it 


Ra cae 


sits Mir ie 


Noaie 
fester tk 
Lat hes 
wanes 
Hi f° fi 
ela at Md 


i nah 
uithe gi) 


ipies 
Naas 
Heh tides, 


+1} 
hate! 
art 


Bi) eat 
opt hss is 3 
ied tah 


eu 
Reet 


ee 


sf ‘f 
a tah 
Lele *) Ay bets 


hae 


A 


tea! 
ali aes } 
+t 
(ie 


Pe pata it 
famitey 


et 
i 


peseh 
iy 4 HH sie 


rae 


Hs ite 
aie ERATE H tes 4 
satie 
tat 
een 
Shaped 


oleh 
Tage! ribs ah ? bien 
ita ; ee 
: pict 


4h bath i 
te ieihas 
iran fick 


fas) 


NitiNeg 
abe Mf ; i 
iy 


at 
brary it 


as 


coareets 


—~ 


ath 
+ 
a 


a 


NH 
Hy i 


P 
4 
se 

oF 


Hah seiseatectn 
Pi acca 


pease 


+ 
haadtbaartts 
itt 


of 
. 
wit) 


Ses 


a 


“vt 
Ana the 
! 


as 


i 
; * 
\ A 
ie 
a 


aT Teak 


THE UNIVERSITY OF CHICAGO PRESS 
CHICAGO, ILLINOIS 


THE CAMBRIDGE UNIVERSITY PRESS 
LONDON 


THE MARUZEN-KABUSHIKI-KAISHA 
TOKYO, OSAKA, KYOTO, FUKUOKA, SENDAI 


THE MISSION BOOK COMPANY 
SHANGHAI 


NH 


THE 


JOURNAL OF GEOLOGY 


A seta Quarterly Magazine of Geology and 


« 


Related Sciences 


EDITED BY 
THOMAS C. CHAMBERLIN AND *ROLLIN D. SALISBURY 
With the Active Collaboration of 


STUART WELLER, ALBERT JOHANNSEN, 
Invertebrate Paleontology Petrology 

EDSON S. BASTIN ROLLIN T. CHAMBERLIN, 
Economic Geology Dynamic Geology 


ASSOCIATE EDITORS 
SIR ARCHIBALD GEIKIE, Great Britain *JOHN C. BRANNER, Leland Stanford Junior Uni- 


CHARLES BARROIS, France versity 

ALBRECHT PENCK, Germany, RICHARD A. F. PENROSE, Jr., Philadelphia, Pa. 
HANS REUSCH, Norway WILLIAM H. HOBBS, University of Michigan 
GERARD DEGEER, Sweden FRANK D. ADAMS, McGill University 


SIR T.W. EDGEWORTH DAVID, Australia CHARLES, K. LEITH, University of Wisconsin 
BAILEY WILLIS, Leland Stanford Junior WALLACE W. ATWOOD, Clark University 


University WILLIAM H. EMMONS, University of Minne- 
CHARLES D. WALCOTT, Smithsonian sota 

Institution ARTHUR L. DAY, Carnegie Institution 

*Deceased. a 


VOLUME XXX 


JANUARY-DECEMBER, 1922 


ties ae 


i mul [en ba 
(peel sama 


THE UNIVERSITY OF CHICAGO PRESS 
CHICAGO, ILLINOIS 


Published 
February, March, May, June, August, September, 
November, December, 1922 


Composed and Printed By 
The University of Chicago Press 
Chicago, Illinois, U.S.A. 


CONTENTS OF VOLUME XXX 


NUMBER I 


AN OUTLINE OF THE GEOLOGY OF NEW ZEALAND. W.N. Benson. 

PENNSYLVANIAN STRATIGRAPHY OF NORTH CENTRAL TEXAS. Raymond 
C. Moore and Frederick B. Plummer 

PLEISTOCENE MoLLuscA FROM NORTHWESTERN AND Conte emeen. 
Frank Collins Baker ’ : ; 

THe Muppy MovuntAINn Ouanaemaen IN Cou AEIEES NEAL. 
Chester R. Longwell. Sas 

GROWTH-STAGES OF THE BLASTOID, Ono phacins Siclliformas ERwAG 
Bather. ; 

POSSIBLE SILURIAN eee IN Sou aren Teer Conners 
Francis P. Shepard 

REVIEWS 


NUMBER II 


THE GREAT FAULT TROUGHS OF THE ANTILLES. Stephen Taber 

THE CHARACTER OF THE STRATIFICATION OF THE SEDIMENTS IN THE 
RECENT DELTA OF FRASER RIVER, BRITISH COLUMBIA, CANADA. 
W. A. Johnston 

THE STRUCTURAL Rane AnAORY OF THE EuRGeET Rasen AND THE REISE 
MOvuNTAINS OF CANADA. Francis Parker Shepard : 

ASPECTS OF ONTOGENY IN THE STUDY OF AMMONITE EVOLUTION. 
A. E. Trueman 5 : 5 i 

A New PHYTOSAUR FROM THE TNS OF Avsieaonan.  Mienaries G. Mehl 

ADAPTING A SHORT-BELLOWS, Roti-FILM KODAK FOR DETAIL WORK 
IN THE FIELD. Chester K. Wentworth 

SEGREGATION GRANITES. Alfred C. Lane 

‘ON THE PRESENTATION OF IGNEOUS ROCKS IN eC one eae 
Albert Johannsen : : 

PETROLOGICAL ABSTRACTS AND Rimes 

REVIEWS 


NUMBER III 


THE REACTION PRINCIPLE IN PETROGENESIS. N. L. Bowen ; 

AN OUTLINE OF THE PHyYSIOGRAPHIC HistToRY OF NORTHEASTERN 
Ontario. W. H. Collins ; 

A New OccCURRENCE OF CRYSTOBALITE IN Ciera Aueen F. 
Rogers. 


V 


PAGE 


177 


199 


211 


VI CONTENTS TO VOLUME XXX 


FAULT FEATURES OF SALTON BASIN, CALIFORNIA. John S. Brown 

MARINE UPPER CRETACEOUS AND A NEW ECHINOCORYS FROM THE 
ALTAPLANICIE OF BoLiviA. Edward W. Berry . 

FORMER COURSES OF THE ANDROSCOGGIN RIVER. Irving B. Grochy, 

NOTES ON THE SAND DuNES OF NORTHWESTERN INDIANA. George B. 
Cressey : : : 

PETROLOGICAL A pamRacts AND REVIEWS 

REVIEWS 


NUMBER IV 


On Contact PHENOMENA BETWEEN GNEISS AND LIMESTONE IN 
WESTERN MAssAcHuseEtts. Pentti Eskola 

A CRITICISM OF THE ‘“‘FAUNAL RELATIONSHIPS OF THE Mecucer 
Group” BY Bruce L. CrarK. Roy E. Dickerson 

Tuer AGE OF THE DOMES AND ANTICLINES IN THE LOST SOLDIER- Fouee 
District, Wyominc. A. E. Fath 


On THE OCCURRENCE OF AN APUS IN THE PERMIAN OF Oxranonem 
Rudolf Ruedemann . 


PETROLOGICAL ABSTRACTS AND eevee 
REVIEWS 


NUMBER V 


Post-GLACIAL LAKES IN THE MACKENZIE RIVER BASIN, NORTHWEST 
TERRITORIES, CANADA. A. E. CAMERON é : ? 
DINOSAUR TRACKS IN HamILtTon County, Texas. W. E. Wrather 


PROBLEMS IN STRATIGRAPHY ALONG THE Rocky MOUNTAIN TRENCH. 
Francis Parker Shepard . ; : ; 

A SCALE OF GRADE AND CLASS TERMS FOR Cragin Sec, Chester 
K. Wentworth : ; : ’ ; 

THE PRE-CAMBRIAN OF WESTERN en E. M. Burwash 

PETROLOGICAL ABSTRACTS AND REVIEWS 

REVIEWS 


NUMBER VI 


Tue Hor WATER SUPPLY OF THE Hor SPRINGS, ARKANSAS. Kirk Bryan 

A DENOVIAN OUTLIER NEAR THE CREST OF THE OZARK UPLIFT. a 
Bridge and B. E. Charles , : 

THE EARLY PRE-CAMBRIAN FORMATIONS OF NOTE Osuene AND 
NorTHERN Manirospa. E. L. Bruce j 

THe TIME oF GLACIAL LoEsSS ACCUMULATION IN Tia Rm AnOR TO THE 
CLIMATIC IMPLICATIONS OF THE GREAT Lorss Derposirs: Dip 
Tuey CuteFrLy ACCUMULATE DURING GLACIAL RETREAT? Stephen 
Sargent Visher . 


PAGE 
217 


227 
232 


248 


252 
257 


450 


459 


472 


CONTENTS TO VOLUME XXX 


MEMORIAL EDITORIAL—ROLLIN D. SALISBURY. T.C. C. 
PETROLOGICAL ABSTRACTS AND REVIEWS 
REVIEWS 


SUPPLEMENT TO NUMBER VI 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS Macmas. N. L. Bowen 
Introduction : : 
Heat Effects of Solution . 
The Question of Superheat a cael eet Saree 
Equilibrium Effects between ‘STraltetone” and Liquids in Investi- 

gated Systems . 

Reaction Series . 
Effects of Magma on Thnlineione af lencoue Onn 
Effects of Magma on Inclusions of Sedimentary Origin 
Effects of Basaltic Magma on Aluminous Sediments 
The Action of Basic Magmas on Siliceous Sediments S 
Effects of Granitic Magma on Inclusions of Sedimentary Origin 
Deductions to be Compared with Observed Results . 
Summary 


NUMBER VII 


THE NEPHELITE SYENITE AND NEPHELITE PORPHYRY OF BEEMERVILLE, 
New Jersey. M. Aurousseau and Henry S. Washington 

TINTRAFORMATIONAL CORRUGATED Rocks. William J. Miller 

THE PHYSICAL CHEMISTRY OF THE CRYSTALLIZATION AND MAGMATIC 
DIFFERENTIATION OF IGNEouS Rocks. V. J. H. L. Vogt 

EDITORIAL. E. 5S. B. 

_ PETROLOGICAL ABSTRACTS AND Remar 

REVIEWS 


NUMBER VIII 


THE PHYSICAL CHEMISTRY OF THE CRYSTALLIZATION AND MAGMATIC 
DIFFERENTIATION OF IGNEOUS Rocks. VI. J. H. L. Vogt 

THE PLEISTOCENE HiIsTORY OF THE LOWER WISCONSIN RIvER. Paul 
MacClintock a : : : 2 

ORIGIN OF THE TRIASSIC TOMES ¢ OF Conn OTe. Wilbur G. Foye . 

In SUPPORT OF GARDNER’S THEORY OF THE ORIGIN OF CERTAIN CONCRE- 
TIons. Leroy Patton = Hee mesa ans aha ke di 

Mup CRACKS ON STEEPLY INCLINED Sian cra. Weert R. MacCarthy 

PETROLOGICAL ABSTRACTS AND REVIEWS 

REVIEWS 

INDEX . 


ce oe EDITED Byes : 
“THOMA C. CHAMBERLIN AND ROLLIN D. SALISBURY 


- With the Active Collaboration of 


ate ce ree ee ALBERT JOHANNSEN, Petrology 
Geo ROLLIN T. CHAMBERLIN, Dynamic Geology 


eve 


oe EDITORS 


ss RICHARD A. F. PENROSE, Jr., Philadelphia, Pa. — 
_.- WILLIAM H. HOBBS, University of Michigan 

pa ess FRANK D. ADAMS, McGill University 

_—sS CHARLES K, LEITH, University of Wisconsin - 

ari 44 - WALLACE W. ATWOOD, Clark University - ee 
? _ WILLIAM H. EMMONS, University of Minnesota ~ 
ARTHUR L. DAY, Carnegie Institution 


FROM NORTHWEST ERN AND CENTRAL ILLINOIS 
FRANK COLLINS BAKER 


CHESTER = ‘LoncweELt 
BLASTOID, OROPHOCRIN US STELLIFORMIS E. A. BATHER 


ERSFLY. OF CHICAGO PI 
CHICAGO, ILLINOIS, U.S. co 


Poa ayo WoW Benson 


RayMonp C. MoorE anp FREDERICK B. PLUMMER 


JOHN €. BRANNER, Leland Stanford Junior University 


18 


43 


January-February 1922 hes Beak a nan) 


EDITED By ras ea 


THOMAS C. CHAMBERLIN*AND ROLLIN D. SALISBURY = 

With the Active Collaboration of ~ 5 . € 

ean WELLER ALBERT JOHANNSEN Pa Save r. 

‘ Invertebrate Pe eOEy. ' Sic: Petrology A tak cot 

EDSON S. BASTIN : ; ROLLIN T. CHAMBERLAIN * oa 
Economic Geology : Dynamic Geology ¥tye 


” 
~. = 


The Journal of Geology is published by the-University of Chicago at the University of Chicago Press, 
5750 Ellis Avenue, Chicago, Illinois, semi-quarterly, on or about the following dates: February 1,March1s5, 
May 1, June 15, August 1, September 15, November 1, December 15. ] The subscription price is $4. 00 2 
per year; the price of single copies is 65 cents. Orders for service of less than a half-year will be charged — 
at the single-copy rate. ] Postage is prepaid by the publishers on all orders from the United States, 
Mexico, Cuba, Porto Rico, Panama Canal Zone, Republic of Panama, Dominican Republic, El Salvader, — 
Bolivia, Colombia, Honduras, Nicaragua, Peru, Hawaiian Islands, Philippine Islands, Guam, Samoan — 
Islands, Shanghai. ] Postage is charged extra for Canada, and for all other countries in the Postal Union 
as follows: 30 cents on annual subscriptions (total $4.30); on single copies 4 cénts (total 69 cents). — 
{ Patrons are requested to make all remittances payable to The Unueety of Chicago Press in postal “ 
or express money orders or bank drafts. 

The following are authorized to quote the prices snetentent 

For the British Empire: The Cambridge University Press, Fetter Lane, London, E. CiA: “Yearly see 
subscriptions, including postage, 29s. 4d. each; single copies, including postage, 4s. 9d. each. ; Be 

For Japan and Korea: The Maruzen-Kabushiki-Kaisha, 11 to 16 Nihonbashi Tori Sanchome, 
Tokyo, Japan. Yearly subscriptions, including postage, Yen 9.00 each; single copies, including — 
postage, Yen 1.50 each, we 

For China: The Mission Book Company, 13 North Szechuen Road, Shanghai. Yearly sub- 4 
scriptions $4.00; single copies 65 cents, or their equivalents in Chinese money. Postage extra, if. 
mailed direct outside of Shanghai, on yearly subscriptions 30 cents, on single copies 4 cents. 

Claims for missing numbers should be made within the month following the regular month of publi- 
cation. The publishers expect to supply missing numbers free nab when losses have been sustained in 
transit, and when the reserve stock will permit. — 

Business correspondence should be addressed to The University of Chicago Press, Ghteaeat Ill. 

Communications for the editors and manuscripts should be addressed to the Editors of THE Jour- 
NAL OF GEOLOGY, the University of Chicago, Chicago, Il. : 

Contributors are requested to write scientific and proper names with particular care and i in citations 
to follow the form shown in the pages of the Journal. 

Because of the high cost of printing, this Journal has found it necessary to adopt the following rules: 

{ 1. If articles exceeding 25 pages in length are accepted for publication in the Journal, the cost of com- 
position beyond 25 pages shall be paid by the author or by the institution which he represents, at the current 
rates of the University Press. {] 2. The cost of illustrative material, in excess of $1.00 per page for the 
article, beyond the cost of type which would have occupied the same space, shall be borne by the author. 
{| 3. The cost of corrections in proof in excess of 10 per cent of the cost of first composition, will be 
charged to the author. {| 4. While the cost of printing remains as now, the Journal cannot supply re- 
prints without charge, but reprints will be furnished authors at cost, if ordered in advance of publication. 


Entered as second-class matter, March 20, 1893, at the Post office at Chicago, Ill., under the Act of March 3, 1870. 
Acceptance for mailing at special rate of postage provided for in Section IT03, Act of October 3, 1917, authorized on ~ 
July 15, 1918. 


PRINTED IN THE U.S.A. 


Among articles to appear in early 


numbers of the Journal of Geology 


are the following: 


The Relation of the Physical Properties of Natural Glasses to T heir 
Chemical Composition. By WILt1AM O. GEORGE. 


The Character of the Stratification of the Sediments in the Recent 
Delta of Fraser River, British Columbia, Canada. By W. A. 
JOHNSTON. 


Notes on a Recent Collection of Fossil P Plants from the Dalles Group 
of Oregon. By RatpH W. CHANEY. 


The Great Fault Troughs of the Antilles. By STEPHEN TABER. 


_A New Occurrence of Crystobalite in California. By Austin F. 
ROGERS. 


Fault Features of Salton Basin, California. By JOHN 5S. BRown. 


A New Phytosaur from the Trias of Arizona. By MAUvrRIcE G. 
MEHL. 


The Geology of Vancouver and Vicinity 
By EDWARD M. J. BURWASH 


14 half-tones, 10 line drawings, and 2 colored maps 


A thorough and authoritative survey of the region, made 
|| more interesting by the remarkable photographs, which 

include a profile of the Grouse Mountain spur, the Lions 
from Mount Brunswick, a view of Red Mountain from Black 
Tusk Mountain, Mount Garibaldi from the south, a glacial 
section from a cutting on Keith Road, and others equally inter- 
esting. One colored map shows the topography of Vancouver 
range and the other is a geological map of Vancouver and 
vicinity. 

112 pages, paper covers; $1.50, postpaid $1.60 


THE ee we ae PRESS 
CHICAGO : ILLINOIS 


THE ORIGIN OF THE EARTH 


By THOMAS C. CHAMBERLIN 


Head of the Department of Geology, The University of Chicago 
One of “The University of Chicago Science Series” 
xii++272 pages, 8vo, cloth; $1.75 net, postpaid $1.90 


FROM THE PREFACE 


“Tn telling the story of this search for the mode by which the earth came 
into being, we have let the incidents that led the inquiry on from one stage to 
another fall in with the steps of the inquiry itself. It is in keeping with the 
purposes of this series of booklets that the motives which set researches going 
should have their place with the quests that arose from them 
story of the birth of the earth will come only after a time, when the vestiges of 
creation have been more keenly discerned and more faithfully rendered than is 
possible now.” 


THE UNIVERSITY CE CHICAGO PRESS 
CHICAGO : : ILLINOIS 


The Geographic Society of Chicago 


BULLETINS 


No. 1. The Geography of Chicago and Its Environs. Illustrated. 


By Rortin D. SALispury and WILLIAM C. ALDEN. 6a pages, 
8vo, paper, postpaid, 53 cents; met 50 cents. 


. The Plant Societies of Chicago and Vicinity. Illustrated. 


By Henry C. CowLes. 76 pages, large 8vo, paper, postpaid, 
60 cents; met 50 cents. 


. Lantern Slide Illustrations for the Teaching of Meteorology. 


Edited by HENRY J. Cox and J. PAuL GoopE. 130 pages, 
8vo, paper, postpaid, 85 cents; net 75 cents. 


. The Weather and Climate of Chicago. By HErEnry J. Cox 


and JoHN H. ARMINGTON. xxvi+376 pages, 8vo, cloth, 
postpaid, $3.25; met $3.00. 


. Animal Communities in Temperate America. A Study in 


Animal Ecology. By Victor E. SHELFORD. xiv-+362 pages, 
8vo, cloth, postpaid, $3.20; met $3.00. 


. Starved Rock State Park and Its Environs. By Cart O. 


SAUER, GILBERT H. Capy, and Henry C. Cowles. x+148 
pages, royal 8vo, cloth, illustrated, postpaid, $2.15; net $2.00. 


. The Geography of the Ozark Highland of Missouri. By CARL 


O. SAUER. xvili+246 pages, 8vo, cloth, illustrated, postpaid, 
$3.15; net $3.00. 


EXCURSION BULLETINS 


. Excursion through the Rivers and Harbors of Chicago. Pre- 


pared by R. E. Blount and C. S. JEWELL. 20 pages, 8vo, 
paper, postpaid, 12 cents; met to cents. 


. Excursion on the Rock River of Illinois between Rockford and 


Dixon. Prepared by R. E. BLount. 4 pages, royal 8vo, 
paper, postpaid, 11 cents; met to cents. 


. Stony Island. By ZontA BABER. 16 pages, 8vo, paper, post- 


paid, 11 cents; met to cents. 


THE UNIVERSITY OF CHICAGO PRESS 
CHICAGO” - - ILLINOIS 


Outlines of Geologic History with 
Especial Reference to North America 


Symposium Organized by 
BAILEY WILLIS 
Compilation Edited by 
ROLLIN D. SALISBURY 


(5 tiertare w and all readers of geologic 


literature will welcome the publication, 
in book form, of an important series of 
essays and discussions on the subject of geologic 
correlation under the title, Outlines of Geologic 
History with Especial Reference to North America. 
They embody the present state of knowledge 
and opinion concerning many of the funda- 
mental problems of North American geology, 
and form an admirable supplement to earlier 
treatises and manuals. 
The value of the book is greatly enhanced 
by the fifteen paleogeographic maps by Bailey 
Willis which accompany the papers. 


316 pages, Svo, cloth; price $1.50, postpaid $1.05 


The University a Chicago Press 


Chicago ou - Illinois 


Starved Rock State Park 


and Its Environs 


By CARL O. SAUER 
Department of Geology, University of Michigan 


GILBERT H. CADY 
Illinois State Geological Survey 


HENRY C. COWLES 
Professor of Ecology, University of Chicago 


This thorough and interesting geographical 
study of the best-known feature of the Illi- 
nois Valley has been prepared for the benefit 
of those who are interested in knowing more 


about the region than can be gathered by 
personal observation in a few hours or a few 
days spent in the park. The physical geog- 
raphy of the park and its surroundings has 
a number of features, such as the beautiful 
little canyons, which are unusual in this part 
of the country. Part I is given up to the 
geography of the park, its surface features 
and their origin, and the exploration, settle- 
ment, and development of the region; Part 
II, to the geology of the park; and Part III, 
to its botany. 
130 pages, cloth, $2.00, postpaid $2.15 


THE UNIVERSITY OF CHICAGO PRESS 


CHICAGO - - - ILLINOIS 


Source Book for the Economic Geography 
of North America 


By CHARLES C. COLBY 


Assistant Professor of Geography, the University of Chicago 


The fundamental idea of the book is to make available to the busy teacher the 
material on this subject which is scattered widely through literature. The course which 
has been given for a number of years at the University of Chicago has resulted in the 


collection and organization of the material most suitable for the purpose. 


The book contains short magazine articles and materials from Canadian, Mexican, 


and United States government publications. 


The selection of the material has been 


governed by the concept that there is a direct relation between the environment and 
the economic activities of the region. The author aims to survey the different sections 
of the country from the viewpoint of industrial opportunity. 


$v0, cloth; 419 pages, $4.00, postpaid $4.15 


THE ELE oe CEU Ne PRESS 


CHICAGO - 


- ILLINOIS 


VOLUME XXX NUMBER 1 


PELE 


FOWRNAL OF GEOLOGY 


“fanuary-February I 922 


AN OUTLINE OF THE GEOLOGY OF NEW ZEALAND: 


W. N. BENSON 
University of Otago, Dunedin, New Zealand 


Pre-Silurian.—In the more recent accounts of the history of 
New Zealand it has been customary to consider as of Pre-Cambrian 
age the complex of gneisses and associated metamorphic rocks in 
the southwestern extremity, but the evidence for so doing is not 
yet conclusive, and various authors have referred the plutonic 
gneisses therein to different periods, extending as late as the Creta- 
ceous. It would appear that the more probable hypotheses could be 
limited to two. 
The first definite point lies in the proved existence of Lower 
Ordovician rocks in the extreme northwest and southwest of the 
South Island (Fig. 1).2 These graptolitic slates are associated 
t This outline is based on the writer’s presidential address delivered to the geo- 
logical section of the Australasian Association for the Advancement of Science in 
1921. For detailed discussion of the most recent work, with full bibliography, reference 
may be had to this address. Earlier extensive accounts have been given by Marshall 
(“New Zealand and the Adjacent Islands,’ Stemmann’s Handbuch der regionalen 
Geologie, Bd. VII, Abt. 1, 1912), and Park (The Geology of New Zealand, Whitcombe 
and Tombs, 1910). 
2The map herewith (Fig. 1) has been based upon one recently issued by the 
Geological Survey (see New Zealand Journal of Science and Technology, 1921) and 
also incorporated in the above-mentioned address, thanks to the generous permission 
of the director of the geological survey, Mr. P. G. Morgan. The present copy has 
been somewhat modified in accordance with the view favored herein concerning the 
age of the Otago schists, etc. 


174 East 


| Neeo nen: | 


Lon 


: : 
COLUMNAR SECTION 


Pe|Acid ae Rocks, some Recent. mastly 

| Basic} Upper Tertiary, little Cretaceous 
se|POST-TERTIARY Gravels,Moraines etc. 
Period of block foulting & differential uplift. \ 
(AAPLIOCENE Marine Sediments & Gravels, 


ul Tranaression tn central area, regression elsewhere. 


Zrsonnst OLIGOCENE Marine sediments 


ZZ 


a Zs (after local erosion)Limestone, 


IAA marine sediments and basal coal measures. 
©) Widespread transgression§ slight unconformity } GS 
= AZ EOCENE Marine beds & basalcoal measures. > S 


©} Erosion interval and transgression . “a \ WELLINGPON 
2 HUPPER CRETACEOUS DanianLimestones y 
HSenonian Greensandetc basal coalmeasures. 
Albian marine beds (in Kaikoura Ranges only). 


Intense orogenic movement except in S.E. 


ZREOR IO 


rm 

EON x 
S OG eee Se 
ENN 
NN Once ee 


ONG. 


Pa: 


Noikaura fanges 


extremity. Intrusion of granite and basic 
rocks and perhaps of diorites in S.W 


re *| Granite. 
Peridotite , Gabbro, Norite etc. 


UO LOWER CRETACEOUS? Greywackes. 


re) 
souls Upper ) Marine greywackes and ; Res tae 
: ‘ z KORE Teninsula 
S34 =<{ Middle conglomerates with plant beds | se 


SOE 
SO 
SGN 


$S 


x 


=| Lower J and thin cal: seams. 


XRY 


& Rhaetic marine greywacke§ plant- beds. 


= Noric grey wacke limestone & basic tuffs. 
= Carnic greywackes with extensive fauna. 


HOKONUI 


SxS 


0: 
> 
QO 
ox 


XX 


Ss Un fossiliferous greywackes. 


Possible break in Sequence of un known extent, 


KS 
xs 
RS 

SHS 


a 


Various authoritiesclaim complete cnfarinity, 
miner regression, or great orogeny with 
intrusion of ¢neissic diorites. 
SLOWER TRIASSIC? Annelia Greywacke. 
ssi PERMIAN Greywacke,limestone, basic tuff: 
\ eee Schists ete of Centra! Otago 
Probably altercd form of the above. 
Conditions of this interval unknown. 
S|UPPER SILURIAN Reefton and 
1 Baton River Series 
ably strong unconformity. 


Pro 


SSJLOWER ORDOVICIAN Aorere Series RS Bg stem 
50 


assing into(?) schists & paragneiss. 
Fi 8 Pp 3 ISLAND 


| = DIORITE More or less gneissic. Age un- 
certain, Palaeozoic Triassic or in part Cretaceous. Miles 1 


00 


166 East 


Fic. 1 


Correction: Insert ? after ‘“‘Erosion interval” between EOCENE and UPPER 
CRETACEOUS in section above. 


THE GEOLOGY OF NEW ZEALAND 3 


with limestones, graywackes, and quartzites and pass with appar- 
ently gradual transition into mica-schists, which are invaded 
by gneissic diorites. A narrow zone of such schistose rocks extends 
intermittently down the west coast of the island. In the southwest 
the gneissic diorites become dominant and are associated in one 
region with “‘granitic”’ gneisses. In addition, there are massive 
plutonic rocks, chiefly granites, occurring especially in the two 
extremities of the island and surrounded by zones of contact 
metamorphism, which is naturally more marked where the regional 
metamorphism is least pronounced. The first hypothesis is 
that adopted by the Geological Survey. The gneissic rocks and 
invaded schists are considered to be of pre-Ordovician, possibly 
pre-Cambrian age, and covered unconformably by the Ordovician 
sediments. No sharp line of separation has, however, been found 
between them, nor other evidence of difference of age, such as the 
inclusion of gneissic pebbles in the Ordovician rocks. The massive 
plutonic rocks are considered to be of much later date, and prob- 
ably later Paleozoic or even in part Mesozoic. 

The second hypothesis regards the more or less gneissic dioritic 
rocks as having invaded the Ordovician sediments in Paleozoic 
or possibly Mesozoic times, before or during the climax of an 
orogenic movement, while the intrusion of the massive plutonic 
rocks followed after that climax. Within the present year massive 
diorite has been found to invade Lower Triassic (?) annelid- 
bearing graywacke. In a variant of this hypothesis advocated by 
Park, the subordinate mass of “granitic” gneiss is held to be 
intensely altered sedimentary rocks and regarded as probably 
' pre-Ordovician, merging upward into Ordovician slates. ‘The abun- 
dance of sillimanite in the gneiss accords with this view. Petro- 
graphical examination of the gneissic diorites on Grubenmann’s 
principles, first applied hereto by Speight, indicates that they form 
the core of a folded range, for structures characteristic of deep- 
seated metamorphism are found in the center of the massif, while 
those produced at lesser depths occur on either side. 

Bartrum has found that pebbles of similar metamorphic rocks 
are widely distributed in the Mesozoic and Cainozoic rocks of the 
North Island, suggesting the existence of an ancient Paleozoic 


4 W. N. BENSON 


platform beneath the younger formations. Their occurrence 
also supports the hypothesis that the dioritic rocks are not younger 
than early Mesozoic. 

Silurian.—The fossiliferous Silurian strata are known in two 
localities only, both in the northwest of the South Island. They 
strike in a direction usually about north-northwest, the same as 
that of the Ordovician rocks, but their relation thereto is as yet 
unknown, nor are the conditions at all favorable for investigating 
this. In the northern of the two areas (Baton River) the rocks are 
calcareous argillites, but in the southern occurrence (Reefton) 
they are of more littoral and varied nature, with a consequent 
difference in faunal facies. Both faunas are apparently closely 
related to the Upper Silurian (Wenlock) faunas of Southeastern 
Australia, but have not yet been very critically examined. 

‘‘ Permo-Carboniferous’ and Lower Mesozoic.—No further infor- 
mation is available until the close of the Paleozoic era, when com- 
menced the more nearly continuous portion of the stratigraphical 
record. The greater part of the mountainous country of New 
Zealand is made up of steeply folded and shattered argillites and 
graywackes, generally devoid of fossils. They contain, though 
rarely, lenticular masses of limestone, and also, especially in the 
upper portions intercalated plant beds. Widespread basic tuffs, 
etc., occur in the lower portion. On one higher horizon basalt flows 
are present. Numerous subdivisions and groupings have been 
proposed for this series, and Trechmann’s recent paleontological 
work, supported by that of Arber, Wilckens, and Boehm, seems 
to place on a firmer basis than formerly the division of the complex 
into a series of formations ranging from Permian to early Creta- 
ceous age. 

Permian.—The oldest fossiliferous rocks seem to rest conforma- 
bly on an extensive series of basic breccias, which are probably coeval 
with similar breccias which elsewhere rest unconformably on Silurian 
rocks. The fossiliferous beds, limestones, and argillites contain 
at one locality (near Nelson) a few poorly preserved shells and 
corals which have been compared with Eastern Australian Permo- 
Carboniferous (Permian) forms. Among these is a large, indefinite 
myalinid shell which was referred doubtfully to Inmoceramus and 


THE GEOLOGY OF NEW ZEALAND 5 


more recently to the Australian Permian form A phanaia, though 
probably both comparisons are erroneous. Fragments of this 
Shell, however, are widespread throughout the South Island of New 
Zealand, and in default of better evidence may be used to indicate 
the extent of Permian rocks. These are succeeded by a great 
thickness of graywackes in which an annelid tube is almost the 
sole indication of organic life. Jaworski considers this form indica- 
tive of a Triassic age. It is perhaps remarkable that there are 
no traces of plant beds known among these rocks. Glossopteris 
does not occur in New Zealand, though a form occurring in Upper 
Triassic and Jurassic beds was for a time doubtfully referred to 
this genus. 
| Triassic.—There is no clear evidence (beyond that mentioned 
above) of a great break, accompanied by plutonic intrusions, 
between the Permian and Triassic fossiliferous strata, though this 
has been assumed by some authorities. Probably, however, a 
regression of the sea occurred during Middle or Lower Triassic 
times, as in New Caledonia. The oldest fossiliferous Mesozoic 
sediments are referred to the close of the Middle Triassic, and are 
succeeded by fossiliferous Upper Triassic rocks, representatives 
of the three divisions of the Alpine-Himalayan Upper Trias, 
Carnic, Noric, and Rhaetic being recognizable, and each divisible 
into subzones. The fauna is Tethyan, with the interesting addi- 
tion of the circum-Pacific form Pseudomonotis ochotica. In general 
the succession of Upper Triassic faunal zones is very similar to 
that of New Caledonia. The local absence of well-marked horizons 
gives evidence of crust-warping during Upper Triassic times, and 
there is a notable development of. basaltic tuffs and lavas among 
the Noric rocks. The alternation of graywackes and argillites, and 
intercalation therewith of plant beds or conglomerates, indicate 
that these sediments were formed on a continental shelf. The 
flora of the region, which has been studied by Arber, was similar 
to the contemporaneous Australian flora, but the total number of 
species known in the Mesozoic rocks of New Zealand is only about 
a quarter of those known in Australia. 

Jurassic and Early Cretaceous——These general conditions of 
sedimentation were maintained during the Jurassic period, in 


6 W. N. BENSON 


which, however, the plant-bearing beds became relatively more 
abundant in the South Island, and thin coal seams occur. Lower 
(Liassic), Middle (Bajoccian), and Upper (Tithonian) Jurassic 
marine faunas are recognized, and a sequence of small floras also, 
but the detailed description of the first two has not yet appeared." 
There are also a series of sediments which are rather widespread 
in the east of the North Island, and are perhaps of early Cretaceous 
age, but may be in part Upper Jurassic. They contain Inoceramus 
in some abundance. In the western side of the island, the Jurassic 
rocks are followed conformably by early Cretaceous (Neocomian) 
plant beds, containing two angiosperms. 

Orogeny.—The orogenic movement which succeeded this long- 
continued sedimentation reached its maximum in Lower Cretaceous 
time. The Mesozoic sediments were forced into broken or overturned 
folds, and zones of shattering were produced. The axis of such 
folding is for the most part approximately meridional and generally 
oblique to the much later warpings and fault-lines which determine 
the northeasterly trend of the present mountain system. Plutonic 
intrusions occurred in connection with this folding, among which 
are a series of masses of ultrabasic and basic rocks appearing at 
intervals from end to end of New Zealand. They are most note- 
worthy near Nelson in the north of the South Island, where is the 
type-locality of the rock dunite. The association of such intru- 
sions with fault-zones 1s sometimes clear. In addition to these 
are granites and syenites especially well developed in the northwest 
of the South Island, but it is as yet problematical how much of the 
more or less gneissic plutonic rocks of the southwestern region 
should be referred to this series of intrusions. Among the minor 
intrusions of this period are a restricted series of lamprophyric 
dikes, monchiquites, etc. 

Undetermined schists —We may here describe one of the most 
puzzling formations in New Zealand, a broad zone of mica-schists 
‘running to the southeast from the main mountain range, through 
the Otago province of the South Island. These have for the 
most part a very flat position, but dip to the southwest and north- 


*Trechmann and Spath have described this fauna since this paper was written. 
See Quart. Journ. Geol. Soc., 1921. 


THE GEOLOGY OF NEW ZEALAND 7 


east of the broad anticlinal axis, which 
reaches the sea near Dunedin. To these 
rocks all ages have been assigned from 
Archaean to Jurassic. ‘They are not as- 
sociated with plutonic rocks, but may 
be traced outward from a central zone 
of maximum schistosity, through gradu- 
ally decreasing metamorphism into 
graywackes indistinguishable from the 
Ordovician or Lower Mesozoic gray- 
wackes. The Geological Survey and 
the majority of other authorities hold 
that the schists are ancient and must 
be separated by obscure disconformities 
from the Mesozoic rocks. Marshall 
considers the schist Mesozoic, and, after 
much hesitation, the writer inclines 


toward a modification of this view. In -- 


this it is suggested that the flat arch of 
metamorphic rocks is the base of a great 
series of recumbent folds of late Paleo- 
zoic and Lower Mesozoic rocks which 
were pressed against a resistant or con- 
tinental mass now concealed beneath 
the southern portion of the island. In 
the front of such a series of recumbent 
folds one would expect to find a sharp 
' fourfold-wrinkle of the crust, beyond 
which the thrusting would die away in 
‘gentle undulations of the strata lying 
on the resistant mass. Within the 
overfolded areas, block-faulting might 
bring down the comparatively unmet- 
amorphosed rocks of the higher recum- 
bent folds into close apposition with 
the more schistose rocks of the lower 
folded sheets (Fig. 2). These expecta- 


NE. 


TIMARU 
Basalt on 


Tertiary 
Marine Beds 


, 


Length, 180 miles 


! 


an? schiicy. 


c 
! 
! 
' 
' 
! 


\ vern 


Fic. 2.—Hypothetical section from Waikawa to Timaru across the broad “anticline” of Otago. 


VWALKAVA 


aoe 


cansenial Ingy 


(A 


8 W. N. BENSON 


tions appear to be realized in some measure in the facts observed in 
field study. The microscopical structure of the schists is indicative 
of strong lateral thrust rather than static metamorphism, and an 
excellent series of gradational structures which link them with the 
graywackes has been obtained by Marshall. Moreover, the 
apparent absence from the graywackes of any material which seems 
to have been derived from the schists is a striking feature, con- 
sidering their close association. 

T he ‘‘Notocene”’ Sediments.—The record of the later Cretaceous’ 
and Tertiary times has as elsewhere received diverse interpretations. 
The simplest among current views considers that the whole series 
is conformable throughout. Overlap on to an evenly subsiding 
but irregular surface accounts for the difference of age of the basal 
beds in different regions. The period of greatest submergence 
was the period when limestone was deposited, so that the lime- 
stones in all regions must be considered as of the same age. They 
are succeeded by beds of a clastic character, indicating the return of 
shallower conditions. On other hypotheses the whole group of 
formations is divided up into several unconformable series, but 
the ages assigned these series, and the horizons at which the uncon- 
formities were recognized, have been different in the statements 
of different writers, or at different times in the statements of one 
writer. While, therefore, there has been little dispute as to the 
succession of strata in any region, the history of the whole period 
throughout New Zealand has remained obscure. 

Two new conceptions have been advanced during the last 
decade. Thomson has suggested the existence of ‘“‘diastrophic 
provinces,” i.e., regions throughout each of which the tectonic 
history has been the same, though differing from that of adjacent 
regions. On this hypothesis, the difference of age of basal beds 
depends on the overlap of formations on a subsiding uneven bed 
but also on the different periods at which subsidence commenced 
in the various provinces. The limestones are not necessarily 
coeval throughout New Zealand, but represent merely the rock 
formed at the time or times of maximum submergence in each 
particular province, for in some districts more than one horizon 
of limestone is present. So also the regression of the sea occurred 


THE GEOLOGY OF NEW ZEALAND 9 


at different times, and the highest beds in one province may not be 
coeval with those in an adjacent province. There is no need to 
consider that any unconformities or disconformities of a general 
nature are present, though local breaks may occur. Thomson 
has, moreover, suggested the convenient term ‘‘Notocene”’ (the 
‘Southern New” formation) to indicate the whole group of sedi- 
ments laid down between the Lower Cretaceous orogenic period 
and that which, occurring possibly at the close of Pliocene time, 
ushered in the present physiographic cycle. 

The areal work of the officers of the Geological Survey has 
given rise to the second conception. Block-faulting and tilting 
of the crust, which was so extensive a process of the post-Notocene 


| 


SSS 
Cl —————s 
NING TANNA AAW, 


Fic. 3.—Diagram illustrating the conception of a general but deceptive conformity 
obscuring an erosion-interval in the deposition of ‘“Notocene” rocks. 


————— 
——SSS=S=S==sJ 
—————— 
SS 
—— 
SS 
—————— wl 
——————— 
ZW 


(Pleistocene?) orogeny, was not confined thereto, but occurred 
at intervals from Middle Cretaceous times onward. The move- 
ments were chiefly vertical, but along fault-planes the strata may 
be much crushed or upturned. Erosion proceeded pari passu 
with elevation, and the beds laid down after such movements may 
rest with apparently perfect conformity on the undisturbed areas, 
but with small or great unconformity where lying on the eroded 
surface of a tilted block, or near a fault-plane, or they may cross, 
without disturbance a zone of fault-breccias in the underlying 
Lower Notocene rocks (Fig. 3). If these two conceptions be 
admissible, we have the explanation of much apparent conflict of 
evidence seen when we attempt to correlate the stratigraphi- 
cal succession in various regions. We must also note that such 
generalizations as are in the columnar statement on the map 
herewith must be subject to modification in detailed application 
to special areas. | 


Ic W. N. BENSON 


Later Cretaceous.—The earliest of the Notocene incursions of 
the sea entered a small part of the northeastern portion of the 
South Island in Middle Cretaceous times, flooding a very irregular 
surface. Sixteen species of fossils have been recognized by Woods 
in the deposits of this period, forms identical or allied with species 
typical of the general Indo-Pacific fauna of the period. What 
follows is not quite clear, but probably the sea retreated, and 
entered again in late Cretaceous times, covering a much more 
extensive area and submerging other portions of the northeast 
and also in the southeast of the South Island and the east and 
northern portions of the North Island. A much more extensive 
fauna was introduced than before, over sixty species being known, 
which, according to Woods, Trechmann, and Wilckens, are of the 
typical Indo-Pacific Senonian types, with a particularly marked 
affinity with the fauna of Southern America and Grahams Land. 
Indeed, it seems as if we must conclude that New Zealand, Ant- 
arctica, and South America lay, at the close of Cretaceous times, 
on the coast of the South Pacific Ocean, involving the hypothesis 
of a land connection at this period between these regions, a con- 
clusion to which biologists have long tended.t In some parts of 
New Zealand the Senonian beds are followed by a great thickness 
of almost unfossiliferous though partly foraminiferal limestone, 
to which a Danian age has been assigned. ‘This stage is, however, 
missing from many districts. 

Eocene.—In “‘Eocene”’ times,? the sea spread in certain regions, 
but was missing from others. Three contrasted developments of 
beds of this epoch may be noted. At this time were formed the 
most valuable coal-measures in New Zealand, those in the north- 
western portion of the South Island, the recent survey of which has 
been in large part due to Morgan. ‘The highlands adjacent to this 

* There is very little affinity between the Cretaceous faunas of New Zealand and 
Australia. 


2 The Lyellian terminology may perhaps be retained for the major subdivisions 
of the Notocene period, but since we do not know the relative rate of evolution of new 
marine forms in New Zealand and Europe, we cannot conclude that strata with like 
percentages of Recent forms in the two regions are coeval, and hence the terms based 
on such percentages can have but a local comparative value. For this reason the 
convention is adopted of placing them between quotation marks. 


THE GEOLOGY OF NEW ZEALAND met 


region were still unreduced, and a great thickness of conglomerate 
was laid down prior to the formation of the coal-measures, which 
in turn are succeeded by marine mudstones, the fossils of which 
contain less than 5 per cent of recent forms. On the other hand, 
in the southeastern portion of the South Island, the Upper Cre- 
taceous coal-measures are succeeded by soft mudstones or impure 
limestones, which Marshall found to contain less than ro per cent 
of Recent forms in rather extensive faunas, including also some 
characteristic early Tertiary genera. In the northeastern part 
of the South Island a series of marly rocks overlying the above- 
mentioned Danian limestone may represent this epoch. 

Oligocene and Miocene.—By the close of “Eocene” times the 
mountains formed during the Cretaceous crust-folding had been 
reduced approximately to a peneplain and the marine trans- 
gressions of ‘‘Oligocene” and ‘“‘Miocene” times submerged the 
greater portion of the South Island, and much of the North Island, 
and a very rich molluscan fauna was present. There was, however, 
considerable local variation in the development of such marine 
rocks, a discussion of which would necessarily lead into much 
detail, with the difficulties of conflicting correlations and a con- 
fused nomenclature. It should, however, be noted that in the 
central region of Otago in the South Island, where, it is suggested, 
the recumbent folds of the Cretaceous orogeny were most highly 
piled, reduction by erosion and subsidence seems to have been 
retarded, so that this region remained above the level of the Tertiary 
sea, but in ‘‘Miocene” and perhaps also in “Pliocene” time was 
largely covered with fluviatile and lacustrine deposits, often richly 
' auriferous. 

Pliocene.—Crust-warpings at the close of “‘Miocene”’ times 
caused the sea to retreat from the greater part of the South Island, 
though it transgressed on to its northeastern angle. The North 
Island, however, was largely if not completely submerged, except 
for the great volcanoes which now commenced their activity. In 
this transgressive sea ‘‘Pliocene” beds were laid down, in some 
cases with quite noticeable unconformity upon “Miocene” beds. 
One district calls for special mention, that of Wanganui in the 
southwestern portion of the North Island which must surely 


I2 W. N. BENSON 


become a classical district for the study of later Tertiary marine 
rocks in the Southern Hemisphere. A nearly continuous line of 
sea cliffs exposes a thickness of 3,500 feet of gently dipping clay 
stones. No break is seen in the succession, nor any sign of faulting, 
though two carbonaceous layers indicate temporary land-surfaces. 
Large collections of shells were made by Marshall and Murdoch 
at four different horizons. In the lowest were 61 per cent of 
Recent forms; in the next 76 per cent; in the next 90 per cent; 
and in the highest 93 per cent. Sands containing only modern 
shells rest with obvious unconformity upon this series, an almost 
dramatic conclusion to the Tertiary record. 

Middle Tertiary faunas—The marine faunas of Middle Tertiary 
times in New Zealand contain many new immigrant forms and 
show signs of South American influence, which scarcely are sufficient 
to indicate coastal connection at this time, though the affinity of 
the New Zealand fauna with that of South America is greater than 
that with Australia. Moreover, since the Middle Tertiary period 
New Zealand appears to have been entirely isolated.t The modern 
marine fauna is but a diminished residue of the Middle Tertiary 
fauna, and its specific and generic distinctness from Australian | 
forms precludes a connection of these areas in “Pliocene” and 
post-‘‘ Pliocene”’ time. 

Vulcanism.—During the Notocene period there was considerable 
volcanic activity. Its earliest manifestation was in a few Middle 
Cretaceous basaltic eruptions in the northeastern portion of the 
South Island, and rhyolitic extrusions, to the east and west of 
Christchurch, which are probably Upper Cretaceous. There are 
also west of Christchurch some post-Jurassic, pre-Senonian ande- 
sites, the relations of which are now being investigated. The 
products of early and middle Tertiary activity are much more 
important and widespread. Economically the most important 
among these were the andesites and dacites*? of the Coromandel 
Peninsula, North Island, in which post-magmatic solutions have 
developed great auriferous deposits. Rhyolites also occur in this 

Tt is of interest to note in this connection that characteristic genera of modern 
New Zealand plants are represented among the Tertiary fossil leaves in Grahams Land. 


2 Included with the basic volcanic rocks in map herewith. 


THE GEOLOGY OF NEW ZEALAND 13 


complex. Somewhat later were the eruptions of basalt, etc., 
which formed Banks Peninsula, described by Speight, and the 
basalts and varied alkaline rocks about Dunedin, which Marshall: 
has studied. In Upper Tertiary times volcanic activity broke out 
in the center of the North Island, where it has continued to the 
present time. Andesites, and more or less pumiceous rhyolites, 
are the chief products of this activity, the principal centers of which 
are arranged on a line following the northeasterly ‘‘grain” of the 
country. Mount Egmont, the great isolated cone in the west of 
the island, also consists of andesite. 

Pleistocene diastrophism.—The period which followed the 
cessation of Notocene sedimentation, though comparatively short, 
was that during which New Zealand assumed its present form as 
a result of a great series of differential movements, warping or 
tilting of crust-blocks, and the denudation of the surface so pro- 
duced. The nature of these processes has been elucidated by 
Cotton. The boundaries of the several blocks are now marked by 
fault-scarps in homogeneous structures, or by faulted contacts of 
older and younger strata, at which the latter are often steeply 
upturned. Usually the faults are oblique to the strike of the 
folded strata they truncate, and very frequently extend in a north- 
easterly direction. The movement was not all due to simple tension 
and differential subsidence of blocks, but strong compressive 
lateral thrusts also occurred, with the occasional production of 
folding passing into faults, of overthrusting and even overfolding 
(Fig. 4). 

The faults present the following characteristics. The movement even 
/ along the same dislocation, may be concentrated in a single fracture with 
walls close together, or perhaps several chains apart, the intervening space 
being filled with comminuted rock. Again the fault may bea shear-zone.... . 

One type of fault constantly recurs: narrow trough-faults in which the 
rock between the main fault-walls belongs to a higher horizon than the walls 
themselves. When the Tertiary beds which overlie the more ancient sediments 


and graywackes are involved, the recognition of this type of fault is very 
easy. [Henderson.] 


The surface of New Zealand at the close of the Pleistocene 
orogeny was thus that of a group of variously elevated earth-blocks 


14 W. N. BENSON 


composed usually of hard graywacke or schist, on the more or 
less planed surface of which rested much less resistant Tertiary and 
sometimes Cretaceous sediments. In many elevated blocks 
the hard basement rocks rose to a higher level than the compara- 
tively weak covering strata in the adjacent relatively depressed 
blocks, which formed broad, fault-bounded, intermontane basins 
or narrow, rectilinear rift valleys, or fault-angle valleys. A conse- 
quent drainage was soon established on the broken sheet of covering 
rocks, which it commenced to remove. Where this process is 
still incomplete, topography is controlled in large measure by the 
variation in the resistance to erosion offered by the different strata 


YAR OS a fe? 


Fic. 4.—Section at Lake Wakatipu showing extensive movement in the post- 
Tertiary orogeny. «, sandstone and conglomerate; b, limestone; c, marly sandstone; 
d, marly clays; ¢, brecciated Tertiary marine rocks; m.s., mica schist. (After Park.) 


in the covering rocks, and dip slopes and scarps are characteristic 
features of the scenery. Where, however, erosion is more advanced, 
portions of the planed surface of the underlying rocks are laid bare, 
and where the cover is cleared away from extensive areas, a stripped 
peneplain is exposed, the further reduction of which is very slow 
(UB 5) 

Late erosion.—An enormous amount of detritus results in the 
stripping of the covering strata from the surface of the elevated 
blocks and from the dissection of the fault-scarps. ‘This is only 
exceptionally removed as it is supplied. In most places deep 
ageradation of the troughs has taken place concurrently with the 
dissection and degradation of the higher blocks. 


THE GEOLOGY OF NEW ZEALAND 15 


As a result of this double process many interesting features 
have been produced in the development of the topographic forms 
and especially valley systems on this complex land surface. In 
some regions, especially those composed of the older pre-Notocene 
rocks, the control of the drainage by structural features is indicated 
by the reticulated plan of the valley systems, the association of 
valleys with marked lines of faulting, or with zones of fault-breccias. 
This is especially noteworthy where narrow masses of the softer 
covering strata have become involved in the fault-zone, but some 
control of drainage by fault-zones may be observed even in regions 
in which the soft covering rocks only are exposed. Probably not 


= ~~ 
=| Ss ~ 


Si 


Mi, 


five 


ih au 
i : Sea Level 
Wc 
Hew iy /\ Wms 


ie: 


J 


Fic. 5.—Diagram to illustrate the evolution of the present topography. 4, effect 
of block-faulting of region covered by weak “‘Notocene” sediments; B, modern to- 
pography resulting from denudation of above surface and aggradation in the troughs. 


only the somewhat reticulate character of the valley systems in the 
argillites, etc., of the northeast of the South Island has been thus 
influenced, but also the rectilinearly branching valleys which now 
form the fiords traversing the gneisses, etc., of the southwestern 
portion of the same island. Some of the more open, though still 
long and narrow depressions may be due, not merely to the differ- 
ential erosion of a strip of soft material among harder rocks, but 
even to actual trough-faulting. The disposition of the covering 
rocks about one lake (Te Anau) in the southwest of the South 
Island affords proof of such in this particular instance. 
Glaciation.—The topography does not, however, depend 
actively on structure and normal differential subaerial erosion, for 


16 W. N. BENSON 


the effects of glaciation are also evident. It seems clear, how- 
ever, that this was restricted to the highlands and the valleys 
among them reaching sea-level only along the west coast of 
the South Island. The ice did not advance to form a large 
confluent piedmont glacier beyond the eastern slopes of the 
Southern Alps, though it deployed into sheets of considerable 
area in the lake basins and associated depressions, notably that of 
the above-mentioned Lake Te Anau. The view that an ice sheet 
extended to the low southeastern portion of the island does not 
appear to be acceptable. As a result of glacial erosion, the valleys 
in the mountainous portions of the South Island, though originally 
determined as explained in the previous paragraph, have since 
been considerably modified. The topographic features character- 
istic of glaciation have been developed, and there are even indica- 
tions of the effects of more than one cycle of mountain glaciation. 
It is not clear what extent of glaciation may have been present in 
the North Island, but there seems to have been little if any. 

In the rain shadow of the Southern Alps an area of very small 
precipitation occurs in the center of the province of Otago, and 
here topographic features characteristic of semi-aridity may be 
seen. 

Piedmont aggradation.—Concurrently with the degradation of 
the mountains, there has been much sedimentation, forming 
piedmont plains. A series of gravels formed during the elevation 
of the Southern Alps have been unconformably covered by the 
gravels of the Canterbury Plains. These consist of a great sheet of 
detritus over 130 miles long and 30 miles wide, the confluent fans 
of the rivers draining the eastern slopes of the mountains, at the 
foot of which they rise to a height of 1,000 feet above sea-level, 
but they have been built out over a sea floor which has subsided 
at least 600 feet during their formation. Less extensive plains of 
gravel and alluvium occur in other districts, each with special 
features of interest. 

Loess.—Of lesser importance are the deposits of loess along the 
central portion of the eastern coast of the South Island. The loess 
is composed of the rock-flour carried by the dominant northwesterly 
winds from the dried pools in the braided valleys of the glacier-fed 


THE GEOLOGY OF NEW ZEALAND 7 


streams, and has accumulated to a depth of 20 feet or more in 
suitable situations. Bones of the moa are found in these deposits. 

Recent movements.—Since the epoch of great differential dis- 
placements of relatively small earth-blocks, which was the essential 
character of the “‘ Pleistocene” orogeny, there have been a series of 
relatively small movements involving much larger earth-blocks, 
which movements have been elevation, depression, or tilting. 
Exceptionally localized differential movement has resulted in 
fault-coasts or strongly warped surfaces. Between these move- 
ments were long periods of crustal stability during which the 
cycles of erosion reached a fairly advanced stage. Hence along 
the coast in different districts there are marked raised beaches, 
while extending far up the valleys are rock-benches or alluvial 
terraces, indicating that the valleys are not monocyclic but were 
rejuvenated from time to time. 


PENNSYLVANIAN STRATIGRAPHY OF NORTH 
CENTRAL TEXAS 


RAYMOND C. MOORE anp FREDERICK B. PLUMMER? 


University of Kansas, Lawrence, Kansas, and Roxana Petroleum Corporation, 
St. Louis, Mo. 


INTRODUCTION 
Previous Work 
Recent Investigations 
Acknowledgments 


GENERAL DESCRIPTION OF THE NORTH CENTRAL TEXAS PENNSYLVANIAN 
Surface Distribution 
Lithologic Character 
Thickness 
Topography 
Divisions 
BEND GROUP 
STRAWN GROUP 
CANYON GROUP 
Cisco GROUP 


PHYSICAL History OF THE TEXAS PENNSYLVANIAN 


INTRODUCTION 


The Pennsylvanian strata of north central Texas furnish a 
beautifully exposed and nearly complete section of the rocks of 
this period, exceedingly variable in lithologic character and prolific 
in well-preserved fossils. Yet because of their isolation and the 
semiarid, somewhat forbidding character of the country in which 
they outcrop, detailed study of them has been long delayed. The 
discovery in recent years of great deposits of petroleum within this 
area has brought to it many geological workers, and has served as 
a great stimulus to the study of the stratigraphy, differentiation, 
and correlation of its formations. 

t Published by permission of the Director, Bureau of Economic Geology and 
Technology, University of Texas. 

18 


PENNSYLVANIAN STRATIGRAPHY OF TEXAS 19 


_ Previous work.—The Pennsylvanian area of Texas was first 
explored by Roemer’ in 1846 and later by Shumard,? Ashburner, 
and others. The most important contributions to the stratigraphy 
of the north Texas country, however, are contained in the writings _ 
of Tarr,4 Cummins,’ and Drake,° who made investigations of the 
coal fields of the Colorado and Brazos river valleys in the late 
eighties and early nineties for the first Texas Geological Survey. 

Cummins’ recognized six divisions in the rocks of the Texas 
Pennsylvanian, (1) Bend, including the black shale and limestone 
typically exposed in San Saba County; (2) Millsap, comprising 
shale, limestone, and sandstone exposed in the Brazos River Valley; 
(3) Strawn, including the strata, chiefly shale and sandstone, 
between the first coal and the base of the massive limestones of the 
succeeding division; (4) Canyon, the dominantly limestone division 
in the middle or upper part of the Pennsylvanian section; (5) 
Cisco, composed of shale, sandstone, and thin limestones above the 
Canyon limestones and below the Red Beds; and (6) Albany, 
consisting of thick limestones and shales above the Cisco, at first 
thought to belong to the Coal Measures, but later referred to the 
Permian. Cummins did not attempt to subdivide these large 
units, but the broad groups which he recognized and the names 
which he applied are those in use at the present time. In the 
main they appear to be very well chosen and the contributions of 
_ this pioneer worker based on work done under great difficulties are 
most important. 

Drake’ was first to make a detailed study of any portion of the 
_ Pennsylvanian in Texas. In mapping the coal field of the Colorado 
River Valley, he differentiated the large groups of Cummins into 
many smaller units which he described and named. He fixed the 

tF. Roemer, Die Kreidebildungen von Texas (Bonn, 1852). 

2B. F. Shumard, Trans. St. Louis Acad. Sci.,; Vol. I (1860), pp. 686-87. 

3C, A. Ashburner, Trans. Amer. Inst. Min. Engrs., Vol. IX (1881), pp. 495-506. 

4R.S. Tarr, Texas Geol. Surv., First Ann. Rept. (1889), pp. 201-16. 

5 W. F. Cummins, Texas Geol. Surv., First Ann. Rept. (1889), pp. 145-82; Second 
Ann. Rept. (1890), pp. 359-94. 

6N. F. Drake, Texas Geol. Surv., Fourth Ann. Rept. (1892), pp. 357-481. 

7W. F. Cummins, Texas Geol. Surv., Second Ann. Rept. (1890), p. 375- 

8 N. F. Drake, loc. cit. 


20 RAYMOND C. MOORE AND FREDERICK B. PLUMMER 


limits of the Strawn, Canyon, and Cisco in the area studied by him, 
designating definite stratigraphic horizons at the top and bottom 
of each. Many of the subdivisions defined by Drake are employed 
in the present classification, but in a number of cases it has seemed 
necessary to depart from his usage and to apply appropriate 
geographic names for units to which he gave merely descriptive 
names, as ‘‘Cherty” or ‘‘Coral’’ limestone. . 

In addition to the investigators whose work has just been 
mentioned, a considerable number of geologists have reported 
studies of value on various portions of the Texas Carboniferous 
beds. Among these may be mentioned Hill,t Gordon,’ Paige,’ 
Udden,‘ Girty,5 Moore,® and Plummer.’ All of these investigations 
have been studied carefully by the writers in connection with the 
present work. 

Recent investigations —In December, 1916, the geologists of the 
Roxana Petroleum Corporation began the systematic mapping of 
the surface and structural geology of the northern portion of the 
Pennsylvanian area in Texas. As the work progressed the maps 
were fitted together and the data slowly compiled for a new, 
detailed, geological map of north Texas. At present almost the 
entire area of the Pennsylvanian outcrops has been mapped, the 
outcrops of all the principal limestones and many other beds having 
been traced by means of plane table and alidade. Several hundreds 
of geological sections have been measured and carefully described. 
Fossils have been collected from most of the stratigraphic divisions, 
those from some horizons having been studied in detail. Finally, 
much information has been obtained from the records of numerous 
wells over a large area in north Texas which have been duulles 
deeply into or through the Pennsylvanian. 


tR. T. Hill, U.S. Geol. Surv., Twenty-first Ann. Rept. (1901), Part VII. 
2C. H. Gordon, U.S. Geol. Surv., Water Supply Paper 276 (1911). 

3 Sidney Paige, U.S. Geol. Surv., Geol. Atlas, Folio 183 (1912). 

4J. A. Udden, Bur. Econ., Geol., and Tech., Bull. 44 (1916). 

5G. H. Girty, Bull. Amer. Assoc. Petrol. Geol., Vol. III (1919), pp. 71-81. 


6R. C. Moore, Bull. Amer. Assoc. Petrol. Geol., Vol. III (1919), pp. 216-52; G. H. 
Girty and R. C. Moore, Bull. Amer. Assoc. Petrol. Geol., Vol. III (1919), pp. 418-20. 


7F. B. Plummer, Bull. Amer. Assoc. Petrol. Geol., Vol. III (1919), pp. 132-50. 


PENNSYLVANIAN STRATIGRAPHY OF TEXAS 21 


Acknowledgments.—Cordial acknowledgment is made of the 
assistance of Dr. W. A. Van Waterschoot Van der Gracht, president 
of the Roxana Petroleum Corporation, and of Mr. Richard Conk- 
ling, head geologist, through whose courtesy the results of the 
stratigraphic studies of the Texas Pennsylvanian are presented. 
Among the geologists who have contributed importantly to the 
progress of:the study are Messrs. John Burtt, Paul Applin, James 
Armstrong, Sam Wells, Chester Hammill, Angus Mcleod, Grady 
Kirby, E. G. Allen, and Miss Linda Green. 


GENERAL DESCRIPTION OF THE NORTH CENTRAL 
TEXAS PENNSYLVANIAN 


Location.—The Pennsylvanian area of north central Texas may 
be described as two great inliers of Carboniferous rocks which 
protrude through the Cretaceous strata on the east and dip beneath 
Permian rocks on the west and north. The two areas are separated 
_ by a narrow tongue of Cretaceous (Trinity) sand, and the southern 
outcrop rests against Ordovician rocks for a short distance along 
the Llano uplift, so that the southern portion does not possess the 
relations of a true inlier. 

The total area covered by the Pennsylvanian is about 7,000 
square miles. It includes the west part of Montague, the south- 
east part of Clay, the greater portion of Jack, Young, Stephens, 
Palo Pinto, Eastland, Brown, the east half of Coleman, the north 
part of San Saba, and the northeast of McCulloch counties. The 
shape and location of the Pennsylvanian area are shown on the 
index map, Figure 1. 

Lithologic character —The lower portion of the Pennsylvanian 
rocks consists of massive blue, gray, or black limestone, and greenish- 
gray to black argillaceous and bituminous shale. As observed at 
the outcrop, these are not interbedded, but a division consisting 
almost wholly of limestone, 400 to 500 feet in thickness, lies between 
two shale formations. This portion of the Pennsylvanian, the 
Bend, contrasts in lithologic character with the remaining rocks of 
the system. It is the chief petroliferous horizon in north Texas. 

The Millsap and Strawn divisions are dominantly clastic, the 
latter, especially, being composed of very massive, more or less 


22 RAYMOND C. MOORE AND FREDERICK B. PLUMMER 


conglomeratic sandstones and alternating sandy shales. The 
resistant beds in this portion of the section are beautifully exposed 
over large areas in the Colorado and Brazos river valleys. 

