17 “Structural and stratigraphic features of the basal gneisses of the Highlands,” Bull.

N. Y¥. State Museum, 107 (Geology 12). 1907. This paper is the first in connection

with this study and is not concerned with the Cortlandt Series.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES. 19

extension is complete in itself, as is that of Hobbs on the Connecticut

areas ; but as for the main area, the most recent work done on it—that of

Berkey on the boundaries—is what should have been the earliest. Owing

to this somewhat unfortunate reversal, there are several erroneous concep-

tions extant concerning the series; and the writer, while not pretending

to the experience of Dana or the erudition of Williams, hopes in the fol-

lowing pages to give a correct general notion of the rocks as a whole, leay-

ing to later investigators that part of the more specialized work which is

yet undone.

CORRELATION

The correlation of the series is, like that of many such igneous masses,

open to considerable speculation. Evidence of its age with respect to the

surrounding rocks is indubitable; in a number of places, marked contact

action is shown, and on Verplanck’s Point especially a number of igneous

dikes are intruded into the schist and limestone, so that the Cortlandt

series is unquestionably younger. ‘The lower limit, then, is dependent on

the age of the Manhattan schist and the associated Inwood limestone.

There is, however, considerable difference of opinion as regards the

correlation of these rocks. A. C. Spencer’® and his associates regard them

as the metamorphosed equivalents of the Cambrian and Ordovician, which

occur in an unaltered condition north of the Highlands. C. P. Berkey,?®

on the other hand, who has been mapping the Tarrytown and West Point

Quadrangles in the Highlands of New York, finds that the Manhattan

schist and Inwood limestone are separated by unconformities from both

the basement gneiss series below and the Poughquag quartzite above, this

latter representing the base of the Paleozoic. The schist and limestone

are therefore pre-Cambrian.

The Cortlandt Series has been intruded into these rocks and is there-

fore at least later than pre-Cambrian time. Moreover, since the schist

and limestone were strongly metamorphosed, probably in the Green Moun-

tain upheaval, the series must be post-Ordovician. There is no criterion,

however, by which we can fix the upper limit. Van Hise and Leith?°

make the following rather cryptic remark: “The rocks of the Cortlandt

Series (the clastics) of the original Taconic area and of the upper series

of the Adirondacks are of the same age, 1. e., Taconic, or Lower Cam-

18 A, C. SPENCER summarizes his own work and that of W. S. Bayley in New Jersey,

and the work of E. C. Eckel and F. J. H. Merrill in New York, in Bulletin 360, U. S.

G. S., “Precambrian Geology of North America,’ by Van Hise and Leith, p. 634. 1909.

12 Op. cit., pp. 361-378.

2 Op. cit., p. 319.

20) ANNALS NEW YORK ACADEMY OF SCIENCES

brian.” The writer is not aware of any clastics in the Cortlandt Series.

nor does there seem to be any justification for calling the series Lower

Cambrian. There is a general lithologic similarity with the upper series

of the Adirondacks, but the almost entire absence of metamorphism in the

Cortlandt Series, coupled with the surprising freshness of the olivine

present, are rather strong indications of a considerably later date. The

Triassic trap of New Jersey is so entirely different in structure that the

Cortlandt Series has probably no relationship with it: and from the litho-

logic character of later igneous masses and from the fact that no meta-

morphism has taken place in the East since the Permian, it is practically

certain that the series is not post-Permian. Jkemp’s study of the Rose-

town extension fully confirms this view; he finds** that the rocks here cut

the (Cambrian) Tompkin’s Cove limestone but have not disturbed the

Triassic sandstones. Hobbs’s work in Connecticut is also entirely con-

cordant. The whole expanse of the Paleozoic is, however, open; and

while nothing can be definitely said, it is the writer’s opinion that the

Cortlandt Series is younger than is generally supposed, more probably

late Paleozoic than early Cambrian.

PETROLOGY

In the following descriptions of the various types, the rocks will be con-

sidered in the order of their acidity, and frequent reference will be made

to the large geological map. In the case of the common rock minerals, at

least, an effort has been made to condense as far as possible the mass of

data which has accumulated from the description of the 260 slides exam-

ined and to render the optical descriptions reasonably brief and succinct.

In every case, the fullest description of the various minerals is given

under the important rocks in which they are the most abundant or of

which they are the most typical. Williams, in the papers cited above, has

sometimes given fuller and more minute optical discussions of the min-

erals than is here considered necessary.

PLUTONIC ROCKS

Granite

The granite member of the series lies to the northeast of the main body

and extends from south of the Crompond Road northward to Lake Mohe-

gan. It covers an area of about 3% miles, and throughout its extent it

appears to be very uniform in texture and composition. The exact bor-

2 Op. cit., p. 253.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 21

ders of the area are not, however, susceptible of close determination; a

heavy blanket of drift covers the district and obscures the geology. Out-

crops are few and are generally considerably altered, except when the rock

is artificially exposed. The land surface is on the whole rather flat for a

rock of this character, as compared to the hilly topography southward ;

this may be due to the fact that it lies in the lee of the Highlands, which

would eause an undue accumulation of till at this point, fading away to

the south. The granite is surrounded in general by mica schist, but to

the northwest lies an area of pinkish granitic gneiss, related to the typical

Highland gneisses.

As stated above, the credit for the recognition of this rock as part of

the Cortlandt series belongs to Dr. Berkey. He has as yet published

only a short note on the subject.2* Dana does not mention this area at

all; and since he apparently guided Williams over the country, the latter

likewise overlooked it. Closer work on the geology, however, reveals its

unmistakable relationship with the neighboring basic rocks. Its entire

lack of (megascopic) metamorphism separates it sharply from the sur-

rounding schists and gneisses and places its age as approximately the same

as that of the rocks to the south. It has undergone about the same de-

gree of weathering as have the basic rocks. Furthermore the latter are

frequently penetrated by aplite and pegmatite dikes, which are closely

akin to the granite; these are especially abundant in the northern part of

the basic area. There are, however, apparently no areas of granite in the

latter district ; the acidic flow seems to have been more sharply separated

than the various intermediate and basic facies of the magma. The pene-

tration of the basic rocks by the acidic dikes would seem to indicate that

the acid extreme of the series is the youngest; the chronological and

chemical relations of the various members will, however, be more fully

discussed at the end of the petrographical descriptions.

In the hand specimen, the granite is practically white when fresh,

being made up almost entirely of quartz and feldspar, with very subor-

dinate amounts of muscovite and biotite. The rock is generally weath-

ered at the surface to a faint dirty brownish color, owing to the formation

of epidote. The grain is medium, and the rock is very firm. A number

of specimens were taken from the surface outcrops; several from the

quarries, and one from a diamond drill boring at the 200 ft. level. The

thin sections show that the rock is remarkably uniform in composition,

Quartz, in large angular grains, constitutes in general about one-half the

rock. It shows numerous rehealed cracks but few inclusions. The feld-

2 “The acid extreme of the Cortlandt Series,’ Science, XXVIII, 575. 1908.

92 ANNALS NEW YORK ACADEMY OF SCIENCES

spar, which makes up perhaps two fifths, is generally orthoclase, with

varying amounts of microcline and plagioclase. The extinction angles

of the latter show it to be albite and oligoclase. In all of the sections

examined, the feldspar was considerably altered to kaolin, this being

much more striking than in the other members of the series. Epidote

also occasionally forms in small amount through the feldspar. In the

four thin sections of granite from the quarries, the alteration of the

plagioclase was especially apparent, as was also the case in the rock from

the drill core; in the surface specimens, however, the orthoclase seems to

be the more kaolinized. Several cases of zonal alteration in orthoclase

were noticed. Muscovite occurs in typical habit; the grains vary consid-

erably in size but are seldom as large as the quartzes. It is altered only

in the surface specimens; in these, it is hydrated and appears to yield

damourite. It is usually much more abundant than the biotite. The

latter is seldom entirely fresh and is often altered to chlorite. Horn-

blende is rare, and when found is generally altered. Magnetite is pres-

ent in very small quantity; apatite and zircon are practically lacking.

In the diamond drill hole above referred to, which was sunk to 425 feet

for water at a point half a mile east of Jacob’s Hill, the ordinary white

granite was first encountered ; this changed gradually to a somewhat finer

grained bluish gray-white granite carrying a larger percentage of quartz.

Through this, several streaks of a basic hornblendic segregation, a foot

or so across, were found.

At several points, the granite is quarried as a building stone. The

three cuts to the northeast are known as the Mohegan Quarry, the one

to the south as the Peekskill or Cornell. The latter is the largest; from

it was taken the rock out of which the bulk of the Cornell Dam across

the Croton River was constructed. The granite from the Mohegan

quarry is used as the chief building stone in the Cathedral of St. John

the Divine, in New York City, which when completed will be the fourth

church in size in the world. The stone has also been employed in several

other edifices in the city. While the alteration of the granite is consider-

able, as above described, it is not serious in character; and the rock com-

mends itself on account of the paucity of the dark minerals. The altera-

tion of the small amount of biotite has proceeded as far as it is likely

to; the muscovite and orthoclase are fairly stable minerals; and the

plagioclase is not abundant, while the large amount of quartz insures

the rock against serious alteration. Pyrite and other notably deleterious

constituents are entirely lacking. The rock is firm and easily worked,

although the abundance of joints of all sizes militates against the ex-

traction of large blocks. The actual amount of granite taken out varies,

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 93

of course, with the demand, but it probably greatly exceeds in value the

emery mined to the south.

Syenite.

Syenite constitutes a member whose areal importance is small and

whose existence even is not well defined. Three small patches were

found, two on the borders and one in the interior of the district. The

patch to the north lies between biotite norite and Manhattan schist, as

does also the small area on the southern border; the interior patch is sur-

rounded by hornblende norite.

In all of these areas, the rock is heavy and dark gray, appearing much

more basic than is actually the case. The rock in the northern patch is

very fine grained and micaceous (biotitic), although the mica has no

regular orientation. Some of the feldspar is evidently glassy oligoclase,

while much of it is faintly pinkish. The thin section confirms these ob-

servations; the feldspar is about half oligoclase, slightly kaolinized, and

half orthoclase in larger and more irregular grains, somewhat more

altered than the triclinic feldspar. Slight recrystallization and wavy ex-

tinction are apparent. Several small grains of quartz, giving unmistaka-

ble interference figures, and evidently original, are present. Biotite,

typically developed, is practically the only dark mineral to be seen; green

augite is present in very small quantity. Apatite is abundant in fairly

large crystals, and ilmenite is present in small grains.

In the southern patch, the rock is also fine grained, gray and mica-

ceous. The specimen sectioned was taken from a small quarry at the

juncture of the two roads and looks entirely fresh. In thin section, it

resembles the rock described above, except that the orthoclase is much

more abundant and that there is more quartz. The triclinic feldspar is

oligoclase, as above. There is also a moderate amount of deep green

hornblende and a little pyrite. The amount of alteration, however, is

surprising: the feldspar is somewhat sericitized, and there is a moderate

amount of chlorite and epidote and considerable calcite in small irregular

patches. This rock is evidently much like the rest, although it is a more

typical syenite.

In the interior patch, the rock is similar to the other megascopically,

except that there is less biotite. In thin section, it appears that the

orthoclase is predominant over the oligoclase but that what quartz there

is is secondary. Hornblende is the chief ferromagnesian mineral, but

alteration to scaly chlorite, starting from the center of the crystal, is very

common. The biotite is typical but somewhat altered and is usually

found clustering around the hornblende laths. Pyrrhotite is present in

small quantity and magnetite, ilmenite and apatite in moderate amount.

24. ANNALS NEW YORK ACADEMY OF SCIENCES

The syenites are thus a fairly well distinguished group, although that

of the northern area is related to the gabbros and that of the interior is

almost a monzonite. It differs from them chiefly in the prominence of

the orthoclase and in the considerable amount of apatite, this mineral

being present in very small quantity in the typical Cortlandt monzonitie

diorites. There is little of interest about this area, but the location of

the other two between mica schist and biotite norite is very suggestive,

taking into account the similarity between these two widely separated

areas. A discussion of the possibilities of contact action must be reserved

for another portion of this paper, but it may be noted here that if it be

possible for the molten norite magma to act upon the mica schist to form

a rock of normal igneous composition, this third rock would probably be

approximately syenitic, both in chemistry and in mineralogy. No analysis

of the syenite was made; but the analyses of biotite norite and of mica

schist, given below,”* may be compared. Williams** suggested the same

idea with reference to the gabbros, supposing that they always appeared

between limestone and norite. His premises are not altogether correct,

but the general idea can scarcely escape one who has worked on these

rocks.

Sodalite Syenite

Sodalite syenite was found in only one place, and then in a very small

place. It is interesting chiefly from a theoretical standpoint, as the only

representative in the series of the feldspathoid group. Its field relations

are somewhat peculiar. It is located at the north end of the road, imme-

diately east of Lake Meahach, on the contact with the inclusion of quartz

schist found at that place. Starting from the south, we have the sodalite

syenite (whose exact extent is indeterminable, but probably small), then

the schist, then a light colored augite monzonite in a strip about five feet

across, following by schist again. This is cut about 20 feet farther by a

much altered minette dike, three feet wide; beyond this the rock is

largely covered, and the exact contact of the schist on the north is hid-

den. It is thus impossible to say whether the sodalite syenite is a thin

contact strip surrounding the schist or not, although it is certain that it

does occur on the southern border. The significance of the augite mon-

zonite is difficult to perceive; it is not a dike and is therefore probably a

tongue of the country igneous rock projecting into the schist and partly

altered by it.

See pages 61 and 65.

% Amer. Jour. Sci., (3), XXXV, 440. 1888.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 25

The sodalite syenite is a very fine grained black rock, in which biotite

is the only mineral visible. It is somewhat brecciated and is considerably

altered. Under the microscope, the biotite is seen to be subordinate in

amount to a green hornblende, the crystals of which are often dislocated

across the cleavage, leaving cracks filled with an isotropic substance.

There is ‘little orthoclase, moderately altered, and no plagioclase. The

sodalite is fairly abundant, and is typically developed, in rounded iso-

tropic grains, n< 1.54. It has none of the structures of analcite, and

the latter moreover could hardly form in so acid a rock and one in which

the orthoclase is so slightly altered. Somewhat more abundant than the

sodalite is a mineral which seems difficult of identification. It occurs in

grains which are often hexagonal, having a relief of about 1.54 and a

medium (second order) birefringence. It is biaxial positive. Its hex-

agonal outline would suggest some derivative of nephelite, which would

be expected in this association, but it is not cancrinite or any of the other

common alteration products. Thomsonite usually occurs in a more

fibrous form, although except for this, the characters correspond rather

closely. The exact determination of this mineral, however, would hardly

affect the name given the rock, which stands as another suggestive occur-

rence of a zone on the border of the mica schist.

Diorite

The diorites are more important in the western part of the district,

where they cover an area of about two miles. Since Dana and Williams

worked chiefly in this section, the diorites appeared more important to

them than they really are, for, except for this area, they appear only as

small isolated patches. here is one at Pleasantside and another about

half a mile southeast of that place; three along the southern border of

the series, and a confused mass on Montrose Point. On Stony Point,

across the river, they are found. There is also a garnetiferous phase a

mile and a half east of Pleasantside associated with the emery; and dio-

rites, often very micaceous, constitute the immediate wall rock in several

of the emery cuttings on the hill just east-northeast of that village.

As stated above, Williams subdivides the diorites entirely on the basis

of their dark minerals, arriving thereby at brown and green hornblende

diorites, mica-hornblende diorite and mica diorite. He states in this

connection that the brown hornblende diorites tend to pass into norites

and pyroxenites, while those with green hornblende show aspecial affin-

ity for the mica-bearing rocks. These observations are quite correct ;

there is every gradation from a diorite whose sole ferromagnesian mineral

26 ANNALS NEW YORK ACADEMY OF SCIENCES

is green hornblende, to one which carried biotite alone, and in the field,

moreover, the two types are very intimately related. The whole of the

large area of diorite is of this type, except in two places along the contact

of the diorites with the pyroxenites, where the hornblende is brown, the

rock carrying biotite as usual. All of the smaller areas of diorite carry

green hornblende excepting two,—the patch just southeast of Pleasant-

side, which appears to grade into the associated hornblende norite, and

the northern part of the area southeast of Salt Hill, which appears to be

closely associated with the neighboring biotite augite norite, both carry

the brownish variety. This latter patch, however, shows green horn-

blende along its southern margin. Willams’s conception of the affinities

of these two main types, derived from the study of a small portion of the

district, appears to hold therefore for the whole series.

On Montrose Point, the brown hornblende is most plentiful, both in

the hornblende pyroxenites shown on the map and in associated diorites

which are not mapped. The extraordinary complexity of the rocks on

Montrose Point makes the geology impossible of adequate representation,

and it was deemed advisable, therefore, to map merely the most abundant

rock, the pyroxenites. To quote from Williams:*° “These rocks (horn-

blendites) have a glistering black color and are most intimately asso-

ciated with the norites, hyperites, diorites and pyroxenites which also

occur there. No more complicated interpenetration of eruptive rock-

types could possibly be imagined than is displayed at this locality—every

rock includes and forms dikes in every other; and at the same time, every .

type passes by gradual changes in its mineralogical composition into

every other one.” It is here, then, that the brown hornblende is best

developed; and in this wonderfully complicated net-work of rocks, it

grades on the one hand into a biotite augite norite and on the other into

hornblendite and hornblende pyroxenite. In the latter rock, it is espe-

cially well developed and often ar gieedanes basaltic hornblende in color,

birefringence and relief.

An attempt was made by the writer to classify the diorites on the basis

of their feldspar, into monzonites and diorites. In most of the diorites,

the unstriated feldspar constitutes from one third to two thirds of the

whole. Williams states that on a number of tests made on the feldspar,

however, the specific gravity ran from 2.648 to 2.67, which would show

it to be plagioclase of the oligoclase-andesine series. Moreover, the re-

casting of the subjoined analysis of diorite shows that all of the potash

must be in the biotite, although the rock would be called a monzonite,

2 Amer. Jour. Sci., (3), XXXV, 441. 1888

ROGERS, GEOLOGY OF THE CORTLANDT SERIES OY

if all of the unstriated feldspar were considered orthoclase. The peri-

cline twinning is usually, of course, very faint, and much of the osten-

sible orthoclase is probably plagioclase twinned according to this law.

While this analysis shows an exceptionally low percentage of silica,?° it

indicates that the identification of the orthoclase must be attended with

great care and that a subdivision by this criterion would be very hazard-

ous. At the same time, microperthite was noticed in one slide and quartz

in another, so that orthoclase is probably often present.

The diorites in the hand specimen vary greatly in appearance. They

are generally of medium grain, although they vai,;.from almost felsitie

to an extremely coarse rock at Crugers in which the »ornblendes are six

inches long. The transitions in texture are very sudden, although there

appears to be no concomitant change in composition. The feldspar is

usually white, and the rock is then distinguishable in the field, but it is

often a dark gray. The hornblendes are occasionally in slender laths,

though usually in poorly defined grains.

The microscopic features have already been largely covered. The

characteristics of the feldspar have been noted; the plagioclase varies

from oligoclase to andesine, although it may be labradorite when the rock

carries hypersthene. ‘The hornblende is usually green with strong ple-

ochroism: X olive green, Y brownish green, Z brownish yellow, though

X may occasionally become bluish green. It is rarely in laths and never

so when it is of the brown variety. The green variety especially often

carries delicate parallel inclusions, presumably of ilmenite, which re-

sembles somewhat the partings of diallage or enstatite. It alters to chlo-

rite and epidote, and in one instance secondary biotite seemed to have

been formed. The biotite is usually of a deep brown color, apparently

high in iron and showing excellent pleochroism. It may be almost want-

ing and may again almost replace the hornblende. In basal section, it is

nearly opaque. The optical angle is extremely small. It is usually asso-

ciated closely with the hornblende, and they appear to have been crystal-

lized simultaneously. Epidote, apparently original, was found in one

slide as figured by Williams.?7 Quartz is rare but occasionally occurs in

small grains. Magnetite and ilmenite, this last usually the more abun-

dant, occur in ordinary quantity, and pyrite and pyrrhotite are not

rare. Apatite, while sometimes abundant in large crystals, is usually

notably less than in the norites. Garnets are abundant in the locality

noted above.

26 WILLIAMS (Op. cit., p. 444) states that an average of four silica determinations was

53.94 per cent. The particular rock analyzed by the writer is probably abnormally basie,

although, under the microscope, it appeared quite typical.

2 Op. cit., p. 445.

98 ANNALS NEW YORK ACADEMY OF SCIENCES.

The chief transition of the diorites is marked by the entrance of hyper-

sthene and enstatite, which leads to the hornblende norites with the de-

crease of biotite and to gabbro and biotite augite norite, with the entrance

of augite. The latter, however, is rare, and in only a few of the diorites

was either of these minerals noticed. In one case, an approach to a

diabasic texture was observed, but this reversal of the normal order is

uncommon.

The alteration of the diorites is a very salient characteristic. Aside

from the ordinary weathering to chlorite, kaolin and epidote, sericite was

noticed in several specimens along the schist contact at the north end of

Lake Meahach. In these rocks, moreover, and in those taken from simi-

lar positions, a very perceptible amount of strain was noticed. The feld-

spars showed pronounced mortar structure and wavy extinction, and the

biotite was twisted and bent. The same phenomena are apparent in less

development in most of the diorite slides examined; either because of

their location, or on account of an inherent weakness in the rock itself,

the metamorphism of this type is more evident than it is in the case of

any other member of the series.

Gabbro

Gabbro is somewhat akin to the last mentioned type, but unlike the

diorite it is areally unimportant. Only one important area was found,

situated on the northern border just to the east of the syenite area there

and lying between the biotite augite norite and the mica schist. Its posi-

tion is thus analogous to that of the syenite, the accompanying norite here

carrying augite, whereas that which adjoined the syenite carried merely

biotite. Moreover, the syenite carries a small quantity of augite, which,

with other characters emphasized below, would seem to indicate its close

genetic relationship with this gabbro. The syenite and gabbro very pos-

sibly adjoin ; but since no outcrops are available for a space of half a mile

between them, it was deemed best to map them as found.

The rock of this area is dark gray and micaceous, closely resembling

the syenite, although the short stout augites may be distinguished on close

examination. The grain is medium fine and the rock quite fresh, al-

though well jointed. In thin section, it appears that the plagioclase is a

basic andesine, but that nearly a third of the feldspar is unstriated and in

rounded irregular grains, indicating orthoclase. The biotite is abundant

and characteristic, with pleochroism from golden yellow to brownish

black. The augite is green, non-pleochroic, and shows extinction angles

from 45° to 50°. In basal section, it shows cleavages at 89° and gives a

good positive axial bar. Only one grain of hypersthene is visible. Apa-

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 29

tite is not very abundant. Several small grains of what seemed to be

quartz were seen.

A small area of a very simple gabbro lies in the extreme southeastern

corner of the district, where it is intimately associated with diorite and

biotite augite norite. The gabbro phase is composed entirely of augite

and feldspar. This rock, or more probably a mixture of the three, was

used for the basement work of the Cornell Dam across the Croton River,

but while the crushing strength was all that could be desired, the abun-

dance of chloritized joints was found to affect seriously the value of the

large blocks required, so that the engineers were finally forced to employ

the granite described above. A large quarry, however, was opened, and

most of it still remains above the level of the lake.

Another type of gabbro is found, however, which is very different. It

occurs in two localities adjoining schist inclusions; one at the north end

of Lake Meahach and the other just west of Montrose. In the field, it

closely resembles limestone, being gray and homogeneous in appearance

and weathering in layers. Under the microscope, it appears that the

rock is made up chiefly of two minerals, feldspar and augite. The latter

is brownish but is non-pleochroic and otherwise typical. Biotite, apatite,

ilmenite and titanite are present in small amount.

The most extraordinary feature of the rock is the amount of shearing

which it has undergone, which far exceeds anything noted elsewhere in

the series. The augite is distributed in bands, in some of which the min-

eral is actually granulated, and these surround “augen” of broken augite

and feldspar. This lenticular effect is visible in the specimen, especially

when polished. The whole rock resembles a badly crushed gneiss, rather

than any of the other members of the Cortlandt Series, and it may pos-

sibly represent one of the inclusions of Highland gneiss which are found

elsewhere in the district. Since, however, it seems to be associated with

the schist inclusion, it is probably a true igneous rock, from the strained

and crushed zone which might be expected to border such a foreign mass,

and movement of the latter just before the consolidation of the igneous

rock may have contributed to this extraordinary crushing.

Norite

The norites are perhaps a trifle more abundant than any other group,

although they do not, as Williams and Dana supposed, constitute the

great bulk of the series. The general norite magma exhibits several

facies, which are distinct rock types and which yet often pass into one

another by insensible gradations. That the classification made upon a

30 ANNALS NEW YORK ACADEMY OF SCIENCES

petrographical basis is not academic is evidenced by the areal distribu-

tion—the various types fall naturally into distinct areas. Their general

distribution is suggestive, for although occasional isolated patches occur,

practically the whole of the norite magma, in its various ramifications, is

confined to the central part of the Cortlandt area, being flanked on the |

west by diorites and pyroxenites and on the east by the latter. At the

north and south, they are often separated from the mica schist by sye-

nites, diorites and gabbro; and if the boundary could be accurately traced,

it is very possible that these or similar rocks would be found continuously

along the margin. .

True norite, which is composed merely of hypersthene and feldspar,

occupies a very peculiar position in the series. It does not appear on the

map, since it never covers an appreciable area; it occurs always as inclu-

sions in the other members of the norite family and generally, if not

always, in those containing biotite. These inclusions are sometimes

streaks and sometimes rounded flow-like patches, and this imparts a very

peculiar appearance to the norites, especially in the northwestern corner

of the area. The inclusions vary in size from a few inches to possibly

forty or fifty feet across, though the smaller ones are the most common.

Absolutely pure specimens are rare, inclusions in which augite and biotite

appear in insignificant amount being the general thing. This pure norite

is found frequently in the district between Spitzenberg Hill and Peeks-

kill, especially along the post-road to the west; at Pleasantside again it is

well developed, and striking cases were also found (see Plate IV, fig. 2)

in the woods a mile south of Spitzenberg. It may occur anywhere in the

biotite-bearing norites, however, and the above cases are merely a few

examples.

The rock is easy to distinguish in the field, since it is always fine

grained. Its color is usually pale pink, owing to the amount of plagio-

clase present, and scattered through this are the black hypersthenes. As

the grain grows coarser, a little biotite and augite may enter, and the

rock conceivably grades into the more complex and coarser norites.

Under the microscope, it appears that the pink color of the plagioclase

is due to the presence of a fine reddish dust, which Williams has figured

and discussed at some length.?* Under a high power, this dust resolves

itself into plates, rods and globulites, presumably of hematite. The

plates may be hexagonal, rectangular or irregular, and they vary in size

up to .01 by .04 mm. The rods are often arranged in rows, forming a

kind of discontinuous needle, but these exhibit no parallelism to any

2 Amer. Jour. Sci., (3), XXXIII, 141. 1887.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 31

structure of the feldspar. The globulites are extremely small and are

scattered promiscuously over the crystal, except at the border and on the

margins of the larger inclusions (7. e., the rods and plates, or crystals of:

magnetite) ; and this clear margin moreover varies directly as the size of

the inclusion. This would indicate that they represent the form in

which the iron of the rock first crystallized, many of them combining

later to form the larger inclusions. The inclusions may be gray in color,

then imparting a dark gray color to the feldspar, and they may be so

plentiful that the mineral even in thin section is gray. They are de-

scribed here at some length, because this reddish dust is very character-

istic of the plagioclase in all of the norites and separates it sharply from

that in the gabbros, diorites, syenites and granites. The feldspar in this

rock is usually in a mosaic of small grains, the lamellz of the plagioclase

occasionally showing strain. The variety is usually andesine, as deter-

mined optically (angles 18°-20° on sections perpendicular to (010) and

by specific gravity (2.674, according to Williams). The presence in con-

siderable amount of orthoclase is attested by the analysis and by specific

gravity determination, but the proportion varies greatly in different

specimens. Carlsbad twinning and zonal growth have been noticed.

A description of the hypersthene had best be reserved, since in the pure

norite it is not altogether typical. The pleochroism is generally some-

what fainter than usual, being from pink to light green. The crystals

are usually small and arranged in clusters, and inclusions are few or

wanting altogether. Magnetite and ilmenite are usually present in

smaller amount than in the other norites. Apatite is fairly well devel-

oped. Williams also mentions garnet and pleonaste in specimens collected

at an emery cutting, but these minerals are associates of the ore and are

usually not found elsewhere. A peculiar variety of this rock, in which

the hypersthene has a rather remarkable habit, is found only at the emery

mines and will be described below.