The Canyon consists of massive limestones, from a few feet to 
as much as 250 feet in thickness, alternating with shales. Although 
commonly designated a limestone division, its character is in no 


| RES Tertiary and Quaternary 
Cretaceous 


= Triassic-Jurassic 
ZZ Permian 


[J PENNSYLVANIAN 
HNNI Undifferentiated Paleozoic 


|_| Pre-Cambrian 


[ts] Igneous 


Fic. 1—Index geological map of Texas 


way comparable to the Bend, for in the Canyon, the limestones are 
thinner and are interbedded with the shales. The limestones are 
hard, fine-grained, in part very cherty, and are not as a whole very 
fossiliferous. The shales are chiefly yellow to gray in color and 
clayey rather than sandy. 

The Cisco is composed of thick shales and more or less con- 
glomeratic sandstones, thin limestones, and some coal. The shale 


PENNSYLVANIAN STRATIGRAPHY OF TEXAS 23 


and sand are much the most important rock types, but the lime- 
stones form persistent escarpments and are prominent, especially 
in the upper part and to the south. Conglomerates are found 
chiefly in the north. They are composed, for the most part, of 
small angular fragments of the resistant materials from other 
sedimentary formations. The shales are commonly sandy. The 
limestones are for the most part fine-grained and yellow to gray 
in color. 

Thickness.—The total thickness of the Pennsylvanian of Texas, 
computed from measurements of the surface outcrops, is as a 
maximum about 6,800 feet. A compilation of average thicknesses 
for each of the divisions gives a total of 5,350 feet. As a matter 


Breckenridge 
Mineral Wells 


Millsap 


= ss 
=S— 


SS = 
__ SOS Sat obs 


Fic. 2.—Diagrammatic section through the Pennsylvanian of Texas 


of fact, the thickness of the total section is somewhat less than 
the total of measurements of thickness at the outcrop, for there 
appears to be a very important overlap of the beds from east to 
west, so that above the Bend the older divisions disappear suc- 
cessively in going to the west. This is illustrated by the diagram- 
matic section, Figure 2. 
The average thicknesses of the divisions of the Pennsylvanian in 
Texas are indicated in the table of formations. 
_Topography.—The outcrop of the Pennsylvanian rocks and of 
each of the major divisions of the Pennsylvanian in north Texas 
is characterized by the topography. The Pennsylvanian area as a 
whole is distinguished by its prominent and very persistent escarp- 
ments. Due to the character of the soil and the climate, there is a 
widespread cover of mesquite, scrub oak, and cactus, with belts of 


24 RAYMOND C. MOORE AND FREDERICK B. PLUMMER 
CLASSIFICATION OF THE TEXAS PENNSYLVANIAN 
niches Colorado River Valley Brazos River Valley . 
pee ti oe {Coleman Junction limestone]! Coleman Junction limestone 
Crmaton T25175 _|\Santa Anna Branch shale | Santa Anna Branch shale 
M Sedwick limestone Sedwick limestone 
ee i x Santa Anna shale Shale 
ormation | 1507209 |\ Horse Creek limestone= ? | Dothan limestone 
Watts Creek shale Shale 
Puebl Camp Colorado limestone Camp Colorado limestone 
ee g i 5 Shale Shale 
Se eae TS0~200 | Stockwether limestone= ? | Eolian limestone 
Camp Creek shale Shale 
Saddle Creek limestone Saddle Creek limestone 
Bla ‘ll Shale 
S Pee £ = Belknap limestone 
b) Oru loi BOATS) Shale and coal Shale and coal 
8 Crystal Falls limestone 
5 Shale and sandstone 
Breckenridge limestone Breckenridge limestone 
Shale Shale 
Thrifty “‘Lower Chaffin” limestone 
formation 100-200 (Drake) = ? Black Ranch limestone 
Shale Shale 
“Speck Mountain” lime- 
stone (Drake) = ? Ivan limestone 
Shale and sandstone Shale 
Avis sandstone 
Wayland shale 
Wayland shale Gunsight limestone 
Gunsight limestone South Bend shale 
Graham Bunger limestone 
formation To0—-600 | Bluff Creek shale Gonzales Creek shale 
Jacksboro limestone 
Finis shale 
~ |Caddo Creek Y {Home Creek limestone Home Creek limestone 
formation SomSe \Hog Creek shale Hog Creek shale 
Bend Ranger limestone Ranger limestone 
a te Gan = Placid shale 
2 eet ™75~25° |\ Clear Creek limestone Seaman Ranch shale 
1S) Cedarton shale 
= 
5 Gerona Adams Branch limestone Adams Branch limestone 
3 ie ee ti ts Brownwood shale Brownwood shale 
oO ial 3 please Capps limestone bed 
Rochelle conglomerate 
Palo Pinto 
limestone 50-100 | Not present Palo Pinto limestone 


PENNSYLVANIAN STRATIGRAPHY OF TEXAS 25 


CLASSIFICATION OF THE TEXAS PENNSYLVANIAN—Continued 


hicks Colorado River Valley Brazos River Valley 
Keechi Creek shale 

oy Turkey Creek sandstone 
2 Salesville shale 
‘} y Lake Pinto sandstone 
~ | Mineral Wells ; : East Mountain shale 
E formation 500-800 | Undifferentiated Brazos River sandstone 
& Mingus shale 
n Thurber coal 

Millsap 

formation | 1,800-3,000 Millsap formation (undiffer- 
entiated) 

a Smithwick 
2 shale 400 Undifferentiated | 
(EY oa Aa 
Marble Falls i : Present, but not exposed 
g limestone 400-500 | Undifferentiated 
2 TEE SRR nu ES Se RN 

Barnett shale 0-50 Undifferentiated | 


cedar locally along the prominent limestone ridges. The area of 
the Bend has a rough, semimountainous topography which results 
from the resistant character of the massive, thick limestone of the 
Marble Falls formation. The Strawn area is distinguished by 
prominent but irregular escarpments which are produced by the 
hard sandstone beds. The weathering of the shales and the 
disintegration of the sandstone produces flat, sand-covered bottom 
lands, but the bold escarpments along Brazos and Colorado rivers 
dominate as topographic features. The Canyon, as the name some- 
what fortuitously suggests, gives rise to a very rough, deeply 
chiseled topography which is one of the prominent geographic 
' features of north central Texas. The massive limestone beds make 
high, sharp-edged escarpments which make travel from east to 
west very difficult. The Cisco produces a topography of gentler 
relief, with broad, open valleys and less prominent, though well- 
defined, escarpments. 

Divisions—In making the new geological map of the Texas 
Pennsylvanian, attempt has been made to present a classification 
which, conforming as closely as possible to the well-known divisions 
of Cummins, will apply equally to the whole area from the north 
to the south. Carefully measured and studied sections across the 


26 RAYMOND C. MOORE AND FREDERICK B. PLUMMER 


Pennsylvanian along a number of selected lines were compared, and 
data gathered from mapping and evidence from invertebrate fossils 
were considered. The resulting classification of the sediments is 
shown in the following table. 


BEND GROUP 


The strata included in the Bend group are exposed in northern 
San Saba, McCulloch, western Lampasas, and Burnet counties, 
areas bordering the Central Mineral Region of Texas. The group 
is named from McAnnelly’s Bend in Colorado River (Fig. 3). It 
consists of three formations, the Barnett shale at the base, the 
Marble Falls limestone in the middle portion, and the Smithwick 
shale at the top. The total thickness in the region of outcrop is 
about 850 to goo feet. 

The Barnett shale, named from springs east of San Saba, is a 
yellowish-gray to black, bituminous shale ranging in thickness up 
to 50 feet, the average being about 30 feet. Its outcrop forms a 
narrow, smooth pathway between the rough, broken terranes of 
limestone on either side, the Ellenburger (Ordovician) below and 
the Marble Falls above. This pathway many of the roads in San 
Saba County follow. Though represented in numerous well rec- 
ords north of the outcrop, the Barnett is not found everywhere 
at the base of the Bend group, as at the type locality of the Marble 
Falls limestone on Colorado River at Marble Falls. Where the 
Barnett is absent the base of the succeeding limestone commonly 
appears to be somewhat conglomeratic, containing débris evidently 
derived from the subjacent Ordovician limestones. The Barnett 
shale is not very fossiliferous, except locally where certain marly 
beds contain numerous fossils. The chief distinguishing features 
of the fauna are the goniatites Gly phioceras cumminsi and G. incisum 
and the brachiopod Liorhynchus carboniferum. ‘The cephalopods, 
which were first described by Hyatt, Smith" referred to the Euro- 
pean species Goniatites striatus and G. crenistria. Girty? included 
G. striatus in the synonymy of G. choctawensis, but it appears that 
the Bend fossils are most probably distinct from these. Liorhyn- 

tJ. P. Smith, U.S. Geol. Surv., Mon. 42 (1903), pp. 68, 80. 

2G. H. Girty, U.S. Geol. Surv., Bull. 439 (1911), p. 97- 


PENNSYLVANIAN STRATIGRAPHY OF TEXAS De 


a 
So 


WY 


sy (da 
ae LY ip 


WY 


Sy 


g eS ae 
GST EGE OREN NN 

ia a ta ca 
Breckenridge BY ~N < NSS 
eae "ip oy Marsy a 
oe NOS Ves NS 

4 , ~ 1 

SON 


>» “AY, So AA, 
CA og 4 
<4 PAuarpersville® 

, Ms: se 


COMANCHE - |: 


PENNSYLVANIAN 


to  NBomervert | | 


SLING 
\ me) 
LEGEND 


CRETACEOUS 
PERMIAN 
ZZ Putnam formation 
CA Moran formation 
Pueblo formation 
(Zz Harpersville formation 
Thrifty formation 
(7 Graham formation 
SSSS9 Caddo Creek formation . 
RA Brad formation 
Graford formation 


1 it} Mineral Wells. formation 


Cisco group 


nal Millsap formation 
‘_|| Strawn Unaifferentiated 
Smithwick shale 


Marble Falls limestone 
==} CAMBRO ORDOVICIAN 


Fic. 3—Map showing outcrops of Pennsylvanian formations in north central 


Texas. 


28 RAYMOND C. MOORE AND FREDERICK B. PLUMMER 


chus carboniferum is reported from the Moorefield shale of northern 
Arkansas, in the Floyd shale of Alabama, and elsewhere in beds 
which are referred to the Upper Mississippian. Accordingly Girty* 
regards the Barnett as late Mississippian in age and therefore 
distinct from the remainder of the Bend which is undoubtedly 
Pennsylvanian. Goldman,’ studying samples of cuttings from two 
wells north of the outcrop, believes that certain lithologic characters 
differentiate the lower shale, and that it is separated by a strati- 
graphic break from the overlying beds. Through the courtesy of 
the United States Geological Survey, Mr. Moore has recently had 
opportunity to examine particularly the contact between the 
Barnett and the succeeding Marble Falls divisions. It has been 
found that although the characteristic Barnett fauna is obtained 
in a limestone which apparently marks the base of the Marble 
Falls, there is really a sharp line of faunal demarcation between 
this and the Marble Falls divisions which is also distinguished by 
a thin zone of glauconite and phosphatic pebbles. The fauna of 
the Barnett is unlike that of the succeeding beds and resembles 
most closely that from formations which are regarded as Upper 
Mississippian. The Barnett shale is therefore tentatively referred 
to the Mississippian rather than the basal Pennsylvanian where it 
was placed in the previous correlation of the writers. ate 

The Marble Falls limestone is a massive, resistant formation 
which is well exposed throughout the region of its outcrop. It is 
somewhat irregularly folded and faulted, and because of the lack of 
continuous exposures or readily identifiable horizons within the 
formation it is difficult to measure an accurate section of the unit 
as a whole. Its total thickness in the region of its outcrop, how- 
ever, appears to be about 400 to 500 feet. In the surface exposures 
no sandstone and practically no shale is observed in the Marble 
Falls formation; but to the north well records show the occurrence 
of some shale interbedded with the limestone and also the presence 
of some limestone in the succeeding Smithwick shale. The Marble 
Falls limestone is but sparsely fossiliferous in some exposures, but 
in parts of San Saba County numerous beautifully preserved fossils 

1G. H. Girty, Bull. Amer. Assoc. Petrol. Geol., Vol. IIL (1919), p. 72. 

2M. I. Goldman, U.S. Geol. Surv., Prof. Paper 129-A (1921). 


PENNSYLVANIAN STRATIGRAPHY OF TEXAS 20 


have been collected. Altogether more than 150 species have been 
identified in studies reported by one of the writers,’ which show 
that in spite of a very considerable proportion of species which 
have not previously been described the fauna is undoubtedly 
representative of the Lower Pennsylvanian. Forms which are 
commonly found only in Mississippian strata, as the coral Paleacis, 
occur in the Marble Falls, but the dominant element of the fauna 
belongs without question to typical Pennsylvanian species. 

The Smithwick shale is typically a fine-grained, black or olive- 
green, rather fissile, bituminous formation which rests conformably 
upon the Marble Falls limestone. It has a total thickness of about 
400 feet. Excellent exposures are found along Colorado River 
near Smithwick, Marble Falls, and Bend, but the outcrop is in most 
places marked by areas of very gentle relief where the shale has 
been covered by débris from weathering. The drainage, as shown 
in the course of San Saba River, appears to have become adjusted 
to the surface geology, the larger streams following the outcrop of 
the easily eroded Smithwick shale. Fossils are few in the black, 
bituminous portion of the Smithwick, but in the gray portion a 
varied fauna has been obtained in which the mollusks are, as might 
be anticipated, the dominant element. The fauna is related to that 
of the Marble Falls formation, but contains a number of forms not 
found in the beds below. 

The Bend group may be correlated with the Morrow group of 
northeastern Oklahoma and northern Arkansas and with the 
horizon of the Wapanucka limestone in southern Oklahoma. Not 
only is there a notable similarity in the faunas of these beds, but 
'a number of species which have been reported only from them 
appear to be in common. However, Pentremites and Archimedes, 
which have been found in both the Morrow and the Wapanucka, 
are not known in the Bend. As has been indicated, the Pennsyl- 
vanian aspect of the fauna is evident. If the Morrow belongs to 
the Upper Pottsville, as indicated by plant fossils studied by 
White,? it would appear that the Bend belongs to a late portion of 


tR. C. Moore, loc. cit. 


2 David White, U.S. Geol. Surv., Nineteenth Ann. Rept. (1898), Part III, p. 469; 
Twentieth Ann. Rept. (1900), Part II, p. 817. 


30 RAYMOND C. MOORE AND FREDERICK B. PLUMMER 


the Pottsville. On account of the unconformity and pronounced 
change in the character and the considerable thickness of the 
succeeding Strawn strata which appear certainly to be not younger 
than Allegheny, the writers are inclined to refer the Bend to a 
somewhat earlier position in the Pottsville, possibly as indicated 
by Ulrich,t the Lower Pottsville. It may be noted that some of 
the Bend fossils, as among the cephalopods Paralegoceras iowense, 
which are reported from other portions of the American Pennsy]l- 
vanian, do not appear to be identified correctly with these species, 
and their significance in correlation is therefore not so important. 


STRAWN GROUP 


The Strawn group includes all the strata between the top of 
the Smithwick shale and the base of the Palo Pinto limestone in 
the Brazos River Valley or its stratigraphic equivalent in the 
Colorado River Valley. The rocks of this group are distinguished 
chiefly by their clastic character, especially the thickness of coarse 
sandstones, and by their irregularity in bedding (Fig. 4). The two 
main areas of Strawn outcrop, one in the valley of Colorado River 
and the other in the valley of the Brazos, are broadly similar, but 
it has not been possible to identify divisions of the one in the other. 
In the northern area there are exposed in the upper part of the 
Millsap division a number of beds of limestone which are found 
nowhere to the south, from which it appears that the waters of the 
Brazos River Valley were farther from the land of Strawn time 
than those of the Colorado. The entire section of the Strawn is 
observable along Colorado River, but in the Brazos Valley a 
considerable thickness of beds belonging to the lower portion of 
the Strawn are not exposed on account of the Cretaceous overlap 
from the east. In both areas there is indication of a very marked 
overlap of the Strawn from east to west, the successively younger 
strata of the group extending farther to the west than the older. 

The Strawn of the Colorado River Valley lies in the area studied 
by Drake.? In his work the alternating subdivisions of sandstone 
and shale are differentiated in some twenty units which he termed 

tE. O. Ulrich, U.S. Geol. Surv., Prof. Paper 24 (1904), p. 111. 

2N. F. Drake, op. cit., pp. 375-89. 


PENNSYLVANIAN STRATIGRAPHY OF TEXAS aie 


Putnam 
formation 


Moran 
formation 


Pueblo 
formation 


CISCO GROUP 


Harpersville 
formation 


Thrifty 
formation 


Graham 
formation 


Caddo Creek 
formation 


=- Home 
Hog Creek sh. 


Brad a) Ss ne 
formation 


Graford 
formation 


LEGEND 
—2=7] Limestone 
Sandstone 
[ae] shale 
conglomerate 
EE Coal 


h sh. 


Seaman Rane 


GROUP 


CANYON 


forma 


East Mountain sh. 


Mineral Wells 


Mingus sh. 


STRAWN GROUP 


Thurber coal 


Millsap formation 


Fic. 4.—Stratigraphic sections through the Pennsylvanian of north central 
Texas. (1) Brownwood-Trickham section; (2) May-Coleman section; (3) Strawn- 


Baird section along Texas & Pacific Railway; 
(5) Jacksboro-Newcastle section. 


(4) Mineral Wells—Moran section; 


32 RAYMOND C. MOORE AND FREDERICK B. PLUMMER 


beds and to which he applied stratigraphic names. The outcrops 
of these beds extend from south to north across the Strawn area, 
the dip being more or less steeply to the west. The sandstones 
form escarpments, but because of variations in the deposits these 
are very irregular and are not readily traceable or easily differ- 
entiated, as are the limestone escarpments higher in the section. 
The thickness of the Strawn in the Colorado River Valley as deter- 
mined from measurements at the outcrop is more than 3,800 feet, 
but the drill shows that the greatest thickness south of Brownwood 
is not more than 1,200 feet. Wells 5 miles west of Brownwood 
show a thickness of the Strawn amounting to 700 feet, and 9 miles 
north of Coleman only 500 feet. Near Brady the Strawn is appar- 
ently not represented, and higher divisions of the Pennsylvanian 
rest directly upon the Bend. 

In the Brazos River Valley two main divisions of the Strawn 
have been identified, the Millsap formation below and the Mineral 
Wells formation above. Only the upper portion of the Millsap 
formation is exposed at the surface, outcrops being found in the 
eastern part of the Strawn area near Millsap and along Brazos 
River in southwestern Parker County. The limestones which 
appear in this part of the section are quite unlike any beds observed 
in the Mineral Wells formation. Cummins‘ defined the Millsap 
division in 1890 to include all the beds in the Brazos River Valley 
below ‘‘coal seam No. 1”’ (Thurber coal) and the top of the black 
Smithwick shale. Thus the lower portion of the Millsap formation 
is known only from drill records. As a whole, the formation con- 
sists mostly of dark blue and clayey shale, limestone, and several 
thin, light-colored, friable sandstones. Locally it contains oil and 
gas in commercial quantities. Well records show its thickness in 
the Strawn area to range from 1,800 to 3,000 feet, and as in the case 
of the Strawn of the Colorado River Valley the thickness diminishes 
to the west: 2,200 feet at Brad in Palo Pinto County; 1,600 feet 
at Caddo in Stephens County; and 800 feet at Breckenridge, 
Stephens County. 

The Mineral Wells formation includes the sandstones and 
shales of the upper part of the Strawn in the Brazos River Valley 

1W. F. Cummins, Texas Geol. Surv., Second Ann. Rept. (1890), p. 372. 


PENNSYLVANIAN STRATIGRAPHY OF TEXAS BR 


above the Thurber coal. It is very well exposed in the vicinity of 
Mineral Wells and along Brazos River, its outcrop extending in a 
belt 10 to 15 miles wide from Erath to Jack and Wise counties. 
Four prominent sandstone members produce prominent escarp- 
ments which are the chief topographic features of the region. The 
shales are sandy and are at least in part very fossiliferous. 

Fossils from the Strawn collected by the writers are chiefly from 
the Millsap and the middle portion of the Mineral Wells formation 
in the Brazos Valley. The fauna of the Millsap, so far as known, is 
not very large, nor does it contain strongly diagnostic elements, but 
it appears to be more closely related to that of the Bend than the 
Mineral Wells. A large and varied fauna is found in the Mineral 
Wells formation, more than go per cent of which is common to the 
Wewoka fauna of southern Oklahoma which has been studied in 
detail by Girty.. Without doubt these faunas are very closely 
related, but since this fauna with some minor changes occurs in the 
Canyon group and in the Graham formation of the Cisco, and since 
in southern Oklahoma it is found as low as the Hartshorn sand- 
stone,” many hundreds of feet below the Wewoka formation, it is 
believed that the stratigraphic equivalent of the upper Strawn in 
southern Oklahoma is below the horizon of the Wewoka. From 
evidence at hand the Strawn of Texas may be correlated with 
beds below the Calvin sandstone of the section northeast of the 
Arbuckle Mountains, with the Vinita and Winslow formations of 
the region farther north, and probably in part the Cherokee shale 
of Kansas and Missouri. It is evidently of Allegheny age. 


CANYON GROUP 


The Canyon group includes the beds formed after the deposition 
of the coarse sandstones, conglomerates, shales, and coal of Strawn 
time, when the land to the east had been worn low, the accumulating 
sediments forming a series of thick limestones and fine calcareous 
clays, with only a few lenses of sandstone. As here defined, the 
Canyon group includes the strata assigned to it by Cummins? in his 

1G. H. Girty, U.S. Geol. Surv., Bull. 544 (1915). 

AG, lel; Guin, WS. Geol. Surv., Nineteenth Ann. Rept. (1898), Part III, p. 541. 

3 W. F. Cummins, Texas Geol. Surv., Second Ann. Rept. (1890), p. 374. 


34. RAYMOND C. MOORE AND FREDERICK B. PLUMMER 


section along Brazos River, that is, from the base of the massive 
limestone on the East Fork of Keechi Creek, now named the Palo 
Pinto limestone, to the top of the massive limestone which outcrops 
near Finis on the line between Jack and Young counties. The 
latter has been found to be equivalent to the Home Creek limestone 
of Drake in the Colorado River Valley. Conditions were more 
uniform in the north Texas region during the Canyon epoch, and 
it is possible to trace many of the stratigraphic subdivisions for 
very long distances. As shown in the table of formations, there 
are four formations in the Canyon of the Brazos Valley, in order 
from the bottom: Palo Pinto, Graford, Brad, and Caddo Creek; 
in the Colorado Valley there are three, the Palo Pinto not being 
represented. The total thickness of the group ranges from an 
average of about 500 feet in the south to 800 or goo feet in the north. 
The outcrop is correspondingly wider in the north. 

The Palo Pinto limestone is a thick, crystalline, dark gray rock 
made up typically of beds 2 to 6 inches in thickness and having a 
total thickness:of 50 to 100 feet. It forms a prominent escarpment 
across Palo Pinto County and has been traced for a long distance in 
the Brazos Valley. It has not, however, been identified south of 
the Cretaceous overlap in Eastland County which separates the 
Pennsylvanian outcrops. The basal beds of the Canyon consist 
here of wave-worked sands, thin, brecciated limestone containing 
pebbles of black limestone, chert, and conglomerate. Evidently 
there was land no great distance farther south, in the region of the 
present Llano Mountains. The chief distinguishing feature of the 
fossils which have been found in the Palo Pinto formation is their 
very robust size, many species being represented by individuals 
more than twice the normal size. | 

The Graford formation, named from the town of Graford in 
northern Palo Pinto County, consists of a thick, locally very 
fossiliferous shale, the Brownwood member, below, and a massive 
escarpment-forming limestone, the Adams Branch member, above. 
These divisions are well developed both in the south, where they 
were differentiated by Drake, and in the north. The Brownwood 
shale is about 160 feet thick near Brownwood, but at least 400 feet 
near Graford. A conglomerate at the base of the formation in the 


PENNSYLVANIAN STRATIGRAPHY OF TEXAS 35 


Colorado Valley has been named the Rochelle conglomerate, and a 
thin lentil of limestone which is traceable for a considerable distance 
in the lower part of the Brownwood in Brown County is named the 
Capps bed. The Adams Branch limestone has a thickness of to 
to 30 feet in the southern Pennsylvanian area, but northward it 
increases locally to more than too feet. A varied fauna has been 
collected from the Graford formation, the most numerous fossils 
coming from the Brownwood member. While it contains some 
species, in part undescribed, not known in the Wewoka fauna, and 
lacks many which occur in the southern Oklahoma formation, it is 
a local modification of this fauna and corresponds to it more closely 
than to any other. <A thin bed about 60 feet below the top of the 
Adams Branch limestone, characterized by an abundance of a very 
elongate, emaciated Fusulina, has been identified throughout the 
extent of this member. 

The Brad formation, named from Brad in Palo Pinto County, 
rests conformably upon the Graford. It consists of limestone and 
shale, a massive, very cherty limestone, the Ranger member, mark- 
ing the top. The Brad formation has been traced from near 
Stewarton in Jack County to Brady in McCullough County, its 
average thickness in the north being about 225 feet, and in the 
south about 200 feet. The lower part of the formation consists of 
shale with thin limestones and some sandstone, and in the north 
all the strata below the Ranger limestone are included in the 
Seaman Ranch member. In the Colorado River Valley, however, 
three members are distinguished below the Ranger: the Cedarton 
shale and sandstone, 20 to 80 feet at the base; the Clear Creek 
‘limestone, to to 25 feet, next above; and the Placid shale, 30 to 50 
feet, beneath the Ranger. The Ranger limestone, which was 
called by Drake the ‘‘Cherty limestone,” is readily distinguishable 
because it is the only uniformly cherty bed in the Canyon. Its 
thickness ranges from to to 50 feet. 

The Caddo Creek formation, named from a tributary of the 
Brazos in Stephens County, has been mapped from eastern Jack 
County southwest and south to a point near Brady. In the north 
its average thickness is 100 to 150 feet, in the south 30 to 50 feet. 
It consists of two members, both named by Drake in the Colorado 


36 RAYMOND C. MOORE AND FREDERICK B. PLUMMER 


Valley, the Hog Creek shale below, and the Home Creek limestone 
above. The shale is rather sandy, and especially in the north 
grades into a thick, cross-bedded sandstone at the top. The lime- 
stone, or rather the series of thin limestones, at the top of the 
Caddo Creek, has a thickness of 10 to 50 feet and in the extreme 
northern portion of the outcrop disappears. 

A review of the paleontologic evidence and the stratigraphic 
relations of the Canyon group leads to the conclusion that this 
portion of the Texas Pennsylvanian is approximately contem- 
poraneous with the upper portion of the section described in the 
Coalgate area* of southern Oklahoma, that is, Calvin to Holden- 
ville, and perhaps some of the subjacent beds. This includes the 
Wewoka formation. Since the horizon of the Calvin has been 
traced northeastward to the Claremore, and this in turn to the 
base of the Marmaton formation in Kansas, it appears that surface- 
mapping checks the paleontological correlation made by Girty? 
and the Wewoka formation may be regarded as equivalent to a 
portion of the Marmaton. ‘The horizon of the Lenapah limestone 
which belongs in the upper part of the Marmaton formation has 
also been traced across most of Oklahoma and appears to lie just 
above the Wewoka. 


CISCO GROUP 


The upper portion of the Texas Pennsylvanian included in the 
Cisco group is characterized by its more clastic sediments, its thin 
but persistent limestones, and the presence of coal. It includes 
all the beds between the Home Creek limestone of the Canyon 
and the lowermost beds containing Permian fossils. The change 
in the character of the rocks in passing from the Canyon to the 
Cisco is evidently the result of a diastrophic movement which made 
shallow the waters in northern Texas and which brought into them 
large amounts of coarse sand and gravel, chiefly from the north, for 
the northern portion of the Cisco is materially thicker and more 
clastic than the southern portion. The total thickness of the Cisco 
group is about 700 to 800 feet in the southern Pennsylvanian area 

tj. A. Taff, U.S. Geol. Surv., Geol. Atlas, Folio 74. 

2G. H. Girty, U.S. Geol. Surv., Bull. 544 (1915). 


PENNSYLVANIAN STRATIGRAPHY OF TEXAS By 


and 1,400 to 1,500 feet in the north. Six formations have been 
recognized in the Cisco, as indicated in the foregoing table of 
stratigraphic divisions, in order from the base: Graham, Thrifty, 
Harpersville, Pueblo, Moran, and Putnam. As a whole, the Cisco 
group is not more fossiliferous than other parts of the Texas Penn- 
sylvanian, but some beds, as the upper shale of the Graham forma- 
tion, are among the most fossiliferous in the mid-continent region. 

The Graham formation, named from the county seat of Young 
County, a place already known to science on account of the unusual 
ammonoids which have been collected there by J. P. Smith™ and 
others, appears to rest disconformably upon the Canyon beds. The 
older or lower members of the Graham are present only in the north, 
pinching out southward and being overlapped by the younger or 
higher members. ‘The formation is distinguished from the under- 
lying beds by its very clastic character and thinner limestones, and 
from succeeding beds by its prolific and characteristic fauna. The 
subdivisions of the Brazos Valley include the Finis shale and sand- 
stone at the base, Jacksboro limestone, Gonzales Creek shale, 
Bunger limestone, South Bend shale, Gunsight limestone, and Way- 
land shale. In the Colorado River Valley there are the Bluff Creek 
shale, Gunsight limestone, and Wayland shale. The limestones 
are in each case very thin and are not so persistent as the limestones 
of the Canyon. The Jacksboro and Gunsight are locally charac- 
terized by an extraordinary abundance of the coral Campophyllum 
torquium. ‘The shales, especially the Wayland shale, contain very 
numerous fossils which for the most part are beautifully preserved 
and weather out readily from the shale. The fossiliferous zone of 
' the Wayland has been traced in some detail from northern Jack 
County to Brown County, a distance of some 175 miles, and at 
every point where suitable exposures were found the characteristic 
fauna was discovered. One locality a mile south of the town of 
Gunsight, in southern Stephens County, contains the most sur- 
prising profusion of well-preserved fossils which has come under 
the observation of the writers. The fauna is very closely related 
to that of the Wewoka formation in Oklahoma, for not only is a 
very large proportion of the species common to both, but a number 

tJ. P. Smith, Joc. cit. 


38 RAYMOND C. MOORE AND FREDERICK B. PLUMMER 


of rare species which may be considered as satisfactory index fossils 
are found in both. One very interesting element in the Graham 
fauna is the group of ammonoids, or, in part, true ammonites, which 
occurs in the Wayland shale. While a number of these have been 
found nowhere outside of the Graham, they are very strongly 
suggestive of the Permian; indeed, so much so that if the associated 
invertebrate fauna and stratigraphic position were less definitely 
certain, the containing rocks would probably be regarded as Per- 
mian. The Graham formation is referred to the Upper Coal 
Measures and is believed to be approximately equivalent in age to 
the Kansas City formation of the Kansas section. There are 
marked similarities in the faunas of these horizons, and each is 
followed by a pronounced change in fauna, marked by the dis- 
appearance of many of the common elements in the previous 
invertebrate groups and by the appearance of new forms such as 
Enteletes hemiplicata, Chonetes granulifer meekianus, and other 
shells, which became widely distributed and abundant in the 
closing portion of the Pennsylvanian. The greater thickness of 
the Graham formation in the north, 500 to 600 feet as compared 
with roo feet or less in the south, and its much more clastic char- 
acter in the north appear to be associated with the uplift of the 
Arbuckle Mountains in southern Oklahoma which independent 
studies indicate was subsequent to the time of the deposition of the 
Wewoka formation and long after the beginning of the Penn- 
sylvanian. . 

The Thrifty formation, named from the town in central Brown 
County, includes the strata from the sandstone which discon- 
formably overlies the Graham formation to the top of the limestone 
which forms a prominent escarpment in the town of Breckenridge, 
Stephens County. The formation consists of thick shales which 
are less fossiliferous and brighter in color than those of the Graham, 
limestones which are thicker and somewhat more massive than 
those of other divisions of the Cisco, and some sandstone and coal. 
It has been mapped from Jermyn in Jack County through Young 
and Stephens counties to the border of the Cretaceous in Eastland 
County. Inliers belonging to the Thrifty appear in the midst of 
the Cretaceous near Romney and Bethel, and it has been traced 


PENNSYLVANIAN STRATIGRAPHY OF TEXAS 39 


south of the sand to the border of the Llano Mountains southwest 
of Brady. In the northern Pennsylvanian area its thickness is 
about 150 to 200 feet, in the southern, 100 to 125 feet. The forma- 
tion has not been subdivided into named members, but three of 
the limestone beds which have been separately mapped for long 
distances have been differentiated. The Breckenridge limestone, 
3 to 5 feet thick, occurs at the top, the Black Ranch limestone, 
about the same in thickness, is found 20 to 4o feet below the top, 
and the Ivan limestone, 50 to 80 feet below the top. The two lower 
beds have not definitely been identified in the Colorado Valley, 
though they are possibly equivalent to Drake’s “Lower Chaffin” 
and ‘“‘Speck Mountain” limestones. 

The Harpersville formation is named from a small town to 
miles south of Breckenridge in Stephens County. It includes the 
strata from the top of the Breckenridge limestone to the top of the 
Saddle Creek limestone of Drake, and is chiefly characterized by 
its content of one of the important workable coals in the north 
Texas Pennsylvanian, “‘Coal No. 6” of Texas reports. The shales 
which are the chief constituent rock are carbonaceous and ferru- 
ginous, containing siderite and limonite concretions, and locally 
abundant, well-preserved fossils. The only important limestone in 
the south is the Saddle Creek member, which consists of two or 
three beds 1 to 3 feet in thickness separated by shale; but in the 
north a second limestone, named the Belknap, 30 to 50 feet below 
the top of the formation, is traceable for a long distance; and in 
the vicinity of Crystal Falls, Stephens County, a third bed, the 
Crystal Falls limestone, occurs 60 to 80 feet lower. The total 
‘thickness of the Harpersville formation is 200 to 275 feet. The 
fauna is a fairly large one, but it is very unlike that of the lower 
portion of the Cisco described in the Graham. The species most 
common there are here missing, and new species, as Enteletes 
hemiplicata, Chonetes granulifer meekianus, Derbya cymbula, and 
other forms, appear. Also species which are present in the lower 
rocks as a rather unimportant faunal element are here very abun- 
dant. The fauna of the Harpersville is indicative of a late portion 
of the Pennsylvanian. It is approximately equivalent to the 
Douglas or Shawnee formations of the Kansas section. 


40 RAYMOND C. MOORE AND FREDERICK B. PLUMMER 


The fourth formation of the Cisco, the Pueblo, is named from a 
town on the Missouri, Kansas & Texas Railway in northeastern 
Callahan County, where the rocks of this division are typically 
exposed. It lies conformably upon the Harpersville formation and 
includes the strata to the top of the persistent Camp. Colorado 
limestone which forms a readily traceable escarpment both in the 
Colorado and Brazos valleys. The thickness of the formation, 
which averages from 150 to 200 feet, is largely made up of gray, 
buff, or reddish shale. The limestone beds at the top are thin, 
3 to 6 feet, but are remarkably uniform and persistent. The fauna 
of the Pueblo is similar to that of the subjacent Harpersville and 
is typical of the uppermost Pennsylvanian. 

The strata above the Pueblo belong to the Moran formation 
which is defined to include the shale and limestone to the top of 
the persistent yellow limestones which outcrop typically in the 
vicinity of Moran, Shackleford County, and in the Moran oil field. 
These limestones, commonly three in number, are closely associated 
and are grouped together as the Sedwick member. The Sedwick 
beds are distinguished by widely distributed and abundant small 
silicified fossil casts, mainly gastropods. The Sedwick may be 
identified with little difficulty by reason of this peculiarity of the 
contained fossils, the lithologic character of the limestones, and the 
stratigraphic position. The average thickness of the formation is 
150 feet in the south and 200 feet or more in the north. 

The uppermost division of the Cisco group is named the Putnam 
formation, from a town in Callahan County. It includes the shale 
and limestone overlying conformably the Moran formation to the 
top of the Coleman Junction limestone of Drake which comprises 
the upper member of the formation. The shale which follows the 
Coleman Junction bed contains fossils which appear to belong to 
the Lower Permian, but those which have been found in the lime- 
stone are not dissimilar from those in the subjacent divisions of the 
Cisco. Asin Kansas, there appears to be here a gradual transition 
from the Pennsylvanian to the Permian, and a line of division is 
more or less arbitrary. The Coleman Junction limestone has been 
selected as a conspicuous, readily traceable line of such division, 
but it is possible that future detailed investigations will show the 
presence of Permian species below this horizon. The thickness of 


PENNSYLVANIAN STRATIGRAPHY OF TEXAS 4l 


the Putnam formation is 125 to 150 feet in the Colorado Valley, 
and about 175 feet in the Brazos Valley. 

The middle and upper portions of the Cisco, that is, the Thrifty 
formation and succeeding divisions, correspond in stratigraphic 
position and faunal character with the upper part of the Kansas 
section included in the Lansing to Wabaunsee formations. The 
Texas divisions cannot be correlated precisely on the basis of 
information at hand with the Kansas formations, but there can be 
no doubt as to their essential equivalence. 


PHYSICAL HISTORY OF THE TEXAS PENNSYLVANIAN 


Sedimentation began in the north Texas area in Pottsville time, 
the sea first advancing into the region probably in the very early 
Pottsville. Black shale and later a thick mass of limestone were 
widely deposited about and over the Llano region and northward 
probably beyond the Red River. The sea of this time was evi- 
dently an invasion from the southwest, and the faunas represent 
the earliest typical marine Pennsylvanian of the continent. The 
epoch closed with more deposition of carbonaceous muds. 

At the end of Bend time the sea withdrew temporarily from the 
Llano region and probably from most of the north Texas area. 
However, there was little erosion of the soft Smithwick shale, and 
the unconformity which is observed at the base of the Strawn is 
defined mainly by the difference in strike of the younger beds, the 
conglomerates and very remarkable change in the character of the 
sediments, and the overlap of the Strawn upon the Bend. The 
ancient land to the east representing the northeast part of Columbia, 
' or,/as it has been termed recently, Llanoria,* was notably uplifted 
and furnished a very great amount of clastic débris to the shallow 
sea lying to the west. This was spread out in irregular lenses with 
a general delta structure, dipping westward and younger deposits 
overlapping the older. Coal was deposited locally. The fauna had 
undergone a marked change from that of Bend time and at least in 
the upper part of the Strawn was very closely related to the southern 
Oklahoma Wewoka fauna. This portion, and perhaps all of: the 
Strawn, is referable to Allegheny time. 


tH. D. Miser, paper presented before the Geological Society of America, Decem- 
ber, 1920. 


42 RAYMOND C. MOORE AND FREDERICK B. PLUMMER 


Clearer waters, in which calcareous muds and limestones were 
the chief accumulating sediments, succeeded, the Canyon group 
representing the second epoch of dominant limestone formation in 
the Texas Pennsylvanian. Some 700 or 800 feet of limestones and 
shales were deposited during this relatively quiet interval, the 
invertebrate faunas showing only changes which might be antici- 
pated with undisturbed, clear waters. It is probable that the 
change to dominant limestone deposition in Texas was approxi- 
mately synchronous with the similar change in the northern part of 
the mid-continent region above the Cherokee shale. 

The beginning of Cisco time is marked by a change in the 
character of the sediments, a great amount of mud, sand, and 
gravel being swept from the north and northeast, where the orogenic 
movement which resulted in the formation of the Arbuckle Moun- 
tains and probably an uplift of the ancient land of Llanoria fur- 
nished an abundance of materials. The thick, relatively soft, 
clastic formations which compose the Lower Pennsylvanian in the 
Arbuckle region would first be exposed to erosion and probably 
furnished a large proportion of the sediment in the Graham forma- 
tion of north Texas. When the Lower Paleozoic limestones of the 
mountain areas were exposed, limestone conglomerates were formed 
locally (Franks) which were strewn across the eroded edges of the 
earlier Pennsylvanian. Although there is an important uncon- 
formity in this part of the Pennsylvanian of southern Oklahoma, 
the seas do not appear to have receded more than temporarily from 
north Texas, for the unconformity at the top of the Graham forma- 
tion is not a large one. However, the faunas underwent a striking 
change and throughout the remainder of the Cisco have a very 
evident late Pennsylvanian aspect. The middle and upper Cisco 
represents a continuation of mud and sand deposition with an 
occasional limestone bed and one or two important intervals when 
coal was formed. There does not appear to have been any marked 
change with the transition to Permian time in part of the north 
Texas region, but to the northwest the presence of continental 
Red. Beds sediments characteristic of the Permian appears. In 
these are found the well-known reptiles and amphibians which lived 
on the early Permian lands and along the old shores in the north 
Texas region. 


PLEISTOCENE MOLLUSCA FROM NORTHWESTERN 
AND CENTRAL ILLINOIS" 


FRANK COLLINS BAKER 
Museum of Natural History, University of Illinois 


Two very interesting and important collections have recently 
been submitted for study by members of the Illinois State Geo- 
logical Survey. ‘They include several new forms, besides a number 
of species not before recorded from Illinois Pleistocene deposits, 
though several of these have been known from Iowa glacial deposits 
for many years. 

In the volume on the Life of the Pleistocene (Baker, 19200, p. 368) 
the author listed as extinct six species of mollusks. That so small 
a number of species in so large a group should have become extinct is 
noteworthy, but is paralleled by the plants of which seven species 
are extinct. It is believed by the writer that careful analysis of 
Pleistocene Mollusca will show that more species or varieties have 
become extinct than is indicated by present records. Many 
species have been linked with existing forms though they vary 
quite enough to constitute distinct species or races. It will be 
found that some species are abundant in Pleistocene deposits, but 
rare living, indicating that the species is approaching extinction. 

In the study of Pleistocene fossils it is of the greatest importance 
that minute differences between species be recorded, for it fre- 
quently happens that two deposits contain species closely related 
but yet sufficiently different to indicate a difference in climate, 
habitat, or general environment. A case in point is Succinea avara 
and Succinea vermeta, two closely related species or races, the latter 
of which is abundant in loess deposits and other Pleistocene beds. 
The so-called “‘lumper”’ of species can do great harm to the study 
of fossil faunas, whereas the so-called “‘splitter,”’ recognizing small 
differences, may be of real help in the discrimination of faunas and 


. Contribution from the Museum of Natural History, University of Illinois, No. 20. 


43 


44 FRANK COLLINS BAKER 


their relation to each other. Of the two evils, splitting is certainly 
the less. In the study of the material from these deposits this 
fact has been kept in mind and careful comparisons have been 
made with Recent and with other fossil species. 


MATERIAL FROM THE LOESS AND OTHER DEPOSITS 
IN NORTHWESTERN ILLINOIS 


This material was collected by Dr. Morris M. Leighton, geolo- 
gist in charge of Pleistocene investigations for the Illinois State 
Geological Survey, during the summer of 1920. Dr. Leighton has 
worked out the stratigraphy of these deposits and his papers appear 
elsewhere. The localities and sections from which the material 
came are indicated below, the information being contributed. by 
Dr. Leighton. The molluscan species contained in these strata 
are listed as to species and abundance. The writer has added 
certain notes to Dr. Leighton’s data, suggested by the zodlogical 
content of the strata. More specific information concerning the 
different species will be found in a later part of this paper. 


STATION NO. I 


Locality: Bluff along Farm Creek, Tazewell County, about 7 
miles east of Peoria, just east of the Toledo, Peoria & Western 
Railway bridge, near the south line of SE. 4, Sec. 30, T. 26 N., R. 
2 W. 

Material: From the deposit of loess underlying the Shelbyville 
till, and overlying an older leached loesslike clay and the weathered 
Tlinoian till. 

Stratigraphic horizon: Early Peorian loess. 


MOLLUSCAN LIFE 
Helicina occulta, common. Succinea ovalis, rare, immature. 


Oreohelix towensis, broken pieces. 


All are typical Peorian loess mollusks. This is near the type 
locality of Leverett’s Peorian interglacial stage,’ and these are the 
first mollusks to be specifically identified from this type section. 


t Leverett, ‘“The Illinois Glacial Lobe,” U.S. Geol. Surv. Mon. 38 (1809), p. 187. 


PLEISTOCENE MOLLUSCA FROM ILLINOIS 45 


STATION NO. 2 


Locality: Road cut 2 mile northwest of Winslow, Stephenson 
County, just southeast of fork in roads. 

Material: Yellow calcareous sandy loess from base of cut. 

Stratigraphical horizon: Probably early Peorian. 


MOLLUSCAN LIFE 
Succinea vermeta, a common loess fossil, usually listed as avara. 


STATION NOS. 3 AND II 


Locality: Stephenson County, SW. 4 of SW G, Sec. 28, T. 27 N., 
R. 9 E., about } mile northeast of Pecatonica River bridge, bank 
along roadside 12 to 15 feet high, mostly grass- and shrub-covered; 
1 mile north and $ mile east of Ridott. 

Material: Exposure at one place shows yellow, loesslike clay, 
calcareous, about 5 feet below top. ‘Topographically a terrace 
resembling the silt terrace of the Pecatonica farther downstream 
and in tributary $ mile east. No show of bedding or lamination; 
possibly obliterated by slumping. 

Stratigraphic horizon: Probably Wisconsin, in back water from 
Rock River during the deposition of the Rock River Valley train. 


MOLLUSCAN LIFE 


Pomatiopsis lapidaria, very abundant. Galba obrussa, seven broken shells. 
Helicodiscus parallelus, common. 


The writer is of the opinion that this deposit may have been 
partly formed by wind. Two of the species are semiaquatic. 
Pomatiopsis now lives in very small streams, usually not exceeding 
2 feet in width and a few inches in depth. Galba obrussa is semi- 
aquatic (or amphibious) and spends as much time out of the water 
as in it. Its habitat is in such streams as above described, or on 
muddy shores bordering rivers. It does not (nor does Pomatiopsis) 
live in large bodies of water such as lakes or rivers. The writer 
has seen several places in Illinois, shallow rivulets in woodlands 
bordering streams, inhabited by these species. One such is near 
Mahomet on the Sangamon River. WHelicodiscus is a typical land 
mollusk. The character of the deposit (loesslike), the contained 


46 FRANK COLLINS BAKER 


calcareous concretions, and the habitat of the contained life all 
point to the possibility of its being low-land loess. 


STATION NO. 4 


Locality: Illinois Central Railway cut, Boone County, about 
i mile northwest of depot at Irene, northeast side of cut, NW. 3, 
SE, AO, Wo aig Ney IS, @ Jee 

Material: Where fossils were collected the drift section was as 
follows: 


Feet 
62. Soildarkeloessialeesparse pe Db lesmear seer ae eee eee About 23 
cy ull leached darks buthy clays land 2s ye ernest 13 


4. Calcareous till, yellow on dry surface, grayish yellow with pinkish cast 
on damp surface, limestone pebbles; dark tillin lower 6feet About 15 
3. Stratified, yellow, calcareous sand, no pebbles; numerous pelecypods 


an@rother mollusks stress very. eerie cuca as oe eee About 5 
2. Blue-gray calcareous silt with numerous mollusks; exposed......... 2 
Tn Bottomvol Cui uncoVercdn see eee wad stale ee 5 


Stratigraphic horizon of Nos. 2 and 3: Post-Illinoian and pre-early 
Wisconsin, possibly Sangamon. 
MOLLUSCAN LIFE 
Fossils were obtained from Nos. 2 and 3 as noted: 


No. 3 No. 2 
Fi SES Jd ae EAR AE Musculium rhomboideum, 1, fragments 
Pisidium costatum, common. Pisidium costatum, abundant. 
Planorbis altissimus, common. : : , s : ; : 
Galba palustris, common. Galba palustris, rare, immature. 


Succinea vermeta, rare. 


The blue-gray silt, No. 2, indicates a quieter body of water 
than the yellow sand, No. 3, and the mollusks bear out this inter- 
pretation. 

STATION NO. 5 
Locality: Ten rods southeast of Station No. 4, northeast side 


of cut. 
Material: The cut shows the following section: 


Feet 
Tt. Soil, dank, loessial, Sparsespebbless cant see see i) een 25 
ro, Weached till; dank butt. clayeyeuiarcl nese sss ieee eee aS 


9. Calcareous till, yellowish when dry, grayish yellow with pink tinge 
when damp, contams) limestone pebbleseriys jenn ee 7 


PLEISTOCENE MOLLUSCA FROM ILLINOIS 47 


Hee 
8. Fine sand, no pebbles, yellow, highly calcareous.................. a 
7. Calcareous till, banded, yellow and grayish, limestone pebbles, 
COMP ACH itr SHU SL aa alee gn sag doe le a eS 
6. Blue, gummy clay, nearly gritless in some places and pebbles up to 
TminchnmMothersy highly icalcareousme sas.) 2. ee a alae 3-3 
5. Fossiliferous loess, grayish to yellowish, rusty streaks near base, 
calcareous throughout, sand streaks, cross-bedded, dip southerly.. 5 
4. Stratified yellow sand and gray silt with few pebbles, few fossils, dis- 
COI alRCal CARCOUS seme conga cists) esc eep aise Gao aereumcieuaae ns Ha senna eee S 1s 
3. Light blue silt, pebbly, some limestone pebbles, highly calcareous; 
TERODT LOS THM ao ol Sea Rly eae Re Eee sess ey ote ets: fee eee oT I 
2. Sandy to gravelly till, yellowish to rusty, limestone pebbles, matrix 
Call CARCOUS EE PN enna No Uc an cca. MRA Rn ceuabln ea ees 2 
1. Probably till, gray with pinkish tinge, calcareous; steam shovel has 
worked stratum giving steeper slope than above section.......... 7 


Nos. 6 to to include the Wisconsin drift sheets; Nos. 1 to 3 
represent Illinoian till; Nos. 4 and 5 indicate an interglacial 
interval between the deposition of these tills. The sand and silt 
(No. 4) may represent the Sangamon, and the loess (No. 5) the 
Peorian, although there appears to be no break between them. 
No. 4 has yielded two species of fresh water mollusks; Planorbis 
altissimus and Calba palustris, represented by but a few specimens. 


STATION NO. 6 


Locality: Same as Station No. 5. 

Material: Taken from the fossiliferous loess No. 5 of the previ- 
ous section. Separate collections were made from both the yellow 
and gray loesses, the mollusks in both being of the same species. 

Stratigraphic horizon: Possibly Peorian loess. 


MOLLUSCAN LIFE 


Succinea vermeta, the common Succinea of the loess, occurred 
somewhat more abundantly in the yellow than in the gray loess. 
Mr. B. B. Cox, a student in the University of Chicago, collected 
mollusks from the loess and other deposits, though he did not dis- 
criminate betweeen the deposits. He found in addition to the 
above Sphyradium edentulum alticola, a typical Peorian loess fossil, 
represented by one specimen. 


48 FRANK COLLINS BAKER 


STATION NO. 7 


Locality: Whiteside County, SE. ¢ of SE. 4, Sec. 15, T. 20 N., 
R. 3 E., stream-cut bank about 50 feet northwest of small bridge. 

Material: Where the fossils were collected the following section 
was exposed: 


6 Solid arks.:\ic Siee ene ities os ets sult <ieole ose a ene eee 3-13 
4. leoesslike clayaidinty but tol butt, leached/arnns arene ene eee 1-24 
3. Weathered zone, brownish, separates upper from lower material..... t 
2. Loesslike clay, yellow with rusty spots, no structure............... 5 

1. Fossiliferous clay, blue silts, dense, sticky, calcareous... ........... 13 


Stratigraphic horizon of No. 1: Probably Wisconsin in age; per- 
haps deposited in backwater of tributaries to the Mississippi 
River during deposition of the Wisconsin Valley train in the 
Mississippi Valley. 

MOLLUSCAN LIFE 


Six species of fresh mollusks (one a bivalve) and one species of 
land mollusks were found in the fossiliferous bed, as noted: 


Fresh Water Land 
Pisidium costatum, rare. Succinea vermeta, common. 


Valvata sincera, abundant. 
Planorbis altissimus, rare. 
Galba palustris, rare, immature. 
Galba dalli, common. 


The Succinea was doubtless washed into the stream from the 
land. The near relative of vermeta, Succinea avara, is frequently 
found on vegetation bordering streams or lakes. 

In the section it would be interesting to know just what the 
weathered zone, stratum No. 3, indicates. It possibly indicates 
Recent wash from the bordering hill slopes since man has disturbed 
the soil conditions by removing the forests. 


STATION NO. 8 


Locality: Carroll County, NEY of NEP 3, Sec: (6) ize 
R. 5 E., south bluff of creek valley southeast of schoolhouse. 
The bluff is 30 feet high and occurs in the end of a terrace spur at 
the mouth of a tributary valley. 


PLEISTOCENE MOLLUSCA FROM ILLINOIS 49 


Material: The bluff shows the following section: 


Feet 
Paollkandsanctilathoat ee acide. soils ae etek ewes ahees : ie; 
4. Silt, structure mostly obliterated, leached, conforms to the present 
SUD DGs 0-6 8 ip cha inc ee atest ere a eae ries Ree A a I- 3 
3. Sands and silts, laminated and thin-bedded, minor lens and pocket 
structure; clay yellow and brown; red granite bowlder 1 foot in 
diameter occurs about to feet above creek bottom on the north 
side near the base of this horizons "maximum. .........+.0.4..- 22 
2. Blue silt, fossiliferous, some brown sand......................... 6- 8 
1. Coarse, angular gravel, discontinuous in the middle portion exposed 
(Gibovemuaterleviel) mer csa nn lacie yee eee an een Bs o- 23 


Stratigraphic horizon: Nos. 2, 3, and 4 are believed to be of 
Wisconsin age, formed in backwater from the Mississippi, but there 
is reason to assign it to Peorian age or earlier. 


MOLLUSCAN LIFE 


Mollusks were collected from both the silts and the sands, as 
follows: 


Silts Sands 
Succinea vermeta, abundant. Succinea vermeta, abundant. 
Vertigo modesta, rare, 1 spec. : 3 : , ; : 5 5 
Pyramidula shimekit, rare, 3 spec. Pyramidula shimekii, rare, 1 spec. 


Galba parva, not common. 


The land shells (including all except Galba) are characteristic 
of the Peorian loess. Their presence (especially Pyramidula 
_ shimekii) in this deposit beneath nearly 25 feet of other material 
would indicate an earlier age than the Wisconsin. The till in 
Carroll County is of Illinoian age, and it seems probable that the 
_ material above the till, especially Nos. 3 to 5 of the section, while 
deposited by the Wisconsin waters, really covered up life-remains 
that lived during the Peorian or Sangamon interglacial period. 

In the sands and silts of stratum No. 3 (Station No. 9) three 
species were found, Succinea vermeta, Vertigo modesta, Galba parva, 
the Succinea being the only abundant species, the others very rare. 


STATION NO. 10 


Locality: Same as Station No. 8. 
Material: Creek bank 5 feet high on south side of meander 
curve shows the following section: 


50 FRANK COLLINS BAKER 


Feet 
ANGray sit itmod enn)! Acck < het cen cae te cee 92 See I 
34 Black, peaty soil, tossiliterous. © ae ae. ioe eee ee Iz 
25, Gray Silt: COMPACL.<c iwi e sree OR ne ee ec ee 1 
r. Fine gravel, cross-bedded to water level.)..72-20..-.s45-54 eee I 


Stratigraphic horizon of No. 3: Wabash stage (pre-Recent). 


MOLLUSCAN LIFE 
Ten species of both land and fresh-water mollusca occurred in 
this peat deposit, as noted: 


Water Land 
Pisidium, species, a few odd valves. Carychium exile, rare. 
Physa gyrina, rare. Vitrea rhoadst, rare. 
A plexa hypnorum, rare. Gastrocopta tappaniana, common. 
Planorbis urbanensis, not common. Strobilops virgo, rare. 
Galba dalli, common. Euconulus fulvus, rare. 


This is a typical swamp or small-pond fauna such as might be 
found in any quiet pond in Illinois. Such small bodies of water 
are common in old flood plains. The presence of Planorbis urban- 
ensis is noteworthy, the species being hitherto known only from 
the original locality at Urbana. 


STATION NO. II 
Same as Station No. 3. 
STATION NO. 12 


Locality: Stephenson County, Sec. 22, T. 29 N., R. 6 E., about 
too feet southeast of the fork of the roads + mile northwest of 
Winslow, east side of road; roadcut 7 feet deep, all grassed except 
13 feet at base. : 

Material: Probably entirely loess. Base shows loess inter- 
stratified with sand: yellow, fossiliferous. . 

Stratigraphic horizon: Not precisely known, although post- 
Tllinoian. 


MOLLUSCAN LIFE 


The mollusk Succinea vermeta appears sparingly in the base of 
the section. 


PLEISTOCENE MOLLUSCA FROM ILLINOIS 51 


STATION NO. 13 


Locality: Whiteside County, west line of NW. 4 of SE. 4, Sec. 15, 
T. 20 N., R. 3 E., road cut on upland spur. 
Material: Shown in the following section: — 


Stratigraphic horizon: Post-Ilinoian, possibly Peorian. 


MOLLUSCAN LIFE 


Seven species of land mollusks typical of the Iowan loess 
occurred in stratum No. 1, as noted: 


Vallonia gracilicosta, common. Gastrocopta armifera, common. 

Vertigo modesta, rare (previously Succinea avara, common. 
known as Pupilla blandt). Pyramidula shimekit, rare. 

Pupilla muscorum, rare. Euconulus fulvus, rare. 


DISCUSSION OF SPECIES REPRESENTED IN THE DEPOSITS 


Thirty species are represented in the material collected by Dr. 
Leighton, three of which have not before been definitely reported 
from Pleistocene deposits in the state, and three others have been 
reported but rarely and from deposits not clearly defined strati- 
graphically. These are: 


New Records Rare Records 
_Gastrocopta tappaniana Vertigo modesta. 
Vallonia gracilicosta. Pyramidula shimekit. 
Galba dalli. Oreohelix towensis. 


These collections indicate that the fauna of the loess and other 
deposits common in Lowa also occur in Illinois, at least in the north- 
western part, and that the deposits may be rather definitely 
correlated with those of Iowa. 

The thirty species have the following distribution as to geologic 
age: 


52 FRANK COLLINS BAKER 


TABLE I 


DISTRIBUTION OF SPECIES ACCORDING TO GEOLOGIC AGE 


Species Sangamon Peorian Wabash 


Bivalves 
Musculium rhomboideum........0.% eee eens > mee neem a (St Soi oo cc 
IEUSTAUIUIIGOSTOLU TID eee eee x x 
PASUdtuUm'S PECles Na cs oricrere. Pines ee eps ee ae eee Ce eee eee Xx 
Fresh-water gastropods 
‘POMALOPSIS LA PLIAOT1G. oe ote ca tents cio ee OE EO ORE Cee x 
VidlGia SinGer arises «xo 2 oo eee tee tok Cer ha one ne Ce eLe oe x 
A PleRO UY PNONUN ns a Biases wien eR EEO Ol ea era | OOO x 
PRYSOsCYTINE i, che mags areas Tks Ceo ee eo ERO ee ne Ge eee x 
x 
x 
x 
x 
x 


eee eee ewes 


Planorbts alitssumus + 2. see oan celeste 


Galbaspalusirisae nee eee Lee een > Cae eee oer orc 
Galbavobrussa. . ste sks ce eee Oe ROE | ee 
COC TLT TCL ee eR es Oli Oi Stats Pak geen Cr re eel ORS eR O CoG IABTANS 6 Bide oc 
Galbarparues foi. hoe kee NO ee aoe aCe eee XX - Weeseeeeee 
Land gastropods 

Carychium: exile...) soo eee OR Oe ee ce ee AG ee Oe x 

EL eliCinGrOcculla:.c05 chussteee Ae BO Roe een CE oe 
VWallonvanoracilicos (Omer eee Cee 
SUECINEDOUAIIS 5 Fee Re Oona Cn eee 


sete ee tees 
Ce er ry 
eee ee tee ae 


Vertigowmodestasminan erences cers tee oes hie ee eee 

RU PUlOrnUs COnUM ee er ree re eo) Oe ee 
Gasinocopiataninticn de eee eer Cee eee eee ere 
Gusivocoplatappantand saan eee eee eee te | eee eee x 
WENODLIOP SPUN Z OREN tan etek arate oa peel ee ined | ere x 
Sphyradium edentulum alticola...............)e0e.-000:- Xx) 3 ee 
JHGUIGRISENS HHORUNGAIS oo oboe ooonbauAcoepanblooccusvocolloooccoccn< x 

IPS KOMIUGUIGES HUINER ities erie ere eee | roe eee x 
Orecheliniwowensisen vo ee eee ee Ce On eee X's 
TL UCONUMIS AFULDUS He ci cncrs Veo Cate cee heater aoe Ree x x 
WVGEK CG RHOGAS Bre cusses rice stairs SOOO ee RE OE | Ger eee Xx 


a a ir 


FRESH-WATER BIVALVES (PELECYPODS) 


Musculum rhomboideum (Say). A single valve of this fragile 
species was found in a deposit of blue-gray silt of post-Lllinoian and 
pre-early Wisconsin age (Sangamon?) near Irene, Boone County. 
It is apparently rare in Pleistocene deposits. 

Pisidium costatum Sterki. This small bivalve was selected 
from three deposits at two localities; in yellow sand and blue-gray 
silt near Irene, Boone County (Sangamon age), and in the blue 
silty clay in Whiteside County (Wabash age). It is abundant 
near Irene, but rare in Whiteside County. Costatwm is known only 
from Pleistocene deposits, not yet having been detected among 


PLEISTOCENE MOLLUSCA FROM ILLINOIS 53 


living Pisidia. It is widely distributed and has been recorded 
from Aroostook County, Maine (Baker, 19200, p. 155), Michigan 
(Sterki, 1916, p. 466), and Urbana, Illinois (Baker, 1918, p. 663). 
These records are all Wabash (post-Wisconsin). The present 
records carry the species back into interglacial time. 

Pisidium, species indeterminate. A few odd valves of a small 
Pisidium occurred in peat material from a flood-plain pond in 
Carroll County, of Wabash age (post-Wisconsin). 


FRESH-WATER SNAILS (GASTROPODS) 


Pomatiopsis lapidaria (Say). A mile north of Ridott, Stephen- 
son County, in loesslike silt, this species is abundant in company 
with Galba obrussa and Helicodiscus parallelus. The species does 
not differ in any way from living forms. 

Valvata sincera (Say). Abundant in blue silts in Whiteside 
County, apparently of Wabash age. These individuals are quite 
typical in everything except size, the largest specimens being 
almost 1 mm. less in diameter than the largest living sincera from 
Lake Winnipeg. The two measurements are: 

ALTITUDE DIAMETER 


Recent species from Lake Winnipeg............... dio '5 Se 
Fossil species from Carroll County............... B35 4. 


5 
5 
Sincera from Urbana deposits (Baker, 1918, p. 663) are like 
the Carroll County forms in size. If this disparity in size should 
prove universal, it might be of advantage to differentiate the latter 
as a fossil race. Not enough material of either the Recent or the 
‘fossil sincera is at hand to settle this question definitely. Like the 
Recent forms, the fossil specimens vary somewhat in the height of 
the spire and in the sculpture of the whorls, there being a tendency 
in a few individuals to form thin, sharp ribs, as in the variety 
nylanderit. This feature is more marked in the Urbana szncera 
than in the specimens from Carroll County. 


PULMONATE GASTROPODS 


Physa gyrina Say. Rare in the Carroll County deposit, the 
specimen found being also immature. 


54 FRANK COLLINS BAKER 


Aplexa hypnorum (Linn.). A few very young specimens, 
3-4 mm. in length, occurred with the other species in the Carroll 
County peat deposit. 

Planorbis altissimus Baker. This small Planorbis occurs in 
sand and silt deposits near Irene, Boone County, in strata believed 
to be of Sangamon age, and in silt in Whiteside County, of Wabash 
age. It was most abundant in the Irene deposits. Altissimus 
is the common small Planorbis of all Pleistocene deposits. As a 
fossil it has recently been recognized in deposits from Maine, 
Michigan, Wisconsin, Indiana, Illinois, New Jersey, and Canada 
(Ontario). It is the species listed as parvus in the writer’s Life of 
the Pleistocene (in most cases, although true parvis does occur in 
Pleistocene deposits) and in most references to glacial fossils. It 
was thought to be extinct, but recently Miss Mina L. Winslow, of 
the Museum of Zodlogy, University of Michigan, collected a large 
number of a small Planorbis in Devil’s Lake and other water bodies 
in North Dakota which are apparently the same species. Condi- 
tions in these lakes are becoming severe, due chiefly to increase of 
alkalinity, and the species appears to be dying out in the region. 

Planorbis urbanensis Baker. Six specimens of a small Planorbis, 
first described from deposits in Urbana (Baker, 1918, p. 664; 
IQIQ, p. 94), occurred in the peat deposit in Carroll County. It 
does not differ from the Urbana individuals and its presence in a 
distant part of the state indicates a rather wide distribution. It 
has probably hitherto been listed under the all-embracing name of 
parvus. Known only from fossil strata at present. 

Galba palustris (Miller). Specimens of this protean species 
were found in deposits near Irene, probably of Sangamon age. 
These do not differ from living examples of the species. Young 
individuals, 4 and 6.5 mm. in length, occurred in Wabash deposits 
in Whiteside County. 

Galba obrussa (Say). Broken specimens of this species occurred 
in post-Wisconsin deposits near Ridott, Stephenson County. 

Galba dalli (Baker). This, the smallest of the lymnaeids, was 
common in the peat deposit in Carroll County, and in silt deposits 
in Whiteside County. The specimens are somewhat larger than 
the types of the Recent individuals from Indiana and there is some 


PLEISTOCENE MOLLUSCA FROM ILLINOIS 55 


variation in the height of the spire and the width of the shell. 
This is the first record for the species in the Pleistocene deposits of 
Illinois, though it is common in the Recent fauna. 

Calba parva (Lea). Three lots of a small lymnaeid apparently 
referable to parva were collected by Dr. Leighton, in brown sand 
and yellow sand, Carroll County, and in silt, Whiteside County, 
the latter in post-Wisconsin deposits (Wabash age). In one lot 
from yellow sand (see Station No. 8) the shell is very wide and con- 
vex on the body whorl. This shell resembles Wollf’s figure of his 
tazewelliana, described from deposits in Tazewell County (Wolf, 
1870, p. 198, Pl. XVII, Fig. 2). This form is much more obese in 
the body whorl than are individuals of this species from the Recent 
fauna. Specimens from silt in the same section have a much 
narrower and more compressed body whorl, and the columella is 
slightly impressed. A single adult individual from Whiteside 
County (silt deposit) has the columella impressed so-as to form a 
slight plait. The material at hand is not sufficient to separate these 
forms satisfactorily, or to indicate whether they are merely local 
sports or larger variations. On the whole, if these are merely 
individual variations, the parva of the late Pleistocene is much 
more variable than its living representative. 


LAND GASTROPODS 


Helicina occulta Say. A number of individuals of this species 
occurred in Peorian loess near Peoria. It is not a pulmonate land 
snail. 

Carychium exile H. C. Lea. A single individual of this small 
snail occurred in the peat deposit in Carroll County. It is quite 
typical. 

Vallonia gracilicosta Reinhard. Seven specimens of a Vallonia 
occurred in the loess deposit of Whiteside County that are referable 
to gracilicosta. They exactly conform to the figures by Sterki (1892, 
p. 256, Pl. XXXIII, Figs. 48, 49) and they agree with his descrip- 
tion, having the fine, distinct ribs characteristic of this species, 
which are finer and differently spaced than those of costata. Tracili- 
costa, according to Shimek, is a common loess fossil in Iowa. It 
has not previously been positively identified from Illinois deposits, 


56 FRANK COLLINS BAKER 


although it probably occurs and has been listed as pulchella or 
costata. McGee’s reference to pulchella from Fulton, Whiteside 
County (Pleistocene History of Northeastern Iowa, p. 448) may be 
this species, as it looks like that species without the aid of a powerful 
magnifier (Baker, 19200, p. 351). It probably occurs widely 
distributed in northwestern Illinois. Living gracilicosta are 
known only from the West and Canada. Its abundance in the 
loess indicates a former greater southward extension in distribution. 

Succinea ovalis Say. Fragments of a large Succinea from the 
Peorian loess near Peoria are believed to be this species. 

Succinea avara Say. Several specimens of a species referable 
to true avara occurred in the loess of Whiteside County. 

Succinea vermeta Say. The great majority of Succineae col- 
lected by Dr. Leighton are referable to Say’s vermeta, which appears 
distinct from his avara, the spire being longer, the sutures deeper, 
the whorls rounder, and the aperture roundly ovate instead of 
long ovate. ‘The shells referred to vermeta vary among themselves, 
but all are easily separable from typical avara. Avara as recorded 
from Pleistocene deposits also includes vermeta, the two forms not 
being differentiated. The two forms are said to intergrade com- 
pletely in the Recent fauna, but this does not seem to be true of 
the Pleistocene fauna, at least as shown by the material examined. 
The localities represented in the collections of Dr. Leighton are: 
Sangamon sand, Irene, Boone County. 

Early Peorian loess, Winslow, Whiteside County. 
Peorian loess, Stevenson County. 


Wabash sand, Carroll County. 
Wabash silt, Whiteside County. 


Vertigo modesta Say. This small land shell (of which the 
Pupilla blandi Morse of the Iowa deposits is a synonym) occurred 
sparingly in three places, in silt and sand, Carroll County, and in 
loess, Whiteside County. The individuals from silt and sand 
deposits are apparently typical with four teeth in the aperture; 
but the loess specimen from Whiteside County is different from any 
form described. There are a columella tooth and two palatal 
teeth, but no parietal tooth. Pilsbry (1919, p. 128) describes a 
toothless and a tridentate form of modesta from Norton Sound, 


PLEISTOCENE MOLLUSCA FROM ILLINOIS 57 


Alaska, and the Whiteside County specimen adds another varia- 
tion. Not enough material is at hand to ascertain whether this 
variation is anything more than local. Modesta (under the name 
of Pupilla blandi) is a common loess fossil in Iowa. In Illinois 
it has been reported from the Yarmouth stage (Baker, 109206, 
p. 271), and from deposits in the driftless area (Baker, 1920), p. 353). 

Pupilla muscorum (Linn.). Two specimens from the loess of 
Whiteside County, apparently typical. 

Gasirocopta armifera (Say). Common in loess of Whiteside 
County, and quite typical. 

Gastrocopta tappaniana (C. B. Adams). Abundant in the peat 
deposit of Carroll County. 

Strobilops virgo (Pilsbry). Two specimens of this small land 
shell occurred in the peat deposit in Carroll County. They are 
typical. 

Sphyradium edentulum alticola (Ingersoll). A single specimen 
of this small species occurred in the calcareous loess in Boone 
County, collected by Mr. B. B. Cox. This species is reported 
from but two other places in Illinois. These are as follows (data 
from Baker, 19200): Aftonian, well boring near Rock Island, 
cited as Pupa alticola (p. 240); Peorian or Wabash loess, near 
Galena (p. 353). Its small size has probably caused it to be over- 
looked in the examination of loess deposits from Illinois. 

Helicodiscus parallelus (Say). Common in yellow sand of 
_ Wabash age near Ridott, Stephenson County. All of the speci- 
mens are typical. 

Pyramidula shimekii (Pilsbry). This characteristic land mol- 
' lusk was found in two localities, both of Peorian age, loess of 
Whiteside County and sands of Carroll County. But one speci- 
men was collected in each place. Pyramidula shimeki is an 
abundant fossil in the loess of Iowa and may be said to be char- 
acteristic of the Iowan or Peorian loess. No authentic records are 
known from strata later than the Peorian. While it is common 
in Iowa, it is rare in Illinois, and records are known from but 
one locality other than those listed (Galena in the driftless area, 
Trowbridge and Shaw, “Galena Folio). It may have been listed 
elsewhere under the name of Pyramidula cronkhiter anthonyi, which 


58 FRANK COLLINS BAKER 


somewhat resembles shimekitz. The specimens collected by Dr. 
Leighton are slightly smaller than specimens from the Iowa locali- 
ties. 

Oreohelix iowensis (Pilsbry). Fragments of a large land shell 
in the Peorian loess near Peoria are believed to be this species. 
This fossil has been reported several times from Illinois deposits, 
and as it is a characteristic loess fossil, it may be of value to list 
‘these localities for comparison. 


Virginia, Cass County (listed as Helix strigosa), Sangamon loess (Leverett, 


Bb WA) 

Fulton, Whiteside County, Peorian loess (McGee, p. 448). 

Savanna, Carroll County (Chamberlin & Salisbury, p. 285), driftless area. 
(See Baker, Life of Pleistocene.) 


Euconulus fulous (Miller). A single specimen occurred in 
each of two deposits, in loess, Whiteside County, and in peat, 
Carroll County. Both are typical. 

Vitrea rhoadsi Pilsbry. Two specimens of this little snail 
occurred in the peat deposit in Carroll County. They were appar- 
ently typical. 


WABASH DEPOSITS IN THE VALLEY OF THE ILLINOIS RIVER 


Mr. Harold E. Culver, geologist of the State Geological Survey, 
recently obtained a very interesting collection of molluscan material 
from marl deposits in Grundy County. Mr. Culver has furnished 
the following information concerning this deposit and its relation 
to the associated strata: 


The shells from the marl deposit were uncovered in coal stripping opera- 
tions near Morris, Illinois. The pit is located in the S.E. quarter of sec. 34 of 
Saratoga Township (34-N., 7 E., 3d. P.M.). 


TYPICAL SECTION OF UNCONSOLIDATED STRATA 


Inches 

5. swamp) silts, medium) todarkienay eesvereeer e aneees erie eee 18 
4. Porous clay, bright yellow-brown in color, lower surface very uneven, 

and thickness, variable: ave motauims «elas! ion e ein oe acd exe eee 2 

3. Marl) lisht:gray im color. sive. ose: tices oie ee eee 8 
2. Lake muds, gray, iron-stained, porous and full of organic matter, no 

shells, some glacial bowlders, and in places till at the base.......... 24 


wT. INO. 2 coal; average sia ny-miare sce sevice rates eel ote ovat eteneoke yc ee ean 30 


PLEISTOCENE MOLLUSCA FROM ILLINOIS 59 


From the character of the deposit and its relation to the underlying 
material, it is clear that it is post-Marseilles in age, and has evidently been 
laid down in a somewhat limited depression, which was probably connected 
with the main Illinois Valley through a narrow outlet three or four miles to the 
southwest. This depression presumably constituted merely an area of over- 
flow for the older Illinois River, and still bears somewhat of the same relation to 
the present stream. The deposit of marl is in places less than three inches in 
thickness, but probably exceeds a foot at the maximum. The areal extent is 
not known, but judging from the topographic relation the basin in which it 
was deposited covers two or three miles. It does not seem probable, however, 
that the marl deposit is even as extensive as this. 


From Mr. Culver’s description the deposit would seem to be 
related to the old glacial outlet from Glacial Lake Chicago, when 
that body of water was at one of its high stages, possibly the 
Calumet stage. During the Glenwood and Calumet stages the 
Illinois Valley was well filled with water and every little cove, inlet, 
or depression near the valley margin was filled with water and 
formed ideal habitats for fresh-water mollusks such as are now 
found in the deposit under discussion. Similar marl beds are 
known from Joliet and are now being studied. The species in the 
Morris deposit are mostly identical with those found in the Chicago 
basin (Baker, 19200), and it is to be presumed that the latter area 
was supplied with life from the Illinois Valley. It is not impossible 
for the Morris deposit to be pre-Lake Chicago in age, as the mol- 
lusks in the deposit could easily have lived during the Glenwood 
stage of Lake Chicago. 


DISCUSSION OF SPECIES 


Pisidium tenuissimum calcareum Sterki. An abundant species 
occurring also in the deposits at Urbana. 

Pisidium compressum Prime. This species is evidently rare in 
this deposit, only a single valve being found in picking over a halt- 
pint of material. 

‘Amnicola leightonit Baker. Abundant. ‘This Amnicola appears 
to be peculiar to glacial deposits. Some of the Ammnicola recorded 
from Pleistocene deposits in the Chicago basin (Baker, 19206) are 
this species and not Ammicola limosa, although specimens believed 
to be limosa occur. It is recorded from Ohio (Baker, 19204, p. 448), 


60 FRANK COLLINS BAKER 


and the species probably has a wide distribution in Pleistocene 
deposits, from which it has been listed as limosa. 

Amnicola walkeri Pilsbry. A common species in this deposit. 
Also widely distributed. 

Amnicola lustrica gelida Baker. ‘This variety of Ammicola 
lustrica was recorded as lustrica variety in a previous paper on 
Ohio glacial mollusks (Baker, 1920a, p. 448). It occurs abun- 
dantly in the Gundy County deposits, at Chicago, in other parts of 
Illinois, in Wisconsin, and in Michigan. It is constantly separable 
from typical lustrica and should have a name to distinguish it 
(see Baker, 1921, p. 22). It is probably peculiar to Pleistocene 
time. 

Valvata tricarinata (Say). Both Recent and Pleistocene 
carinate Valvatae show a large amount of variation in the degree 
of carination. Most of these variations have been named and 
but one possible combination seems unrecognized (supracarinata), 
and this has been characterized from specimens in the Grundy 
County deposit (Baker, 1921, p. 24). Of these possible variations, 
seven in all, five occur in the deposit under discussion. ‘These are: 
Valvaia tricarinata (Say), common. 

Valvata tricarinata perconfusa Walker, abundant. 
Valvata tricarinata infracarinata Vanatta, rare, two specimens observed. 


Valvata tricarinata supracarinata Baker, rare, four specimens observed. 
Valvata tricarinata simplex Gould, rare, one specimen observed. 


Physa anatina Lea. A common Physa in this deposit appears 
to be Lea’s species, although the spire is not as high as in normal 
anatina. 

Physa walkeri Crandall. Common, associated with anaiina. 
Apparently typical. Both species are widely distributed in Pleisto- 
cene deposits. 

Planorbis campanulatus Say. Rare, but two specimens observed. 

Planorbis antrosus Conrad. 

Planorbis antrosus striatus Baker. Both anirosus and its 
variety striatus occurred commonly, in about equal numbers. A 
very common species in Pleistocene deposits. 

Planorbis deflectus Say. Not common. ‘Typical of the species 
as found Recent. 


PLEISTOCENE MOLLUSCA FROM ILLINOIS 61 


Planorbis exacuous Say. <A single specimen of this characteris- 
tic species was observed. It differs from Recent exacuous in having 
a bluntly acute periphery, hence higher whorls, and in the decided 
deflection of the aperture. It may be an individual variation, as 
but the one specimen was found. 

Planorbis altissimus Baker. This common small Planorbis 
occurred abundantly in the deposit. It is the most abundant 
species of the genus in Pleistocene deposits. 

Calba obrussa decampi (Streng). Common, but most of the 
material is immature. 

Mr. Culver also collected a number of shells in Section 23 of 
the same township, in a slight depression which drains fairly well 
to the southwest. Mr. Culver says of this material: 

The shells lie scattered through about a foot of black mucky soil, which 
overlies clay which is presumably weathered Marseilles till. Although lying 
at an elevation considerably above the present alluvial flats of the Illinois 
River, this appears to be merely an accumulation of recent muds, and hence 
cannot be referred to the glacial epoch. 


Four species are represented in the lot: 


* Galba reflexa Say. Planorbis trivolvis Say. 
Physa gyrina Say. Planorbis pseudotrivolvis Baker. 


All were abundant except Planorbis pseudotrivolvis, of which but 
one specimen was present. This is an interesting find, because 
this form of Planorbis is present in the Chicago deposits and is now 
living near Urbana (see Baker, 1920, p. 123). The fauna is one 
that we now find in swales or summer-dry ponds in many parts of 
Illinois, or in the small branches of rivers near their sources. 


LIST OF WORKS CITED 


BAKER, FRANK COLLINS. 

t918. ‘‘Post-glacial Mollusca from the Marls of Central Illinois,” Journal 

of Geology, Vol. XXVI, pp. 659-71. 
to19. “Description of a New Species and Variety of Planorbis trom 

Post-glacial Deposits,” Nautilus, Vol. XXXII, pp. 94-97. 

1920. ‘A New Planorbis from Illinois,” Nautilus, Vol. XX XIII, pp. 123- 
Dis 

19204. ‘‘ Pleistocene Mollusca from Indiana and Ohio,” Journal of Geology, 
Vol. XXVIII, pp. 439-57. 


62 FRANK COLLINS BAKER 


BAKER, FRANK COLLINS. 
1920). ‘““The Life of the Pleistocene or Glacial Period,” University of 
Illinois Bulletin, Vol. XVII, No. 41, pp. i-xiv, 1-476. 
1921. “‘New Forms of Pleistocene Mollusca from Illinois,” Nautilus, 
Vol. XXXV, pp. 22-24. 


CHAMBERLIN, T. C., and SALISBURY, R. D. 


1888. ‘Preliminary Paper on the Driftless Area of the Upper Mississippi 
Valley,” U.S. Geological Survey, Annual Repori, Vol. VI, pp. t99- 
B22. 


LEVERETT, FRANK. 
1899. ‘The Illinois Glacial Lobe,” U.S. Geological Survey, Monograph 
38. 


McGEE, W. J. 
1891. ‘‘The Pleistocene History of Northeastern Iowa,” U.S. Geological 
Survey, Eleventh Annual Report, Part I, pp. 189-577. 


PitsBry, HENRY A. 
1919. Monograph on Pupillidae in Tryon’s Manual of Conchology, II, 
Vol. XXV. 


SHaw, E. W., and TROWBRIDGE, A. C. 
1916. “‘Galena-Elizabeth Folio, Illinois-Iowa,”’ U.S. Geological Survey, 
Geological Atlas, No. 200. 


STERKI, VICTOR. 
1892. Monograph on Vallonia in Tryon’s Manual of Conchology, IT, Vol. 
VIII, pp. 247-61. 
1916. “‘A Preliminary Catalog of the North American Sphaeriidae,” 
Annals, Carnegie Museum, Vol. X, pp. 429-77. 


Wotr, JOHN. 


1870. ‘Descriptions of Three New Species of Shells,” American Journal 
of Conchology, Vol. V, p. 198. 


THE MUDDY MOUNTAIN OVERTHRUST IN 
SOUTHEASTERN NEVADA? 


CHESTER R. LONGWELL 
Yale University 


INTRODUCTION 


Geologists have long recognized the fact that the region of the 
Basin Ranges has passed through more than one period of deforma- 
tion since the middle of the Mesozoic era. The most recent impor- 
tant disturbance found expression chiefly in great normal faults, 
which divided the crust into essentially parallel mountain ranges, 
Within many of these mountain blocks the structure is complex. 
Tertiary formations are strikingly unconformable among them- 
selves, and rest with profound unconformity on the older rocks. 
This great unconformity is a record of prolonged erosion following 
an epoch of crustal compression in Mesozoic time. The youngest 
rocks involved in this disturbance appear to be of Jurassic age; and 
therefore it is probable that the movement was contemporaneous 
with the folding of the Sierra Nevada and Humboldt Ranges. 

The degree of deformation due to this mountain-making varies 
greatly in the different ranges. In some the Mesozoic and older 
strata appear to have been nearly horizontal before the last impor- 
tant period of faulting. Other ranges preserve a record of the 
-Nevadian movement in folds of varying intensity. In the Muddy 
Mountains of southeastern Nevada the record consists of over- 
turned folds and an overthrust of considerable magnitude. These 
structural features were recognized by the writer in the course of 
field work during the summer of 1919, and a brief description has 
appeared in another article? The present paper will give a more 
detailed description of the overthrust (Figs. 1 and 2). 


t Published by permission of the Director of the United States Geological Survey. 
2 Amer. Jour. of Sci., Vol. L (Jan., 1921), pp. 39-62. 
63 


64 CHESTER R. LONGWELL 


TOPOGRAPHY OF THE MUDDY MOUNTAINS 
The Muddy Mountains are a complex range bounded by wide 
valleys of structural origin. From north to south the length of the 
range is nearly 40 miles, and the maximum width, measured from 


z—Bluepoint 


Ji) Pennsylvanian?) é 4 
Ej Pennsylvanian Mis- EsPG 7 
E— 1 sissipptan, Devonian iy SN 


FN Fautt 
Probable fault 
I] Overthrust contact’ 


Index Map 
Oo 10 209% 
Miles 


Sea Level— 


SECTION ON LINE A-A 
LEGEND 


te) 5 0 10 


Horizontal Scale (Miles) Vertical Scale 
(Thousands of feet) 


YEVEATLLON 
(Pennsylvanian, Mis- 


L—1 sissippian,4Devonia 


SECTION ON LINE B-B 


Probable location of 
fault or thrust fault 


Fic. 2.—Cross sections showing structure in the Muddy Mountains 


east to west, is approximately 15 miles. The intermontane valleys 
form a border several miles in width, with elevations ranging from 
less than 1000 to slightly over 2,000 feet above sea-level. The 
northern part of the mountains consists of ridges developed by 


THE MUDDY MOUNTAIN OVERTHRUST IN NEVADA 65 


differential erosion on tilted strata. These ridges trend generally 
north and south and are elevated only a few hundred feet above the 
bordering valleys. Similar ridges, trending northeastward, form 
the southern part of the mountains. The central division consists 
of faulted portions of an overthrust block, and is the dominating 
part of therange. This division of the mountains is shaped roughly 
like a letter C, with the opening toward the south. It is a dis- 
tinct unit, made conspicuous by its superior average elevation, by 
the abrupt faces bounding it on practically all sides, and by the dark 
Paleozoic limestones that form the greater part of the mass. For 
convenience in reference the unit has been called Callville Moun- 


Fic. 3.—View of White Basin and part of Callville Mountain, looking northwest 
from the south end of the White Basin fault scarp. 


tain. Its upper surface is rugged, and a number of sharp peaks rise 
prominently above the general elevation. Muddy Peaks, the high- 
est points, are nearly 5,800 feet above sea-level. Immediately 
north of the peaks the continuity of the mountain has been some- 
what broken by faulting and erosion, and Jurassic sandstone is 
exposed in a depression known as Muddy Peaks Basin. White 
Basin, a low area floored by light-colored Tertiary deposits, lies 
within the curve of the C formed by Callville Mountain (Fig. 3). 


STRATIGRAPHY 


Except for small amounts of igneous materials, all of the rocks 
exposed in the Muddy Mountains are sedimentary. They include 
strata of Paleozoic, Mesozoic, and Cenozoic age. The Callville 
Mountain block consists of four formations, chiefly heavy limestones, 


66 CHESTER R. LONGW ELL 


recognized as Devonian, Mississippian, and Pennsylvanian. The 
lowest horizon at which fossils were found lies nearly 1,000 feet 
above the base of the series, and therefore a part of the lower beds 
may belong to periods older than the Devonian. At the top of the 
block an unknown thickness of the Pennsylvanian beds has been 
removed by erosion. Outside the borders of Callville Mountain the 
oldest strata exposed are the upper part of the Supai formation, 
which is mapped (Fig. 1) with the Permian. Thus there is a gap of 
undetermined value in the stratigraphic succession. The following 
table summarizes the nature of formations and shows their relation- 
ships: 


5 Thickness 
Age Formation Nature (in feet) 
Quaternary Alluvial deposits Gravel, sand, and silt 0-400 
Unconformity 
P.iocene ? Intermontane clays Clay, silt, and sand, with salt, 200-1800 
gypsum, and other saline 
deposits 
—Unconformity: 
Horse Spring formation Limestone, clay, sandstone,| 1000-2800 
Tertiary conglomerate, volcanic ash, 
magnesium carbonate, gyp- 
Miocene? sum, and other saline ma- 
terials 
Overton fanglomerate Coarse alluvial fan deposits, 
in part firmly cemented 
—Great unconformity = 
Jurassic Cross-bedded sandstone Red and gray sandstone, with 750-2000 
(equivalent to the LaPlata extreme cross-bedding 
group) 
Chinle formation Sandstone, shale, and gypsum, 800-3 200 
with some silicified wood 
Shinarump conglomerate Conglomerate and _ cross- 10-200 
Triassic bedded sandstone, with 
abundance of silicified wood 
—Disconformity — 
Moenkopi formation Limestone, shale, sandstone,| 1200-1600 
and gypsum 
Disconformity . 
Permian Kaibab limestone Heavy beds of limestone, with 400-700 
some sandstone and gypsum 
Pensylvanian ? Supai formation Red and gray sandstone, in 100-800 
part cross-bedded, with} (incomplete 
some gypsum and gypsifer-| exposures) 
ous shale 
Unknown thickness not ex- . 
pose . Heavy beds of dark gray lime- I1I00 
Pennsylvanian Callville limestone stone, cherty in part. (incomplete) 
Bluepoint limestone Dark carbonaceous limestone, goo 
with chert nodules 
Mississippian == 
Rogers Spring formation Cherty limestone and dark 600 
gray quartzite 
Disconformity — 
Devonian Muddy Peaks limestone Heavy beds of gray limestone, 1200 
(and older ?) in part carbonaceous 


THE MUDDY MOUNTAIN OVERTHRUST IN NEVADA 67 


STRUCTURE OF CALLVILLE MOUNTAIN 


Fault boundaries —Very evidently the steep sides of Callville 
Mountain are in large part due to normal faulting. Portions of the 
actual fault surface are well exposed, and at many points the adja- 
cent younger formations may be seen in fault contact with the 
Paleozoic rocks of the mountain. The White Basin fault scarp is 
exceptionally well preserved. Adjacent to White Basin it is sev- 
eral hundred feet high, and its remarkable regularity gives it the 
appearance of an artificial wall. At the base of the scarp, Ter- 
tiary beds are turned steeply upward along the edge of the down- 
thrown block, and large slickensided surfaces are numerous on the 
hard Paleozoic limestone of the footwall. Slickensides are distin- 
guishable even at a height of more than 100 feet above the base of 
the cliff. These fault surfaces have an average westward inclina- 
tion of 50°, and near its southern end the scarp as a whole has 
the same inclination to a height of nearly 1,000 feet. The un- 
dissected character of the scarp is due in part to the eastward tilt 
of the footwall block. Consequent streams on the block flow 
eastward, and have cut deep canyons across the Rogers Spring 
fault scarp, which accordingly has a much older appearance 
than the White Basin scarp. At the base of the Rogers Spring 
front the Tertiary beds are upturned steeply toward the moun- 
tain, and portions of the fault plane are exposed, dipping 60° 
eastward. The block east of the fault has been thrown at least 
several hundred feet. Thus the part of Callville Mountain included 
between the Rogers Spring and White Basin faults is a typical horst. 
. Similar faults, all with large throw, bound White Basin on the 

west, and at least a part of the west boundary of Callville Mountain 
is formed by normal fault scarps. It is very natural, therefore, to 
infer that the mountain is an irregular horst or group of horsts left 
elevated by the sinking of adjacent blocks. For the northern 
boundary, however, this explanation is not tenable. This side of 
the mountain is high and precipitous, and has the general appear- 
ance of an eroded fault scarp; but the fault plane, exposed at many 
points near the base of the cliff, has an average dip of nearly 70 to 
the south, and the older strata of the mountain have very evidently 
moved down with respect to the younger formations on the north. 


_ 68 CHESTER R. LONGWELL 


Therefore the original topographic expression of the fault has been 
reversed by differential erosion. This anomalous relation indicates 
that the structure is more complex than is apparent at the borders of 
the mountain (Fig. 4). 

Overthrust relations —The key to the older structure of Call- 
ville Mountain is furnished by Muddy Peaks Basin. Pro- 
nounced doming has affected the western part of the mountain, 
the strata on the sides of the dome dipping at a maximum of 
45°. Erosion has stripped the thick Paleozoic beds from the 


Fic. 4.—Portion of the north front of Callville Mountain. In the foreground, 
Triassic sediments are masked by recent waste. Outcrops of Jurassic sandstone 
partly covered by talus near base of cliff. 


summit of the dome, exposing cross-bedded Jurassic sandstone 
beneath. The sandstone forms the floor of an erosional basin, 
which has been divided into two lobes by a faulted tongue of 
Tertiary sediments. Limestone walls rise abruptly around the 
margins of the basin, and are retreating rapidly due to sapping 
of the comparatively weak sandstone beneath. Remnants of the 
limestone, lying almost horizontally, cap sandstone hills near 
the center of the basin. The contact of limestone and sandstone, 
well exposed at many places, is plainly one of overthrust, and is 
essentially parallel to the bedding both below and above. Immedi- 
ately above the thrust surface a breccia has been developed, the 
thickness ranging from a few feet to at least 75 feet. Most of the 
fragments are of limestone, but near the base some pieces of sand- 


THE MUDDY MOUNTAIN OVERTHRUST IN NEVADA 69 


stone are included. The breccia is well cemented with calcium 
carbonate, and the base is smoothly polished. Above the zone of 
typical breccia the limestone is intensely shattered through a thick- 
ness ranging from too to 500 feet. The shattered condition of the 
rock is responsible for the development of numerous grotesque forms 
by the agents of erosion. 

There can be no reasonable doubt that the formation beneath 
the overthrust is the Jurassic sandstone, corresponding to the 
LaPlata group. No fossils have been found in this formation, but 
its lithologic features are very distinctive. The only other for- 
mation in the region that in any way resembles it is the Supai; 


Fic. 5.—View to the northwest across Muddy Peaks Basin, showing remnants 
of dark Paleozoic limestone capping hills of Jurassic sandstone. 


and to anyone who has studied the two formations carefully the 
differences are sufficiently apparent to prevent confusion. A thick- 
ness of approximately 700 feet of the sandstone is exposed in Muddy 
Peaks Basin, and throughout this thickness the rock has large scale 
cross-bedding, both straight and tangential; it has alternating 
laminae with coarse and fine grains; and it shows an irregular blend- 
ing of red coloring with gray. All of these features are typical of 
the Jurassic sandstone. 

Extent of the overthrust.—It is quite certain that the entire Call- 
ville Mountain mass is included in the overthrust block, although 
the actual surface of thrusting is exposed only in the “window” of 
Muddy Peaks Basin. The mountain is a unit in stratigraphy, and 
in all parts the lower beds show the intense shattering which is a 


70 CHESTER R. LONGWELL 


characteristic feature in the vicinity of Muddy Peaks. Moreover, 
the unusual relation of older strata to younger along the entire 
north side of the mountain is strong evidence of the unity of the 
block. The known portion of the overthrust plate, therefore, ex- 


Fic. 6.—Devonian limestone, at the north end of Muddy Peaks Basin. A 
thickness of about 600 feet is shown. Gray Jurassic sandstone beneath the overthrust 
appears at the left of the view. 


Fic. 7.—Close view of the overthrust contact. The hammer rests on the Jurassic 
sandstone. Well-cemented breccia lies above the surface of. overthrust. 


tends from the west side of the Muddy Peaks mass to Bluepoint, 
a distance of nearly 18 miles, and from Piute Point to the south 
side of White Basin, a distance of to miles. It is not possible to 
estimate the original dimensions of the plate. An indication that 
the movement affected a large area is given by the fact that the 
stratigraphic displacement apparent in Muddy Peaks Basin is ap- 


THE MUDDY MOUNTAIN OVERTHRUST IN NEVADA FT 


proximately g,ooo feet, and yet the thrust surface is almost paral- 
lel to the bedding. The portion of the plate that formerly lay 
north of the Arrowhead fault has been removed by erosion, whereas 
on some other sides of Callville Mountain the strata involved in 
the movement have been carried downward by normal faulting. 

Structural features adjacent to Callville Mouutain.—North of 
Callville Mountain the Mesozoic rocks form a broad anticline 
with its axis extending slightly north of west. Lateral move- 
ment on the White Basin fault has caused displacement of the 
strata within the anticline, corresponding to the offset in the 
north boundary of Callville Mountain. To the east the fold 
plunges beneath Tertiary sediments, and to the west it dies out on 
meeting a stronger north-south anticline which is overturned toward 
the east. The north-south fold is affected by a strike fault, and 
near Callville Mountain the structure is further complicated by 
overthrusting. An overturned block of Kaibab and Moenkopi 
limestones has been thrust toward the northeast and now lies di- 
rectly on the Jurassic sandstone. The thrust plane dips to the 
southwest at an angle of 60°; but it is possible that its original atti- 
tude was less steep, and that its present inclination is in part due to 
tilting during normal faulting which has affected the block. 

One structural feature south of Callville Mountain deserves 
special mention. Nearly a mile southeast of the Muddy Peaks 
mass an isolated block of Callville limestone forms a prominent 
ridge. The beds dip northwest at an angle of 30°, and on the south 
side the block is faulted down against Jurassic sandstone. This 
relation, which is similar to that on the north side of Callville Moun- 
tain, suggests that on the south side also the overthrust plate has 
been faulted down against younger formations. To what extent 
this is true cannot be definitely determined, because the Tertiary 
deposits conceal all except small portions of the older rocks. Pos- 
sibly the isolated exposure of Callville limestone is only a local 
feature, with exceptional structural relations. 


DIRECTION OF OVERTHRUSTING 


No striations were found on the polished thrust surface to indi- 
cate direction of movement; but crumpling and minor thrusting in 


72 CHESTER R. LONGWELL 


the vicinity of Muddy Peaks Basin suggest that the overthrust plate 
moved from southwest to northeast. Further evidence favoring 
this conclusion is furnished by the strike of the minor overthrust 
north of the Arrowhead fault. Very probably this thrust was asso- 
ciated with the major movement, and had the same general direc- 
tion. Some caution must be used, however, in drawing conclu- 
sions from the present strike of the smaller thrust plane, for it may 
have been changed considerably by the normal faults that cut the 
block. 
GENERAL DEDUCTIONS 


The east side of Callville Mountain is less than 40 miles from the 
Grand Wash Cliffs; and the original eastern edge of the overthrust 
plate was probably much nearer the western edge of the plateau. 
Thus there is evidence of intense compressive deformation near the 
border of the Basin Range region, closely adjacent to a wide area in 
which the effects of crustal compression are practically absent. 
Further evidence of the intensity of the post-Jurassic revolution 
in this part of Nevada is seen in the closely compressed and over- 
turned fold extending north from Callville Mountain. This close 
folding and the overthrusting involve the same Mesozoic strata, 
and therefore may be different expressions of the same general crus- 
tal movement. 

Rowe has reported an overthrust in the Spring Mountain Range, 
40 miles west of the Muddy Mountains, stating that ‘“‘the Carbon- 
iferous limestone is thrust over the red Mesozoic sandstone.’* This 
description suggests the possibility that the feature is contempora- 
neous with the Muddy Mountain overthrust. It is a noteworthy 
fact that in a large part of the Spring Mountains and in neighboring 
ranges folding is gentle or practically absent. The writer ventures 
the suggestion that detailed mapping may show some of these 
ranges to be faulted portions of a great overthrust plate. Evidently 
the post-Jurassic revolution had a profound effect in at least a part 
of southern Nevada, and further field work may be expected to 
reveal additional evidence of the movement. 


™ Quoted by J. E. Spurr, U.S. Geol. Survey Bulletin No. 208, p. 177. 


GROWTH-STAGES OF THE BLASTOID, OROPHOCRINUS 
STELLIFORMTIS 


F. A. BATHER, F.R:S. 
British Museum (Natural History), London 


The references to literature are given in the List of the Genera 
and Species of Blastoidea in the British Museum, 1899. 

Among the specimens entered in that List are some referred by 
Etheridge & Carpenter to Orophocrinus stelliformis (E840 and 
E 1055), which, however do not agree well with the description by 
those authors or with the original figures of the species, but seem 
to agree more nearly with the form which Etheridge & Carpenter 
distinguish as var. campanulatus Hambach. 

A set of sixteen specimens collected by Mr. D. K. Greger in 
the Lower Burlington Limestone at White Ledge, Marion Co., 
Missouri, and presented to the British Museum in 1918 by the 
late Dr. G. B. Longstaff (E30091-30106), if arranged in order of 
size, shows a complete transition from the campanulatus form to 

the typical stelliformis, the latter being the larger. The dozen or 
- so specimens previously in the museum fit into their places in this 
series. ‘There can therefore be no doubt that the alleged species 
_ or variety O. campanulatus is merely the young of O. stelliformis. 

Apart from mere size the growth-changes consist in: 

1. An increase in the greatest diameter as compared with the 
height of the theca. 

2. The lowering of the ambitus from a position but little below 
the oral pole to one less than half the height of the theca from the 
base. 

3. A consequent change from a tumbler outline, through a 
bell outline, to a parachute outline, the sides changing from straight, 
or in part convex, to concave, and the summit changing from almost 
flat to flattened-convex or convex. 


73 


74 F. A. BATHER 


4. With the extension of the ambulacra in length, the radials 
grow outward, and the cross section at the ambitus becomes more 
stellate. 

s. Concurrently the number of ambulacrals (side-plates, etc.) 
increases. 

6. A change in the outline of the ambulacra outside the deltoids 
from broadly lanceolate, through narrowly lanceolate, to the 
characteristic outline of the adult, which is broad in the proximal 
region, then rather rapidly tapering to a region from one-third to 
two-thirds the distance from the proximal end, then narrow and 
gently tapering. 

7. The distal region of the ambulacrum is in all stages depressed 
below the edges of the radial sinus, but the proximal region, which 
in the young is flush with the sinus, gradually rises above it, until 
in the largest individuals it projects conspicuously; and this pro- 
jecting portion increases in length, until it nearly (though never 
quite) attains the distal lip of the sinus. 

8. Concomitantly with the preceding change, the edges of the 
sinus become sharply beveled, and the two beveled faces of 
adjacent limbs of the radials meet in a ridge on the deltoid. ! 

9. The curve of the side of the theca below the ambitus, though 
described above as straight or concave, is never a simple curve, 
but consists of three parts: (a) the almost vertical sides of the 
basal circlet, (0) the outwardly spreading lower part of the radials, 
which is straight or feebly convex, (c) the part of the radials below 
the lip, which in the young bends outward almost imperceptibly, 
but gradually lies at a sharper angle to region 6, until it is at last 
horizontal. 

10. The distal parts of the radial above this region c come to 
be marked off from the body of the radial by a vertical depression, © 
which at its upper end meets the distal end of the beveled surface 
mentioned under (8). 

11. Various changes of proportion in details are scarcely worth 
setting out, since all follow from the one great change, namely the 
extension of the ambulacra outward, arching upward, and bending 
downward, unaccompanied by equally rapid growth of the other 
elements. 


GROW TH-STAGES OF “OROPHOCRINUS STELLIFORMIS” 75 


It follows from the preceding that such a description as that 
given by Etheridge & Carpenter for O. stelliformis could not apply 
even to all the forms which they themselves labeled with that name. 
Some of the necessary modifications will be gathered from what 
has been said. There are other variations which may be mentioned. 

The deltoids are said to be constricted at one-third of their 
length from the proximal end. In the large E 30091 the measure- 
ments are 4.6/8.4 mm., i.e., more than one-half. In the medium- 
sized E 30099 they are 1.8/3.8 mm., i.e., rather less than one-half. 
In the smallest E 30106 they are 1.1/1.9 mm. 

The anal deltoid is said to have its distal margin rounded. This 
seems to be the case in the two specimens from Burlington registered 
E 841, as also in E 30096; but in most of the specimens the sutures 
are straight, and, even when curved, nearly always meet in a 
- decided angle. 

The number of side-plates on each side of an ambulacrum is 
given as fifty, on the authority of Meek & Worthen. It did not 
reach this number in any of the specimens before Etheridge & 
Carpenter. In Ex1oss the total is thirty-five or less. In those 
registered E 840, the total is forty or less, and in one case twenty 
orless. In E 8172a, it is about twenty-five; and in E 8172), about 
forty. In the very young E 30106 from White Ledge, the number 
is about nine. From this it increases to fifty in E 30093, sixty in 
- E 30092, and something over sixty in E 30091. In E 8426, referred 
to var. campanulatus by Etheridge & Carpenter, the number is 
thirteen or fourteen, and was probably the same in E482a. In all 
- individuals the number of side-plates to 3 mm. is nine or a fraction 
more. 

The ornament, when preserved, as in E 30094, consists, on the 
radials, of the usual growth-lines; these also occur on the upper 
part of the basals, but on their lower part, where the basals form 
a cylinder, the lines are coarse, irregular, and rugose. 

The cover-plates pass right up to the oral pole in E 30099, 
apparently without fusion or enlargement. Since no one has ever 
doubted that they could open on the ambulacra, there is no reason 
why precisely similar plates should not have opened over the mouth. 
Possibly these plates may have fused over the mouth in some 


76 F. A. BATHER 


individuals, as supposed by C. A. White, Meek & Worthen, and 
Etheridge & Carpenter. In E 30106 they are not preserved, but 
the notches indicating a mobile articulation extend round the peri- 
stone; in so young a form one would not expect these plates to 
have been fused. 

The new material shows no traces of plates covering the anal 
aperture, or of facets for plates of a tube raised above it as suggested 
by Etheridge & Carpenter. 

The horizon of all specimens in the British Museum is given as 
Lower Burlington. I do not find either the species or its supposed 
variety recorded from any other horizon. It would, however, be 
interesting to work out its vertical distribution with more accuracy, 
and to see whether the earlier representatives are closer to the 
tumbler or bell-form of the smaller, and supposed younger, indi- 
viduals in our series; and whether the later representatives are 
closer to the parachute-form of the larger, and supposed older, 
individuals. 


-POSSIBLE SILURIAN TILLITE IN SOUTHEASTERN 
BRITISH COLUMBIA 


FRANCIS P. SHEPARD 
The University of Chicago 


From the eastern slopes of the Purcell Range, in the vicinity of 
Lake Windermere, the view across the deep depression known as 
the Rocky Mountain trench to the Rocky Mountains on the east, 
shows a series of bands of rocks varying in color and running 
roughly parallel to the trench. One of these is a red conglomerate, 
which can be traced almost uninterruptedly for ten miles between 
the Sinclair River and Shuswap Creek, both of which are tributary 
to the Columbia River in the Rocky Mountain trench. The red 
formation was examined in two localities, and in a third a yellow 
conglomerate was found, which is undoubtedly a part of the same 
formation. In the last instance evidence was found which points 
toward a glacial origin of the conglomerate. 

_ The conglomerate was first studied at Sinclair Springs. Here 
an outcrop of red rock, with a dip of about 90°, forms a high wall 
running down one side of the canyon and up the other (Fig. 1). 
Closer examination shows that the rock is not made up entirely of 
red material, but consists of bowlders, which are predominantly red, 
imbedded in a yellow calcareous matrix. The bowlders are lime- 
stone with the exception of a very few pieces of quartzite. The 
limestone varies considerably, having such varieties as purple, 
black, white crystalline, blue with white veins, and gray. The bowl- 
ders are mostly sub-angular and in some cases are as large as four 
feet in diameter (Fig. 2). The material of the conglomerate is unas- 
sorted, and shows no trace of stratification. Two bowlders were 
seen which had polished surfaces. It was difficult to remove the 
bowlders from the matrix, and the exposed surfaces of the bowlders 
were so weathered that any striations they may have once pos- 
sessed have been destroyed. 


il 


78 FRANCIS P. SHEPARD 


The thickness of the conglomerate is about 200 feet: Against 
it on the east, with apparent conformity, is gray limestone. This 
contains a scattering of pebbles in the portion next the conglomerate. 
Six hundred and fifty feet below the top of the limestone is a bed 
containing fossils of Richmond age. As these beds are vertical, 
the age relations of the conglomerate to the limestone are not 
shown by the structure, but since somewhat older Ordovician faunas 


Fic. 1.—The red conglomerate near the Sinclair Hot Springs 


occur farther east, the conglomerate is known to be 650 feet above 
strata containing the Richmond fauna. Above the bowlder bed 
there is a break in the succession, and above the break is massive 
crystalline limestone in which no fossils were found. Farther west 
there is severe faulting which cuts out further evidence from fossils 
on that side. From the evidence on the east it was concluded that 
the age of the formation probably is Silurian because of the 650 
feet of limestone between it and the fossil bed which represents 
the top of the Ordovician. 


SILURIAN TILLITE IN BRITISH COLUMBIA 79 


Two miles up the canyon of Shuswap Creek the southward 
continuation of the formation was examined and found to be prac- 
tically identical with that seen at Sinclair Springs. The bowlders 
are not so large on the average, but are equally angular and un- 
rounded. In two places there are thin lenses of crumpled strata 
included in the formation. Both of these were only a few inches 
thick and did not extend more than 4’or 5 feet before being cut off 
by bowlders. The formation here is 150 feet thick. To the west 
there is crystalline limestone very similar to that found on the 


Fic. 2.—A typical section of the conglomerate. The hammer illustrates the 
size of the bowlders. 


same side at Sinclair Springs. On the east there is a thin bed of 
quartzite, followed by massive limestone. To the south, up Winder- 
mere Creek in the vicinity of the Blue Lakes, there is conglomerate 
in direct line with the red bowlder bed. Here the character of the 
bowlders is quite different, as are also the distribution of the out- 
crops and the type of weathering, but there is sufficient evidence 
to show that the two conglomerates are parts of one formation. 
The formation in Windermere Creek valley is probably 400 or 
500 feet thick and extends along the north side of the canyon for at 
least three-quarters of a mile. It is in the form of an anticline 
plunging to the south, and disappears in that direction. 


80 FRANCIS P. SHEPARD 


The bowlder bed here consists of a firmly cemented aggregation 
of limestone, quartzite, and shale blocks with a few fragments of 
igneous rocks. Limestone is the most abundant constituent, but 
in some places quartzite bowlders predominate. Many of the 
pebbles and bowlders are distinctly faceted. Many of them are 
polished and some rounded as if by stream action. In some cases 
the quartzite surfaces are so polished as to be dazzling in the sun- 
light. Markings which suggest glacial striations were found on at 
least one bowlder. A large block of quartzite was seen to have on 
its surface a distinct gouge about ten inches long, three inches 
wide, and half an inch deep. A number of quartzite pebbles had 
concave surfaces. Evidence from fossils in the bowlders indicates 
the age of the conglomerate to be younger than Richmond. 

The contact of the bowlder bed with the underlying calcareous 
shale is distinctly seen. The topmost layer of the shale is contorted 
and contains pebbles ranging up to six inches in diameter. ‘They 
undoubtedly came from the same source as the overlying conglom- 
erate and indicate that the shale was soft when the conglomerate 
was deposited. 

The bowlder bed is overlaid to the west by massive limestone, 
with a slight gradation between the two. It is difficult to find the » 
contact because it is covered by débris in most places. 

Before considering the evidence that the formation is tillite, 
other possible modes of origin may be considered. Is the con- 
glomerate in question (1) a stream conglomerate, (2) an autoclastic 
breccia, or (3) an undercliff breccia? The size and sub-angularity 
of the bowlders are difficult to explain if the conglomerate is fluvia- 
tile. Such bowlders are commonly found only in the valleys of 
mountain streams. As there is no evidence for a change in dip 
between the underlying and overlying limestone at Sinclair Springs, 
it is very improbable that there was any intervening uplift which 
could have led to the development of mountain streams. Further, 
the gradation between the bowlder bed and the limestone shows 
that the conglomerate could not have been deposited by mountain 
torrents. 

Autoclastic breccias are more or less local phenomena. They 
are generally not confined to one stratigraphic horizon, but are 


SILURIAN TILLITE IN BRITISH COLUMBIA 8I 


likely to cross bedding planes, while the formation in question con- 
stitutes a distinct horizon traceable for a number of miles. The 
diversity of types of rock among the bowlders in the Windermere 
Creek occurrence is not characteristic of an autoclastic breccia, and 
the thickness of the formation is also against this interpretation. 

Undercliff breccias have angular unsorted material such as 
found in this formation, but the possibility of adjacent cliffs at this 
time is slight, because of the gradation into overlying and under- 
lying limestone, which shows only slight interruption of marine 
conditions. The great continuation of the formation both along the 
range and in a vertical direction (it has a known vertical range of 
more than 3,000 feet) are altogether against a cliff-base origin. 

On the other hand, evidence for glacial action is strong. It 
includes (1) the finding of probable striations, (2) the sub-angular- 
ity of the bowlders, (3) the distinct faceting of some of the bowlders, 
(4) the heterogeneous mixture of the bowlders without assortment 
or stratification, (5) the occurrence of grooving and concave faces 
on pebbles, and (6) the occurrence of Silurian tillite in southeastern 
Alaska,’ 900 miles away, also is significant. 

Thus it seems possible that glacier ice existed at this period in 
British Columbia and southeastern Alaska. Since there are marine 
deposits on both sides of the formations in Alaska as well as in 
British Columbia, it is possible that the glaciers came out into 
shallow seas in these places. The absence of weathering at the top 
of the conglomerate points to encroachment of the sea before such 
weathering took place. The gradation into limestone in several 
instances in British Columbia suggests that icebergs coming from 
the retreating or advancing ice fronts distributed occasional bowl- 
ders among the limestone deposits. It is not surprising that this 
formation should continue over such great distance without a 
_ break, when it is considered that sea deposits covered it before it 
could be attacked by erosion. 


tE. Kirk, Amer. Jour. Sci., Fourth Series, September, 1918, p. 511. 


REVIEWS 


Source Book for the Economic Geography of North America. By 
CuHartes C. Cotsy. The University of Chicago Press. 1921. 


Even a brief study of this volume shows that it will serve a good 
purpose, for it is a step—and an important one—in the organization of 
material which is scattered so widely that its utilization is out of the 
question for those not in touch with good libraries, and sometimes 
discouraging to those who are. Every student of college grade who 
studies the geography of North America will find the book of great service, 
if not indispensable. ‘The teacher of such students should welcome the 
volume, not as a textbook, but as the basis for a course on the economic 
geography of the continent. He must not expect that the book is to do 
away with his own work, but he may expect it to give him many sugges- 
tions as to methods of procedure and as to sources of material. Intelligent 
use of the book will make it serviceable for courses of various grades of 
advancement, and even for courses with somewhat diverse aims. A good 
feature of the book is that it gives publicity to much excellent material 
published in government reports and largely lost to view by many who 
could utilize it to advantage. 

One cannot help wishing that the book were somewhat fuller, and 
that a well-annotated bibliography had been added. This may come 


with future editions. 
R.. Digs 


The Malagash Salt Deposit, Cumberland County, Nova Scotia. By- 
A. O. Haves. Ottawa: Canadian Geological Survey, Memoir 
UAW, MOS, LEO, Mal, Joie. Th, TOROS Dy 
Mining development has proved the presence of a sufficient quantity 

of pure salt in this area for the establishment of an important industry. 

The deposit is favorably located with respect to the supply of raw 

material and fuel, and has exceptional advantages for transportation 

either by rail or water. 

Outcrops are not very abundant in the area, but sediments belonging 
to the Mississippian, early Pennsylvanian, and late Pennsylvanian or 
early Permian are identified, and are separated by two marked uncon- 


82 


REVIEWS | 83 


formities. The Mississippian is represented by the Windsor series and 
the salt is interbedded with gypsum and shale in this series. The 
Windsor series has been intensely folded and faulted and is now exposed 
on the crest of an anticline where the main workings are located. The 
salt bed is about 75 feet below the surface. A shaft has been sunk to the 
salt horizon and the salt is removed by mining. One bed of rock salt 
over 21 feet thick is already being worked. Sylvite (potassium chloride) 
is present as fairly pure lenses in the halite and probably is a replacement 
of halite. Chemical analyses of rock salt show from 2.5 per cent to 11 
per cent sylvite, and this potassium salt will probably be an important 
by-product from this deposit. Very little is known concerning the 
lateral extent of these salt beds, but since they were formed by the 
evaporation of sea water during the recession of the Mississippian seas 
in Windsor time, they are likely quite extensive. 

The Windsor series is probably much more extensive than formerly 
thought, because limestone in Kings County, New Brunswick, formerly 
correlated with the Albert series, which is unconformable below the 
Windsor series, was found to contain Windsor fossils. A number of salt 
springs occur in the area where these fossils were collected and in other 
localities in New Brunswick and suggest the presence of salt beds in the 
Windsor series of this province, although up to the present no development 


work has been done near any of the springs. 
J. F. W. 


The Limestone and Phosphate Resources of New Zealand. Part I. 
Limestone. By P. G. Morcan. Geological Survey Branch, 
Department of Mines, New Zealand, 1919. 


This is primarily a discussion in detail of the limestone resources of 
New Zealand, considered particularly in their relation to agriculture, and 
_as such is mainly of local interest. The principal features of general 
interest are several excellent plates that illustrate remarkable fluted 
and pinnacled forms developed through erosion by solution. 

As bearing upon the origin of odlitic and pisolitic structures in 
limestone, relations at Kotuku near Greymouth are of interest. Here 
several drill holes, sunk in search of oil, discharge in geyser-like fashion 
salt water highly charged with CO, and dissolved calcium carbonate. 
The greater part of the CO, at once escapes and, in consequence, abun- 
dant carbonate of lime is deposited wherever the water touches any 
solid object. As the water flows away it forms numerous little balls of 


carbonate of lime, in size and shape resembling marbles. Se 


84 REVIEWS 


Maps and Sections to Accompany Report on the Geology and Ore- 
Deposits of Meekatharra, Murchison Goldfield. Geological Sur- 
vey of Western Australia, Bulletin No. 68, 1916. Pls. I-XXYV. 


The bed rock of the area consists chiefly of amphibolite and other 
varieties of schists, into which numerous granitic masses have been 
intruded. There are also present some extrusives and a few local 
occurrences of Devonian shales, arkoses, and dolomites. The bulletin 
consists of numerous geologic maps and sections, and mine maps for 


the area. 
A. (C) Mick 


Maps and Sections to Accompany Report on the Geology and Mineral 
Resources of the Vuilgarn Goldfield. Geological Survey of 
Western Australia, Bulletin No. 1, Pt. III, 1917. Pls. I-XIX. 


The geology as shown on the maps and sections is essentially the 
same as found in the other gold fields of Western Australia, consisting 
chiefly of igneous and metamorphic rocks and a minor amount of sedi- 
mentary. ‘The report includes detailed maps to the scale of one mile to 
the inch, as well as more generalized maps on a much smaller scale. 

A. Co Meck 


The Cambrian and Ordovician of Maryland. By R. S. BASSLER. 
Systematic Report, Maryland Geological Survey, 1919. Pp. 
424, pls. 58, figs. 27. 

The Ordovician and Cambrian are considered together because of 
the continuity of the limestone which forms the top of the one and the 
base of the other. The proposed Ozarkian and Canadian systems are 
not recognized. Three great phases of sedimentation are recognized, 
(1) the Lower Cambrian siliceous phase, (2) the great limestone deposits 
of the Upper Cambrian and Lower Ordovician, and (3) a shale phase 
covering the Middle and Upper Ordovician, the three forming an 
ageregate about 16,000 feet thick, of which the limestones form more 
than half. The first phase mentioned is more or less confined to the 
Blue Ridge, the limestones to the Great Valley and Allegheny Ranges, 
and the shale phase chiefly to the latter. 

The description of formations includes detailed sections, faunal lists, 
discussion of correlations, topographic features, areal distribution, and 
economic resources. In many of them distinct faunal zones are recog- 
nized. There is appended a general faunal list showing the geographic 
distribution of the forms. 


REVIEWS 85 


The part devoted to systematic paleontology contains the descrip- 
tion of some two hundred species and varieties, of which sixteen are 
new. Most of these are figured on the fifty-eight plates. There is also 


a bibliography of Cambro-Ordovician literature. 
i (Cs IME 


Contributions to the Pre-Cambrian Geology of Northern Michigan and 
Wisconsm. By R. C. ALLEN and L. P. BARRretr. The Geology 
of Limestone Mountain and Sherman Hill in Houghton County, 
Michigan. By E. C. Case and W. I. Ropinson. Michigan 
Geological and Biological Survey, Publication 18, Geological 
SSMS US, WOKS, IO, wso, jos, wa, ES) ick. 

Part I consists of a number of contributions on the geology of an 
area lying west of the Crystal Falls and Iron River districts, and extend- 
ing south of the Marquette and Gogebic iron ranges to the state line and 
into Wisconsin beyond. The region is a great Huronian interior in 
which the better-known and structually distinct ranges on its north and 
east borders coalesce and lose their identity. The authors are unable 
to say that rocks older than the Huronian are present, but find that the 
area heretofore mapped as Archaean and undifferentiated pre-Cambrian 
is in reality Huronian. The outstanding feature of the region is the 
presence of an enormous granite mass intruded into the Lower and 
Middle Huronian sediments, constituting a great batholith which seems 
to occupy many thousands of square miles in northern Wisconsin, and 
is represented by outlying remnants in the east end of the Gogebic range 
and southwest and east in Michigan. ‘The contributions consist of the 
descriptions of the geology of a number of districts and ranges, together 
with discussions and some revisions of the correlations of the pre- 

. Cambrian rocks of the region. 

An important conclusion arrived at by the authors is that the Symmes 
tions heretofore included in the Upper Huronian are separable into two 
groups by an unconformity of the first magnitude. For the upper one 
of the two the name Copps is proposed. 

-Part II: Limestone Mountain and Sherman Hill are outliers of 
Paleozoic dolomite found in Houghton County. The stratigraphic 
range is from the Cambrian to the Middle Devonian, including the 
Cambrian Upper Black River, Decorah, Upper Galena, Richmond, 
Niagaran, and Middle Devonian. Belief in the presence of the Devonian 
is based on a piece of float found on the hillside. The structure has 


- 86 REVIEWS 


been worked out in detail and includes numerous minor faults. The 
great importance of the work lies in the establishing of the Ordovician, 
Silurian, and Devonian seas in the region. The similarity of the 
Ordovician fauna of this region with that of Minnesota and Wisconsin 
indicates they were all part of the same invasion. Good faunal lists 


are given. 
A.C. Mick 


Maps and Sections to Accompany Report on Contributions to the 
Study of the Geology and Ore Deposits of Kalgoorlie, East Cool- 
gardie Goldfield. Perth: Geological Survey of Western Aus- 
tralia, Bulletin 69, Part III, 1917. 


The gold deposits at Kalgoorlie, Australia, are of two types, the 
gold-quartz veins and the gold-telluride deposits.. The bed rock of the 
region outside of some minor metamorphosed sediments consists of 
granite and amphibolite schists. It is with the last-named that the 
gold is usually found. Bulletin 69 consists of fourteen plates showing the 


areal and structural geology of the region. 
A. Cx Meck 


Eleventh Biennial Report of the State Geologist on the Mineral Indus- . 
tries and Geology of Vermont. By GrorGE H. PERKINS, eé¢ al. 
Burlington, 1917-18. Pp. 209, pls. 18, figs. ro. 


The present report consists of a number of contributions by different 
authors on the geology of the state of Vermont. These are as follows: 

I. “Physiography of Vermont,” by G. H. PErKins.—The discussion 
and description of the physiography is given in a popular though thorough 
style. The physiographic history, which is rather complex, is well 
summarized. The mountain areas include regions of complex igneous 
and metamorphic history. Most of the large rivers are old and ante- 
cedent in character. The author believes that most of the lakes of the 
state are glacial in origin. 

II. “The Ordovician Terranes of Central Vermont,’”’ by CHARLES 
H. RicHarpson.—The formations discussed are all pre-Trenton in age 
and include, from base upward, the Irasburg conglomerate, the Mem- 
phremagog slates, and Waits River limestone. A brief summary of the 
geologic history of the state is given. 

III. Post-Glacial Sea-Level Waters in Eastern Vermont,” by H. L. 
FAtrRcHILD.—Mr. Fairchild describes the post-glacial marine features 
found in the eastern part of the state, thus supplementing an earlier 


e 


REVIEWS 87 - 


article dealing with the western portion of the state. With regard to 
the crustal movements which such features involve, he states: 

“The earth’s superficial or crustal portion is sensitive to unequal loading, 
or differential pressures, and: it appears that northeastern North America 
was depressed by the weight of the long persistent ice sheets of the recent 
glacial time. With the removal of the weight by the melting of the ice body, 
the land rose. The amount of Post-Glacial land uplift, and the area affected 
seem to have been proportionate to the calculated thickness and extent of 
the latest ice sheet.” 

The article is accompanied by a map representing the Ors es 
amount of post-Glacial uplift as shown by isobases. 

IV. ‘“‘A Report of the Geological Work within the Rochester Ver- 
mont Quadrangle,” by W. G. FoyrE.—Rochester quadrangle lies within 
the heart of the Green Mountains, midway between the Champlain and 
Connecticut valleys. The structure is rather complex, with the bed 
rock chiefly pre-Cambrian. 

V. “The Terranes of Northfield, Vermont,’ by CuHaries H. 
RICHARDSON and SamuEL H. Camp.—The geology of Northfield is 
intricate. The terranes consist of a series of highly crumpled, folded, 
and faulted metamorphic rocks, dipping at high angles and cut by 
intrusive masses. The discovery of Ordovician fossils in the Northfield 
slate quarries and in the limestone and quartzose marbles, has made 
possible the accurate stratigraphic correlation of these beds and con- 
stitutes the most important result of the work. 

VI. “The Terranes of Roxbury, Vermont,” by CHARLES H. RICHARD- 
SON.—The work here was very similar to that at Northfield, important 
graptolite faunas being found forming the basis for the correlation of 
the beds of the area. The origin of the verd-antique marble, which is 
of commercial importance, is carefully studied. 

XI. ‘Eruptive Rocks at Cuttingsville, Vermont,” by J.W. EGGLESTON. 
_ —The general character of the main body of the igneous rock mass 
indicates that it is a composite stock. The order of eruption found was 
from the more basic essexite through the syenites to the more acid 
nordmarkite. In the complementary dikes the order was reversed, 
tinguaite dikes antedating the camptonite dikes. 

_ Other chapters are: VII. “Progress in Copper Mining and Milling,” 
by E. C. Jacops; VIII. “Progress in Talc Production,’ by E. C. 
Jacogps; IX. “The Lime Industry in Vermont,” by E. C. Jacoss; 
X. “Additional Corrections to the List of Altitudes in Vermont,” 
XII. “Provisional Report of the Areal and Structural Geology of the 
West Flank of the Green Mountain Range,” by N. C. Date; XIII. 


88 3 REVIEWS 


“The State Cabinet;”’ XIV. “Mineral Resources”’; XV. “Progress of 
Stream-Gaging in Vermont during the Two-Year Period Ending Sep- 
tember 30, 1918’’; XVI. ‘Records of Stream Flow for the Two-Year 
Period Ending September 30, 1918,” by the United States Geological 
Survey in co-operation with the State Survey. 

It may be noted that Vermont led in tale production in 1917, the 
grade of talc, however, being low. In the production of granite ao 


marble it leads, and in slate it is second only to Pennsylvania. 
A, Cy Men: 


An Annotated Index of Minerals of Economic Value, to Accompany 
a Bibliography of Indian Geology and Physical Geography. 
Compiled by T. H. D. la loucuz, M.A., F.G.S., Fellow of the 
Asiatic Society of Bengal. Calcutta: Geological Survey of 
India, 1918. Pp. 490. 

The purpose of the author is to furnish a guide to the literature on 
Indian minerals of economic importance and at the same time to indicate 
as concisely as possible the nature of the information given by each of 
the various writers. These notes (Part II) are supposed to be used in 
conjunction with the bibliography which forms Part I of the work. 
The minerals are arranged alphabetically, or frequently under group 
heads where they are chemically or economically related to one another. 

A. C. Mc 


The Geology of the Tuapeka District, Central Otago Division. By 

P. Marsuatt, M.A., D.Sc., F.G.S. New Zealand Department 

of Mines, Geological Survey Branch, Bulletin No. ro (N-S.). 

12 Oey DIS, 2 

This report consists of a description of the general geology and 
physiography of the region, together with its various cultural and natural 
features. The formations present include the pre-Jurassic Tuapeka 
series, which, so far as now known, may be of any age between pre- 
Cambrian and Jurassic, and the Waitahuna series of Upper Cretaceous 
and Lower Eocene age, together with some Pleistocene and Recent 
deposits. The Tuapeka series represents deposits of littoral sands and 
muds and the Waitahuna true marine sediments, with some associated 
volcanics. 

The chief economic resource of the region is gold, which occurs 
disseminated in conglomerate beds or as lodes. It is believed to be 
primary in the conglomerate. Other minor mineral resources include 


antimony, copper, cinnabar, and scheelite. 
A. C2Mickt 


For Everyone ~ 
who has 


studies in Minor I Folds 


By CHARLES E. DECKER | ||. Personal 


Hes 


iM - ‘The asc Ss purpose is to alnstvate by. diagrams ° 
{| and photographs a series of types of minor folds; 
to illustrate and study briefly a few minor folds in 
their relation to major ohes; to illustrate a larger 
number of small folds in the midst of horizontal 
|} or. gently dipping strata, showing their character- 
| istics, methods of origin, age, and relation to 
3 faults; and finally, to connect these minor . 
deformations, in so far as possible, with larger 
; movements, and show their significance, as 
ihe indicating the presence of compressional stresses tc 
in the rocks in the interim between the great 
periods of deformation and mountain-building. 