Biotite Norite

Biotite norite is one of the three more important members of the

norite family, covering perhaps three and one-quarter square miles. It

occurs in two large areas and at least three small ones. In the area back

of Montrose Point, a number of small emery pits have been opened. This

patch may possibly be connected with the larger one to the east of the

railroad. One of the other areas is in biotite augite norite in the south

central part of the district and the other in pyroxenite in the extreme

eastern end. It is thus always associated with the other norites, as stated

32 ANNALS NEW YORK ACADEMY OF SCIENCES

above, and shows particular relationship with the biotite augite variety.

Williams records it from only one place (along the road to Montrose

Point), which is so small an area that it cannot be mapped.

In the hand specimen, the biotite is, of course, distinguishable, but

these rocks cannot be told from those which carry also augite. The

feldspar often imparts a dark pink color to the rock, but it is as often a

dark gray. The grain is medium, being never very fine or very coarse.

A faint gneissoid structure is occasionally apparent, but the rock is gen-

erally quite massive.

Under the microscope, the feldspar resembles that described under

norite proper. The reddish dust is usually, but not always, visible. The

andesine carries it most frequently; when the plagioclase becomes labra-

dorite, the inclusions are more grayish. The lamelle are sometimes bent

or broken by strain, and secondary twinning is also sometimes induced,

but such a degree of metamorphism is rare. The orthoclase may be al-

most entirely wanting, or it may again make up one third of the feldspar.

It sometimes becomes quite gray from inclusions.

The biotite is present in typical habit; it seems to be a somewhat less

ferriferous variety than is characteristic of the diorites, and the axial

angle is larger. Its pleochroism is X golden yellow, Y dark brown, Z

brownish or greenish black. Rarely the colors are lighter than this. The

biotite is an especially delicate indicator of the amount of strain which

the rock has undergone, but, as stated above, any marked amount is rare.

The hypersthene is, of course, the essential constituent of the whole

norite group. It usually occurs in stout, rounded prisms, which are idio-

morphic unless the ferromagnesian minerals are unduly crowded. In the

finest grained rocks, the crystals are small and often clustered together,

and even in the coarser varieties, the hypersthene is sometimes present in

large aggregates of small irregular grains. Sometimes there is a ten-

dency for the ends to fray out, as it were, and alteration then begins at

this point. The depth of color and pleochroism vary directly with the

amount of iron. Enstatite rarely occurs in the norites; bronzite, with a

fains pleochroism, is more common; but hypersthene, of varying degrees

of pleochroism, color and relief, is the typical orthorhombic pyroxene.

X varies from deep red to reddish yellow or pink; Y from yellowish

brown to dirty yellow or pale yellow, and Z from bright to pale green.

Extinction is, of course, parallel. In basal section, it shows the charac-

teristic crossed prismatic cleavages parallel to (110) ; the pleochroism is

then rather faint, but the figure of the obtuse bisectrix distinguishes the

mineral from monoclinic pyroxene. The fine parting of hypersthene

parallel to (010) is usually well developed, and the characteristic ilmenite

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 33

inclusions are oriented with respect to this. These inclusions are prac-

tically always visible with the high power objective, and they often be-

come very large. They are usually reddish and are seldom opaque, al-

though when abundant they may almost obscure the pleochroism of the

mineral. By reflected light, they exhibit an excellent metallic luster,

resembling that of phosphor-bronze. They are in many cases recti-tri-

angular in shape, but sometimes four-sided and often irregular. Altera-

tion of the hypersthene is fairly common, usually to bastite, which is

then geometrically oriented on the altered pyroxene. Alteration may

begin at the ends, or the bastite may be flecked over the surface of the

erystal. Its color is yellowish green, and it practically lacks pleochroism.

Uralite may occasionally form on the hypersthene; it is distinguished

from the bastite by its pleochroism, deeper color and more fibrous habit.

The opaque mineral in these rocks is chiefly ilmenite, in large irregular

masses, magnetite being usually markedly subordinate. Pyrite is not

uncommon, but pyrrhotite is rare. Apatite occurs in all quantities,

usually in small, well defined rods, but sometimes in large crystals, sev-

eral millimeters long. The crystals show a sharp hexagonal outline in

cross section, with frequently a member of fine parallel rod-like inclu-

sions—so fine in fact that they resemble the parting which is sometimes

developed in corundum parallel (1011). The figure is the same in the

two minerals, and the relief and birefringence seem to be higher in these

apatites than usual. In one roek in which the identity of the mineral

was doubtful, a test for phosphoric anhydride was made, but the high

percentage (2.45) removed all suspicion of its being corundum. ‘This

latter mineral was not noticed in any of the normal igneous rocks of the

series.

In several slides, hornblende (usually light brown) was present in sub-

ordinate quantity, marking a passage to the biotite hornblende norites.

Augite may also enter, to indicate the transitional phase to the biotite

augite norites, but biotite norite appears to be a fairly constant type.

Biotite Augite Norite (Hyperite)

Biotite augite norite is the most important of all the norites, and from

its central position it gives the impression of being the fundamental

norite magma. There are three large areas, two of which are probably

connected as mapped; two smaller ones, one on the west border and the

other on the eastern, and a third located in the pyroxenite area. It also

frequently occurs in the complex on Montrose Point, associated chiefly

with the pyroxenites. Williams’s specimens are taken mostly from this

34 ANNALS NEW YORK ACADEMY OF SCIENCES

locality, although he mentions the western portions of both of the other

areas.

This rock closely resembles the biotite norites, being medium grained

and either dark pink or dark gray. One or two light pink specimens

have been taken, but this variety is uncommon. ‘The rock seldom shows

any metamorphism.

The feldspar is identical with that in the biotite norites, the orthoclase

being present in about the same quantity. The plagioclase is usually

andesine, but in one of the light pink specimens mentioned above, it was

oligoclase, and when the rock is dark gray it may be labradorite. ‘The

hypersthene and biotite occur as described under biotite norite. Green

augite is the other essential mineral in this rock, and it appears to be

closely related to the hypersthene. In shape, the crystals resemble each

other strongly, and the augite often carries inclusions identical with those

occurring in the hypersthene. These are clustered in the center of the

crystal and are often so abundant as to render it dark reddish brown, but

the margin usually is free and retains its clear green color. The color

is about that of the Z ray of the orthorhombic variety. In many cases,

an intimate intergrowth of augite and hypersthene was observed, a patch

of pleochroic hypersthene appearing in an augite crystal, with the cleay-

age and inclusions coincident. The augite may, of course, be distin-

guished by its lack of pleochroism, its extinction angles of 40°-50°, its

higher birefringence and its interference figure in basal section. Both

simple and polysynthetic twinning were occasionally noticed in the same

region, although these are both more common in the pyroxenites. Altera-

tion is similar to that of the hypersthene, except that chlorite and horn-

blende are the common products.

The ferromagnesian minerals constitute in typical specimens from

one-third to one-half the rock, but their mutual proportions vary consid-

erably. The hypersthene makes up usually about one-half, with the bio-

tite and augite in subecual amounts; frequently, however, these will ex-

ceed the hypersthene in quantity. Original hornblende is rather rare.

The apatite may become very abundant in this rock; in one case, the

little rods were so plentiful in the orthoclase as to constitute one-third

or more of its bulk. Ilmenite is much more abundant than magnetite,

and pyrite and pyrrhotite are not uncommon.

The order of crystallization of the ferromagnesian minerals is some-

times hypersthene, augite and biotite, and sometimes hypersthene and

augite, and then biotite. The orthoclase usually crystallizes last of all.

Williams does not recognize this member as a distinct variety; he groups

all of the augite norites together. That the subdivision is a rational one,

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 35

however, is indicated by its areal distribution; and, as remarked above,

it seems to be at once the most common and the most resistant member

of the norite family. It is itself a transition phase from biotite norite

to augite norite; it rarely shows relations with other types.

Quartz Norite

The remarkable rock, quartz norite, was found in only one area, at the

extreme southern point of the series, where it lies between biotite norite

and mica schist. This and similar quartz-bearing rocks were also found

in the emery mines, but the peculiar conditions which undoubtedly gov-

erned the formation of these deposits tend to discredit the occurrence of

such rocks over larger areas, and they will therefore be described in the

economic section of this article. The patch in question might perhaps

be relegated to the same category, as purely a contact development; but

since there is no emery present, and since quartz is known to occur in

other gabbros as a result of purely igneous”® action, it is thought best to

notice it here.

The rock in hand specimen is of a basic, micaceous, brownish gray ap-

pearance. In the slide, the feldspar is seen to make up about one-third

‘of the rock, and about one fifth of it is unstriated. It is distinctly gray

from inclusions and shows somewhat wavy extinction. In the case of

the plagioclase, the twinning is often irregular. The quartz is present in

‘considerable quantity and appears to be undoubtedly original. It is in

fairly large grains, which fill the interstices between the other minerals.

It is often cracked, and these cracks are only occasionally rehealed. The

hypersthene constitutes about one-quarter of the rock. Its color is deep

and its pleochroism strong, and it carries numerous inclusions. It is

altering to bastite and apparently to hornblende also, in places. Only a

few grains of the typical green augite were noticed. The biotite makes

up about one-quarter of the rock; it is typical and somewhat altered, so

that it has a greenish tint. The hornblende is largely secondary, but one

or two (deep green) pieces may be original. Apatite is abundant, and

there is considerable ilmenite and pyrite. Chlorite, bastite, hornblende

and kaolin are the alteration products.

The rock has undergone a small amount of shearing, but the alteration

is probably largely superficial. It appears to the writer that this rock is

2 J. P. IppiInGs, “Origin of Primary Quartz in Basalt,” Amer. Jour. Sci., (3), XXXVI,

208. 1888. This paper deals with a somewhat different rock, but quartz norite itself

has been described by TELLPR and vON JOHN, “‘Beitrage zur Kenntniss der dioritischen

-Gesteine von Klausen in Siidtirol,’’ Jahrb. der k. k. geol. Reichsanstalt, XXXII, 589-684.

1882.

36 ANNALS NEW YORK ACADEMY OF SCIENCES

due to a reaction between the mica schist and the biotite norite, although

it is, of course, impossible to explain why the same conditions should have

given rise to the syenite to the west of it or the diorite to the east. That

the quartz in the rock of this little border patch is entirely adventitious

is quite improbable; its rather frequent occurrence in the rocks asso-

ciated with the emery would indicate that certain peculiar conditions

are requisite for its formation and that these conditions had been satis-

fied here.

Augite Norite

Augite norite is a basic and comparatively unimportant facies of the

biotite augite norite magma, having been found in two small patches on

the borders of the latter area and as streaks in pyroxenite on Montrose

Point. The smallest is at Montrose; the other, about half a mile east

of Pleasantside. Its position would thus seem to indicate that it is

merely a local segregation, derived by the loss of biotite. In the hand

specimen, it resembles hornblende norite most closely, since there is no

biotite. The rock is compact, dark gray and of medium grain. Under

the microscope, there is little of note. In the patch at Montrose, there

is a trace of biotite, while the augite and hypersthene are subequal in

amount. There is a remarkable development of apatite in this rock,

sharply crystallized in very large grains. In the rock from the other

patch, there is no biotite; the hypersthene is somewhat less in amount

than the augite, and both are characterized by such an abundance of

inclusions as renders them almost opaque. Bastite, chlorite, epidote

and kaolin are the well-developed alteration products. The streaks on

Montrose Point are similar to the last, except that the plagioclase is

more acid (approaching oligoclase) and the orthoclase abundant, being

about one-third of the feldspar. The hypersthene is light in color and

in pleochroism and is often in small elongated grains. The augite is in

larger crystals, with numerous black inclusions, and there are a few

grains of brown hornblende.

Hornblende Norite

Hornblende norite is the third important member of the norite group.

The bulk of it occurs in the large area to the southwest of Pleasantside;

the rest in a small patch half a mile to the southeast. The former is

entirely surrounded by norites, while the latter is on the edge of the

norite district.

The rock in the hand specimen can usually be identified as a norite by

the pink or (when altered) brownish color of the feldspar. The absence

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 37

of biotite in a norite may generally be taken to indicate this variety.

The identification of the hornblende in the presence of hypersthene is

difficult, owing to the peculiar luster of the latter on cleavages. Occa-

sionally, however, the hornblende occurs in slender black rods, whose

long axes are roughly parallel, when it is unmistakable.

In thin section, the feldspar resembles that of the other norites. The

plagioclase is usually a trifle more basic than in the others, being gener-

ally either a basic andesine or a labradorite. In one case, microperthite

was found, but in another, the plagioclase was almost bytownite. The

orthoclase is perhaps less abundant in hornblende norite, though almost

always present. The hypersthene is usually deep in color and pleochro-

ism. The hornblende seldom resembles that of the diorites, being gener-

ally coarser and of the brownish variety. The greenish tint, is usually

present, however, though faint. The pleochroism is good: X dark

brownish green, Y dark brown, and Z yellow, with the usual absorption

formula. The alteration is to chlorite. Biotite and augite are not in-

frequently present in relatively small amount. Apatite is often found

in great abundance, size and perfection, and ilmenite is also common in

large irregular masses. ‘Titanite is rarely present in moderate amounts.

The order of crystallization exhibited in this rock is often peculiar.

In many cases, the hypersthene, biotite, augite and even plagioclase seem

to have preceded the amphibole, which then occurs in large irregular

plates containing all of the other minerals excepting orthoclase. The

structure then approaches the poikilitic, this term having been coined by

Williams in fact to describe a similar feature in the hornblende of the

hornblende pyroxenites of Stony Point.*° In other cases, the plagio-

clases are outlined by irregular streaks of ilmenite, so that in two respects

a somewhat extraordinary reversal seems to have taken place.

From the frequent presence of augite and biotite in small amounts,

the hornblende norite seems to have its most pronounced relationship

with biotite augite norite. It is, however, most frequently in contact

with this rock. A larger increment of biotite than is usually seen re.

sults in the less common biotite hornblende norite. In the smaller area,

however, it appears to grade into the diorite by loss of hypersthene, the

variety of hornblende being the same in both rocks.

Biotite Hornblende Norite

An area of hornblende norite which carries biotite in addition, hes

just southwest of Dickerson Hill. It may be placed in the same cate-

%0 Amer. Jour. Sci., (3), XXXI, 30. 1886.

38 ANNALS NEW YORK ACADEMY OF SCIENCES

gory as the augite norite, both being derived by subordinate change from

a much more important variety. The rock, however, is important as the

connecting link between the norites and the diorites; it finds a further

transitional phase in some of the latter, which carry hypersthene or

enstatite.

The rock is the customary dark gray and is of medium grain. The

feldspar usually constitutes two thirds or more of the rock. It is chiefly

labradorite, there being little unstriated feldspar present. The hypers-

thene is a little more abundant than either the biotite or the hornblende.

The latter mineral is of the brown variety. Apatite is generally abun-

dant, and ilmenite is present in the ordinary amount and habit. Magne-

tite and pyrite are not abundant.

Olivine Augite Norite

Olivine augite norite, which is the most basic of all the norites, has

been found in only one small area, half a mile south of Pleasantside.

It is situated on the border of the norite district, where it adjoins the

pyroxenites. No olivine was noticed in them at this point, although it

may be occasionally present. The rock is evidently a local basic segre-

gation of the biotite norite member.

The feldspar constitutes about three fourths of the rock and is almost

entirely plagioclase, of the andesine-labradorite series. Of the dark

minerals, green augite is somewhat the most abundant; it is occasionally

intergrown with the hypersthene and always contains the olivine. The

latter may be in large masses, or again it occurs in small grains. It is

always strongly cracked, but it is less altered than would be expected.

What alteration there is has formed chiefly antigorite; there is little

separation of magnetite. This makes up about one fourth of the ferro-

magnesian minerals, and the hypersthene, which has begun to alter to

bastite, nearly one third. There is a little biotite, generally in small

flakes. Ilmenite, magnetite and apatite are not abundant.

Hornblendite

Hornblendite is not common, being known in four small areas only.

One is located at the northern end of Lake Meahach, another a mile

south of Montrose; these are both in the main diorite area, and there is

little question as to their derivation from the diorites by local segrega-

tion, due to loss of feldspar. The others are in the pyroxenite district,

located respectively about a mile east and half a mile west of Dickerson

Hill. The latter appears to be an altered pyroxenite, the hornblende

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 39

having been derived by paramorphism from augite. In the former area,

the coarse green hornblendes often contain patches of altered but still

recognizable augite, and these are undoubtedly secondary; but some of

the hornblende may well be original. If this be so, the primary rock

was a hornblende pyroxenite. Finally, hornblendite occurs in the net-

work on Montrose Point, apparently there also derived by paramorphism

from pre-existing pyroxenites. These two modes of origin thus appear

to have both been operative in the formation of the hornblendites.

The rocks are easily recognizable in the hand specimen; they are

usually medium coarse in grain and glistening black in color, with prac-

tically no white minerals. In the rock of the first patch referred to (at

Lake Meahach), there is a little feldspar present, less than one-fifth of

the rock; a small proportion of it is unstriated. The hornblende consti-

tutes the great bulk of the specimen; it is pleochroic from yellow to dark

brownish green in some cases, while in others it is distinctly brown,

with a very small extinction angle. These two varieties may occur to-

gether in a single specimen. Sometimes the magnetite inclusions become

so abundant, especially in the green variety, that all light is shut off ex-

cept at the margin of the crystal. Hypersthene may be present in small

amount and biotite in larger quantity. The grain of the rock in this

area changes very suddenly from coarse to unusually fine.

In the area south of Montrose, the rock is similar to the above. The

hornblende is always brownish green, with extinction averaging about

20°. Biotite is a trifle more abundant and occurs as aggregates of crys-

tals in the hornblende. Irregular masses of pyrite and pyrrhotite are

common. Calcite has developed to some extent.

In the more westerly of the two other areas, the rock is much the

same, but the brown hornblende is apparently pseudomorphic after au-

gite, and traces of the latter may still be seen. There is a moderate

amount of biotite and several distinct alteration products, such as chlo-

rite, calcite and quartz. There is no feldspar.

The structure in the easternmost area has been described; the horn-

blende is sometimes in large coarse green crystals, and sometimes in a

confused lighter green mass, which is partly chlorite. The remnants of

the augite may still be seen.

The process of paramorphism is thus clearly indicated, and there can be

no doubt that this is the common mode of origin of the hornblendites.

In the two areas in the diorite, this is by no means impossible, since

pyroxenite is apt to occur around inclusions in the main body of igneous

rocks, and these two patches are both near inclusions. There is shown,

however, in the writer’s suite of specimens from this district, a fairly

40 ANNALS NEW YORK ACAD#MY OF SCIENCES.

complete gradation from a very feldspathic diorite to one in which the

feldspar is distinctly subordinate to the ferromagnesian minerals; the

hornblendites, moreover, resemble diorites in their sudden changes in

grain, so that on the whole these two areas may best be regarded as basic

segregations.

Pyroxenite

Pyroxenites form the most basic group found in the Cortlandt area,

with the exception of an occasional specimen in which olivine runs over

one-third and which may thus be classed as peridotite. Williams classifies

all rocks containing olivine as peridotite and devotes the whole of his first

paper to this group; most of them are considered by the present writer

under the heading “Olivine Pyroxenite,” below. The hornblende pyrox-

enites form a distinct group which Williams does not consider at all,

per se. He merely mentions, moreover, the pyroxenites proper, which, as

will be seen from the map, constitute nearly one-fourth of the whole

series, and are thus the most important single rock type of all. They

may be considered first.

Under this head are included all those rocks which consist almost en-

tirely of pyroxene, whether it be monoclinic or orthorhombic. As a

matter of fact, nearly all of them contain both of these minerals, and only

occasionally would there be any warrant for calling a-specimen augite

rock or hypersthenite. Enstatite is quite common, moreover, and in one

case what is probably hedenbergite was found, so that a multiplicity of

names may be avoided by considering the pyroxenites as one group; and

the field distribution seems to indicate that this is a rational procedure.

The pyroxenites may be conveniently described under two heads, accord-

ing as the color of the monoclinic pyroxene in thin section is pink or

white, or green.

White pyroxenite, in which the augite is a light pinkish or greenish

white, is by far the commoner variety. It occurs as mapped on Montrose

Point (although the numerous other rocks which interpenetrate there are

not indicated) and constitutes the bulk of the large area to the east. The

patches of pyroxenite at Buchanan and just west and south of Montrose

are also of this variety,** and it was only in the first two of these that the

pinkish augite was noticed. By the entrance of olivine, it becomes the

olivine pyroxenite of the series; and it may be noted here that while the

areas in which the olivine is segregated are probably fairly definite, they

are at the same time probably not as well defined as depicted on the map.

%. The “fine black rock’”’ which Dana describes as usually associated with the ‘limestone

areas” is merely a fine pyroxenite.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 41

All of the important emery developments are located in the eastern

pyroxenite area, although a local development of a more acid rock often

constitutes the immediate wall rock.

These rocks cannot be distinguished megascopically from those carrying

small amounts of olivine, but their general identity as pyroxenites is at

once apparent. When fairly fresh, they are dull black; upon moderate

alteration, they assume, in many cases, a reddish brown color, and

strangely enough when badly altered, they become gray and look like an

entirely fresh rock. The reddish rock is common in the region around

Salt and Dickerson Hills; the gray, farther to the east. They always

retain their toughness, but the rock, especially when it contains olivine,

will often disintegrate under the hammer, rather than break. That the

rock is the most resistant in the series is evidenced by the fact that the

two highest hills in the county (Dickerson and Salt Hills) are composed

of it. Very perfect jointing is often developed in the pyroxenite. The

grain varies greatly but is generally medium fine. In the district just

west of Dickerson Hill, however, it becomes very coarse, containing crys-

tals of hypersthene and diallage one inch or more long and frequently

masses as large as one’s fist; and this coarse facies occurs in other places

also.

Under the microscope, it is seen that the bulk of the rock is generally

made up of a light colored pyroxene, which is probably augite (fassaite).

It is a uniform light greenish gray in color, with extinction of 42°-45°

in the vertical zone (in diopside Zac= 38°). The alteration, moreover,

is usually to antigorite or to brown hornblende or chlorite, so that too

much FeO is indicated for a diopside. Furthermore, the analyses indi-

cate a mineral carrying 2 to 4 per cent. Al,0,, so that it seems to be an

augite. The mineral sometimes occurs in idiomorphic grains, but it is

ordinarily in an entirely allotriomorphic condition. The grains are some-

times clear and coarse, and sometimes exhibit the fine parting of diallage

(generally parallel to 100) in great perfection. They often carry a

notable amount of fine black rod-like inclusions, which are very different

from the inclusions in the green augite of the norites, although occa-

sionally the latter are found and constitute a typical diallage. Simple

and polysynthetic twinning are common, and the latter especially reaches

a high degree of perfection in these rocks. The other and much less

common variety is similar in every respect excepting color; this is a

light brownish pink, with very faint pleochroism. Its color is very pos-

sibly due to TiO,, but the rock containing this mineral appears to be

intimately associated with the normal variety. The alteration of the

augite is almost always perceptible and often far advanced; the common

49 ANNALS NEW YORK ACADEMY OF SCIENCES

products being antigorite and brown hornblende, or more rarely chlorite.

The light green antigorite forms in a confused fibrous mass over the

original mineral, the alteration starting from the border of the crystal.

The antigorite may form in considerable amount before the polarization

of the augite is destroyed. The brown hornblende forms in sharply de-

fined masses along the cleavage rocks, in strong contrast to the antigorite.

Calcite and quartz are accessory products.

The orthorhombic pyroxene in these rocks is less ferriferous than in the

norites. Hypersthene in rather small rounded grains set in the augite,

and less altered than that mineral, is nearly always present. Its color and

pleochroism are usually not strong, however, and it grades into bronzite

and enstatite by a loss of iron and a concomitant decrease in refringence

and birefringence and in the optic angle about Z. These colorless varie-

ties are difficult of distinction from the white augite in plane polarized

light, but under crossed nicols, of course, the orthorhombic symmetry

becomes evident. Like the augite, the enstatite is usually (but not al-

ways) allotriomorphic. Fine parting is generally well developed, and

the characteristic small inclusions, regularly arranged, are common.

The rock is usually at least half made up of the augite, although the

proportions of the monoclinic and orthorhombic pyroxenes vary greatly.

Sometimes an almost pure enstatolite will be developed; pure augite rock

is known, but hypersthenite is rare. The most common type consists of

augite to the extent of from one half to two thirds, with hypersthene and

enstatite—the latter usually the more abundant—for the other chief con-

stituent. Hornblende in small quantity, and generally basaltic, is very

common, especially in the Montrose Point area. Biotite is rarely present.

Traces of feldspar, chiefly unstriated, are not uncommon. Apatite is

practically absent. The chief ore is pyrrhotite; pyrite and chalcopyrite

are less common; ilmenite and magnetite are always present, usually in

small quantity, and chromite is known.

Green pyroxenite was found in only one patch, where it adjoins and

probably surrounds a limestone inclusion. It is located on the post-road

between Montrose and Buchanan. The available outcrops were all very

near the contact, the rock being greenish black and felsitic. Under the

microscope, it appears that the great bulk of the rock is composed of an

apple-green pyroxene in small crowded grains. The extinction angles run

from 30°-38°. In basal section a somewhat eccentric bisectrix figure is

obtained. The refringence is a trifle higher than usual in a pyroxenite,

and the birefringence lower. Pleochroism is more distinct in some grains

than in others, though it is nearly always visible; when most distinct X is

pale brownish, Y pale green and Z apple green, or X may be pale green.

ROGERS, GHOLOGY OF THE CORTLANDT SERIES 43

Some of the larger grains show almost colorless centers. The cleavage is

always coarse, and there is little suggestion of a diallage parting ; in basal

section, the cleavages intersect at 87°. p > v. It appears from these

characters that the mineral is a member of the diopside series, probably

salite or hedenbergite, or possibly (from the pleochroism) a lime dial-

lage. Titanite is scattered through the rock rather plentifully in small

wedge-shaped grains. Very small grains of what is probably wernerite

are not uncommon, and there is a very little quartz. Opaque minerals

are almost lacking.

The peculiar combination of minerals in this rock indicates that its

relation with the associated limestone is very close. All of them, except

the very small amount of quartz, are lime minerals, and all of them may

be contact minerals. There is little doubt that the rock represents a clear

case of contact between limestone and probably diorite; and similar,

though entirely abnormal, mixtures are not uncommon in the district and

will be described below.

Hornblende Pyroxenite

Pyroxenites containing sufficient amounts of hornblende to place them

in this class are most common on Montrose Point, although two other

small areas exist: one just south of Salt Hill, and the other two miles

north of the first. Williams describes this class only in their peridotitic

phase ; they are then most abundant on Stony Point, though known in the

extreme eastern end of the main area. The hornblende pyroxenites have

originated in all the cases examined by local increase of hornblende in the

typical pyroxenite described above, although conceivably a partial para-

morphism of the pyroxene would give rise to a similar rock.

These rocks cannot be distinguished in the field unless, as is often the

case, the hornblende occurs in very large individuals with the other min-

erals imbedded in it. This is the type development of the poikilitic struc-

ture which led Williams to coin the term. The hornblendes may reach

three or four inches in width, and their glistening black surfaces, inter-

rupted by small included grains, is an unmistakable feature. On Mon-

trose Point, however, this structure is often lacking entirely, the horn-

blende having crystallized simultaneously with the other minerals.

Under the microscope, the proportions of the various minerals are seen

to vary considerably, according to the degree of completeness which the

segregation of hornblende has attained. The typical greenish gray augite

is always present and may constitute as much as two thirds of the rock.

Hypersthene, often in the beautifully pleochroic clear rounded grains,

44 ANNALS NEW YORK ACADEMY OF SCIENCES

and enstatite are present as stated under pyroxenite. Small amounts of

olivine may be present, but if at all important the rock is classified as

olivine pyroxenite. Calcite, quartz, chlorite, antigorite and hornblende

may occur as alteration products; and in one more acid variety, contain-

ing some feldspar, zoisite was found, typically developed. The horn-

blende is the essential constituent in this class, making up from one third

to two thirds of the rock, and yet it seems to differ in the various patches.

. It is usually of the basaltic variety, rich brown in color, and pleochroic

from dark to light brown. The refringence and birefringence are higher

than in the green variety, and the extinction angle is smaller. It is this

variety that usually gives rise to the poikilitic structure; it seems gen-

erally to have crystallized last. In one specimen, the hornblende was

apparently basaltic but had become a pale brown, and the crystals were

often decolorized in the center. Again it may be of the coarse greenish

brown variety which is characteristic of the hornblende norites, so that the

hornblende pyroxenites do not seem to be a very well defined member of

the series.

Olivine Pyroxenite

Olivine pyroxenite forms the last class of any importance; rocks in

which the olivine runs over one third and which are then classed as perido-

tites, are rare. They seem to occur, moreover, in the centers of the areas

of olivine pyroxenite; and since they merely mark the culmination of the

segregational process which has led to the formation of these areas, it was

thought unnecessary to differentiate them on the map. As stated above,

the areas occupied by the chrysolitic rocks are rather vaguely defined, and

owing to the difficulty of distinguishing these rocks in the field, their

extent on the map is only as close an approximation as could be attained

by collecting and sectioning a large number of specimens.