Remington Portable 


The most compact of typewriters— 
case only 4 inches high; weight, complete 
with case, only 11% 


The data on which these studies are based have - 


_ been secured from a narrow area south of Lake 
| Erie, extending from: Cleveland across north- 
}} eastern Ohio, northwestern Pennsylvania, and 
- into New York as far as Dunkirk. A few folds 


on Lake Ontario were studied, ‘both in northern 


|New York and southern Canada. A few folds 
' were studied in the folded areas of the Arbuckle 
Fi TES the Wichita Mountains pe OR lahama. 


pounds. 

Standard in every fea- 
ture, including standard 
keyboard. Simple, swift, 


_ easy to operate and does 


beautiful work—always. 


Price, complete with 
case, $60. 


Remington 


ee dees: Loi: $1.50, Canon $r.05 7 : C 
_——— ewriter Co. 
ad de hose 


374 Broadway N.Y... 


Branches Everywhere 


‘The University of Chicago Press 
Pye, = - - Illinois 


The Geography of Illinois 
ae By DOUGLAS C. RIDGLEY 


Professor of Geography in the Illinois State Normal University 
ys 


The author’s aim is to provide an authoritative study of the state, presenting in 
compact form the wide range of physical influences which make up the geographic 
environment in which men live. | 


The natural features and natural resources of the state are treated in some detail. 
- The great occupations of mankind—agriculture, mining, manufacturing, trans- 
portation, and trade—are discussed with sufficient fulness to give an adequate 
_ idea of their development and present importance within the state. The popula- 
tion census of 1920 for Illinois is treated in a final chapter. The book is designed 
to be of interest to the busy citizen who wishes to know his state as a unit in its 
present-day activities; to teachers and pupils who would know Illinois well enough 
_ to interpret other regions in comparison with the home state; to all who wish to. 
learn the reasons for the high rank of Illinois in many lines of human endeavor. 


CN _ Abounds ih ces maps, illustrations, and colored insert maps 
: xvita-+ 385 LES r6mo, cloth; $2.50, postpaid ee 


P THE UNIVERSITY OF “CHICAGO PRESS 
PEHICAGON = + : ) - ILLINOIS 


Microscope “‘S.M.” with synchronic “ae equipment. : 
rotation of the nicols ae 
abate 


3 


The complete assortment of . ur petro- 


some of the prominent featares of thei new ; model 
mmielpeecere are: 


a a ae centered. 


Anastigmatic Tube Analyzer which Tee a most crisp definit 


Special Illuminating Apparatus, whereby a wider range of ead is 
afforded. be . 


Enlarged Microscope Tube which accommodates eyepieces of 
whereby the field of view is ee 50% over other types of ins 


Ask for Catalog ILB (F) 


NUMBER 2 


JEOLOGY 


ALBERT JOHANNSEN, Petrology 
ROLLIN T. CHAMBERLIN, Dynamic Geology — 


ASSOCIATE EDITORS © : 


JOHN C. BRANNER, Leland Stanford Junior University 
RICHARD A. F. PENROSE, Jr., Philadelphia, Pa. 
WILLIAM H. HOBBS, University of Michigan 

_ FRANK D. ADAMS, McGill University — 

_ CHARLES K. LEITH. University of Wisconsin 
WALLACE W. ATWOOD, Clark University — 
WILLIAM H. EMMONS, University of Minnesota 
ARTHUR L. DAY, Carnegie Institution d 


-FEBRUARY-MARCH 1922 : 
SS TROUGHS OF THE ANTILLES - - - ~ Srepnen Tamer 8% 


\CTER OF THE STRATIFICATION OF THE SEDIMENTS IN THE RECENT 
cA OF, FRASER RIVER, BRITISH COLUMBIA, CANADA - W. A. JOHNSTON TTS) 


R CTURAL RELATION. OF THE PURCELL RANGE AND THE ROCKY 
NT AINS UO OUNINION ar eee heed one ote FRANCIS PARKER SHEPARD 130 


3 OF ONTOGENY IN THE STUDY OF AMMONITE EVOLUTION 
bd A. E. TRUEMAN 140 
V PHYTOSAUR | FROM THE TRIAS OF ARIZONA + = Maurice G. Ment 144 


A SHORT-BELLOWS, ROLL-FILM KODAK FOR DETAIL WORK IN THE 
BS * = By hae ais CHESTER K. WENTWORTH 158 


a es 2 = me aes = 3 Atrrep C. LANE 162 


e _ ALBERT JOHANNSEN 167 
DO Uo pe a) atone 
SLE Ree ee ieee ee aes ie Sle 
| THE UNIVERSITY OF CHICAGO PRESS 
5: pa ae CHICAGO, ILLINOIS, on A. 


x: 


: _ ‘THE CAMBRIDGE UNIVERSITY PRESS, Lonpon 
4 oe MARUZEN-KABUSHIKI- KAISHA, Toxvo, Osaka, Kyoto, Fukuoka, SENDAI 
: THE MISSION BOOK COMP ANT ADSANCHEL 


February-March 1922 


EDITED BY | 
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY | 


With the Active Collaboration of : a ges 


STUART WELLER oe ALBERT JOHANNSEN Nee: Ge fe 
‘Invertebrate Paleontology : Petrology ve 
EDSON S. BASTIN : ROLLIN T. CHAMBERLIN : 
Economic Geology _Dynamic Geology 
i : : = [7 me 


The Journal of Geology is published by the University of Chicago at the Uhivesttg of Chicice Bias 4 
5750 Ellis Avenue, Chicago, Illinois, semi-quarterly, on or about the following dates: February - 1, March 15, 
May 1, June 15, August 1, September 15, November 1, December 15. {| The subscription price is $4.00 
per year; the price of Soot copies is 65 cents. Orders for service of less than a half-year will be charged © oe 
at the single-copy rate. {| Postage is prepaid by the publishers on all orders from thé United States, — : 
Mexico, Cuba, Porto Rico, Panama Canal Zone, Republic of Panama, Dominican Republic, El Salvador, 
Bolivia, Colombia, Honduras, Nicaragua, Peru, Hawaiian Islands, Philippine Islands, Guam, Samoan — 
Islands, Shanghai. [ Postage is charged extra for Canada and for all other.countries in the Postal Unio 
as follows: 30 cents on annual subscriptions (total $4.30); on single copies 4 cents (totak 69 cents) 
{| Patrons are requested to make all remittances seine to The Leiesity of Chicago Press i in’ BEu 
or express money orders or bank drafts. ; 

The following are authorized to quote the prices indicated: ’ 

For the British Empire: The Cambridge University Press, Fetter Lane, London, E. C. as Veni 
subscriptions, including postage, 29s. 4d. each; single copies, including postage, 4s. od. each. 
For Japan and Korea: The Maruzen- Kabuchike Kaisha, 11 to 16 Nihonbashi Tori Sanchome, 

Tokyo, Japan. Yearly Spe ons, Hgts postage, Yen 9.00 each; nal copies, including 

postage, Yen 1.50 each. *, 

For China: The Mission Book Company, 13 North Szechuen Road, Shanghai. ea Sub- 
scriptions $4.00; single copies 65 cents, or their equivalents in Chinese money. Postage extra, if iq 
mailed direct outside of Shanghai, on yearly subscriptions 30 cents, on single copies 4 cents. ae 

Claims for missing numbers should be made within the month following the regular month of publi- “a 
cation. The publishers expect to supply missing eet free only when losses have been sustained in 
transit, and when the reserve stock will permit. or : 

‘Business correspondence should be addressed to The University of Chicago Press, Chicago, Ml : : 

Communications for the editors and manuscripts should be addressed to the Editors of Tar Joma 4 
NAL OF GEOLocy, the University of Chicago, Chicago, II. von 

Contributors are requested to write scientific and proper names ae particular care sane | in citations a 
to follow the form shown in the pages of the Journal. 

Because of the high cost of printing, this Journal has found it necessary to adapt the ‘olla rules: 
| 1. If articles exceeding 25 pages in length are accepted for publication in the Journal, the cost of com- : 
position beyond 25 pages shall be paid by the author or by the institution which he represents. at the current 
rates of the University Press. {[ 2. The cost of illustrative material, in excess of $1.00 per page for the 
article, beyond the cost of type which would have occupied the same space, shall be borne by the author. 
{| 3. ‘The cost of corrections in proof in excess of 10 per cent of the cost of first composition, will be 
charged to the author. { 4. While the cost of printing remains as now, the Journal cannot supply Te= — 
prints without charge, but reprints will be furnished authors at cost, if ordered in advance of publication, 


Entered as seenad. class matter, March 20, 1893, at the Post office: at Chicago, Ill, under the Act of March 3 ee 
ral Recep tence for mailing at special rate of ‘postage provided for in 1 Section 1193, Act of October 3, Tor7, authorized on 
uly 15, ror ‘. 


PRINTED (IN TI THE U.S.A. 


y 


Source Book for the Economic Geography 
of North America 


By CHARLES C. COLBY 


Assistant Professor of Geography, the University of Chicago 


The fundamental idea of the book is to make available to the busy teacher the 
material on this subject which is scattered widely through literature. ‘The course which 
has been given for a number of years at the University of Chicago has resulted in the 
collection and organization of the material most suitable for the purpose. 


The book contains short magazine articles and materials from Canadian, Mexican, 
and United States government publications. The selection of the material has been 
governed by the concept that there is a direct relation between the environment and 
the economic activities of the region. The author aims to survey the different sections 
of the country from the viewpoint of industrial opportunity. 


&vo, cloth; 419 pages, $4.00, postpaid $4.20 


THE LS A elle we CHICAGO PRESS 
CHICAGO - : ILLINOIS 


The Geography of Illinois 


By DOUGLAS C. RIDGLEY 


Professor of Geography in the Illinois State Normal University 


The author’s aim is to provide an authoritative study of the state, presenting in 
compact form the wide range of physical influences which make up the geographic 
environment in which men live. 


The natural features and natural resources of the state are treated in some detail. 
The great occupations of mankind—agriculture, mining, manufacturing, trans- 
portation, and trade—are discussed with sufficient fulness to give an adequate 
idea of their development and present importance within the state. The popula- 
tion census of 1920 for Illinois is treated ina final chapter. The book is designed 
to be of interest to the busy citizen who wishes to know his state as a unit in its 
present-day activities; to teachers and pupils who would know Illinois well enough 
to interpret other regions in comparison with the home state; to all who wish to 
learn the reasons for the high rank of Illinois in many lines of human endeavor. 


Abounds with drawings, maps, illustrations, and colored insert maps 
xvitit 385 pages, 16mo, cloth; $2.50, postpaid $2.65 


THE SS Bao oF CHICAGO PRESS 


CHICAGO - - ILLINOIS 


Cambridge Antversity Press 


The Study of Geological Maps. By Gerrrune L. Exres, D.Sc., Fellow of Newn- 
ham College, Cambridge. Crown 4to. With 7 plates, 64 figures,and a map. Price on 
application. Cambridge Geological Series. 


A Manual of Seismology. By CuHartEes Davison, Sc.D. With too illustrations. 
Demy 8vo. Price on application. 


Petrology for Students: an Introduction to the Study of Rocks under 
the Microscope. By Atrrep Harker, M.A., F.R.S., F.G.S. Fifth edition, revised. 
Crown 8vo. $3.75. 


A Treatise on Crystallography. By W. J. Lewis, M.A. Demy 8vo. $5.00. 


The Principles of Stratigraphical Geology. By J. E. Marr, ScD., F.R.S., F.G.S. 
Crown 8vo. $2.50. 


An Introduction to Geology. By J. E. Marr, Sc.D., F.R.S., F.G.S. Crown 8vo. 
$2.50. 


Published by the Cambridge University Press (England) 
The Macmillan Co., Agents in the United States 
64 and 66 Fifth Avenue, New York City 


THE ORIGIN OF THE EARTH 


By THOMAS C. CHAMBERLIN 


Head of the Department of Geology, The University of Chicago 
One of “The University of Chicago Science Series” 
xii+-272 pages, 8vo, cloth; $1.75 net, postpaid $1.90 


FROM THE PREFACE 


“Tn telling the story of this search for the mode by which the earth came 
into being, we have let the incidents that led the inquiry on from one stage to 
another fall in with the steps of the inquiry itself. It is in keeping with the 
purposes of this series of booklets that the motives which set researches going 
should have their place with the quests that arose from them... .. The final 
story of the birth of the earth will come only after a time, when the vestiges of 
creation have been more keenly discerned and more faithfully rendered than is 
possible now.” 


THE UNIVERSITY OF CHICAGO PRESS 
CHICAGO - - - : Sie ILLINOIS 


Starved Rock State Park 


and Its Environs 


By CARL O. SAUER 
Depariment of Geology, University of Michigan 


GILBERT H. CADY 
Illinois State Geological Survey 


HENRY C. COWLES 
Professor of Ecology, University of Chicago 


This thorough and interesting geographical 
study of the best-known feature of the Illi- 
nois Valley has been prepared for the benefit 
of those who are interested in knowing more 
about the region than can be gathered by 
personal observation in a few hours or a few 
days spent in the park. The physical geog- 
raphy of the park and its surroundings has 
a number of features, such as the beautiful 
little canyons, which are unusual in this part 
of the country. Part I is given up to the 
geography of the park, its surface features 
and their origin, and the exploration, settle- 
ment, and development of the region; Part 
II, to the geology of the park; and Part III, 
to its botany. 
130 pages, cloth, $2.00, postpaid $2.15 


THE UNIVERSITY OF CHICAGO PRESS 


CHICAGO - - - ILLINOIS 


The Geology of 


Vancouver and Vicinity 


By 
EDWARD M. J. BURWASH 


14 half-tones, 10 line drawings, and 2 colored maps 


A thorough and authoritative survey of the 
region, made more interesting by the re- 
markable photographs, which includea profile 


of the Grouse Mountain spur, the Lions from 
Mount Brunswick, a view of Red Mountain 
from Black Tusk Mountain, Mount Garibaldi 
from the south, a glacial section from a 
cutting on Keith Road, and others equally 
interesting. One colored map shows the 
topography of Vancouver range and the other 
is a geological map of Vancouver and vicinity. 


Ir2 pages, paper covers; $1.50, postpaid $1.60 


The University of Chicago Press 
Chicago - - Illinois 


The Geography of the Ozark Highland of Missouri 


By CARL ORTWIN SAUER, the University of Michigan 
(Publications of the Geographic Society of Chicago) 


The Ozark Highland of Missouri was selected for investigation because of its 
unusual wealth of geographic responses and because little is known concerning its con- 
ditions and possibilities. 

The purpose of such a study is twofold: to furnish an adequate explanation of 
the conditions of life in a given area, and to contribute proved statements which will 
aid in working out fundamental principles. 

The book is divided into three parts. The first is an outline of the environment. 
Only those things which are pertinent to an understanding of the conditions under which 
the people live are introduced. 


The second part considers the influence of smvinoanment on the settlement and develop- 
ment of the different parts of the highland. 


The third part is a study of the economic conditions as they exist today. In conclusion 
a forecast is offered of the lines along which the future of the region will be worked out. 


A valuable feature of the volume is the 44 figures in the text and 26 plates. 


xvitit246 pages, 8vo, cloth; $3.00, postpaid, $3.20 


The University of Chicago Press 
Chicago, Illinois 


Among articles to appear in early 
numbers of the Journal of Geology 
are the following: 


A New Occurrence of Crystobalite in California. By AvustIN F. 
ROGERS. 
Fault Features of Salton Basin, California. By JoHN S. Brown. 


A Criticism of the “‘Faunal Relationships of the Meganos Group,”’ 
by Bruce L. Clark. By Roy E. Dickerson. 


Former Courses of the Androscoggin River. By Irvinc B. Crossy. 


Notes on the Sand Dunes of Northwestern Indiana. By Grorce B. 
CRESSEY. 


Marine Upper Cretaceous and a New Echinocorys from the Alta- 
planicie of Bolivia. EDWARD W. BERRY. 


Post-glacial Lakes in the Mackenzie River Basin, Northwest 
Territories, Canada. ‘By A. E. CAMERON. 


The Occurrence of an Apus in the Permian of Oklahoma. By 
RUDOLF RUEDEMANN. 


On Contact Phenomena between Gneiss and Limestone in Western 
Massachusettes. By PENTTI ESKOLA. 


The Early Pre-Cambrian Formations of Northern Ontario and 
Northern Manitoba. By E. L. Bruce. 


The Reaction Principle in Petrogenesis. By N. L. BowEn. 


An Outline of the Physiographic History of Northeastern Ontario. 
By W. H. Co.t.ins. 


Dinosaur Tracks in Hamilton County, Texas. By W.E.WRATHER. 


VOLUME XXX NUMBER 2 


THE 


JOURNAL OF GEOLOGY 


February-March 1 922 


THE GREAT FAULT TROUGHS OF THE ANTILLES’ 


STEPHEN TABER 
University of South Carolina, Columbia, S. C. 


INTRODUCTION 


The region of the Greater Antilles is characterized by a marked 
elongation of the principal geographic features in an east-west 
direction and by extremely high relief (PlateI). The highest point 
above sea-level, Mount Tina on the island of Haiti, has an eleva- 
tion of over 3,100 m. while the deepest sounding in the Atlantic 
Ocean (8,526m.) was obtained only 320km. to the northeast.? 
The Antillean mountain ranges are among the most precipitous in 
the world, but their slopes are, in large part, submerged beneath 
the surface of the Atlantic Ocean and Caribbean Sea. Off the 
north coast of St. Croix the descent is 4,348 m. in a distance of 
8 km., and for shorter distances the slopes are very much steeper. 

The physiography of the Greater Antilles is very complex and 
as yet little of the region has been studied in detail. The dominant 
tectonic trend is approximately east and west along arcs convex 
toward the north, in part following the margins of the great trough- 
like depressions which are such a striking characteristic of the 

t Presented in part at the Chicago meeting of the Geological Society of America, 
December, 1920. 


2 All elevations and soundings have been taken from the maps of the Hydrographic 
Office of the United States Navy. 


89 


go STEPHEN TABER 


region. These arcs cut directly across the earlier tectonic lines and 
are everywhere marked by extremely precipitous slopes. 

The evidence outlined in the following pages indicates (1) that 
the east-west arcs delineating the major relief features of the Greater 
Antilles are zones of normal faulting developed in late geologic 
time; (2) that this faulting has resulted in the formation of the 
great troughs of the region; and (3) that the displacements are 
continuing at the present time. 


CRITERIA OF FAULTING 


The fault-zone hypothesis rests partly on geologic evidence, but 
largely, since most of the region is under water, on evidence that 
is either topographic or seismologic. Topographic evidence of 
faulting is unusually well preserved because the fault scarps are 
mostly below sea-level where they have been protected from erosion. 
The characteristics of these scarps which indicate an origin through 
faulting are (1) great height; (2) extremely precipitous slopes; 
(3) abrupt changes in slope at top and bottom; (4) rectilinear 
course for long distances; and (5) subordinate depressions and 
elevations (troughs and horsts) near the base of the main 
scarps. — 

The last-mentioned features develop within fault zones as a result 
of the differential displacement of long narrow blocks or wedges 
formed by the branching and intersection of nearly parallel faults. 
The formation of these subordinate troughs and horsts is a common 
accompaniment of normal faulting on a large scale and especially 
of trough faulting. On land these minor topographic effects of 
faulting are soon obliterated by erosion and by the accumulation 
of rock-waste in the troughs; therefore the significance of these 
criteria in the recognition and interpretation of faulting has been 
generally overlooked. Because of their short life these topographic 
features are seldom found on land except in arid regions and where 
very recent displacements have occurred along old fault zones. 
Beautiful examples of this type of fault topography on a small 
scale have developed along the Wasatch fault zone in Utah as 
a result of post-Pleistocene displacements. Similar topographic 
effects were produced in California by the faulting that accompanied 


JouRNAL OF GEOLOG 


PLATE I 


300 


= 


Miles 
Kilometers 


io 
me) 


se. 


=~ 


ST<MAATIN I. 
= 8 


Qiparsupa 1. 


AST. CHRisTpPHER 1. 


GUADELOUPE I. 
= 


i 
a 


’ 


XN 
SS 


23 


Jovrnat or Grotocy, VoL. XXX, No. 2 


_-- 


“. 
ace BROWN SON? 


ANEGADA TROUGH 


oo --- 


fC wes=-y 
Noes? 


‘ST. CROIX 1. 
=) 


oa 


: QiearBupA 1. 
2500 


AST. CHRISTPPHER 2, 


ee Q> 
SEA 


~~. maa 


“€ARIBBEAN 


Miles 
300 Kilometers 


Se _SUADELOUPE 1 Uf 
\ 


pus--------- 


~ 


Map oF THE GREATER ANTILLES, SHOWING FAULT TROUGHS 


Contour interval, 1,000 fathoms (6,000 feet). Areas within the dotted lines, less than 100 fathoms. 


THE GREAT FAULT TROUGHS OF THE ANTILLES QI 


the Owens Valley earthquake of 1872; and there are many 
examples along the line of the San Andreas fault zone of Cali- 
fornia. 

Seismologic evidence, when available, is especially valuable in 
locating active faults which are not exposed to direct observation. 
Many earthquakes have been recorded in the Greater Antilles 
and Virgin Islands during the last four centuries, and, while the 
published catalogues of these earthquakes are far from complete, 
it seems probable that few, if any, of the very destructive shocks 
have been omitted. Most of the weak shocks of a district originate 
in the same localities as the stronger ones, and therefore in this 
investigation attention has been focused on the destructive earth- 
quakes. In the region of the Greater Antilles the epicenters of 
destructive earthquakes, instead of being scattered at random, are 
almost entirely limited to a few well-defined belts, which also possess 
the topographic characteristics of fault zones. 

Most of the earthquakes occurred before seismographs were 
developed, and, therefore, in determining the location of their 
epicenters, it has been necessary to rely chiefly on the distribution 
of intensities and the evidence derived from a study of accompany- 
ing sea waves. Fortunately, for present purposes an approximate 
location of an epicenter is, in most cases, sufficient. The seismo- 
logic data are only briefly summarized here as they are discussed 
in more detail in another paper which will be published shortly 
in the Bulletin of the Seismological Society of America. 


DESCRIPTION OF FAULT ZONES AND FAULT TROUGHS 


The principal troughs of the Antillean region are: the Bartlett 
Trough, which lies between Cuba and Jamaica and extends from 
the Island of Haiti westward into the Gulf of Honduras; the 
Brownson Trough, lying immediately north of Porto Rico and the 
Virgin Islands; and the Anegada Trough, which separates Porto 
Rico and the Virgin Islands Bank from St. Croix and the Lesser 
Antilles. In addition to these there are some minor troughs and 
probably one fault zone that is not associated with any trough. 


*G. K. Gilbert, ‘‘Lake Bonneville,” U.S. Geol. Surv. Monograph I (1890), 
p- 361. 


Q2 STEPHEN TABER 


THE BARTLETT TROUGH* 

The Bartlett Trough is probably the most striking physiographic 
feature of the Antillean region (Fig. 1). It is a long, narrow trench 
stretching from the Gulf of Honduras eastward into Gonaive 
Gulf between the two western peninsulas of Haiti, a distance of 
15, or 1,570km. Its width, where widest, between Cuba and 
Jamaica and near the Cayman Islands is 150 to 160km. Its 
deepest sounding, 3,506 fathoms (6,412 m.) was obtained less than 
50 km. from the coast of Cuba and is over 8,275 m. below the 
higher peaks of the Blue Mountains in Jamaica and the Sierra 
Maestra in southern Cuba. The six deepest places (all over 3,000 
fathoms) are close to the inclosing scarps rather than near the center 
of the trough. The floor of the trough seems to be relatively flat 
over quite large areas, but in longitudinal profile rises and falls 
throughout its length. The Bartlett Trough lies between two 
great fault zones which may be traced longitudinally far beyond 
the limits of the trough itself. They are here designated the Swan 
Island—Jamaica—South Haiti fault zone and the Cayman Islands— 
Sierra Maestra—North Haiti fault zone. 

Swan Island—Jamaica—South Haiti Fault Zone—This fault 
zone may be traced westward from its juncture with the 
northern escarpment of the Caribbean Basin in the vicinity of 
Ocoa Bay, Santo Domingo, across Haiti, and along the southern 
side of the great Bartlett Trough. On the Island of Haiti it is 
marked by a trough-shaped valley 15 to 20 km. in width, which 
extends from Barahona on Neyba Bay to Port-au-Prince, a dis- 
tance of about 150km. Throughout its length this depression 
is confined between the precipitous fronts of two lofty mountain 
ranges. ‘The floor of the trough comprises the plain of the Cul de 
Sac in the west, the plains of Neyba in the east, and the lake region 
lying between. Two of the lakes contain salt water, and 
W. F. Jones states that they are both below sea-level, but at differ- 
ent elevations.? A very slight subsidence of the land would com- 


tA brief description of this trough was given in a previous paper, ‘‘ Jamaica 
Earthquakes and the Bartlett Trough,” Bull. Seis. Soc. of Amer., Vol. X (1920), 
pp. 84-88. 

2W. F. Jones, ““A Geological Reconnaissance in Haiti, A Contribution to 
Antillean Geology,”’ Jour. Geol., Vol. XX VI (1918), p. 730. 


THE GREAT FAULT TROUGHS OF THE ANTILLES 93 


pletely submerge the floor of the trough and separate the Tiburon 
Peninsula from the rest of Haiti. 

The depression has been described and figured by Jones as a 
normally faulted block, buried under Quaternary deposits which 
have been removed at one place exposing highly tilted Oligocene- 
Miocene sediments. Limestones of Eocene—Early Oliogcene age 
are exposed along the flanks of the inclosing ranges while late 
Tertiary intrusives form the core of the mountains on the south. 
Another “‘large fault is indicated well up on the south range of 
Haiti, a well-marked depression extending east and west along the 
course of this fault.’ 

Along the north side of the salt lakes Jones found basalt-flows, 
which he thinks came from fissure eruptions; and he states that — 
between the Cul-de-Sac depression and Ville Bonheur (Saut d’ Eau) 
there is a well-defined crater from which extend basalt-flows, 
so recent in origin that they have not been appreciably modified 
by erosion.? In the Province of Azua, Santo Domingo, there are 
rocks of igneous origin, which C. W. Cooke states are not older than 
Pleistocene.’ 

During historic times the fault trough of southern Haiti has 
been the locus of more earthquakes of high intensity than any similar 
area in the Greater Antilles. The high seismicity of this depression 
is noted by Scherer in his excellent article on “Great Earthquakes 
- in the Island of Haiti’* from which most of the data on Haitian 
earthquakes given in this paper have been abstracted. 

The cities of Azua and Santo Domingo, located on alluvial 
ground a short distance north of the fault zone and near its juncture 
with the Caribbean escarpment, have been damaged repeatedly 
by earthquakes. They suffered from severe shocks in 1673, 1684, 
and 1691, Azua being entirely destroyed in 1691. On October 
18, 1751, an earthquake threw down all houses in Azua and a 
sea-wave overwhelmed the town. It was rebuilt farther inland. 


Dbide Ps 75a 2 T[bid., pp. 750-51. 
3C. W. Cooke, ‘‘Geological Reconnaissance in Santo Domingo,” Bull. Geol, 
Soc. Amer., Vol. XXXI (1920), p. 218. 


4 Rev. J. Scherer, ‘‘Great Earthquakes in the Island of Haiti,” Bull. Seis. Soc. 
Amer., Vol. II (1912), pp. 161-80. 


94 STEPHEN TABER 


‘Other towns near the coast were severely damaged, Santo Domingo 
City losing many of its finest buildings. The earthquake of 1751 
and several other shocks, assigned by Scherer to the central valley 
of Haiti, are here correlated with the southern fault trough because 
of high intensities near the coast and the phenomena of the sea- 
wave. The earthquake of May 11, 1910, cracked walls in Azua 
and Santo Domingo City. Scherer states that “‘the strongest part 
of the earthquake occurred in the Bay of Ocoa where the sea-wall 
was broken.’ 

The fault trough apparently continues westward into the Gulf 
of Gonaive, for there is a marked depression between Gonaive 
Island and the straight north coast of the Tiburon Peninsula. The 
earthquake of November 9, 1701, threw down masonry houses on the 
plains near the western end of the trough, and the road along the 
north shore of the Tiburon Peninsula from Léogane to Petit Godve 
sank into the sea. The severe earthquakes of November 21 and 22, 
1751, destroyed the recently founded town of Port-au-Prince and 
overthrew buildings on the plain of the Cul de Sac. Lyell states 
that “‘part of the coast 20 leagues in length sank down and has ever 
since formed a bay of the sea,’” but the writer has found nothing 
which would confirm this assertion. 

The earthquake of June 3, 1770, was one of the strongest 
shocks recorded on the Island of Haiti, the area of greatest destruc- 
tion extending from Croix de Boquets through the plain of the 
Cul de Sac to Port-au-Prince and along the north coast of the 
Tiburon Peninsula as far as Mirago4ne. Southey states that 
“‘the sea rose a league and a half up into the island’’s but does not 
mention where this occurred. Scherer states that at Grand Goave 
the foot of the mountain of La Saline was partly submerged and 
at Arcahaie north of Port-au-Prince, a wave was also recorded. 

The earthquake on the night of April 8, 1860, originated a little 
farther west than the disturbances previously described, the inten- 


SING Me SIME, Os Cilio (De UD: 

2 Sir Charles Lyell, Principles of Geology, Vol. I, p. 440, London, 1830. 

3Thomas Southey, Chronological History of the West Indies, Vol. II, p. 407, 
London, 1827. 


4 Rev. J. Scherer, op. cit., p. 178. 


A 


THE GREAT FAULT TROUGHS OF THE ANTILLES 95 


sity being greatest from Petit Goave to Anse 4 Veau, but towns 
as far east as Port-au-Prince had houses thrown down or badly 
damaged. In the vicinity of Anse 4 Veau the sea withdrew and 
then broke with a crash on the shore. 

Soundings indicate that the fault trough of southern Haiti con- 
tinues westward as a topographic feature at least as far as the end 
of the Tiburon Peninsula; and, while it cannot be traced farther, 
the trend of the entire depression 1s in alignment with the southern 
scarp of the Bartlett Trough north of Jamaica. Evidence that 
the north coast of Jamaica is determined by a fault zone has been 
given elsewhere.* Briefly summarized, it is as follows: 

1. The coast is an almost straight line from Port Maria to 
Montego Bay, a distance of nearly 113 km., where it is offset about 
8 km. to the south and then continues westward to Pedro Point. 
The east-west line between Pedro Point and Montego Bay is con- 
tinued eastward by the valley of Montego River which runs parallel 
to the coast for 16 km. 

2. The land rises steeply from the sea to the plateau surface 
which has an elevation of 300 to 400m. Wave erosion could not 
have produced these bluffs in the relatively short time that it has 
been active, and there is no broad, wave-cut terrace either above or 
below sea-level. The steep slopes continue below the sea and 
depths of 1,000 to 2,251 fathoms (1,829 to 4,117 m.) are attained 
within less than 15 km. of the shore. 

3. There is a sudden change in slope both at the top and bottom 
of the escarpment. 

4. The Tertiary beds terminate abruptly along the coast and 
in places the uplift has exposed the older underlying rocks. 

5. The occurrence in modern times of several severe earth- 
quakes with their epicenters a short distance off-the coast indicates 
that there is here a zone of instability along which adjustments 
are still going on. The earthquake of 1692—one of the great catas- 
trophes of history—and the destructive earthquake of 1907 
both originated off the north coast of Jamaica and both were 
accompanied by sea-waves. 


1 Stephen Taber, “Jamaica Earthquakes and the Bartlett Trough,” Bull. Seis. 
Soc. Amer., Vol. X (1920), pp. 55-89. 


96 STEPHEN TABER 


Passing westward from Jamaica, soundings show that the steep 
southern escarpment of the Bartlett Trough continues toward 
Swan Island, and everywhere with abrupt changes in slope at top 
and bottom. Swan Island has the topographic characteristics of 
a horst which has remained standing within the zone of subsidence 
(see Fig. 1). 

Great Swan Island is only 2.5 km. in length and about 20m. 
in height; Little Swan Island is scarcely more than a reef. The 
submarine slopes in the vicinity of the islands are precipitous: 
on the south a sounding of 1,053 fathoms (1,926 m.) was obtained 


Swan I Misteriosa__Bank 


fo) 100 Kilometers 200 300 


Jamaica Cuba 
- Bartlett Trough 
fo) 100 Kilometers 200 500 


St Croix St.Thomas 


Anegada Trough 


(e} 20 Kilometers 40 60 


Fic. 1.—Profiles of the Bartlett and Anegada Troughs. Vertical and horizontal 
scales the same. Additional soundings would show the scarps to have steeper rather 
than gentler slopes. 


within 10 km., the descent continuing until a depth of 2,136 m. 
is reached 16km. from the island, after which the sea-bottom 
rises rapidly to the edge of the submerged Honduras-Jamaica 
Plateau; toward the north, a depth of 3,o10 fathoms (5,505 m.) 
is attained within 32 km. of the island, thus giving an average 
slope of about one in six. 

The latter sounding was obtained in a long, narrow depression 
lying at the foot of the scarp. The 3,000-fathom contour surround- 
ing the depression (see map, Plate I) extends eastward from Swan 
Island along the base of the scarp for 200 km. or more. It has an 
average depth, below the floor of the Bartlett Trough in this 
vicinity, of over 1,000m., the deepest sounding recorded being 


THE GREAT FAULT TROUGHS OF THE ANTILLES 97 


_ 3,206 fathoms (5,858m.). The general form of this depression, 
in so far as it has been revealed by soundings, and its alignment 
with the fault trough of southern Haiti, together with the presence 
of other depressions of unknown shape nearer Jamaica, suggest 
that there has been more or less trough faulting all along the arc 
from Ocoa Bay to Swan Island. 

Swan Island is built of limestone, mostly Pleistocene or Recent 
in age. A specimen from the higher part of the island consists 
chiefly of the coral, Orbicella cavernosa (Linn.). The particular 
form present in the specimen now lives in those waters and has 
not been found in deposits older than Pleistocene. In places the 
rock is a loosely cemented, coarse, calcareous sandstone. Pebbles 
and bowlders of hard limestone, containing terrigenous material, 
are abundant along the beaches. A specimen, showing well- 
defined, closely spaced bedding-planes and made up largely of 
Globigerina with amorphous silica filling the cavities, contained 
flakes of muscovite one millimeter in diameter and subangular 
quartz grains. Another specimen stained with ferric oxide, 
contained about 1.5 per cent ALO, as well as some quartz. The 
land mass from which these terrigenous materials were derived must 
have been close at hand, but it has entirely disappeared. 

It has not been possible to obtain seismologic data for Swan 
Island extending back over a long period of time, but during the 
summer of 1920 two shocks were felt.2, On July 11 at 10:40 A.M. 
a distinct rumbling was heard, and this was immediately followed 
by a pronounced shock, apparently vertical, lasting five seconds. 
On August 18 at 6:04 P.M. a rather violent earthquake of 
‘approximately fifteen seconds’ duration was experienced. The 
motion was reported as vertical and similar in character to that of 
the earthquake of July 11. No unusual sea conditions were 
noticed. Tools were dislodged in the engine-room and shop of the 

-t A set of rock specimens was procured from Swan Island through the courtesy of 
George S. Davis, General Manager of the Radio Telegraph Department of the United 
Fruit Company. The writer is also indebted to him for his kindness in furnishing 


seismologic data for Swan Island. The specimen of coral was identified by Dr. T. W. 
Vaughan, and the Globigerina by Dr. Jos. A. Cushman and Dr. Vaughan. 


2 These earthquakes were observed and reported by George H. Rogers, Chief 
Operator of the Swan Island Radio Station. 


93 STEPHEN TABER 


Radio Station, some falling from racks to the floor, but no damage 
resulted. The descriptions of these two earthquakes and the fact 
that they were not recorded on the seisomographs at Panama, 
Port-au-Prince, or Vieques indicate that they were local, probably 
originating close to Swan Island. j 

An earthquake at 115 o1™ 46° G.M.T. on January 1, 1910, was 
of sufficient intensity to throw two men out of bed. Instrumental 
records indicate an origin about 35 miles south of Swan Island 
but this determination is not regarded as very accurate. 

The submerged scarp bounding the Bartlett Trough on the south 
continues beyond Swan Island in a southwesterly direction and is 
especially well defined north of the Bay Islands (Utilla, Ruatan, 
and Bonacca) where it has a height of about 4,000m. The Bay 
Islands are probably the eastern continuation of the Sierra de Omoa 
of Honduras? and they are also in alignment with Swan Island. 
Basalt-flows on the Island of Utilla and in the Sierra de Omoa at 
Chameleconcito near Puerto Cortez are believed by Powers to be 
Recent, perhaps Pleistocene in age;3 and their presence is indica- 
tive of profound faulting such as has permitted the effusion of 
basalts in the fault trough of Southern Haiti. 

Topographic evidence suggests that the separation of Jamaica 
and Haiti has resulted from faulting along a branch of the Swan 
Island—Jamaica—South Haiti fault zone, extending from a point 
near the north coast of Jamaica eastward along the south coast of 
Haiti until it joins the northern escarpment of the Caribbean 
Basin. The Tiburon Peninsula of Haiti continues westward 
beyond Cape Dame Marie as a submarine ridge for 150 km., pass- 
ing through Navassa Island and Formigas Bank; a parallel ridge 
extends eastward from Morant Point, Jamaica, through Albatross 
Bank for a distance of over 100 km.; between the two lies a narrow 
channel having a nearly uniform depth of 2,000 m. Immediately 
south of the Tiburon Peninsula there are very steep slopes, espe- 
cially near Vache Island and near Jacmel where soundings of over 


« These data were supplied by Professor Harry Fielding Reid. 

2 Karl Sapper, ‘“‘Uber Gebirgsbau und Boden des siidlichen Mittelamerika,”’ 
Petermann’s Mitteil., Bd. 32, Heft 151 (1906), p. 17. 

3 Sidney Powers, “‘ Notes on the Geology of Eastern Guatemala and Northwestern 
Honduras,” Jour. Geol., Vol. XXVI (1918), p. 514. 


THE GREAT FAULT TROUGHS OF THE ANTILLES 99 


3,932 m. have been obtained within 20 km. of the coast. There 
is no evidence of important seismic activity along this branch 
zone east of Jamaica, and therefore, if the channel be due to fault- 
ing, as seems probably, a condition of at least temporary stability 
has been attained. 

The Cayman Islands—Sierra Maestra—North Haiti Fault Zone.— 
This fault zone extends along the north side of the Bartlett 
Trough, crosses the Island of Haiti between Manzanillo Bay and 
Samana Bay, and, in the vicinity of Mona Passage, joins the fault 
zone which forms the south side of the Brownson Trough. 

The Sierra Maestra range, after closely following the straight 
east-west coast of southern Cuba, continues westward as a sub- 
marine ridge, which, bearing slightly toward the south, reappears ~ 
above the surface in the Cayman Islands and almost reaches the 
surface in the Misteriosa Bank. It culminates in the Pico del 
Turquino which rises abruptly from the sea to an altitude of 
2,000m. ‘The precipitous southern slope of the range is continued 
below sea-level to depths of over 6,000 m., forming one of the most 
magnificent fault scarps known. At the base of this scarp, which 
forms the north side of the great Bartlett Trough, there is a narrow 
subordinate trough or series of elongated depressions containing 
four of the deepest soundings obtained in these waters; and it is | 
interesting to note that these soundings are located opposite points 
where the top of the scarp is relatively high. The four soundings 
are: 5,501 m. immediately southeast of Misteriosa Bank; 6,269 m., 
32 km. south of the Cayman Islands; 5,724 m., 35 km. south of 
Cape Cruz; and 6,412m., 50km. south of Turquina Peak. In 

‘passing south from Turquina Peak there is a precipitous descent 
to something more than 2,286 m. below sea-level, and then an ascent 
to 1,582 m. below sea-level before the descent to 6,412 m. is made 
(see profile, Fig. 1). Farther east there are other submerged peaks 
or ridges near the base of the scarp. 

Vaughan has referred to the precipitous profiles along the south 
shore of Cuba as indicative of faulting and states that this interpre- 
tation is supported by the geologic structure." The La Cruz marl 


tT. W. Vaughan, “‘Some Littoral and Sublittoral Physiographic Features of 
the Virgin and Northern Leeward Islands and Their Bearing on the Coral Reef 
Problem,”’ Jour. Wash. Acad. Sci., Vol. VI (1916), p. 56. 


smele) STEPHEN TABER 


(Miocene) is abruptly cut off by faulting at the shore line near the 
mouth of Santiago Harbor.* 

Seismologic data are not available for the Cayman Islands, and 
because of sparse population little is known about the effects of 
earthquakes along the Sierra Maestra scarp except at Santiago de 
Cuba. This city, founded in 1514, has been repeatedly damaged 
by earthquakes.? Strong earthquakes were recorded at Santiago 
in 1578,1675,and 1677. Kimball states that the city was destroyed 
by the shock of 1675,3 but the writer has not been able to verify 
.this statement. The earthquake of February 11, 1678, known in 
Cuban tradition as the great earthquake, caused enormous destruc- 
tion in Santiago; and exactly one year later the cathedral was 
destroyed by another shock. The severe earthquake of 1755 was 
accompanied by a sea-wave which almost completely inundated 
the town. The strongest earthquake recorded at Santiago, accord- 
ing to Salterain, occurred June 11, 1766, and it was followed by a 
large number of aftershocks. Many buildings were completely 
destroyed and others were badly damaged. Between 1777 and 
1852 eighteen important earthquakes are listed by Salterain, the 
earthquake of August 20, 1852, and its aftershocks being especially 
severe. 

In contrast with the Sierra Maestra region the north and central 
parts of Cuba have been virtually free from seismic disturbances. 
A severe earthquake followed by aftershocks originated in the 
Sierra de los Organos of western Cuba in 1880, but, before that 
time and since, earthquakes have been almost unknown in that 
section. 

Passing eastward from the Bartlett Trough the fault zone is 
marked by a depression which obliquely crosses the Windward 
Passage and extends between Tortuga Island and the coast of 
Haiti. The depression enters Haiti through Manzanillo Bay, and, 

tT. W. Vaughan, “‘Geological History of Central America and the West Indies 
during Cenozoic Time,” Bull. Geol. Soc. Amer., Vol. XXIX (1918), p. 626. 


2 Most of the facts concerning Cuban earthquakes given in this paper have been 
abstracted from ‘“‘Ligera Resefia de los Temblores de Tierra Occuridos en la Isla de 
Cuba” by P. Salterain, Boletin de la Comisién del Mapa Gedlogico de Espana, Vol. X, 
Pp. 371-85, Madrid, 1883. 

3R. B. Kimball, Cuba and the Cubans, p. 20, New York, 1850. 


THE GREAT FAULT TROUGHS OF THE ANTILLES IOI 


as a land valley, crosses the island to Samana Bay, a distance of 
240km. ‘This great valley was named the Véga Real by Columbus 
in 1494, but now the name is usually limited to the eastern portion. 
It is very similar topographically to the great valley of southern 
Haiti. ‘The floor of the valley, less than 25 km. in width, consists of 
a series of broad plains lying between the Sierra de Monte Christi 
on the north and the Grande Hilera on the south. The valley is 
drained by two rivers, the Yaqui del Norte and the Yuma, flowing 
in opposite directions. 

The geologic structure of the valley has not been studied in 
detail. Cooke,’ who has recently visited the region, states that’ 
the south front of the Cordillera Sententrional (Sierra de Monte 
Christi) is a fault scarp which brings up Eocene and Cretaceous 
rocks high above ‘upper Miocene. The displacement is therefore 
in the same direction as along the Sierra Maestra fault. Cooke 
found several faults of slight displacement along the Gurabo River 
near the southern side of the Véga Real and states that these are 
probably normal, with uplift on the north. There are other faults 
of much greater magnitude near the southern side of the Véga 
Real and in the adjacent part of the Cordillera Central, but no 
data are now available concerning them. Cooke believes that 
faulting has been an important factor in shaping some of the 
boundaries of the Véga Real, but he does not regard the valley as a 
_ simple down-dropped fault block. 

The high seismicity of the great northern valley of Haiti was 
recognized by Scherer, who states that the first great earthquake 
- mentioned in histories of Haiti occurred here in 1564. It resulted 
in the destruction of Conception de la Véga and Santiago de los 
Caballeros, two cities in the Véga Real about 30 km. apart. In 
1783 the principal church at Santiago was partly thrown down by 
an earthquake and several buildings were destroyed. 

The earthquake of May 7, 1842, was one of the most severe 
recorded on the island. Destruction was greatest at Cap Haitien, 
a city of ten thousand inhabitants, which lost half of its population. 
The intensity was almost equally as high at Méle St. Nicolas, 


t Dr. C. W. Cooke kindly furnished information concerning the geologic structure 
of the Véga Real in letters dated January 26, and October 5, 1920. 


102 STEPHEN TABER 


Port de Paix, Fort Liberté and Santiago de los Caballeros, all of 
which were destroyed with the loss of several thousands of lives. 
At Cap Haitien waves dashed against buildings along the quay; 
at Port de Paix the sea withdrew 60 m. and, upon returning, buried 
the city under 4 or 5m. of water. The bed of the Yaqui River is 
said to have been suddenly elevated, driving its waters both up 
and down stream. ‘The shock was severe at Santiago de Cuba and 
was felt throughout the length of the Sierra Maestra. The dis- 
tribution of intensities and the phenomena of the sea-wave indicate 
that the earthquake was caused by a vertical displacement along a 
fault passing between Tortuga Island and St. Nicholas Peninsula. 

The earthquake of December 29, 1897, seems to have originated 
in the central part of the Yaqui Valley, the intensity being greatest 
between Guyaubin and Santiago and at Altamira. At Guyaubin 
and Santiago great cracks were formed and subsidence of the ground 
was reported. 

Scherer correlates the disastrous earthquake of September 23, 
1887, with the great northern valley, but the writer thinks that it 
was caused by a vertical displacement along a fault near the east. 
end of the Bartlett Trough, a short distance south or southwest. 
of Méle St. Nicolas. The destruction was greatest at Mole St. 
Nicolas, nearly all of the houses being thrown down. Buildings 
were also damaged or destroyed at Cap Haitien, Port de Paix, 
Gonaives, and other places. A sea-wave followed the earthquake 
and was reported from points along the shores of Gonaive Gulf 
as far as Anse d’Hainault on the end of the Tiburon Peninsula; 
at Jérémie on the north coast of the Tiburon Peninsula the sea 
withdrew 20 m. and returned with a rush. Along the north coast 
of the St. Nicholas Peninsula, however, the wave seems to have 
been of no importance. The shape of the St. Nicholas Peninsula 
suggests that both its north and its south coasts may have been 
determined by faulting. 

Three other Haitian earthquakes of moderately high intensity 
cannot be correlated with either of the great fault valleys. One 
originated in the eastern part of the island in 1882 and damaged 
the churches at Seybo and Higuey; another, which occurred 
October 6, 1911, was of highest intensity in the central part of the 


THE GREAT FAULT TROUGHS OF THE ANTILLES 103 


island and probably originated on the south flank of the Grande 
Hilera or in the central valley of Haiti. This valley, according to 
Jones, was formed by folding and erosion rather than by faulting. 
The earthquake of April 23, 1916, which damaged buildings at 
Boya, Guerra, and Bayaguana, seems to have originated near 
latitude 18° 45’ N., longitude 69° 45’ W. 

The separation of Cuba and Haiti has been attributed by 
Vaughan to the downthrow of a block between two faults; one 
forming the north side of the Bartlett Deep, the other forming the 
south side and converging toward the former in the Windward 
Passage.? The evidence summarized in the present paper indicates 
that the separation of the two islands is to be correlated with the 
formation of the series of depressions, including the Véga Real, 
which mark the Cayman Islands-Sierra Maestra—North Haiti 
fault zone, and which have probably originated through differential 
displacement of relatively narrow blocks or wedges within that zone. 

A narrow trough, 3,058 to 4,353 m. in depth, separates the 
Island of Haiti from the Bahama Banks. It extends eastward, 
in a great arc convex toward the north, from a point north of the 
Windward Passage and enters the Brownson Trough northeast of 
the Samana Peninsula. The topographic characteristics of this 
trough and its parallelism to the two Bartlett Trough fault zones 
suggest that it has originated through faulting; but no geologic 
- evidence of faulting is known and little seismologic evidence is 
available except for the extreme eastern end where it joins the 
Brownson Deep (see page 106). 


\ 


THE BROWNSON TROUGH 


The Brownson Trough, containing the deepest sounding made 
in the Atlantic Ocean, parallels the Porto Rico—Virgin Islands ridge 
on the north. Its shape is not known in detail for as yet few 
soundings have been made in these waters. The 4,o00o-fathom 
contour surrounds a narrow area, about 320 km. long in an east- 
west direction; and the 3,o00-fathom contour which is approxi- 


tW. F. Jones, op. cit., Jour. Geol., Vol. XXVI (1918), pp. 735-36 and Plate V. 
2T, W. Vaughan, ‘‘Geologic History of Central America and the West Indies 
During Cenozoic Time,’ Bull. Geol. Soc. Amer., Vol. XXIX (1918), pp. 625-26. 


104 STEPHEN TABER 


mately concentric extends nearly three times as far. The trough 
is deepest near its western end. 

On the south the trough is bounded by a great scarp that rises 
steeply to the plateau-like ridge on which Porto Rico and the 
Virgin Islands stand, the average slope being about one in thirteen. 
The rectilinear north coast of Porto Rico and the steep submarine 
scarp descending from it are indicative of faulting. This coast is 
bordered, near the west end of the island, by a long line of high 
cliffs which evidently represent a fault scarp, for the youthful 
topography of the plateau back of the cliffs and the gorge-like 
valley of Guajataca River near its mouth testify to a recent eleva- 
tion of the land, whereas a long period of time would be required 
for the sea to cut such cliffs in gently dipping rock strata. A 
wave-cut bench at the foot of the cliffs marks the latest uplift of 
the land, amounting to several meters; and a similar sea-terrace 
bordering Desecheo Island proves that this recent elevation 
extended at least that far westward.t | The abrupt change in the 
slope near the 1oo-fathom contour about 8 km. north of the sea- 
cliffs and the occurrence of earthquakes at points lower down along 
the slope indicate the presence of a zone of faulting, and suggest that 
the descent into the trough is not accomplished by a single fault 
scarp. 

The fault zone along the south side of the Brownson Trough is 
possibly the eastward extension of the Cayman Islands—Sierra 
Maestra—North Haiti fault zone, but if so, there is a sharp flexure 
in the trend of the zone immediately north of Mona Passage, and 
the downthrow changes to the opposite side. The topography 
of this critical region, in so far as it has been revealed by soundings, 
suggests that the ends of the two zones overlap and intersect; and 
this may explain the origin of the peculiar submarine valley, which, 
heading in Aguadilla Bay, extends northwestward into the Brownson 
Trough. This valley was described and figured in a previous 
paper. At Aguadilla the coast has been slowly sinking during the 
last half-century, while at Mayagiiez, 23 km. farther south as well 


t Harry Fielding Reid and Stephen Taber, ““The Porto Rico Earthquakes of 
October-November, 1918,” Bull. Seis. Soc. Amer., Vol. EX (1919), p. 120. 


2 Tbid., pp. 118-21 and Plates 13 and 14. 


THE GREAT FAULT TROUGHS OF THE ANTILLES 105 


as along the north coast the land has been rising. The seismicity 
of the region immediately north of Mona Passage is high. 

Many earthquakes have been recorded in Porto Rico and the 
Virgin Islands.*| Most of them have had a low intensity and have 
been reported from only one or two places, so that it is impossible 
to determine accurately their epicenters, but it is usually possible 
to locate, approximately at least, the epicenters of the stronger 
shocks. If consideration is limited to earthquakes which have had 
a probable maximum epicentral intensity of above VI in the 
Rossi-Forel scale, it is found that, with very few exceptions, they 
have originated along the steep slopes descending into the Brownson 
or the Anegada troughs. 

The earthquake of April 16, 1844, which damaged buildings at 
Isabela on the north coast of Porto Rico, probably had its origin 
a short distance north or possibly northwest of the island. The 
shock of November 28, 1846, was most strongly felt in the north- 
western part of Porto Rico where some buildings were injured, the 
distribution of the intensity indicating an origin off the northwest 
coast. The earthquake of October 11, 1915, which was felt over 
most of Porto Rico and as far west as Puerto Plata, Santo 
Domingo, probably originated a short distance north of Mona 
Passage. 

The destructive earthquake of October 11, 1918, with its 
accompanying sea-wave, and the strong aftershocks of October 
18 and 24 and November 12, as well as a host of weaker shocks felt 
during 1918-10, all originated a few kilometers west of Point 
Borinquen on the northwest coast of Porto Rico. 

Other earthquakes have originated at points farther east along 
the southern scarp of the Brownson Trough. On the night of 
December 8, 1875, an earthquake, which probably had its epicenter 
a short distance north of the coast, damaged buildings in Arecibo. 
The earthquake of September 27, 1906, having an epicentral 


t A catalogue of earthquakes felt in Porto Rico and the Virgin Islands from 1772 
to 1918 was given in “The Porto Rico Earthquake of 1918 with Descriptions of Earlier 
Earthquakes. Report of the Earthquake Investigation Commission,” by Harry 
Fielding Reid and Stephen Taber, Document No. 208 U.S. House of Representatives, 
66th Congress, 1st Sess. (1919), pp. 53-06. 


106 STEPHEN TABER 


intensity close to IX R.-F., originated about 50 km. north of the 
coast and opposite the middle of the island; and the shock on 
September 5, 1908, with slightly lower intensity seems to have had 
its origin in approximately the same locality. On July 7, 1860, 
two light shocks were felt on board the ship Esther and Sophie 
when about 20km. north of Culebra Island. The earthquake of 
February 17, 1909, felt over the greater part of Porto Rico and the 
Virgin Islands, originated along the steep submarine slopes north 
of Culebra and St. Thomas. On October 29, 1886, an earthquake 
was felt on board the British brigantine Wilhelmina while over the 
steep scarp 75 km. north of Anegada Island. 

The north side of the Brownson Trough is entirely under water 
few soundings have been made in its vicinity, and it is so far from 
the land that earthquakes originating along its slopes could cause 
little or no damage; therefore evidence of faulting is not abundant. 
A broad ridge rising 3,200 to 5,600 m. above the floor of the trough 
separates it from the North Atlantic Basin. The trough-like 
depression separating Haiti from the Bahama Banks apparently 
joins the north slope of the Brownson Trough at a slight angle, in 
much the same way that the Cayman Islands-Sierra Maestra— 
North Haiti fault zone joins the fault zone on the south side of the 
Brownson Trough. 

On February 19, 1883, the bark Siddartha experienced a sharp 
earthquake, lasting 25 seconds, while over the north side of the 
Brownson Trough in lat. 20°04’ N., long. 67° 41’ W. The bark 
trembled as if dragging over a hard bottom although the depth 
here is more than 6,000 m. 

On November 29, 1916, and on July 13 and 26, 1917, severe 
earthquakes originated in the vicinity of lat. 10° 30’ N., long. 
68° 30’ W. The last of these shocks was felt over most of Haiti 
and Porto Rico and probably had an intensity of IX R.-F. near the 
epicenter. It was followed by a series of aftershocks lasting several 
days. ‘These earthquakes of 1916 and 1917 should be correlated, 
perhaps, with the trough that separates Haiti from the Bahama 
Banks for they originated in the area where it joins the Brownson 
Trough (see page 103). 


THE GREAT FAULT TROUGHS OF THE ANTILLES 107 


THE ANEGADA TROUGH 


The Anegada Trough, separating Porto Rico and the Virgin 
Islands group from St. Croix and the Lesser Antilles is the deepest 
of the many passages connecting the Caribbean Sea with 
the Atlantic Ocean. A description of this trough accompanied 
by a map and transverse profile was given in a previous 
paper.* 

It is deepest midway between St. Croix and Vieques where a 
sounding of 2,501 fathoms (4,574 m.) was obtained; here it extends 
nearly east and west for too km. Between St. Croix and the islands 
of St. Thomas and St. John the depth is over 2,000 fathoms. 
Farther east the trough rises until the depth is a little over 1,000 
fathoms and there it bifurcates. One branch seems to extend 
northeast and join the Brownson Trough about 60 km. northeast 
of Anegada Island, where the depth is over 3,000 fathoms; the 
other extends eastward in the direction of St. Martin but is not 
well defined. In its deeper parts the trough is about 4o km. in 
width between the tops of the inclosing scarps, which show abrupt 
changes in slope at both top and bottom. The floor of the trough 
is relatively flat. 

The south side of the trough near St. Croix and westward 
therefrom is bounded by a tremendous fault scarp (see profile 
Fig. 1). Near Harms Bluff at the northwest corner of St. Croix 
the scarp descends 4,348 m. in 8km., an average slope of 30°. 
For a distance of 4.4 km. the slope averages over 37°, and for shorter 
distances it is much steeper. Vaughan states that the faulting has 
taken place so recently that the sea has barely cut a niche into the 
fault plain The northern scarp of the trough does not touch the 
coast of any of the islands; it lies 15 km. south of St. Thomas and 
6 or 7 km. from Vieques. It is not so precipitous as the opposing 
scarp, the average slope being about 12°, though in places it is much 
steeper. 


« Harry Fielding Reid, and Stephen Taber, “The Virgin Islands Earthquake of 
1867-1868,” Bull. Seis. Soc. Amer., Vol. X (1920), pp. 20-25. 

2T. W. Vaughan, ‘“‘Some Features of the Virgin Islands of the United States,” 
Asso. Amer. Geog. Ann., Vol. [IX (1920), pp. 78-82. 


108 STEPHEN TABER 


Several severe earthquakes have originated in the Anegada 
Trough and many light shocks are recorded by the seismographs on 
the Island of Vieques. <A severe earthquake was felt in Antigua, 
St. Christopher, and Tortola July 11, 1785. Since the shock was 
strongly felt on all three islands its epicenter was probably between 
them; and the report that cracks were formed in the ground in 
Tortola suggests an origin in the vicinity of that island. 

A severe shock was felt in St. Thomas, April 20, 1824. The 
precise location of the epicenter cannot be determined from the 
data available but is not far from St. Thomas, probably in the 
Anegada Trough, though possibly on the opposite side of the island 
along the southern scarp of the Brownson Trough. The strong 
shock reported from St. Thomas September 19, 1853, and the two 
felt on May 12, 1865, appear also to have originated near that 
island. 

The great earthquakes of November 18, 1867, originated along 
the northern scarp of the trough 15 to 20 km. south of St. Thomas. 

.There were two strong shocks separated by an interval of about 
to minutes, both being followed by sea-waves. Aftershocks 
continued for several months, the strongest occurring December 
rt and 12, 1867; January 5 and March 17, 1868, and September 
17, 1869. The shock on March 17, 1868, was accompanied by a 
small sea-wave and seems to have had its epicenter a little farther 
west than the others, though all originated along the same scarp. 

On July 24, 1913, a severe earthquake was felt in the Virgin 
Islands and throughout Porto Rico except along the west coast. 
The pute aa was apparently in the Anegada Trough near lati- 
tude 18° 30’ N., longitude 64° W. 


THE NORTHERN ESCARPMENT OF THE CARIBBEAN BASIN 


The Caribbean Basin is bounded on the north by an escarpment 
that descends steeply from Porto Rico and Haiti to a depth of 
4,572 m. within 90 km. of the islands. This is the only important 
scarp in the region that does not form one side of a relatively narrow 
trench. It gives the Porto Rico—Virgin Islands ridge, extending 
eastward from Haiti, the topographic characteristics of a great 
horst. 


THE GREAT FAULT TROUGHS OF THE ANTILLES imeye) 


Near Ponce on the south coast of Porto Rico, Berkey has 
traced an east-west fault which he thinks is of very late Tertiary 
age.t It brings the older rocks of the pre-Tertiary into contact 
with the younger series forming the present coastal margin. This 
fault may be one of a series by which the descent to the floor of 
the Caribbean Basin is accomplished. ‘There is, however, no topo- 
graphic evidence of recent displacements along it; and the Seismic 
history of the region indicates that, if the Caribbean scarp be due 
to faulting, it is now relatively stable. The epicenter of no earth- 
quake can be definitely assigned to this scarp, although some of 
the shocks reported only from the south coasts of Porto Rico and 
Haiti may have originated there. Certainly no disastrous earth- 
quakes and comparatively few light shocks have occurred there in 
several centuries. 

On August 4 and 13, 1908, moderately strong shocks were felt 
at Ponce and neighboring points. The intensity was about 
VIR.-F., and some buildings were slightly damaged. The ori- 
gin was probably not far from Ponce. 

Earthquakes of low intensity have originated at many places 
in Porto Rico, but in contrast with the north coast none has had 
an intensity of more than VI or VII. A shock on August 30, 1865, 
which slightly damaged churches at Manati and Ponce, probably 
had its origin near the center of the island. On October 23, 1860, 
an earthquake with intensity of VI-VII caused some damage at 
Mayagiiez on the west coast, and the epicenter was probably not 
far away. Several light shocks have originated in the eastern part 
of Porto Rico near Cidra, the strongest, on October 22, 1901, having 
an intensity of about VI. 


RELATION OF FAULT ZONES TO EARLIER TECTONIC TRENDS 


The fault zones described in this paper collectively form’ a 
great fault system which extends from the eastern end of the 
Brownson Trough westward to the Gulf of Honduras, a distance 
of 2,500km. The two fault zones inclosing the Bartlett Trough 
are apparently in alignment with the Matogua Valley and the 


tC. P. Berkey, “‘Geological Reconnoissance of Porto Rico,” Ann. N.Y. Acad. 
Sci., Vol. XXVI (1915), p. 4o. 


IIO STEPHEN TABER 


Polochie—Lake Izabal Valley in Guatemala, but Powers who 
has recently visited the region states that he believes these 
valleys are due to erosion dependent on folding. Sapper,? in 
mapping the tectonic lines of Central America attributes the 
east-west ranges of Guatemala to folding, and this view has been 
adopted by other geologists. If the Antillean fault zones continue 
westward beyond the Gulf of Honduras their curvature is reversed, 
for in Central America the mountain ranges extend in approximately 
parallel arcs convex toward the south. 

The Central American ranges, as they approach the Gulf of 
Honduras, curve gradually toward the northeast, and therefore 
meet at an angle the north coast of Honduras which extends east 
and west. Powers states that this coast is evidently a fault-zone 
area. Vaughan correlates the tectonic lines of Honduras with 
the submarine ridge or plateau connecting the Honduras Peninsula 
with Jamaica, but this feature is abruptly terminated by the fault 
scarp north of Jamaica. The tectonic lines of Central America, 
according to Sapper, are of several different ages; some having 
been formed in the Paleozoic, others as late as the Tertiary. Possibly 
some of these lines continue into the West Indies with curvature 
convex toward the south while others turn gradually southward 
and become convex toward the north. Faulting in Central 
America may have followed along the old structural lines due to 
folding. The location of Guatemala in a belt of high seismicity 
that may be traced from the Greater Antilles through Central 
America and Mexico into California, together with the precipitous 
topography of the region suggest that the great Antillean fault 
system may extend into Central America, and, gradually curving 
toward the north, continue into California. These questions can- 
not be decided, however, until more field data are available. 

The Sierra de los Organos, which form the backbone of western 
Cuba, have a northwest-southeast trend and may be genetically 


t Sidney Powers, written communication dated September Io, 1920. 
2 Karl-Sapper, op. cit. 
3 Sidney Powers, written communication dated September 10, 1920. 


4T. W. Vaughan, “‘Geologic History of Central America and the West Indies 
during Cenozoic Time,” Bull. Geol. Soc. Amer., Vol. XXTX (1918), p. 618. 


THE GREAT FAULT TROUGHS OF THE ANTILLES EADIE 


related to the tectonic lines of Central America and southeastern 
Mexico. 

Another group of tectonic lines, prominent in the Greater 
Antilles, trends northwest and southeast. The principal mountains 
of Jamaica, composed of intensely folded Cretaceous rocks trend 
about N. 70° W. parallel to the northeast coast; the Grand Hilera 
or Sierra Cibao extend across the Island of Haiti in a northwest- 
southeast direction from a point near Cape Engano to Méle St. 
Nicolas, and the axis of elongation of Cuba from the Windward 
Passage almost to Habana also follows this trend. Vaughan infers 
that a fault runs northeast from Cape Cruz but the evidence is 
meager.* 

Another old structural direction, which seems to have no relation 
to the present topography, is indicated by the north-south strike of 
the crystalline rocks exposed on the Isla de Pinos lying south of 
western Cuba.? 

The Antillean fault zones cut across the earlier trend lines which 
seem to be due chiefly to folding, and, because of their recency as 
well as the magnitude of the displacements, they dominate the 
present topography. ‘Transverse faulting seems to be rare, and 
it has had little or no effect on the topography. ‘The irregular 
coast lines at the east and west ends of the various islands are in 
marked contrast to the rectilinearity of many of the coast lines that 
run east and west. 

Berkey has suggested that Porto Rico is a large fault block 
uplifted along the southern margin and tilted toward the north 
with breaks at each end. In support of this hypothesis he mentions 
the unsymmetrical position of the main drainage divide, which is 
nearer the south coast; the comparatively abrupt termination of 
the island at both ends; and the absence of the younger limestone 
margin from these coasts although it is fairly continuous along the 
north coast and about half of the south coast. 


tT. W. Vaughan, ‘Fossil Corals from Central America, Cuba, and Porto Rico, 
with an Account of the American Tertiary, Pleistocene, and Recent Coral Reefs,” 
U.S. Nat. Mus. Bull. 103 (1919), pp. 288-89. 

2C. W. Hayes, T. W. Vaughan, and A. C. Spencer, ‘‘Report on a Geological 
Reconnaissance of Cuba,”’ Washington, 1go1. 


3C. P. Berkey, op. cit., pp. 40-41. 


112 STEPHEN TABER 


In opposition to this view it may be urged that the east and 
west ends of Porto Rico are rather irregular; that there is neither 
topographic nor seismologic evidence of faulting; and that the 
absence of Tertiary limestones from the ends of the island may be 
due to submergence. The numerous reefs, projecting rocks, and 
small islands extending eastward from Porto Rico as far as Anegada 
are proof of subsidence at this end of the island; and at the other 
end the water is relatively shallow between Porto Rico and Haiti, 
averaging less than 450 m., while Mona and Desecheo islands rise 
above the surface. Semmes attributes the present position of 
the drainage divide to greater rainfall on the north side of the 
island.t 

ORIGIN OF THE ANTILLEAN TROUGHS 


The great troughs of the Antilles are bounded by fault zones, 
and are characterized by great depth, by precipitous inclosing 
scarps which show abrupt changes in slope at top and bottom, and 
by relatively flat floors that, instead of being graded like river 
valleys, rise and fall throughout their length. Many of the deepest 
places are close to the inclosing scarps instead of near the centers 
of the troughs, and occasional horst-like elevations are also found 
within the fault zones. The troughs are parallel rather than 
convergent and in places are roughly arranged en echelon. 

All of the evidence is in accord with the conclusion that the 
troughs have been formed through faulting. Moreover, they must 
be attributed to subsidence of long narrow blocks rather than 
elevation of the areas on each side; for the rock strata exposed 
in the marginal scarps standing above sea-level were laid down in 
relatively shallow water, and portions of the larger islands have 
remained continuously above sea-level since pre-Tertiary time, while 
the faulting is more recent. The subordinate depressions at the 
base of the main scarps are also indicative of normal faulting, and 
there is an absence of lateral pressure effects such as would be 
expected to accompany thrust faulting. 

Vaughan is of the opinion that the faulting took place in late 
Tertiary, probably Pliocene time, and he cites three kinds of evi- 


*D. R. Semmes, “‘The Geology of the San Juan District, Porto Rico,”’ Scientific 
Survey of Porto Rico and the Virgin Islands, N.Y. Acad. Sci., Vol. 1, Part I (1919), p. 38. 


THE GREAT FAULT TROUGHS OF THE ANTILLES Te 


dence, namely: (1) The faulting along the south coast of Cuba is 
subsequent to old or middle Miocene, as the Miocene La Cruz 
marl is abruptly cut off at the shore; (2) the sea has had time to 
cut only narrow benches into the fault scarps that form shore lines; 
and (3) the biological evidence demands land connection in late 
Tertiary time between Cuba, Santo Domingo, Porto Rico, and 
thence to South America.* 

The faulting may date from the Pliocene, as suggested by 
Vaughan but the frequent occurrence of earthquakes along the 
fault zones proves that the displacements are still continuing; and 
the sea-waves, which have accompanied nearly all of the great 
earthquakes with the exception of those originating in the Haitian 
land valleys, indicate that these latest displacements have also 
been vertical. Another evidence of recent displacements is 
furnished by the character of the material on the sea-floor along 
the fault zones. Off the northwest coast of Porto Rico broken 
coral is found at a depth of 1,864 m. and the bottom is “rocky”’ 
ies 22 Tem. 

The Antillean fault system, whether it continues into Central 
America or not, forms one of the major structural features of the 
earth; and the displacements now going on are similar to those 
which have occurred in the past. The earthquakes are not limited 
to the immediate vicinity of land areas and hence cannot be attrib- 
-uted to adjustments resulting from changes in surface-loading 
due to erosion. That the faulting extends to profound depths is 
shown by the persistence of the fault zones, by the great height of 
the scarps, and by the extrusion of lava from some of the fault 
fractures. 

The total volume of lava that has reached the surface along the 
Antillean fault zones is insignificent as compared with the size of 
the troughs; consequently there can be in this instance no causal 
relation between volcanic activity and the foundering of the narrow 
blocks, such as has been suggested for the Christiania Fiord in 
southern Norway, the “rift valley” of eastern Africa and similar 
troughs elsewhere. 


«T. W. Vaughan, “Geologic History of Central America and the West Indies 
during Cenozoic Time,” Bull. Geol. Soc. Amer., Vol. XXIX (1918), p. 626. 


2 United States Coast and Geodetic Survey Chart gor. 


114 STEPHEN TABER 


Fault phenomena of the type exhibited in the Antilles testify 
to a-state of tension rather than compression in the earth’s 
crust. This tension is difficult to explain. It may be due to 
warping, for there is some evidence that the formation of the 
troughs has been accompanied and perhaps, in part, preceded 
by an upswelling or arching of the region together with more 
or less subsidence of the border portions. The uplift has 
culminated in the Island of Haiti, and the Bartlett Trough 
has developed along the axis of uplift, the surface being in most 
places higher near the borders of the trench than farther back. 
Sea-terraces, marking intermittent elevations have been observed 
along many coasts of the Antilles; in Cuba they extend entirely 
around the east end of the island but attain their finest develop- 
ment along the south coast between Cape Maisi and Guatanamo.' 
On the opposite side of the trough near Montego Bay, Jamaica, 
they are also well developed, Hill having observed six distinct 
benches.” 

While the evidence is extremely meager it is at least in accord 
with what is known concerning the structural relations of fault 
troughs in the other parts of the earth. Most of these troughs ap- 
pear to have been formed as a result of the upswelling of broad arches 
and plateaus of the downwarping of basin-like areas under the in- 
fluence of forces that probably have acted vertically. 


tR. T. Hill, Cuba and Porto Rico with the Other Islands of the West Indids, p. 44, 
New York, 1899. 

2R. T. Hill, ‘The Geology and Physiography of Jamaica: Study of a Type of 
Antillean Development,” Bull. Mus. Comp. Zool. Harvard College, XXXIV, Geol. 
Ser. 4 (1899), p. 31. 


THE CHARACTER OF THE STRATIFICATION OF THE 
SEDIMENTS IN THE RECENT DELTA OF FRASER 
RIVER, BRITISH COLUMBIA, CANADA‘ 


W. A. JOHNSTON 
Geological Survey of Canada, Ottawa, Canada 


OUTLINE 
INTRODUCTION 
FRASER RIVER AND ITS RECENT DELTA 
Pitt LAKE DELTA 
CHARACTER OF STRATIFICATION OF THE SEDIMENTS 
SUMMARY OF CONCLUSIONS 


INTRODUCTION 


During the course of an investigation, carried on during parts 
of rg19 and 1920 by the Geological Survey of Canada in co-operation 
with the Department of Public Works, of the characteristics of 
Fraser River, British Columbia, an opportunity was afforded of 
studying the character and mode of origin of the stratification of 
the sediments forming the Recent delta of the Fraser. Bottom 
samples were obtained by means of a bottom-sampling machine 
- which furnished core samples (1) from the sea-bottom deposits in 
the Strait of Georgia, and (2) from the fresh-water deposits in 
Pitt Lake, a tidal lake tributary to the Fraser. The object of this 
- paper is to describe the character and mode of origin of the stratifica- 
tion of the Recent delta deposits of the Fraser River, and to contrast 
the character of the stratification of the sediments being formed in 
fresh water with that of the sediments being formed in salt water. 

The writer is indebted to Dr. W. Bell Dawson, superintendent 
of the Tidal and Current Survey, Department of the Naval Service, 
Canada, for information regarding the tides and tidal currents in 
Fraser River and in Pitt Lake, and to Mr. C. C. Worsfold, district 
engineer, Department of Public Works, Canada, for information 
regarding the tides and freshets in Fraser River. 

t Published by permission of the Director of the Geological Survey, Canada. 


TIS 


116 W. A. JOHNSTON 


FRASER RIVER AND ITS RECENT DELTA 


The Fraser River is the largest river in British Columbia whose 
basin lies entirely within the province. It has a length of 7090 
miles and drains an area of 91,700 square miles. The maximum 
discharge of the river at Hope (100 miles above its mouth) for the 
past nine years, as determined by the Dominion Water Power 
Branch of the Department of the Interior, was in July, 1920, when 
it amounted to 380,000 c.f.s. During the great flood of 1894, 
however, the discharge as estimated by officials of the Department 
of Public Works was nearly 500,000 c.f.s. The average minimum 
discharge of the river at Hope for the past nine years was 17,000 
c.f.s., the lowest being 12,000 c.f.s. There is thus considerable 
variation from high to low water in each year. The water usually 
begins to rise in May and reaches its greatest height in June or 
July. Extreme low water may occur in any month from November 
to March. During the low-water period the river is affected by 
ocean tides for a distance of 60 miles above its mouth. The 
current is reversed in the river by the flood tide, however, only for 
30 miles above the mouth. ‘The tides in the Strait of Georgia are 
declinational in character and are characterized by marked diurnal 
inequality of the semidaily tides. The mean range at the mouth 
of the Fraser River is 6.4 feet. The maximum range of the great 
flood tides is 15 feet and of the great ebb tides 14 feet. These 
maximum ranges, however, are rarely reached, the average for the 
large tides—which occur for a few days twice a month near the 
time of maximum declination of the moon—being about 12 feet 
for the ebb tides and 13 feet for the flood tides. During the 
balance of the month the ranges are less. The fact that the river 
is tidal for a considerable distance above its mouth renders it 
navigable; for the lack of river water during the low-water stage 
of the river is largely compensated by the tidal water. The 
combined ebb-tidal and river discharge of the river, in the main 
channel near its mouth, is rarely less than 100,000 c.f.s., even 
during the extreme low-water stage of the river. 

The water of the Fraser River is turbid to some extent through- 
out the year, and is markedly so during freshets; but the amount » 
of material carried in suspension is comparatively small. The 


SEDIMENTS IN THE DELTA OF FRASER RIVER Iie 7) 


average amount of material carried in suspension, as shown by 
analysis of composite samples of the river water collected triweekly 
at New Westminster for the period of one year from June 5, 1919, 
to June 5, 1920, was 62 parts per 1,000,000, and the maximum 
230 parts per 1,000,000. The high turbidity of the water is due 
to comparatively large amounts of very fine material carried in 
suspension. About half the material carried in suspension during 
the freshet is composed of silt and clay, a part of which is so fine 
that it can with difficulty be removed by filtering. The river in 
its seaward part carries in suspension little material coarser than 
very fine sand (0.1 to 0.05 mm.), but considerable quantities of 
medium and coarse sand are moved along the bottom. During 
the freshet the turbid river water flows out over the sea water in 
the Strait of Georgia and extends for distances of ro to 15 miles 
from the mouth of the river. The color line marking the contact 
of the river water and the sea water is a striking feature in the 
Strait of Georgia, near the mouth of the river, where it oscillates 
backward and forward according to the varying influences of the- 
river, tidal, and wind-induced currents. 

The velocities of the currents in the lower part of the river 
near its mouth vary greatly with the tides and according to the 
stage of river water. During the freshet the maximum velocity 
of the combined river and ebb-tidal currents reaches, but rarely 
- exceeds, 53 knots per hour. During the low-water stage the 
maximum velocity is about 33 knots. The inward-flowing flood- 
tidal current rarely, even during the lowest stage of river water, 
- equals the outward current, although the rise of tide on the flood 
is more rapid than the fall on the ebb. The ebb-tidal current is 
the dominant current because it is reinforced by the river current. 
The sea water enters the river during the freshet only to a very 
small extent, and only along the bottom of the main channel. 
During the low-water stage the sea water at times extends on the 
flood tide as much as 8 miles up the river. The flood-tidal current 
in the Strait of Georgia runs north and is the dominant tidal 
current. It has a maximum velocity of about 2 knots. Slack 
water is very pronounced in the lower part of Fraser River, because 
of the lack of a deep channel where it enters the river, which 


118 W. A. JOHNSTON 


prevents sea water from entering the river in flood tide until the 
tide has reached a considerable height. 

The Recent or modern delta of Fraser River has its head, as 
defined by the point where the first distributary is given off, at 
the city of New Westminster. It extends west for 19 miles to 
the Strait of Georgia and, across its seaward front, is 14 miles wide. 
The surface of the delta is all below the level of high tide, except 
in a few places, where the surface of peat bogs is a few feet above 
the general level and the delta land high enough to be reclaimed; 
this is diked to exclude the flood-tidal and freshet waters. At 
New Westminster the river is confined between drift ridges or 
upland areas, which rise 200 to 300 feet above the river. The 
upland area south of the river marks the inner edge of the delta 
and extends from a point on the river 33 miles below New West- 
minster nearly straight south to Boundary Bay. The delta is 
bounded on the north by the highland area extending from New 
Westminster nearly west to Point Grey. In its seaward part on 
the south side it is interrupted by the highland area of Point 
Roberts, an island-like drift hill, which has been joined on to the 
mainland by the construction of the delta. ‘The North Arm of the 
Fraser flows along the northern border of the delta, the main 
Fraser flows through the central part, and in the southern part 
there are a number of smaller outlet channels. 

There can be no doubt that the materials composing the delta 
have been transported to their present resting-place very largely 
by the river; for it has been shown that the river transports con- 
siderable material during its freshet stage, and there is no other 
apparent source for such a large amount of material. A small 
part, however, has been and is being derived from wave erosion 
of the shores, chiefly of the highland area of Point Roberts. A 
large part of the material composing the delta is sand and silt 
derived by stream erosion from the vast accumulations of glacial 
drift which border the banks of the Fraser and Thompson rivers 
for long distances above the canyon of the Fraser, in smaller part 
from erosion of the bedrock (granite) in the canyon of the river, 
and from erosion of the drift deposits in the upper part of the delta 
area. 


SEDIMENTS IN THE DELTA OF FRASER RIVER IIg 


The delta is building out into fairly deep water in the Strait of 
Georgia in spite of the facts that the river is tidal for a considerable 
distance above its mouth, with a mean tidal range of 6.4 feet and a 
maximum range of 15 feet at its mouth, and that the seaward front 
of the delta is swept by fairly strong tidal currents. The out- 
building occurs because of the dominance of the river currents 
over the tidal currents. The river currents are strong enough to 
transport large quantities of coarse and fine material which the 
tidal currents cannot transport into deep water. The tides and 
tidal currents have the effect, however, of giving an unusual form 
to the seaward portion of the delta. In its seaward portion the 
river flows for several miles through sand banks exposed in large 
part at low tide, but completely submerged at high tide. These 
banks extend seaward on an average of 4 to 5 miles from the higher 
delta land and are for the most part nearly level except where 
channeled by outlet streams. The flood tides and tidal currents 
tend to cause a diffusion and diversion of the river currents in this 
part of the delta and prevent the river from building up its banks. 
The coarse material in the sand banks is gradually moved seaward 
by the dominance of the combined ebb-tidal and river currents, 
which tend to carry it seaward, over the flood-tidal currents moving 
inward. The fine material is kept in suspension until it is carried 
seaward by the river currents and settles by gravity and through 
the influence of sea water in causing it to flocculate, or is carried 
inshore by the flood-tidal or wind-induced currents and lodges in 
the grass-covered marshes and slack-water channels, tending to 
' build them up and extend the higher delta land seaward. The 
flood-tidal currents also have the effect of giving to the greater 
part of the subaqueous front of the delta a smooth, curved 
outline lacking the finger-like projections characteristic of many 
deltas. 

In structure, the delta presents the forms characteristic of a 
high-grade delta built into fairly deep water. The fore-set beds 
are well developed and extend from the 3-fathom line to about the 
30-fathom line and have an average dip of about 10°. The dip, 
however, is irregular, and in places the subaqueous front of the 
delta is nearly vertical for heights of 1 to 3 fathoms. Below the 


120 W. A. JO“NSTON 


30-fathom line the beds slope more gradually seaward, the t1oo- 
fathom line being reached at from 1 to 2 miles from the outer edge 
of the sand banks. The horizontal bottom-set beds, consisting of 
very fine material, occupy the bottom over a considerable part of 
the Strait of Georgia, off the mouth of the river. ‘The strait, 
which is 10 to 15 miles wide, varies in depth in its central part 
from 100 fathoms to about 225 fathoms. The top-set beds are 
thinnest and in most places only a few feet in thickness. In 
places, however, they may be upward of too feet thick, as, for 
example, where a deep channel of the river has been abandoned 
and has been gradually filled and built up to high-tide level. The 
top-set beds are sandy in character in their lower part up to about 
half-tide level, the upper part being silty in character. The silty, 
fine-grained beds, formed by deposition from flood waters over- 
flowing the river banks, become thicker, progressively, upstream, 
owing to the fact that freshet waters do not affect the level of high 
tide in the lower part of the river as far up as Steveston, 6 miles 
above the mouth, but appreciably affect the level in the upper part 
of the river. The shore face of the delta is very weakly developed 
because wave action has little effect owing to the shallowness of 
the water over the sand banks. 

The river, for a distance of 40 miles above New Westminster, 
is in places bordered by extensive flats, mostly below high-tide, 
freshet level, which have been formed by the meandering of the 
river and its tributaries, and by deposition from flood waters. The 
flats, for the most part, have been carved out of the raised 
Pleistocene marine clays and delta deposits and form the alluvial 
flood plain of the river. 


PITT LAKE DELTA 


An unusual kind of delta is being formed at the lower or south 
end of Pitt Lake, which drains south by way of Pitt River into the 
Fraser above New Westminster. The lake is 4 to 23 miles wide 
and 15 miles long, and is tidal. During the freshet stage of the 
Fraser the surface of the water in Pitt River and Pitt Lake is 
raised several feet above the normal level and falls only slightly | 
with the tides, so that during this time there is very little current 
into or out of the lake. During the low-water stage, the tides in 


SEDIMENTS IN THE DELTA OF FRASER RIVER 121 


the lake have a maximum range of about 35 feet, and there are 
strong currents both ways. ‘The delta is being built into the lake 
at the south end, although the drainage of the lake basin is outward 
or toward the south. This is the case because the inward, flood- 
tidal current is stronger than the outward current, because the 
rise of tide on the flood is more rapid than the fall on the ebb, and 
the outflow of river water is not great enough to counteract this 
effect. 

The delta is partly submerged and extends about 4 miles into 
the south end of the lake. The top-set beds are exposed in places 
at extreme low tide, but are for the most part covered by 3 to 6 feet 
of water. A single, straight river channel extends through the 
delta. The delta has advanced along the line of the channel and 
left unfilled a deep depression on its west side. The fore-set beds 
are well developed and extend to depths of 20 to 30 fathoms, the 
slope being in places as much as 30°, but for the most part only 
about 10°. The lake has a maximum depth near its south end of 
about 50 fathoms. In the narrow parts near the north end the 
maximum depth is about 80 fathoms. 


CHARACTER OF STRATIFICATION OF THE SEDIMENTS 


A number of core samples of bottom deposits in the Strait of 
Georgia, off the mouth of the Fraser and in Pitt Lake, were obtained 
by means of a bottom-sampling machine. The machine consisted 
of a hollow casting, 3 feet long, sharpened at the lower end and 
fitted at the top with a ball valve which worked automatically. 
~The machine weighed 55 pounds and was operated from a motor 
boat by means of a sounding machine of the ordinary type. The 
samples were mounted in plaster of paris and sectioned vertically 
to show the character of the stratification. A number of the 
samples are shown in Figures 1 and 2. 

The stratification of the top-set beds of the Fraser Delta and 
alluvial flood-plain deposits differs in different parts of the vertical 
section. Below the level of mean low tide the top-set beds are 
composed largely of sand, are for the most part horizontally bedded, 
but in places are cross-bedded and current-ripple-marked. In 
places there are thick and thin lenses or beds of silt and clay which 


on chamegmmeet 


W. A. JOHNSTON 


py, 2b 4 OL, Yt 


SuLoe shod) Of 


od ek. DE eth odf OG) 


soduies oy, ‘epeues “erqumo5) ysnug ‘JOAN IASI JO VIap Judy oy 


| 


oe 


soyxef 


ffa 7 b425 fo pIvt4o 


ay 


Sayorey yu 27 Gx YI 20 

e) A714 4gsvLg f? wireyy, 
ie widiosg Jo preys 
gv aff aire dk op) ogy 12 Ww frees iLLoyy ogy 


ers P 2m 


YP ye v0 


29% aAly tosrty fo soos 


Crk f Bw $ayzu; 


on 

LE wIPA 
“wary tosnig fo vy 0¢ 
eed Bib anee lo Ware 


9 oy af FLD tee ony 29, 


0 x93se{d ut payumour : 
} UL Pawo; Sutaq syuouMpas jo saydures e109—'r ‘ong 


(2 pLs 


Pes) tatty 4IS BF 72 Yj ay a 
f oy ¥SGPULES te oy 


Soy th 


SEDIMENTS IN THE DELTA OF FRASER RIVER m8 


show no definite lamination. ‘The stratification of the sandy beds 
is in places rendered visible by thin partings of silt and clay. The 
stratification of the beds in the intertidal zone is very marked, the 
characteristic feature being a very thin tidal lamination. Bottom 
sample No. 1, shown in Figure 1, illustrates the character of the 
stratification of part of the beds in the intertidal zone. The light- 
colored parts are silty in character and the dark parts are sandy. 
The seasonal character of these beds is indicated in places by 
layers of vegetable matter, and in other places by the different 
character of the laminae formed during the freshet stage of the 
river, from those formed during the low-water stage. During 
the freshet the water holds in suspension considerable quantities of 
silt, and during the low-water stage it is nearly clear. Hence the 
laminae formed during the freshet are largely silt, and those formed 
during the low-water stage are largely fine sand. The tidal laminae 
average possibly 15 to 20 to the inch. ‘The seasonal layers average 
about 5 or 6 to the foot. But as contemporaneous erosion and 
deposition are constantly going on, any one section gives a very 
incomplete record of either the tidal laminae or seasonal layers. 

The stratification of the alluvial flood-plain deposits differs 
from that of the tidal flood-plain deposits because of the fact that 
during the freshet the surface of the part of the river above New 
Westminster is raised above the normal level and falls very little 
- with the tides. Hence, during the freshet, a single layer of silt, 
which may be as much as $ inch in thickness, is deposited on the 
alluvial flood plain. Neither the tidal nor the alluvial flood-plain 
' laminae show any marked gradation upward from coarse to fine 
particles in the individual laminae. 

The character of the stratification in the fore-set beds of the 
Fraser Delta is indicated approximately in samples Nos. 6, 7, and 8 
shown in Figure 1. The beds are composed of a mixture of fine 
sand (0.1 to 0.05 mm.), silt (0.05 to 0.005 mm.), and clay (less 
than o.oco5 mm.). They are, for the most part, thinly laminated, 
in a somewhat similar manner to the tidal flood-plain deposits. 
_In places the lamination is even, in other places it is markedly 
irregular, and cross-bedding as well as inclined bedding occurs. 
The lamination is tidal in character, but is the result of the com- 


124 W. A. JOUNSTON 


bined effects of flocculation in sea water, river and tidal currents, 
and slack water. <A leaf bed is shown in sample No. 7 at about mid- 
length. As it is known from a comparison of soundings made in 
1859 and shown on the 1860 chart with those made in 1919 by the 
Hydrographic Survey of Canada that sedimentation takes place in 
the fore-set beds, at the points where the samples were taken, at the 
average rate of about 20 feet per year, it is obvious that the samples 
would not show evidence of seasonal banding. It is known that 
coarse material is deposited off the mouth of the river during the 
freshet, and that this is overlain by finer materials deposited during 
the low-water stage. Hence the bedding probably has in places a 
seasonal character, but the annual layers vary in thickness, in 
different parts of the delta front, from a few inches or even less to 
as much as 50 feet. 

Bottom samples Nos. 9 and ro, shown in Figure 1, are from the 
fine-grained bottom-set beds of the delta, in the Strait of Georgia. 
Sample No. 9, from a depth of 70 fathoms, is largely silt, and 
sample No. to, from a depth of 92 fathoms, is largely clay, which, 
because of excessive shrinkage in drying, is badly cracked. The 
samples as mounted and shown in the figures are smaller in size 
than in the original state because of compression and shrinkage in 
drying. Sample No. to is most compressed, and, judging by the 
depth of penetration indicated by the mark left on the outside of 
the sampling machine, is one-half to two-thirds of its original 
length. The other samples are more nearly representative of the 
original thicknesses. 

A marked feature of the bottom samples from the fine-grained 
bottom-set beds is the absence of any trace of lamination either in 
the beds composed of silt with a mixture of very fine sand, or in 
the beds composed for the most partiof clay. The lack of lamina- 
tion is due to the well-known effects of flocculation in sea water, 
which causes the fine silt and clay particles in suspension in the 
river water to settle to the bottom together when flocculation 
takes place as the result of the mixing of the two waters. In 
places also very fine sand settles to the bottom along with the silt 
and clay. Hence, in the case of the fine-grained materials deposited 
in sea water, there is a lack of sorting and no definite lamination, 


125 


SER RIVER 


SEDIMENTS IN THE DELTA OF FRA 


“UOT VIGIIVITS OY} JO JoJIVILYD oY} MOYS 0} AT[OI}IVA pouor}as Udeq savy pue sited Jo 
qosvjd ut pazunowi oie sojdures oyy, ‘“epeury ‘eIquinjoD YsHug ‘oyxe'] Iq Ul poulsoy Sutod S}uswrpas jo so[dures a10)—z “Ol 


- 
= 
oo 
& 
oe % 
= « 
— 
ee ae bean, : 
6 9267 77M PD gos) pe M aeyau) : serene 
ty df ORES yf 2. . ey A O° yideg Yyr4 OF 4 2g FrAp pow Te Soy su} ae £ Sale: By Sof a TEE 
¢ 7 Ye UF ee g2 DE PY 27 the PUES WIPES 
geerr7 ne Pe 20D y6j'ar07 44 pus yirrs Hyg eye] Hey pu Woes | 71 og uto UES ies BE iy : 


Jig 27°] He PES HOPS ela ag wes Mera ; 
OY AT LEDED MO es 
BLO Oe oe 


ae 


J 


20 


3 ) IS PYV7 Hos PUP ors | 
2 OLS HL Of 91% of {mre wou EB o/ejosfiurs weoyog VALU YIUOS aoyog 


126 W. A. JOHNSTON 


and the beds are probably thick and massive. As the samples 
from the bottom-set beds were taken at points where sedimenta- 
tion is taking place at the rate of at least 1 foot per year, they 
cannot be taken as criteria of the presence or absence of seasonal 
banding. It is possible that in places the bedding is seasonal in 
character, for coarser material is carried out into the strait during 
the freshet than during the low-water stage of the river. 

Samples from the bottom of Pitt Lake are shown in Figure 2. 
Sample No. 12 is from the submerged top-set beds, samples Nos. 
13 and 14 are from the inclined fore-set beds, and samples Nos. 
15 and 16 are from the bottom-set beds of the delta at the south 
end of the lake. Sample No. 17 is from the bottom-set beds of 
the delta formed by inflowing streams at the north end of the lake. 

The character of the stratification of the deposits formed in 
the tidal fresh-water body differs in several ways from that of the 
deposits formed in salt water in the Strait of Georgia. Tidal 
laminae are shown to some extent in all the samples from Pitt 
Lake. They are exceedingly fine, being in places as many as 100 
to the inch. The bedding of the fore-set beds is much more even 
than that of the fore-set beds of the Fraser Delta, apparently 
because of the absence of tidal currents along the front of the 
delta. Samples Nos. 13, 14, and 15 (Fig. 2) were taken at the 
south end of the lake near the subaqueous front of the delta. 
They all show a fairly regular recurrence of silty layers (light- 
colored) separated by sandy layers (dark colored). This banding 
is probably tidal in character and may be explained as follows. 
There are two tides daily in Pitt Lake, but one is so small that as 
a rule the current is not reversed; so that there is nearly always 
only one period of outflow and one of inflow each day. Large 
tides occur twice a month near the times of maximum declination 
of the moon, and small tides occur near the times when the moon 
is over the equator. The inward current is greater at the period 
of large tides than at that of small tides. During the times of 
large tides sand tends to be deposited, and during the times of 
small tides silt tends to be deposited. The sandy bands separating 
the silty bands are very thinly laminated and in places contain 
eight or ten laminae. The silty bands also are in places thinly 


SEDIMENTS IN THE DELTA OF FRASER RIVER i, 


laminated, but not so definitely as the sandy bands. ‘The rate of 
deposition in the Pitt Lake Delta is not definitely known, but 
probably is very small, for a comparison of a number of soundings 
taken in 1920 with those taken in 1860 shows no great difference. 
The soundings taken at the point, about 1 mile from the front of 
the delta, where sample No. 16 was obtained, showed that deposi- 
tion had taken place to a depth of 5 to 10 feet during the past 60 
years, or at the rate of 1 or 2 inchesa year. ‘The rate of deposition 
near the front of the delta is probably much greater. The beds 
probably show therefore both tidal lamination and tidal banding. 
They apparently do not show seasonal banding. From the fact 
that there is very little current in the entrance to Pitt Lake during 
parts of May, June, and July when the Fraser is in flood, it might 
be expected that this period would be indicated in the bottom 
samples by a thick layer of silt. During this period, however, the 
turbid water of the Fraser does not enter Pitt Lake to any extent 
and the lake water is nearly clear. 

Bottom sample No. 16, taken at a depth of 50 fathoms, is from 
the bottom-set beds of the delta, in the deepest part of Pitt Lake, 
near the south end. It is composed of silt and clay and differs 
markedly in the character of its stratification from the fine-grained 
samples from the bottom-set beds of the Fraser Delta in the 
Strait of Georgia in that it is thinly and gradationally laminated. 
- The lamination is probably due to tidal action and to the absence 
of flocculation of the fine material in suspension in the fresh water, 
which permits the silt and clay to settle to the bottom at different 
rates. 

Bottom sample No. 17, taken at a depth of 35 fathoms, is from 
the bottom-set beds or from near the contact of the fore-set bottom- 
set beds of the delta at the north end of Pitt Lake. It shows in 
places a tidal lamination and probably also a seasonal banding. 
There is a fairly regular recurrence of silty layers $ to 1 inch, 
separated by sandy layers 1 to 2 inches thick. ‘The silty layers 
were probably deposited during the freshet when the surface of 
the lake did not rise or fall appreciably with the tides, and the sandy 
layers during the periods when the lake was affected by the tides; 
for at ebb tide the currents in the streams flowing through the 


128 W. A. JOHNSTON 


delta at the north end of the lake are greatly increased. This 
appears to be borne out by the fact that in places the sandy layers 
are thinly laminated. 


SUMMARY OF CONCLUSIONS 


The tidal flood-plain deposits of the Fraser River are char- 
acterized by a very thin lamination which is tidal in origin. The 
seasonal character of the bedding is shown in places by layers of 
vegetable material and in other places by alternations of silty 
layers and sandy layers, both of which show tidal lamination. 

The alluvial flood-plain deposits of the Fraser are characterized 
by seasonal layers of silt and vegetable material. 

Neither the tidal nor the alluvial flood-plain deposits show to 
any marked extent gradational lamination. 

The fore-set beds of the Fraser Delta, deposited in sea water, 
are characterized by a thin lamination which is the result of the 
combined action of flocculation in sea water, river and tidal cur- 
rents, and slack water. The beds have in places marked cross- 
bedding as well as inclined bedding. ‘The fine-grained bottom-set 
beds are thick and without lamination, because of the effect of 
flocculation in sea water. 

The Pitt Lake Delta deposits formed at the south end of Pitt 
Lake (a tidal fresh-water lake) are characterized by a very thin 
lamination and by a banding, both of which are probably tidal in 
origin. The fine-grained bottom-set beds have a very thin and 
gradational lamination, which is due to tidal action and to the 
absence of flocculation in fresh water. They show no definite 
evidence of seasonal banding. ‘The deposits now forming in the 
delta at the north end of Pitt Lake have a tidal lamination and 
probably also a seasonal banding. 

A conclusion which may be drawn from the fact that the fine- 
grained sediments deposited in salt water are not laminated, 
because of the effects of flocculation, while those formed in fresh 
water are, is that fine-grained sediments which are evenly and 
gradationally laminated are fresh-water sediments in origin or, if 
marine in origin, are glacial; for it is probable that sea water would 


SEDIMENTS IN THE DELTA OF FRASER RIVER 129 


be sufficiently dilute to prevent flocculation and hence to permit 
of lamination of the sediments only at times when and in places 
where large volumes of water from melting ice sheets were being 
poured into the sea. A possible exception might occur in places 
where large volumes of river water are being poured into an estuary 
or into a nearly land-locked part of the sea; but flocculation takes 
place in the Strait of Georgia in spite of the large volume of fresh 
water brought down by the Fraser, and where the density of the 
‘surface sea water is only 1.010, as compared with the density of 
normal sea water which at.17.5° C. is 1.027. 


THE STRUCTURAL RELATION OF THE PURCELL RANGE 
AND THE ROCKY MOUNTAINS OF CANADA 


FRANCIS PARKER SHEPARD 
University of Chicago 


It has been found repeatedly, in recent years, that on the 
flanks of many elongated mountain ranges there are thrust faults 
dipping in under the range. In the absence of these faults, over- 
turned folds, with their axial planes dipping into the range, and 
indipping slaty cleavage are common. Partly from observation of 
this sort, Dr. R. T. Chamberlin developed his hypothesis of the 
wedge-shaped zones of deformation in mountain building.t With 
this idea in mind, I attempted to determine whether the Rocky 
Mountain trench, a long, narrow depression between two mountain 
ranges, would show a connection with this type of structure. 

This great valley has a length of about 800 miles. It starts 
somewhat south of the boundary line between the United States 
and Canada and runs in almost a straight line to the northwest 
between the Rocky Mountains on the east and a succession of 
ranges on the west. From what is known about the trench it 
has been estimated to have an average width from wall to wall 
of 5 miles. ‘The portion of the trench which I studied was from 
Gateway, Montana, at the boundary line, to Golden, British Colum- 
bia, 180 miles to the north. Study of the trench has been hitherto 
quite limited. Dr. Daly’s section at the 49th parallel? crosses the 
trench, Dr. Schofield’s ‘Cranbrook Area’? borders the west side of 
it for about 60 miles, and the section along the main line of the 
Canadian Pacific across the Cordillera, made by Allen and Daly, 
crosses it about 200 miles north of the boundary. Others have 
made rapid reconnaissances of the trench. 

=R. T. Chamberlin, Jour. Geol., Vol. XVIII (1910), p. 228. 

2R. A. Daly, Geol. Surv. Can. Mem. 38 (1912). 

3S. J. Schofield, Geol. Surv. Can. Mem. 68 (1915). 

4 Guidebook No. 8, Congrés Géologique International, 1913. 


130 


THE PURCELL RANGE AND THE ROCKY MOUNTAINS 131 


In the 180 miles which I studied, there are two distinct geo- 
graphic provinces. In the northern province, which extends as 
far south as Canal Flats, the trench is drained by the Columbia 
River which has its source in Columbia Lake just north of the 
Flats. The valley is narrow, for the most part not more than 
3 or 4 miles wide. In part of this northern zone there is a wide 
valley west of and parallel to the Columbia Valley, which joins 
the trench at both ends and is only separated from it by a narrow 
ridge, broken in one place. South of Canal Flats the valley is 
drained by the Kootenay River, which comes into the trench from 
the east and flows south. In this portion the trench becomes 
much wider till it reaches its maximum width of 16 miles at Fort 
Steele. South of there it contracts rapidly to 5 miles at Bull 
River, but widens again at Elko and at Gateway. 

From what could be deciphered from the stratigraphy, it was 
found that there are two principal series of rocks represented. 
One is a great succession of limestones ranging from the upper 
Cambrian to the Mississippian, but with rather large stratigraphic 
disconformities. The other series is dominantly clastic, consisting 
chiefly of various metamorphic phases of sandstone, shale, and 
conglomerate. It has recently been determined, in at least some 
cases, to be Cambrian, but probably includes much pre-Cambrian. 
Probably its age overlaps somewhat that of the limestone series. 
_ As shown on the map (Fig. 1), this last series makes up most of 
the west side of the trench in the northern zone, while the limestones 
are on the east side and occasionally overlap to the west side. 
- South of Canal Flats the limestone series retreats into the Rocky 
Mountains, and the clastic series crosses the trench, and occupies 
a zone about 12 miles wide on the east side. Farther south, 
between Bull River and Gateway, the limestone series again appears 
in various hillocks in the trench and in narrow zones along its 
sides. 

Structurally the trench has hitherto been considered as a sort 
of Graben‘ or at least the product of normal faulting.? The results 


tR. A. Daly, Geol. Surv. Can. Mem. 38, p. 600. 
2S. J. Schofield, ‘“The Origin of the Rocky Mountain Trench,” Trans. Roy. Soc. 
Can., Sec. 4 (1920), pp. 73-81. 


. Beavermouth 
Golden 

. Harrogate 

. Brisco 

. Sinclair 

. Lake Windermere 
. Columbia Lake 
Canal Flats 

. Fort Steele 

. Cranbrook 

tr. Bull River 

12. Elko 

13. Gateway 


Hi 
(oo) SOI AN BW ND 


Ly 


mT 


LI] Pleistocene 

G7 Mississippian 

Fay Devonian 

NS Upper Ordovician 

RS Lower Ordovician 

tl Upper Cambrian 

b= Pre-Cambrian & Cambrian 


jae! 
Co 
v 


mans 
om Mit 


Scale 1” = 20 mi. 


Fic, 1.—South portion of the Rocky Mountain Trench 13 


THE PURCELL RANGE AND THE ROCKY MOUNTAINS 133 


of my study do not harmonize with this view. The structure 
should be considered in three divisions. 

1. The northern part.—In the northern part, the trench is 
certainly defined by structural features. While the structure 
varies considerably, there are some features which hold for the 
entire distance. There is, for example, intense folding on both 
sides of the trench. The axes of the folds in most cases dip away 
from the trench, and in places are overturned (Fig. 2). 

Most of the faults are thrust faults. On the west side most of 
these have west-dipping fault planes. On the east, most of the 


\ 


Fic. 2.—Ideal section of folding 


thrusts have east-dipping planes. There is a very large thrust 
fault, which either borders or is found in the trench from Canal 
Flats almost to Parsons. This fault causes the semimetamorphic 
series of pre-Cambrian and early Cambrian age to be thrust over 
against the topmost beds of a thick Ordovician series on the east. 
At Harrogate (Fig. 3) there are two west-dripping thrust faults 
each with a displacement of more than 6,000 feet. This great 
fault probably does not exist north of Parsons. At Beavermouth, 
25 miles northwest of Golden, Daly considered that the upper 
Cambrian-Ordovician series on the east was separated from the 
Cougar Quartzite on the west by a normal fault of 15,000 feet 
displacement. In a section in Canyon Creek (Fig. 4) near Golden, 
the same formations were found as those occurring on the two sides 
of the fault at Beavermouth, and here there is no fault and the 
formations follow each other in normal stratigraphic succession. 


FRANCIS PARKER SHEPARD 


134 


SOON o*T=YOUL I sapwos *(1 “Bl J—{ yYSnoryy) yoorg uokueg ye uonyoos—v ‘org 


UDINIANOPIC) 1ah07) ree UDIIGUO)  Uolsquog-sf Fe 


Lens 


“aM ‘abonoag ——— UDIDIAOPIO) uaddy Fs 


Wes 


Mt Mb 


‘Youad | wibpunow Ay oo 


Soft QO°T=YOur I :apeog “(1 “Sty “Y—p yYSnory}) oyeSo11exT 30 UW01}99S—E *o1T 


‘uosainopsc) wadd p) fea} UoLIGUod—ayy [eee 


UDIAIAD PI) 4AM0 7) 


SSS = LSS SSS? 


——_=S 
SSS 


Un 
| 
oH? 


THE PURCELL RANGE AND THE ROCKY MOUNTAINS 135 


At Beavermouth there is a wide zone separating the two series in 
question, and the fault was inferred entirely on lithological grounds. 
It seems unlikely that such a fault exists. As will be seen from the 
section, thrust-faulting on the east is important in this region, 
which is rather significant in view of the dying out of the west- 
dipping thrust faults. 

In the southern portion of this northern zone there are cases 
on the east side of the trench where there are intersecting east- and 
west-dipping thrust faults. In all such cases the west-dipping 
thrusts cut those from the east. There is an interesting example 


i i 


ay 


a) i 
mg 


Roeky M ountain 


La mt 


fit ce a a i : a — Cy - 
ve rH if i a a 
Hn ee HZ X ue a eee a " 


Fic. 5.—Section at Sinclair 


of this at Sinclair Springs (Fig. 5). To the east the west-dipping 
thrust faults always become less conspicuous, and finally die out. 
At Brisco an east-dipping thrust fault was found on the west side 
of the trench. Thus there is considerable intersection of the east- 
and west-dipping thrust faults. 

If the general rule for thrust faults is to have dips of much less 
than 45°, then these faults are exceptional since their dips average 
about 50° and in some cases are as high as 70°. This is in contrast 
to the low dips of the thrust faults on the east side of the Rockies 
in the northern United States and Canada. Experiments, on 
folding and faulting, which I have made, show that faults which 
develop at the intersection of two folds may have dips which 
exceed 45° due to the upward relief being much easier than the 
lateral. The faults developing between two separate mountain 
ranges might be subjected to the same influences. 


136 FRANCIS PARKER SHEPARD 


The age of the faulting is not known definitely, but there are 
several factors bearing on the age, which should be mentioned. 
In the first place the faulting must have been subsequent to most 
of the folding in the Rocky Mountains, because in almost no 
instances are the fault planes involved in the severe folding found 
on the east side of the trench. The folding of the Rockies is 
thought to have been mostly during the Laramide revolution. In 
several cases thrust faults of very great throw were found to exist 
in such a location that the upthrow side of the thrust is in the 
valley, while the downthrow side forms the mountain wall (Fig. 3). 
This feature was found in so many cases that it is obvious that 
faulting does not directly control the topography. Therefore this 
faulting must be old. It seems probable that it took place in early 
Eocene times, at the close of the Laramide deformation. If the 
overthrusts on the east side of the Rockies are contemporaneous 
with those on the west, the great amount of subsequent erosion 
lends some weight to the problem of the age of the Lewis over- 
thrust, which has been considered either early Eocene or 
Miocene.* 

The slaty cleavage found in the vicinity of the trench in this 
section is very constant in character in that it persistently dips to 
the east on the east side of the trench, while on the west side it dips 
to the west. Up Canyon Creek, where the section was almost 
uninterrupted, the cleavage, as shown by the dotted lines (Fig. 4), 
dips to the west in the Purcell Mountains, becomes gradually 
vertical toward the middle of the trench, and farther east dips at’ 
high angles to the east. Across the trench the cleavage dips to. 
the east at lower angles. 

Thus the folding, the thrust-fault system, and the slaty cleawane 
all point toward the indipping structures and are evidence in favor 
of Chamberlin’s hypothesis of the wedge-shaped deformation 
masses formed in mountain-building. Here in the trench two 
intersecting wedges occur, one the Rocky Mountain and the other 
the Purcell-Selkirk system. The great overthrusts on the east side 
of the Rockies, such as the Lewis thrust and the Cascade Mountain 


1B. Willis, Bull. Geol. Soc. of America, Vol. XIII (1902), p. 333; R. A. Daly, 
Geol. Surv. Can. Mem. 38, pp. 90-95. 


THE PURCELL RANGE AND THE ROCKY MOUNTAINS 137 


thrust, are examples of indipping structure on the other side of the 
Rocky Mountain wedge. 

The cause of this valley in the northern section is partly the 
zone of weakness produced by the intersection of such a large 
number of fault planes, causing erosion to be more rapid here than 
elsewhere. Also the type of folding on the sides of the trench 
produced a depression between (Fig. 2). 

South of Canal Flats the zone which separates the structure of 
the Rocky Mountains from the Purcell Range leaves the trench, 
and swings to the east. This zone does not die out, but appears in 
the Hughes Range of the Rockies some 12 miles from the trench. 
It was traced as far south as Elko. In at least two places, this 
zone has overthrust faults from the west. Fifteen miles up into the 


——N Rocky Mountain Trench. 
’ rel Clas ic Saries Purcell Clastic Series. 
SW. 


Fic. 6.—Section at Fort Steele (through C—C, Fig. 1). Scale: 1 inch=14.63 
miles. 


AA > NON 
Paleozio Limestfones. 
=a lRR\AR AWA 


NE 


Rockies from Fort Steele, there is an overthrust with a throw of 
several thousand feet (Fig. 6). No valley exists along this fault, 
which shows that the faulting does not always control topography. 

2. The central part.—From Canal Flats to Bull River the trench 
does not follow structural lines, neither in regard to the strike of the 
bordering formations, which is roughly parallel on both sides, nor, 
- so far as could be found, in regard to the faults. The conclusion 
to be drawn from this point is that this part of the trench was 
developed by a stream flowing on an old erosion surface which 
was flat enough to allow the river to develop its course inde- 
pendently of structure. 

This conclusion is strengthened by several factors. To begin 
with, the Purcell Range is thought by Schofield' to have been 
deformed in the Jurassic and peneplained in the Cretaceous, and 
at this time to have developed many valleys out of accord with the 
structure developed in the Jurassic mountain-building. After the 


tS. J. Schofield, Geol. Surv. Can. Mem. 76, pp. 101-2. 


138 FRANCIS PARKER SHEPARD 


uplift of the Purcell Range, these streams maintained their courses 
and developed antecedent drainage. ‘That the old Purcell Range 
existed on both sides of the trench in this zone is evidenced by 
finding the typical Purcell strike on the east as well as on the west 
side of the trench. Also the Purcell series (pre-Cambrian) is 
present on both sides of the trench in this zone.* 

The great width of this portion of the valley, 16 miles, is out of 
accord with the rest of the trench, which might be explained by 
greater age. Also there was found a continuation of the antecedent 
portion of the trench up into the Purcells leaving the trench in the 
vicinity of Cranbrook, and it is just south of here that the trench 
is found to be again connected with structure. 

3. The southern part——The southern structural unit extends 
from Bull River down to Gateway. In this zone there is a thrust 
fault on the east side of the trench which causes the Mississippian, 
or in some cases the Devonian, to be brought into contact with the 
Galton series, partly Cambrian and partly pre-Cambrian,? on the 
east. Large disconformities in this general region make it difficult 
to estimate the displacement of the fault. It probably is not more 
than 2,000 or 3,000 feet north of Elko, but at the boundary line it 
was estimated by Daly? to be as much as 10,000 feet. ‘Ten miles 
south of Bull River it virtually dies out. Here the Devonian lime- 
stone comes down from the mountains into contact with the 
Mississippian in the valley. 

This fault, which appears ‘to limit the east side of the trench 
from Bull River to Gateway, was considered to be a normal fault 
by both Daly* and Schofield.s The writer has no definite proof that 
this is not the case, but there are several facts which suggest that 
it is a thrust fault. In the first place the trace of the fault plane at 
Gateway was followed with considerable certainty for several miles, 
and it was found that the fault plane curved decidedly up a tribu- 
tary valley to the east (Fig. 7). This, of course, suggests an east- 

t§. J. Schofield, ‘“The Origin of the Rocky Mountain Trench,” Trans. Roy. Soc. 
Can. (1920). P. 76. 

2S. J. Schofield, Science, Vol. LIV, p. 666. 

3R. A. Daly, Geol. Surv. Can. Mem. 38, p. 118. 

4 hid. 


5S. J. Schofield, ‘The Origin of the Rocky Mountain Trench,” Trans. Roy. Soc. 
Can. (1920), p. 76. 


THE PURCELL RANGE AND THE ROCKY MOUNTAINS 139 


dipping fault plane. As the upthrow is on the east, the relation 
is that of a thrust fault. 

Also along the eastern wall south of Bull River there is a place 
where the Mississippian appears to dip under the Galton series on 
the east. The common occurrence of the east-dipping thrust faults 
in the northern zone on the east side of the trench suggests the 
probability of this fault being one of the thrust type. It was 
thought by Dawson! that the fault continued north of Bull River 
because of a depression along the east side of the valley. The 


Jefferson Limestone. Galeway Formation. 


Fic. 7.—Sketch contour map showing trace of fault east of Gateway 


writer considers that this depression is one of many instances of 
drainage along the flanks of the retreating valley glaciers. The 
valley in this southern section was probably originally defined by 
this fault plane, and subsequent erosion has broadened the valley 
chiefly to the west. 

In conclusion, the Rocky Mountain trench does not appear to 
be the unit in its development and structure that it has been 
thought to be by Daly and Schofield. It appears instead to have 
been produced partly by normal erosion, partly by erosion along 
lines of structural weakness, and partly by the escarpment of a 
fault. Even in this last instance the faulting is probably of the 
thrust rather than of the normal type. 

1G. M. Dawson, Geol. Surv. Can., Ann. Rept. (1885), p. 190 B. 


ASPECTS OF ONTOGENY IN THE STUDY OF 
AMMONITE EVOLUTION 


A. E. TRUEMAN 
University College of Swansea, Swansea, Wales 


It is well known that the development of an individual generally 
recapitulates to some extent the stages passed through during the 
evolution of its ancestors, or in other words, that “‘ontogeny repeats 
phylogeny.” In the paleontological work of the last quarter of a 
century this principle of recapitulation has frequently been used in 
determining the relationships of fossils. and its general truth has 
been demonstrated among many groups of Invertebrates; for 
example, by Mr. R. G. Carruthers and others among the Rugose 
corals,* by Professor Grabau among the Gastropods,? and by 
numerous workers, particularly Mr. S. S. Buckman, among the 
Ammonites.3 . 

That ontogeny is not always merely a recapitulation of phy- 
logeny, however, has been recognized by zodlogists for many 
years; the development is often modified to fit the embryo for a 
special mode of life, as in many modern Arthropods, so that ances- 
tral stages are obscured to a greater or less extent. In other 
cases on account of the individual hurrying through its development 
in order to attain maturity as early as possible, certain phylo- 
genetic stages are “‘skipped”’ in development. 

Numerous paleontologists have recognized the importance of 
this skipping of stages in development; Mr. S. S. Buckman dealt 
with it some years ago and has pointed out several examples. 


t See, for instance, R. G. Carruthers, ‘‘The Evolution of Zaphrensis delansuei,” 
Quart. Jour. Geol. Soc., Vol. LXVI (1910), p. 523. 

2A, W. Grabau, “‘Phylogeny of Fusus and Its Allies,”’ Smithsonian Misc. Coll., 
No. 1417 (1904), pl. xviii, fig. 2. 

3See, for instance, “Yorkshire Type Ammonites.” Introduction to Vol. I 
(1909-12). 

4S. S. Buckman, ‘‘ Monograph of the Inferior Odlite Ammonites,”’ Palaeont. Soc., 
p. 280. 

I40 


ASPECTS OF ONTOGENY IN AMMONITE EVOLUTION 141 


More recently, he proposed to call it by the name of ‘“‘saltative 
palingenesis”;' Dr. W. D. Lang suggested the simpler term 
“‘lipogenesis’* but as this term may convey a wrong idea of the 
skipping of stages im ontogeny, Mr. Buckman has now proposed 
the term “‘lipopalingenesis.’”3 

The study of fossil lineages affords a ready means of investigat- 
ing the various ways in which the skipping of stages has been 
brought about. This is particularly true in the case of ammonites, 
among which it is comparatively easy to study the ontogeny of 
successive members of a lineage. 

From such studies it appears that there are several slightly 
different ways in which phylogenetic stages come to be skipped in 
the development of later members of a series. The object in this 
article is to draw attention to some of these aspects of lipopalin- 


genesis. 


THE OMISSION OF EARLIER CHARACTERS 


In many cases the characters omitted are those of the earliest 
ancestors, and the ‘‘skipping?’ means rather the omission of the 
earliest phylogenetic stages. This occurs as a direct result of the 
acceleration of development (tachygenesis). A simple example 
may be found in the evolution of the Ammonoids or Nautiloids. 
It is generally accepted that these have evolved from straight 
Orthoceras-like forms, through curved Cyrtoceras-like forms, to 
closely coiled forms like Nautilus itself. Now it may be admitted 
that the Orthoceras-like stage is repeated in an abbreviated manner, 
in the early portion of the Cyrtoceras form, but in succeeding 
members of the series this stage becomes progressively shorter, 
and in Mesozoic Ammonites it can scarcely be said to be represented 
at all. 

In most cases of lipopalingenesis the stage or character omitted 
is not the most primitive.4 This is illustrated by the skipping of 


tS. S. Buckman, “ Yorkshire Type Ammonites,” Vol. I (1909-12). 
2W. D. Lang, Proc. Geol. Assoc., Vol. XXX (1919), p. 60. 
3§. S. Buckman, “‘Type Ammonites,”’ Vol. III (1920), p. 11. 


4It should be noted that in the case just given, the skipping of the Orthoceras 
stage merely represents the skipping of the earliest skeletal stage. 


142 A. E. TRUEMAN 


certain ornament stages in the development of Ammonites. The 
normal order of appearance of ornamentation in Ammonites, as 
in many other groups is (a) striae, (6) costae, (c) tubercles. But 
although this order is followed in phylogeny almost invariably, 
and often, in a general way, in ontogeny, in certain ‘‘accelerated”’ 
members of phylogenetic series the striate and occasionally also the 
costate stages are omitted in development, the tuberculate stage 
having been accelerated so that it directly succeeds the smooth stage. 
The smooth embryonic stage of an Ammonite may be restricted to 
the first few whorls, but however accelerated the specimen, the 
inner whorls appear to be always smooth; the smooth stage is 
never skipped, while the later costate or striate stages may be." 
Similar examples may be found among the Brachiopods and 
Lamellibranchs. Dr. Lang points out to me that in certain 
advanced corals twelve septa appear simultaneously in the embryo, 
the early phylogenetic stages, with fewer septa, being skipped.” 


THE OMISSION OF COMPARATIVELY LATE CHARACTERS 


A more interesting form of lipopalingenesis is illustrated by 
certain Ammonite groups; perhaps the clearest example is to be 
found in the evolution of the Liparoceratidae. Slender ‘‘capri- 
corn’? Ammonites of this family evolve through intermediate stages 
to stout bituberculate forms, but the ontogeny of the latter shows 
no stage in any way comparable with their immediate ancestors, 
the slender capricorn forms. ‘This skipping of the slender-whorled 
stage is represented diagrammatically in Figure 1, which shows 
sections of three Ammonites typical of (A) the slender capricorn 
forms, (B) the intermediate forms, (C) the stout forms, from one 
genetic series. In the ontogeny of the capricorn form the depressed 
early whorls (i and ii) pass into round whorls (iii-v) ; in (B) the stage 
with depressed whorls is shortened (by acceleration of the slender 
whorl stage) but the latest whorls are stouter than in the adult of 
A. This stout form of whorl is more characteristic of C, and has 

tIn certain accelerated Gastropods striae may be present on the protoconch 
and the smooth stage is there omitted (Grabau, op. cit.). 

2 See Geological Magazine, N.S., Dec. V, Vol. IX (1912), p. 557- 


3 A. E. Trueman, “‘ The Evolution of the Liparoceratidae,”’ Quart. Jour. Geol. Soc., 
Vol. LXXIV (1919), pp. 2, 7. 


ASPECTS OF ONTOGENY IN AMMONITE EVOLUTION 143 


been accelerated; since the depressed early whorls are also pro- 
longed the slender whorl stage is skipped entirely in the develop- 
ment of this form. 

The reason for this skipping is perhaps best appreciated when 
it is remembered that the Ammonite C was thus able to produce 
the adult form directly, instead of developing from stout (embryo) 
to slender (youth) and back to stout (adult). The tendency to 
omit in development the stage 
which appears unnecessary for 
the production of the adult form 
would perhaps be expected; the 
most interesting fact is that this 
skipping is accomplished not 
simply by the acceleration of 
the stout form of whorl but Fic. 1.—Sections to show the changes 
partly by an apparent retarda- of whorl-shape in three ammonites from a 

3 single lineage of the Family Liparocera- 
tion of the early characters, not- 


tidae: 
withstanding that the series is A, a slender capricorn ammonite. 
distinctly progressive. C, an involute stout ammonite. 


B, a form intermediate between these 
two. 


Without discussing here the 
biological interest of these obser- 
vations, it is perhaps not out of place to notice that if lipopalingenesis 
is of frequent occurrence, the tracing of phylogeny from ontogenetic 

evidence is not so easy as has been thought by some workers. It 
is comparatively easy to follow cases of skipping of stages when an 
early phylogenetic stage is omitted, but in cases like the one just 
‘described, where the characters omitted are some of the more 
advanced, there is obviously much difficulty in tracing descent 
from developmental evidence alone. In such cases, descent can 
only be proved by the finding of series of intermediate forms 
connecting the various species. 


A NEW PHYTOSAUR FROM THE TRIAS OF 
ARIZONA 


MAURICE G. MEHL 
University of Missouri, Columbia, Missouri 


Among the vertebrate remains from the Triassic of the western 
states no other group is so abundantly represented as the Phyto- 
sauria. Relatively abundant as fossils of this group are, however, 
there is much to be learned of each of the several distinct types that 
have been described. For the most part it is the skull that is 
available for study, but even this is imperfectly known. 

Some time ago the writer described a well-preserved phytosaur 
skull from Arizona, now in the geological museum of the University 
of Wisconsin. This skull was considered a new form and was made 
the type of the genus Machaeroprosopus.* While several of the 
doubtful details of the phytosaurian skull were made known by 
the study of the specimen, especially the relations of the bones of 
the posterior side, the palate, as is usually the case, was left in 
doubt. 

Through the kindness of the University of Chicago the writer 
was permitted some time ago to study a phytosaur skull in the 
collections of Walker Museum; a skull very similar to the Uni- 
versity of Wisconsin specimen in many points. The study of this 
material has made evident several pointed suggestions especially 
concerning the structure of the palate. 

The specimen herein described is No. 396 of the Walker Museum 
Vertebrate Paleontology Collections. It is the gift of Professor 
J. E. Anderson, formerly of the School of Mines, at Socorro New, 
Mexico. The name of the collector is unknown and the exact 
locality has not been recorded. However, the skull is known to 

™M. G. Mehl, “New or Little-known Reptiles from the Trias of Arizona and 
New Mexico, with Notes on the Fossil-Bearing Horizons near Wingate, New Mexico,” 
Bull. University of Oklahoma, New Series No. 103, University Studies Series No. 5, 
1916, pp. 5-24. 

144 


A NEW PHYTOSAUR FROM THE TRIAS OF ARIZONA 


have come from the Triassic of Gua- 
daloupe County, near Santa Rosa, 
New Mexico. 

The material collected consists of 
a few large, well-preserved pieces of 
bone representing a fairly complete 
skull (Fig. 1). In the skilful hands 
of Paul C. Miller the missing por- 
tions have been restored in plaster 
and the major features are almost 
as certainly determined as though 
all the fragments had been collected. 
On both the dorsal and the ventral 
sides the skull is complete along the 
median line save for the occipital 
condyle proper. On the right side 
the jugal, except for the portion that 
forms the posterior border of the 
antorbital fenestra and the process 
that takes part in the border of the 


orbit, is missing, as is the quadra- 


tojugal, the squamosal, and all but 
the anterior end of the postorbital. 
-On this side, too, the outer end of 
the paroccipital and the quadrato- 
pterygoid bar are missing. On the 
‘left side the missing portions are 
much the same as on the right, 
except that the restoration extends 
farther forward. On the left side 
the posterior end of the jugal and 
all of the quadratojugal are pre- 
served. 

The skull (Fig. 1) is large, about 
865 mm. long, and massive. It isa 
crested form of the “broken out- 
line” type. From an oval cross- 


Lat 
aS 
on 


Sutures indicated by broken 


Fic. 1.—Machaeroprosopus andersont, side view of restored skull, slightly over one-sixth natural size. 


lines cannot be determined. 


146 MAURICE G. MEHL 


section just behind the expanded and abruptly downturned tip, 
the rostrum gradually assumes an A-shaped cross-section. Back 
from the tip for a distance of about 293 mm. the rostrum does 
not increase in depth materially. At this point, however, the 
outline of the crest bends in a pronounced curve upward for a short 
distance and then extends in an essentially straight line up to the 
anterior border of the nares, 512 mm. from the tip of the rostrum. 
As the crest approaches the nares after the abrupt rise, it loses 
much of its angular cross-section. 

The narial “hump,” while actually rising above the plane of the 
cranium proper about half an inch and while somewhat accentuated 
by the depression of the cranial roof in front of the orbits, is not 
so conspicuous as in several other forms. This is due to the fact 
that the crest is but slightly depressed Immediately in front of the 
nares. 

OPENINGS OF THE SKULL 

The position of the nares is that of the most highly specialized 
phytosaurs, high on the skull and not far from the eyes. The 
posterior border of the nares is about 105 mm. in front of the center 
of the orbit. The openings are 63 mm. long and about 21 mm. 
wide. They are separated by a moderately thin partition that 
does not reach the level of the outer borders. The plane of the 
combined openings is directed upward. 

The orbits are slightly longer than wide, about 62mm. in 
diameter. 

The antorbital fenestrae are somewhat distorted. They appear 
to have been about 104 mm. long and perhaps 45 mm. wide. While 
they do not extend forward beyond the nares, as is the case in 
similar forms, this is due to the moderate length of the antorbital 
fenestrae and not to a unique position. ; 

The sides of the lateral temporal fenestrae are restored in part. 
Because of this and some distortion it is not possible to give exact 
dimensions. It is thought, however, that the greatest diameter 
was not over 120 mm. 

The supratemporal fenestrae show a development similar to 
that of Mystriosuchus.* The posterior border, i.e., the parieto- 


tJ. H. McGregor, ‘‘The Phytosauria, with Especial Reference to M ystriosuchus 
and Rhutidodon,” Mem. Am. Mus. Nat. Hist., Vol. IX (1906), Part II. 


A NEW PHYTOSAUR FROM THE TRIAS OF ARIZONA 147 


supramosal arcade is markedly depressed. The depression of this 
arcade has not advanced so far as in Machaeroprosopus validus.* 
The opening is very inconspicuous in a superior view because of the 
backward extension of the postorbital. The opening is directed 
out and back. It is slitlike, but its length cannot be determined 
because the outer posterior border is missing. 


SEPARATE BONES OF THE DORSAL ASPECT 


The premaxillae are gradually expanded laterally at the anterior 
ends for the accommodation of the large terminal teeth. The 
expanded portion is abruptly down-curved and extends about 
30 mm. below the palate surface of the rostrum. At the margins 
of jaw, the premaxillae unite with the maxillae at about the twenty- 
fifth tooth. From this point the suture extends back and up in an 
irregular line. At the median line each premaxilla sends back a 
process to within 45 mm. of the nares. This process and a lateral 
posterior process extending back the same direction clasp the 
anterior process of the septomaxilla. 

The septomaxillae are larger than in any other form that the 
writer has examined. ‘They are united along the median line for a 
distance of 48 mm. and thence extend forward between the posterior 
process of the premaxillae a distance of 45 mm. The septomaxillae 
form a large area about the front and sides of the narial prominence, 
_but their exact posterior extent cannot be determined. It is 
thought that they make up but a small part of the narial septum. 