The weathering of the olivine rocks is beautifully shown in the two

eastern areas. The decay of the olivine causes a disintegration of the rock

into a coarse red sand, whose fragments are the grains of augite and

hypersthene. This is much prized as road metal, and it is used exten-

sively through the eastern part of the district. It is in effect a fine, re-

sistant, homogeneous gravel, found ready crushed and sifted, and is excel-

lent as a road covering. It is not, however, of like benefit to the farmer,

being too ferriferous to be fertile. On the hills, it favors especially a

peculiar flora, with an abundance of such trees as cedar and hemlock.

The topography of this region is rather striking, there being numerous

rounded hillocks, whose shapes are due to the melting away of the rock

masses which compose them. These ledges when artificially exposed show

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 45

their peculiar softened shape, with rounded masses often projecting high

above the general level of the rock. This is well shown at Chase Corners,

just northwest of Dickerson Hill, as pictured in Plate III, fig. 1; the

effect is rather striking, when it is recalled that the other rocks of the

series are usually found planed smooth by glacial action.

As above stated, these rocks resemble in general aspect the pyroxenites,

although closer examination may reveal the small yellowish green grains

of olivine. It is only in thin section, however, that the rock may be ac-

curately studied. The proportions of augite (usually the gray variety,

but occasionally the pink), enstatite and hypersthene vary as recorded

under the pyroxenites; almost pure olivine-augite rocks and olivine-

hypersthenites are known. MHornblende, usually basaltic, is a fairly con-

stant component, though it seldom occurs in important amount. The

olivine is present in amounts varying from one-fifth to one-third of the

rock. It occurs in typical grains, usually rounded, but sometimes ex-

hibiting a distinct crystal outline. Usually besides these large grains

(which may reach 3 mm. in size), there are numerous little ones scattered

throughout the augite. The mineral is colorless but often carries magne-

tite inclusions in the form of minute rods; this feature is, however, better

developed in the peridotites of Stony Point. The alteration is exhibited

in all stages. Sometimes the mineral is nearly all serpentinized, with

only occasional little bullets of olivine remaining, while again it may be

almost perfectly fresh, even when strongly cracked, and when the augite

shows distinct alteration. The characteristic mesh structure is always

developed on alteration. The product is, of course, serpentine. Antig-

orite is the most common variety, chrysotile is somewhat more rare and

bowlingite less frequent. Magnetite separates out in varying quantity.

According to Williams, whenever the mineral comes in contact with feld-

spar a diopside-actinolite reaction-rim is formed, such as he has figured ;**

but feldspar has not been found in contact with the olivine in the rocks

of the main area, so that this effect is not a common one. A similar

phenomenon, between olivine and biotite, is described and figured below.

Peridotite

Peridotites are merely exceptionally basic phases of the rock last de-

scribed; and their relations to that type have already been discussed.

They do not differ in any respect except the proportion of olivine; this

mineral, however, in several cases constituted about three fifths of the

rock, and in one instance was almost unaltered, though badly cracked.

% Amer. Jour. Sci., (3), XXXI, 35. 1886.

46 ANNALS NEW YORK ACADEMY OF SCIENCES .

These cases were all found in the two eastern areas, although less pro-

nounced ones occur in the Montrose Point district. All of these rocks

would be classed by Williams as augite peridotites (picrites) which he

thought occurred in the highest development on Montrose Point. He

describes at considerable length another group from Stony Point, how-

ever, which is.not typically represented in the main area—the hornblende

peridotites. The nearest approach to these rocks is to be found in the

hornblende pyroxenite area on the south side of Montrose Point; but

these contain olivine to the extent of more than one-third only exception-

ally. Oddly enough, however, Williams, who, after all, studied only a

small part of the whole area, felt that they were so important a member

of the Cortlandt Series that he proposed the name “Cortlandtite” for

them, and rocks consisting of olivine

and hornblende are now widely

known by this name.

A rather extraordinary rock, some-

what analogous to the so-called cort-

landite, does, however, occur in a

small patch on the extreme eastern

edge of the series. This in the hand

specimen exhibits the poikilitic

structure par excellence, but biotite,

instead of hornblende, is the host.

This occurs in individuals often sey-

eral inches across (although by rea-

son of the abundance of the included

minerals, it makes up somewhat less

than one-third of the rock), and under the microscope it is extraor-

dinary for the abundance of its inclusions. These are of a sharp,

brownish black substance, probably magnetite, and are so abundant in

the center of the grain as to render it opaque, but the periphery and a

margin around each included olivine is entirely clear. On the extreme

edge of the biotite in the latter case, however, there is a narrow band of

magnetite grains, and surrounding, or nearly surrounding, the olivine is

a wider zone of pale green pyroxene in rectangular blocks (fig. 1). This

appears to be a reaction rim, although the pyroxene zone is not entirely

universal. Basaltic hornblende, without inclusions, is fairly abundant,

and hypersthene and enstatite make up nearly one-third of the rock.

The olivine is also present in about this amount; white augite is almost

lacking. Alteration is apparent, but is not severe.

As the culmination of the basic development along this line, we have

Fig.1. Poikilitic Relation of Biotite

and Olivine in Peridotite. Sl. 166

ROGERS, GHOLOGY OF THE CORTLANDT SERIES 47

several occurrences of serpentine. These are probably generally, if not

always, dikes and they will be described under that head. They always

occur in the pyroxenite area, however, and may represent local segrega-

tions of olivine. They would thus once have been dunites, since only

serpentine, showing mesh-structure, is contained in them. If they are

genetically related to the peridotites, however, it is difficult to under-

stand their complete alteration; and while their field relations cannot be

definitely established, they may be provisionally considered as dikes.

DIKE ROCKS

Dikes are fairly common through the district and are usually small ;

and the following list is therefore probably by no means complete. Aside

from those within the area itself, an excellent development of chiefly

basic varieties occurs on Verplanck Point, where the dikes ramify into

the limestone and schist. The abundance of pegmatite dikes in the north-

ern part of the district has already been remarked. The complex of rocks

on Montrose Point, in which, as Williams says, “each rock includes and

forms dikes in every other” is interpreted as a case of true differentiation,

rather than as a network of dikes; and the rocks have therefore been

considered above. A careful study of the dikes of the series with regard

to their mutual time relations would probably be of great value in indi-

cating the relations of the plutonic types; for if the dikes had followed

the laws which govern their successive differentiation in other districts,

it might be inferred that the plutonics had done the same.

A plite

The aplite dikes are usually small and not especially common. They

occur chiefly north of Pleasantside. They are, of course, fme grained and

consist of muscovite, orthoclase and quartz, with small amounts of plagio-

clase and biotite and with zircon and magnetite as the common acces-

sories. They are, perhaps, slightly more acidic than the granites, but

their composition does not differ greatly from the latter.

Pegmatite

Pegmatite is probably the most common dike rock of all. It was no-

ticed especially in narrow veins or dikes in many of the more northerly

emery cuttings. The pegmatitic structure is generally well developed, and

the component minerals are as usual. The largest and most striking

occurrence of the rock appears about 200 feet south of Montrose station,

48 ANNALS NEW YORK ACADEMY OF SCIENCES

where a large irregular mass outcrops on the west side of the railroad.

Its relations are rather obscure, and its size if fully exposed would prob-

ably be about 60 to 80 feet, but there can be little doubt as to its identity

as a dike. Occasional blotches of a fine grained gray substance are in-

cluded in it. The true pegmatitic structure is not well shown, this name

being applied because of the coarseness of its texture. About 30 feet to

the south lies one of the schist inclusions described below; between the

two the country rock (biotite augite norite) is altered to a diorite, some-

what gneissoid, but this is probably due to the schist rather than the

pegmatite.

Dacite Porphyry

The very remarkable rock, dacite porphyry, is known from only one

locality, viz., about 300 feet up the hill east of Montrose station. It out-

crops in the gutter of the road for about ten feet (though it may be traced

for 75 feet), and this outcrop is about three feet wide. By differential

weathering, it projects about three feet above the ground, and its jointing

and its white color, combined with its odd position, cause it to look at a

glance exactly like a stone wall. In thin section, the rock is seen to be

composed of a moderate number of large idiomorphic feldspars set in a

mosaic of quartz and feldspar grains. The phenocrysts are almost never

striated, but Williams** states that all of the feldspar of the rock has a

specific gravity of 2.63-2.67, so that it must be oligoclase and andesine.

The phenocrysts never show resorption, but usually exhibit beautiful

zonal growth. The quartz is all in the groundmass, and some of the

small feldspars are striated and seem to be oligoclase. Biotite, hornblende

and muscovite, some of the latter probably damourite, but not all, occur

in rather small quantity. In the groundmass, the grains often appear to

be interlocked, as though the rock were a kind of augen gneiss, but the

perfect outlines of the phenocrysts and the distribution of the muscovite,

as well as the field relations, preclude this.

Dioritic and Gabbroic Dikes

Various types of the dioritic lamprophyres constitute by far the most

common group of dike rocks. They are especially abundant on Verplanck

Point, although found quite abundantly elsewhere, especially through the

southern part of the district. The different types, of course, exhibit no

preference for any particular plutonic rock. Only in the limestone on

83 WILLIAMS Calls this rock a bed of porphyritic quartz mica diorite, (Amer. Jour. Sci.,

(3), XXXV, 446. 1888.) and Dana a granitoid micaceous quartzite (Idem, XX, 218).

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 49

Verplanck Point was any contact metamorphism observed: here the dikes

are gabbroic (camptonite), and cause the formation of diopside, actino-

lite or hornblende and pleonaste in the limestone. Only occasionally is a

typical minette, vogesite or kersantite found; usually the orthoclase and

plagioclase, and the biotite and hornblende, are subequal in amount,

Green augite has been found only in the dikes on Verplanck Point, where

they radiate into the limestone; these, however, are typical camptonites.

Furthermore, there is seldom any serious alteration, such as is described

by Rosenbusch in the type lamprophyres, although this is a less essential

feature. Minette has not been found typically developed: the usual

variety is a basic one and might almost as well be called either of the

following. Vogesite is quite typically developed in a number of places;

the orthoclase constitutes two thirds or more of the feldspar, and horn-

blende, which is usually greenish brown, though sometimes green, is dis-

tinctly predominant over the biotite. Augite, however, has been seldom

found in them. Kersantites, though not typical, are found abundantly

on Verplanck Point, and also on Montrose and Stony Points. They are

merely fine grained mica diorites with considerable hornblende. Camp-

tonite occurs on Verplanck Point, as stated above. The classification of

these rocks according to Rosenbusch is thus a difficult matter, and they

had best, perhaps, be regarded as dioritic and gabbroic dikes.

Hornblendite

Only two dikes of hornblendite were found, one on Verplanck Point,

in limestone, and the other in one of the emery pits on the hill east of

Pleasantside, in diorite. In both cases, the rock was composed almost

entirely of bright green hornblende, with a very subordinate amount of

biotite. Apatite, which is ever-present in the diorite dikes, is practically

lacking.

Serpentine (Peridotite)

Three outcrops of serpentine (peridotite) were found, one two miles

east of Montrose, another four, and the third about half a mile farther.

The first is in hornblende norite; the others in olivine pyroxenite. The

first and last are undoubtedly dikes, the former three feet wide and the

latter about ten feet. The second outcrops on an old log road southeast of

Dickerson Hill, and its relations cannot be definitely determined. If a

dike, it must be at least twenty feet wide. In all three cases, the rock is

greenish black, fairly soft and strongly jointed. In the first case, the

structure of the serpentine is hard to work out. There are traces of

50 ANNALS NEW YORK ACADEMY OF SCIENCES

typical mesh-structure, though most of it has neither the antigorite nor

the bastite habit. It most resembles the former, however, and a hypers-

thenite dike so completely altered would be a curiosity. There is also a

small amount of sericite, quartz and biotite. In the other cases, the

typical antigorite or mesh-structure, indicating the former presence of

olivine, permeates the whole rock. Traces of altered white augite are

still visible; but the rock was evidently almost entirely olivine. In all

three cases, a considerable amount of magnetite is present.

ABNORMAL CONTACT ROCKS

The true igneous character of the series is proved by the frequent

presence of contact action. This usually occurs along its borders, but

several very peculiar and abnormal developments are to be found within

the area itself which can only be explained by considering them as due to

the contact action on the inclusions which are fairly common in the

western part of the area.

Williams* has given such a minute description of what seems to be the

characteristic type of border contact action that it would be superfluous

to describe in detail the work which the writer has done along the same

line. The general result seems always to be that the mica schist increases

in alumina content as the massive rocks are approached; and this is

attended by the formation of such aluminous minerals as staurolite,

sillimanite, cyanite and garnet. A very striking feature, and one whose

importance will be emphasized later, is the great increase in the amount

of biotite, which at a point ten yards from the contact at Crugers consti-

tutes the bulk of the schist; and magnetite is thickly scattered through

the mica. Contained in the diorite itself at this point are numerous small

schist inclusions, and these in like manner furnish excellent opportunity

for the study of contact action. They are largely changed to pleonaste

and corundum in some cases; in others to quartz; in others to staurolite

and green mica. The first case resembles the emery from the mines very

closely.

Along the limestone contact, which exists only on Stony Point and at

Verplanck, contact action of a different kind has taken place. The com-

mon minerals developed are usually pale green amphibole and pyroxene ;

more rarely titanite, zoisite and scapolite are formed (these last were

observed on Stony Point). Similar effects have been recorded by Kemp

from the Rosetown area,** where the limestone contact is especially well

shown.

% Amer. Jour. Sci., (3), XXXVI, 254. 1888.

= Op. cit., p. 252.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 51

The writer has observed several other instances, however, which would

seem to merit description. The pale-green pyroxenite recorded above, .

occurring on the border of a limestone inclusion, is undoubtedly a contact

facies, and may be recalled in this connection.

Wernerite Schist

Wernerite schist is found in a small patch at the cross-roads on Mon-

trose Point, lying in the pyroxenite area. It has been only partly ex-

posed by the excavation of an old clay pit at this point, but it probably is

not more than several hundred feet long, extending northeast. In the

field, it is distinctly gneissoid in appearance, the black bands being very

distinct. It is very tough and has been planed off into a series of roches

moutonnées, so that it is hard to sample. Under the microscope, it is

seen to be made up of scapolite, pyroxene, titanite, calcite and pyrite.

The scapolite seems to be wernerite, and the appended analysis as recast

indicates that it is Me, Ma,, which is a very calcareous wernerite. The

mineral gives conclusive tests. It makes up almost three fifths of the

rock. Associated with it is the calcite, which is very subordinate in

amount. The pyroxene is similar to that found in the green pyroxenite

described above, except that the grains are less crowded and have a pris-

matic shape. They are regularly aligned, and impart the gneissoid ap-

pearance to the rock. The pyrite and titanite are very variable in amount,

occasionally becoming quite abundant. Since all of these minerals have

been observed on the limestone contacts, there is little doubt that this

patch represents an entirely absorbed inclusion of limestone. It is pos-

sible, of course, that the limestone is still present, though concealed, but

the rock described exists over a zone 75 feet wide, at least.

Lying between this rock and the pyroxenites, in several places at least,

a typical diorite is to be found.

On Verplanck Point, just west of the brick yards, and about 200 feet

from the schist contact, a pyroxenite exists whose peculiarities might be

taken to indicate a similar history. The rock is very hard and black,

although in thin section a surprising amount of alteration is revealed.

There is considerable pinkish augite in the rock, which appears to have

altered largely to a greenish hornblende. Wernerite is present to the

extent of nearly one third,-in large and small grains.** Calcite is less

abundant, and in part at least, it is derived from the augite. There are

also biotite, apatite and plagioclase in small amounts. Although no

limestone is visible near this rock, the wernerite would point to an in-

cluded patch which had been partly or wholly absorbed.

36. C, CALKINS (Science, XXIX, 946. 1909.) notes the occurrence of “primary”

sscapolite in igneous rock, but also concludes that it is due to the absorption of limestone.

52 ANNALS NEW YORK ACADEMY OF SCIENCES

Garnet Rocks

On the southern border of the district, along the road at the foot of

Salt Hill, is found a contact facies of very interesting composition. The

outcrops are badly cracked and jointed, and the rock is friable and stained

brown along these cracks. When fresh, two varieties may be distin-

guished, one pink and the other an ash gray, these two occurring within

30 yards of each other.

The former variety is of about the color of rhodonite; it is fine grained,

but is thickly penetrated by flat gray tremolite rods. Under the micro-

scope the pink mineral, which makes up more than half the rock, is seen

to be garnet. It is practically colorless and always isotropic and is prob-

ably grossularite. The tremolite*’ is typical and constitutes nearly one

third, and quartz makes up the rest. Rutile, in small sharp grains, with

very high relief and birefringence, and a deep yellow color, is quite com-

mon. This rock is thus probably derived from an impure limestone, or

possibly a calcareous schist, and presents an interesting contrast to the

wernerite rock. The abundance of titanic oxide present is a curious

feature in the contact metamorphism of the limestones; either as ilmen-

ite, titanite or rutile, it appears to be always present, whereas iron and

alumina are more typical in the case of the mica schists.

The gray rock in thin section is resolved chiefly into quartz and garnet.

The latter is identical with that in the last, except in its color in mass.

The most interesting and suggestive thing about this rock is the presence

in rather small amount of both corundum and pleonaste; and magnetite

is, moreover, thickly scattered over the slide. Notwithstanding its pro-

pinquity to the previous phase, the aluminous character of its components

seem to indicate derivation from a mica schist rather than a calcareous

rock.

The very abnormal developments which are associated with the emery.

may or may not be contact rocks; they had best be described, therefore,

in conjunction with the emery.

INCLUSIONS

Inclusions** of schist (and occasionally of limestone and gneiss) have

been alluded to frequently in the previous pages; they are quite abundant

37 KHMP mentions an analogous rock from the border of the Rosetown extension, com-

posed entirely of tremolite (Op. cit., p. 252.).

88 Accidental xenoliths of this kind are of wide occurrence in other igneous districts.

They have been exhaustively studied by Lacroix, ‘‘Les Enclaves Des Roches Volcan-

iques,’”’ 1893. The tendency of intruded rocks to carry them is discussed in “Geology of

the Castle Mountain Mining District, Montana,” Bull. U.S. G.S.,139. 1896.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 53

in the district and are of especial interest from the standpoint of contact

action. They were first noticed by Dana, who mapped*® several of the

larger limestone areas in the western part of the district. They appear

to be most frequent in this area; in the region around Salt Hill they

occur less commonly. In the main diorite area, however, they are much

more numerous than is indicated on the map; there are many, especially

near Crugers, that are too small to be shown; and near Salt Hill, they are

too small and their relations too obscure for adequate representation on

the map.

Dana maps three inclusions of limestone in the large diorite area, and

all of these have been found by the writer to be a very quartzose schist,

closely resembling limestone in appearance, but nevertheless of an entirely

different composition. Hither Dana, who was looking for limestone,

mistook its identity on account of its peculiar appearance, or else it does

contain calcareous layers which he happened to find and which the writer

overlooked; but it is certain that it is essentially a quartz schist. It

weathers in layers, however, which also tends to give it the appearance of

limestone. It appears to be chiefly quartz in most places, with some

feldspar and traces of muscovite; but it occasionally contains layers which

are entirely similar to the Manhattan schist. It is usually considerably

decomposed. In one case, it resembled a very fine gneiss, but other parts

of the same inclusion were normal, and it was moreover very similar to

an outcrop of schist on Broadway, the straight road running north from

Verplanck. An outcrop of the schist (see map) is to be seen on the post-

road, a short distance below Montrose, and the cliff behind Mackey’s store

at Buchanan contains two stringers of this rock imbedded in a somewhat

gneissoid diorite, which to the west becomes pyroxenite (Plate IIT, fig. 2).

The inclusions are generally associated with rocks which show more or

less evidence of metamorphism. This is so apparent that these rocks

were first interpreted by the writer as parts of the schistose inclusions

themselves, but their peripheral position would seem to indicate that they

are true igneous rocks which have undergone some shearing; and this is

borne out by the composition, which is that of a mica diorite. The

strongly crushed gabbro is also found on the border of two of these in-

clusions.

The abnormal mineralogy of the schist is peculiar, although there is

little doubt that it is Manhattan schist. The mica has evidently been

largely altered or absorbed by the igneous rocks, leaving a very quartzose

residue ; and it would seem, therefore, that these inclusions may thus en-

% Amer. Jour. Sci., (3), XX, 195. 1880. A minute description of a number of inclu-

sions is here given.

54 ANNALS NEW YORK ACADEMY OF SCIENCES

rich the surrounding magma in certain constituents, such as alumina.

It will be recalled that the numerous small schist inclusions—at Crugers,

for example—are also distinctly altered, but often in exactly the opposite

direction, so that magnetite, corundum, biotite, etc., are developed.

Only one inclusion of limestone was found, which was either over-

looked by Dana or else its position on his rough map is not accurate. It

occurs along the post-road a short distance above Montrose, and on the

north the contact with the green pyroxenite described above is visible.

It is about seven hundred feet wide and is a medium grained, thoroughly

crystallized patch of the white Inwood limestone. The wernerite schist

described above may be mentioned in this connection, since its characters

would indicate that it is a thoroughly worked-over inclusion of limestone.

A single inclusion of a rock which appears to be undoubtedly a basal

Highland gneiss was found. It is located on the shore of Lent’s Cove,

and it seems to be only about seventy feet wide and of a slightly greater

length. It is almost entirely feldspar, which is very strongly kaolinized,

but which seems to be albite. There are also streaks of chlorite in the

rock; and the whole has a decidedly gneissoid aspect. It could not be a

dike or a member of the Cortlandt Series; it is probably a xenolith of an

ancient gneiss, although it might also be thought that the igneous rocks

are thin over this area and that this outcrop represents a projection of

the gneiss which underlies the series.

STRUCTURAL GEOLOGY

There is little that might be called structural geology that has not been

already discussed. Jointing usually occurs in these rocks, but always in

moderate development. The weathering of the olivine pyroxenites has

been noticed above, and the only other features of note may be now

briefly considered.

DYNAMIC METAMORPHISM

The amount of shearing undergone by the Cortlandt Series is, as has

been shown, slight but constant. It appears to vary somewhat among

the different members. The granites, although they carry microcline,

show strain in general to a very small degree: It is only occasionally that

the norites exhibit any trace of the mortar structure, bent plagioclase

lamelle, wavy extinction, and twisted biotite that characterize the most

metamorphosed types. In but one or two instances was any crushing

seen in the pyroxenites.

In the diorites and gabbros, however, the case is different. The former

group, as remarked above, shows perceptibly more strain than is to be

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 55

found in the granites, norites or pyroxenites, although at the same time

it is not severe and is apparent in the hand specimen only in exceptional

cases. The gabbros in two cases show a rémarkable degree of crushing,

such as would be characteristic of an augen gneiss, and such as was de-

scribed by Lehman*® in the “Flaser Gabbros” of Saxony.

Both of these cases, however, occur directly on the border of an inclu-

sion; and those of the diorites which show a gneissoid structure are simi-

larly located. Williams has described the crushing which is apparent in

some of the types found in the “Butler Section”; it is here also very

severe microscopically, but this section is located 150 yards west of the

inclusion which runs northwest from near Montrose. In many of these

instances, it is easy to take the strike and dip, but the cases of straining

which were noticed elsewhere were relatively insignificant. It would thus

seem to appear that dynamic metamorphism is usually or always confined

to the borders of foreign inclusions; and if the district had undergone

any amount of regional pressure, it would naturally be concentrated

along the lines of weakness which would develop on the contact. This

is the only explanation that would seem to account for such a localization

of metamorphic effects.

It may be noted in this connection that dynamic action is also very

perceptible at most of the emery developments in the district; it is evi-

denced not only by microscopic and megascopic shearing, but by faulting

and veining.

ORIGINAL GNEISSOID STRUCTURE

Dana** describes in some detail a structure which the present writer

believes to be of an original gneissoid character. As already stated, on

Montrose Point, in particular, several very different kinds of rocks are

associated in the most intricate way, often as successive bands ;*? and

Dana cites one case in which biotite augite norite and olivine pyroxenite

are found in alternate layers of constant grain only three or four inches

wide. There are many other cases of less pronounced character, which

are referred to by both Dana and Williams (e. g., the Butler Section)

and which have been seen by the writer. In another instance, a streak

of the coarse dark pink norite was seen in a cliff of the black pyroxenite ;

the norite was coarser, if anything, than the latter, and was coarse, more-

over, to its very edge, having thus none of the characteristics of a dike.

The analyses of these types show their great chemical differences.

40 Uber die Entsthehung der altkrystallinen Schiefergesteine, p.190. Bonn, 1884.

“1 Amer. Jour. Sci., (3), XX, 211 et seg. 1880.

42 WILLIAMS refers to these as “‘dikes,’’ but Dana evidently recognized that they were

not of this character.

56 ANNALS NEW YORK ACADEMY OF SCIENCES

In the norite family, however, this structure is commonest and best

developed. As remarked above, it seems to be a general rule that the

finer grained a norite is, the simpler it is, 1. e., a very fine-grained norite

is composed chiefly of feldspar with considerable hypersthene, while the

coarser varieties carry in addition either hornblende, or biotite and augite.

The fine-grained simple norite is never found in large areas, but always

as inclusions in the coarser, and therefore more complex varieties. It

often occurs in biotite norite, for example, as small, rounded flow-like

patches, or again as streaks; or it may be banded with the coarser rock.

In this case, the chemical difference is not so great, as the accompanying

analyses show.

Now, if the simpler norite be not quite so fine grained, it will not be

entirely pure; and this is the case in most places. In Plate IV, fig. 1, the

mass of the rock is a biotite augite norite, while the white streak is

merely a finer and therefore simpler facies, containing only small amounts

of biotite and augite. In one place, it shows an included patch of the

coarser rock. The ledge shown is about 20 feet high. Plate IV, fig. 2,

shows the same thing in better development, so that the rock might easily

be mistaken for a real metamorphic gneiss. A number of other equally

good instances might be shown, for the structure is quite common; but

these suffice to show its general aspect.*

From what has been said above, it is evident that this structure cannot

be due to ordinary shearing, and we are therefore obliged to search for

another explanation.

Eliminating Dana’s idea of worked-over volcanic ashes, and Williams? 8

suggestion of the ordinary regional metamorphism of igneous rocks, we

are thrown back on some force concomitant in its action with the cooling

of the rock. Since the several layers or streaks are always quite different

in mineralogical composition at least and more or less so in chemical, it

is evidently a question of magmatic differentiation. It is inconceivable

that the structure be due to the differentiation of a magma in situ—after

it had reached its present position—since the differentiation is into bands

which bear no definite relation to the borders of the magma; and the

idea of successive intrusions—first of a light band and then of a dark—is

equally inapplicable, since even when there is a sharp line of demarcation

# The classic locality for this structure is on the Isle of Skye, in Tertiary gabbro,

where it was first described as such by GIEKIE and TEALL. (Quart. Jour. Geol. Soc., I,

646. 1894.) It has also been described in wonderful development by A. G. HédcsBom,

from the Island of Orné, just south of Stockholm. (‘Zur Petrographie von Orné Huf-

yvud,” Bull. of the Geol. Instit. of Upsala, X, 150.) F. D. ADAms has noted a striking

development of the structure near Montreal, which he will describe in a forthcoming

paper.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 57

separating two bands, the individual grains seem to interlock across the

line. The only remaining hypothesis, therefore, is that of the intrusion

of a molten mass already heterogeneous.

Harker** appears to favor the view that the structure is due to the

approximately simultaneous intrusion of two different magmas, which

would give rise to an interpenetration of the two. This would account

for the banded structure, which always shows evidence of flowage and is

seldom straight and clear cut. The assumption would be then, of course,

that the mass would soon begin to cool and harden while resting quietly,

as otherwise the two magmas might combine to form a third and homo-

geneous one. Harker’s alternative view is that the mass was intruded as

a unit, already heterogeneous, the two different magmas having been

partly mixed before intrusion. Whichever be the correct theory, it is evi-

dent that in the Cortlandt Series the simple norite magma was very small

in comparison with the more complex norite magmas, since the former is

always as included bands in the others, while these latter cover con-

siderable areas.

RELATIONS OF THE TYPES, WITH ANALYSES

From the above description of the more salient characters of the Cort-

landt Series, it appears that we have in it a fairly complete and very

intricate complex, one susceptible 4nd worthy of the most detailed study.