The maxilla has its greatest anteroposterior extent along the 
_ alveolar margin. In this respect it differs from the University of 
Wisconsin specimen referred to above. 

The zasals form the posterior, median, and most of the lateral 
border of the nares. The lower anterior margin extends some 
distance beyond these openings, essentially as far as the septo- 
maxillae. Posteriorly they extend about 35mm. beyond the 
posterior border of the antorbital vacuity. 

In the shape and extent of the lachrymals and prefrontals there 
is the normal phytosaur development. 

The parietals differ from those of Machaeroprosopus validus? in 


t Mehl, op. cit., p. 8, Fig: 2. 
27bid. p. Iz. 


148 MAURICE G. MEHL 


that their combined width is equal to the interorbital width. 
Their length is 67 mm. and combined width the same. 


Fic. 2.—Machaeroprosopus anderson, details of the posterior end of the skull as 
seen from above. Restored portions are indicated by lighter shading and less heavy 
outlines. Two-thirds natural size. 


The postfronials are small and make but a small part of the 
orbit boundary, as is usually the case. 

The relation of the postorbitals is not clearly shown. The 
posterior extent of these bones cannot be determined, for that 
portion of the skull is restored. 


A NEW PHYTOSAUR FROM THE TRIAS OF ARIZONA 149 


The jugals form the 
lower posterior border of 
the antorbital vacuity, 
the lower anterior border 
of the lower temporal 
vacuity, and enter 
slightly in the boundary 
of the orbit. Along the 
lower border of the skull 
their posterior extent is 
such as to practically ex- 
clude the quadratojugal 
from that margin. 

Little or nothing can 
be said of the squamosal 
development. 


POSTERIOR VIEW OF THE 
SKULL 


So much of the skull 
as seen from the pos- 
terior side is restored 
that little of value can 
be given in description. 
The skull has been tele- 
scoped downward, but 
it.1s evident that it was 
considerably wider than 
high. In so many de- 
tails is the skull like the 
type of Machaeroproso- 
pus validus that it is 
assumed the similarity 
extends to the posterior 
side. Some of the dif- 
ferences in the develop- 
ment of the region about 
the supratemporal open- 


Fic. 3.—Machaeroprosopus andersoni, restored 
palate surface, slightly over one-sixth natural size. 


150 MAURICE G. MEHL 


ings are evident in a comparison of Figure 2 with that of M. 
validus.* 


THE PALATE 


The palate of this specimen (Fig. 3) seems very little distorted, 
but although the configuration of the bones is readily determined, 
some of the sutures are obscure. The bones are thin and over- 
lapping, and the slightest abrasion tends to obliterate the unions. 

The palate is marked by the characteristic phytosaurian ridges 
just within the alveolar border. These start indistinctly near the 
third tooth back of the down-turned terminus of the rostrum and 
increase in prominence to a point a little in advance of the internal 
nares. From here they flatten out posteriorly and lose their 
identity before the last tooth is reached. A slight median ridge, 
low and rounded, continues forward from the internal nares to 
about mid-length of the rostrum. 

In the region of the twenty-third tooth, there is a conspicuous 
lateral expansion or swelling of the rostrum. The alveolar ridges 
broaden in this region to fill in the increased width. 

Among the most striking features of the palate is the develop- 
ment of the narial arch. In Angistorhinus, from the slight arching 
formed by the alveolar ridges far forward, the height increases 
gradually until it is broad and marked just in front of the nares. 
Here the height increases rapidly to about 42 mm. at the posterior 
end of the nares. At this point the arch has a width of 82 mm. 
In this genus there is not the slightest suggestion of the con- 
striction of the arch at the lower palate plane, the suggestion of a 
primitive false palate. 

In Mystriosuchus, as McGregor has shown,; the constriction of 
the narial arch at the lower palate plane is marked. To quote: 

This arched condition of the palate suggests two questions of great impor- 
tance in their bearing upon the genetic relationships of the group, namely: 


(1) Do the Phytosauria exhibit the incipient formation of a secondary palate ? 
and (2) if so, is this the first step in a phyletic series, culminating in the highly 


«Mehl, op. cit., p. 8, Fig. 2. 
2M. G. Mehl, “‘The Phytosauria of the Trias,” Jour. Geol., Vol. XXIII (1915), 
No. 2, pp. 129-65. 


3 McGregor, op. cit., pp. 42-43. 


A NEW PHYTOSAUR FROM THE TRIAS OF ARIZONA 151 


modified palate of the eusuchian crocodiles? As for the first of these questions 
I feel that there is no escape from an affirmative answer; an examination of 
the palate of either Mystriosuchus or Phytosaurus |Lophoprosopus] shows a pair 
of longitudinal palatine ridges . . . . which plainly represent the beginnings of 
a secondary palate..... 

The rounded inner border [of the portion of the palatine about the base of 
the narial vault] projects very slightly toward the middle line, so that the 
palatal aspect of the cranium exhibits two elongate ridges which approximate 
each other within 25 mm. at the level of the anterior border of the nares, 
diverging gradually behind this region. These palatine ridges partly obscure 
the outer part of the narial cavities and are continued anteriorly on the maxil- 
laries, but fade out posteriorly without involving the pterygoid..... It 
should be explained that these ridges do not present a sharp edge, but are 
broadly rounded. Nevertheless it is an approximation of the palatines, 
ventral to the internal nares and the pterygoids, and it seems to me that it must 
be interpreted as a tendency toward the formation of a secondary palate. ... . 


In the present form this tendency is so marked that there can 
be no doubt as to its meaning. The narial vault has its greatest 
width somewhat back of the nares. The cavity is still partly 
filled with matrix, but is at least 80-90 mm. wide and somewhat 
over 50mm. high. So restricted is the arch at the lower palate 
plane by the median extension of the palatines, that the opening 
as seen in a palate view is little more than a slit, not over 24 mm. 
wide near the posterior border of the nares and gradually widening 
to the region of the interpterygoid vacuities. This constriction in 
the narial region is in the form of sharp-edged, thin, horizontal 
plates extending beneath the narial vault at the plane of the 
palate. The suggestion of an unfinished or rough edge bespeaks 
a cartilaginous or fleshy continuation. It is believed that the 
air passage was thus completely closed below for a considerable 
distance back of the nares. 

The alveolar ridges cannot be considered as part of this “‘false 
palate,” as is suggested by the foregoing quotation. In the present 
form they are quite distinct from the palatine extensions beneath 
the nasal vault. These alveolar ridges, always conspicuously 
developed in the phytosaurs, should probably be looked upon as 
buffers to prevent the breaking or interlocking of the teeth through 
the sharp snapping of the jaws in an unsuccessful attempt at seizing 


prey. 


152 MAURICE G. MEHL 


OPENINGS OF THE PALATE 


The length of the zwternal nares is 71 mm., about 8 mm. greater 
than that of the external openings. The internal nares are about 
20 mm. in advance of the externals at their posterior border. 

The postpalatine foraminae are exceptionally small and incon- 
spicuous. They are, in fact, little more than slight depressions 
or cracks along the palatine-ectopterygoid union. They are not 
over 30 mm. long and if actually perforating the palate are not 
more than 3 mm. wide. 

The interpterygoid openings are exceptionally small. Their 
exact anteroposterior extent cannot be determined, but they could 
have been but little more than 25 mm. long. 


THE BONES OF THE PALATE 


Some of the details of the relations of the bones of the palate 
are not at all certain. In the palate restoration, the writer has 
attempted to show the relation as the weight of the evidence seems 
to indicate and not as indisputably determined. 

The premaxillae apparently have a remarkable posterior extent 
on the palate surface. They seem to form the anterior border 
and the entire inner boundary of the internal nares and extend a 
short distance back of these openings along the median line. An 
unpublished drawing of this specimen by S. W. Williston doubtfully 
places the posterior end of the premaxillae about 40 mm. in front 
of the internal nares. The bone in this region is platey and brittle 
and has been somewhat abraded in preparation. The writer 
recognizes the possibility of the premaxilla-vomer suture in this 
region, but cannot verify this point. 

The maxillae extend forward on the palate surface to the 
twenty-fourth tooth, numbering from the front. Their width 
on the palate is at no place much greater than that required for 
the alveolar ridges. The union with the jugal as seen in a palate is 
not determinable. 

The vomers seem to be exceptionally small and confined to the 
posterior and possibly the posterolateral borders of the internal 
nares. ‘This is a condition decidedly unlike that usually attributed 
to the phytosaurs and is merely suggested. What is assumed to be 


A NEW PHYTOSAUR FROM THE TRIAS OF ARIZONA 153 


the union of the premaxillae and the vomer at the inner posterior 
border of the nares may be a fracture. If so it is remarkable for its 
symmetry in relation to the median line of the skull. Of the 
posterior boundary there can be little doubt. It joins the ptery- 
goid in a forward convex line about 32 mm. back of the internal 
nares. i 

The palatines apparently form a goodly portion of the lateral 
borders of the internal nares. Aside from this, however, they 
seem to be largely confined to the lower palate plane. It is thought 
that their union with the pterygoids and vomers is at or near the 
base of the narial vault, but of this one cannot be sure because of the 
matrix-filling part of the cavity. Posteriorly the palatines come 
to an acute angle, the point of which prevents the articulation of 
the pterygoids with the ectopterygoids on the palate surface. 

The ectopterygoids are small, triangular bones. The posterior 
borders are down-curled so as to be conspicuous in a lateral view 
of the skull. 

The pterygoids are exceptionally large, but are almost entirely 
confined to the sides and roof of the narial arch. Their extent 
along the pterygo-quadrate bar cannot be determined. 

The parasphenoid has been destroyed, but it must have been 
very similar to that shown in the palate restoration. The union 
between the basi-sphenoid and the pterygoids and basi-occipital 
‘is as indicated in the restoration. ‘The relation of the other bones 
of the posterior part of the skull as seen in the palate view cannot 
be determined. The relations shown are those determined from 
‘the University of Wisconsin specimen mentioned above. 


THE TEETH 


With the exception of a few roots and one partially erupted 
tooth in the middle of the series that shows the crown, all the 
teeth have been lost. The alveolar margin of the jaw is splendidly 
preserved for the most part and gives a good indication of the 
number of the teeth and of their variations in size. 

The alveolae are all distinct and usually separated by a space 
of at least 3 to5 mm. In each maxilla there are apparently twenty- 
four teeth and twenty-three in each maxilla, a total of ninety-four 


154 MAURICE G. MEHL 


in the upper dentition. In the downward portion proper of the 
premaxilla there are two alveolae of a size and shape to indicate 
long conical teeth of about 21 mm. diameter at the base. Immedi- 
ately behind the down-turned portion is another large alveolus, 
about 15mm. in diameter. This is followed closely by a fourth, 
considerably smaller, alveolus. Between the fourth and fifth 
alveolae is a conspicuous space. This is marked on one pre- 
maxilla by a depression as though for the reception of a tooth 
from the lower jaw. The following alveolae, with the exception 
of the last few on the premaxilla, increase gradually in size from 
about 6mm. tor2mm. There is a marked lateral expansion of the 
rostrum near the posterior end of the premaxillae for the accommo- 
dation of three or four exceptionally large teeth. The root of one 
of these is preserved and measures 13 mm. in diameter. The 
crown preserved in the maxilla series as mentioned above is laterally 
compressed with sharp, slightly serrate, anterior and posterior 
edges. 

It would seem that the dentition was very much like that of 
Machaeroprosopus validus,’ greatly enlarged seizing teeth in the 
front, grading through smaller, sharp, conical to laterally com- 
pressed slicing teeth behind. The space between the fourth and 
fifth alveolae indicates likewise a shorter jaw with large terminal 
teeth directed sidewise. 


RELATIONSHIPS AND HABITS 


Of the close affinity between the present specimen and that 
described as Machaeroprosopus validus? there can be no doubt; 
both skulls are crested forms of the ‘‘broken outline” type; both 
have the depressed posterior border of the supratemporal openings; 
and both have enlarged, conical terminal teeth with laterally 
compressed, sharp-edged, slicing teeth behind. The skulls differ 
in SO many minor points, however, that they can scarcely be placed 
in the same species. The writer will designate the present form, 
therefore, as Machaeroprosopus andersoni in honor of Professor 
Anderson, who presented the material to the University of Chicago. 


™ Mehl, Bull. Univ. of Oklahoma, op. cit., p. 20. 
2 Thid. 


A NEW PHYTOSAUR FROM THE TRIAS OF ARIZONA 


155 


The following table comparing the two forms will show the chief 
differences on which the new species is established. 


Machaeroprosopus validus 


Postero-median border of  supra- 
temporal fenestrae completely 
depressed. 


Anterior border of nares elevated 
above lateral borders. 


Terminal expansion of rostrum 


abrupt. 


Nasals not extending to anterior 
border of nares. 


Greatest length of maxillae above 
alveolar margin. 


Approximately seventy-four teeth in 
upper dentition. 


Six large teeth in terminal expansion 
of rostrum. 


Alveolae crowded, occasionally con- 
fluent. 


No lateral expansion of rostrum at 
- posterior end of premaxillae. 


Machaeroprosopus andersoni 
Postero-median border of  supra- 
temporal fenestrae not completely 
depressed. 


Anterior border of nares not elevated. 


Terminal rostrum 


gradual. 


expansion of 


Nasals extending some distance in 
front of anterior border of nares. 


Greatest length of maxillae at alveolar 
margin. 


Approximately ninety-four teeth in 
upper dentition. 


Four large teeth in terminal expansion 
of rostrum. 


Alveolae not crowded. 


Lateral expansion of rostrum at 
posterior end of premaxillae. 


The Phytosauria are a unified group in that they all tend to de- 
velop certain peculiarities that set them off from other groups in a 


striking manner. 


skull, an elongation that left the nares far from the tip. 


All developed the snoutlike elongation of the 


In some 


cases there seems to have been an actual retreat of the nares. 
There seems also to have been a tendency to elevate the nares 
above the plane of the cranium proper, a goal very conspicuously 
reached in several forms. All of the phytosaurs showed a more 
or less marked development of greatly enlarged terminal teeth 
and most of them developed at least a crude slicing dentition 
farther back in the jaws. 

These modifications or tendencies are so distinctive that they 
have called forth considerable speculation as to the habits of the 


156 MAURICE G. MEHL 


phytosaurs. The seemingly common goal in the modifications 
of all the phytosaurs suggests similar habits. Of the probable 
habits of Machaeroprosopus the writer has stated in another 
place: 


All of the phytosaurs were supposedly more or less amphibious in habit. 
Still, the heavy dermal armor of several of the forms would indicate that a 
very considerable part of their time was spent on the land. The posterior 
position of the nares is usually not considered a distinct aquatic adaptation 
in the case of the phytosaurs. It is rather explained, along with the long 
snout and large terminal teeth, by attributing to the slender rostrum the 
function of a prod or rake with which the possessor searched out worms and 
other soft bodied invertebrates, etc., in the mud of shallow waters while the 
nares, by virture of their position, were above the surface of the water. 

There is a marked tendency in most of the phytosaurs, a tendency in 
which M. validus surpasses all other known forms, for the nares to rise on a 
considerable prominence. This would seem to be a modification entirely 
uncalled for were the position of the nares due solely to the use of the rostrum 
asa prod. So situated are the nares in M. validus that it could submerge the 
body save for the narial hump and lie in wait admirably concealed from its 
enemies, or more likely, from its prey. Certainly the teeth of this form are 
more fitted for tearing large vertebrates than for small, mud-burrowing crea- 
tures. Then too, Machaeroprosopus could scarcely pick small objects from 
the ground because of the difference in length between the upper and lower 
jaws. When the jaws were closed the terminal teeth of the lower jaw were 
functionless and the upper terminal teeth were but little better. It was not 
until the jaws were wide open that the terminal teeth could be used effectively 
either as a rake or for seizing any sort of prey. With the jaws separated, 
however, the upper and lower terminal teeth were directly opposed and admi- 
rably fitted for seizing and tearing large animals. It seems likely that M. validus 
was wont to lie in wait concealed close up to the shore in shallow waters ready 
to seize its prey when the latter came down to drink. Once the prey was 
dragged into the water it was at the mercy of its captor, for the dentition of 
the latter probably matched that of any carnivorous land form of the time and 
the phytosaur had the distinct advantage of being entirely at ease in the water. 


If these were the habits of the more highly specialized phyto- 
saurs, and the assumption seems logical, the group was remarkably 
well adapted to the conditions of the times. The adaptation was, 
moreover, of a very peculiar nature. In most cases adaptations 
seem to do little more than tend to counteract adverse conditions. 
For instance, the development of speed in ambulatory forms of 


t Mehl, Bull. Univ. of Oklahoma, op. cit., pp. 23-24. 


A NEW PHYTOSAUR FROM THE TRIAS OF ARIZONA 157 


arid regions permits a wider range about the limiting water supply 
to compensate for the diminished food supply and the distance 
between water-holes. - 

In Machaeroprosopus the modifications not only met an emer- 
gency but actually turned it to advantage; the phytosaur waited 
for the rigorous conditions to bring his food within reach. If our 
interpretation of the conditions in the western interior of North 
America during Triassic times is correct the scarcity of food necessi- 
tated a wide range for many forms. This meant speedy forms 
ordinarily difficult of capture, but it also meant the periodic 
crowding of the limited water-holes; it compelled the food to 
walk to the captor. 

It is not easy to see why a group so remarkably fitted to their 
environment as was the phytosaur group should not prosper 
better asarace. They were very short lived, confined to the latter 
part of the Trias, apparently. One can scarcely appeal to over- 
specialization unless this means “fitting in” rather than marked 
structural change. The phytosaurs were scarcely less generalized 
than the living crocodilians. Perhaps perfection alone is enough 
to condemn a race. A consideration of this possibility will form 
the basis for a future paper. 


ADAPTING A SHORT-BELLOWS, ROLL-FILM KODAK 
FOR DETAIL WORK IN THE FIELD 


CHESTER K. WENTWORTH 
University of Iowa 


Most geologists, when in the field, have occasionally felt the 
need of a portable camera with which they might take large-size 
detail photographs. ‘The ordinary equipment for such work, the 
long-draw, box camera, is far too heavy and bulky to be carried by 
the geologist as a part of his daily accoutrement. ‘The roll-film 
kodak with the customary short bellows which is most convenient 
for his usual needs, fails when he wishes a picture larger than about 
one-tenth natural size, and the auxiliary portrait lens does not 
greatly increase the scope of his outfit. If he wishes a detail picture 
of a fossil in place, a curious marking or texture of a rock which 
cannot be conveniently sent to his headquarters, he must note the 
place and come again later with the more bulky equipment. He 
almost never does. 

The device described below has been found by the writer to be a 
simple and convenient solution of the problem. An auxiliary por- 
trait lens was first secured to fit the kodak he wished to use. This 
attachment consists of a simple plano-convex lens of about 52-inch 
focus. When placed close in front of the kodak lens this causes 
parallel beams of light to become slightly convergent when they 
strike the main lens system and thus shortens the focus of the com- 
bination (See Fig. 1, Scale II). With the principal focus now some- 
what nearer the ground glass, the slight additional excess of bellows- 
length permits focusing on closer objects and hence larger pictures 
than with the kodak lens only. 

Next, there were procured from an optician two more lenses of 
foci approximately 16 inches and 8 inches respectively. These 
were ground to the proper diameter to fit the cell belonging to the 
original auxiliary. The three lenses are readily interchangeable by 
unscrewing the ring which holds in place the lens used. By using 

158 


ADAPTING A KODAK FOR WORK IN THE FIELD 159 


the two additional lenses, one at a time, the focus of the combina- 
tion is further reduced, as shown in Scales III and IV of Figure 1. 
With the three used successively, the entire range of object-distance 
from infinity to less than eight inches can be accommodated, using 
only the two inches adjacent to the bellows-limit for adjustment. 

After obtaining the lenses, the several combinations were cali- 
brated for different distances by focusing on a ground glass placed in 
the image-plane and the readings made on a white celluloid metric 


Bellows Lirtrit: 
Image Flare. 


Fic. 1—Diagram showing four focusing scales in relation to the image-plane 
and bellows-limit as follows: Scale I, with kodak lens of 6.75-inch focus; Scale II, 
using 52-inch focus auxiliary lens; Scale III, using 16-inch focus auxiliary lens; 
Scale IV, using 8-inch focus auxiliary lens. The figures at the right show the foci 
‘of the several combinations and the diagram shows clearly the increase in scope of 
the bellows as the auxiliaries of shorter focus are added. 


scale which was attached to the bed, as shown in Figure z. From 
these data were constructed the tables I and II which were photo- 
reduced and pasted in place on the bed of the kodak as shown. 
Here they are out of the way, but always convenient. Table I 
shows the setting on the metric scale for different object-distances, — 
the three parts of the table referring, respectively, to the three aux- 
iliary lenses, I, II, and III. Table II gives the object-distances 
for different ratios of object to image. For example, suppose one 


160 CHESTER K. WENTWORTH 


wishes a picture which is one-half natural size. In Table II oppo- 
site the ratio 2 is given the object-distance 15 inches. Then in 
Table I, interpolating between 14 inches and 15 inches, we find that 
the focus should be at the scale-reading of 10.85 and that auxiliary 
lens II should be used. The kodak is then set with the lens 15 
inches from the object to be photographed. Table II gives also at 
the bottom the foci of the three combinations, the focus of the kodak 
lens alone being about 6.75 inches. 


Fic. 2.—Photograph of bed of kodak showing attached metric scale and two 
tables as described in the text. (Somewhat reduced.) 


The celluloid scale may, if desired, be so set as to read inches of 
focus from the ground glass directly; that shown was set arbitrarily 
and the readings indicate position only. With this device, sharp 
and undistorted photographs up to natural size may be taken by 
using a tripod or one of the numerous types of clamp support, and 
an aperture of F/32 or smaller. Since in pictures of this kind where 
the object is not plane, it is almost always necessary to reduce the 
aperture and give a time exposure in order to get depth of focus, 
there is no additional difficulty due to the auxiliary lens. Instan- 
taneous pictures at full aperture may be made if need be, but the 


ADAPTING A KODAK FOR WORK IN THE FIELD 161 


distance must be estimated with considerable accuracy to secure 
good definition. Such rapid exposures are rarely necessary or even 
desirable in the case of geologic subjects. 

Anyone desiring to use this device will find that auxiliary lenses 
of 52-inch, 16-inch, and 8-inch foci will be a suitable series for a 3A 
kodak having a lens of about 6.75-inch focus. For the use of others 
with different kodaks, the following formula is given: 


1E I I 
res (+p) I.08 
where F,=focus of kodak lens, F,=focus of auxiliary lens, and 
F,+:=focus of the combination. This formula is only approximate 
but is sufficiently accurate to guide one in the selection of a series of 
auxiliary lenses. These lenses can be secured from any well- 
equipped optician who does his own grinding. They should then 
be calibrated by focusing on a ground glass in the image-plane and 
taking readings on the scale. Tables can then be constructed, as 
shown in Figure 2. | 

The extra lenses can be carried in any convenient pill-box. The 
cost for the three lenses and cell is less than five dollars and the out- 
fit has been found by the writer during two seasons’ field work to 
give highly satisfactory results. 


SEGREGATION GRAN ITES 


ALFRED C. LANE 
Tufts College, Mass. 


The deposits of asbestus of the Eastern Townships in Canada 
are the most exploited of the world. In 1919 $18,000,000 worth 
or 3, 082,384 tons were mined. This means a great deal of develop- 
ment and has given J. W. Dresser, the geologist who has for years 
followed the region with the closest attention, a chance to make 
observations which are of general theoretical interest. Fifteen 
years ago* Dresser suggested that certain granites seemed to have 
been differentiated from the same magma from which were derived 
the serpentines in which the veins of asbestus chrysotile occur. 
This would tend to support Daly’s idea? that granite is a product 
of differentiation from a basaltic substratum, but not necessarily that 
ut 1s ““ syntectic,”’ that is, due to previous assimilation of something 
like a quartzite. 

Some twenty years ago I had occasion to study with critical 
care what I then called “‘acid interstices” which practically always 
occurred near the middle of every diabase dike of over ten meters 
thickness or so. I had over two hundred thin sections, and while 
Bayley and Irving and Wadsworth were inclined to believe the 
interstices filled with micropegmatite were secondary, I was inclined 
to agree with A. C. Lawson in believing them primary. By a 
careful study I convinced myself that such was the case. Later 
studies showed that the filling of similar interstices in effusive 
rocks was characteristically different, and I called the effusive 
texture doleritic. The conception I gathered was that these 


tSee Bull. Geol. Soc. Am., Vol. XVII, p. 510; also Canadian Survey Memoir, 
Vol. XXII, and other papers by Dresser given in Ferrier’s finding list. 

2 Igneous Rocks and Their Origin, p. 361. 

3 Geological Survey of Michigan, Vol. VI (1809), pp. 235-42, the work was largely 
done in 1890; see Report of the State Board of Geological Survey of Michigan for the Years 
1891 and 1802, Pp. 177. 

162 


SEGREGATION GRANITES 163 


cavities were “filled with the residuum of the molten magma. 
Enough of the rock was formed to make it perfectly solid, for no 
further motion could take place without disturbing the micropeg- 
matite borders of the felspar laths and fracturing the excessively 
delicate apatite needles. The remaining interstices seem, agreeing 
with the general law of increasing acidity in residual magmas, to 
have been filled with the final concentration of an acid aqueo- 
igneous magma which had been corroding the olivine and forming 
the less basic augite from it. In this magma were also concentrated 
the absorbed gases, aqueous and otherwise. which the dike margin 
originally contained and which, as the dike solidified at the margin, 
would probably be driven from it and concentrated at the center. 

‘““The acid magma thus left seems to have proceeded to produce 
brown hornblende upon and out of the augite; brown mica upon 
and out of the iron oxides, as Smyth has suggested; and pegmatite 
growths on or out of the feldspar, while apatite needles formed 
across the cavities.”” The convincing arguments were that these 
interstices did not occur in marginal sections, and showed no signs 
of being more abundant in uralitic sections, but were best developed 
in otherwise fresh dikes, and were not present with superficial 
textures such as the amygdaloid textures. 

The facts I have seen in the Medford diabase and elsewhere 
in the intervening thirty years have only strengthened the convic- 
- tion that this interpretation is substantially correct, though it 
should be expressed not so much in terms of acid and basic, as in 
terms of those eutectic lines and troughs which the geophysical 
' laboratory at Washington has worked out, the theoretical bearing 
of which has been developed by N. L. Bowen.* 

The fluid magma from which the minerals of these interstices 
crystallized out would be closely held by capillarity if they were 
not large. But with increasing coarseness and size of interstices 
there would be more chance for it to drain out like honey from the 
honeycomb, or as Bowen and Harker have suggested, be squeezed 


1 “The Problem of the Anorthosites,”’ Jour. Geol., Vol. XXVIII (1919), pp.393- 
434; ibid., Vol. XXV, 3 (1917), pp. 210-44, ‘‘Crystallization—Differentiation in Igne- 
ous Magmas,” and literature there cited; Am. Jour. Sci. (1915), p. 407; (1914), 
p. 207, etc. 


164 ALFRED C. LANE 


out like fluid from a filter press. When this happens we have the 
aplitic red rock associated with the Duluth Gabbro described most 
fully and lately by F. F. Grout* or with that of Mount Bohemia, 
so fully described by F. E. Wright.?_ If one supposes this segrega- 
tion conducted on a still larger scale or so that the crystallization 
shall be somewhat coarser and the grain coarser, the logical outcome 
would be a granite. This is just what Dresser finds that the 
advance of mining operations has conclusively proved’ by showing 
that masses of granite have no separate connection with the earth’s 
interior. 

The largest mass of differentiated hornblende granite mentioned 
by Dresser is three-quarters of a mile long by one-quarter of a mile 
wide, with a coarser pegmatite border two or three feet wide and a 
porphyritic texture throughout. This granite “‘is an indication 
rather than a cause” of the presence of acid waters in the mag- 
matic residue needed to produce serpentine and asbestos. It is 
interesting to note that Bowen starting with peridotite as a mono- 
mineralic differentiate suggested that the region “‘may” furnish a 
“complementary granitic differentiate.”” He is apparently right! 

Now the laws of physics and chemistry are universal. If this 
has happened in the Lake Superior region and the Eastern Townships 
it must have happened in many other places. One is tempted to 
consider the Mull pitchstones* the “‘leidleites and inninmorites”’ 
which (a) apparently occur only as intrusions, (6) are high in 
primary water, (c) are glassy at the center and stony at the sides, 
i.e., coarser at the margins, as representing such a magma as that of 
the red rock of Wright, of Grout, etc., chilled more quickly and 
without loss of water, but yet like many aplites or the granite de- 
scribed by Dresser, consolidating more slowly at the margin. This 
I have shown is more likely to happen when the initial magma tem- 


x “A Type of Igneous Differentiation,” Jour. Geol., Vol. XXVI (1918), p. 618. 
2 Report Michigan Geological Survey (1908), p. 387. 


3 “Granitic Segregations in the Serpentine Series of Quebec,” Transactions of the 
Royal Society of Canada (1920), pp. 7-13. Compare also N. L. Bowen, “ Differentia- 
tion by Deformation,” Proc. Nat. Acad. Sci. (1920), p. 160. 

4. M. Anderson, and E. G. Bailey, Quart. Jour. Geol. Soc., Vol. LXXI (1916), 
pp. 205-16, London. 


SEGREGATION GRANITES 165 


perature is low. In other words they are “‘secundine’’ to the 
great series of basalts and gabbros. 

The primary water-content (exceeding the average for rocks of 
similar composition) which Anderson and Bailey emphasize as 
characteristic is very suggestive. The analysis (1) of the glassy part 
of the leidleite may be very nearly that of the anchieutectic solu- 
tion that once filled the acid interstices and that wandering off by 
itself would make red rocks and even hornblende granites. That 
though central it is glassy may be due partly to great viscosity and 
less power of crystallization, due to less lime, iron, and magnesia, 
and more silica, partly to a more rapid passage through the crystalli- 
zation range of temperature, even though later. 

If we study these analyses (see following table), 1 and 2 given 
by Anderson and Radley, and compare with 3 and 4, Bowen’s 
analyses of Canada diabase and dike granite and with the average 
igneous rock as given by Clarke and Daly we find, (1) that they 
are not very far from the average igneous rock, (2) that they are 
not very far from the eutectic trough or line to which I called 
attention in 1904,” (3) that they are not very far from the analyses 
of the red rocks given by Grout and Wright, etc., (4) that the 


t J have found the following classification of dikes with the appropriate adjectives 
to have some value: 

1. Invasive-—Forced more or less slowly into cavities formed by the extra hydro- 
static pressure of the invading magma; contacts irregular and often close-welded. 

2. Suctive—Forced quickly into a crack otherwise opened, by fault or earthquake 
relieving strain, aided by gravitative suction, owing to the condensation by cooling 
of the gases from the magma; contacts generally fairly straight and not close-welded. 

Nearly parallel is a classification according to the hot or cold condition of the 
’ country rock as follows: 

1. Secundine (Latin secundine=aiterbirth).—Injected into a hot country rock; 
contact generally irregular, close-welded, the grain generally equal throughout, 
either finer or coarser at or near the margin. 

A characteristic mode of occurrence of lamprophyres, aplites and pegmatites. 

2. Subsequent.—Injected into a cold country rock, with fine-grained selvages 
(the zones of increasing grain amounting to from one-quarter to one-tenth of the 
breadth of the dike) with straight and not close-welded contacts. 

A characteristic occurrence of dikes not closely connected with larger masses or 
volcanic centers, and in composition close to Bunsen’s normal basaltic magma or 
mofe basic. 

The anchieutectic rocks which we are discussing are generally invasive secundine, 
but in case of shrinkage cracks, due to loss of heat or contact metamorphism, may be 
suctive and secundine too. 


2 Jour. Geol., (1904), p. gt. See also Wet and Dry Differentiation, ‘‘Tufts 
College Studies,” Vol. III, Pt. I. 


166 ALFRED C. LANE 


stony margin (An.z) in its less silica, more lime and magnesia is not 
so far on the toboggan slide toward the goal of ‘‘wet differenti- 
ation’’—the magma from whence crystallizes pegmatite veins, and 
not so near to a granite (An. 4) as the glassy center (An. 1). We 
seem thus to have caught two stages in a differentiation which 
carried on on a large scale would lead to the hornblende granites 
described by Dresser. 

As to the syntexis upon which Daly lays stress, there is no 
doubt that the inclusion and absorption of fragments of sandstone, 
especially uf they contained water in the interstices, should promote 
the formation of micropegmatite, and zones of micropegmatite 
around such fragments in process of absorption are found in the 
Medford diabase, as Jaggar and others have seen. Possibly the 
water is quite as important as the silica. Very likely much of 
the red rock of Pigeon Point is of this nature. But the association 
of micropegmatite with diabases is too widespread for me to agree 
with Daly that most granites are differentiates of syntectics. It 
is well worth considering how many are, like that described by 
Dresser, direct differentiates with the help of juvenile juices or 
mineralizers. 


I 2 3 4 
Oxides f i 
Caalete | Redes | Diabase Cobalt mene. 

SiOsiars src 61.69 59.21 50.12 M283 
DIO scan 1.00 1.06 55 74 
TAQ Rott exaors pice 14.43 14.00 1570 12.99 
ING Oneg cio.ol6 node 123 2.66 1.42 none 
HeORe neues 5.86 4.87 6.89 2.50 
IMGTOe ao eeao cos 0.30 Or2ar i Wi de. ekihe ate 2e lee 
CaORR eee 4.07 5-95 II.30 178 
MgO. 2) foi Bi 9.50 0.07 
Na, Oe ached B52 2.06 2.91 7.60 
KAO ace iercrneeere 72 2.83 T.07°) > Jeane eee 
IBAOs son vanes 0.04 0.03 none... | saareareer areas 
180) ae TOE Coo ©.25 2.05 1.03 I.09 
H.,Oabove 105° C 2.36 Tho byl 0.21. “| Se eee 

AO tal eee 100.12* 100.47 T 100.84 { 100.95 § 

* Including 0.02 Cl. 0.24 P20; + Including 0.2 P20;. 


t Bowen in Can. Mining Inst., Vol. XII (1909), p. 523, diabase, including 14 S. 
§ Ibid., granite cutting diabase, including 1.00 CO:. 


© OD: cit, Pa 3u2, etc: 


ON THE REPRESENTATION OF IGNEOUS ROCKS IN 
TRIANGULAR DIAGRAMS 


ALBERT JOHANNSEN 
University of Chicago 


It has been customary to represent three components in a 
triangular diagram by a dot, or four components by a triangle 
within the triangle. The writer, by means of only a dot and a 
line, plots without the use of a slide rule or any calculation what- 
soever, the actual percentages of quartz, orthoclase, plagioclase, 
and dark minerals as well as the relative percentages of quartz, 
orthoclase, and plagioclase among the light constituents. 

Let it be required to plot a rock with the following mode: 


(OUR HAs hice cl coe ci ilo Se a 14= 18.7 
Orthoclasemae errs sn ate cates 5 T5—= )2080 
las OClASG see ene ce et rate age kasd ao 46= 61.3 
75 100.0 
MD arkerimeral smectic skye) eccluimey she ti. eile 25 
100 


The usual method is to reduce to too, by means of a slide rule, 
the sum of the percentages of the minerals represented in the three 
corners of the diagram, and plot the point so obtained. The 
writer’s method is shown in the figure. Draw three lines, parallel 
to the sides of the triangle, through the points representing the 
amounts of the corner minerals in the rock, in this case a horizontal 
line through 14 for quartz, a line sloping northeast and southwest 
through 46 for plagioclase, and a line sloping northwest southeast 
through 15 for orthoclase. Lay the side of a straight-edge on the 
apices of the two triangles and draw a short line (6d) through 
the small triangle from its apex to its base. Connect either of the 

The seme method may be used in other ways to show the relative percentages 
of thre components whose sum is not a hundred. 


167 


168 ALBERT JOHANNSEN 


lower corners of the large triangle with the similar corner of the 
small one, and indicate its intersection with the line first drawn 
by a dot, F. The line bd and the spot F represent the rock. The 
point Ff is the same as that which would have been obtained by 
the slide rule, and gives the relative proportions of the light con- 
stituents (quartz [mF], orthoclase [RF], and plagioclase [zF]), in 


Quartz 


TITS 
AA TAS AVA AVA 
LN TRIN NEVA 

[\f\/NJ SEP RAIS 
LEPTIN VLAN 


Orth. Plag. 


the rock by its distances from the sides of the triangle opposite 
these names at the corners. The actual percentages of the minerals 
in the rock are also represented: bc or pk for orthoclase, ad or zo for 
plagioclase, bd for the dark constituents, dm for quartz. ° 

In a similar manner the composition of the rock at L may be 
read directly from the diagram, giving the values: 


CUA TEZ, ai ste oe ie Se nya talent Sct ate Uae 23= 27.7 

Orthoclase we nsGsk ocecwhlc eee eyecare 45= 2 

Pla IOClASE sie ee ca'ho teas cae a: supe Aenea I5= 18 
83 100.0 


THE REPRESENTATION OF IGNEOUS ROCKS 169 


The rock at S is seen to consist of orthoclase 27 per cent, 
plagioclase 42 per cent, quartz o per cent, and dark minerals 
31 per cent. The relative ratios of the light constituents are: 
orthoclase 39.1 per cent, plagioclase 60.9 per cent, and quartz 
-o per cent. The rock at T consists of orthoclase o per cent, 
plagioclase 66 per cent, quartz 24 per cent, dark constituents 
ro per cent. Relative ratios of the light minerals, plagioclase 
73.5 per cent, quartz 26.5 per cent, and orthoclase o per cent. 

The actual plotting takes much less time than the telling. In 
fact the small triangle is never plotted. The upper end (0) of the 
line is located at the dividing point between orthoclase (6c) and 
plagioclase (ab) on the horizontal line whose length (ac) is equal to 
the sum of the feldspars, and through this point the inclined line 
is drawn to the percentage position of quartz (d).. One of the 
lower corners of the small triangle is located, without drawing it, 
and the intersection of the inclined line with the ferromagnesian 
mineral line used to determine the locus of the rock. 


PETROLOGICAL ABSTRACTS AND REVIEWS 


ALBERT JOHANNSEN 


ADAMS, SIDNEY F. ‘“‘A Microscopic Study of Vein Quartz,” 
Econ. Geol., XV (1920), 623-64. 48 figs. on 8 pls. 
This study of the microscopic characteristics of vein quartz is con- 
fined to quartz of hydrothermal origin; magmatic, metamorphic, and 
replacement quartz are not included. A well illustrated and instructive 


paper. 


ALLEN, E. T., and LomBarp, Ropert H. ‘‘A Method for the 
Determination of Dissociation Pressures of Sulphides, and 
Its Application to Covellite and Pyrite,” Amer. Jour. Scz., 
XLII (1917), 175-95. 
A secondary enrichment investigation of the Geophysical Laboratory, 
in which methods and apparatus are described. 


ANDERSEN, OLAF. ‘‘On Aventurine Feldspar,”’ Amer. Jour. Sct., 
IG (GOrs), ssroos Imes, me, jollss 2. 

The schiller in certain feldspars was determined as being due to 
oriented, lamellar inclusions of hematite of various shapes and sizes. 
They originated through the unmixing of an originally homogeneous 
feldspar which contained iron oxides in solid solution. The lamellae 
were found always to be oriented after simple crystal forms. 


BACKLUND, HEtcE. “Petrogenetische Studien an Taimyrge- 
steinen,” Geol. Foren. i Stockh. Forhandl., XL (1918), 101-203. 
Figs. 11, map I. 

This is a petrological study based upon about 500 specimens col- 
lected by the unfortunate Baron v. Toll in his expedition to Taimyr 
Peninsula, in northern Asia. The region is made up of gneisses, mica, 
and other schists, contact hornfeld-like rocks, and three types of granite. 
Many of the rocks were analyzed, and they are computed into norms 
and into the Osann system, while under granite the percentage modes 


170 


PETROLOGICAL ABSTRACTS AND REVIEWS 171 


also are given. Besides detailed descriptions of the various rocks, there 
is a good discussion of movement in solidified rock, illustrated by numer- 
ous diagrams. 


BaRRELL, JOSEPH. ‘Relations of Subjacent Igneous Invasion to 
Regional Metamorphism,” Amer. Jour. Sci., I (1921), 1-10, 
AO OF 2515-017: 

This paper, written by the late Professor Barrell in 1913 or 1914, 
has been edited and seen through the press by Frank F. Grout, and no 
better abstract of it can be given than Mr. Grout’s own summary. 

Evidence is presented that batholithic invasions widen downward and may 
occur close below many rocks where they have not been suspected. Batholiths 
like those in the American Cordillera seem to come to place without crustal 
compression, but those of the Archean shield and those of the later Appalachian 
invasions are accompanied by compression. A detailed study of three or 
four regions shows the metamorphism to be related to the igneous invasion 
more than to the depth and pressure. One of the regions of deepest burial 
and close folding in Pennsylvania shows slight metamorphism. 

The action of magmas, both by heating and metasomatism, is reviewed. 
The solutions are not meteoric in origin. The results in minerals depend on 
equilibria—largely on the presence of HO and CO,. The depth of 
anamorphism may be small, due (1) to weakness of some rocks, (2) to invasion 
of batholiths. An argument for shallow depth is based on the completeness 
of Archean metamorphism and the salt of the ocean as a measure of erosian. 

The features of metamorphic rocks are reviewed and interpreted as due to 
one or another factor. Major factors are batholithic invasion and compression. 
Movements of solutions, selective crystallization, lit-par-lit injection gneisses, 
and the alternation of injection and mashing, each leaves its marks. 


Bartrum, J. A. “The Conglomerate at Albany, Lucas Creek, 
Waitemata Harbour,” Zrans. New Zealand Institute, LII 
(7920), 422-30. Pls. 2. 


The conglomerates of the Albany Riverhead district (probably of 
Upper Miocene age) contain pebbles of greywacke, argillite, granodiorite, 
quartz-diorite, diorite-gneiss, diorite, anorthosite, dolerite, andesite, 
trachyte, and rhyolite. In this paper the igneous rocks are briefly 
described and four photomicrographs are given. It is suggested that 
the gneissic rocks in conglomerates, here and elsewhere in the North 
Island, perhaps furnish evidence of a terrain injected by batholitic 
intrusions, subjected to compressional stresses, and eroded before the 
deposition of the main mid-Mesozoic sedimentaries. 


172 PETROLOGICAL ABSTRACTS AND REVIEWS 


Bartroum, J. A. ‘‘ Additional Facts Concerning the Distribution of 
Igneous Rocks in New Zealand,” Trans. New Zealand Institute, 
DOIG ro r0)), 46245 ses ae 

Brief descriptions are given of hypersthene-basalt, troctolite, 
granodiorite with epidote (which the writer, not the reviewer, thinks 
primary), basalt with biotite (character of the feldspar of the rock not 
given), hornblende-basalt, andesite, diorite, and trachyte. Most of 
the descriptions are too brief and incomplete to permit passing judgment 
on the names. 


Bartrum J. A., ‘Additional Facts Concerning the Distribution of 
Igneous Rocks in New Zealand: No. 2,” Trans. New Zealand 
Institute, LII (1920), 416-22. Figs. 5. 


Here are brief descriptions of norite, dolerite, basalt, and hypersthene- 
andesite, and one more extended of quartz-norite. The dolerite is 
apparently the diabase of United States usage. The quartz-norite is 
described as ‘‘a moderately typical norite but for two considerations: 
first the plagioclase . . . . is somewhat acid, being in the main andesine- 
labradorite; secondly, there is... . a little interstitial quartz.” <A 
third objection which might have been given is the fact that besides 
hypersthene there is abundant augite, biotite, some hornblende, and 
“probably a third pyroxene.” With a feldspar more acid than Ab,.An,o 
why not call the rock quartz-hypersthene-diorite? The “third pyrox- 
ene’”’ is described in considerable detail. All but one of its properties, 
including orientation, agree with hypersthene, the exception being the 
“extinction angle which was found to be from very small to 42°.” May 
this mineral not be hypersthene? ‘The reviewer has found that this 
mineral, in many cases, gives apparently inclined extinction in sections 
which are cut at right angles to the principal sections and yet show only 
one set of prismatic cleavage lines brought out by the grinding. Meas- 
suring then from the cleavage lines, the extinction is inclined, but it will 
usually be seen, if the stage is turned an equal number of degrees be- 
yond the point of extinction, that there are here traces of the other 
cleavage. Random sections cutting all three axes in orthorhombic 
crystals do, of course, show inclined extinctions when measured from 
cleavage lines. Zoisite is mentioned as a mineral originally identified 
as apatite, and surprise is expressed that apatite does not occur al- 
though the analysis gives considerable P,0;. The sketch given of this 


PETROLOGICAL ABSTRACTS AND REVIEWS 173 


mineral may well represent a corroded apatite of a form not uncommon, 
and its occurrence surrounded by magnetite is indicative of the same 
mineral. Unless the fragments were large enough to permit of absolute 
determination as zoisite, the reviewer is inclined to think the original 
identification correct. 


BEcKE, F. Ueber den Monzonit. Festschrift C. Doelter. Dres- 
SecGen and Leipzig, ro20, «14. Fig. 1. 

Monzonite, originally a collective name for all the different rocks 
found on Predazzo and Monzoni, has been variously used. Becke gives 
the following definition: Monzonite is a plutonic rock of granular 
texture. Its essential constituents, in the order of their crystallization, 
are magnetite, augite, hornblende, biotite, plagioclase (average composi- 
tion andesine), and perthitic orthoclase. Small amounts of quartz or 
aegirite-augite or nephelite may occur. Accessory minerals are apatite 
and titanite. The dark and light constitutents, also the orthoclase and 
plagioclase, are of approximately equal amounts. Chemically the An: 
Ab: Or ratios are about 20:45:35; and the Ca:Mg:Fe=30:30:40. 
SiO, fluctuates around the point of saturation, so that quartz or a feld- 
spathoidmayoccurinsmallamounts. Becke objects to the use of the term 
monzonite as applied to rocks of the “Pacific type.” The silica content 
must be below 63-64 per cent. ‘The term thus used is dependent upon 
the chemical composition. Used in the foregoing sense, many rocks 
described as quartz-monzonites are not such at all. 


Benson, W. N. ‘Report on the Petrology of the Dolerites 
Collected by the British Antarctic Expedition, 1907-1909,” 
Geology, Vol. II, of Shackleton’s Report. London, n.d., 
753-00. Figs. 6. 


Dolerites, in this report, are rocks composed essentially of basic 
plagioclase and pyroxene, with varying amounts of quartz. Following 
British usage they are called dolerites (diabases, United States). Since 
the name was originally given by Hauy to coarse-grained basalts, the 
use of the expression aphanitic dolerite seems anomalous. Several 
analyses, recast in the C.I.P.W. system, are given, and the rocks and 
component minerals are described in detail. 


174 PETROLOGICAL ABSTRACTS AND REVIEWS 


Benson, W.N. ‘The Origin of Serpentine, a Historical and Com- 
parative Study,” Amer. Jour. Sci., XLVI (1918), 693-731. 


The author agrees with the general opinion that chrysotile- or 
antigorite-serpentine of large ultrabasic masses is derived from the altera- 
tion of originally intrusive peridotite; the hydration, in some cases at 
least, having been brought about by the agency of waters emanating from 
the same magma as that which produced the peridotite. Whether a 
peridotite which escaped hydration during the igneous epoch can subse- 
quently be changed to serpentine by the action of deep circulating 
epigene waters is regarded as less clear though thought not improbable. 
The reviewer is glad to see followed here his own mode of indicating by 
asterisks, in the bibliography, works not seen by the writer. 


Benson, W.N. “The Geology and Petrology of the Great Serpen- 
tine Belt of New South Wales,” Parts VI (Appendix) and VII. 
Proc. Linnean Soc. New South Wales, XLII (1918), 693-700, 
XLIII (1918), 320-94. Map 1, pls. 10, figs. 5. 


The preceding papers of this series were reviewed in Jour. Geol., XXV 
(1917), 493-95. In the present papers the rocks of the Attunga and 
Loomberah districts, and of a portion of the Goonoo Goonoo Estate 
are described. Two analyses of dolerites are given and keratophyres, 
dolerites, albite-dolerites, and granophyres are described rather com- 
pletely, although much-to-be-desired model percentages are not given. 


Berek, M. and Jentzscu, F. ‘Ein kleiner lichtstarker Mono- 
chromator, besonders fiir mikroskopische Beobachtungen,” 
Zeitschr. f. Instrumentenkunde, 1914, 47-51. Figs. 2. 


Describes a small monochromator for the production of mono- 
chromatic light. The instrument is only half as large as a petrographic 
microscope, and is so arranged that the whole spectrum may be made to 
pass through the opening without adjusting the light source or the 
microscope. The visible portion of the spectrum is normally 12 mm. 
but the emergence slit may be adjusted to any width. The emerging 
ray may be made parallel, convergent, or divergent by means of a lens 
in a sliding tube. 


REVIEWS 


Life of James Hall, Geologist and Paleontologist, 1511-1898. By 
J. M. Crarxe. Published by S. C. Bishop, Albany, N.Y., 
1921. 8vo, pp. 565, pls. 14. Price $3.70 postpaid. 


In this handsome volume of 565 pages, the deft pen of Dr. Clarke 
has set forth the life and times of James Hall in a most illuminating and 
fascinating way. While the story holds with tenacity and strict fidelity 
to the realities of Hall’s remarkable career, the handling of its varied 
incidents is none the less most skilful, and the book is worthy of being 
read as a work of art. All the way through it bears evidence of unspar- 
ing labor in searching out from voluminous correspondence and tedious 
official records the essential details that give precision and life to the 
sketch. It is written from a deeply sympathetic and appreciative point 
of view, but the angularities and irascibilities of a strong and intensely 
earnest man are given their due place in the portraiture. These are 
touched with wonderful skill, so that they seem merely incidental 
shadows artistically designed to set off the high lights of the picture. 

The book, however, is far more than a biographical sketch; it 
carries the reader into the very heart of investigative work as it was in 
those early times, while incidentally not a little light is thrown on the 
status of the allied sciences of the day. It gives most interesting glimpses 
of the leading scientific men of the time, including many of the foremost 
men in other than the geological sciences. Dr. Hall had a wide acquaint- 
ance with these, and the abundant extracts from his intimate correspond- 
ence with them shed suggestive side lights on the personal aspects of 
pioneer investigation and the troubles, political and otherwise, which 
attended it. We doubt if there is any other text that carries the reader 
so close home to the inner history of our own and allied sciences in this 
country during the early and middle stages of the last century. No 
student of geological and paleontological progress should miss the 
opportunity to read this thesaurus of information on a most vital 


stage of early American science. 
DREN? 


175 


1700.2 REVIEWS 


The Geology of Northeastern Rajputana and Adjacent Districts. By 
A. M. Herron, B.Se., F.G:S., Assoc. Inst; ‘C.E 3) Assist 
Superintendent Geological Survey of India. Calcutta: Memoirs 
of the Geological Survey of India, Vol. XLV, Part I, 1917. 
Pp. 128, pls. 26. 


This report gives the results of a general geological survey of the 
region started in 1908, made with special reference to stratigraphy. 
The formations present include only the pre-Cambrian overlain by 
unconsolidated post-Tertiary beds. ‘The chief physiographic feature is 
that of an ancient, folded, mountain complex, in the last stages of 
denudation, surrounded by gently sloping plains. There is a general 
discussion of local correlations and nomenclature. 

There is no mineral wealth of any great importance. Quartzite is 
quarried rather extensively, but chiefly for local use. The report is 
accompanied by a geologic map and a number of structure sections. 

A.C. Mick: 


The Phosphate Deposits of South Australia. By R. LOCKHART 
Jack, B.E., F.G.S. Adelaide: Geological Survey of South 
Australia, Bulletin No. 7, t919. Pp. 135. 


The author considers briefly the mode of occurrence, sources, method 
of working, and the preparation of the rock phosphate. Both guana 
deposits and rock phosphate are worked. The latter is associated with 
sedimentary rocks of Cambrian or even possibly pre-Cambrian age 
which are usually highly metamorphosed. It is almost invariably 
found in close association with the limestones and marbles or the adjoin- 
ing argillaceous rocks. Whether the phosphate is primary or secondary 
is not known. Brecciation of the associated rock is conspicuous even 


at considerable depth. 
AC Miche 


Systematic Report on the Cambrian and Ordovician of Maryland. 
By R.S. BAsstER. Maryland Geological Survey, 1919. 


In a review of this report in the last issue of this Journal, it was 
stated that the proposed Ozarkian and Canadian systems “are not 
recognized”’ instead of “are recognized.”’ It is the desire of the reviewer 
to call attention to this correction. The two systems are not only 
recognized but are discussed in some detail, the Conococheague lime- 
stone being referred probably to the former, and the Beekmantown 


limestone to the latter. 
A. C. McF. 


PIIYSIOGRAPHIC DIAGRAM 


OF THE 
UNITED STATES 


” 
ALK, LOBECK 
onty 


0 


Tar 


AJ.Nrstrom & Co, 


Puncisness = Chicago 


-PHYSIOGRAPHIC DIAGRAM OF 
_ THE UNITED STATES 


Size 6244 inches, Scale 50 miles to one inch 


By A. K. LOBECK 
Department of Geography, University of Wisconsin 


This. vices is intended to be used asa guide in teaching the physiography. and 
geology of the United States. The topographic features are expressed with the idea 
of revealing or suggesting the underground structure, and the general character of 
the underlying rock. 


An introductory drill such as having students draw cross sections through the - 
Adirondack uplift, the Cincinnati or Nashville Domes, across the Appalachians in 
different places, through the Black Hills, Central Texas, the Ozarks, Wisconsin and 
many other regions is followed up with the use of other available material, espe- 
cially state geological maps and the United States Geological Survey topographic 
sheets. 


The student frequently fails to-note the significance of a locality with relation to 
the physiographic province or geological region as a whole. . This is overcome by 

' requiring the student to locate upon the diagram the position of the topographic 
quadrangles under consideration. As practically every topographic map published 
by the Survey was consulted in the preparation of the Diagram, most of the signifi- 
cant features are recognizable in the Diagram too. Seventeen geological provinces 
are discussed briefly on the map. 


PRICES 


Student edition (one color) with one-half of U.S. on each side of sheet, size 31 mee 
inches. 


Single copy fe RUE $1.25 Fifty copies, each......... $1.00 
Ten copies, each...... I.10 One hundred copies, each.. .90 
Two color edition with entire map on same side of sheet, size 62 X44 inches. 
3 Single copy,..... . $2.25 Fifty copies, each Sn $1.75 
Ten copies, each...... 2.00 One hundred copies, each.. 1.50 


Prices on two color edition, mounted on cloth for permanent class room use, size 
6244 inches. 


With Gan teniter ab Ropyandsbsottomie sa... ans cstaereee ee $4.50 
On Spring Roller and Board with Dust Proof Cover Sena eae: 2n 7150 
‘In SteeluS pring wWoller: @asecas cet Wher asc ele v0 owieieia ele . 8.75 


AL. OS. NYSTROM & COMPANY 


Publishers —- 2249 Calumet Ave., Chicago 


It makes all Writ- 
ing Swift and 
Easy 


Remington 
Portable 


A little brother of the big 
Remington. Small but 
strong. Light but efficient. 
Compact but complete. 


EOLOGISTS and all readers ge 
literature will welcome the pub 
in book form, of an. imp ‘se. 

Has every feature common to the big 


machines—including automatic ribbon reverse 
and standard keyboard. 


and opinion concerning many bag hed 
mental problems of North American 
and form an admirable supplement to 
treatises and manuals. 
The value! of the book is coe e anc 
by the fifteen paleogeographic maps by Bail 
Wis which accompany the ae : 


Price, with case, $60 


Remington 
Typewriter Company 


(Incorporated) 


374 Broadway 
New York 


Branches 
Everywhere 


The University of Chicago P 
Chicago + NOT Vine gto a 


-LEITZ PETROGRAPHICAL 
MICROSCOPE “K. Mg 


for routine investigations 


It is provided with the same features as our large research 
model viz. 
Wide tube and eyepieces of enlarged dianiehers which render 
a field of view 50% larger than those of other manufactu 
Improved illuminating systems. 
Its liberal size permits the accommodation of all existing acc 
sories, including “‘Federow” Universal-Rotating-Stage. MF 


accessories and are able to meet all requirements for reseai 

or elementary. investigations. : &. 
The microscope “‘S.M,” for sya rotation of bo he 

nicols i is our latest achievement. on 


Leitz Microscopes are recognized throughout the SF, as Jeadit ‘ 
and are invariably selected by petrographists who Paes quality 


60 East lO@St. 


es ies ey Cts a NUMBER 


ee oe THE 


EDITED BY 
THOMAS Cc. _ CHAMBERLIN AND eo EST D.S 


With the Active Collaboration of. 


WE LLER, Beater Paleontology P ALBERT JOHANNSEN, Petrology 
ASTIN, Economic Geology ROLLIN T. CHAMBERLIN, Dynamic Geology - 


ASSOCIATE EDITORS ‘ 


B. GEIKIE, Great Britain ; JOHN C. BRANNER, Leland Stanford Junior University 
\RROIS, France os RICHARD A. F. PENROSE, Jr., Philadelphia, Pa. 
CHT PE CK, Germany WILLIAM H. HOBBS, University of Michigan 
U eae AEN FRANK D. ADAMS, McGill University S: 
) EGER, Sweden j ; CHARLES K. LEITH. University of Wisconsin 
ED EWORTH DAVID, Australia : WALLACE W. ATWOOD, Clark University 
eland Stanford Junior University WILLIAM H. EMMONS, University of Minnesota 
TA coTT, Smithsonian Institution — _ARTHUR L. DAY, Carnegie Institution 


ION PRINCIPLE IN PETROGENESIS Deby SAS Nop owen aie 


OF a PHYSIOGRAPHIC HISTORY OF NORTHEASTERN ONTARIO 
fi W. H. Corrs 199 


( RRENCE ¢ OF CRISTOBALITE IN CALIFORNIA - Austin F. Rocers 211 
E ‘URES OF SALTON BASIN, CALIFORNIA - = = Joun S. Brown 217 


PER CRETACEOUS. AND A NEW ECHINOCORYS FROM THE ALTA- 
, OF BOLIVIA - - - - - iia - Epwarp W. BERRY 227 


R COURSES OF THE ANDROSCOGGIN Biv - - TrvING B. CrosBy 232 


HE: UNIVERSITY OF CHICAGO PRESS 
: CHICAGO, ILLINOIS, U.S. A. 


; THE CAMBRIDGE ‘UNIVERSITY PRESS, Tousen 
fy THE MARUZEN-KABUSHIKI- KAISHA, Toxyo, Osaka, Kyoto, FUKUOKA, SENDAI age 
} THE ae 2s BOOK COME SHANGHAI 


aie 


April-May 1922 agg ES eel aa nen 


THE gOURNAES ae 


i 


. EDITED BY » 
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY 
With the Active Collaboration of : \ Wie: f . 
seen WELLER ni ' ALBERT JOHANNSEN 
Invertebrate Bajeorentase ; ~ Petrology 
EDSON. S. BASTIN ROLLIN T. CHAMBERLIN 
Economic Geology — Dynamic Geology 


% 


The Journal of Geology is published by the University of Chicago at the University of Chicago Pr ess 


May 1, June 15, August 1, September 15, November 1, December 15: {[ The subscription price is $. 
per year; the price of single copies is 65 cents. Orders for service of less than a half-year will be char: 
at the single-copy rate. | Postage is prepaid by the publishers on all orders from the United States 
Mexico, Cuba, Porto Rico, Panama Canal Zone, Republic of Panama, Dominican Republic, El Salvador, 
Bolivia, Colombia, Honduras, Nicaragua, Peru, Hawaiian Islands, Philippine Islands, Guam, Samoan 
Islands, Shanghai. { Postage is charged extra as follows: for Canada, 25 cents on annual subscriptions 
(total $4.25); on single copies 3 cents (total 68 cents); for all other countries in the Postal Union 
50 cents on annual subscriptions (total $4.50), on single copies 6 cents (total 71 cents). {] Patrons a1 re 
requested to make all remittances payable to The University of Chicago Press in posta. or HEADERS money 
orders or bank drafts. a 
The following are authorized to quote the Brees indicated: “a 
For the British Empire: The Cambridge University Press, Fetter Lane, London, E.C. 4. Yearly 
subscriptions, including postage, 27s. each; single copies, including postage, 4s. 3d. each. i 
For Japan and Korea: The Maruzen-Kabushiki-Kaisha, rr to 16 Nihonbashi Tori Sanchome, 
Tokyo, Japan. Yearly subscriptions, snelucine, postage, Yen 9.04 each; single copies, including 
postage, Yen 1.42 each. I 
For China: The Mission Book Company, 13 North Szechuen Road, Shanehai: Yearly an 
scriptions $4.00; sirigle copies 65 cents, or their equivalents in Chinese money. Postage extra, 
mailed direct outside of Shanghai, on yearly subscriptions 50 cents, on single copies 6 cents. 
Claims for missing numbers should be made within the month following the regular month of publi- 
cation. The publishers expect to supply missing numbers free only when losses have been ‘sustained ir 
transit, and when the reserve stock will permit. ~ ro Nie al 
Business correspondence should be addressed to The University of chicane Press, Chicago, Ill ke 
Communications for the editors and manuscripts should be addressed to the Editors of Tue Jour 
NAL OF GEOLOGY, the University of Chicago, Chicago, Ill. . e 
Contributors are requested to write scientific and proper names with particular care and in citations 
to follow the form shown i in the pages of the Journal. : 
_ Because of the high cost of printing, this Journal has found it necessary to adopt the following rule: 
1. If articles exceeding 25 pages in length are accepted for publication in the Journal, the cost of com- 
position beyond 25 pages shall be paid by the author or by the institution which he represents. at the current 
rates of the University Press. {] 2. The cost of illustrative material, in excess of $1.00 per page for the 
article, beyond the cost of type which would have occupied the same space, shall be borne by the author. — 
{ 3. The cost of corrections in proof in excess of 10 percent of the cost of first composition, will be 
‘charged to the author. {] 4. While the cost of printing remains as now, the Journal cannot supply 
prints without charge, bul reprints will be furnished authors at cost, if ordered in advance of Bae 


Entered as second-class matter, March 20, nabs, at the Post office at Chicago, Ill., under the Act of March 3, 1870. 
' Acceptance for mailing at special rate of poste provided for in Section 1103, Act oy October 3, 1917, authorized o 
July 15, 1918. : 


PRINTED IN THE U.S.A. 


Among articles to appear in early 
numbers of the Journal of Geology 
are the following: 


A Criticism of the ‘‘Faunal Relationships of the Meganos Group,”’ 
by Bruce L. Clark. By Roy E. DIcKERSON. 


Post-Glacial Lakes in the Mackenzie River Basin, Northwest 
Territories, Canada. By A. E. CAMERON. 


The Occurrence of an Apus in the Permian of Oklahoma. By 
RUDOLF RUEDEMANN. 


On Contact Phenomena between Gneiss and Limestone in Western 
Massachusetts. By PENTTI ESKOLA. 


The Early Pre-Cambrian Formations of Northern Ontario and 
Northern Manitoba. By E. L. BRUCE. 


Dinosaur Tracks in Hamilton County, Texas. By W.E. WRATHER. 


The Age of the Domes and Aniticlines in the Lost Soldier-Ferris 
District, Wyoming. By A. E. Fatu. 


A Scale of Grade and Class Terms for Clastic Sediments. 


By CHESTER K. WENTWORTH. 


Cambridge Gntversity Press 


The Study of Geological Maps. By GertruvE L. ELtes, D.Sc., Fellow of Newn- — 
ham College, Cambridge. Crown 4to. With 7 plates, 64 figures, and a map. Price on — 
application. Cambridge Geological Series. "4 


A Manual of Seismology. By Cuartes Davison, Sc.D. With 100 illustrations, _ 
Demy 8vo. Price on application. . 


Petrology for Students: an Introduction to the Study of Rocks under A 


the Microscope. By Atrrep Harker, M.A., F.RS., F.G.S._ Fifth edition, revised. 
Crown 8vo. $3.75. 


A Treatise on Crystallography. By W. J. Lewis, M.A. Demy 8vo. $5.00. 


The Principles of Stratigraphical Geology. By J. E. Marr, Sc.D., F.R.S., F.G.S. 
Crown 8vo. $2.50. 