In some places—e. g., Montrose and Stony Points—the complexity of the

mass is bewildering, while again we may have several miles of a fairly

uniform rock. At times, as shown in the original gneissoid structure, the

contacts are sharp and clear, although never showing contact metamor-

phism on each other, while again numerous cases have been noticed of

one rock grading into another. Thus, for example, the biotite norite area

at Lent’s Cove grades into the biotite augite norite to the south by a

perfect series of steps; similarly the augite norite at Montrose. Diorite

becomes hornblendite, and pyroxenite becomes chrysolitic, by imper-

ceptible gradations. Moreover, the different larger groups are similarly

connected ; the diorite just southeast of Pleasantside grades into the ad-

joining hornblende norite, and norite passes into olivine pyroxenite by

way of olivine norite. An infinite number of species might be differen-

tiated within this small area of twenty-five miles; in this paper only the

unavoidable ones have been mentioned. The accompanying diagram

(fig. 2), greatly modified from Williams, is designed to show in a rough

way the relations of the species. The lines connecting the circles indi-

cate the direction in which gradation has been most frequently observed.

“ Natural History of Igneous Rocks, pp. 138 and 341. New York, 1909.

58 ANNALS NEW YORK ACADEMY OF SCIENCES

The variation diagram of the series (fig. 3) indicates a rather complex

relationship. The pyroxenites alone, however, are responsible for the

sharp curves; if they be disregarded, the variation curves of the several

oxides will be smooth and simple. This would seem to indicate that the

differentiation which led to the formation of the pyroxenites was of a

Norite

Fig. 2. Diagram of the Relations of the more important Types of the Cort-

landt Series

peculiar character: that either they are derived directly from some other

type, as the norites, or else that the primal parent magma underwent

great changes between the times of intrusion of the pyroxenites and the

other rocks.

An effort has been made to indicate under each type, in a general way,

some of the more suggestive points which the map brings out. Thus the

granites appear to have little connection with the main body, except as

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 59

they all belong to the same series. The norites occupy a central position

in the area, being flanked by diorite and pyroxenite, and often, at least,

protected on the north and south by more acid rocks. The biotite augite

norite magma is the largest and is closely connected with the norite proper

and the biotite norite; while the hornblende norite shows closer affinities

with the diorites. The pyroxenites comprise the eastern third of the

series, and possibly likewise extend to the west, since they are found on

As BCODEFG’ HI K

Fig. 3. Variation Diagram of the Cortlandt Series

A=Diorite; B=MHornblende Norite; C=Biotite Norite; D= Olivine

Pyroxenite; E=—= Pyroxenite; F=—Biotite Augite Norite; G—=Norite; H=

Gabbro; I= Augite Norite; K—Granite. (The rocks of the New Jersey and

Connecticut extensions of the series are not included.)

Montrose Point and again across the river, on Stony Point. They be-

come chrysolitic in apparently irregular patches. Finally, the diorites

occupy an area to the west which contains most of the inclusions of the

district ; and in several places where there is reason to believe that inclu-

sions have existed but have been absorbed, diorite is also found. Gabbro

is found in two cases next to the limestone, and in two others adjoining

schist, while syenite is in an analogous location.

The problem suggested by the location of these last rocks is a difficult

one to solve. According to Harker,** the hybrid formed by the absorption

45 Jour. Geol., VIII, 389-399. 1900. Also Nat. Hist. of Ign. Rocks, Chap. XIV. 1909.

60 ANNALS NEW YORK ACADEMY OF SCIENCES

of a sediment by an igneous rock is always of abnormal composition and

constricted extent and is usually marked by such minerals as sillimanite,

cordierite and wollastonite. By variation diagrams he proves that such a

commingling would give rise to rocks whose composition might be pre-

dicted and which would not even approximate any known igneous flow.

In the Cortlandt Series, we have this latter type well developed in the

instances described; whether a normal igneous rock, such as syenite and

gabbro, may be formed, despite Harker’s arguments, is a question requir-

ing more data. The rocks in question must be analyzed, and the variation

diagrams constructed for them and for the surrounding magmas, before

anything can be definitely said. It is interesting in this connection, how-

ever, to recall Daly’s*® theory of overhead stoping, which postulates exten-

sive assimilation of the blocks which drop into a magma as it ascends.

He has further developed this idea of abyssal assimilation,’ until it re-

sembles the marginal assimilation hypothesis which is supported by many

French geologists.** If something of this kind has actually taken place

in the Cortlandt Series, it is probably a fairly common process, since it

has suggested itself in many parts of the world.*® On the other hand, of

course, the location of these rocks may be merely an accident of intrusion.

Little can be said regarding the mutual relations of the four funda-

mental magmas, since they are not yet sufficiently clear. The volcanic

phase, the first in the normal cycle of igneous activity, is wanting. The

normal order of intrusions in the plutonic phase is that of decreasing

basicity, and this seems to have been followed in the Cortlandt. As al-

ready stated, the granites seem to have been intruded last, both from the

evidence of the acid dikes in the more basic rocks and from their almost

isolated position. The pyroxenites and norites, from their intimate rela-

tion in several places, were probably formed at about the same time,

although not simultaneously, for their association is only local. The

pyroxenites probably preceded the slightly more acid norites but had not

haidened when the latter appeared. At the same time, however, the evi-

dence of the variation diagram, as indicating a much more complex rela-

tionship, must not be lost sight of. The diorites were probably still later.

Here again, however, we are confronted by the suggestion of hybridism in

the occurrence of gradation from hornblende norite into diorite, for ex-

ample, and a further discussion of these problems would be fruitless in

view of the comparative scantiness of our data.

«6“'The Mechanics of Igneous Intrusion,’ Amer. Jour. Sci., (4), XV, 269, and XVI,

107. 1903. Also “The Geology of Ascutney Mountain, Vt.,’’ Bull. 209, U.S.G.S. 1903.

47 Amer. Jour. Sci. (4), XXII, p. 195. 1906.

8 See for example ‘Contribution 4 l’etude du granite de Flamanville,” Bull. carte géol.

France. 18938.

See M. WepeER, k. Bayerischen Akad. der Wissenschaften, Dec., 1910.

ROGERS, GHOLOGY OF THE CORTLANDT SERIES

—_

. Granite.

He OO bo

. Gabbro,

Rocks, 1908, p. 72, No. 3. Symbol, IT.

. Norite proper, 14 miles S. of Peekskill.

. Granite, Cornell Dam Quarry.

municated by C. P. Berkey. Magmatic symbol, I.

ANALYSES.

Mohegan Quarry. Idem.

S. E. of Salt Hill.

5. 3.4. <Andose.

Lassenose.

Analyst, Elwyn Waller.

4. 2. 4.

Privately com-

Lassenose.

(Near Toscanose. )

. Diorite, Buchanan. Analyst, G.S. Rogers. Symbol, III. 6.3.4. Limburgose.

Analyst, H. T. Vulté. J. F. Kemp, Handbook of

5. 8. 3. Shoshonose.

Analyst, G. S. Rogers.

Symbol, II.

. Biotite Norite, 2 miles E. of Montrose Point. Analyst, G. S. Rogers. Symbol,

Il. 5.4.3. Hessose.

7. Biotite Augite Norite, Montrose Point. Analyst, M.S. Munn. J.D. Dana.

Amer. Jour. Sci. (3), XXII, 104. Symbol, II. 5.3.4. Andose.

Hie 7s 3: 4. 5. 6. Uf

SG ae eee 73.54 73.32 | 42.45 54.72 | 51.49 | 46.10 | 55.34

JIAO Senaeeec 15.20 15.01 16.36 WEIS B20S72 18.66 16.37

Ber. 2. 50 AG 3.20 2.08 1.80 3.00 Ary,

1, 0 eee 81 1.19 9.85 6.038 7.28 9.58 7.04

ESO) oa 0 .03 15 oz 5.85 3.82 6.71 5.05

CAO. 38: 1.69 1.35 9.57 | 6.84 6.71 8.26 7.51

INGE O areet : 4.99 4.27 2.61 3.02 | 3.70 2.57 4.06

HO Rea 2.31 S24 2e29 3.01 2.14 1.59 2.03

H,O+..... .06 les SOS Weeds outa ill STON S Jo aces

pee os ee ey od tae (1 Nee NN 10 10 58

OOP Pree otisie te cats alle etessienaiouh = NOME. Pais S315 soy trace NOME. || pee

Seer Bis .08 .06 12 | ea a 2.26 2IB88 NY. seeseae

Re ee Pace clase pce cilh atorcaw Sys ous BAO NS cia cscyate 3 15 OU. Lape Sete

“Ulign saith anes Big cl Gears ao lie Aes ae eee aera ee 202% | on eer

See reer craraitetiievs sviaeis) WAS asectesae tVACE Hers e/aae s sul EDO | iaista carers

CID face lland Ge da DAIS 2 eS aire Cee neon ae | UA VOPR er eer ieee ee

WinOe se oases trace ACER ereey- valet secyei teae = cle rs cortex SL Stes seen ers 40

1 SEO) Been Oeics Roem Cena eee ok) Get Aelia aiec sarc ETACE osu. ae hteey

SIRO) Scope dallopeeeacied ences mCi bars ait vaya eel eae ErACE le ay etre

Sum 99.21 99.67 99 55 99.34 100 72 100.57 99.65

Mopes

18 2. 3. 4. 5. 6. | 7

ae oon SOAS Ae ce ols is obae ave (ae alraeretas [eldewan 2 ath ws ae ak

Od) ae mee oe 8.3 115).(0) ine eee 17.8 10.6 5.0 10.0

Allpite: 3)... 41.9 36.2 22.0 25.2 30.9 21.5 34.1

Anorth 8.3 6.7 24.5 26.1 33.4 34.8 | 20.9

Moscoy 4.9 AG SM Whe Sc 4. OSS ole Rdamee fa efi oe Neale entero emrarmare

Biotite..... oe 4.7 SSSe as cake 4.6 17.0 5.6

Telia teens ots flata cd oles pri Meee Parr ha neeh teh ce PION RPh ent k

Jalili Ooi pel ha Bs Bee or aus © Coal er Dieae| Garner s lth cerns eas

PDS SLY ee ans votes |i a ainda [isis « oo areas orale veretehe « 13.0 10.2 15.9

PONG s-aicycthieveadecr ver Aone hive atdheyanc = [AV Jee Beene, A Ft ty Os Le iTie: | heen ee

ANTHETIN IS A eel Gee othe ae Ges 6 GRO Oe mea Ol Sirk Some eel Gams ohoaal ec ost ea

ic oe 2 | 2 2.6 3.0 1.9 2.6 7

Ilmen...... 2 yy Oita alteetscctetcr eins 4.2 eae ll. cusporesraers

ATTA eA odllG datatoa Gets |6 GE Oo (CHE REA! een ACA eee tb ictoena moe

ANNALS NEW YORK ACADEMY OF SCIENCES

Norms.

1 2. 3 4 5. 6. 7

CR Raa ae 32.0 BOLD S|. she'd serste allie s'e/aible bua] ele aie Glee ke [lore eee mie ee cer

Comet iak ie 1.5 Dd Gree Aloe elle aaron se 2B. s.th ae ee een

OPLie 6s se 13.3 21.7 13.3 17.8 12.2 9.5 12.2

BD oie ates 41.9 36.2 7.3 25.2 30.9 21.5 34.1

BN sciccs s\s'3s 8.3 6.7 26.4 26.1 32.5 34.8 20.3

Ie prerparltae darn &5 Gallloincto.om.n.0.5 SiO ys arkieles ere 2 Slee afate oni mies © rel ee

GIS, ie i4is S cuate ee eee elope 14.8 Gis lewens tee 5.0 14.0

10 asa pe Te 7a dal RRO cas te 20.9 13.8 3.0 11.4

Olssas tes eae seen Roars Bt pee ee a 2.9 15.9 5.3

Tiles ec ouherats 7 7 4.6 3.0 2.6 4.4 1.2

Hl Sse IS ta Ae a aes ae NED “tae 4.2 B.S eb see

AD: cece al aaeieines el os eee ore EO eee ite .4 122.5 e eee

8. Biotite Augite Norite, 13 miles S. of Peekskill. Analyst, G. S. Rogers.

Symbol, II. 5.3.4. <Andose.

. Hornblende Norite, } mile N. of Montrose. Analyst, G. S. Rogers. Symbol,

15. 4. 3. -Hessose:

. Pyroxenite, Montrose Point. Analyst, G. S. Rogers. Symbol, IV. 17. 1”. 2.

Hudsonose.*°

11. Olivine Pyroxenite, 3 mile N. of Dickerson Hill. Analyst, G. S. Rogers.

Symbol, IV. 1%. 17.2. Hudsonose.*°

12. Augite Peridotite, Montrose Point. Analyst, W. H. Emerson. G. H.

Williams, Amer. Jour. Sci. (3), XX XI, 40. Symbol, IV. 27.172. X.

13. Wernerite Schist. Montrose Point. Analyst, G. S. Rogers.

8 9. 10. ID 12. 13.

SiQss silos sete | 60.74 45.91 49.95 49.73 47.41 42.96

Gee: eaten 20.30 | 19.83 6.52 4.01 6.39 | 19.49

Re:Or Ssceen ses 1.23 ASIII 1.50 .70 7.06 1.08

BG ORE acta Orel tele 1.27 8.09 10.41 6.66 4.80 4.89

Mp Omri 3.78 5.13 17.02 20.06 15.34 1.50

CaO eran 7.28 9.32 ei 16.71 14.32 19.32

INGORE occ er 3.82 2.78 .98 .63 .69 2.25

KOE Lee & 2.21 1.23 .o2 ol 1.40 26

EOE es ol 40 65 .o9 2.10 1.24

EGO. oe. ae .10 14 16 SOE? cieete ete 06

CO Seat e ne oe heer .04 25 by STi nah eRe Ors. 4.85

DORN lt lee 2.31 2.71 15 "64.0 |e 2 ees 1.72

JEG Tage ee 15 1.03 none O02) Ss), i: steer trace

OUR nad ames an aerotac .08 TORE N oz. oa ss ciao the | ee eRe 31

Diane <a ckowes 12 .30 odull .30 .49 .64

CRO octal ee pus as 01 10 | {TAace:/) liksecee see trace

MTG eo he 3 B17 (iy Pa ae Ph he | Pe hen oe .05

CuO i, eas cs tes Js i- eas Col oan sec teem ert tYACe: 5) sec een Soe

SILLS S O10 mor 99.75 100.03 100.09 100.738 100.00 | 100.62

% The writer ventures to suggest the name Hudsonose for this rock, in view of its

abundance and typical development in the Cortlandt Series, and inasmuch as Cortland-

tose has already been assigned to a type apparently not found in the series.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES

Mopss.

> tonnatHtox

Pee its: jeu Ce we le Wy [voo) > i 4

; atid 2 iii ita i i iaNndiisdtin

s Sy ee ee ee re ke

ONS deme omy eh ccpmian TS. Sieh toe teelte

. C1 bX ONWayialien Gia 80) sO OT i Os 0m 96

N He Ar ee ee ib be SS iais kb e ere tan ae

re eee Kd eles 6 Me wee he ba, «808 ee) ote

AN SMe perimlleNeioy 6G 6 o oO ¢

; is Neon Ne OE et oe le a at

— AYO a bmn ae ea ee ee

—

POON (HOM CAMHMO 7 ttl:

. rk ©1090 CO =H re ai eS le iochage

f= Sar) fon eae a

eee otk

CN heientoy Cer Wohi 2 OU oe on 6

: YAaNCeyes we) UNiGlive | 2 6 Oo 0 9 0 85

o> N oO fad eee ei ele (6 ere pee Ay 8 haw Bie

BEV CN ISN SSCA RE Be ae Ue Gn amet yor a0

2 ONAIGD +00 Natl ie De OMA I nA GPRIOMG

[o a) CA Of . . . . : . R : .

renee eet ast

aie Rene eae ee “LEE so

. s . 5 . s ~~

“ops S5as ca B8aauss

SROBQHAS EE TEAS QEOgS

ele cates kmwabe coach ss

OAdaeHt qtOdes ea nosed

Pyrite...

Norms.

13.

12.

er ee oe 8, a ers

Swe Ow oN Vi Chit metre SC ON

A Ge Oe Ch ed Ce) mec

A OC, Be Re, et PO ahr

ni 6! 0) els ee aye Ve ee

Ue BO eC, i) aT cre

Ok dhe th SPC eT othe OMe

a welaser ep ene) te) we cemuce

Wipe tes yee Xen Wet tevs Ne! 67 fe

10.

rH AIDOMN OR 1

Ve) N

NODDODAOMH :

rio of

CLO a) OMmLnn Lees Cuma)

Th. Oe CRS DF OP Os oO

mie <6) Xe Ne) ee pe = seve Vel nec fel

ee FL Oa: AP te emai maa oS

We te. ale! Me OS Pe) Monee s Se bev

Cette oe Vem th ety gh Ske tv

pee, wt Ge ee

A phpee> ee Ce Ciba) UCase

bat Condber Sel se, 6). el w re 6

Gg i QO wren

Cie Dine 0, See Se Om Ora:

14. Spinel Emery, high grade, Buckbee Mine. Analyst, G. S. Rogers.

15. Pure Emery, Dalton Mine.

16. Feldspathic Emery, Salt Hill.

17. Quartz Emery Schist, Salt Hill.

18. Emery. Analyst, T. Egleston.

ANNALS NEW YORK ACADEMY OF SCIENCES

p. 197 (No. 1).

19. Idem (No. 4).

20. Idem (No. 7).

Analyst, G. 8. Rogers.

Analyst, G. S. Rogers.

G. H. Williams, A. J.S. (3), XXX@IE

14, 15. 16. 18. 19! 20.

SiOn sere ia $8454, 10.60") ) AVG As a sal cet 25 elle ee

SA Ose we 68.14 59.22 43.72 31.93 37.43 46.53

IRONS Ge oco- 1.438 16.66 13.75 18.19 Al aily/ 32.31

INS OG sooar 16.25 14.02 LOAM, S2AO} es wisstetamia ltr ee ercinreel| ste coeaeete

any Fe ORE Pc = 10.02 3.54 8.59 7.41 7.20 9.43

CaQy. iaks~e trace trace MOB 4") “2iOB Gace aie Sales om dae eile ee eae

NacOP a trace trace BOO Sie oe ROD Weve erst aeral|'s, caste een a fisre eeee

Ow snese trace trace Be | dO ll ieee tas aleee ae nese eee

H,O+ 1.15 2.65 SBD = TW ental ee as ltes cates hee eee

RO snaee SiR .05 S07 NE ALO) ates Set aky sad ricio celts

RI Ok Beer 1.41 3.28 1.43 Wel .65 ol

LENO} dao 06r trace trace trace trace trace trace

Site cere 05 .06 .07 .04 .02 01

Cr Ogee ace 04 trace AG ¥) ~ ¢ GEACE Tie is.:5 5 sic, |lneets 2 oie

MnO cee trace .06 SUE a LOS eee REP es 5 occ

Magnetic

=o Sian a eres Perea, 3 ER 34.200 | 19.81 | 8.98

Siliceous

Bee aoe [eet wa tae Sem, A 6.42 | 1314 | 2.42

SUE eel ee 100.54 100.38 100.51 99.31 99.42 | 100.19

Moves

14. 15. 16.

Oize 1.9 8 3.9 36.7

Orie eles tote several emer cike eee 2.8 Ee)

SAN DIG See Sale aye ene aielllo ake sexta tate 4.7 4.7

PAGVOTUI: S52) sralete craters sivas eocers Be) 10.0 |

SOUT CET 8) RAS acl iaeioenc [es aon'5 9 5 4.9 |

Spinel ..... 75.0 Dili 61.3 24.3

Corundum.. 19.5 45.2 9 5.0

MiGic cus Dal pA ON 8.8 |

Ilmenite... 2.6 6.2 2.6 2.8

51 Probably derived in considerable part from the agate mortar.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES

21. Brown Norite, Buckbee Mine. Analyst, G.S. Rogers. Symbol, IV. 11. 1’. 2.

Cookose.

22. Sillimanite Schist, Dalton Mine. Analyst, G. S. Rogers.

23. Manhattan Schist. Composite analysis of five specimens from various points

beyond the borders of the series. Analyst, G. 8. Rogers.

24. Manhattan Schist (least altered), Crugers. Analyst, F. L. Nason. G. H.

Williams (3), XXXVI, p. 259.

25. Manhattan Schist (nearer contact), Crugers. Idem.

26. Manhattan Schist (about 750 feet from contact), Crugers. Idem.

27. Manhattan Schist (on contact), Crugers. Idem.

Pel. pe 23. 24, 25. 26. PA

SION Rasa: 52.27 28.81 57.94 62.98 61.57 55.12 40.16

aN Og efi s 3 6.81 46.95 21.70 16.88 19.53 24.32 29.50

He Owes ae 2.48 10.00 1.57 2.48 5.44 6.13 19.66

IRON Sooaee 10.01 7.40 5.90 5.00 2.61 4.99 5.80

MeO-...3 2: 23.29 2.08 2.49 1.58 1.90 trace trace

CAO Sa sch: 2.04 trace .08 trace trace trace .85

Ine Oana 87 50) 1.74 3.02 3.48 Zeal 1.46

Ke Ope a 25 sath 4.68 7.45 2.14 2.83 1.36

EOS oe 5s 61 .64 OL ThA Ss Buren dka rane tall cera a ea cee es ore Reco

H,O— 14 .07 BAD [tsps oh < ee ee eneh| es feeo aya ver rele ose av oreonl [ioe a mens

CORE eres 13 none TENG Sica lites ct seers cx cvl | atens\ ny esenerli mcr eoe te suki | el ears aaa

ERY OR ers ores 40 2.76 TCD Fe ae a 1.53 ZAG Es ee

lea Or deere trace none trace trace trace trace trace

Sata icteis eve 05 SOFA sll here recite .08 85 1223 .82

CRO, aac 2 22 Eilers crete Hs. S/d Mel tee uthee We allen re to Aes ake etonan

WhiOwog sues trace .29 ola ee eee ee te al leks a Protein (arbre ener IIe ra Ges. c

Sule ee 99.57 100.06 100.26 99.47 99.05 99.79 99.61

Mopes.

Bille i. 2B}

GZ ae cece a | ses Sarena iovel| atershst tts 61 19.2

@Ortheenen HTL 52 0.6

IN SiKE so co oe ies 6.3 14.7

Anorth INE Sia | eetaa wreteye 2.5

Whoa ssllogosde gaclonenede cs 38.5

BIOUILERE al eee atertote al eeterencvekons 14.9

Bronzite... USVEP No eran Oo. | SRDS ee

Cordieriters|acer ae sc ZS) cea

Sillimanite:|2...-...< Sica)” | Roxane

Corundunga eee 15} ae ene

1 ie eer 3.5 14.4 sft

Ilmenite... 8 5.1 1.9

66 ANNALS NEW YORK ACADEMY OF SCIENCES

GEOLOGY OF THE EMERY

Emery has been mined in this district for the last twenty-two years;

and at a previous period, an attempt had been made to use it as an iron

ore, but being so high in alumina, when used by itself, it of course hope-

lessly clogged the furnace which was erected, and the undertaking was

abandoned. Since 1889, however, many thousand tons have been taken

out for use as an abrasive, although the so-called Peekskill emery is the

poorest variety on the market.

Isaac McCoy was the first to mine emery, and on his land, on the central

and southern part of the hill east nortiveast of Pleasantside, is the most

pretentious mine in the district. He worked in a small way himself for a

while and then leased the property to the Tanite Emery Company of

Stroudsburg, Pa., who kept twelve or fifteen men busy at intervals for

about eighteen years. The Blue Corundum Mining Company of Boston,

Mass., has also done work on his property. The Keystone Emery Com-

pany of Frankfort, Pa., started work a little later on the Oscar Dalton

property (on the northeast slope of the hill above mentioned) and on the

land of John H. Buckbee, which is located on a hill a mile southeast of

the last. The work here was later taken over by H. M. Quinn of Phila-

delphia, Pa. The Tanite Company also leased the land of David Chase,

just southwest of Mr. Buckbee’s property. The work done on the deposits

which are found on Dickerson and Salt Hills is mostly of a private nature,

although the Lombard mine was once quite productive; and the same

applies to the small pits which are located in the biotite norite area north-

west of Crugers.

The work on the emery deposits has been so scattered and irregular

that no estimate of the total amount mined can be made. Work is at

present going on in the Dalton pits, in a new cut at the McCoy mine

(Plate V, fig. 1) and in the southeastern area. A thousand tons or more

are lying in the pile at the Dalton property and an equal amount in the

emery yard at the Chase mine. The ore is hauled by teams to Peekskill

and is thence shipped chiefly to Easton, Pa., as the market demands.

Most of the mining is of a very primitive nature, being chiefly open

cutting and stripping; after a small amount of work has been done, the

emery usually pinches out. It occurs chiefly in veins and pockets but at

the Buckbee mine in fairly well-developed lenses. Here more extensive

mining was carried on; tunnels were driven in several places, and a small

. shaft was sunk, but the timbering has given way, and the whole work has

caved in. At the McCoy mine, the largest pit is about 75 feet long by 40

feet wide and is 80 feet deep. The work here was stopped partly on ac-

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 67

count of the water, which fills the pit to a depth of about 30 feet. Just

at this level, two adits have been run, at either end of the pit; one of

these opens a 30-foot pocket of emery, the other merely runs out on the

side of the hill. A derrick was used to raise the ore, which was run down

to the road on cars. This was the only place at which an attempt was

made to use a steam drill, and it proved a nuisance. The usual method

consists of enlarging the cracks in the ore by hand drilling, shaking it up

with dynamite and extracting it with pick and bar.

Williams was the only geologist to discuss the emery in any detail;

Dana*’ only briefly mentions it, since in his time it had not been mined

except for iron. He notes the thin magnetite beds in diorite on Crugers

Point, which Williams has shown to be merely metamorphosed inclusions

of schist,®* and states that the ore is chloritic, whereas Williams®* has

shown in his discussion of the emery that the green mineral is pleonaste.

The latter, in his first description, states that the veins are segregations

in the norite. He describes the ore from a microscopical standpoint,

mentioning the various minerals which the present writer describes below ;

and gives seven analyses of the emery, partly chemical and partly physical,

but makes no mention of the localities from which the specimens were

taken, or what varieties they were, etc. In his paper on contact meta-

morphism, however, he abandons his former view. To quote: “The

isolated inclusions (7. ¢., the metamorphosed schist inclusions at Crugers)

“of spinel and corundum are almost identical with the more extensive de-

posits of the same character occurring near the southern border of the

norite region farther to the east and described at length in a former

paper. Their origin in both cases is without doubt essentially the same.”

It is somewhat unfortunate that his previous view, based on fewer

‘observations, should have found its way into the literature.

J. H. Pratt®* gives a condensed account of Dana’s work and of Wil-

liams’s work on the norites and evidently relies largely on the latter for

his statement that the emery is probably due to segregation in the norite

along the borders of the magma, in a fashion analogous to that of pyrrho-

tite, etc. The fact which sustains this view is given as the “gradual

transition of the spinel, iron ore and emery into the normal norite’—a

fact observed by Williams. He does not appear to have consulted Wil-

liams’s other papers; but from the description of the series given above,

52 Amer. Jour. Sci., (3), XX, 199-200. 1880.

68 Amer. Jour. Sci., (3), XXXVI, 261 et seg. 1888.

54 Amer. Jour. Sci., (3), XX XIII, 194-199. 1887.

5 Loc. cit.

58 Corundum and its Occurrence and Distribution in the United States, Bull. 269, U. S.

‘G. S., pp. 41, 93 and 137. 1906.

68 ANNALS NEW YORK ACADEMY OF SCIENCES

it is evident that no broad deductions can safely be made from the study

of one member, aside from the fact that in only one place does the ore

occur in a norite area. F. W. Clarke®* quotes Williams as regarding the

ore as a segregation, as does also J. F. Kemp.°* Lagorio,®°*® however,

refers to the deposits as in contacts, although his reference is to Wil-

liams’s first paper.

PETROGRAPHY OF THE EMERY AND ASSOCIATED ROCKS

Perhaps the description of the several types of occurrence may be

facilitated if the ore and some of the rocks peculiar to the mines are

considered first.

Spinel Emery

In the hand specimen, spinel emery is a heavy black fine grained aggre-

gate, with dark gray crystals of corundum appearing in the best varieties ;

sometimes the corundum is pink

(approaching ruby) and sometimes

blue (approaching sapphire). Un-

less these crystals are large, however,

the amount of the mineral present

cannot be distinguished. ‘The emery

often has an excellent fracture and

closely resembles the Turkish emery.

In thin section, it is seen to consist

of an aggregate of pleonaste, corun-

dum and magnetite. The pleonaste,

except in the highest grade of ore,

constitutes the bulk of the rock. It

Fig. 4. Relations of Spinel, Corundum 1S 10 rich green grains of irregular

and Magnetite in Emery. Sl. 326 — shape, closely crowded. They are, of

course, isotropic and have a high in-

dex of refraction, though the surface appears smooth. They often carry

magnetite inclusions in the form of very fine parallel rods. The corundum

is present in all amounts; it may constitute half the specimen, or it may

be lacking entirely. The grains, which may occasionally reach an inch in

size, are in this variety colorless in thin section. They exhibit a prismatic

shape, which is usually quite sharp and apparently always crystallize

before the spinel. They are always strongly cracked and are often con-

57 Data of Geochemistry, Bull. 330, U.S. G.S., p. 278. 1908.

58 Ore Deposits of the U. S. and Canada, pp. 61 and 173. 1906.

5° Zeit. Kryst. Mineral, vol. 24, p. 288. 1894-95.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 69

siderably altered to a hydrated mica (damourite or margarite), so that

their double refraction is destroyed. They give the usual tests. The

magnetite is in somewhat smaller grains scattered among the spinels, and

it is usually not nearly so abundant. It seems to have crystallized simul-

taneously with the corundum (fig. 4).

Williams mentions an analysis of the spinel “whose results were unfor-

tunately lost,” but from the fact that the analyst remembered that the

MgO ran only about 9 per cent., he concluded that the mineral closely

approached the ferrous aluminate, hercynite. In the analyses (Nos. 14

and 16) of the emery, made by the writer and given above, the spinel was

chemically separated from the magnetite and ilmenite, and the analysis

of the mineral as recast gave the following results:

(14) (16)

INOW spico pbc po0do A OOCUOOOOED ED ODIO COaOOS 64.86 65.19

INXD) 6.650ib000000 00 bo CORO COO ME RCO aE OO ODIOES 21.78 20.78

WIE) pcecedaccos0o cos DObOOUOOUOboeoGnd50 DO 13.36 14.03

From the analyses of pleonaste given in Dana’s Mineralogy, there

appears to be no reason for doubting that the spinel in question is of this

variety, and not a hercynite.

Pure Emery

Pure emery, which has a peculiar reddish black tint, is rare. It has

the fracture and other characteristics of the other variety, but no corun-

dum can be distinguished. In thin section, however, it appears that

corundum is the chief constituent. It occurs in small square grains which

contain reddish brown inclusions in great abundance. These may com-

pletely fill the center of the crystal, or they may form a ring, as is often

observed in leucite. They are probably ilmenite. The corundum is

badly altered in this rock, as shown by the high percentage (2.65) of

water in the subjoined analysis. The magnetite constitutes about one

third of the rock. Spinel rarely occurs; biotite has been noticed in small

amount.

Feldspathic Emery

Feldspathic emery resembles the spinel variety when well developed ;

and a surprising amount of feldspar may be present without being no-

ticeable in the hand specimen. Streaks of almost pure magnetite may

occur in this ore, which was the one formerly mined for iron. In thin

section, the plagioclase is seen to make up from one third to one half of

the rock. It is usually basic, though not always. The corundum is scat-

70 ANNALS NEW YORK ACADEMY OF SCIENCES

tered through the rock, usually associated with the more basic minerals ;

it is generally of the typical colorless variety, occurring in medium-sized

idiomorphic crystals. In ore from certain places, however, the corundum

is deep bluish green in color, resembling glaucophane, with strong pleo-

chroism: E light greenish yellow, and O dark greenish blue. It then

occurs in stouter crystals, with less distinct outline. Once recognized as

corundum, it is always unmistakable. Spinel is typical and abundant, as

is also magnetite. Sillimanite, in long blades and in a confused fibrous

mass (fibrolite), is often present.

Quartz-Emery Schist

Quartz-emery schist is most closely associated with the last variety,

which grades into it; it is also, however, found in the spinel emery mines.

It is usually consigned to the dump as too poor in emery. Quartz streaks

run through it in great abundance and give it a distinctly gneissoid ap-

pearance, and it was this resemblance to a schist or gneiss which first

suggested (to Dr. Berkey and the writer independently) that it might

possibly be an altered inclusion. This texture is brought out to a certain

degree in the accompanying photograph of a typical hand specimen

(Plate V, fig. 2). It is much like the feldspathic emery under the micro-

scope, differing chiefly in the abundance of quartz. This is in large inter-

locking grains, and it may compose one third or more of the rock. It

sometimes carries an abundance of rutile inclusions in sharp yellowish

needles. The corundum in this rock is often of the peculiar variety de-

scribed under the last; and the rock may (at the Buckbee mine for ex-

ample) become very sillimanitic.

Besides these distinct varieties of ore, corundum or spinel, or both,

may occur in the wall rocks, such as diorite, norite, etc. In these rocks,

the corundum is usually in more irregular crystals, and the common

colorless variety may show pleochroic blotches of an ultramarine blue.

Norite Proper

While norite proper must bear the same name as that given the more

widely spread variety, it is in reality a somewhat different rock. In the

hand specimen, it usually is of a soft brownish color, being apparently

composed of enstatite and feldspar, with the former generally in excess.

The rock is perfectly massive. In thin section, the dark mineral is seen

to be orthorhombic pyroxene of a peculiar type. It is usually more or

less idiomorphic, occurring in rounded oblong crystals, a habit not seen

in typical norite. The pleochroism is weak (X yellowish pink, Z grayish

ROGERS, GEOLOGY OF THE CORTLANDT SERIES ral

green), giving the mineral a curious washed-out appearance. The analysis

shows it to contain 11.5 per cent. of FeO, so that it is bronzite. It may

become very abundant, so that the rock is almost a bronzitite, which was

never noticed in the typical norite proper; and magnetite, moreover, is

almost lacking. The writer has three slides containing the contact of this

rock with spinel emery. In every case, the feldspar is continuous over the

line; in two the spinel and magnetite are sharply segregated from the

norite, while in the other the minerals mingle for the space of about 1

mm. The feldspar of the emery is somewhat serictized, and its amount

is surprising, considering the black heavy aspect of the ore. In one of

the slides, a line of spinels diverged from the main mass and ran out over

the norite, crossing the bronzites indiscriminately.

Sillimanitte Schist

Sillimanite schist is usually light gray in the specimen and of a fine,

somewhat fibrous texture. The blades of sillimanite, however, can gen-

erally be distinguished, and they

sometimes reach one half an inch in _ Cordierite

length, or rarely an inch or more.

This rock is extremely tough and so

hard that it will turn a drill at times.

The “ore” from the Dalton property

and from the latest cutting of the

McCoy mine is largely this rock, al-

though it is a poor abrasive, powder-

ing when ground.®® Under the mi-

croscope in typical cases, it is seen

to be made up chiefly of sillimanite.

This is largely fibrolite, with the

blades scattered more or less abun- Fig. 5. Sillimanite Schist. Sl. 304

dantly through it. These latter are

blotched with a fine brown dust in places, and they also carry abundant

magnetite inclusions. Cordierite is almost always present in varying

amount; its biaxial figure and the strongly pleochroic yellow halos which

surround its inclusions serve to distinguish it from quartz. Allanite, in

large brownish prismatic crystals, is often present. It shows parallel

extinction and good pleochroism from yellowish brown to greenish brown,

but the color, pleochroism and birefringence often vary even in one

Sillimanite

60 Mr. John H. Buckbee rather aptly characterized this rock to the writer as being

from a practical standpoint, “like gristle-—neither bone nor meat.”

ire ANNALS NEW YORK ACADEMY OF SCIENCES

crystal. Corundum, in any of the three varieties, may be present, and

garnet (almandite) has been observed. Quartz was found in one instance.

Magnetite and ilmenite are usually abundant.

In the rock just described, the important minerals are of a peculiar

type, but the relation of this abnormal schist to true igneous rocks is indi-

cated by the fact that it often contains hornblende, biotite, hypersthene

and even feldspar, chiefly orthoclase. In fact, a pure sillimanite schist

is more rare than one containing a little biotite or hornblende, and in two

cases a hornblende pyroxenite was found which carried merely subordi-

nate amounts of the sillimanite and its associates.

Other rocks are found associated with the ore, such as quartz gabbro,

etc., which have not been before considered, but their peculiarities’ are

indicated by the names given. The statement that shearing is often

apparent at the mines may be recalled in this connection; in several cases

the wiiter mistook for mica schist rocks which appeared under the

microscope to be diorite and norite, and the schistosity was so striking

that he was led to record the dip and strike. Faulting has also often

been seen, and the ore itself is always at least cracked.

TYPES OF OCCURRENCE

Gradational Type

In the area of biotite norite which lies northwest of Crugers, a number

of small emery pits have been sunk. The ore seems to be widely devel-

oped in this district, but the veins are all small and rapidly pinch out.

The ore is of the spinel emery type; it is sometimes in sharp veins, pet

more often it grades into the quartz-emery schist.

The small pit which offers the best opportunity for study lies 250 feet

due south of the point at which the mica schist inclusion outcrops on the

post-road. In this pit, an unmistakable quartz emery schist with streaks

of coarse red garnet lies to the north, 7. e., nearest the schist inclusion ;

it is adjoined on the south by the remnants of the ore, about two feet in

width. This ends abruptly as far as could be ascertained in the field;

but specimens of the adjoining rock taken one, two and four feet from

the contact were respectively spinel norite with garnet, spinel norite and

norite. This rock was of the ordinary variety described on page 30, with

the spinels merely superposed upon it. This gradation appears to be

the typical association of the emery in this district, although the cuttings

are not confined to this distance from the large schist inclusion. The

ore is important chiefly from a theoretical standpoint; only a very lim-

ited amount has been taken out.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES we

Emery Schist Type

Emery schist is a very common type in the southeastern part of the

Cortlandt area, in the Dickerson and Salt Hill region. The country had

here been widely prospected for iron at the time when it was thought

that the ore was magnetite. The emery has been mined on the roadside,

at the southeast corner of Salt Hill; and it is said to run more or less

continuously over the mountain to the north. Here it appears to trend

more to the east, being mined in several places on the eastern slope of

Dickerson Hill. This seems to be the chief vein, but there are several

other outcrops, some of which are along the border of the district to the

west and east.

In all of the outcrops visited by the writer, the ore is of the same type;

a vein of black emery (which in thin section is seen to be the feldspathic

variety) is bordered on either side by the quartz emery schist. More

often, perhaps, no distinction can be made between the two; the emery

merely becomes quartzose in streaks. Bands of pure quartz a foot or

more across are not uncommon. The quartz streaks are not always

straight; they are often strongly contorted and crumpled. The opening

on the road above referred to is within a hundred feet of the border of

the series, and possibly less; and so closely does the emery in all but

_ color resemble a coarse schist, that it was difficult to decide on its iden-

tity offhand. To the north of this, about a quarter of a mile, is a small

inclusion of mica schist, strike N. 20° E. 80° E.; and 75 yards south of

this outcrop, emery is again taken out. The strike of this inclusion is

very close to the general strike of the schist in this district and also of

the emery schist. The latter varies from N. 35° E. to N. 60° E., but it

generally approximates the former. On the west side of Salt Hill, ore

was found in a tongue of pyroxenite which seemed to be projecting

directly into another inclusion of schist.

The ore of this type, therefore, is sometimes on the border of the

series, near the contact with the mica schist, sometimes near an inclusion

of the same in the igneous rocks, while again, although it shows exactly

the same structure, no association with schist was observed.

Norite Type

Norite and the following type are found chiefly in the northeastern

area, where all of the larger workings are located; but in this area, the

emery is always associated with this peculiar variety of norite, or else as

is indicated below under the following type. The succession and rela-

"4. ANNALS NEW YORK ACADEMY OF SCIENCES

tions of the various rocks are, however, much more complicated than in

the previous types, and three varieties may be distinguished.

In one of the Dalton pits, a simple succession from the brown norite

to ore to sillimanite schist was observed. The former is typical; the

latter is coarser than usual, carries some garnet and sometimes shows

crumpling and contortion. ‘The ore varies in width from two to ten

feet and is of the pure variety.

At the Buckbee mine, the underground work is inaccessible, having

all caved; there are two open cuts, however, which seem to indicate the

structure. These are both lenticular in ground plan, curving into the

hill and out again. ‘The ore is an excellent one, being often nearly half

corundum in large gray-brown crystals, and practically fresh and un-

altered. It may at times carry garnet and occasionally pyrite in half

inch crystals. In both of these pits, the wall rock is the brown norite,

which in this case, however, resembles ordinary norite, in thin section,

more closely. It is altered to chlorite in patches. Such of the ore as

has not been taken out occurs in this rock in irregular veins, which often

“break” into pockets. The poorest grades of ore resemble the streaky

quartzose emery, but this in thin section resolves itself into a quartz

sillimanite schist, often garnetiferous. There is a broad irregular band

of coarse pyroxenite running horizontally through the cliff of brown

norite, but it seems to be a local segregation, having nothing to do with

the ore. A vogesite dike appears in both of the pits. The succession

here may or may not be as in the gradational type; it would seem rather

that the sillimanite schist encloses the ore and is itself surrounded by

the norite.

In the two other large pits on the Dalton property, a more complicated

relation exists. The wall rock is a quartz gabbro; this becomes a much

chloritized pyroxenite of the usual type, which passes into brown norite.

The latter is at first chiefly hypersthene, but it becomes feldspathic near

the ore, which is of the pure variety. Sillimanite schist is plentiful; in

one case, it occurs in irregular fine-grained masses on the other side of

the ore, while in the second case, it seems to occur mingled with the ore.

Kssentially the same succession is to be observed in the latest cut of the

McCoy mine. This is about 30 feet wide by 45 feet deep, and 35 feet

high and gives the complete cross section. At a short distance away,

the whole ledge resembles a sandstone with vertical dip, owing to the

regular zones in which the types occur (Plate V, fig. 1). Starting from

the west, the succession is biotite norite (typical), sillimanite schist,

“ore” (sillimanite schist with spinel, very fissile, or “slatey’ as the

miners call it), then the typical sillimanite schist again; this succession

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 5

is then repeated on a smaller scale, with an exceptionally fine grained

biotite norite. This is evidently an apophysis of the larger mass project-

ing into the schist. On the east, the wall rock is a hornblende pyrox-

enite, which seems to be the real country.

Micaceous Type

Two of the three larger pits of the McCoy mine and most of the

smaller prospects scattered over the hill are of micaceous type, in which

a sheared micaceous rock is associated with the ore. The ore from the

large pit of the mine, and the chamber connected with it (described

above), was the best mined anywhere in the district, except possibly

Buckbee’s. Small areas of pink and blue corundum emery were found

in this pit. In the other pits, it is a lower grade of spinel emery, seldom

showing the corundum crystals.

In the large pit, which is about 80 feet deep, of which depth 30 feet is

below water, the sides are not quite vertical, dipping 70°-85° S. This

is most apparent at the eastern end, where the wall rock is very mica-

ceous, dipping as above and striking N. 80° E. At times, it strongly

resembles Manhattan schist; again, the mica may be all biotite, and the

feldspar may diminish until the rock is practically an aggregate of ande-

sine, with less orthoclase and a great amount of biotite. Scattered

through these minerals is abundant spinel, with varying amounts of

corundum. At the west end of the pit, the wall rock is diorite, with

strongly pleochroic hornblende, which is in idiomorphic crystals often

absorbed and embayed and considerably in excess of the biotite. To the

east and south of the mica rock lies a pyroxenite, carrying hornblende

and some biotite; and it may also contain sillimanite and garnet. The

large veins of ore, if not worked out, are below water level and inacces-

sible; but the mica rock carries small veins of spinel emery. Men who

had worked this pit stated that the ore was always found in this black

mica rock, often associated with garnets; that occasionally white mica

was found with it, and that in the ore itself would sometimes be found

little bunches of black mica or green mica (clinochlore ?). On the

dump, a large amount of the quartzose-streaked emery was found, but it

was not observed in place.

About 50 feet to the east of this pit is the head of a cut extending 300

feet east. It is 30 to 60 feet wide, becoming at its head 50 feet deep.

The wall rock on either side is a typical looking diorite, as in the large

pit. Then, starting from the diorite, comes a rock which entirely resem-

bles a mica schist (N. 80°-90° E., 75°-90° S.), but which proves in thin

6 ANNALS NEW YORK ACADEMY OF SCIENCES

section to be a biotite norite carrying abundant spinel and some corun-

dum. Adjoining it is a poor ore, which is a sillimanite schist. with abun-

dant cordierite. Then comes a brown norite, which is followed by a

good spinel emery; and hornblende pyroxenite separates this from the

diorite. This is, however, an ideal cross-section, found only in one

place; in other places, the succession is less regular and often repeats

itself. In a 40-foot cut to the northeast, the structure is exactly the

same, except that the spinel norite is replaced by spinel diorite, as in the

large pit, and that the brown norite is lacking.

In these three pits of the McCoy mine, then, the ore occurs in a

schistose, spinel-bearing micaceous rock, either diorite or norite, and is

associated with sillimanite schist and hornblende pyroxenite.

The Chase mine, which is the last to be described, appears to be similar

in structure to these; but garnets in great abundance are found there.

he ore occurs in a ledge, flanked on one side by a corundum-bearing

sillimanite schist, which at times becomes very coarse; and on the other

by a very biotitic rock, similar to that in the large pit of the McCoy mine,

except that it carries corundum in irregular masses instead of spinel.

In irregular association with the ore occurs a feldspathic rock contain-

ing abundant garnets. From a rotten zone in this rock, well-formed

trapezohedrons one and a half inches in diameter can be readily ex-

tracted. This rock in thin section is seen to be chiefly a basic altered

feldspar, with a little biotite, spinel and corundum and the large garnets.

It may occur in the ore in irregular masses, but is chiefly separate. In

the ore itself, garnet occurs abundantly, both as crystals and as rounded

flow-like masses, often four or five inches across. The ore is somewhat

peculiar aside from this, being more than half corundum in fairly fresh

white crystals; for the rest, it is made up of spinel, magnetite, allanite

end garnet. This is a high grade ore, but it is injured by the abundance

of brittle garnet and allanite. Around the cutting for a few hundred

feet, a very garnetiferous rock extends, whose dark mineral is probably

an altered hornblende, constituting a diorite. The true country rock is,

however, a pyroxenite, which varies an indicated on the map.

The foregoing description of the emery deposits may be epitomized as

follows:

1. The ore usually occurs in a region in which mica schist inclusions

are abundant and often within a hundred feet or so of such an inclusion;

and the largest mines (McCoy, Dalton, etc.) are within 1000 feet of the

border of the Cortlandt Series.

2. The ore is always in sharply defined veins, pockets or lenses, but its

constituents often occur disseminated through the rocks immediately

adjacent.

ROGERS, GHOLOGY OF THE CORTLANDT SERIES vara

3. The ore is immediately associated with abnormal rocks, containing

sillimanite, cordierite, garnet, quartz or allanite, which are found no-

where else in the area, except around certain schist inclusions near

Crugers; or more rarely, it adjoins rocks which are normal except for

the spinel scattered through them. There is often a great abundance of

biotite around the ore, which is also characteristic of these inclusions.

4, These rocks often exhibit evidences of shearing, faulting and crack-

ing, which is rare in other parts of the district, except around schist

inclusions.

ORIGIN

In reviewing the possible modes of origin of the emery in the Cort-

landt Series, it appears evident that only two are plausible; the emery

may be pyrogenic, due to magmatic segregation, or due to the absorption

of some of the numerous schist inclusions in the area. The evidence

may perhaps be more effectively weighed if we first briefly consider the

artificial production of corundum and its occurrence in other districts.

Artificial Production of Corundum

Several French geologists have been especially active in the field of

synthetic mineralogy,®* but their results in this connection are of little

importance. Many of them worked with alumina and artificial salts,

-so that their processes find no equivalent in nature; others dissolved

alumina in various melted minerals, adding in some cases a trace of

ammonium fluoride or some similar salt. These last experiments are of

more importance in a geological way, but they appear to have no appli-

cation to the Cortlandt Series.

The only relevant case that the writer has been able to find of the

production of corundum by the simple fusion of a mineral is described

by Vernadsky.®*? He observed the formation of corundum and silli-

manite in a melt of muscovite, in the course of his experiments on the

genesis of sillimanite.®

Probably the most important investigation from a geological point of

view is that undertaken by J. Morozewicz.** Basing his work on the

6 FouguE and L&vy. Synthese des Minéraux et des Roches, Paris. 1882. See also

L. Bourcsors, ‘Reproduction artificielle des minéraux,’’ in Fremy’s Encyclopédie

chimique, II, pt. 3, p. 63.

On the Sillimanite Group, and the Role of Alumina in Silicates. Moscow, p. 83.

1891. (Russian.) He also describes the formation of sillimanite from clay in “Sur la

reproduction de la Sillimanite,’’ Bull. Soc. france. de minéral., p. 256. 1890.

* Corundum is commercially manufactured by the Norton Emery Company at Niagara

Falls by the fusion of bauxite, previously rendered anhydrous by thorough heating.

& “Experiment. Untersuchungen itiber d. Bildung d. Mineralien im Magma.”’ Tscher-

mak’s mineral. und petrog. Mittheil., XVIII, pp. 1-90 and 105-240. 1898.

738 ANNALS NEW YORK ACADEMY OF SCIENCES

principles of Vogt and Lagorio, he rejected the French synthetic methods

as unnatural and studied the deposition of corundum in the melts of a

large glass factory, where he could control his temperatures up to

2100° C. His melts varied in size up to one hundred pounds or more.

They were made up, sometimes of the pure salts and sometimes of pow-

dered minerals, to approximate the composition of certain natural erup-

tive rocks. They were contained in clay crucibles, whose position in the

furnace might be varied according to the degree of heat desired. A point

often overlooked in connection with this investigation, and one which he

states to be its most serious defect,®° is that 30 per cent. of his melts

were spoiled because they attacked the clay crucibles. Melts high in

magnesia with low alumina and alkalis, were especially prone to this,

deriving alumina from the crucible to form spinel, while those rich in

lime, alumina and alkalis had no effect at the highest heat. The results

of this well-known study prove that the role of alumina in a magma is

entirely analogous to that of silica; if it be saturated with respect to

alumina, corundum will separate out just as quartz does in a granite.

An alumino-silicate magma is saturated when it has the general compo-

sition MeO, m Al,0O,, n SiO, (Me—K,, Na,, Ca, and n=2 — 13)

and m is more than one. If silica is also in excess. (over 13), sillimanite

is formed. If magnesia and iron are present in excess, they will form

spinel; or if silica is also in excess, cordierite will separate. The sepa-

ration of corundum from a magma, therefore, is governed by definite

laws; and minerals, moreover, which have previously been thought of as

due to contact action are thus shown to be pyrogenic under certain

conditions.

Morozewicz in a previous paper® had shown that alumina will readily

dissolve in a molten magma whose composition approximates that of

basic magnesian rocks, and that on cooling, the excess alumina separates

out as corundum and spinel.

Corundum may, therefore, be formed by the action of “agents miné-

ralisateurs” on alumina, or by the fusion of certain aluminous minerals,

or by the solution of alumina to supersaturation in molten magmas.

Origin of Corundum in Other Localities

Since the emery of the Cortlandt Series is always in igneous rock,

only a few similar occurrences of corundum in other localities will be

noticed.*”

& Op. cit., p. 18.

% Zeitsch. fur Krystall., XXIV, 281. 1895. See also LaGorio, idem. p. 285.

87 For a full discussion, see J. H. Pratt, Bull. 269, U.S. G.S., pp. 71-96. 1906.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 79

Corundum is found in North Carolina in peridotite segregated at the

contact of the igneous rock with gneiss. There is a sharp contact with

the gneiss, but the corundum grades into the peridotite. Pratt®* he-

lieves that it separated out of the magma as an original constituent, ac-

cording to the laws cited above, and segregated at the border in a fashion

analogous to that of pyrrhotite or ilmenite. The earlier writers,®? how-

ever, believe that it is due to contact action on the gneiss. Corundum

is also found in plumasite in Plumas County, California, and there is

no doubt that it is an original pyrogenic constituent.”°

Corundum has been found on the contact of granite and clay slate on

Dartmoor in Devonshire, apparently due to contact action," and at con-

tacts between granite and micaceous quartzite near Morlaix, France,

having the same origin and associated with sillimanite and andalusite,

spinel, etc.*?

Corundum in sapphire crystals has been found in a monchiquite dike at

Yogo, Montana. The rock is too low in alumina, however, for it to have

separated out as an original constituent, and it is probably due to inclu-

sions of argillaceous sediment derived from the underlying rocks.7* Co-

rundum and zircon found in basalt and other basic rocks in Haute Loire”

are thought also to be due to the destruction of gneiss inclusions.

Corundum, then, in igneous rocks, may be due to magmatic segregation,

or to contact action, or to the absorption of aluminous inclusions.

Evidence as to Its Formation in the Cortlandt Series

It has been shown above that the emery in the Cortlandt Series is often

streaky in appearance; usually associated with inclusions of mica schist

or found near the borders of the area but not on them; immediately asso-

ciated with sillimanite rocks and other abnormal varieties; showing no

marked preference for any country rock as previously supposed, though

found most generally in pyroxenite, and often associated with rocks which

have undergone metamorphic action. Spinel and corundum may be scat-

tered through the adjoining rocks for a few yards or less whether it be

an abnormal rock or a normal igneous flow.

688 Tdem, p. 81.

6 See for example T. M. Chatard, Bull. 42, U.S. G.S., p. 45. 1887.

7 Bull. 269, U.S. G.S., p. 94. 1906.

71K. Busz, Geol. Mag., p. 492. 1896.

72 A, K. COOMARA—SWAMy, Quart. Jour. Geol. Soc., LVII, 185. 1901.

737,. V. Pirsson, “Petrography of the Igneous Rocks of the Little Belt Mountains,”

Montana, 20th Ann. Rep. U. 8. G.S., pt. 3, p. 554. 1900.

74 A, Lacroix, “Sur l’origine du zircon et du corindon de la Haute Loire,” Bull. Soc.

franc. de minéral., p. 100. 1890.

80 ANNALS NEW YORK ACADEMY OF SCIENCES

The Theory of Magmatic Segregation

That magmatic segregation is possible is proved by the experiments of

Morozewicz, who has shown that corundum may separate out of a super-

saturated magma and that the associated minerals, which were formerly

supposed to be due in every case to contact action, may have a similar

origin. If it has so formed, we should expect it to show probably a close

association with the border of the igneous area, as in the case of the Caro-

lina peridotite, simce such an abundance of early-crystallizing minerals

would be carried to the borders by convection currents, or by some process

akin to Soret’s principle. It shows a rather close association, in some

cases occurring practically on the border, while the most important, mine

is located within a thousand feet of it. Other occurrences are, however, a

mile or more removed. ‘The ore is disseminated to a certain degree

through the surrounding rocks, which are sometimes otherwise normal

and sometimes extraordinary aggregates of the minerals which have been

shown to be associated with pyrogenic corundum under certain conditions.

Moreover, the magnetite which is associated so plentifully with the ore is

known to be in other localities a basic segregation,”® although in such cases

it is usually highly titaniferous. It is possible, therefore, that the ore

has separated out of the magma as an original pyrogenic constituent.

There are certain phenomena, however, which this hypothesis does not

account for. If the emery shows any preference for the borders, as in the

case of the Carolina peridotite, it should show a distinct preference, occur-

ring directly on the borders as a sharp contact but grading into the

igneous rock. In no case, however, was it observed directly on the border

or even on the margin of an inclusion. Moreover, in none of the cases of

pyrogenic corundum cited by either Morozewicz™® or Lagorio™ is magnet-

ite a very important constituent; certainly in none does it occur in excess

of the corundum. Basic segregations of magnetite, furthermore, are

usually very titaniferous,* while in the Cortlandt Series, the average of

Ti0, in the four different emery analyses is 1.90 per cent, while in seven

analyses of the important igneous rocks of the district, the average is 2.15

per cent., running much the lowest in the pyroxenites. Nor does this

theory explain the dynamic action which is so evident at the mines and

almost confined to them; nor the streaked appearance of the ore in a very

7% See for a summary Kemp’s Ore Deposits of the U. S., p. 171, ete. 1906.

7 Min. pet. Mitth., XVIII, 212-219. 1898.

7 Zeit. Kryst. Min., XXIV, 285. 1895.

7 Thus, for example, the ore from the classic locality for this process, Taberg, is

almost an ilmenite; similarly the Cumberlandite of Rhode Island, the titaniferous mag-

netites of the Adirondacks, ete.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 81

important variety, unless indeed flowage after the separation of the emery

is postulated. Finally, the abundant garnets found at the Chase mine

are not explained, nor the biotite at the McCoy and Chase mines; nor is

the association of the emery with quartz accounted for. It would be diffi-

cult to explain how corundum could separate in the presence of quartz,

since according to Morozewicz the two would combine to form sillimanite.

Theory of the Absorption of Sedimentary Material

The possibility of the effectiveness of the absorption of sedimentary

material is evidenced by the fact cited above from Morozewicz, viz., clay

crucibles are attacked at a high temperature by melts rich in MgO and

poor in Al,O, and alkalis, such as pyroxenite. If this could take place at

2100° C. with a low pressure, it appears certain that a molten magma of

such composition could readily attack a more or less attenuated inclusion

of mica schist; and the abundance of these inclusions in the district has

already been shown.”® Lacroix,®® moreover, states the following general

rule: “Si l’enclave différe beaucoup de la roche eruptive par sa teneur en

silice, elle est facilement detruite et l’on n’en trouve plus, en général, que

des traces.” This principle had been applied by him in Haute Loire and

by Pirsson at Yogo, Montana.