An Introduction to Geology. By J. E. Marr, Sc.D., F.R.S., F.G.S. Crown 8vo. 
$2.50. 


Published by the Cambridge University Press (England) 
The Macmillan Co., Agents in the United States 
64 and 66 Fifth Avenue, New York City 


A Living Monument to 


AGASSIZ 


A Joy and Benefit to Students 
and Lovers of Nature 


Please co-operate in the great work of 


THE AGASSIZ ASSOCIATION, Inc. 


Forty-seven years of effective service 


Science, Education and Humanity 


The Guide to Nature 


Monthly Illustrated Magazine 
EDWARD F. BIGELOW, Editor 
Subscription, $1.50 per year; single copy, 15 cents; 


three months’ trial, 25 cents 


Please write for particulars 


ARCADIA 
SOUND BEACH CONNECTICUT 


Starved Rock State Park 


and Its Environs 


By CARL O. SAUER 
Department of Geology, University of Michigan 


GILBERT H. CADY 
Illinois State Geological Survey 


HENRY C. COWLES 
Professor of Ecology, University of Chicago 


This thorough and interesting geographical | 


study of the best-known feature of the TIli- 
nois Valley has been prepared for the benefit 


of those who are interested in knowing more |] 


about the region than can be gathered by 
personal observation in a few hours or a few 
days spent in the park. The physical geog- ~ 
raphy of the park and its surroundings has 
a number of features, such as the beautiful 
little canyons, which are unusual in this part 
of the country. Part I is given up to the 
geography of the park, its surface features 
and their origin, and the exploration, settle- 
ment, and development of the region; Part }j 
II, to the geology of the park; and Part III, }} 
to its botany. 


130 pages, cloth, $2.00, postpaid $2.15 


THE UNIVERSITY OF CHICAGO PRESS |] 


CHICAGO - - - - ILLINOIS | 


A NEW BOOK 3 o‘GFr NATURE 


A NATURALIST IN THE 
GREAT LAKES REGION 


By ELLIOT R. DOWNING 


Associate Professor of Natural Science in the 
School of Education of the University of Chicago 


SIME person who loves the out-of-doors and 
wants to know more about Nature “in her 
| visible forms” will be interested in this new 
“| book by Mr. Downing, who is recognized as 
one of the leading naturalists of the United States. He 
knows the region ‘‘inch by inch” and presents many inter- 
esting facts about the various forms of life which inhabit 
prairie, forest, and dunes. 


-@ The volume traces the geological and physiographic 
development of some of the more important features of the 
region, showing how the distribution of plants and animals 
_ is dependent upon these features. 


Profusely illustrated, flexible binding, xxv1+325 pages 
$3.50, postpaid $3.65 


Take this book with you on your vacation 


THE UNIVERSITY OF CHICAGO PRESS 
CHICAGO - - - - ILLINOIS 


The Geography of the Ozark Highland of Missouri 


By CARL ORTWIN SAUER, the University of Michigan 
(Publications of the Geographic Society of Chicago) 


The Ozark Highland of Missouri was selected for investigation because of its 
unusual wealth of geographic responses and because little is known concerning its con- ~ 
ditions and possibilities. j 

The purpose of such a study is twofold: to furnish an adequate explanation of 
the conditions of life in a given area, and to contribute proved statements which will — 
aid in working out fundamental principles. } 

The book is divided into three parts. The first is an outline of the environment. — 
Only those things which are pertinent to an understanding of the conditions under which 
the people live are introduced. 


The second part considers the influence of environment on the settlement and develop- — 
ment of the different parts of the highland. ie 


The third part is a study of the economic conditions as they exist today. In conclusion ~ 
a forecast is offered of the lines along which the future of the region will be worked out. r: 


A valuable feature of the volume is the 44 figures in the text and 26 plates. 


xviti+246 pages, Svo, cloth; $3.00, postpaid, $3.20 


The University of Chicago Press 
Chicago, Illinois 


THE ORIGIN OF THE EARTH 


By THOMAS C. CHAMBERLIN 


Head of the Department of Geology, The University of Chicago 


One of “The University of Chicago Science Series” 
xli-+272 pages, 8vo, cloth; $1.75 net, postpaid $1.90 


FROM THE PREFACE 


“Tn telling the story of this search for the mode by which the earth came 
into being, we have let the incidents that led the inquiry on from one stage to 
another fall in with the steps of the inquiry itself. It is in keeping with the 
purposes of this series of booklets that the motives which set researches going 
should have their place with the quests that arose from them The final 
story of the birth of the earth will come only after a time, when the vestiges of 
creation have been more keenly discerned and more faithfully rendered than is 
possible now.’ 


THE UNIVERSITY OF CHICAGO PRESS 
CHICAGO . : ILLINOIS 


PHYSIOGRAPHIC DIAGRAM 


oF THR 


UNITED STATES 


AJ.Nystrom & Co. 
Crcaao 


PUBLISERS, 


PHYSIOGRAPHIC DIAGRAM OF 
THE UNITED STATES 


Size 62X44 inches, Scale 50 miles to one inch 


By A. K. LOBECK 


Department of Geography, University of Wisconsin 


This diagram is intended to be used asa guide in teaching the physiography and 
geology of the United States. The topographic features are expressed with the idea 
of revealing or suggesting the underground structure, and the general character of 
the underlying rock. 

An introductory drill such as having students draw cross sections through the 
Adirondack uplift, the Cincinnati or Nashville Domes, across the Appalachians in 
different places, through the Black Hills, Central Texas, the Ozarks, Wisconsin and 
many other regions is followed up with the use of other available material, espe- 
cially state geological maps and the United States Geological Survey topographic 
sheets. 

The student frequently fails to note the significance of a locality with relation to 
the physiographic province or geological region as a whole. This is overcome by 
requiring the student to locate upon the diagram the position of the topographic 
quadrangles under consideration. As practically every topographic map published 
by the Survey was consulted in the preparation of the Diagram, most of the signifi- 
cant features are recognizable in the Diagram too. Seventeen geological provinces 
are discussed briefly on the map. 


PRICES 


Student edition (one color) with one-half of U.S. on each side of sheet, size 31 X44 
inches. 


Singlecopys see ee $1.25 Fifty copies, each......... $1.00 
Ten copies, each...... I.10 One hundred copies, each.. .90 
Two color edition with entire map on same side of sheet, size 62 44 inches. 
Singlercopyseneeee eee: $2.25 Fifty copies, each Jahan $1.75 
Ten copies, each...... 2.00 One hundred copies, each.. 1.50 


Prices on two color edition, mounted on cloth for permanent class room use, size 
62 X44 inches. 


With Common Rollers at Top and Bottom.................-- $4.50 
On Spring Roller and Board with Dust Proof Cover........... 7.50 
In Steel Spring Roller Case.............------ se seereceeeee 8.75 


A. J. NYSTROM & COMPANY 


Publishers 2249 Calumet Ave., Chicago 


Ward’s Natural Science Establishment 


ROCHESTER, NEW YORK 


Relief Maps 


This map, 29" x 46", scale 32 miles to the inch, with a vertical exaggeration of 20 to 1, costs 
only $30.00. The same map, hand colored to show the geology as given in United States 


Geological Survey professional paper No. 71, is $125.00. Write for circular G-233 describing our 
relief maps. 


Fossils 


We are better prepared than ever before to furnish index fossils and exhibition material. A few 
of our recent acquisitions follow: 


Cambrian material collected by Dr. Wolcot in British Columbia. 

Rare Ordovician material from New York and Bohemia, such as Aeglina, Proetus, Cryptolithus, 
Illaenus, Cystids, etc. 

Rare Silurian forms from Bohemia such as Ascoceras and Ophidioceras, Cromus, Acidaspis, etc. 


Rare Devonian material from Germany such as Harpes, Cerarges, Goniatites showing sutures, 
Hexacrinus, Cupressocrinus, etc. 


Fine Solenhofen fish, crustaceans and insects, 
Comanchic material from Texas, 

Eocene leaves from Texas and Louisiana, 
Insects in amber. 


Rock Sections 


We have designed and had made a machine for the preparation of microscopic slides of rock 
sections, and believe we can fill your requirements quickly and in a satisfactory manner. Let 
us send you our special announcement circular. 


WARD’S NESUISEN UE ESTABLISHME® 
ROCHESTER : : NEW YORK 


ERRATA 


Journal of Geology, Volume XXX, on page 164, line 2 from the bottom, 
and page 165, line 2 from the top, read Radley, instead of “Bailey.” 


VOLUME XXX NUMBER 3 


Welle 


JOURNAL OF GEOLOGY 


April-May 1922 


THE REACTION PRINCIPLE IN PETROGENESIS 


N. L. BOWEN 
Geophysical Laboratory, Carnegie Institution of Washington 


INTRODUCTION 

It is now many years since petrologists first began to think of 
the crystallization of a molten magma in terms of the physico- 
chemical principles governing the behavior of solutions. In the 
study of ordinary solutions a condition frequently found was that 
known as the eutectic relation. In the simple case of two com- 
ponents, each lowered the melting-point of the other to the temper- 
ature of the eutectic point, at which temperature both solids 
separated side by side from a solution of fixed composition, the 
eutectic mixture. This case and the analogous condition in sys- 
tems of more components are now so familiar to petrologists 
that they need not here be enlarged upon. The concept of the 
eutectic was early seized upon by petrologists and has been one of 
great utility in petrogenic theory. It accounted for the low 
melting temperatures of mixtures of minerals that are individually 
highly refractory. It threw light on some of the factors governing 
the separation of minerals from their mutual solution. But most 
of all, it stimulated the tendency to think of magmas in the light 
of the laws of solutions, or, better, of phase equilibrium, and 
encouraged experimental research whose expected result was the 
location of the composition of the eutectics for chosen mineral 


177 


178 N. L. BOWEN 


mixtures. In investigation of this kind, both on its theoretical 
and experimental sides, Vogt took a leading part and the importance 
of his pioneer work cannot be overestimated. 

As might have been anticipated, it was soon found that the 
eutectic relation did not always obtain, for, after all, it is but one 
of many possible relations between two or more components in 
any system. Very early it was realized that the existence of solid 
solution between two components might eliminate the eutectic 
relation between them. Later, as more accurate methods of 
experimentation were introduced, examples of incongruent melting 
were found, which again lead to the elimination of the eutectic. 
Indeed, it seems that relations between the components such that 
the eutectic disappears are so common in minera] systems that it 
may be doubted whether considerations based upon the doctrine 
of eutexia can longer be regarded as of any great service to petro- 
genic theory, in spite of their great suggestive utility in the past. 

It should be realized that these non-eutectic relations should 
not be termed anomalies nor even apparent anomalies. They are 
just as definitely the result of the laws of solution as are eutectics. 

This paper has been written with the object of readjusting the 
emphasis in our conception of mineral relations. It will be pointed 
out that another relation between the phases, here called the 
reaction relation, is very common in silicates, and is at the same 
time a very much broader concept, with which to approach the 
study of petrogenesis, than the simple doctrine of eutexia, so 
that it might well be set up as a principle to serve as a guide and 
stimulus to the search for facts. 


OF CONTINUOUS REACTION SERIES 


The reaction relation in a common form is illustrated by the 
case of the plagioclase feldspars. The equilibrium diagram of the 
plagioclases (Fig. 1) is of a simple and familiar type.* Those 
features of it that require emphasis in the present connection may 
be brought out very briefly. The crystallization of any random 
liquid, say Ab,An,, takes place in the following manner. At 1450° 
crystals of the composition Ab,An, begin to separate. As the 


tN. L. Bowen, American Journal of Science, Vol. XXXV (1013), p. 583. 


THE REACTION PRINCIPLE IN PETROGENESIS 179 


temperature falls the amount of crystalline matter increases and at 
the same time changes in composition along the solidus curve. 
Thus at 1370° the crystals have the composition Ab,An,. In other 
words, the first-formed crystals have suffered a change of composi- 
tion; indeed, they are continually modified in composition by 
reaction with the liquid. Crystals and liquid mutually influence 
each other throughout 

the course of crystal- 
lization. 

All of this is now an 
old story and is repeated 1400 
here merely to empha- 
size ‘the marked differ- 
ence in crystallization 
between a mixture in 
such a system and one 
in a eutectic system. 
In this latter, a crystal 
once separated is no "°F 
longer concerned in the 
equilibrium, which 
makes it in every respect 
a special case whose 
occurrence might be expected to be comparatively infrequent, 
particularly in more complicated solutions (magmas). 

On account of the continual reaction relation between crystals 
and liquid in a solid solution series such as the plagioclases, it is 
proposed, for the purposes of the present paper, to call such solid 
solutions a continuous reaction sertes. ‘The term will apply to any 
solid solution series, whether to a complete series such as the plagio- 
clases or to each separate series in the case of incomplete solid 
solution. 

The essential feature of a reaction series, the reaction relation 
of crystals and liquid, is retained when the series becomes a part 
of amore complex system. This is true even when the end members 
of the series bear a eutectic relation to the newly added component, 
as is well shown when diopside is added to plagioclase. The 


1500 


1300 


1200 


ALBITE WT. PER CENT ANORTHITE 


Fic. 1.—Equilibrium diagram of the plagioclase 
feldspars. 


180 N. L. BOWEN 


equilibrium diagram of this system is given in Figure 2. Again it 
is a familiar diagram that has been fully discussed in other places,? 
but certain features of it require emphasis in the present connection. 

A mixture of composition F (diopside 50 per cent, Ab,An, 50 
per cent) begins to crystallize at 1275° with separation of diopside. 
As the temperature falls the crystals merely increase in amount. 


OIOPSIDE 


D 


ALBITE T AD,AN, t R H ANORTHITE 


Fic. 2.—Crystallization diagram in the system, diopside-anorthite-albite 


At 1235°, when the liquid has the composition G, plagioclase of 
composition Ab;An, begins to separate and thereafter the change 
of composition of the liquid is represented by the boundary curve 
ED. And as crystallization proceeds along the boundary, the same 
reaction relation exists between the plagioclase crystals and the 
liquid as in the binary system. Vogt would call the curve ED a 
eutectic boundary curve or eutectic line and, to be sure, it has some 
properties in common with such a line.? It marks the composition of 


«N. L. Bowen, op. cit., Vol. XL (1915), p. 175. 
2 J. H. L. Vogt, Journal of Geology, Vol. XXIX (1921), p. 420. 


THE REACTION PRINCIPLE IN PETROGENESIS 181 


the liquids that are in equilibrium with two solid phases. Moreover, 
it lies along a valleyin the fusion surface and when the object is to 
emphasize these features no serious objection can be raised to the term 
used by Vogt. On the other hand, such a usage would tend toward 
a non-recognition of the importance of the reaction relation be- 
tween plagioclase crystals and the liquids lying along the boundary 
curve and, in the writer’s opinion, petrology will better be served by 
emphasizing this relation—calling the curve, say, a reaction curve. 
But quite apart from the question of nomenclature, the distinction 
is really one of fundamental importance in petrogenesis. In the 
case of a true eutectic line in a ternary system, the two kinds of 
crystals separating bear what may be called a mere subtraction 
relation to the liquid and once subtracted they are no longer 
concerned in the equilibrium. Moreover, the true eutectic line 
ends in a ternary eutectic point which represents the composition 
that all liquids of the system must finally attain and beyond which 
they never pass. No such rigidity obtains in a system where 
the reaction relation enters. The composition of the final liquid 
depends upon the freedom with which reaction between crystals 
and liquid may take place. If the reaction is complete in the 
example taken, the crystals are continually made over and the last 
of the liquid is used up by the reaction at 1200°, when the composi- 
tion of the liquid is M and all the crystals have the composition Z. 
But if for any reason early crystals do not participate to the full in 
the reaction—and this may occur when zoning of crystals or sinking 
of crystals supervenes—the composition of the liquid may then 
pass beyond M and the final liquid have a composition represented 
by some point between M and D, with corresponding effect on the 
composition of the final crystals. 

This flexibility in the behavior of the liquid is entirely the 
result of the reaction relation between liquid and crystals. The 
course of crystallization is rendered responsive to the conditions 
under which crystallization takes place, for removal of crystals by 
zoning, sinking, or otherwise is entirely dependent on these condi- 
tions. No such response is possible in a eutectic system, for, 


tA similar effect may of course be brought about in the binary system of the 
plagioclases, but it was considered unnecessary to discuss it for each case separately. 


182 N. L. BOWEN 


whatever the conditions, the eutectic is the goal of all liquids; all 
attain it; none pass it; it is a “fen of stagnant waters.”’ 


OF THE REACTION PAIR AND DISCONTINUOUS REACTION SERIES 


The kind of reaction relation introduced by the existence of 
a solid solution series is not the only kind that is of importance 
in crystallizing magmas. Another type of crystallization phenom- 
enon, ordinarily to be thought of as quite distinct from solid 
solution, has, nevertheless, 
consequences of a similar 
nature. This type is illus- 
trated in several systems 
that have been investi- 
Hcg Sys gated experimentally, but 
for our present purpose 
we shall discuss only cases 
showing phases closely 


CRISTOBALITE & LIQUID 


154 ; i [sy A £ 
| rorszenre related to rock minerals. 
CLINO-ENSTATITE CLINO-ENSTATITE & CRISTOBALITE aks é 

The equwiltbiiia 
Mg,Si0, 20 MgsiO, 40 WT. PER CENT 80 SiO. diagram of the system 


Fic. 3.—Equilibrium diagram of the system, MgO-SiO, is shown in 
forsterite-silica. i Fi : 
Figure 3. Aj liguad ior 
composition 42 per cent MgO, 58 per cent SiO, begins to crystallize 
with separation of olivine Mg,SiO, and this continues until 1557° 
when the olivine reacts with the liquid to form the pyroxene clino- 
enstatite. If the opportunity for reaction" is perfect the liquid is 
completely used up by the reaction and the mass consists entirely 
of olivine and pyroxene. If, on the contrary, liquid and olivine are 
not free to continue in intimate contact for any reason, then upon 
completion of all reaction possible under the circumstances, some 
liquid will be left over. This will proceed to crystallize in the 


«The phrase, opportunity for reaction, is used throughout this paper without 
explanation. It is considered that the various factors involved have been sufficiently 
discussed elsewhere. The removal of crystals from a part of the liquid by their 
sinking is a factor limiting the opportunity for reaction as far as that part of the liquid 
is concerned. And so it is with the formation of a reaction armor about crystals or 
the squeezing out of liquid from a crystal mesh. Usually the rate of cooling is the 
fundamental control over these factors and therefore over the opportunity for reaction. 


THE REACTION PRINCIPLE IN PETROGENESIS 183 


ordinary way giving a mixture of the pyroxene MgsSiO, and silica. 
We thus obtain a mass consisting of olivine, pyroxene, and silica. 

Again it is found that the existence of the reaction relation 
introduces a flexibility such that different products are obtained 
with varying opportunity for reaction. Under certain conditions 
no free silica is formed; under others a little may be formed; and 
under still others a considerable amount may appear.. The com-. 
pound Mg,siO, and MgsiO, may be called a reaction pair. By 
this is meant that crystals of the first compound react with the 
liquid to produce the second during the normal course of crystal- 
lization. 

A reaction relation of this latter type may exist between three 
or more compounds and the compounds, arranged in proper order, 
may then be said to constitute a discontinuous reaction series. 
An example of this kind of series is given by the system H,O-K,SiO,- 
SiO, as worked out by Morey and Fenner.* The equilibrium 
diagram is given in Figure 4. A liquid of composition A begins to 
crystallize with separation of K,SiO;. This is joined later by K,Si,O, 
and the two crystallize side by side, the liquid changing in composi- 
tion along the boundary curve between their fields. At Q, the liquid 
reacts with K,SiO, converting it into K,SiO,-5 H,O and if the oppor- 
tunity for reaction is perfect the whole mass solidifies. On the other 
hand, if the opportunity for reaction is limited, some liquid Q, is left in 
excess and it proceeds to crystallize along Q,-Q, with separation of 
K,Si,0, and K,SiO,-5H.O. At Q, the crystals of K.Si,O; react with 
the liquid to produce K,Si,O,-H.O and solidification is complete if 
_ there is complete freedom of reaction. Here again there may be 
some liquid (Q.) left over if reaction is for any reason limited and 
it will proceed to crystallize along Q,-Q, with separation of 
K.Si,O,-H,O and K,SiO,-3H,O. At Q, the crystals of K,SiO,-3H,O 
are converted into K,SiO,-H,O. In this case there is always some 
liquid in excess of that required for the reaction even with perfect 
opportunity for reaction, but there is more of it when the reaction 
is limited. The liquid then crystallizes along the boundary between 
the fields of K,SiO,-HO, and K,Si,O,-H.O, with separation of those 
compounds, until the isotherm of 200° is reached, and if there is no 

t Journal American Chemical Society, Vol. XX XIX (1917), pp. 1173-1229. 


184 N. L. BOWEN 


further cooling some solution is left that does not crystallize 
at all. 

It is plain from the outline of crystallization given that we have 
a reaction pair, K,Si,O, and K,Si,O,-H,O and also a discontinuous 
reaction series, K,SiO,-K,SiO,-$H,O-K,SiO,-H,O. Each member 
of the series is produced from the preceding member by reaction 
with the liquid. The series is distinct from a continuous reaction 


K.Si0. 


K25103sH20 


KoSig0 5 


oe "$1205.40 


H.0 Si0: 
Fic. 4.—Equilibrium diagram of the system, H,O-K.Si0,-SiO, 
series in that changes of composition are discontinuous, taking 
place in steps and not by insensible gradations as in the continuous 
series. 

The consequences of the existence of this reaction series have 
been pointed out in the foregoing. Under certain conditions of 
crystallization we may get simply the three phases K,SiO,, K.Si.0;, 
and K,SiO,;-3H,O. Under other conditions we may get a part or 
all of a long chain of products including the foregoing three and in 
addition K,Si,O,;-H,O and K,SiO,-H,O. Moreover, when crystal- 
lization runs far down in the series we may obtain a water-rich 


THE REACTION PRINCIPLE IN PETROGENESIS 185 


liquid that fails to crystallize at all. The formation of this liquid 
under such conditions is important in connection with the abundance 
of analogous liquid in certain types of magma, say, the granitic. 


RELEASED MINERALS 


A feature of reaction series that is important in its consequences 
is the fact that a compound, which otherwise would not be formed 
from a given liquid at all, may, as a result of failure of complete 
reaction, be stored up in the liquid and appear as a mineral at 
some later stage. Such a mineral may be regarded as released 
because of the existence of the reaction relation and it is at the 
same time complementary in composition to the mineral disappear- 
ing during reaction. ‘Thus in the case of the reaction pair Mg,SiO,- 
MgsiO, free silica is released as a result of failure of complete 
reaction and it is complementary to Mg,siO,. 


THE EFFECT OF THE REACTION SERIES ON THE 
ORDER OF SEPARATION 


We have now examined several examples of the manner in which 
reaction series may render the course of crystallization responsive 
to external conditions. In another important particular the exist- 
ence of the reaction series, continuous and discontinuous, causes 
the process of crystallization to depart from that obtaining in the 
eutectic system. In the crystallization of the plagioclase feld- 
spars a plagioclase always separates before any other plagioclase 
that is less calcic. There is no such thing as the separation of 
calcic plagioclase first from mixtures rich in calcic plagioclase and 
of sodic plagioclase first (followed by calcic plagioclase) from 
mixtures rich in sodic plagioclase, as there would be in the eutectic 
system. And so with the reaction pair and the discontinuous 
reaction series the higher member of the series always separates 
before the lower, if at all. We do not have a condition in mixtures 
of forsterite and clinoenstatite such that forsterite separates first 
from mixtures rich in forsterite and clinoenstatite first in mixtures 
rich in clinoenstatite as there would be in eutectic mixtures. On 


the contrary, forsterite, however small in amount, always separates 
first. 


186 N. L. BOWEN 


Thus the existence of reaction series tends to introduce a fixity 
in the order of crystallization, calcic plagioclase before sodic 
plagioclase, if at all; forsterite (olivine) before clinoenstatite 
(pyroxene), if at all; K,SiO, before K,SiO,-3H,O before 
KeESiO;- LO: 

SERIES WITHIN SERIES 

Both the continuous and discontinuous types of reaction series 
may be much more complicated than the simple examples chosen 
for illustration. The continuous reaction series may embrace 
more than two components, and the reaction between liquid and 
crystals will then be concerned with the adjustment of the relative 
concentrations of all of the components. Likewise the discontinu- 
ous reaction series may consist of any number of members. More 
than this, any member of a discontinuous reaction series may itself 
be a continuous reaction series. This case is exemplified by the 
system diopside-forsterite-silica.. Here we have the discontinuous 
reaction series olivine-clinopyroxene-silica in which the member 
clinopyroxene is itself a continuous reaction series, that is, a solid 
solution series. The crystallization of a liquid of this system will 
illustrate the complexity to which the reaction relation may lead 
even in this comparatively simple system. The liquid D (Fig. 5) 
begins to crystallize with separation of the olivine, forsterite. At 
K the olivine crystals begin to react with the liquid to form clino- 
pyroxene of the composition L, and as the temperature falls the 
liquid reacts, not only with the remaining olivine crystals to make 
them into pyroxene, but also with the pyroxene crystals already 
present to make them more calcic. Thus at F the olivine has 
been completely changed to pyroxene all of which now has the 
composition R. In this case, too, lack of opportunity for complete 
reaction will have an important effect upon the course of the liquid 
and the kind of crystals produced from it. Failure of reaction 
will enrich the liquid not only in silica but also in the more calcic 
pyroxenes. 

CRYSTALLIZATION SERIES IN ROCKS 

A sufficient number of examples of the reaction relation have 
been given to illustrate the more important aspects of it. More- 

1N. L. Bowen, Amer Jour. Sci., Vol. XX XVIII (1914), p. 207. 


THE REACTION PRINCIPLE IN PETROGENESIS 187 


over the examples have in most cases dealt with members of com- 
mon rock-forming groups and the prevalence of reaction series of 
one kind or the other among the rock-forming silicates is indicated 
by these few examples. The data are not at hand—and are not 
likely to be for some time—for a quantitative discussion of reaction 


Fic. 5.—Crystallization diagram of the system, diopside-forsterite-silica. A. fors- 
terite; B, silica; C, diopside; H, clinoenstatite, and H-C, clinopyroxene series. 


series in mixtures corresponding to natural magmas. Neverthe- 
less it is believed that much is to be gained from a qualitative con- 
sideration of this feature of rock-minerals. 

It should be frankly stated that the existence of the reaction 
relation between two phases in a simple system is no guarantee 
of the persistence of an identical relation between them in a more 
complex system. In the case of the phases olivine and magnesian 
pyroxene, for which such a relation exists in the binary system 
MegO-Si0., it is not unreasonable to expect that the relation might 
be modified in more complex systems. Actually, however, it is 
found that the relation persists in all the more complex systems 


188 N. L. BOWEN 


examined, which fact renders it more likely, but by no means 
certain, that the reaction relation obtains in magmatic systems." 
The service rendered by experimental investigation, so long as it 
is confined to a limited number of components, must lie in its 
indicating where a reaction relation is to be expected. We are, 
moreover, instructed as to what we may expect in the way of 
indications of reaction and thus enabled to extend our inferences to 
phases not formed under laboratory conditions. This brings us to 
the question of the criteria of the reaction relation. 

A criterion of the reaction series, common to both the continu- 
ous and discontinuous type, and serving to show their fundamental 
likeness, is simply the tendency of one mineral to grow around 
another as nucleus. In the case of the continuous series this is 
commonly known as zoning of mix-crystals and in the discontinu- 
ous series as the formation of reaction rims, coronas, etc. Thus we 
have plain evidence of this kind, from a wide range of rocks, that 
that the plagioclases constitute a continuous reaction series and 
that pyroxene, amphibole, and mica form a discontinuous series. 
The development of these special structures is, however, dependent 
on particular conditions of consolidation and the lack of such struc- 
ture in an individual case should not be regarded as indicating a 
lack of the reaction relation in that case. It must be, in part, 
from a general survey of mineral relations in igneous rocks that the 
reaction series are deduced. 

Fortunately the continuous reaction series are easily picked 
out, for the mere existence of solid solution or variability of composi- 
tion in a crystal phase is sufficient to establish that phase as a 
continuous reaction series. Their number is legion, all the impor- 
tant igneous rock minerals with the single exception of quartz 
being members of solid solution series. The detection of the 
discontinuous reaction series is not always so easy, and the element 
of judgment enters to some extent. 

As an example of the information to be obtained on this point 
from a general survey of an igneous sequence let us reproduce a 


1 Vogt believes there is no reaction relation but a eutectic relation at higher 
pressures between olivine and pyroxene. Jour. Geol., Vol. XXIX (1921), p. 528. 


THE REACTION PRINCIPLE IN PETROGENESIS 189 


table given by Harker showing the mineral relations in the sequence 
at Garabal Hill.* 
TABLE I 


MINERAL AND Rock SEQUENCE AT GARABAL Hitt (AFTER HARKER) 


oO © 
e/a] 8 
a | S| SS 
a & 3 ) 
oS S) a n o 
9 & || Be || fel ay ee 
Se Sel ep eres 
Le ais |e Sy eed & 
ES | ae | 6 Pa te ie |S 
Hi a Coy + w Noy n ee) fo) 
A. Olivine-diallage-rock ............. sei se se se] =] j= fH] H | = 
eeBIOtite-Gionite.. 0.1.6.5. e520 ess so ff =] = |] sb | ae ole fe 
C. Hornblende-biotite-granite........ t+i—}]—}]—-7}+y4+)+)] ++ 
D. Porphyritic biotite-granite....... so f= | = | = | = | se] se | ar | ae 
BAPE UTILS VEL. 5 cig wie 6 ee cue conve Scie +}—}]—}]—]—]—]+})+)+ 


It will be noted that the minerals appear in a certain order, as 
they might in a system where simple eutectic relations prevailed, 
but they also disappear in a similar order, a feature that is altogether 
foreign to a eutectic system. In a eutectic system no mineral ever 
disappears.? The first-formed mineral is simply joined by another, 
the pair by a third, and so on until all the minerals appear together 
in a final eutectic product. Very different from this is the condi- 
tion actually found, namely, the disappearance of minerals in the 
order in which they appear which is of the very essence of the reac- 
tion series. 

Upon examination in detail it is found that 2, 3, and 4 disappear 
in B, 5 disappears in D, and 6 in E. From this we conclude that 
these phases bear a reaction (not a mere subtraction) relation to 
the liquid and that, as a result of the reactions, phases appearing 
later are formed. We arrive at the definite conclusion that 4, 
5, and 6 constitute a reaction series and at the same time note 
indications that they are but a part of a series containing more 
members. 

* The Natural History of Igneous Rocks, 1909, p. 131. 


2 A case of simple inversion without change of composition would require to be 
excluded from this statement, but it has no importance in rocks at any rate. 


190 N. L. BOWEN 


By piecing together the information to be obtained from the 
examination of such sequences and observation of the structural 
relations of the minerals a conclusion as to the reaction series in 
rocks is to be arrived at. Without going into detail as to the evi- 
dence, an attempt is made below to arrange the minerals of the 
ordinary subalkaline rocks as reaction series. Neither rigid accu- 
racy nor finality is claimed. It is regarded merely as a framework 
upon which others may build, making such modifications and 
additions as may be found necessary. 


TABLE II 
REACTION SERIES IN SUB-ALKALINE ROCKS 
olivines as 
a 
x o calcic plagioclases 
cA 
Mg pyroxenes a oe 
Sy calci-alkalic plagioclases 
Mg-Ca pyroxenes i 
alkali-calcic plagioclases 
amphiboles I 
\gusalic plagioclase 
biotites 


potash feldspar 


muscovite 


’ 


Quartz 


Beginning at the upper end of the series in the more basic 
mixtures we have at first two distinct reaction series, the continuous 
series of the plagioclases and the discontinuous series, olivines- 
pyroxenes-amphiboles, etc. As we descend in these series, how- 


THE REACTION PRINCIPLE IN PETROGENESIS IQI 


ever, they become less distinct, in the aluminous pyroxenes and 
amphiboles a certain amount of interlocking begins and they 
fmally merge into a single series. This is expressed diagrammati- 
cally by a convergence of the series, with a dovetailing of the mineral 
names at first, and finally a joining of the two series by the arrows 
converging upon potash feldspar. Just where the two series merge 
completely is more or less a question, .but it is given closely enough 
for our present purpose in the figure. 

That the series, olivines-pyroxenes-amphiboles-biotites, con- 
stitutes a reaction series is well attested in many rock varieties. 
By this is meant that liquid reacts with olivines to produce pyrox- 
enes, with pyroxenes to produce amphiboles, and with amphiboles to 
produce biotites. In the continually increasing water content of the 
series it is related to the series K,SiO,-K.SiO,-3H.O-K,SiO,-H.O. 
The continuous reaction series of the plagioclases is perhaps the 
best understood series of rock minerals. This is fortunate, for 
the series happens to be of particular importance in that it runs 
through a wide range of conditions and compositions in the rock 
series. We simply have a continual enrichment of the liquid in 
alkaline feldspar with the separation of the potash variety of 
alkaline feldspar as a separate phase when it has exceeded its 
solubility in the plagioclase mixture. With the formation of pot- 
ash feldspar in the one series and of biotite in the other, the two 
‘series are now so intimately intermingled as to constitute a single 
series. 

There is a little of the nature of eutectic crystallization in the 
‘crystallization series given in the foregoing. At early stages and 
as between the two series there is some suggestion of the eutectic 
relation in that a member of one series lowers the melting ‘‘point”’ 
of a member of the other series. Moreover the one or the other 
begins to separate first according to which is present in excess over 
certain fixed proportions. There the analogy with eutectic crys- 
tallization ends for the simple reason that there is no eutectic, 
no inevitable end-point where final solidification must take place 
when the liquid has attained a certain composition. The minerals 


t The two series are bridged at the very outset by spinel but this has, on the whole, 
no great practical importance. 


192 N. L. BOWEN 


have a reaction relation to the liquid, not a mere subtraction rela- 
tion. Each separated mineral tends always to change into a later 
member of the reaction series. This change of composition is 
effected by reaction with the liquid, and according to the oppor- 
tunity for reaction the liquid is entirely used up, in some cases 
sooner, in others later, and only then is solidification complete. 

Thus we see that rock series cannot be partitioned off into 
such divisions as gabbro, diorite, etc., each having a eutectic of 
its own. All of these belong to a single crystallization series, to 
a single polycomponent system, which is dominated by reaction 
series. 

Even after what is ordinarily termed complete consolidation 
of the rock many changes of a reaction nature may occur between 
the minerals. Usually a small amount of liquid is a factor in these 
changes which are commonly described under the head, meta- 
morphism. Certain aspects of these changes have been described 
by V. M. Goldschmidt in a paper on metasomatism.’ It is impor- 
tant to note in the present connection that metasomatic reactions 
are not confined to the metamorphic stage but are, as we have seen, 
a constant feature of the whole process of crystallization of the 
magma. Eskola has presented evidence for believing that, at 
times, the original crystallization of a magma may itself take place 
under the conditions that are ordinarily prevalent during meta- 
morphism. ‘The rock then presents a “mineral facies” identical 
with that of a metamorphic rock of the same composition formed 
under the same conditions. ‘Totally different reaction series would 
be concerned in the crystallization of these rocks and we have not 
gone far toward an understanding of these, though Eskola has 
done much to set us on our way.? The variation of conditions 
necessary to produce different facies is of a different order of 
magnitude from the variation of conditions here considered. 
This is largely a variation of the rate of cooling and we are here 
concerned mainly with the diverse rocks in the common and defi- 
nitely established sequence that can be produced under the 
conditions of his diabase facies principally. With adequate 

* Economic Geology, Vol. XVII (1922), p. Ios. ' 

2 P. Eskola, Norsk Geologisk Tidskrift, Vol. VI (1920), pp. 143-04. 


THE REACTION PRINCIPLE IN PETROGENESIS 193 


fractionation under conditions of slow cooling we may, however, 
have such concentration of mineralizers and consequent lowering 
of consolidation temperatures that conditions correspond to those 
of other facies. 


EMPIRICAL RULES REGARDING ORDER OF CRYSTALLIZATION 


As the result of a most extensive knowledge of the relations of 
minerals in rocks Rosenbusch formulated certain rules regarding 
the order of crystallization. When petrologists began to think of 
the crystallization of rocks in terms of eutectics these rules seemed 
to be quite at variance with theory, for this stated that those 
minerals should separate first that were in excess over eutectic 
proportions. To reconcile this with Rosenbusch’s rules giving a 
fixed order of crystallization was a very difficult problem, and yet 
it was conceded that his rules represented the facts in very remark- 
able degree. In the light of the reaction principle no such difficulty 
is encountered. It is characteristic of the reaction series, as we 
have already seen, that however small an amount of any member 
may form, it always forms before a lower member of the series. 
In so far as reaction series control the crystallization of rocks they 
tend to produce a fixed order of crystallization. 

Another generalization concerning the order of crystallization 
came from the French school of petrology. It was that the minerals 
separate in the order of their fusibilities, the least fusible first. 
This generalization, too, in the light of the doctrine of eutectics 
seemed to be quite absurd, but here again there is a very consider- 
able accord with the facts. Such substances as spinel and chro- 
mite are among the most refractory materials known. Magnesian 
olivine has the highest melting-point of the common rock-forming 
silicates. The more magnesian pyroxenes stand next in the list, 
with the more calcic feldspars about on a par with them. The 
more complex pyroxenes, the amphiboles, and in particular the more 
alkalic feldspars are lower still. And this list corresponds very 
satisfactorily with the order of separation of the minerals from 
magmas. Again we are dealing with a tendency that is introduced 
by the presence of reaction series. It is a familiar character of a 
reaction series of the continuous type that the higher melting 


194 N. L. BOWEN 


members should crystallize first (witness the plagioclases), and 
there is ano less distinct tendency in this direction in the discontinu- 
ous réaction series. 

This explanation and vindication of the doctrine of the French 
school is not, however, to be regarded as a proposal that we should 
reverse the attitude arising from eutectic considerations and accept 
unreservedly the statement that minerals separate in the reverse 
order of their fusibilities. There are too many obvious exceptions 
to such a rule. However, these exceptions are themselves such as 
might be anticipated in a system dominated by reaction series. 
We have seen on an earlier page that certain minerals that may be 
called released minerals may appear in such a system. It is appar- 
ently minerals that belong in this category that constitute the more 
notable exceptions. The principal of these is quartz. It is a 
released mineral of the reaction, olivine-pyroxene and no doubt 
has a similar relation in the reaction, potash feldspar-mica, a sort 
of hydrolysis requiring an adequate concentration of water. It 
is a mineral of high melting-point but separates very late—a fact 
that is to be connected with its character as a released mineral. 
In other words, it may be regarded as not actually present in the 
liquid at an early stage, but as released later in consequence of 
the reactions mentioned. 

Some minerals may separate early when intrinsically present 
and may also appear as released minerals at later stages. Magne- 
tite is a particularly good example. It may separate very early 
from a gabbro and it may be formed very late as a result of, say, 
the pyroxene-hornblende reaction and in particular of the mica- 
feldspar reaction. 


PROGRESS TOWARD AN UNDERSTANDING OF REACTION 
SERIES IN MAGMAS 


At this place it may be well to point out where we are in par- 
ticular need of increased knowledge in order to understand the 
exact nature of the reactions occurring. It will be noted that each 
member of the discontinuous reaction, series, olivines-pyroxenes- 
amphiboles-biotites, is itself a continuous reaction series. Of the 
olivines and the pyroxenes we know a fair amount, but practically 


THE REACTION PRINCIPLE IN PETROGENESIS 195 


nothing is known as to the details of the amphiboles or the biotites 
as reaction series, and this means not merely their possible range 
of composition but also how those compositions are arranged in the 
series. The continuous reaction series of the plagioclases occupies 
a more conspicuous place than the other continuous reaction 
series, not merely because we know more about it, but because, as 
already pointed out, it enters into the rock series through a much 
wider range of conditions. Of the plagioclases we require to know 
how the potash feldspar enters into their make-up and, in the present 
connection particularly, at what stage in its concentration the 
potash molecule must appear as a separate phase. We require to 
know, too, the exact composition of pyroxene that forms by reac- 
tion from olivine of a certain composition, the exact composition 
of amphibole that forms from a certain pyroxene, and so on. 
These are questions on which much light may be shed by systematic 
equilibrium studies in synthetic minerals and by careful separation 
and analysis of associated minerals in various rock types. Studies 
of this kind are now being carried on at this Laboratory by Aurous- 
seau on olivines, by Washington and Merwin on pyroxenes, and 
by Buddington on melilites. Eskola points out the importance 
of such work on the natural minerals, referring to it under the 
special name, facies petrology. The results of these and kindred 
investigations may be profitably regarded in the light of their 
bearing on the reaction principle, for in their manifold details they 
are likely to prove excessively wearisome unless considered in 
relation to some such co-ordinating principle. 


REACTION SERIES AND MAGMATIC DIFFERENTIATION 


As the information gained from such studies of reaction series 
accumulates, our knowledge of igneous rock differentiation should 
increase conspicuously, for the reaction principle is the very life- 
principle of differentiation. We have seen, in our examination of 
the simple systems used to illustrate reaction series, how the exist- 
ence of the reaction relation lends a flexibility to the behavior of a 
cooling liquid, renders it capable of giving different products 


=P. Eskola, ‘‘The Eclogites of Norway,” Videnskapsselskapets Skrifter I. Math. 
Naturv., Klasse No. 8 (1921), p. 6. 


196 N. L. BOWEN 


according to the conditions under which it cools. This is equally 
true of the complex reaction series that dominate the crystalliza- 
tion of magmas an outline of which, for the subalkaline magmas, 
is suggested in Table II. Asa result of the existence of the reaction 
relations there indicated, it may come about that a magma pre- 
cipitating olivine may, at a later stage, react with the olivine and 
convert it into pyroxene, and, according to the extent to which 
controlling conditions facilitate the reaction, the whole future 
course of the magma is modified. If conditions are particularly 
favorable the reaction may be complete and the olivine may dis- 
appear, its place being taken by pyroxene. The liquid then cools 
further with deposition of later members of the crystal series. If 
conditions are not so favorable some olivine may be left unchanged 
and, in consequence, a somewhat different liquid is left to pass on 
down the crystal sequence. So it is with the reaction, pyroxene- 
amphibole. Not only that, but within each mineral group the 
reactions may be variously facilitated under different conditions. 
This is true not only of the members of the olivine-pyroxene- 
amphibole series but also of the plagioclase series, and the liquid 
may be entirely used up by reaction, sometimes earlier, sometimes 
later. It need not be surprising, therefore, that the differentiates 
derived from a gabbro magma may vary both in scope and quality. 
We may have only gabbro and diorite in one case, but a longer 
sequence including granite in another. We may have a potassic 
granite in one case and a sodic granite in another. Rapid crystal- 
lization at an early stage, with subsequent slowing-up, seems to 
bind up much of the potash feldspar in the early formed plagio- 
clase and give a sodic granite. On the other hand slow cooling 
throughout seems to avoid this factor and give potash granites. 
Too little is known as yet of the details of the various reaction 
series to make possible any very definite statements on these points. 

In offering an outline of crystallization series in rocks (and 
therefore of differentiation) the list has been carried only as far 
as the constituents of granite. This should not be interpreted as 
indicating that there is anything final about the granite with respect 
to differentiation series. We have seen in the system K,SiO,- 
SiO.-H,O that when any mass is cooled to 200° and no farther 
there will commonly be a liquid left over that never crystallizes. 


THE REACTION PRINCIPLE IN PETROGENESIS 1Q7 


The composition of this liquid lies at some point on the 200° 
isotherm. If the cooling were continued to 1oo° and no farther 
there would again be liquid left whose composition would be rep- 
resented by a point on the 100° isotherm, and this liquid would never 
crystallize. And so it would be for any temperature except some 
negative temperature at which the water itself would appear as ice. 

A similar condition will prevail in the aqueous system repre- 
sented by magmas. ‘The cooling of rocks is always limited by the 
temperature of their surroundings. There is always a liquid left 
that does not crystallize and it is probable that to the very last 
the rock constituents bear a reaction relation to this liquid. 

In this connection the behavior of liquids crystallizing along 
the boundary Q,-H.O (Fig. 4) may be instructive. At the higher 
temperatures crystallization consists of the simple separation of 
both KHSi,O, and quartz, but at lower temperatures KHSi,0; 
separates and quartz redissolves, or, better, the liquid reacts with 
quartz, converting it into KHSi,O,. If the reaction is not free to 
take place the liquid leaves the boundary curve, passes into the 
KHSi,O, field, and thereafter only KHSi,O,; separates from it, no 
quartz. Thus we have a reaction relation between liquid and 
crystals at the very latest stages and, in consequence of it, alter- 
native behavior of the liquid. ; 

There are some indications that in granitic liquid at the peg- 
matite stage, more or less similar conditions prevail. Quartz may 
react with liquid to produce feldspar and sometimes the graphic 
structure may be a result of this reaction (replacement). Why 
replacement should produce graphic structure in some instances and 
not in others is not apparent, but the possibility should be con- 
sidered that graphic structure may often be the result of replace- 
ment both in rock minerals and in ore minerals. The graphic 
intergrowth of spinel and pyroxene seen in certain rocks can scarcely 
be interpreted otherwise than as due to reaction.t Experimental 
work on systems involving these phases shows pretty definitely that 
anything suggesting a eutectic relation between spinel and pyroxene 
is not to be entertained.’ 


xW.N. Benson, Jour. Proc. Roy. Soc. N. S. Wales, Vol. XLIV (1910), p. 521. 


2 Rankin and Merwin, ‘“‘The Ternary System MgO-Al,03-SiO:,”’ Amer. Jour. Sci., 
Vol. XLV (1918), pp. 301-25; also Olaf Andersen, ‘‘The System: Anorthite-Forster- 
rite-Silica,”” Amer. Jour. Sci., Vol. XX XIX (1915), p. 437. 


198 N. L. BOWEN 


If a reaction of the kind suggested (involving quartz, feldspar, 
and mica molecules) is a definite feature of pegmatitic granite it is 
possible that failure of reaction may produce a liquid deficient in 
silica, just as such a liquid is produced in the system above. From 
this liquid feldspathoid may be deposited instead of some of the 
feldspar. Thus it is possible that alkalic rocks may, in some cases, 
be formed as a result of the reaction relation among the minerals 
noted. Foye has noted a very intimate relation between the granite 
pegmatites and nephelite syenite of Haliburton, Ontario.’ © 

NoTE.—Since the foregoing was written it has been demonstrated by Morey and 
Bowen that leucite and orthoclase constitute a reaction pair with quartz as a released 
mineral. The importance of the reaction principle in connection with the origin of 


some alkaline rocks is confirmed by these facts but the full significance cannot be 
discussed here. 


SUMMARY 


Petrogenic theory has passed beyond the stage where the con- 
ception of eutectics can longer be regarded as of any considerable 
service. Experimental investigations and the study of the rocks 
themselves, in the light of such investigations have made it clear 
that the eutectic relation is unimportant but that another relation 
between liquid and crystal phases, here called the reaction relation, 
is of fundamental significance. The ordinary solid solution series 
such as the plagioclases may be regarded as a continuous reaction 
series because during crystallization each member is produced 
from an earlier member by reaction with the liquid, the variation 
of composition being continuous. There are also discontinuous 
reaction series exhibiting related characters but with discontinuous 
changes of composition. The series olivine-pyroxene-amphibole- 
mica is a prominent example among the rock-forming minerals. 

On the basis of these considerations the minerals making up 
the rocks of an igneous sequence can be arranged as reaction series 
and it is the existence of such series that controls the crystalliza- 
tion and differentiation of the rocks of the sequence. Even the 
graphic structure, usually regarded as a eutectic structure, is prob- 
ably to be considered the result of reaction between the phases in 
many examples. 


January, 1922 


1 Amer. Jour. Sci., Vol. XL (1015), p. 436. 


_AN OUTLINE OF THE PHYSIOGRAPHIC HISTORY 
OF NORTHEASTERN ONTARIO™ 


W. H. COLLINS 
Canada Geological Survey, Ottawa 


The region here referred to as northeastern Ontario is one about 
250 miles square, lying between Lake Superior and the Quebec 
boundary and reaching from Lake Huron north to the Trans- 
continental Railway. Its topography has been described so many 
times in reports of the Geological Survey and of the Ontario Bureau 
of Mines that no extended account of it is needed in this place. 
The region is referred to by geologists and topographers as a glaci- 
ated peneplain. As a matter of fact it is a hilly country and only 
conveys an impression of flatness when a large area is seen at one 
time. Viewed from the top of one of the more commanding hills 
the horizon appears as an even line and the intervening hills 
can be seen to stand at about the same elevation. The rivers have 
slopes of only 5 to ro feet a mile, and the highest and lowest points 
on the railways which cross the region are 1,350 and 600 feet 
above sea. 

This condition holds true for most of northeastern Ontario. 
There are a few places, however, where the relief is greater, and 
where significant traces remain of a former topography of more 
rugged character. There is, for example, one hilltop, about 40 
miles west of Sudbury, from which one can see, on a fine day, a 
line of white hills on the southern horizon that certainly remind 
one of mountains. These are the Lacloche Mountains, to which 
further reference will be made. It also requires considerable 
scientific faith to acknowledge the wild country bordering Lake 
Superior as plainlike or nearly plainlike. 

In the case of a true plain, like the Mississippi basin, or the 
western plains of Canada, the impression of tranquillity conveyed 
by the landscape is borne out by the rocky structure underneath. 


t Read before Section E, American Association for the Advancement of Science, 
December 27, 1921. 


199 


200 W. H. COLLINS 


The rock formations are all sediments, deposited by the subdued 
agencies of water or of wind. They lie in horizontal beds, free 
from stress. 

The mild topographic aspect of most of northern Ontario might 
lead one to expect somewhat the same undisturbed condition under- 
ground. ‘This is not found, however. It is not asserted here that . 
some of the rock formations are not flat-lying or nearly so. Some 
of them are; but the prevailing condition is quite different. Char- 
acteristically, the stratified formations of northern Ontario are 
folded, twisted, sheared, and broken in the most extraordinary 
manner. ‘They spring from deep underground in arches that are 
abruptly truncated by the surface of the peneplain. In places 
they have been broken by faults so great that the two sundered 
parts of a formation, that once lay face to face, are now found 
from a few hundred feet to a mile apart. They are cut by dikes 
and sills of rocks that were once molten, and a large share of the 
present surface of the region is now underlain by batholithic 
masses of granite, once molten, a hundred miles or so in diameter, 
and of unknown depth. The condition is one of suspended geo- 
logical turmoil of incredible magnitude and violence. 

There is something dramatic and stirring to the imagination in 
this contrast between the mild landscape of the present northern 
Ontario and the evidence of telluric convulsion that lies just beneath 
it; a contrast which impels one to speculate about the cause of 
this violence, in what manner it expressed itself at the surface of 
the earth, and what have been the processes of change between 
then and now. 

The characteristic geological features of this region, as con- 
trasted with those of a true plain, are the extreme deformation of 
the stratified formations, and the presence of igneous formations, 
particularly the batholiths of granitic rocks. If one studies a 
geological map of North America (Fig. 1), it is soon apparent that, 
outside of the great Precambrian Shield region, to which northern 
Ontario belongs, the greatly folded and faulted rocks are confined 
to the mountain regions—the Western Cordillera and the Appa- 
lachian Mountains along the Atlantic Coast. It can be seen at 
the same time that great granite masses, the batholiths, are present 


PHYSIOGRAPHIC HISTORY OF NORTHEASTERN ONTARIO 201 


in these mountain regions, and nowhere else in the continent, 
except, of course, in the Precambrian Shield. 

This conjunction of batholithic intrusions with mountain 
systems is no new observation. It has been long noted and there 


legend 


Sediments not highly folded 


Precambrian areas 


Geological Survey Caneda 


Fic. 1—Map of North America to show the distribution of batholithic intrusions 
in relation to mountain-building. 


has been much speculation as to the nature of the association— 
whether the compression and upheaval of the earth’s crust caused 
the igneous invasions or whether the latter were causes or mere 
concomitants of mountain-building. Nor has the analogy between 


202 W. H. COLLINS 


the deformed strata and intrusive batholiths in mountain regions 
and in the flat, Precambrian Shield, been overlooked; but the 
apparent total absence of actual mountains in the Precambrian 
Shield, as well as its enormous area, exceeding that of any known 
mountain system, have led geologists to accept the theory of Pre- 
cambrian mountain-building with reservations, or even to prefer 
the explanation that the Precambrian Shield is part of the original 
crust of a once molten world. 

Let us compare the mountains of North America in respect to 
their ages, their topographic aspects and the abundance in them 
of batholithic rocks. The accompanying illustration (Fig. 2), 
compiled by J. F. Wright, of the Geological Survey, Ottawa, shows 
a series of representative cross-sections through the Rocky Moun- 
tains, the Coast Mountains, the Appalachians, the stumps of the 
ancient Acadian Mountains of Nova Scotia, and through a part 
of the peneplain of northeastern Ontario. The sections are 
arranged in order of age from youngest to oldest, are all referred 
to sea-level and drawn to the same horizontal and vertical Scag 
the relief being exaggerated twenty-one times. 

The Rocky Mountains, which date from early Tertiary time, 
reach elevations of 10,000 to 12,000 feet, have an extremely sharp 
profile and reveal no batholithic intrusions, although some distance 
south of the line of section the Ice River syenite mass does reach 
the surface. The Coast mountains, of Jurassic age, reach about 
7,000 feet above sea, and more than one-third of the section is of 
batholithic granitic rocks. These mountains are scarcely less sharp 
in profile than the Rockies; but that is due to the fact that they 
have been elevated twice and are now being carved a second time. 
The Appalachians were formed about Permian time, and have 
suffered erosion for a correspondingly longer time than the Coast 
and Rocky mountains. They reach only 2,800 feet above sea, 
have a comparatively gentle skyline, and erosion has laid bare a 
larger proportion of their granite core. The Acadian Mountains 
can be called such only in a genetic sense. They possess a relief 
of only a few hundred feet, and their granite core is freely exposed. 
They are the barely recognizable roots of mountains, and not more 
deserving of the designation than those shown in the next and last 
section. 


PHYSIOGRAPHIC HISTORY OF NORTHEASTERN ONTARIO 203 


‘ppyueuquirese g sy} Jo ASojoo3 oy} pure surpying 
-ureyunour usemjoq drysuoryejer orjoues oy} OsTe MOYS 07 pUP ‘SUOISNIZUT SIYITOYIVq 91eq SUTAL] UT UOISOIO JO 
qoayja VAIssaiz01d oY} MOYS 0} VOLIOUIY YON JO SUTeYD UTeJUNOU ay} JO SUOTJIVS SSOID Jo SaTIagG—z “OI 


epeuey fansne 62/60/0895 


OldE4UC “LOIN 

qaa4 aye] a) Aingpns Jeau Woly Buipying ufejunow 

ajeog yeo1q4a, (UeIUIE/ IY) UELIQUIEDAg BJE/ JO JIG §S0JIC U0/492S 
ae : aTeSSn —JIAI|-BIG 


oc oe oO 


ooo: 2001 


SovIW ri j 
ajeog jequozisoy o0ce 


6 


SOASNLJU PY QYOYREOG S,UaUuIpaS 
4 a 3 E402 GAO 
eal ey ‘SUIRJUNOfy ULIPEIp IY; SS0IIE WOl4IIS 


TG 1-828 


2901 


pusgey 


{@Ao}-8a aS x a Jaaaj-BaG 
LS S - =~. 
9001 SA \ 3 y NX 0.00! 
< Z fe } oooe 


Zsa) 


ee 


APM/OEl ULAISET 7103/9 I1I984 ayy Guoje Aempley II9e4 uepeue7 ay, buoje 
PIQUNOD YSiidg ‘ABUeY ISLOD Aly} SS04I2 UO/JIAS EIQUIYOD YSHIIG SUIPJUNOP AYIOY ai S50/98 LOW295 


J9Aa]-B9G 


Reference was made in the first paragraphs of this paper to 


Lacloche Mountains, south of Sudbury, as one of the topographic 
Ss 
S 
SH 
3 
‘ 
: 

is 
R 
8 
¢ 
iS 
& 
x 
5 
S 
S 
S 
IS 
& 
g 
& 
S 
S 
§ 
3 
& 
& 
S 
S 


features that can scarcely be included as part of the Precambrian 


They rise in a solid, broad-based range 60 miles 


long and as much as 800 feet above the surrounding country. 


Shield peneplain. 


204 W. H. COLLINS 


Their mountainous aspect is enhanced by the fact that they con- 
sist of white, Lorrain quartzite and, when seen from any consider- 
able distance, look as if they were covered with snow. ‘The last 
of the five cross-sections in Figure 2 is through Lacloche Mountains. 
The sediments forming these mountains are Precambrian (Huro- 
nian), and the intrusive granitic batholiths are late Precambrian 
(Killarney). ' 

Figure 2 has been introduced for the purpose of demonstrating 
that the associated phenomena of regionally deformed strata and 
batholithic intrusion are results and criteria of mountain-building 
and that by sufficiently protracted erosion a mountain system may 
be reduced to a peneplain having the topographic appearance and 
geological characteristics of northern Ontario. Endeavor will now 
be made to apply these criteria to the Precambrian Shield and to 
outline the chief events in the physiographic history of the north- 
eastern part of Ontario, proceeding from the present as far back in 
time as the geological record has been interpreted. 

Figure 3 shows the distribution of the essential geological ele- 
ments in northeastern Ontario with which this thesis is immediately 
concerned. - It also contains a tabular record of the geological 
periods. Periods not represented by rock formations in the region 
are shown in brackets. The larger spaces represent gaps in the 
record of intervals of erosion, and an attempt has been made to 
indicate, conservatively, by the size of the space, the length of 
time unrepresented by rock formations. 

Let us take that part of the record from the Ordovician period 
to the present. Only a few of the series of formations are present— 
the Ordovician, Silurian, part of the Devonian, a little of what Mr. 
Keele regards as Cretaceous,’ and the Pleistocene. All these forma- 
tions lie horizontally or dip at very low angles. The Ordovician, 
Silurian, and Devonian are limestones, shales, and sandstones that 
contain fossil remains of marine animals. ‘The Cretaceous contains 
beds of lignite and was presumably laid down on land or very near 
land. The Pleistocene consists of glacial bowlder clay and strati- 
fied clays and sands laid down in postglacial lakes. The careful 
studies of glacialists, Coleman, Johnston, Taylor, and others, have 

* Summary Report, Geol. Surv., Can. (1919), Part G. 


PHYSIOGRAPHIC HISTORY OF NORTHEASTERN ONTARIO 205 


“surpyinq 


ulejUNOU Jo sassad01d Sutpuodsas109 3a1Y} 0} pe}ead a1e pue sa8e jout}sIp aa1y4} Jo 


d1e YOIYM (SIO}UIeG) SOAISNIZUL ITYIPOYILG 9Y} JO UOTJNQIYSIP UMOUY 94} MOYS 0} OLIeJUQ Uda}sveyIION jo depy—F ‘org 


WISIURI[OA 

JO polled 
(Sattes a10q) 
UOI}eJUSUII pas 

JO pollog 
(¢UIeMoay 
wistued 

“JOA JO portog 
(sr) 
‘urlingpns 

SUIWe YSU 7) 
UOlZ2JUIUIIPIS 

JO pollag 


sonig 
Heq0D 
} 


(197e@M0971Y M) 
Seay Ul 
I 
ueMeuaIaMay | 
I 
! 


(ueraquie)) | 
UPRIDIAOPIO 
ETON 
UeIUOAIG 
(uerddisstsstyl) 
(uvtueAlAsuue dg) 
(UeruIIeg) 


(oIsserty,) 
(oIssein [) 
snoadejaly 


(29007) \ 
(9u9903110) 
(Glienlong\p) 
(aua0I[d) 


9U1990}SI9[q 
quay 


polleg 


UUIN]OD [voIsojoay 


ueluoInyy 


Aqeyiay |, 


Areuloyengd 


elg 


YaFON en] 


og OF OF Of OF ¢ O 


SOT JO a[29S 


aqui 
uvolog-a1g 


es) 


ayes 
uvtuoinyy-a1g 


firme) 
voc 


aylueis AdUIe][IS 


SOTUBO[OA 
pue s}UsWIpas 


Wh 


206 W. H. COLLINS 


shown that the northeastern part of the continent, released from 
its load of ice, has readjusted itself, since the Glacial period, by 
gentle rising and tilting movements. There are no violently dis- 
turbed formations among these. There can be no doubt, then, that 
from Ordovician time until now northeastern Ontario has experi- 
enced only gentle oscillations in level and shallow marine submer- 
gence. 

Let us take next the section of time from the beginning of the 
Bruce period until the end of the Keweenawan. This time is 
represented by a blanket of sediments aggregating over 30,000 feet 
in thickness, in the southern part of the region of northeastern 
Ontario, but thinned out to nothing by erosion, toward the north. 
The Bruce series is in part a land formation, and was for the remain- 
ing part apparently laid down at the edge of a continent; the Co- 
balt and Keweenawan series are both continental. In the northern 
part, these formations (represented by the Cobalt series) lie almost 
flat, or in open folds. Consequently it can be said with assurance 
that from the time of their deposition this northern part of the 
region has experienced no movements but those of rise and fall and 
mild compressional folding. Tracing these sediments southward, 
however, they become progressively more closely folded. South 
of a line drawn eastward and westward through Wanapitei Lake, 
the folds are tight and have a definite eastwest, axial course. Small 
masses of granite, intrusive in all these formations, appear at 
Cartier, at Sudbury, and at other points and, finally, in a line 
running northeasterly from Killarney, the Bruce series and later 
Precambrian sediments are abruptly truncated by a granite batho- 
lith of undetermined extent. How far southward this condition 
extends cannot be ascertained, because, from Lake Huron south- 
ward, the Precambrian formations are covered by the horizontal 
Palaeozoics. 

It is abundantly clear that—if the associated phenomena of 
close regional folding and batholithic invasion indicate mountain- 
building—in late Precambrian time the southern part of the region 
shown on this map was compressed to form a system of mountains. 
These mountains occupied only the southern part of northeastern 
Ontario. They extended west along the south side of Lake Superior 


PHYSIOGRAPHIC HISTORY OF NORTHEASTERN ONTARIO 207 


at least as far as Duluth. How far eastward they extended is not ~ 
known, nor, owing to the cover of Palaeozoic sediments, how far 
south. 

Between Keweenawan and Ordovician time these late Pre- 
cambrian (Killarnean) mountains were worn down to a condition 
probably not greatly different from that of the Adirondacks today, 
and certainly greater relief than the Lacloche Mountains of today, 
for these latter have since endured additional erosion from Palae- 
ozoic time onward. A few peaks of the Killarnean Mountains 
now project through the Palaeozoics on Manitoulin Island; and 
bore-holes that have been made near by in the search for oil indi- 
cate a submountainous local relief of more than 800 feet in the floor 
upon which the Ordovician formations were laid. 

Let us next consider the period of time that elapsed from the 
commencement of deposition of the Doré sedimentary series until 
the Timiskaming series and Sudbury series were completed. The 
Doré series is almost certainly of continental origin. It is thought 
to have been deposited under conditions like those which governed 
the deposition of the silts and gravels in the interior plateau of 
British Columbia. The Keewatin volcanics are believed by some 
geologists to be of submarine origin, owing to the abundance among 
them of ellipsoical greenstone; but the writer can find no evidence 
to support this view. It appears, rather, that they were deposited 
on land and in small bodies of water. Recent studies of the 
Timiskaming series by Mr. H. C. Cooke’ indicate a continental 
origin for this series also. Judging by the extremely variable 
thickness of the sediments throughout this group, their coarse 
texture, and apparently local deposition, the region may have been 
one of considerable topographic relief during the entire time. 

Except in the southern portion of northeastern Ontario, these 
formations are extraordinarily folded, schistified and faulted, and 
invaded by granite batholiths of huge size. In fact, they are 
represented now only by irregular patches which were down-folded 
low enough to escape erosion. If intense regional disturbance and 
batholithic invasion indicate mountain-building, it must be con- 


t “Kenogami Lake and Larder Lake Areas,” Geol. Surv., Can., Memoir in course 
of publication. 