In such a case, the process would involve the separation of a small mass

of aluminous sediment and its more or less complete absorption by the

subterranean magma. This would give rise to a local enrichment in

alumina, which would then be deposited according to the laws enunciated

by Morozewicz. The theory thus involves certain features of both mag-

matic segregation and contact action, but it is essentially neither; nor

does it involve the old view which Morozewicz and Lagorio have shown to

be false, that corundum is infusible in a magma and that it merely repre-

sents the unattacked portions of the xenolith. The occurrence is taken by

the writer to be a strong confirmation of the laboratory experiments of

these geologists, rather than to oppose them.

The accompanying analysis of mica schist was made with a view to de-

termining the extent to which this rock could have contributed alumina.

The percentage found (21.70) is only slightly above that in the norites;

pure norite (which is chiefly feldspar) reaches 20.72 per cent., being the

highest. The pyroxenites in which the ore is generally found run about

7JIn this connection, also, the limestone inclusions of the district, altered to lime

pyroxenite in one case, wernerite schist in another and a garnet-tremolite rock in a third,

may be recalled as indicating the capability of the igneous rock to absorb and work over

an included mass. The fact that most of the inclusions of schist practically lack mica,

which must have been absorbed by the magma, may also be recalled.

8 Op. cit., p. 101.

82 ANNALS NEW YORK ACADEMY OF SCIENCES

four to six per cent. The specimens of schist analyzed had undergone

little or no contact action; but it will be recalled that at Crugers, for

example, the schist shows a steady increase in alumina as the igneous rock

is approached, until on the contact the per cent. is 29.50, as shown by

Williams’s analyses. The same, moreover, holds with regard to iron; the

unaffected schist runs 7 or 8 per cent., but at the contact shows 25 per |

cent. Even at a point 700 yards from the contact, the schist runs 24.32

per cent. in Al,O, and 11.12 per cent. FeO and Fe,0,. The reason* for

this increase is a question beyond the province of this paper; the fact

remains that the mica schist in contact with the igneous rock is abund-

antly able to contribute alumina and iron and that even the unaltered

schist is much higher in alumina than the most common country rock of

the emery. The silica of the schist presumably goes to form the quartz

streaks so generally found with the emery. That the emery may be formed

by the absorption of schist is also indicated by its occurrences in the cliff

of diorite at Crugers. As described above, every gradation, from the

unaltered mica schist to one composed chiefly of quartz, biotite, magnetite,

sillimanite, garnet, etc., at the contact and so to the inclusions in the

igneous rock itself, composed of magnetite, spinel and corundum, may be

traced. These thin lenses of magnetite and spinel are so completely

changed that they were once mined for iron. Moreover, pleonaste was

found in the garnet rock from the contact at Salt Hill; and it was found

in one case in an igneous dike cutting limestone.

The sillimanite, etc., with which the emery is associated, may, accord-

ing to Morozewicz, be found when the ore is pyrogenic, although N. H.

and A. N. Winchell state that sillimanite “in igneous rocks occurs only

as a result of absorption of foreign material.” *? It is at least certain that

it occurs abundantly on the borders of the district, where it can only be

due to contact action. Garnet, which is also found as a reaction mineral,

is a similar case. The great abundance of biotite in some of the mines is

paralleled by its abundance at the Crugers contact, and the green mica

which is said to occur sometimes in the emery is also found by Williams

in certain inclusions and identified by him as margarite. Quartz is com-

mon on the contact and also constitutes the bulk of some of the inclusions

at Crugers ; its association with the emery is thus paralleled, whereas it is

apparently not known in association with true pyrogenic corundum.

If this theory be correct, anything like a sharp contact of the emery or

its associates with the mica schist would not be expected, nor, in fact,

81 There seems to be no warrant for thinking that the igneous rock could have con-

tributed this iron and alumina, as is often the case in garnet zones in limestone.

82 Optical Mineralogy, N. Y., p. 363. 1909.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 83

would the mineral be looked for directly on the contact. It is evident

that if the emery is the result of the absorption of the schist, larger masses

of the latter might be looked for in the neighborhood, but if the inclusion

were small and therefore easily assimilated, they would, of course, not be

found. Thus, in some cases we find inclusions of schist near the emery,

but never to the writer’s knowledge nearer than 200 feet; and often they

are lacking.

The resemblance of the Salt Hill types of ore to a black quartzose

schist has already been remarked upon, and the rough approximation of

its strike to that of the schist has also been noticed. In the light of the

foregoing facts, it would appear that these occurrences are merely par-

tially assimilated inclusions of schist entirely analogous to those so well

exposed at Crugers; whereas when the schist is quite absorbed, a much

more complex aggregation of minerals results, containing no quartz except

in a similar streaky mass at the borders. In the case of the emery pit in

the biotite norite area north of Crugers, where the spinel grades into the

surrounding rock, this gradation may be thought of as a quick lessening

of the effects produced by the absorption of the schist, 7. e., the enrich-

ment in alumina is naturally merely a local development, although at the

same time it can have nothing like a sharp contact. This same statement,

of course, applies to similar cases at the other mines. Reference to

Morozewicz’s paper, summarized above, will furnish the conditions under

which the several minerals must have formed.

It must not be supposed, however, that anything like a simple direct

linear variation exists between the schist and any of the igneous rocks; a

segregational process of some kind must be postulated, for even a richly

aluminous schist could not form as a direct hybrid a rock carrying the 59

per cent. of Al,O, found in the richest emery. Or, since at times the

schist retains something of its form and is evidently not entirely absorbed,

it may be supposed that the hybrid formed contains the schist constitu-

ents in greater quantity than those of the igneous rock. The variation

diagram as constructed between mica schist and quartz emery schist and

a typical norite shows a distinct approach to linear variation, but if

pyroxenite be used instead of norite, the curves are often sharp; and it is

evident that more analyses must be made before the crude process out-

lined above can be reduced to an exact determination.

The final point to be considered—the dynamic action at the mines—is

one whose significance is not clear. The fact remains that similar effects

are common, though not universal, around the schist inclusions of the

district, thus emphasizing a further similarity. It was suggested above

by the writer that it might be due to shearing along zones of weakness in

84 ANNALS NEW YORK ACADEMY OF SCIENCES

the area, which would naturally occur chiefly along the borders of foreign

inclusions and conceivably in places where these inclusions had been

absorbed. In the case of the mines, it is also possible that the schistose

rock is an altered inclusion of the schist itself; but the entire absence of

quartz and muscovite and the presence at times of such minerals as

hypersthene tend to refute this view. Finally, it is possible that these

rocks may be inclusions of Highland gneiss; but if they were, their action

on the igneous rocks would be undoubtedly similar to that of the schist,

and, moreover, their associations indicate that they are not foreign masses.

This completes what the writer has to say concerning the origin of the

emery. From the points brought out, it is evident that the ore has in

most cases at least been formed by the action of the igneous rock upon

included material, thus giving rise to a magma supersaturated with re-

spect to iron and alumina; and out of this the corundum and its asso-

ciates crystallized according to the laws formulated by Morozewicz. The

evidence, while sufficient to justify a fairly positive statement, is, of

course, not entirely satisfactory, and the process is not as simple as indi-

cated. ‘There seem to be no definite objections to this theory, however,

while those which confront that of magmatic segregation either through

the mass or at the cooler borders, appear to be insuperable. It might

also be suggested that the highly aluminous Hudson River slates, which

probably once covered the district, furnished the alumina, but no trace

of them has been found in the field. There are inclusions of schist in

the diorite area which are generally associated with a fine black pyrox-

enite, but around which no emery has been found to the writer’s knowl-

edge; although the farmers, judging from its toughness and color, be-

lieve that it is “full of emery.” If it does not contain emery in any

place—except possibly the very narrow zone almost on the contact, which

we would expect by analogy from the effects at Crugers—it merely in-

dicates that these inclusions were particularly quartzose and compact and

retained their identity to an exceptional degree. As implied above, it is

not the large, well defined inclusions from which we would expect emery

to be derived, but rather the smaller fragments torn from these inclu-

sions and probably entirely absorbed.

From this fact, it is evident that little can be predicted concerning

the probable extension of the emery. While pyroxenite, being low in

Al,O, and high in MgO, may readily attack the schist, the usual associa-

tion of the ore with this rock is not an essential limitation; and its rela-

tions to the borders of the area are vague and irregular. Possibly, a

study of the mica schist at the borders nearest the mines might throw

light on the particular variety of schist most susceptible of absorption,

but this question has not been investigated by the writer.

ROGERS, GEOLOGY OF THE CORTLANDT SERIES 85

The technology of corundum is discussed by J. H. Pratt.’* From the

composition of the Cortlandt emery as described above, it is evident that

it is low grade ore. The presence of spinel, sillimanite, garnet, feldspar,

quartz, etc., greatly lowers its cutting efficiency and materially affects its

toughness. At the same time, the presence of spinel, which is 8 in hard-

ness, would not interfere with the manufacture of a vitrified emery

wheel; and for many purposes where extreme hardness is not required,

the spinel emery furnishes a convenient transition between garnet and

true emery. The unsuccessful attempt to smelt the more ferriferous

varieties for iron has been noted above. Even earlier than this, it had

been suggested*®* that the ore might be used as a refractory lining for the

puddling furnace, open hearth and Bessemer converter, as well as an

aluminous flux in the blast furnace in admixture with a siliceous stock,

but apparently these sugestions have not proved practicable.

SUMMARY

The Cortlandt Series is an igneous complex situated just southeast of

Peekskill, N. Y., and is about 28 square miles in extent. Its correlation

cannot be definitely settled, although it is probably late Paleozoic. The

rocks have been described by J. D. Dana and G. H. Williams, the former

attacking them from the standpoint of genesis, and the latter treating

them rather as an aggregation of petrographic types. In the present

paper, they are treated more from a geological standpoint, the differen-

tiations made in the various broad types being directly proportionate to

their areal importance. Reference to the appended map will impart the

distribution and extent of the varieties determined, and on page 58 is

given a diagram which indicates the most salient relations between the

several types. The various differentiations of the norite magma are

most centrally located; they are flanked on both sides by pyroxenites

which extend to the west, and an unknown distance under the river, pos-

sibly as far as Stony Point, where they again outcrop. Between the

norites and the western area of the pyroxenites lies a diorite area. The

granites are more isolated, though unquestionably part of the series.

The most basic members at least grade into one another in many cases,

while at times, sharp contacts have been found. The analyses of the

more important types indicate an unmistakable serial relationship,

although complicated by the entrance of the pyroxenites. It is probable

that the latter were intruded first, followed closely by the norites, so that

8 Bull. 269, U.S. G.S., p. 159. 1906.

% J. P. KIMBALL, Amer. Chemist, IV, 321. 1874. This paper contains four analyses of

the emery.

86 ANNALS NEW YORK ACADEMY OF SCIENCES

sometimes these varieties are found banded together in flow-like masses.

The diorites must have come next, since the granites were evidently the

last to form.

Dikes are abundant throughout the series and vary. in composition

from pegmatite to peridotite.

Many inclusions of schist, and some of limestone and gneiss, are found

in the district. These are most prominent in the large diorite area and

in the district near Salt Hill. At times, they have been absorbed by the

molten magma, giving rise to very abnormal rocks. Contact metamor-

phism is also apparent around them, and on the borders of the series in

places, developing in the schist an abundance of biotite, quartz, mag-

vetite and many aluminous minerals; and in the limestone wernerite,

diopside and other lime compounds.

Dynamic metamorphism is indicated by all of these rocks to a slight

though constant degree; but this metamorphism sometimes .assumes very

marked proportions around the foreign inclusions in the district and in

places from which the emery has been mined.

The economic geology is confined chiefly to granite quarrying and

emery mining. The emery has been generally considered as an example

of magmatic segregation, but more detailed work would seem to indicate

that it is due to the absorption of sedimentary xenoliths. This would

give rise to a magma supersaturated with respect to iron and alumina,

from which emery would separate out according to the laws formulated

by Morozewicz. This view of the origin of the emery is supported by

its frequent, somewhat remote association with visible inclusions of

schist ; by its association with exactly the same suite of minerals as those

developed on the borders of the district and in certain inclusions whose

relations are unmistakable, and by its frequent occurrence banded with

quartz, sometimes contorted and resembling exactly a black quartz schist.

PLATH III

PYROXENITE AND DIORITE

Fig. 1. Rounded form of olivine pyroxenite outcrop at Chase Corners, caused

by the weathering of the olivine. This rock is never found in roches

moutonnées, as are the other rocks of the series

Fig. 2. Stringers of Quartz Schist in Diorite, Buchanan

peear S08 Coen. rile, ache

ee ire vies tie Toucst tecke

B45 ce Spr fre wen? . Hes tag .

Mea is Stic,”

Pee <2 D- Bye <eet AArscay Tie The. “aera

eee Fe tg sort xe

Bf tery a MR epe9d woh gt, Sieh: Snr eee

Be Saeie sing, Sa ore nat eine en te nee

polka Naeioe year Saleh Eis Pein: ; ot a

— 23 at pas, = 4 FZ Tix 1%. Rei

vA ae 3 ah on oohtsh. on ae ”

Bed eed trons ue Sad paieredes ae ie ne

eee aH. ofi.sF ee 5 sSpsgundas 2

a eS Det) ee pape indestet. wees

Se WEES oy tee as; ry this. ciate :

_ SS ees Pree wi The Canieievinial f

ee eS € Tey: tas puesto

ih a

Psy Be) eats. “tee = = a b Sa Pes . ‘he whey) ee

<i Sotets Srp oor hd = ils Poavtegion: are viable” aru

Raa 8S gee ee See sivite’ “sce or

emer iieas . 45h hese aw Tecate: ais ofr inte i

,. <>

ine ages

ater) : ber SRK . ae eS

ke < M4

¥~

= a

~_ - ‘

. ~ = : -

‘ .

ey - =— .

> : i rs ¥

- . = ;

— ad 4 a

‘ .

- M4 73

» « *

é - “

pa '

P dy

es Py a

, ee

4 4 . To

3 —> -

¥ rs aA ol

wg are oF

} oh a eat 2:

AcAD. §

¥-

ANNALS

Fic.

PLATE IV

GNEISSOID STRUCTURE

Vig. 1. Original Gneissoid Structure, on road near Pleasantside

Wig. 2. Original Gneissoid Structure, one mile south of Spitzenberg Hill

PLATE V

MCCOY MINE AND SCHIST

Fig. 1. The latest cutting in the McCoy Mine, showing the emery banded with

sillimanite schist

Fig. 2. Quartz Hmery Schist, Salt Hill, showing the streaked appearance of

this low grade ore

: i

bye ‘Syatoetent

ANNALS N. Y. ACAD. SCI. VOLUME XXI, PLATE V

_ PLATE VI

: =

“GEOLOGICAL MAP OF THE CORTLANDT SERIES ses

PUBLICATIONS

OF THE

NEW YORK ACADEMY OF SCIENCES

(Lyczum or Natura History, 1817-1876)

The publications of the Academy consist of two series, viz. :

(1) The Annals (octavo series), established in 1823, contains the

scientific contributions and reports of researches, eee with the records

of meetings and similar matter.

A volume of the Annals coincides in general with the calendar year

and is sold at the uniform price of three dollars per volume. The articles

composing the volume are printed separately, each in its own cover, and

are distributed in bundles on an average of three per year. The price of

the separate articles depends upon their length and the number of illus-

trations, and may be learned upon application to the Librarian of the

Academy. ‘The author receives his separates as soon as his paper has

been printed, the date of issue appearing above the title of each paper.

(2) The Memoirs (quarto series), established in 1895, are issued at

irregular intervals. It is intended that each volume shall be devoted to

monographs relating to some particular department of Science. Volume

I is devoted to Astronomical Memoirs, Volume II to Zodlogical Memoirs,

etc. The price is one dollar per part as issued.

All. publications are sent free to Fellows and Active Members. The

Annals are sent to Honorary and Corresponding Members desiring them.

Subscriptions and inquiries concerning current and back numbers of

any of the publications of the Academy should be addressed to _

THE LIBRARIAN,

New York Academy of Sciences,

care of

American Museum of Natural History,

New York, N. Y.

fi ae eT

ANNALS OF THE NEW YORK ACADEMY OF SCIENCES

Vol. XXI, pp. 87-117, pll. VII-IX

Editor, EpmMunpD OTIs HovrEy

THE INFLUENCE OF HEREDITY AND OF

ENVIRONMENT IN DETERMINING THE

COAT COLORS IN MICE

T. H. Morgan

NEW YORK

PUBLISHED BY THE ACADEMY

5 Jury, 1911:

THE NEW YORK ACADEMY OF SCIENCES

(Lyceum or Natura History, 1817-1876)

OFFICERS, 1911

President—Franz Boas, Columbia University

Vice-Presidents—GrorcE F. Kunz, Frepertc A. Lucas,

R. S. WoopwortH, WILLIAM CAMPBELL

Recording Secretary—EDMUND OtT1s Hovey, American Museum

Corresponding Secretary—HeEnny EK. Crampton, American Museum

Treasurer—EMERSON McMItirn, 40 Wall Street

Inbrarian—RatpuH W. Towser, American Museum

SHCTION OF GEOLOGY AND MINERALOGY

Chairman—GuorceE F. Kunz, 401 Fifth Avenue

Secretary—CuHar es P. BerKey, Columbia. University

SHCTION OF BIOLOGY

Chairman—F reveric A. Lucas, Brooklyn Museum

Secretary—L. Hussaxor, American Museum

SECTION OF ASTRONOMY, PHYSICS AND CHEMISTRY

Chairman—WILLIAM CAMPBELL, Columbia University ~

Secretary—Epwarp J. THarouer, Teachers’ College

SECTION OF ANTHROPOLOGY AND PSYCHOLOGY

Chatrman—R. 8. WoopwortH, Columbia University

Secretary—FReEDERIC LyMaNn WELLS, Columbia University

The sessions of the Academy are held on Monday evenings at 8:15

o’clock from October to May, inclusive, at the American Museum of

Natural History, 77th Street and Central Park, West

Jeet

iis Fore LA.

[ANNALS N. Y. Acap. Sct., Vol. XXI, pp. 87-117, pll. VII-IX. 5 July, 1911.]

THE INFLUENCE OF HEREDITY AND OF ENVIRONMENT

IN DETERMINING THE COAT COLORS IN MICE

By T. H. Morean

CONTENTS

Page

Mini tO TUNE TIO MEL ee toler wiaetotehadacalel < havelaloie sok ous s'6.«,5,/02)6' a's oe een: feels 'e evareter neato te 88

Crosses between a wild sport of Mus musculus and domesticated varieties. 88

Deseription of the wild sport. :......... cece ccc ceccces wide Giatetee ares ' 8&9

Crosses between the wild sport and domesticated races with uniform

COO dace od AA OB eae a ator: BP REET M. Sat iocattaleto ed Sheiaiesd aie at's che Seheeienene 90

Crosses between the sport and yellow mice...................... 90

Crosses between the sport and gray mice................-...-20- 92

Crosses between the sport and black mice...............-.-2+--0- 92

Crosses between the sport and chocolate mice.................-4. 92

Crosses between the sport and albinos. ..............00-00+20s00 93

Crosses between F,, hybrid sports and yellows........-.....-... 93

@rassestpeiweenr +, By BEIGMSDOTES a:- 10 <1s << « 3) <teae «, re syn ssa eerie 93

New type of gray with a yellow belly. ............s0.cccscceec cr cceseree 93

Crossing extracted gray and black to test the hypothesis of alternate domi-

MPTICC Ge FECOSSIOM sete hci a ye aietge lee teyers: aie) cree lov'oree 6)'3 6, 3.6% ja 0. 0 0.6) she eee aan 94

Crosses between the spotted and the uniform coat...............--.0++00. 95

Influence of the spotted coat on the white belly of the sport.............. 98

Crosses between the black spotted waltzer and mice with chocolate coat... 100

PAGAN Cie WALLACE? arverets cieteta sielateisin eto scaleial Co occ! <) a: << sree © biavaieve, oe otever euatdyelah amare 102

Are black and chocolate different pigments or stages in the development of

BHCESATHE ISITE itetet aie a sloetak ey ctoldvets sitiaersisin le. so clsiels lege eleys 6 alate tian everatenstene 103

PAU TENE (CA EAV\ CRCOLOLS wets ich eke o areiis. ore tcicayraiie: Slslie: aie sys “008 fee, er eriere,oie ish dabetelobeleas 104

OUTED aoa are ycrtete crc CU beter eid ci aiela lore! cicvers ave laiele 0.6 /¢\ se 'wrerete aiieledo ls atataenet 4 104

Crosses between black and white spotted waltzers and yellows............ 105

ATI DUVE- COOLS VOCS POM belecys istsisia ss «c's. «(= /fsle,c eles) = dvale/a ale) gheld ahelquciotete, diarels 106

The influence of the environment on the color of Peromyscus leucopus am-

TOOUTNB B58 Gime Se ctery LOM Ot OTE COED O Cr SE Coan oer, F 106

SSEBEE AR CONGIUI SION caicha's ete tha cic ce 6) ois. 0's ais) ode erate ee erevetia, a ehelefete ehaterarelsvaieia'e he 108

The ticked or gray as a unit character. ...........cc-csees aie ld vmilaiela 108

ERNE AS SOG ELOMM TV IONESIS: (4°5.5° 5 a: is: S:0,070)))svalelal age aleietere ctagcla ee eunteve Cateye 111

Unit characters and factors in Mendelian inheritance................ 114

EAA STDS UPS Cp aT Pam boost oh aS yo SERRA a Shsisare. 5) o: Siaca’ Grats, dom coeee aie RC aleseeh chee alae cio ate Rue nla iy

LIBR,

NEW y

BOTAN.

GaRD:

88 ANNALS NEW YORK ACADEMY OF SCIENCES

INTRODUCTION

During three years I have carried out some experiments in heredity

with mice. My intention was to familiarize myself at first hand with the

process of Mendelian inheritance, and for this purpose few groups of

animals offer as many advantages as do the races of domesticated mice,

not only because of the ease and rapidity with which they breed in con-

finement, but also because the relation of the colors has been more fully

exploited here than in any other group. The varieties of color also offer

an extended field for study. In the course of the work, some crucial

experiments bearing on the theory of inheritance of coat color in mice

were made, and these are given in the following pages. The discovery

of a wild sport of the house mouse led to a study of its inheritance when

crossed with domesticated varieties, and the more interesting results of

these crosses are here given. Incidentally, I describe some experiments

bearing on the theory of the inheritance of acquired characters as tested

by artificial waltzers; a few cases of asymmetrical eye color are also de-

scribed, and their meaning considered. The occurrence of a second wild

sport is briefly mentioned. I have ventured to offer some suggestions

concerning the factors involved in heredity of coat color in mice. In 1908

I brought forward a tentative hypothesis to account for the segregation of

characters in Mendelian inheritance, and I pointed out at the time how

this hypothesis might be tested. This test I have now completed, and the

evidence that shows that it must be abandoned is here offered for the first

time.

CROSSES BETWEEN A WILD Sport oF Mus MUSCULUS AND DOMESTICATED

VARIETIES

DESCRIPTION OF THE WILD SPORT

- Although many “sports” of wild species have been recorded, there are

few cases describing their inheritance when bred either to the typical wild

forms from which they arise or to domesticated races of the same species,

when such exist. In the summer of 1907, I caught in a house at Woods

Hole, Mass., a house mouse, Mus musculus, that had a pure white belly,

chestnut sides and dark gray back. The mouse had come in from outside,

since the house was new and had been closed all winter. Later, I caught

two more such mice and, in the same closet, another typical house mouse.

In the neighborhood, I have caught a few other white-bellied mice. They

must be, therefore, not uncommon in the locality. In 1908, I had sent to

MORGAN, COAT COLORS IN MICE 89

me from Iowa some mice caught in the fields, which were supposed to be

Peromyscus because of the white belly. These were the same sports that I

had obtained in Woods Hole. I have also found in the collection of house

mice at the American Museum of Natural History in New York a few

skins with the characteristic white belly. The sport must be, therefore,

wide-spread. Since it appears to be found less often in the cities than in

the country, it may possibly be a variety that occurs most often in the

fields, or the conditions there may be more favorable for the appearance

of the sport; and since, as will be shown below, it is a dominant type, it

may ‘in time spread and supplant the ordinary gray-bellied house mouse.

Cuénot has recorded the appearance of this or of a similar sport in his

domesticated races, but whether it arose there as a sport or was already

present in some of the stock he used (in the yellow, for example) cannot

be determined from his data. The sport resembles in all respects the

house mouse except in color. The hair under the chin, neck and belly is

pure white, the white extending to the flanks. Each hair has a dark base,

t. e., it is “ticked” and differs in this respect from the white spots seen in

varieties of domesticated mice. This difference shows that the new sport

has not arisen from escaped spotted mice. A narrow yellow band extends

along the flanks between the fore and hind legs. The yellow band is

strongly marked off from the white below, but less so from the gray of

the sides above. The rest of the body is gray, but in most of my speci-

mens, the gray is somewhat darker than that of the ordinary gray house

mouse. Under the microscope, the hairs of the yellow flank are seen to

have a yellow tip and a black base. The gray back also contains some

hairs with a yellow tip and a black base, also hairs entirely black, while

other hairs have a brown tip and a yellow mid-band. The white hairs of

the belly have a white tip and a black base, while the ventral gray hairs

of the gray-bellied mice have black extending nearly to the tip. The tip is

yellow or nearly colorless. Even here, however, one meets with hairs that

are yellow all the way to the base.

There are no intergrades between the white-bellied and the gray-bellied

mice. There has never been any doubt concerning the nature of the

several hundreds of mice descended from the wild sport that I have ob-

tained. On the other hand, it should be stated that there is a wide range

of fluctuation of the color of the ventral surface in the house mouse.

Individuals are not uncommon that show a distinctly paler belly, but

these belong to the conimon type, and since these lighter forms do not

intergrade with the white-bellied sports or produce them, they rank with

the common type with gray belly.

Cages containing the original wild sports and their offspring have been

90 ANNALS NEW YORK ACADEMY OF SCIENCES

maintained for three years. They have produced but few young, behaving

in this respect like the ordinary house mouse when confined in a small

cage. The males have, however, crossed freely with domesticated races

and this happens also in the case of the house mouse. In most cases, my

sports were bred to gray and produced white-bellied and gray-bellied

young. In a few cases, the white-bellied forms when bred to each other

produced white-bellied and gray-bellied offspring. On the other hand,

two gray-bellied mice of this strain produce only gray-bellied young.

These facts show that the white belly is dominant to the gray belly.

CROSSES BETWEEN THE SPORT AND DOMESTICATED RACES WITH UNIFORM

COAT

The sports have been crossed with all of the ordinary varieties of

domesticated mouse. The results in the first generation are shown in the

following table, in a generalized statement.

Gray white-belicd by yellow = yellow, and gray with white belly.

“ gray with gray belly = gray with white belly.

black = gray with white belly.

chocolate = gray with white belly.

white = gray with white belly.

“ “ “ if 4

oe ce ce (<9

Like the wild gray mouse, the sport is dominant to all other colors

except yellow. The sport, moreover, transmits its entire coat pattern

when it dominates, and even when it becomes recessive in the yellow cross,

the coat pattern reappears in the next generation.

We may now proceed to examine in detail the crosses summarized in

the above table.

Crosses between the Sport and Yellow Mice

Six litters were obtained from male grays with white belly and yellow

females, consisting of fourteen yellows and eleven grays. Of the grays six

were recorded with white bellies, one with a gray belly, four not recorded

for ventral color. Assuming that half of the yellow gametes bear yellow

and half any other color, or colors, the results accord with theory. This

cross was of particular interest because by it I wished to test whether

yellow is allelomorphic to the other colors—gray, black and chocolate. If

the yellow-bearing gametes bear only yellow, then all the F, yellows from

a cross with gray should produce, when inbred, only yellows and grays as

three to one (or two to one). If, on the other hand, yellow stands alone

and is allelomorphic to its absence, then the offspring of the F, yellows

MORGAN, COAT COLORS IN MICE 91

might contain other colors than yellow and gray. Owing to the sterility

of the yellows, I have had poor success with this cross. Three litters

only of the F, yellow by yellow were obtained. They gave ten yellows and

two grays with white belly. In another case yellows were paired with

grays with gray belly (from Iowa). Two of the yellows that resulted

from the cross produced five yellows, five grays with gray belly and two

cinnamon agoutis. In all therefore the F, yellows have given fifteen

yellows and nine grays.

These results, while not so extensive as I should wish, indicate that the

yellow-bearing germ-cell is pure for yellow in the sense that yellow is

allelomorphic to gray. In other words, yellow has for its allelomorph in

the hybrid with gray the combination of yellow-black-*chocolate that

stands for gray. Whether chocolate and black are separate factors, or

whether black is a development through chocolate, and if so whether an

independent factor (not contained in the complex) brings about this

change need not be discussed now, but will be examined later. Whichever

of these views seems more satisfactory, the results of this experiment, as

far as they go, show that the yellow-bearing germ-cell is pure for that

color factor. The numbers are not large enough perhaps to put the

matter beyond question. As far as they go, however, they indicate that

the yellow gametes of the yellow mice carry only yellow, and that gray

(and perhaps chocolate too) is allelomorphic to yellow. I lack the evi-

dence to show the nature of the yellow pigment in mice, whether it is a

lipochrome, as commonly assumed, or belongs to the melanin series. I

must rest my case on the experimental evidence that seems to show that

yellow is the allelomorph of gray (1. e., of the colors that go to make up

the gray). Since yellows may carry other colors latent, gray as well as

chocolate or black, but only gray (black-yellow-chocolate) when gray has

been introduced as such, either the ticking factor is a separate factor, or

else some special combination (union) of black, chocolate and yellow

exists which when present will give the ticked or barred condition to the

hair. It may seem more probable that the barring factor is a separate

factor that behaves independently of the other factors and when present

produces its effect, except when dominated by the yellow factor. That

this factor may act with chocolate and black alone without yellow is indi-

cated below in the black-chocolate mice out of gray parents.

The F, white-bellied mice were crossed extensively with mice of other

colors, and they produced white-bellied grays and other colors according

to expectation. Gray bellies also appeared among the progeny when yel-

lows were used, which is explained on the ground that the yellow mice

1The presence of black in this combination will be considered later.

92 ANNALS NEW YORK ACADEMY OF SCIENCES

carried the combination for gray belly. That yellow mice may have intro-

duced into their composition the factor for white belly also is shown by

the F, yellow crosses; also by an experiment in which one of these F, yel-

lows bred to black produced some gray mice with white bellies.

I hoped to find out whether yellow mice bearing this factor for white

belly have themselves in consequence a white belly, but owing to the

great variability of yellows in this regard, I can make no positive state-

ments. Most of my yellow mice have a lighter belly. I have seen some

with a pure white belly superficially like those of the sport, but I have not

tested them. In some of them, the hair was yellow proximally and white

at the tips, suggesting possibly the presence of the ticking factor. I have

also had yellows whose ventral surface was as yellow as the rest of the

pelage. :

Two of the F, gray, gray belly, were crossed with ordinary yellows and

gave some gray, gray belly, descendants; none with white bellies.

Crosses between the Sport and Gray Mice

As stated, the wild sports that I obtained seemed to be heterozygous,

since they produced both white-bellied and gray-bellied offspring. Sports

V and VI, for instance, gave three gray, white-bellied, and one gray,

gray-bellied young. Sport V crossed to a wild gray gave two gray, white-

bellied and two gray, gray-bellied young. Sport I crossed to a pure race

of extracted gray dominants gave four gray, white-bellied and one gray,

gray-bellied young. One of the F, grays with gray belly, crossed to an

extracted dominant gray, gave six grays with gray belly.

I have produced races of gray with white belly that breed true to this

condition.

Crosses between the Sport and Black Mice

Sport III bred to a pure black female gave one gray with gray belly

and two grays with white belly. Sport I (or F, ?) to black gave two grays,

gray belly and one gray, white belly. Sport III, bred to spotted black

gave four grays with white belly and one gray with gray belly. These

results show again that the wild sports were heterozygous. When an F,

gray with gray belly was bred to black there were produced three grays

with gray belly and one black.

Crosses between the Sport and Chocolate Mice

Sport I crossed with a chocolate female gave five grays with white

bellies. Sport V bred to chocolate gave three grays with white belly and

two grays with gray belly.

MORGAN, COAT COLORS IN MICE 93

Crosses between the Sport and Albinos

Sport I bred to a white female gave three grays with white belly and five

grays with gray belly in one litter; and in a second litter, one gray with

white belly and one gray with gray belly. Sport III bred to a white

female gave four grays with white belly and one gray with gray belly.

In all, therefore, there were eight grays with white belly, and seven

grays with gray belly, which is approximate equality and confirms the

view that the sports used were heterozygous. None of the whites used

appear to have contained yellow.

Crosses between F,, Hybrid Sports and Yellows

One of the hybrid white-bellied mice (F,) crossed to yellow gave four

yellows, two grays with white belly, three grays with gray belly, one black

and two whites. One of the hybrid grays with gray belly crossed with

yellow gave two grays with gray belly, one golden agouti and two yellows.

Crosses between F,, Hybrid Sports

Two hybrid grays with white belly, bred together, gave three grays with

white belly. They may appear to have been pure, but the number of

young is too small to make this certain. One hybrid gray with gray

belly, bred to gray with white belly, gave three grays with gray belly and

one golden agouti with gray belly.

New Type oF GRAY WITH A YELLOW BELLY

As ordinary gray mice grow older, the hair on the ventral surface often

becomes yellowish. In some of the many offspring descended from

crosses between the wild sports and domesticated mice, individuals have

appeared again and again with the whole ventral surface distinctly yel-

lowish in color. Some of those with deeper color have been paired and a

race produced most members of which have a deep yellow belly replacing

the white belly in the sport. This color is not due to age, for while it is

true that it often becomes yellower as the mice grow older, I have found

young mice, three weeks old, with a distinct yellow belly.

Under the microscope, the yellow hairs of the belly are seen to have a

yellow outer end and a black inner part. The hair may be said to be

ticked with yellow and black.

The introduction of yellow into the white belly suggests the possibility

that the yellow is the result of crosses with ordinary yellows, but I can

94 ANNALS NEW YORK ACADEMY OF SCIENCES

find no records in my notes that show this to be the case. The color has

appeared sporadically and does not appear to come through the yellows

of the yellow crosses that have yielded relatively few mice compared with

other combinations. Moreover, if the result had been due to yellow, we

would expect it to appear only in the mice directly derived from yellow

crosses, but this is not apparent.

Crossinc EXTRACTED GRAY AND BLACK TO TEST THE HYPOTHESIS OF

ALTERNATE DOMINANCE AND RECESSION

In order to account for the two kinds of germ cells produced for each

pair of characters by Mendelian hybrids, I suggested in 1905 that if

instead of treating the question of segregation as the result of the sepa-

ration of material factors, we treat the process as due to alternate domi-

nance and recession of the paired characters in the gametes, we arrive at

the same end results, viz., the formation of three classes of individuals in

the second generation. This was an attempt to give a dynamic conception

of the process of segregation instead of the conventional idea of separation

of material particles as the basis of Mendelian inheritance. I advanced

this view primarily to escape Cuénot’s hypothesis for the inheritance of

yellow which called in the aid of selective fertilization. Cuénot assumed

that a yellow-bearing sperm never fertilized an egg bearing this same

factor for yellow, hence all yellow mice arise through the union of two

germ-cells only one of which carries the factor for yellow. Selective fer-

tilization seemed to me highly improbable in itself, and if true in general

for other characters it would lead to all sorts of irregularities in Men-

delian inheritance. It now appears that the particular ratio of three to

one, given by Cuénot? as the probable ratio for yellow in the second

generation, is in reality a ratio of two to one, as Castle and Little have

recently pointed out, so that the grounds for the original assumption by

Cuénot fall to the ground. I pointed out at the same time certain logical

consequences that followed, if my argument for alternate dominance and

recession were valid, and I set to work to put the hypothesis to the test

of fact.

The example that I used in my paper to illustrate the hypothesis of

alternate dominance and recession, namely, the inheritance of albinism

versus pigmented coat, was unfortunate, since it is generally conceded now

that albinism is not due to the absence of the color determiner (gray or

black, etc.), but to the absence of another factor, the color producer. It

2 Cuénot discussed also the possibility of the two to one ratio.

MORGAN, COAT COLORS IN MICE 95

is incontestible that this latter conception of albinism is better in accord

with the facts.

The hypothetical interpretation that I offered involved primarily the

idea, as stated above, that segregation may be a dynamic function of

division in the germ-cells in exactly the same sense that specification of

the cells to produce the organs and tissues of the embryo takes place dur-

ing embryonic development. In this process, the results of experimental

embryology had seemed to show that specification (segregation during

cleavage) is not due to a separation of particles of chromatin, but to the

dynamic action of the cells on each other at the time of or just before the

division stages. Such a conception of embryonic segregation is still held

by most embryologists, and I am still of the opinion, that if this is true

for the ontogeny, it is true for the segregation that takes place in the

formation of the gametes. My hypothesis was complicated, however, by

the further supposition that such a dynamic segregation in the gametes

carries with it the idea of impurity of the gametes in the sense that it

allows as a possibility that the extracted recessive may under certain con-

ditions give rise to the dominant, and conversely that the dominant may at

times produce the recessive type. One especial condition that I assumed

to call forth the latent character in the recessive form was the process of

hybridization. I pointed out how this interpretation could be tested and

its truth or falsity established. If, for example, we cross an extracted

dominant gray mouse (one that has had white in its ancestry) with a

pure black mouse, the offspring in the first generation should all be gray;

but if the presence of the black can call forth the latent white condition,

some white mice should appear in subsequent generations of the gray

hybrids.

The gray mouse used in my experiment was from a race of extracted

dominants that produced only grays. The black mouse was given to me

by B. B. Horton, and was from black stock that he had formed which gave

only blacks. In the first generation grays only were produced. These

inbred produced grays and blacks. The third and fourth generation of

some of these mice were bred in several combinations and continued to

produce only grays and blacks (Plate VIII, figs. 2,3). It is evident that

the hypothesis failed when tested and must therefore be abandoned.

CROSSES BETWEEN THE SPOTTED AND THE UNIFORM CoAT

The experiment was undertaken originally to examine the inheritance

of spotted coat versus uniform coat. Owing to the difficulties in obtain-

ing crosses in large numbers between waltzers and other races, this side

96 ANNALS NEW YORK ACADEMY OF SCIENCES

of the experiment has not yielded as extensive results as I had hoped for.

In conformity with the statement of Cuénot, I found that when the

spotted mice (waltzers) were crossed with races having a uniform coat all

the mice in the first generation were uniform. In the second generation

(F,) both uniform and spotted coats appeared, the former in excess.

Cuénot gave the usual Mendelian ratio 3 :1 for this generation. I have

found difficulty in deciding how to classify the spotted mice, for they

range from those spotted like the original waltzers to those with only a

few white hairs on the belly or even at the tip of the tail. It is apparent

that, if the original coat characteristic of the spotted waltzers be taken as

the recessive character, this particular coat or its equivalent in spottedness

reappears in much less than one fourth of the second generation. I chose

a race of waltzers obtained from Prof. .R. M. Yerkes that had a known

history. This race has not, it is true, a fixed spotted pattern, but within

certain limits the amount of white to black is fairly constant. In marked

contrast to this condition is the series of forms in the F, generation that

range, as stated, from the uniform coat to the original spotted condition.

It is perfectly evident that, as a result of crossing, the uniform coat has

encroached on the spotted coat, so that the latter now has a far greater

range in one direction than before. In a word, the spotted coat in the F,

generation more often approaches the uniform coat than does the original

race. It is clear therefore to my mind that the relation of spotted coat to

uniform is far more complicated than the Mendelian ratio requires and

that hybridizing introduces a new factor or modifies the old one. The

spotted coat may in fact be said to have been contaminated by the cross,

so that in most cases segregation, if the process can be said in fact to be

one of segregation, is less complete than before.

The problem of the spotted coat is one that has not been clearly brought

into line with other interpretations applied to Mendelian inheritance in

mice. For instance, if, as Cuénot has suggested, the albino condition is

due to the loss of a color producer (C), then this factor must be absent

from the white regions of the spotted skin, yet the factor must be present

in other parts of the body of the same spotted animal that produces color.

Hence both the color producer and the color determiner must have been

present in the fertilized egg. It does not seem to me to render the diffi-

culty any less by introducing a spotting factor, unless its mode of action

can be made to conform to the physiological action involved to explain

the presence or the absence of color in other races, for there can be little

doubt that the physiological process that produces yellow, gray, black or

chocolate spots is the same as that producing the same colors in the

uniform coat. On the other hand, it is evident that spotted mice are

MORGAN, COAT COLORS IN MICE Ov

different from mice with uniform coat on the one hand and from albinos

on the other, and that they carry something that causes spotting, however

variable, to reappear in the second generation. It is well known that

crossing spotted animals to albinos does not increase the amount of white

in the spotted coat. On the contrary, if the albino is derived from a

colored race with a uniform coat, the first generation may be uniform and

the second variable to the same extent as though a uniform coat had been

used. It is clear that the albino condition is not related as such to the

spotted condition. The paradox seems to me to find its solution in the as-

sumption that the spotted coat is not segregated from the uniform coat as

a “unit character” in the germ cell, but that it represents a process which

occurs in the early cleavage of the egg when the A factor is separated

in certain cells from the C factor. In other words, some of the cells lose

the C factor and the regions derived from these cells are therefore white.

The well-known variability of the coat pattern that has been so difficult

to fix as such can be explained in this way, owing to the irregularities in

the process that leads to the loss of C, or to subsequent shifting of the

cells. Three corollaries follow from such an assumption: First, that there

is no “spotting factor” as such that segregates in the germ cells of the

hybrid animal. All the germ cells contain the factor for color producing

and color determining—the loss comes later in the ontogeny. It is this

tendency to separate late that is inherited, and I have no objection to

calling such a tendency a spotting factor, provided that we do not confuse

its method of action with what occurs in ordinary segregation.

A second corollary is that the tendency to dissociate in the cleavage

stages may also be supposed to occur in the hybrid of the first generation,

but, since the other factor*® that stands for uniform coat (1. e., no disso-

ciation) is also present, the separation may not normally be evident, al-

though in rats, as I have pointed out, even in the first generation the

occurrence of dissociation is manifest in certain parts of the body.

A third corollary also follows. Since crossing spotted and uniform coats

leads to a production in the F, generation of a long series of intermediate

forms, we must infer that the tendency to dissociate has been modified by

crossing with the uniform coat. It has become modified in the sense that

it is often shifted to a later stage in the cleavage, hence the great varia-

bility observed in the second or F, generation. Cuénot and, later, Castle

have shown that the spotted condition may be gradually shifted in the

direction of selection. Cuénot showed by continued selection of

lighter colored mice (i. e., spotted mice with more white) that the char-

8 The doubleness of each factor makes such an assumption possible.

98 ANNALS NEW YORK ACADEMY OF SCIENCES

acter of the coat shifted in later generations in the direction of selection ;

and Castle has shown for rats similar shifting by selecting the darker

coat pattern. If my interpretation is correct selection here affects the

stage at which the “segregation” occurs in development. The cause of

this shifting must still obviously go back to some change in the germ cells,

but the change so brought about does not, from this point of view, cause

a change in the character of the coat carried by the germ cells, but a

change in the factor that determines when the segregation takes place in

the cleavage of the egg.

The uniform coat appears to bear the same relation to the spotted coat

that the uniformly colored hairs (yellow, black, chocolate or white) bear

to the gray or ticked hairs. From this point of view the “ticking factor”

has the same theoretical value as the “spotted factor.” It is, of course, a

fiction to assume that all the hairs of a gray mouse are alike and consist

of the three bands—black, yellow, chocolate—in the same relation. A

casual examination of the hair under the microscope will suffice to show

the wide range of variability that exists and the numerous departures

from this rule. The variability is almost as great as that shown by the

spotted coat. Nevertheless, in a general way, most hairs of the body may

be said to conform to this scheme.

I have met with this same problem in certain mutations in Drosophila.

In one case, a heterozygous fly had one white eye and one red one. Here

the separation must have occurred after the egg had been fertilized. The

tendency was not inherited. The point will be discussed below. On two

occasions, I have had a heterozygous fly that had one normal wing and

one small “proportionate” wing. These flies when bred to males of the

race with proportionate wings (the recessive) have produced both long-

and short-winged individuals, showing the heterozygous nature of the

original asymmetrical fly. I then crossed again the long and short flies

to bring about the same heterozygous combination, in the hope that the

“asymmetrical factor’ would come to light again, but of several hundred

offspring not one was asymmetrical.

INFLUENCE OF THE SPOTTED COAT ON THE WHITE BELLY OF THE SPORT

The wild sport was crossed on several occasions with spotted mice. The

hybrids were gray mice with white or gray bellies. The former when

inbred produced some spotted mice with white bellies. An examination

of these mice showed that whenever a white spot extended into the region

of the white belly, the black at the base of the hair that is present, as

stated, in the ordinary white belly is absent. In other words, the spot

MORGAN, COAT COLORS IN MICE 99

affects the “white” belly-hair in exactly the same way as it affects the hair

elsewhere—all color is lost.

As is known, the spotted condition affects all mice of whatsoever color,

whether yellow, gray, black or chocolate, in the same way. If a yellow-

spotted mouse is crossed to a mouse with uniform coat of gray, black or

chocolate the spotting disappears in the first generation and the uniform

coat is yellow in some of the offspring. If these yellows are inbred, the

spotting recurs, not only in some of the yellow offspring but in the other

colors as well. The spotted condition is not therefore associated with any

peculiar color, but affects all equally and can be transferred from one to

the other as stated above. If we interpreted the absence of color in the

spotted white areas as due to a factor that suppresses color in these parts,

it would follow that it affects yellow in the same sense in which it affects

black and chocolate, and the ticked condition of the gray also. On such

a view we might argue logically that yellow belongs to the same series as

the other colors, since it is suppressed by the same factor. Furthermore,

if, as is generally believed, the spotted condition is due to the loss of the

color producer, so that the color determiner can no longer act to make the

color, then we are equally led to the conclusion that the color producer

for yellow is the same as that for the other colors—a conclusion of some

general importance.

Returning to the change in the white hair of the belly of the sport,

when a white spot crosses the white belly, it is evident that the effect is

due to the loss of the color producer, so that the hair is white to its base.

This view is, of course, in line with the interpretation of the white belly

as a ticked region.

This leads to a consideration of the ticking or barring factor. Obvi-

ously it is the same sort of thing as the spotting factor, only it appears in

each hair instead of in different regions of the body. If we push our

hypothesis to its logical limits, we must conclude that when the cells that

form the hair are laid down, during the earlier divisions of the cell (or

cells) from which each hair arises, there occurs a series of losses in the

different cells, so that the terminal cell and its descendants have lost the

determiners for yellow and black, leaving chocolate, the next cell or cells

lose the determiners for chocolate and black leaving yellow, and the last

cell of the chain retains only the black determiner. Similarly in the

white belly, the outer cells of the follicle have lost the color determiner

either for yellow or black, or the color producer, while the basal cells re-

tain the determiner for black as well as the producer. Since yellow may

appear in the yellow-bellied race in the outer ends of the hair it appears

simpler to suppose that it is the color determiner that is lost.

100 ANNALS NEW YORK ACADEMY OF SCIENCES

CROSSES BETWEEN THE BLACK SPOTTED WALTZER AND MICE WITH

CHOCOLATE CoAT

In a litter of mice from a chocolate female and a male black and white

waltzer, there were present young mice that are described in my notes as

gray. At that time, I was perfectly familiar with the different coat colors

of mice, yet I did not hesitate to describe the color as gray, although

aware that gray could not, theoretically, arise in this way. As the mice

Fie. 1

FIG. 2

Crosses between Black-spotted Waltzer and Chocolate-colored Mice.

grew older, they became spotted with black (like the mice of Plate VII,

figs. 3, 4), and later entirely black. I thought therefore that I had prob-

ably made a mistake in my determination of the color of the young mice,

but such mice have later appeared not infrequently in the offspring of such

crosses (see Text-figures 1 and 2), and a study of their hair has shown that

there are some grounds at least for confusing their color with gray. The

MORGAN, COAT COLORS IN MICE 101

gray-looking black-chocolate mice often show a stratification of these

colors in many of the hairs. The outer end is chocolate and the base

black. The yellow band of the gray mouse alone is missing. In other

words, the ticking factor has reappeared and affects the distribution of

the only two colors present, namely, black and chocolate. The mice may

be said to be “grays without yellow.” On the other hand, these mice

often turn into black mice, often in spots, so that some of the curious pat-

terns shown in Plate VII, figs. 2, 3, 4, appear temporarily at least. These

patterns usually last several weeks or months and then generally change

to a uniform coat or to a different arrangement of the color area. Con-

versely these same or other mice often change from a uniform color to

a black-chocolate spotted animal. The condition of the animal—its

“Hhysiological state’—induced by change in the environment at the time

of moulting, is probably the cause of these differences. It seems probable

that the “ticking factor” is able to assert itself only when the conditions

are of a particular kind.

When the hair of these mice is studied under the microscope, a very

complicated series of relations becomes manifest. Many of the hairs are

ticked as described, but to various degrees; others may be uniformly

black, etc. Among the mice that have originated through crosses between

strong and dilute races, I have often met with color changes in the coat,

often local, sometimes general, that are related in some way to the pres-

ence of the two factors, that stand for dilute and strong color (Plate VII,

fig. 1). Gray mice may at times also show areas of lighter and darker

color, and these may come and go with each molt (Plate VIII, figs. 1, 2).

When examined under the microscope, the light hairs show a lessening

of the pigment granules in different regions, a lessening that is a charac-

teristic feature of the so-called dilute condition. Here again we can only

refer these effects to the conditions, often transitory, that affect the

heterozygous animals. This variability raises the question as to whether

the ticking factor is in reality present in the “waltzer-chocolate” hybrids.

May not the result be expressed in terms of strong and dilute? I am not

disinclined to such an interpretation, provided that black and chocolate

bear this relation to each other. This question will also be deferred for

treatment to the final conclusions, but one further question may be briefly

mentioned here. The supposed origin in China of the waltzing mice, the

peculiar shape of the head, their size and the proportions of the body

strongly suggest that they belong to a different race (or species?) from

that from which our common domesticated mice have arisen. If this can

be proven, it may be that they have not lost entirely the ticking factor,

but that black is epistatic. By the introduction of chocolate, the possi-

102 ANNALS NEW YORK ACADEMY OF SCIENCES

bility of the ticking expressing itself may be realized, so that the “gray

without yellow” may in reality be a ticked hair, as first suggested.

Although I have made no measurements or weighing, it appears that

the hybrids between the waltzer and common mice are larger than the

waltzer and possibly not as large as our common domesticated races. In

the second generation, I have obtained both large individuals and small

mice, as well as intermediates, the first as large as our common form, the

latter as small as the waltzer. It appears that the size character also

segregates, at least to some extent.

Yerkes has pointed out that the older waltzers are deaf. In order to

study the inheritance of deafness, I tested many of my domesticated races

of mice and found that in them also there is a large percentage of animals

that seem to be entirely deaf. It becomes evident, therefore, that the

problem would be complex, unless races were first produced like the

waltzers that are uniformly deaf, but as I had no such races and did not

succeed in crossing wild mice and waltzers, it does not appear worth while

to give here the result of my observations on my hybrid mice. I hope to

continue this experiment later.

ARTIFICIAL WALTZERS

The curious effect of acetyl-atoxyl on mice has been described by Ehrlich.

It causes them to run about in circles in much the same way as do the

true Japanese waltzing mice. This coincidence in behavior led me to test

whether mice that have been made to waltz by artificial means would

transmit their acquired character.

The injections were made for me by Dr. B. T. Terry, of the Rockefeller

Institute for Medical Research, and I am greatly indebted to him for the

time and skill expended in producing these waltzers. In some cases, one

injection sufficed to brmg on in a few days the circus movements; in

others, two or three doses were necessary, and some mice failed entirely to

become waltzers. In one case, the mouse was pregnant and the injection

caused abortion of two of her young, but the rest of the litter developed

to full term. None of these waltzed later.

I have made a number of pairings between the artificial waltzers and

normal mice. None of the offspring waltzed, but since the waltzing of

waltzing mice is recessive, this might be expected to occur even if the

habit is inherited. Therefore I inbred these F, mice, but obtained only

normal mice. In two or three cases, I have had young from a pair that

were both waltzers. None of these or their progeny waltzed. I conclude,

therefore, that so far as the evidence goes, the acquired character is not

transmitted. Although the results were negative, nevertheless the evi-

MORGAN, COAT COLORS IN MICE 103

dence seems peculiarly significant because of the occurrence of a race of

domesticated mice that breeds true to this character. The test is even

more significant, when we recall that the waltzing habit when induced

artificially seems to be permanent. I have kept artificial waltzers for

more than a year and at the end of that time they waltzed as well as at

first. It seems probable, therefore, that the drug has caused a permanent

change in the animal, and since it is known to produce degeneration of

certain nerve fibres in man, it seems not unlikely that the effects in mice

are of the same nature. Yet, despite the permanency of the effect, the

acquired character is not transmitted through the germ cells.

Dr. Terry has also produced for me a number of waltzing rats by

means of larger doses of atoxyl. Unfortunately, none of these have bred

when both parents were waltzers.

ARE BLACK AND CHOCOLATE DIFFERENT PIGMENTS oR STAGES IN THE

DEVELOPMENT OF THE SAME PIGMENT?

There are two methods of treating the colors black and chocolate in

mice. They are generally considered to be different unit characters and

the product of different factors, but I have found some evidence that indi-

cates that these two pigments may be intimately related, and I suggest

that they may represent different developmental stages of the same

process. From this standpoint, black is a further development of the

same chemical process that produces chocolate, and since pure black and

pure chocolate races exist, there must be some special condition or factor

that leads to the further development of chocolate to make a black mouse.

Let us represent this higher stage by the symbol M. A black mouse is,

therefore, Ch Ch MM, and its gametes are Ch M and Ch M. A chocolate

mouse is Ch Ch. When black is mated to chocolate, the hybrid is Ch Ch

M, which gives in the next generation three blacks to one chocolate.

Similarly when chocolate Ch Ch is crossed to gray (Y Ch) M, the first

generation will be gray, and the second will contain gray. (including cin-

namon agouti), black and chocolate in the proportion 12 : 3:1.

The dilute colors, blue (or dilute black) and silver fawn (or dilute

chocolate) require special consideration. Dilute black is not chocolate,

as Cuénot supposed, but the dilute effect is due to the sparseness of the

black granules. Similarly dilute chocolate owes its peculiarity to the

fewness of the chocolate granules, rather than to their weakening. The

experiments of Miss F. M. Durham show that the factor producing

sparseness in black produces the same effect in chocolate. This factor, S,

is not to be confused with the factor M that transforms chocolate to a

higher stage—i. e., into black. When dilute black is crossed with choco-

104 ANNALS NEW YORK ACADEMY OF SCIENCES

late, the offspring are black. This result is obvious on the explanation

here offered; for the factor that produces the abundant development of

the granules is supplied by the chocolate mouse, while the color of the

granules is determined by the dilute black parent. Such an interpreta-

tion is, as far as I can see, completely in accord with the facts, and I

venture to think that it offers some advantages over the schemata so far

offered, in that the peculiar relation between black and chocolate in

heterozygous black becomes more intelligible. In these mice, most of the

pigment is carried on to its higher stages, .viz., black, some remaining,

however, at the lower stage; and the extent to which the process is car-

ried out depends on the physiological condition of the mouse at the time

when the hair is laid down.

ASYMMETRICAL EYE Coors

In the winter of 1907, I procured from Dr. H. L. Wood, of Groton,

Conn., a dilute colored mouse that had one pink eye and one black eye.

The same condition reappeared again in one of my own mice, probably

related to Dr. Wood’s. It is known that angora cats often have one blue

eye and one green one, and that coach dogs and other races, and man also,

sometimes have eyes of different colors, but I do not know of any other

records where one eye is black and the other pink (albino). The mouse

belonged to a strain of colored mice with pink eyes that Dr. Wood had

imported from England. Some of the mice were spotted with white,

which may enter also into the problem. I bred this mouse to other black-

eyed mice of the same stock, but obtained no asymmetrical offspring,

either in the first or in subsequent progeny. The condition appears, there-

fore, not to be inherited in the ordinary sense, but to be rather what I

have called an ontogenetic process of segregation which takes place for

this combination only rarely. Nevertheless, the phenomenon rests on the

same basis as that for the spotted condition in general, but in this instance

it is not fixed and appears only sporadically. I have had one other mouse,

as stated above, with one black eye and one pink eye from related stock,

but in none of its progeny did the condition reappear. In another com-

bination, I have found one mouse with one pink eye and one ruby eye.

The mouse itself was chocolate.

DILUTE GRAYS

In the course of the large number of crosses that I have made between

different races of domestic mice, there have been produced some dilute

forms of gray. I mention these facts here not only to put on record the

MORGAN, COAT COLORS IN MICE 105

production of such forms, but also to show that, being aware of their

occurrence, I have been less liable to confuse the unusual colors that I

have described—such, for instance, as the black-chocolate that simulates

gray.

The lighter races were made by crossing dilute chocolates with cinna-

mon agoutis and then extracting, or by crossing light black with cinna-

mon agoutis and extracting. In the former case, some of the F, mice

were extremely pale, and the chocolate band appeared to be dilute. In

the latter case, one might expect to obtain some mice in which both the

black band and the chocolate band were dilute. In fact, I have had sev-

eral kinds or at least colors of dilute mice with ticked hair, but I hesitate

to class them under these two groups without further anid more careful

examination, because these dilute forms appear to be variable and the

color of the coat may change with every moult. One of these changing

coats is shown in Plate VIII, fig. 2, where the combination gives a re-

markable contrast.

CROSSES BETWEEN BLACK AND WHITE SPOTTED WALTZERS AND YELLOWS

In order to test in another way whether the yellow gametes of yellow

mice carry the factor for other colors than yellow, I crossed a yellow

with a black and white waltzer of known pedigree that carried only black.

Two yellow mice from this cross were inbred and produced a litter con-

taining yellows, blacks and chocolates. I concluded that the yellow

gametes may carry in addition to yellow the factor for chocolate. Cuénot

objected to this conclusion, on the ground that my original yellow con-

tained the diluting factor, which, according to his view, changes black to

chocolate. Cuénot’s objection will not hold for two reasons. First, be-

cause chocolate is not due to the diluting factor for black, since Miss

Durham has shown that the dilute form of black is not chocolate, but

dilute black or blue. Second, because if the diluting factor were present,

the later generations of these mice should make evident its prsence,

which was not the case. Nevertheless, I now believe that Cuénot’s objec-

tion holds good in principle... If, as I have attempted to show, black is a

higher development of chocolate, the black waltzer must have contained

in duplex this factor that changes chocolate to black, and his spermatozoa

must have contained this factor only in the simplex condition. The yel-

low mouse in question must not have contained this factor, hence the

yellow offspring (F,) contained it only in simplex form. It will be

present or absent, therefore, in their gametes. Those containing it, and

the factor also for chocolate, will give black; those without it will give

106 ANNALS NEW YORK ACADEMY OF SCIENCES

chocolates. The result can therefore be brought into line with those of

the gray-yellow crosses, but only if we treat chocolate and black in the

way here suggested.

A MAUVE-COLORED WILD SPorT

Through the kindness of L. C. Bragg, I obtained the skin of a mouse

caught in Colorado far removed from any possible source of contamina-

tion with domesticated races. Several of these mice were caught in the

same locality and were kept alive by Mr. Bragg for some time. All at-

tempts to procure others have so far failed. The color of the sport is so

peculiar that I believe it worth while to record it here, for we know very

little concerning the origin of the domesticated races of colored mice or

of sports of the house mouse. The general color of the mouse is yellowish

brown or mauve. A microscopic examination of the hair of the back or

sides shows the presence of black pigment granules in the proximal half

of each hair, and yellow pigment in the outer part, except at the tip,

which is colorless. In addition, some of the large hairs contain black

pigment nearly to their tips. Comparison with the hair of the ordinary

house mouse shows that the base of the hair of the mauve mouse is much

less black than is the ordinary hair, and that the yellow brown pigment is

identical with that in the yellow band of the gray mouse. The mauve

mouse appears, therefore, to owe its color to the loss of chocolate, and

perhaps to a lightening of the black color at the same time; but the latter

effect may be due to the absence of chocolate in the black portion, where

it appears to be mixed with black in the ordinary mouse.

THE INFLUENCE OF THE ENVIRONMENT ON THE COLOR OF PEROMYSCUS

LEUCOPUS AMMODYTES

The intergradations between many of the species of Peromyscus has

made their identification and classification difficult. The occurrence of

intergrades in regions of overlap may be explained either as the result of

interbreeding of the more typical forms found elsewhere or as the effects

of the environment. Both of these questions might be studied, I thought,

by breeding the mice in confinement. In the winter of 1907, and during

1908-9, I kept several species of Peromyscus in the laboratory—species

from the Eastern States, from Florida, Texas, California, Colorado,

Indiana, Ohio and elsewhere. The results have been meager because in

most cases the mice failed to breed, both when like was bred to like and

when different species were kept together. In the summer of 1908 and

again in 1909, I collected the local races of Peromyscus found on Mono-

MORGAN, COAT COLORS IN MICE 107

moy Island and on Marthas Vineyard. Experience had taught better

methods of keeping the mice, so that I have been able to rear these types

in confinement and have obtained crosses between them. The results of

these crosses may be described at another time; here I wish to record cer-

tain facts connected with a change of color in confinement. The change

in color has been more marked in the Monomoy variety than in any other

form. This species lives on the sand hills of the small island of Monomoy

off the southern coast of Massachusetts. In its natural habitat this variety

is in general much lighter than the mainland form. The white hairs of

the belly are white to the base, while in the mainland type, P. novebora-

censis, the inner end of each hair is dark. There is a good deal of varia-

tion in the degree of lightness of the upper surface of different individ-

uals, which Bangs suggests is the result of mixing caused by occasional

migrations, when the island has been for a time connected with the main-

land. Some of the specimens that I have collected and have obtained

alive from Monomoy are as dark above as the mainland form, while

others are quite pale. Similarly, in the laboratory, great variation in

color in these Monomoy mice exists, and the mice have shown themselves

more susceptible to change than the other species and varieties kept

under similar conditions. The most extreme change is shown in Plate

IX, fig. 2. For comparison, one of the lighter individuals found in

nature is also figured (Plate IX, fig. 1). The modified form has lost

even the light gray color of the island type, except for a patch on each

side of the body and another on the head, and has assumed a pale bluish-

pink color that is well shown without exaggeration in the figures. When

seen with other mice this mouse looks almost white. A microscopic ex-

amination of the light hairs shows that the outer two thirds of the hair

from the middle of the back is entirely devoid of pigment, but near the

base of the hair some black pigment is present. The color of the mouse

is due, therefore, to the darker base showing through the clear outer ends

of the hairs. The effect, as stated, is to give a faint bluish-pink color to

the mouse, when seen from above or from the sides.

The stock in which this mouse appeared came from Monomoy in the

summer of 1909. It was kept in a large cage in the attic of the labora-

tory throughout the winter and in an upper room in the laboratory for a

part of the summer of 1910. The mouse was found in the condition

figured at the end of the summer and has remained in the same condition

for six months. The cage in which it was kept had produced young mice

in the spring which had not been removed, and, as I was absent during

the summer, I can not state whether this particular individual was one

of the original mice from Monomoy or one of their young that had be-

108 ANNALS NEW YORK ACADEMY OF SCIENCES

come full grown. Several other mice in the same cage showed a similar

change, but none so great as this one. The others showed patches of

lighter color and these may disappear again after several months or new

patches may appear. Similar effects, though not so great, have also been

seen in the mainland forms kept under the same conditions and also in

some of the mice from Marthas Vineyard. Other species also kept in

confinement have shown similar lightening of the coat color. Osgood‘

states for Peromyscus leucopus noveboracensis that the coat is being con-

tinually moulted or “changed,” although it appears to be entirely renewed

only once a year. The changes in color that come and go are undoubtedly

connected with the partial moulting of the individual, and it seems rea-

sonable to suppose that they are expressions of the physiological condition

of the animal at the time of moult.

The important question as to which factor or factors in the environment

is responsible for the changes in color here recorded can not, I regret, be

given. Light seems to be excluded, since the animals are crepuscular, and

the room in which they were kept was lighted by a window, so that in

this regard the conditions were not very different from those in nature.

Food or temperature or humidity might be supposed to be the factors

that are probably involved. The food was varied, but contained less

green food than the animals generally obtain. The temperature was

much higher on the average than that to which the animals are normally

exposed. The air of the attic was extremely dry, yet it is to be remem-—

bered that this lighter-colored race has arisen on an island where the

moisture is extremely high, so that if there is any relation between the

color and the environment, we should expect dryness to produce the re-

verse effect. Until control experiments can be carried out, it will be best

not to assign the change to any one of these possible agents.

How far these changes may be carried in later generations remains

also to be shown. It is, however, sufficiently obvious that, if changes as

great as these may take place, the results of crossing different species in

confinement must be carefully controlled by studies of the influences of

the confinement itself.

GENERAL CONCLUSIONS

THE TICKED OR GRAY HAIR AS A “UNIT CHARACTER”

Under the microscope, the hair of the common gray mouse is found to

contain three pigments, black at the base, yellow in the middle and brown

(or chocolate) at the tip. This banded condition is obviously like that

4Revision of the mice of the American genus Peromyscus. North American Fauna,

No. 28. 1909.

MORGAN, COAT COLORS IN MICE 109

seen in many animals in which the tip of the hair or feather or scale is

brightly colored and the base dark or black. Obviously the peculiar fea-

ture of such structures is the stratification of the color bands. The cells

first formed in the follicle must produce one color, then another, and later

a third. The possibility of producing all three must be present at the

beginning, and a “morphogenetic factor’ of some sort determines the

activity that leads in turn to the formation of one after the other of these

colors. It must be at once granted that we know no more of the nature

of such a factor than we do of the differentiation that appears in the

development of any of the characters of the embryo. To ascribe it to a

“ticking factor” is no more than to describe the phenomenon in general

terms, since we know nothing of the method of action of such a factor.

The fact that the ticking factor may be carried by the yellow mice that

also carry black or chocolate (or both) and yet not come to expression is

one of the peculiar features of the yellow inheritance.

That the development of the ticked coat is a specific element in heredity

can not be doubted, and, admitting the purely symbolic nature of its

representation, I see no objection to calling it in the ordinary sense a unit

character.° It is interesting to see how it has been treated in Mendelian

literature.

Cuénot uses the letter G as a symbol for gray and treats it as allelo-

morphic to the other colors—yellow, black or chocolate. Its presence pro-

duces a gray mouse, which in his formulas is never treated on the presence

and absence theory, but always as allelomorphic to yellow or black or

chocolate. Yet for its expression the presence of all these three pigments

must occur. It is in fact a bundle of these three pigments with the added

condition of their stratification. There is obviously an obscurity here,

unless the group of colors that stand for ticking is inherited as a unit in

allelomorphic relation to either of the three colors considered by them-

selves.

Bateson’s symbolism for gray in mice is GBChC, to which yellow

must, I believe, be added, giving YG@BChC. Miss Durham omits the C

unit and makes Ch its equivalent, or at least suggests such a possibility.

In these formule, G is called the gray determiner which is equivalent to

ticking. Bateson adds, “The important question, what the effect of the

gray determiner, for example, actually is, remains undecided.” Bateson

represents the black mouse as C g BCh and chocolate as CgbCh. In

these formulz small g and b stand for the absence of the gray and the

black determiners.

°*This statement leaves open the nature of the factor that produces such a “unit

character.”

110 ANNALS NEW YORK ACADEMY OF SCIENCES

Two questions arise in this connection. The determiner for chocolate

is not represented in the presence or absence scheme, since it has never

been found absent—+. e., no white mice have been recorded from which

all determiners are absent. Such a mouse would be represented by C gb

ch. Why, it may be asked, is chocolate excluded from the presence or

absence relation? If the chocolate determiner is equivalent to the color

producer C, then it too might be absent from some albinos (or from all) ;

yet it can be shown that albinos that do not carry other factors carry

chocolate, which is contradictory to the assumption. It may be claimed,

however, that the chocolate factor has never dropped out in the past;

hence its presence in all mice and hence the impossibility of treating it

on the presence or absence scheme. Such an interpretation is logical at

least.

In treating the heredity of color in guinea-pigs and rabbits, Castle

considers gray as the result of the presence of a ticking factor. In these

rodents the gray or agouti may be produced by crossing a black and a

red (yellow) individual. Only certain blacks and reds give the result.

Castle concludes that some blacks and some reds contain the ticking fac-

tor. So long as only one color is present, either black or red, the action of

the ticking factor is rendered impossible, since it takes at least two factors

to produce the stratified layers. Hence a black animal with the ticking

factor is still black, and so is a red only red, but when combined, the

presence of red and black and the ticking factor gives again the agouti;

but it is not clear why a black animal with the ticking factor should not

have the outer ends of the hair (where the red lies in the ticked hair)

colorless and the base black. This condition actually exists in the white

belly of my wild sports. On the other hand, yellow mice do carry the

ticking factor, and yet the hair is uniformly or at least continuously

colored yellow, although here too some of my yellow mice with white

bellies have a white tip and a yellow base to each hair. If the red guinea-

pig is equivalent to the yellow mouse, it may be that the ticking factor

can be carried by this type alone and not by the blacks. This view would

make the two groups conform, but Castle’s evidence seems to show that a

yellow that gives agouti with one black may not do so with another. If

his cases are sufficiently numerous to show that this is not an accidental

result owing to his reds being heterozygous for G, then the suggestion

here offered is unavailing.

In mice, I have found clear evidence that black and chocolate may be

stratified to produce a resemblance to gray. Such cases clearly show that

in these animals the factor for ticking may be present and effective in

the absence of yellow. To prove that this is due to a true ticking com-

MORGAN, COAT COLORS IN MICE 1h

parable to that of the wild gray mouse, such mice should be crossed to a

yellow strain that contains no ticking factor and gives rise to some gray

mice. This test I have been unable as yet to apply.

Castle has also suggested that melanistic forms of gray animals arise

through the loss of the ticking factor. In consequence, the colors are not

laid down seriatim and the darkest color becoming epistatic gives the

black color to the hair. The implication is that the yellow and the choco-

late are still present but obscured by black. By a further loss of black

the chocolate stains the hair and a chocolate animal results. Two diffi-

culties with reference to this view suggest themselves. It is assumed that

the loss of the ticking factor will lead to the uniform spread of the other

colors, but this is not self-evident or proven. I have tested a few black

mice and found no yellow present.

Secondly, yellow itself may completely color the hair in a yellow

mouse that carries black or chocolate, and these latter pigments may even

‘sparsely develop. It might seem, therefore, that it is the yellow that

must disappear to produce a black mouse, and this, in fact, seems to be

realized in the black-chocolate ticked mouse that I have described. If

the ticking factor were then lost, pure black might result. It is possible

that black mice may have originated in this way; but even were this true,

it still remains not clear why the black should then spread into the choco-

late region of the hair. Moreover, the sudden appearance of black animals

in nature without an intermediate black-chocolate condition indicates that

the transfer takes place at one step. These considerations seem to me to

show that our assumptions are still too crude to offer a reasonable inter-

pretation of the facts in hand.

THE ASSOCIATION HYPOTHESIS

If we suppose that black, chocolate and yellow and the ticking factor

are carried by a common carrier in the gray mice, and that black is borne

by a different (7. ¢., not homologous) carrier in the black mouse, and

chocolate by a third carrier in the chocolate mouse, we can not give a

consistent hypothesis for the inheritance of these colors. For example,

when gray is crossed with black, the presence of the triune body gives

gray in the first generation. In the gametes of the gray hybrid, the gray

carrier finds no mate on the hypotheses. Its allelomorph is therefore its

absence. Similarly the black carrier has no allelomorph. The gametes

will then be GB, GO, BO,00. There would result twelve grays, three

blacks, one white; but white mice do not result, as I have shown, from

this combination. Moreover, they would contain no color determiner ;

and we know of no such mice.

ARO ANNALS NEW YORK ACADEMY OF SCIENCES

It seems more reasonable, therefore, to assume that, if the three colors

are combined in a single carrier, it may have lost successively one after

another of its components as the black and chocolates arise. If we

assume that by the dropping out of the yellow and the ticking factor a

black mouse results, then gray would remain allelomorphic to black,

since the material body carries one or the other. This is so far consistent

with the results; but on the same hypothesis, chocolate is due to loss of

black in the black mouse. If, then, gray is crossed to chocolate, not only.

should grays result in the F, generation, but only grays and chocolates in

the second generation ; yet Miss Durham has shown that grays, blacks and

chocolates appear, which is inconsistent with the hypothesis. If, how-

ever, black and chocolate are stages in the development of the same sub-

stance, and the stage reached depends on another factor not present in

the common carrier, but in another one, then the situation is clearer.

Chocolate must be supposed to arise from black by the loss of this factor

M, and since the factor is present in gray, then when gray is crossed to

chocolate, the determiner for black is present as simplex, 7. e., once only.

Its allelomorph is its absence. The formula for gray* would be (Y, Ch)

M, and that for chocolate Ch. The formula for the hybrid will be, there-

fore, (Y Ch) MCh. The allelomorphs are a a, and the gam-

etes (Y Ch) M, (Y Ch) O, Ch M, Ch O. Two such mice crossed will

give twelve grays (including three cinnamon agoutis), three blacks and

one chocolate, as the following table shows :

Y Ch M Y Ch M Y Ch M Y Ch M

Y Ch M YiChi@ Ch M Ch O

gray gray * gray gray

Y ChO Y Ch O Y ChoO Yih O

Y ChM Y ChoO Ch M Ch O

gray cin.-agouti gray cin.-agouti

Ch M Ch M Ch M Ch M

Y ChM Y Cho Ch M Ch O

gray gray black black

Ch O Ch O Ch O Ch O

Y ChM Y ChoO Ch M Ch O

gray cin.-agouti black chocolate

6 Omitting the ticking factor from (Y Ch).

MORGAN, COAT COLORS IN MICE 113

In the same way, the cross between yellow and gray would be repre-

sented as follows: If the yellow gamete is represented by Y and the

gray gamete by (Y Ch) M, the resulting yellow mice would have the

formula Y, (Y Ch), M; and its gametes would be Y, (Y Ch), M, and

no Mor O. The allelomorphs are then oe oh, -

ing gametes would be Y M, Y O, (YCh) M (Y Ch) O. The following

table gives the composition of the resulting hybrids. There result twelve

yellows to four grays. Now if the pure yellow combinations do not de-

velop, as Castle suggests, viz: YO YO, YO YM (twice), YM YM,

there will be left eight yellows and four grays, or 2: 1.

; and the result-

eM YM Y Ch M Y Ch O

YM WO WORE YM

yellow yellow yellow yellow

YO VEO ¥ O VEO

ye A wo Y ChM Y ChoO

yellow yellow yellow yellow

Y Ch M Y Ch M Y Ch M Y Ch M

YM Ae @ Y ChM ¥ Chg

yellow yellow gray gray

Y Cho Y ChoO Y¥ ChO ¥-Chr@

Yo Mi YO Y Ch M Y ChoO

yellow yellow gray cin,-agouti

The “presence and absence” theory has been extensively used by modern

writers on Mendelian inheritance. Bateson states that it gives consistent

results in all cases so far studied, and he seems inclined, with some reser-

vations, to adopt it as a method applicable to Mendelian treatment in

general. Its usefulness and even necessity is apparent for many cases ;

but it does not seem to me, therefore, that it is advisable to extend it to all

cases. If, as I have suggested, certain mutations result by or through the

loss from certain material bodies that ordinarily carry a group of such

bodies (such as gray, for example), the allelomorphs of the original body

will be the new body, lacking the particular factor in question, and not

the absence of the original body as a whole. The hypothesis of presence

and absence appears originally to have been invented to explain the gray

chocolate cross in mice. I have tried to show how the facts may be ex-

plained in a somewhat different way, so that the original combination for

gray is allelomorphic with the factor for chocolate, while presence and

114 ANNALS NEW YORK ACADEMY OF SCIENCES

absence applies to another factor that concerns the change from chocolate

to black. A combination of the two hypotheses seems to me to be more

useful than the adoption of either alone. Moreover we can see, on the

view of the allelomorphic relation of the factors that I have used here,

why in some cases one and in other cases the other way of expressing the

facts should be used.

UNIT CHARACTERS AND FACTORS IN MENDELIAN INHERITANCE

I understand by the term unit character any particular structure or

function that may appear in heredity independent of other characters.

Such unit characters may in themselves be extremely complex and include

the possibility of further splitting up. By factor I understand somé

special condition in the germ plasm whose presence is necessary for the

development of a particular unit character which in its absence fails to

develop. There is unquestionably a tendency in much of the Mendelian

writing to identify the unit character with one only of the factors whose

presence is necessary for its development. This view seems to be as un-

necessary in the present state of our knowledge as it is unproven by the

evidence at hand. It is this point of view that underlies the entire Weis-

mannian conception of the process of heredity. The acceptance of the

theory of factors does not seem to me to lead necessarily to such a con-

clusion ; for, while a factor may be essential to the development of a par-

ticular part, it may be only one condition, several or many conditions

combining to produce the effect, the absence of any one of them leading to

the same result. Thus, while it may be that a particular chromosome or

a small part of a chromosome is essential for the formation of a unit

character, it may be that all the other chromosomes and all or many of

the elements that constitute the cell also take part in the final elaboration

of the special organ in question. This may represent the extreme case,

and there may be all degrees in which different parts of the cell combine

to produce any given organ. The essential point is that the acceptance

of the factor hypothesis in heredity by no means leads to the conception

of each character being located in a special unit or biophore or pangene in

the egg, for, as I have said, many parts may be equally essential for the

production of every part.

Concerning the mechanism of segregation in the germ-cell we know

very little at present. Many facts seem to point to the conclusion that

the simplest solution of segregation is that a sorting-out of material par-

ticles takes place during those processes that come under the general head

of maturation in the egg or spermatozoon. This sorting process involves

MORGAN, COAT COLORS IN MICE 15

the units or factors, and most of those who have written on the subject

give the impression that this sorting involves the actual materials out of

which each character is built up. I do not think that we are justified in

this interpretation of the factor hypothesis, for all the facts can be equally

well explained, if the material particles sorted in maturation represent

only parts of the material basis, whose entire activity is essential for the

formation of parts of the organism. The difference in point of view is,

I believe, fundamental, even if the practical outcome is the same, for our

entire conception of the mechanism of heredity and development is

involved.

On another occasion’ I have discussed the view as to how far the mech-

anism of segregation in the germ cells is comparable to the processes that

take place in the later development of the egg itself, when different re-

gions develop into different parts at the time of the specification of the

organs and tissues. I pointed out that there is a good deal of analogy to

support the view that the two processes are the same. The Weismannian

conception practically identifies the two processes, for the biophores are

supposed to be sorted out in the maturation process in the same way that

they are sorted out as the development of the embryo proceeds.

On the one hand, if the facts lead us to interpret the process of segre-

gation as due directly to a separation during the process of maturation

of material particles (factors) essential for development of particular

organs, on the other hand the facts of experimental embryology and of

regeneration have seemed to experimental embryologists to speak equally

emphatically in favor of the view that the localization factors are essen-

tially different from those assumed for Mendelian segregation.

An examination of the facts of sex-limited inheritance has seemed to

me to show with some probability that the simplest conception of the

mechanism of segregation is that material particles carried in the chro-

mosomes are separated at these divisions and lead to the Mendelian classes

of gametes. I do not deny that the facts may be conceived as due to

other kinds of processes that lead to the formation of two classes of

gametes for each pair of allelomorphs, but when we take into account the

evident random segregation of the factors any other hypothesis than that

based on particulate separation seems to me far more improbable than the

one here suggested. If this is admitted, the next logical step would seem

to require a similar process in development of the embryo. I need only

mention by way of example the occasional appearance in heterozygous

individuals of the dominant and the recessive characters in different parts

7 American Breeders’ Association, V, 1909.

116 ANNALS NEW YORK ACADEMY OF SCIENCES

of the body, as when one eye of a mouse is black and its mate is pink, or

one wing of Drosophila is long and the other short.

It may therefore be worth while, perhaps, to attempt to recast our

conception of development in the light of these facts. Obviously it may

not be necessary to attempt to explain all embryonic phenomena as due

to segregation. There are several facts of experimental embryology that

seem to indicate that there are other relations between the cells that are

important factors in determining their fate, but the essential facts of

specification and differentiation may nevertheless be explained by the par-

ticulate theory. No fact is more evident than that in most cases a region

once determined has lost its power to produce other parts, and this fact

bears a strong resemblance at least to the segregation process in the germ

cells that produce recessive characters.

This argument may seem to lead back to Weismann’s theory of develop-

ment, but it differs from his view in two essential respects: Firstly, it

admits that development involves not only a process of separation of

particles, but that other factors also are involved, especially those that

regulate the symmetry of the body and its parts. Secondly, it does not

look upon the particles of chromatin as each representing a different part

of the body—of its characters, in short—but for the development of

each part certain parts of the complex of particles form the necessary

foundation and for other parts other complexes. The germ material from

this standpoint loses in the course of development now one, now another,

of its materials, and the absence of these materials is responsible for what

each region can produce. Weismann’s view postulates the presence of

particular particles that produce each part. On his view, each part has a

representative in the original germ plasma, while on the view here sug-

gested, most of the entire material is necessary for the development of

each part, but the loss of one or another particle of the chromatic com-

plex turns the developmental processes into different channels. I am

aware that speculation concerning the processes of development and

Mendelian segregation may seem premature at present, but I can not but

think that it may be worth while to attempt to bring under the same

point of view the process of gametic segregation and the processes that

take place in the development of the embryo.

MORGAN, COAT COLORS IN MICE * 117

LITERATURE

ALLEN, G. M. Heredity of coat color in mice. Proc. Amer. Acad. Arts and

Science, XI. 1904.

Bateson, W. Mendel’s principles of heredity. Cambridge. 1910.

CasTLe, W. EB. Yellow mice and gametic purity. Science, XXIV. 1906.

CASTLE and ALLEN, G. M. The heredity of albinism. Proc. Amer. Acad. Arts

and Science, XXXVIII. 1908.

CastTLe, W. E., and C. C. Litrtre. On a modified Mendelian ratio among yel-

low mice. Science, XXXII. 1910.

Cutnot, L. L’hérédité de la pigmentation chez les souris. Arch. de Zool. Exp.

et Gén. 1902, 1903, 1905, 1907.

Sur quelques anomalies apparentes des proportions Mendeliens. Arch.

de Zool. Exp. et Gen. IX. 1908.

Recherches sur l’hybridation. Fourth Inter. Zool. Congress. 1907.

DaRBISHIRE, A. D. Note on the results of crossing Japanese waltzing mice with

European albino races. Biometrika, II. 1902.

Davenport, C. B. Color inheritance in mice. Science, XIX. 1904.

DurHaAm, F. M. A preliminary account of the inheritance of coat-colour in

mice. Report Evolution Committee, IV. 1908.

Note on Melanins. Jour. Physiol., XXXV. 1907.

Hagerpoorn, A. L. Inheritance of yellow color in rodents. University of Cal-

ifornia Publications in Physiology. 1909.

Mendelian inheritance of sex. Archiy. Entw. mech. der Organismen,

XXVIII. 1909.

Morean, T. H. The assumed purity of the germ cells in Mendelian results.

Science, XXII. 1905.

Are the germ cells of Mendelian hybrids pure? Biol. Centralb., XX VI.

1906.

Some experiments in heredity in mice. Science, XXVII. 1908.

Breeding experiments with rats. American Naturalist, XLIII. 1909.

Recent experiments on the inheritance of coat colors in mice. Ameri-

can Naturalist, XLIII. 1909.

What are factors in Mendelian explanations? American Breeders’ As-

sociation, V. 1909.

Chromosomes and heredity. American Naturalist, XLIV. 1910.

Witson, E. B. Mendelian inheritance and the purity of the gametes. Science,

XXIII. 1906.

{

td i rH nie i

Fw bs at i!

Bie ent ao if K

Pree eoin

PLATE VII

CHANGES IN COAT COLORS OF MICE

Fig.1. No. SW. A heterozygous mouse showing dilute and chocolate areas.

Fig. 2. No. 32144. A heterozygous black-chocolate mouse showing areas of

black, chocolate and lighter chocolate (mixed).

Vig. 3. No. PC. A heterozygous black-chocolate mouse with “ticked” hair.

Fig.4. No. D'. A black-chocolate heterozygous mouse.

SF rR

ao! G0 agore> vacy MLADEN

* > ei : : ;

ANNALS N. Y. ACAD. SCIENCES VOLUME XXI, PLATE VII

G. Rrecrout, Del.

PLATE VIII

CHANGES IN COAT COLORS OF MICE

Vig.1. No. NMA. A hybrid out of a black-gray (extracted) cross.

Fig. 2. No. AAbb. A heterozygous white-bellied gray showing light gray and

some gray areas.

Fig.3. No. NMA. A heterozygous mouse showing black and gray areas.

lec HIN

*\*

(iv a

ANNALS N. Y. ACAD. SCIENCES VOLUME XXI, PLATE VIII

G. Riccroui, Del.

, 9d Te

sere

“IM 30 aaoi6d T4097

- s an

yads

apis OF Beilin

S = , 2 = +

bot Sina t¥ouY edt 101 Ds

ehdt

ai Selon

“goat ehilt

euls gobhrode

Be Seles

oie teouin

i ;

ANNALS N. Y. ACAD. SCIENCES VOLUME XXI, PLATE IX

G. Riccrout, Del.

PUBLICATIONS

OF THE

NEW YORK ACADEMY OF SCIENCES

(Lyceum oF NatoraL History, 1817-1876)

The publications of the Academy consist of two series, viz. :

(1) The Annals (octavo series), established in 1823, contains the

scientific contributions and reports of researches, together with the records

of meetings and similar matter.

A volume of the Annals coincides in general with the calendar year

and is sold at the uniform price of three dollars per volume. The articles

composing the volume are printed separately, each in its own cover, and

are distributed in bundles on an average of three per year. The price of —