208 W. H. COLLINS 


cluded that following Timiskaming time the whole region, shown in 
Figure 3, was mountain-built. The southern portion was long 
afterward involved in the Killarnean mountain-building and, in 
that part, any evidence of this earlier orogeny was pretty completely 
demolished; but there is little doubt that the older mountains 
extended over the whole region. 

Between the finish of Timiskaming sedimentation and the 
commencement of the Bruce period these older, pre-Huronian 
mountains must have been uplifted, carved, and completely des- 
troyed, for the Bruce series lies upon a peneplained surface evidently 
quite as maturely eroded as the surface of today. The evidence 
for this has been discussed at length in various Geological Survey 
reports,t and need not be repeated. Attention need be directed 
here only to the length of time required for this complete physio- 
graphic cycle; the Appalachian Mountains, which are not yet 
nearly so completely leveled, have existed since Permian time. 

Just as the Killarnean mountain-building demolished in the 
southern part of the region the evidence of earlier geological history, 
so it might be expected the pre-Huronian mountain-building, which 
extended over the whole region, would have obliterated all of the 
geological record anterior to the deposition of the Doré series. 
This is not wholly the case. When the Timiskaming series was 
first described by W. G. Miller,? he noted that it contained granite 
pebbles. Many years before, Logan noted the presence of granite 
bowlders in the Doré series, though, at that early stage in Canadian 
geological work, he could not be sure of the position of this series 
in the geological scale. These granite inclusions in the oldest 
known sediments have afforded ground for much speculation regard- 
ing the nature of the surface from which they were derived, and 
upon which they were deposited. 

It was not until 1920 that the ancient granite which supplied 
these pebbles was identified 7m situ beneath the Doré series, near 
Michipicoten Harbor. At that place the Doré series lies tilted at 
an angle of 50° or more against a basement of older rocks, from 

tM. E. Wilson, Geol. Surv., Can., Memoir No. 39; W.H. Collins, of. czt., Memoir 
Nos. 33, 95, and another not yet published. 

2 Ann. Rep. Ontario Bureau of Mines, Vol. XIX, Part 2, Dp. 62. 


PHYSIOGRAPHIC HISTORY OF NORTHEASTERN ONTARIO 209 


which it derived most, if not all, its pebbles and bowlders. This 
basement is in part composed of schistose volcanics and in part of 
gray granite. Its extent, away from the Doré series, has not been 
worked out and will be very difficult to determine; but there is 
quite enough to indicate that a third granite, older than the Killar- 
ney and the pre-Huronian granites, exists, and had been laid bare 
in large volume in pre-Doréan time.*' Was this pre-Doréan granite 
indicative, like its successors, of a third mountain-building activity 
that took place in a time too remote for the mind to compre- 
hend ? 

Regarding the physiographic conditions governing the deposi- 
tion of the Doré series the authors of the report mentioned specu- 
late as follows: 

The writers are inclined to visualize Michipicoten district at the commence- 
ment of Doréan time as a region of rugged relief and waning volcanic activity. 
During early Doréan time this land surface, devoid of vegetation, was eroded 
rapidly, some of the deep-seated granite masses being uncovered to provide 
the great quantities of granite pebbles and bowlders found in the Doré con- 
glomerate. The rude granite and porphyry conglomerate and associated 
formations in the lower part of the series were formed nearly in place by rock 


disintegration, with a consequent aggradation, or wearing down of summits 
and filling in of depressions in the surface of the country. 


SUMMARY OF PHYSIOGRAPHIC HISTORY 


1. The Doré series, the earliest known Precambrian sediment 
in northeastern Ontario, was deposited as a continental formation 
upon a surface of rugged topography. An earlier period of 
'mountain-building and erosion is suggested by the presence, sub- 
jacent to the Doré conglomerate, of an older granite mass, appar- 
ently of large dimensions. 

2. From Doréan time until the end of Timiskaming time the 
region was apparently a land area of considerable, if not high relief, 
and the seat of prolonged volcanic activity. 

3. A period of mountain-building (pre-Huronian) and complete 
reduction to a peneplain followed, affecting the whole region. A 

*W. H. Collins and T. T. Quirke, Geol. Surv., Can., Memoir in course of 
preparation. 


2 LOG. Git 


210 W. H. COLLINS 


conception of the time involved is obtainable by comparison with 
the rate of formation and reduction of the Rocky, Coast, and 
Appalachian Mountains. 

4. From the Bruce period until the Keweenawan the region was 
apparently a land area or one of shallow and intermittent sub- 
mergence. 

5. Following or during the Keweenawan another (third) process 
of mountain-building (Killarnean) took place which affected only 
the southern portion of the region. ‘These Killarnean Mountains 
were reduced to a state of fairly low relief before the Ordovician 
marine sediments were deposited upon them. 

6. From Ordovician time until the present the region has been 
subjected only to gentle epeirogenic oscillations and shallow sub- 
mergence; its relief was low throughout this interval and it now 
forms a glaciated peneplain. 

7. From the earliest time of which a geological record remains, 
the region of which northeastern Ontario is part, has behaved as a 
positive element of the earth’s crust. 


A NEW OCCURRENCE OF CRISTOBALITE IN 
CALIFORNIA 


AUSTIN F. ROGERS 
Stanford University, California 


Cristobalite, one of the rarer high-temperature forms of silica, 
was first described from the andesite of Cerro San Cristobal, near 
Pachuca, Mexico, by vom Rath,’ and was established as a distinct 
mineral species by Mallard.? Artificial cristobalite has been pre- 
pared in the Geophysical Laboratory at Washington and its proper- 
ties and its relation to the other forms of silica have been studied 
by Fenner.’ Silica bricks are made up of artificial cristobalite and 
tridymite, and so these minerals are of some importance to the 
metallurgist. 

The work of the Geophysical Laboratory is extremely valuable, 
but of especial interest to the mineralogist and petrographer is the 
actual occurrence of these minerals in nature. The laboratory 
production of such artificial minerals is largely to aid in the inter- 
pretation of natural occurrences. Cristobalite has been known 
from five or six foreign localities for some time and in 1918 the 
writer? described it from two American localities, both in California. 
Since that time he has obtained specimens from five additional 
American localities. 

_ The cristobalite described in this paper, like the Yellowstone 
Park occurrence recently described,’ is found in a rhyolitic obsidian. 
The exact locality of the specimens could not be ascertained. The 
material was placed in the writer’s hands by Mr. R. M. Wilke, 
mineral dealer of Palo Alto, California. Mr. Wilke obtained the 
specimens from a prospector, who would not divulge the locality. 
Mr. Wilke, however, says ‘‘I am sure the place is somewhere in the 

* Neues Jahrb. Min. (1887), pp. 1, 198. 

2 Bull. Soc. Min. France, Vol. XIII (1890), p. 172. 

3 Amer. Jour. Sci. (4), Vol. XXXVI (1913), pp. 331-84. 

4Ibid., XLV, pp. 222-26. 

5 Amer. Mineralogist, Vol. VI (1921), pp. 4-6. 


211 


22 AUSTIN F. ROGERS 


northwest corner of San Bernardino County, to the northwest of 
Barstow, probably within 40 to 50 miles from Barstow.” The 
geological map of California compiled by Professor James Perrin 
Smith and published by the California State Mining Bureau shows 
a number of outcrops of volcanic rocks northwest of Barstow, 
and so Mr. Wilke is probably correct as to the approximate locality. 


THE OBSIDIAN AND ITS LITHOPHYSAE 


The specimens consist of dark gray to almost black obsidian 
(n=1.483= .003) with a faint banded structure and bright vitreous 
luster. In thin fragments the obsidian shows on microscopic 
examination numerous, minute, rod-shaped crystallites arranged - 
in parallel lines and a few microlites. 

The obsidian contains many light gray spherulites which range 
from 5 to 25 mm. in diameter and are fairly evenly distributed, 
though sometimes two of them coalesce. The spherulites are more 
or less flattened; some of them are elongate, and most of them are 
somewhat irregular in outline. With a few exceptions they are 
hollow and hence may be called lithophysae. It is estimated that 
the spherulitic material occupies only about a fifth to a half of the 
original space. The spherulites are for the most part ruptured, 
and deep cracks extend almost to their margin, so that the spaces 
left are usually very irregular in outline. The cracks give some of 
the lithophysae a superficial resemblance to mud-cracks. The 
thickness of the spherulitic material is often only about a tenth of 
the radius of the original cavity. 

Accompanying the obsidian specimens were a number of loose 
spherulites or lithophysae, which had been broken out of the 
obsidian. They are much more regular than the imbedded litho- 
physae just described. One of these is shown in the photographic 
enlargement (Fig. 1). They have a more or less well-defined 
hollow space at the center and show a slight tendency to concentric 
structure, as may be noticed in the photograph. 

It seems probable that both the expansive force of the gas 
within the cavity and the tension due to cooling of the surrounding 
glass were responsible for the formation of the lithophysae.* 


1 For a discussion of the origin of lithophysae, see paper by Wright, Bull. Geol. 
Soc. Am., Vol. XXVI (1915), pp- 255-86. 


CRISTOBALITE IN CALIFORNIA PUR 


MINERALS OF THE SPHERULITES 


The solid portions of the spherulites consist of two minerals, 
orthoclase and cristobalite, and of these orthoclase predominates. 
The orthoclase and cristobalite occur in fibrous aggregates, which 
on microscopic examination prove to be intergrowths. In spots, 
which are probably cross-sections, they have an intricate interlock- 
ing structure which suggests graphic texture. The orthoclase is 


Frc. 1.—(X5) Cristobalite and Fayalite in a Lithophysa of the Obsidian from 
San Bernardino County, California. 


cloudy, while the cristobalite is clear. The cristobalite is identified 
by its index of refraction, which is slightly greater than 1.480. 
Most of the cristobalite areas are isotropic, but occasionally they 
show very weak double refraction. The double refraction is uni- 
form, which distinguishes the cristobalite from the glass which has 
practically the same index of refraction. 

It seems probable from the descriptions in the literature that 
such orthoclase-cristobalite spherulites are characteristic of rhyo- 


214 AUSTIN F. ROGERS 


litic obsidians, but have not been recognized before on account of 
the difficulty of distinguishing the cristobalite. It was with great 
difficulty that the writer succeeded in proving that the mineral 
intergrown with the orthoclase was cristobalite. 

Imbedded in these larger spherulites are numerous, very minute, 
spherical bodies, which consist largely of orthoclase with small 
amounts of cristobalite. These probably represent earlier formed 
spherulites of orthoclase which have been recrystallized during the 
later main period of spherulite production. Such composite 
spherulites have been described by Rutley’ from another locality 
in California. Rutley speaks of “‘isotropic matter”? between the 
rods or fibers of orthoclase, which in all probability is cristobalite. 

There is a little magnetite imbedded in the spherulitic material. 


MINERALS OF THE LITHOPHYSAL CAVITIES 


The following minerals were identified in the lithophysae: ortho- 

clase, cristobalite, tridymite, opal, magnetite, and fayalite. These 

minerals occur on the walls of the lithophysal cavities and the 
" specimens make beautiful microscopic mounts (Fig. 1). 

Orthoclase—On the free surface of the cavities the orthoclase 
takes on the form of branching rods with weak birefringence and 
positive elongation (parallel to y). The index of refraction is 
I.517+ .003. | 

Cristobalite—The cristobalite occurs in minute (0.2-0.5 mm.) 
translucent white spherulites, which are scattered over the surface 
of the lithophysae as illustrated in the photograph. In fragments 
the cristobalite shows a well-defined spherulitic structure with 
elongation of the crystals parallel to the faster ray a. There is 
intricate polysynthetic twinning and the extinction is oblique. The 
double refraction is very weak. 

The index of refraction of the cristobalite is 1.483.003, 
which was determined by means of immersion liquids. This deter- 
mination positively distinguishes it from tridymite which has 
indices of refraction of 1.469-1.473. 

The specific gravity of the cristobalite is 2.36.04, which was 
determined by means of a mixture of acetylene tetrabromid and 


« Quar. Jour. Geol. Soc., Vol. XLVI (1800), pp. 423-28. 


CRISTOBALITE IN CALIFORNIA 215 


benzol. (It sank in the liquid in which gypsum [sp. gr. =2.32] 
remained suspended and floated in the liquid in which brucite 
[sp. gr. = 2.39] remained suspended.) 

In still another respect did the mineral agree with cristobalite. 
It is infusible and on heating to a high temperature by means of 
a blowpipe it became semi-transparent and on cooling it suddenly 
became translucent again. This sudden change in appearance, 
which may serve as a blowpipe test for cristobalite, is due to the 
conversion of the high-temperature (-cristobalite to the low- 
temperature a-cristobalite, which, according to Fenner, takes place in 
artificial cristobalite at a temperature varying from 198 to 240° C. 

Tridymite.—The exterior of most of the cristobalite spherulites 
when examined with a high-power binocular microscope are seen 
to be studded with very minute six-sided crystals of tabular habit. 
These are identified as tridymite by their index of refraction, n= 
1.475.005. The tridymite is transparent in contrast to the 
translucent cristobalite. Some of minute cristobalite spherulites 
apparently are converted into tridymite, but careful microscopic 
examination shows that the cristobalite is simply covered with the 
tridymite.) 

Opal.—Several of the lithophysal cavities are coated with a thin 
layer of hyalite opal (x slightly less than 1.45), which was evidently 
the last mineral formed. No chalcedony or quartz could be 
detected in any of the specimens. 

Fayalite—A few, small (1 mm.), brown, transparent crystals 
of fayalite (Fe,SiO,) are found on the free surfaces of the lithophysae. 
‘Three crystals (dark areas) are shown in the photograph. They 
are tabular in habit and resemble the fayalite from the spherulitic 
obsidian of Yellowstone National Park. The fayalite has an index 
of refraction greater than 1.74 and gives an iron borax bead test. 
Some of the crystals have been altered to a black opaque substance. 

Magnetite.—Very minute octahedral crystals of magnetite are 
visible with the high-power microscope. 


FORMATION OF THE MINERALS 


The minerals lining the walls of the lithophysae evidently were 
formed by hot gases liberated by the crystallization of the spheru- 


216 AUSTIN F. ROGERS 


lites. The orthoclase and cristobalite are largely products of 
recrystallization of the spherulites. The cristobalite probably was 
formed at a comparatively high temperature and as the temperature 
became lower, tridymite was formed, and at about the same time 
fayalite and magnetite appeared. The opal is the only mineral 
that was formed by ordinary solutions. The oxidation of the faya- 
lite, a slight staining of the lithophysae, and whitening of the 
orthoclase rods are the result of weathering. 


CRISTOBALITE IN SPHERULITIC OBSIDIAN 


The Yellowstone Park occurrence, the San Bernardino County 
occurrence of the present paper, and two undescribed Californian 
occurrences, all taken together, point to the conclusion that cristo- 
balite is one of the characteristic minerals of spherulitic obsidian. 
In fact it seems to be more common in such rocks than tridymite. 
It is probable that in some cases cristobalite has been identified 
incorrectly as tridymite. For example the minute pellets referred 
to and figured by Iddings' in the spherulitic obsidian of Yellowstone 
Park are probably cristobalite instead of tridymite. 


t Seventh Ann. Rept. U.S. Geol. Survey (1888), p. 264, Pl. 12, Fig. 3. 


FAULT FEATURES OF SALTON BASIN, CALIFORNIA: 


JOHN S. BROWN 
U. S. Geological Survey 


OUTLINE 
GENERAL FEATURES 
SURROUNDING MouNTAINS 
THE BASIN FLOOR 
STRATIGRAPHY 
FORMATION OF SALTON BASIN 
THE InpIo FAULT 
FAULTS SOUTHWEST OF SALTON BASIN 
VALLEYS FoRMED BY FAULTING 
TYPE OF FAULTING 
AGE OF FAULTING 
REFERENCES 


GENERAL FEATURES 


The Salton Basin is a great natural depression in the south- 
eastern corner of California (Pl. I). Its lowest elevation is 273.5 
feet below sea-level,” and the lowest part of the basin is now occupied 
by the more or less ephemeral lake known as Salton Sea, which was 
created in 1905 by the accidental escape of Colorado River during 
the diversion of irrigating water to Imperial Valley. Physio- 
graphically, Salton Basin is continuous with the depression occupied 
by the Gulf of California, and the only topographic barrier that 
prevents the access of sea-water to the basin is a delta dam built 
up in recent geologic time by Colorado River. At its lowest point 
this dam is less than 50 feet above sea-level. 


SURROUNDING MOUNTAINS 


The axis of the basin trends southeast-northwest, and at its 
northwest extremity the basin narrows nearly to a point, being 
connected by San Gorgonio Pass to the Pacific Slope. Mountain 
ranges from 2,000 to 10,000 feet in height flank the basin on the 


t Published by permission of the Director of the United States Geologic Survey. 
2 There is some question as to whether the exact elevation of the lowest point of 
the basin, now covered by Salton Sea, is —273. $ or —287 feet. See U.S. Geol. Surv. 
Water-Supply Paper 213, p. 30, for discussion and statement that former figure is 
probably correct. 
217 


218 JOHN S. BROWN 


northeast and southwest, reaching their highest elevation in the 
vicinity of San Gorgonio Pass, where the basin is narrowest. The 
mountains on the southwest are generally known as the Peninsular 
ranges. They extend down into Mexico, and form the crest of 
the peninsula of Lower California. They include the San Jacinto, 
Santa Rosa, Vallecito, and Laguna ranges, and their crest separates 
the drainage of Salton Basin from that of the Pacific Slope. The 
summit of the Peninsular Mountains is well watered and generally 
timbered. Along it there are numerous tracts of gently rolling 
land sculptured into only a moderate relief. Such are Pinyon Flat, 
Montezuma Flat, and the region near Jacumba. ‘These tracts are 
doubtless remnants of the surface of the region before it was uplifted 
to form the present mountain chain. The eastern slope of the 
Peninsular ranges, however, is precipitous, and its dissection deep. 

The San Bernardino Mountains and other ranges form the 
northeast wall of the basin. To the southeast these break up into 
a series of low, disconnected ranges. All of them are barren rock 
masses, at whose feet are gathered great débris fans. 


THE BASIN FLOOR 


A great part of the basin floor is a monotonous plain formed of 
alluvium brought in from every side. In the center the alluvium 
is fine silt, but at the mountain borders it is a coarse fanglomerate. 
Over some of the basin floor there is a considerable deposit of sand 
in the form of dunes. About the borders there are, at various 
places, exposures of soft, folded, sedimentary beds, which have 
been carved into picturesque badlands. Since the basin is prac- 
tically rainless, vegetation is scant, and restricted to desert plants, 
such as creosote, cacti, ironwood, and, at places, mesquite. 

One other feature of the basin floor deserves mention. Near 
its center, at the southeast end of Salton Sea, is a large group of 
active mud volcanoes, associated with small exposures of recent 
lava (obsidian). 

STRATIGRAPHY 

The rocks of Salton Basin have been studied in detail at only a 
few places. For the purposes of this paper, however, they fall 
naturally into three classes. The first class comprises pre-Tertiary 


eA IL 


114-30’ 


EXPLANATION 


Quaternar 
alluvium 


Pre- ETT encinny 
geneous and 
metamorphic rocks 


Fault 


Relief shading by J.H.Renshawe 


from data supplied by author 


It 


JouRNAL OF Grotocy, VoL. XXX, No. 3 


EX PLANATION 


Quaternary 
alluvium 


Late Tertiany 
y sN sedimentary rocks 
x 
R q ~ N pee, 
: Kells 
| NSA as \ sie 
‘ 


SN 
Lord Weel 
= oy 


» 
Sh Ne ee 


SNA 
SS 


ISneous and 
metamorphic rocks 


33 
30. ~ — 


» Fault 


‘Oasis 


Relief shading by J.H.Renshawe 
trom data supplied by author 


Ss 

S 
[smritheh ell 
Te, 


Niud Valicanos 


Valipatria 


> 


EEE 


<< 


EC 


Uf, "Westmoreland 
VA 


GH 


S Nc x Y 
OLTTMLL ILL” 


FAULT FEATURES OF SALTON BASIN, CALIFORNIA 219 


crystalline rocks, and forms the basement rock of all the region. 
There are large bodies of schist and gneiss, and smaller amounts of 
marble and quartzite. The age of these metamorphics is unknown. 
Most of them are Paleozoic, but a part probably is pre-Cambrian. 
Then there are still greater masses of granite, monzonite, and diorite 
intruded into the metamorphic rocks. ‘These intrusives are prob- 
ably Mesozoic. The various crystalline rocks constitute the mass 
of all the mountain ranges. 

The second division of rocks consists chiefly of sedimentary beds 
of late Tertiary age. The series has a total thickness of several 
thousand feet, and consists of sand, shale, and conglomerate, with 
some salt and gypsum. Generally the beds are more or less folded 
and broken, but they are everywhere soft and poorly consolidated. 
On them is developed the badland topography mentioned above. 
In the Carrizo Creek region these beds contain an abundant late 
Tertiary marine fauna. Elsewhere they are apparently unfossilif- 
erous, and probably terrestrial in origin. 

Interbedded with the Tertiary sediments at some places are 
beds of basaltic lava and tuff sometimes too to 200 feet in thickness. 
These volcanics constitute only a small portion of the series. 

Quaternary alluvium is the last division of the rock section. 
It forms most of the basin floor, and fills many of the adjacent 
valleys. Its areal extent is greatest of the three classes. 


FORMATION OF SALTON BASIN 


Salton Basin has evidently existed as a marked depression since 
times previous to the latter part of the Tertiary period, for it has 
received sediment during the formation of the last two rock series. 
Neither the Quaternary nor the Tertiary sediments are displaced 
by any such vertical movements as must have occurred during the 
formation of the basin. The Tertiary beds are seldom found at 
elevations above 1,000 feet, and the Quaternary shows but few 
recognizable displacements and these are of small size. 

It has been tacitly assumed by geologists that the basin origi- 
nated as a dropped fault block, or graben. This hypothesis, 
suggested by the topographic features, is greatly strengthened by 
the fact that the basin lies directly on the course of the San Andreas 


220 JOHN S. BROWN 


rift, a notable fault along which occurred the displacement that 
caused the San Francisco earthquake of 1906. The California 
Earthquake Commission traced this fault continuously from San 
Francisco southeastward to the very tip of Salton Basin. In its 
report? the commission suggests that the rift probably continues 
southeast, and is connected in some way with the formation of the 
basin. In the Peninsular Mountains, southwest of the basin, there 
are also several well-recognized faults, portions of which have been 
mapped, but these have generally not been traced eastward to 
their visible terminations. These faults and their relation to the 
faults described here are shown in Figure 1. 

The evidence of faulting in this region is as a rule, derived from 
the topography, and can only rarely be established by stratigraphic 
criteria. Igneous and metamorphic rocks, such as compose the 
mass of the mountain ranges, are so homogeneous in nature that it 
is often impossible to determine differences in the rocks exposed on 
either side of a fault, while the Quaternary material which fills 
most of the valleys has been deposited since the faulting occurred 
and obscures it in many places. 


THE INDIO FAULT 


At the base of the crystalline ranges (Little San Bernardino, 
Orocopia, etc.) bordering the upper part of the northeast side of the 
basin, lies a conspicuous chain of low, badland hills in which are 
exposed the late Tertiary sedimentary beds, and occasionally small 
patches of the underlying bedrock. This chain of hills is separated 
by a slight break into two ranges known as the Indio Hills and 
Mecca Hills. The Indio Hills have a total length of 20 miles, and 
a width of 2 or 3 miles. The Mecca Hills are also about 20 miles 
long, and at some places are 4 or 5 miles wide. The general slope 
of the hills is southwest, where they dip beneath the basin floor. 
They are separated from the crystalline ranges northeast by a 
trough from 1 to 3 miles in width filled with coarse alluvial débris. 
At some places the canyons that head in the main mountains cut 
entirely across the hills, as at Shaver Canyon and the Thousand 


t California Earthquake Commission Report, Carnegie Inst. of Wash., 1910. 


221 


FAULT FEATURES OF SALTON BASIN, CALIFORNIA 


Ce a : = Nisva NOLIVS 8¥3N GNV 


(woIssttmWIOD ayenbyyeq erusoped I91fV) 


NL elle, ake 40 SNOILYEOd | 


“NOIssIMWwo2 -pivaduuva 


“PIULOJI[R) UIOYJNOS JO sz[Nvy yUIUTWIOIG— ‘I ‘OTT 


222 JOHN S. BROWN 


Palms. At other places they are deflected by the foothills into 
long, longitudinal valleys between the crystalline and Tertiary rocks. 

The Indio and Mecca Hills have undoubtedly been lifted from 
the basin floor by a fault for which the name of Indio fault is sug- 
gested. This fault passes along the northeast border of the hills 
formed by the Tertiary rocks. Near Indio there is for several 
miles a steep and very perfect scarp, generally not more than 100 
to 300 feet in height. The broken Tertiary beds dip southwest from 
the present face of this escarpment at angles of about 45°. Along 
the northeast side of the Mecca Hills the scarp is less perfect, but 
still constitutes a very well-defined ridge. West of Dos Palmas it 
is a low bluff along the east base of a point of clay hills. 

Although the Indio fault as here described was observed only at 
a few scattered points, it appears reasonably certain that it extends 
throughout the length of the Indio and Mecca hills. The most 
critical point at which it should be examined is the contact with the 
crystalline bedrock in the Orocopia Mountains, where there appears 
to be no marked separation of the crystalline range from the bor- 
dering sedimentary foothills. The fault doubtless continues at 
least as far south as an isolated hill of Tertiary material near 
Durmid, which has probably been elevated in the same manner as 
the hills farther north. Its total length, then, is about 50 miles, 
and its form an arc of gentle curvature. Its northwestern end has 
a trend nearly west, and lies in the direct line of probable extension 
of the San Andreas rift, from which the actual separation is only a 
few miles in length. The conclusion seems justified that the Indio 
fault represents the extension of the San Andreas rift. At the 
southeastern end the fault-trend is more nearly south, and points 
suggestively toward the mud volcanoes. 

It should be noted that the throw of this fault is opposite to 
that of the San Andreas rift farther northwest, and also to the 
original displacement which must have formed the Salton Basin. 
Such reversals of throw along prominent fault lines are, however, 
neither impossible nor uncommon,’ and occur at several places 
along the course of San Andreas rift farther northwest. 


«The Bright-Angel fault in the Grand Canyon is a case in point. Sczence, 
April 24, 1908, p. 667. 


FAULT FEATURES OF SALTON BASIN, CALIFORNIA 223 


FAULTS SOUTHWEST OF SALTON BASIN 


The faulting along the southwestern side of the basin is appar- 
ently of two ages, and took place in somewhat different directions, 
the intersection of the two fault systems probably accounting for 
the present irregular outline of the basin. Thus great mountain 
salients, such as the Santa Rosa Mountains, the Vallecito-Fish 
mountain spur, and the projection of the Peninsular Mountains 
along the Mexican Boundary, are separated by big re-entrant valleys, 
such as San Felipe Valley and the Carrizo Creek Valley. 

The oldest of these fault systems, if this inference is correct, has 
a strike about N. 10 W. and is represented by three notable escarp- 
ments. One lies along the east base of the San Jacinto Mountains, 
passing up Palm Canyon. Another is at the west side of Borego 
Valley. The third extends from Agua Caliente Springs southward 
up Carrizo Gorge, along the east face of the Laguna Mountains. 

Evidence of a fault along the northeast and east face of Mount 
San Jacinto was obtained near Whitewater. Just west of White- 
water Point the mountain face is composed of pink and gray granite, 
and of a grayish marble. The marble and granite are arranged 
in layers turned on edge with a strike about N. 20 W., and a dip of 
75° or more to the northeast. Although the quantity of marble is 
much less than that of the granite it constitutes a considerable part 
of the mountain mass, the layers ranging from a few inches to 50 
feet in thickness. The alternation of rocks is well exhibited in a 
prospect tunnel in Sec. 23, T. 3 S., R. 3 E., where the material pene- 
trated has the appearance of a gigantic fault breccia, and the 
contact surfaces are abundantly slickensided. This intermixture 
of material probably resulted from step-faulting, the successive 
breaks along many parallel lines causing an intimate mixture of the 
different rocks. Associated with this prominent fault line is the 
warm spring at Palm Springs, whose water is believed to be derived 
from granitic rocks'. 

West of Borego Valley no observations were made to confirm 
the existence of a fault except to note that the mountain front in 
that region is a very steep and straight escarpment from 2,000 to 
4,000 feet in height. 


t The inference is based on unpublished analytical data. 


224 JOHN S. BROWN 


The southernmost fault scarp of this series was observed in 
Canebrake Canyon, and at Agua Caliente Springs. At both places 
the mountain front for several hundred feet from the lowland bor- 
der consists of rotten, grayish granite, broken into minute joint 
blocks, kaolinized, and altered. Farther in the interior of the 
mountains the rock is dense and unweathered. There is much 
evidence of hydrothermal alteration, a very natural thing to expect, 
at Agua Caliente Springs, where a large number of springs, part 
of which yield warm water, issue from the granitic rocks. 

Cutting across this first system of faults is a system which 
strikes approximately N. 45 W., and which is represented by several 
prominent faults. The most northerly fault of this system is the 
San Jacinto fault, which passes south and west of San Jacinto 
Mountains, extending through Hemet Valley and down Coyote 
Canyon. For several miles it traverses the northeast side of Borego 
Valley.1 The uplift along this fault was on the northeast. Coyote 
Mountain on the northeast of Borego Valley is part of a prominent 
spur elevated in this uplift, and is bordered on the southwest by 
well-defined scarp, which displaces Tertiary beds. It is probable 
that the San Jacinto fault extends at least as far as Borego Moun- 
tain, but it is much obscured in that direction by recent alluvial 
deposists. Movement occurred along this fault at the time of the 
San Jacinto earthquake of 1899. 

Several faults which have been recognized in the vicinity of 
Warner Valley? extend southeastward into the western part of this 
region. One of these which passes nearly through Warner Hot 
Springs traverses Grapevine Canyon, turns nearly east along a part 
of San Felipe Creek, and disappears near The Narrows. Its uplift 
was on the northeast, and the tongue of granitic rock south of 
Borego Valley and in the vicinity of The Narrows is believed to 
have originated from the uplift. A fault extends from Warner 
Valley down the headwaters of San Felipe Creek, and its eroded 
scarp forms the northeast side of San Felipe Valley, being a promi- 


tH. W. Fairbanks, California Earthquake Commission Report, Carnegie Inst. of 
Wash., 1910, p. 47. 

2A. J. Ellis, and C. H. Lee, ‘Geology and Ground Waters of the Western Part of 
San Diego County, Cal.,” U.S. Geol. Surv. Water-Supply Paper 446. 


FAULT FEATURES OF SALTON BASIN, CALIFORNIA 225 


nent mountain wall for 12 or 15 miles. Another fault passes 
through Banner Canyon and Rodriguez Canyon, and extends along 
the north side of Mason Valley and Vallecito Valley, the mountain 
walls of these valleys probably representing considerably eroded 
fault scarps. The last two faults unite in the vicinity of Agua 
Caliente Springs, and are not known to continue farther, but may 
extend along the north side of Carrizo Valley at the base of Valle- 
cito Mountains and Fish Mountains. 


VALLEYS FORMED BY FAULTING 


Associated with the second system of faults are several peculiar 
valleys for whose formation the faults have been responsible. The 
largest of these valleys are Borego Valley, San Felipe Valley, Mason 
Valley, and Vallecito Valley. Collins Valley, adjacent to Borego 
Valley, and a little valley less than a mile in extent at Banner, were 
formed in the same way. All of these valleys have for their north- 
east boundary a high, steep mountain wall which originated as a 
fault scarp along some one of the faults mentioned above. Thus 
Borego Valley and Collins Valley lie southwest of the San Jacinto 
fault. The general shape of each valley is triangular, and the 
south and west sides are much more irregular in outline than the 
northeast side, the mountainous borders on these sides being also 
somewhat less abrupt than those on the northeast. Most of the 
valleys are high at the southwest, and the drainage is to the north- 
east. This has probably been the natural result of the tilting of 
the faulted strips, which have all been dropped down on the north- 
east and elevated on the southwest. 

Most of the faults have forced the drainage to follow northwest- 
southeast directions, particularly in the various canyons such as 
Coyote Canyon, Grapevine Canyon, and Banner Canyon, but some 
streams, such as San Felipe Creek, northeast of San Felipe Valley, 
and Banner Creek at Banner, occupy deep gorges which cut directly 
across the fault scarps at the northeast border of these valleys. It 
is probable that these streams existed before the faulting, and that 
. the faulting took place gradually, the streams cutting down as fast 
as the rocks were lifted across their beds. A further suggestion 
that the earlier drainage lines may have had a northeast trend is 


226 JOHN S. BROWN 


afforded by the granite ridges which divide some of the valleys. 
San Felipe Valley, for instance, is cut nearly in two by a low spur of 
granitic rock projecting from the southwest, while Mason Valley 
and Vallecito Valley are entirely separated by such a ridge, except 
for a very narrow, rock-cut canyon. ‘These ridges may very likely 
represent drainage divides which existed before the faulting. 


TYPE OF FAULTING 


Most of the faulting observed is of the normal type. The Indio 
fault is associated with much folding, and it is possible that it may 
be in part due to thrust movements. 


AGE OF FAULTING 


The age of the various faults is very difficult to establish, since 
the age of the rocks displaced is so indefinitely known. The original 
settling of the Salton Basin must be pre-late Tertiary in age, 
because in the resulting basin great thicknesses of late Tertiary 
sediments were deposited. Considerable faulting has occurred 
since the deposition of these beds, such as that along the Indio fault, 
and the San Jacinto fault where Tertiary beds are displaced. The 
fact that movement has occurred along the recognized fault lines 
during recent notable earthquakes in California indicates that some 
of those faults are still active; and the excellent state of preserva- 
tion of many of the scarps suggests that much of the displacement 
along some of them has occurred in Quaternary time. 


REFERENCES 


ARNOLD, RALPH, “‘Tertiary and Quaternary Pectens of California,” U.S. Geol. 
Surv. Prof. Paper 47 (1906). 

BiakE, WILt1AM P., “Geological Report of California,” H. Balliére, New 
York (1858). Enlarged from Pacific Ry. Reports, Vol. V. 

FarrBANKs, H. W., ‘‘Geology of San Diego County, California,” Cal. State 
Mining Bureau, XI Ann. Rept. (1892), pp. 76-120. 

Kew, W. S. W., “Tertiary Echinoids of the Carrizo Creek Region in the 
Colorado Desert,” Univ. of Cal. Bull. Dept. of Geol. Vol. VIII, No. 5 (1914). 

MENDENHALL, W. C., “Notes on the Geology of Carrizo Mountain and 
Vicinity,” Jour. Geol., Vol. XVIII (1910), pp. 336-55. See also, ‘Ground 
Waters of the Indio Region, Cal.,”” U.S. Geol. Surv. Water-Supply Paper 
225 (1909). 

Vaucuan, T. W., ‘‘The Reef Coral Fauna of Carrizo Creek, Imperial County, 
Cal., and Its Significance,” U.S. Geol. Surv. Prof. Paper 98 (1917), pp- 
355-80. 


MARINE UPPER CRETACEOUS AND A NEW ECHINO- 
CORYS FROM THE ALTAPLANICIE OF BOLIVIA’ 


EDWARD W. BERRY 
Johns Hopkins University, Baltimore 


It is not surprising that rocks of Upper Cretaceous age should 
be present on the high plateau of Bolivia since the gypsiferous 
shallow water phase of the Upper Cretaceous which represents the 
several typically developed marine horizons that are so abundantly 
represented in the more northern parts of the Andean geosyncline 
(e.g., in central Peru), is conspicuous in the Eastern Andes of 
Bolivia and furnished the writer with marine fossils at two localities 
in the Department of Potosi. 

It may be true also that these marginal Upper Cretaceous 
deposits of Bolivia, which are essentially reddish sandstones and 
gypsiferous shales, with subordinate beds of limestone, are present 
in considerable amount beneath the late Tertiary and more recent 
deposits that make up most of the surface of the high plateau 
except where older rocks are folded and project through these 
thick, surficial deposits. 

- No Cretaceous rocks have heretofore been known from the 
Altaplanicie, however, although it is true that Steinmann called 
the rocks at Corocoro Cretaceous. This was based solely on the 
fact that the Corocoro rocks were red, and as the red rocks near 
Potosi, 375 km. distant, were known to be Cretaceous, the unwar- 
ranted assumption was made that the Corocoro rocks also were 
Cretaceous. Many have followed Steinmann’s opinion, as, for 
example Douglas in his recent geological sections across the Andes,’ 
although it would seem that if color is to be an age criterion, an 
Englishman would consider red as indicative of Old Red or New 
Red age, as did Forbes in his classic studies of Bolivian Geology. 


t George Huntington Williams Memorial Publication No. 9. 


2J. A. Douglas, Quart. Jour. Geol. Soc. Lond., Vol. LXXVII (1914), pp. 1-53; 
Vol. LXX (1920), pp. 1-61; Vol. LX XVII (1921), pp. 246-84. 


227 


228 EDWARD W. BERRY 


As a matter of fact the writer showed in 1917 that the Corocoro 
rocks were of Pliocene age™ and this age determination is fully and 
completely established by detailed field studies made by Singewald 
and Berry in 1919, the results of which are now awaiting publica- 
tion.” 

The presence of true Upper Cretaceous in this region rests on 
the species of Echinocorys or Ananychites described below (Figs. 
1-3), for which I am indebted to Sefior Arturo Poznansky, of 
La Paz. It is said to have been collected at Pefias, which is just 
east of the southern end of Lake Titicaca in the Department of La 
Paz, and 55 km. northwest of the city of thatname. It was not pos- 
sible for me to visit the locality, but there is no reason for doubting 
its correctness since the specimen was newly collected at the time of 
my visit to La Paz and must have come from the near vicinity of 
the place named. 

The specimen in itself is of great interest since it is, to the best 
of my knowledge, the first specimen of this interesting genus, so 
excessively abundant in the Upper Cretaceous of Europe, to be 
recorded from South America, and the third or fourth to be recorded 
from the Western Hemisphere. It does not agree with any of the 
numerous described species and may be called in honor of its 
progressive collector Echinocorys (Ananychites) poznanskii sp. nov. 

It may be described as follows: Ambital outline elliptical, roun- 
ded in front and somewhat narrowed behind. Profile (i.e., transverse 
section) subcircular, the apex broadly rounded. Peristome ellipti- 
cal 7 mm. in length by 1o mm. in width, situated about two-fifths 
back from the anterior margin, hence relatively larger and slightly 
more posterior in position than is commonly the case in this genus. 
Periproct posterior, subambital in position, large, elliptical in form, 
with a length of 8mm. anda width of 6mm. The apical system 
is largely obscured by calcitic incrustations. As near as it can be 
made out it consists of four larger genitals and five smaller oculars 
arranged as indicated in the accompanying sketch (Fig. 1a). The 
plastron appears to have been smooth and to comprise the sixteen 
plates shown in the accompanying sketch (Fig. 1), and is unusual, 

1E. W. Berry, U.S. Nat. Mus. Proc., Vol. LIV (1917), pp. 103-64. 

2 Singewald and Berry, Bull. Geol. Soc. Am., Vol. XXXII (1921), p. 66 (abstract). 


A NEW ECHINOCORYS FROM BOLIVIA 220 


if my interpretation is correct, in having a posterior pair of plates 
occluded, asindicated. Centrally the plastron becomes increasingly 
convex toward the periproct in which region it projects about 3 mm. 
below the oral plane of the test. The abulacral plates number 


a 


3 1b 


Fics. 1-3.—Echinocorys (Ananychites) poznanskii sp. nov. 1, lateral view; 1a, 
sketch of arrangement of apical system; 1b, sketch of plastron, comprising sixteen 
plates; 2, dorsal view; 3, ventral view. 


eighteen or nineteen normal pairs from the apex to the ambitus, 
very small at first and increasing regularly and rapidly in size down- 
ward. The pores are very much obscured by incrustation, the 
pore pairs being clearly seen only toward the equatorial region, 
where they are large and central in position, that is, some distance 


230 EDWARD W. BERRY 


within the outer margins of the ambulacrals. The inter-abulacrals 
are larger and there are ten normal pairs between the apex and the 
ambitus. Length of the type 5.1 cm.; width 4.25 cm.; height 
3.95 cm. 

These interesting and characteristic echinoids were figured as 
early as 1565 and are a striking element in the later Upper Cre- 
taceous faunas of Europe. They appear somewhat abruptly in 
the Turonian, undergo but slight diversification in the Santonian, 
and become exceedingly abundant and diversified in the Campanian. 
They have dwindled to two known forms in the Maestrichtian and 
have a single Danian and a single Eocene survivor. They evidently 
found their optimum environment in the relatively clear and 
shallow waters in which the chalk was deposited, which may account 
for their singular rarity in the North American Upper Cretaceous 
where such a large proportion of the sediments are muddy, thus 
offering obstacles to both migration and colonization. In keeping 
with this theory there is a single small species known from the 
Vincentown lime sands of the Rancocas formation of New Jersey, 
Ananchytes ovalis Clark’, and a second large species, Ananychites 
texana Cragin’, from the Austin and Annona chalk in Texas and 
Arkansas. Both are exceedingly rare and, so far as I know, are 
represented by only the type specimens. The Rancocas formation 
was considered to be of Danian age by Clark? although it is probably 
Maestrichtian, and the Austin chalk should probably be correlated 
with the Santonian or Campanian substages of Europe. A single 
additional record of the genus in North America is an incidental 
and queried reference by Aquilera* to the presence of Ananchyies 
sulcatus Goldfuss in the Upper Cretaceous of Mexico. 

Lambert’s excellent monograph’ of the genus renders compari- 
sons easy and this new Bolivian species is seen to somewhat resemble 


rW. B. Clark, U.S. Geol. Survey Bull. 97 (1893), p. 74, Pl. XXXVI, Fig. 1 a-h. 


2F. W. Cragin, Fourth Ann. Rept. Geol. Surv. Texas (1893), p- 145, Pl. XXV, 
Fig. 12; Pl. X XVI, Figs. 1, 2. 


3 W. B. Clark, Geol. Soc. Am. Buill., Vol. VIIL (1897), pp. 315-58. 
4J. G. Aquilera and E. Ordofiez, Datos para la geologio de México (1893), p. 27- 


sJ. Lambert, “Etude monographique sur le Genre Echinocorys,” Mem. Mus. 
Roy. d. Hist. Nat. Bélge, Tome 2 (1903). 


A NEW ECHINOCORYS FROM BOLIVIA 231 


the European Danian species, E. sulcatus Goldfuss, but it appears to 
be most similar. to the European Campanian and Maestrichtian 
forms, EL. Cotteaut and E. Duponii, described by Lambert, who 
prefers to use the generic term Echinocorys proposed in pre-Linnean 
time (1732) by Breynius, and re-adopted in 1778 by Leske, rather 
than Ananchites Lamarck (1801) which has been adopted in recent 
American texts. The age of the Bolivian deposit would appear 
to be Campanian, although more definite confirmation is desirable 
before this correlation can be accepted as final. 


FORMER COURSES OF THE ANDROSCOGGIN RIVER’ 


IRVING B. CROSBY?, 
Boston, Massachusetts 


INTRODUCTION 


The Androscoggin River is one of the most interesting of the 
many examples of disarranged drainage in New England and 
merits more attention than it has hitherto received. 

The present study has involved a critical examination of the 
topographic relations of the Androscoggin and several adjacent 
rivers, differentiating the topography controlled by underlying 
bed-rock formations from the topography of surficial deposits. 

Nearly all that part of the problem south and east of Berlin, 
New Hampshire, is covered by the topographic sheets of the 
United States Geological Survey; but the region north of Berlin 
is best shown on Hitchcock’s map of New Hampshire; and, finally, 
the hydrography of the entire area included in this investigation is 
shown in outline on the accompanying drainage map (Fig. 1). 


TOPOGRAPHY 


The drainage basin of the Androscoggin River, approximately 
125 miles long in a northwest-southeast direction and 30 to 4o 
miles in normal breadth, may be divided into five topographic 
provinces, as follows: headwater, upper, middle, lower, and 
coastal; and in each province the river has characteristics peculiar 
to that province. 

The Androscoggin River is formed by the ‘unetion of the Magal- 
loway River and the drainage of the Rangeley, Lakes, near Errol, 
New Hampshire. The headwater province extends north and east 
from this point to the ultimate sources of the river and will not be 
further considered in this paper. 


™The writer wishes especially to express his indebtedness. to President_ 
W. W. Atwood, under whose direction this work was done. 
2 C. H. Hitchcock, atlas accompanying the Report on the Geology of New Hampshire. 


New York, 1878. 
232 


FORMER COURSES OF THE ANDROSCOGGIN RIVER 233 
The upper province includes the region between Errol and Berlin. 
It is bounded on the southeast by the Mahoosuc Range,’ the 


| 
q\ \ 
s | 4 Parmafhenee DRAINAGE MAP 
S12 ae ee 
zz NG NORTHERN NEW HAMPSHIRE 
zi" fe AND 
\A , Wass 
Wel Subs RIT WESTERN MAINE 
Soe | J SHOWING CERTAIN PRE-GLACIAL 
y i DRAINAGE COURSES 
ey yo} , AS 
s gid ; 
S ae | \\Sec, te 7 OYA 
2 Write “Ss | $ NS NG é Rumford a J a 
3A Nod Sh sats ji Ramen 
\ Sa! F 
1 Torham xe oh fi 
So He eA 


q 
Me 
©. 
% 
SS 0575 
as 


Q a) 
' 
\ ¥ 
N ‘ 
! 
Q 
f 
v 
> 
ke < ey 
| Statlon “4, 
Scale | % ee BAY 
ry ° oy “er 
0) 5 10 Miles 2. aN PORTLAND 
; ; . * ZS 
saesess Pre-glacial drainage lines * Sy 
N. 
— — — —Pre-glacial drainage lines 


inite} Old Ofchara 
less definitely located oats 


Fic. 1.—Drainage Map 


northeastern extension of the White Mountains, and is character- 
ized by broad valleys and mountains of moderate elevation. 


Slaty 
The range, extending from Gorham, New Hampshire, to Grafton Notch, Maine, 
and including the following mountains: Baldcap (3,100), Ingalls (3,570), Goose Eye 
(3,854), and Speckled (4,250), has recently been named the Mahoosuc Range. 


234 


IRVING B. CROSBY 


rocks prevail in the northern part and granitic rocks in the southern 


part. 


The country is heavily mantled with glacial drift; but the 


existing bed-rock contours indicate that the preglacial topography 


BELERMELS 
002 
00r 
009 
009g 
ooo) 
o0el 


Fic. 2.—Profiles of the present 
and former courses of the Andros- 
coggin Drainage 


sunerenranseuetnaras 
ia fn Secoce 
SE RGeEoeeE Becooo 
Beeeseosec latent 
EGE ooooge genes 
ODOMoooon aie ieee 
fn seas fee 
fSoVoR0ecoeoeseooene 
sue Joy eerazouatiienee 
aaedre SuaeSRicnne 
PEE PGES fect 
iio of 
BSS SEe Suu EP reco ag 
Seale coecoNnss2eoS 
nin faa 
OCUSoSeo0s5ncneoe5 
sf off 

Oo K Boo 3 | BATH 
SeotNeSooooe 

Sig) Webede ated Io 1a a 
Penn Bs eiaaieiala 
eSigog0 aa 
[oni a N death w ATIER ton D | | 

sa RunHOEIGEeHEEE 
PH stale aie ee 
Po 

OID Geel 

ey SeueoauedTauotoee 
Fekete 
OO 
2 S000 0000 SSS 
DEOOOOOCoSotme Ease 
st of fo ee 
HEEhty Rees! COO OOOodS 
Race ogoo oloeeooo 
Nee i H 
ooo 
SSESHOE0GE0 ge Gog 
PCCCCeeeee poe 
AB SDOOSeCoteeesCooeo 
PREC eee 
SEOSCOOoEoNoSeceooon 
So SSoOsocerseesooocso 
ono 
SRR EASE CS e Sees ee 
FEBS CSSos eeeeesoeS 
EEE rE xan s re 
SeCeCooocnneoe A 
SEER SoCooreyScaeooo 
PESSSSSSSSSRRs000509 
oo | 
snk tof fof ea eo 


was in the mature stage of dissection. 
The river flows southerly through this 
province, in a broad valley, descend- 
ing 180 feet in 30 miles, or 6 feet to 
the mile. 

The middle province extends from 
Berlin, New Hampshire, on the west 
side of the main ranges of the White 
Mountains, to Bethel, Maine, on the 
east side, its high mountains and deep 
valleys contrasting strongly with the 
lower mountains and broad valleys 
of the upper province. The Andros- 
coggin flows southerly from Berlin to 
Gorham through a narrow valley in 
granitic rocks, and then turns abruptly 
to the east and passes, in a deep valley, 
through high mountains of massive 
siliceous schists. These schists, the 
most resistant rocks of the region, 
form the main ranges of the White 
Mountains, and extend northeasterly 
into Maine. 

In the middle province the river 
descends 420 feet in 30 miles or 14 
feet to the mile, thus having a gradi- 
ent two and one-third times as great 
as in the upper province. At Berlin 
the river drops 200 feet in several 
falls; and there are a number of 
other falls and rapids in the first ten 
miles below that city. Profile 1 
(Fig. 2) shows the distribution of fall 


on the river and emphasizes the contrast of the river gradients 


above and below Berlin. 


FORMER COURSES OF THE ANDROSCOGGIN RIVER = 235 


Comparison of the middle and upper provinces shows that in the 
former the valley is younger and deeper, suggesting a great change 
in the history of the river. ‘These facts, together with the peculiar 
behavior of the river at the great bend near Gorham, where the 
river turns abruptly from less resistant rocks into high mountains 
of more resistant rocks, lead to the belief that the upper part of 
the valley has been occupied much longer by the river and that 
it formerly had another route to the sea. 

The lower province, extending from Bethel to tidewater at 
Brunswick, is marked by broad valleys, low passes, and low, rounded 
mountains. As the sea is approached the mountains disappear and 
the hills are very low, with the exception of a few monadnocks. 
The preglacial topography of the region near the mountains was in 
late maturity and near the coast it was in old age. 

In the lower province the course of the river differs greatly from 
its course in the other provinces in being very irregular, with many 
sharp turns, as can be seen on the drainage map (Fig. 1) or on the 
topographic sheets. In this province the river descends 635 feet 
in 98 miles or 6.5 feet per mile; but this descent consists of a number 
of falls with quiet stretches between. ‘The numerous sharp turns, 
narrows and falls indicate that the present course of the river is 
not normal, and lead one to look for a more direct route. 

The coastal province extends from the head of tidewater to 
the open ocean and is characterized by drowned valleys and by 
islands. In this province, about 8 miles below Brunswick, the 
Androscoggin joins the Kennebec, with a retrograde course, that 
is, the direction of the tributary is reversed in relation to the main 
stream, the apex of the acute angle between the two streams point- 
ing up the main stream. This abnormal junction of the two rivers 
is good evidence that the Androscoggin is not here flowing in a 
normal course. 

The higher ranges of the White Mountain system have a north- 
east-southwest trend; but seaward from these ranges the ridges 
have a north-south or northwest-southeast direction. This general 
trend of the topography seaward from the mountains is well shown 
by the numerous north-south elongated lakes, valleys, and ridges, 
and the absence of similar features with an east-west trend. There- 


236 IRVING B. CROSBY 


fore the northerly-southerly direction may be considered as normal 
for a valley in the region between the mountains and the sea; and 
the few cases of west to east courses are exceptional. 

The writer believes that the present route of the Androscoggin 
through the mountains in the middle province is not the original 
course of the drainage of the watershed above Gorham, and also 
that the route of the river below Bethel is entirely postglacial, the 
present stream occupying parts of several preglacial drainage sys- 
tems. 

The ice sheet crossed this region in a southeasterly direction, 
widening the valleys and passes, and developing the characteristic 
U-profile. 

In the mountains till is widespread, but well-defined moraines 
and drumlins are few. The valleys are relatively narrow, the 
divides are generally high, and streams are principally controlled 
by the bed-rock topography and not by glacial deposits. 

In the lower province, just east of Bethel, the valleys are 
broader, the divides lower; and the streams are partly controlled 
by glacial deposits, while nearer the coast moraines and drumlins 
are common, rock hills are scarce, and stream control is largely by 
surficial deposits. 


FORMER COURSE OF THE UPPER ANDROSCOGGIN 


The present divide between the drainage to Long Island Sound 
and to the Gulf of Maine extends northerly from the Presidential 
Range between the valleys of the Connecticut and Androscoggin 
rivers. This divide has several low passes, one between Berlin 
and Groveton having a summit less than 100 feet above the Andros- 
coggin River or 1,100 feet above sea-level. ‘The mountains forming 
the divide are composed of granitic rocks and are less high and 
rugged than the schist mountains of the main ranges. The 
irregular northerly-southerly trend of the divide, and its broken 
character, contrast strongly with the Presidential, Carter-Moriah, 
and Mahoosuc ranges, which have a regular northeast-southwest 
trend and have only one low place, the valley of the Androscoggin 
at Gorham. 


FORMER COURSES OF THE ANDROSCOGGIN RIVER = 237 


These facts coupled with the character of the Androscoggin 
valley near Gorham indicate strongly that formerly the divide 
was along the Carter-Moriah and Mahoosuc ranges. If this were 
the case, the upper Androscoggin had another outlet and the only 
possible alternate course would have been through the valley now 
occupied by the Upper Ammonoosuc to the valley of the Connecti- 
cut at Groveton. 

The broad valley of the Upper Ammonoosuc extends eastward 
from the Connecticut and connects with the valley of the Andros- 
coggin by three broad, low passes. 

The most southerly of these passes (A on Fig. 3), between 
Berlin and West Milan, is traversed by the Grand Trunk Railway 
and its summit is less than 100 feet above the Androscoggin at 
Berlin. Dead River, a small, sluggish stream, heads in this pass 
and joins the Upper Ammonoosuc at West Milan. Dead River 
appears to have been beheaded near Berlin and before that accident 
it was undoubtedly large enough to make the broad valley which 
_ it now occupies. 

The next pass to the north (B on Fig. 3) lies between West 
Milan and Dummer. It is broad, with a summit approximately 
100 feet above the Androscoggin. No bed rock is visible in the 
saddle or on its slopes, and it has every appearance of being deeply 
drift-filled. 

The third pass (C on Fig. 3) is about three miles farther north, 
where a broad branch valley connects with the Androscoggin 
valley by a pass approximately too feet above that river. This 
pass is very broad with gentle slopes, and no bed rock is visible. 

‘The broad valley between Groveton and West Milan is occupied 
by the Upper Ammonoosuc River. The narrowest part of this 
valley is 6 miles west of West Milan, but it is probable that a lake 
a few miles to the north marks the preglacial course of the river. 

The valley of the Upper Ammonoosuc was deeply filled by 
glacial deposits, which have been only partially cleared out by the 
present stream; and it is obvious that the preglacial valley was 
much too wide to be accounted for by the erosive action of the 
present river. The branch valley toward pass C is floored with 


238 IRVING B. CROSBY 


stratified drift. If all this drift were removed from these valleys, 
there would be a broad valley, ample for a large river, connecting 
the Androscoggin with the Connecticut. 

Northeast of Berlin is a nearly level plain, rising gradually to 
the southeast and ending abruptly against the Mahoosuc Range. 


Fic. 3.—Map showing present and former drainage lines about Berlin, Dummer, 
and Milan, New Hampshire. 


SSS55 Former Drainage Lines 


Several streams flow northwesterly across this plain and enter the 
Androscoggin with retrograde courses. The descent from the 
plain to the Androscoggin at Berlin is abrupt, indicating that 
the river has been rejuvenated enabling it to lower its bed faster 
than is possible with the smaller streams. 


FORMER COURSES OF THE ANDROSCOGGIN RIVER 239 


There are remnants of a bed-rock bench at an elevation of 
approximately 1,200 feet on the valley sides south of Berlin. At 
Gorham the elevation of the bench remnants is 1,300 feet (z in 
Fig. 4), and two miles to the east their elevation is 1,400 feet 
(2 in Fig. 4). The summit of Mount Winthrop (3 in Fig. 4), a 
long spur projecting into the valley five miles east of Gorham, 
has an elevation of 1,575 feet. 


Fic. 4.—View eastward from Pine Mountain, Gorham, New Hampshire 


Looking eastward from Mount Winthrop, long, even-crested 
ridges extend into the valley; and two with elevations of about 
1,100 feet appear completely to block the valley. These are evi- 
dently remnants of the floor of a valley which descended to the 
east from a divide where Mount Winthrop now is. 

The series of bench remnants rising to the east from Berlin and 
culminating at Mount Winthrop mark the level of an old valley 
which extended from the valley now occupied by the Upper Ammo- 
noosuc to a low pass at Mount Winthrop. 

Observations from Mount Winthrop give indication of an old 
land surface descending on either side. On the opposite side of the 
present valley are benches which correspond to Mount Winthrop 


t From a photograph by the Shorey Studio, Gorham, New Hampshire. 


240 IRVING B. CROSBY 


and represent the opposite spur of the old divide. There appears 
to be ample field evidence to prove that eastward from Mount 
Winthrop Divide the drainage was into the Gulf of Maine, and that 
westward it flowed to the Connecticut. 

This divide prevented the Androscoggin from flowing to the 
east, as it now does, and the broad valley of the Upper Ammonoosuc 
represents its course when tributary to the Connecticut. 

Of the three passes connecting the Upper Ammonoosuc and 
Androscoggin valleys, either B or C offers a direct route without 
sharp angles or reversals of direction. Pass A would necessitate 
a much longer course with a sharp turn at Berlin, and it will not 
be considered as the former route of the river. 

A river appears, therefore, to have flowed through the present 
Androscoggin valley to Dummer, through pass C or B, thence down 
the valley now occupied by the Upper Ammonoosuc, into the 
Connecticut. The name “Mahoosuc River” is suggested for 
this stream. 

Profile 2 (Fig. 2) shows the grade of this course, the full line 
representing the present surface and the broken line the probable 
bed-rock surface; and profile 1 (Fig. 2) shows the grade of the 
present river. A comparison of these two profiles shows that the 
former route has a fairly constant grade, a normal condition for 
a large river, while the present course drops several hundred feet in a 
series of falls and is characteristic of a younger river which has not 
yet eroded its valley to grade. 

During the existence of the Mahoosuc River a stream from 
Mount Winthrop Divide joined a stream from Mount Washington 
and flowed north to the point where Berlin now stands, and then 
through pass A into the Mahoosuc River. Three miles north 
of Berlin was a low divide (D on Fig. 3) between the drainage 
to pass A and a stream which joined the Mahoosuc River at Dum- 
mer. Between divide D and Dummer all tributaries joined the 
present river with retrograde courses; and at the time when 
the stream in that valley flowed north instead of south, as at 
present, these tributaries had normal junctions with the main 
stream. ‘The old drainage is shown in Figures 1 and 3. 


FORMER COURSES OF THE ANDROSCOGGIN RIVER 241 


In explanation of the existing drainage it may be stated that 
the modern valley of the Androscoggin below Berlin represents 
more erosion than could have occurred since the last ice retreat. 
Undoubtedly this valley was greatly widened, straightened, and 
deepened by the movement of ice through it, and possibly the 
Mount Winthrop Divide may have been removed and the present 
drainage brought about by ice action. But river piracy also offers 
a simple explanation. 

The stream on the east side of Mount Winthrop Divide had 
a much steeper grade and thus far greater erosive power. Bethel, 
174 miles east of the divide, has an elevation of 635 feet, and the 
elevation of West Milan, 23 miles west of the divide, is 1,015 feet. 
The descent eastward from Mount Winthrop is 57 feet to the 
mile, while the westward descent is 24 feet to the mile. Allowance 
for erosion does not materially alter the relations. 

From the divide to the Gulf of Maine by the preglacial course 
of the lower Androscoggin is 90 miles and from the divide to 
Long Island Sound by the Upper Ammonoosuc and Connecticut 
rivers is approximately 300 miles. ‘These facts show that favorable 
conditions for piracy existed. 

The Androscoggin River wore its valley back through the 
low Mount Winthrop Divide, captured the Peabody River, be- 
headed the Dead River at Berlin, wore through the low divide at 
D, captured several streams from the Mahoosuc Range, finally cap- 
tured the Mahoosuc River, and thus started the present drainage 
system. 


, FORMER COURSE OF THE LOWER ANDROSCOGGIN RIVER 


It has been pointed out that the present course of the Andros- 
coggin, in the lower province, between Bethel and the ocean, is 
abnormal and is of postglacial origin. 

For the preglacial course two possible routes are to be con- 
sidered: eastward along the line of the Grand Trunk Railway 
or southward via the valley of Crooked River. The latter appears 
to be the more probable route on account of its lower summit and 
freedom from obstructions. 


242 IRVING B. CROSBY 


The Grand Trunk Railway occupies a pass in the hills near 
Bryant Pond with a summit elevation of 720 feet above the sea 
or 85 feet above the Androscoggin at Bethel. From Bryant Pond 
the Little Androscoggin River flows southeasterly and joins the 
Androscoggin at Lewiston with a retrograde course. At Snows 
Falls in the town of Paris the valley is blocked by a rock hill and the 
stream is crowded into a very narrow gorge. The height of the 
pass and the obstruction at Snows Falls make it appear that this 
is not the preglacial course of the Androscoggin. However, glacial 
outwash material in the valley and pass indicate that in glacial 
times a large stream flowed this way. 

The postglacial origin of the lower part of the Little Andros- 
coggin is shown by its retrograde course and its interruption 
by two falls. The preglacial course was probably through Poland 
into Casco Bay. 

The present course of Crooked River and the basin of Sebago 
Lake will now be considered. Crooked River joins the Songo 
River a little north of Sebago Lake and the joint stream is tributary 
to that lake. If Crooked River is followed upstream, a surprising 
discovery is made: its source is only two miles from the Andros- 
coggin near Bethel and the divide separating the two rivers is only 
35 feet above the Androscoggin (Fig. 5). 

The examination of Crooked River and the basin of Sebago 
Lake makes it appear that this route was occupied during pre- 
glacial times by a large river which flowed southerly from Bethel. 
This stream was formed in the vicinity of Bethel by the junction 
of two major streams. One of these headwater streams came 
from the Mount Winthrop Divide and the other from the north, 
probably receiving tributaries from the valleys of Sunday and 
Bear rivers. For convenience this preglacial stream will be referred 
to as the “Sebago River.”’ 

The character of the valley of Crooked River becomes of great 
importance when considered as a possible course of a large river. 
The valley is broad and has a direct course with no sharp turns. 
The name of the stream implies a crooked course, but the windings 
are all of minor magnitude; they are the meanders of a small stream 
flowing in a. broad valley of low gradient. 


FORMER COURSES OF THE ANDROSCOGGIN RIVER 243 


Til ONY Ja \ 
l WN 
Z \ Waa : 
5) ; IO \““\e 
= S GN —— AS Jey et af — 
B\ N\A a 
7): =. ——— Ny I 
Th r ~S Gs N) Mf if f 
aid 3 Z go By f\ 
- je, Sy i be, 
Se LOE "\\ 
= 7 SIAN) Bl He 
0 


Fic. 5.—Map of the region about Bethel 


244 IRVING B. CROSBY 


On the lower course of Crooked River there are several low 
falls. The valley is filled with drift and in meandering across this 
drift plain the stream has discovered the ledges that now break 
its bed. The old river was probably free from falls. 

At several places, in its upper part, the valley is rather narrow, 
but in estimating the size of valley required by the Sebago River 
it must be remembered that the upper Androscoggin was then 
tributary to the Connecticut, and the river below Bethel was 
smaller than the present Androscoggin. It follows that the 
narrow places which are in the mountains were ample for the 
Sebago River. 

In the town of Albany, at the place marked L on Figure 1, 
granite bed rock outcrops in the stream at an altitude of 610 feet. 
The grade of the Sebago River from this point to Bethel was 3 feet 
per mile or the same as the present grade of the Androscoggin 
above Bethel. 

The entire valley of Crooked River has been carefully studied 
with regard to narrow places, sharp bends, and elevation of bed 
rock, and nothing has been found tending to disprove the belief 
that this was the course of a large river. 

There are four passes between the headwaters of Crooked River 
and the Androscoggin, marked F, G, H, and J on Figure 5. The 
lowest pass (F on Fig. 5) is a mile north of the head of Songo Pond. 
Its summit elevation is 670 feet or only 35 feet above the Andros- 
coggin. The pass is broad, flat, and swampy, and no ledge is 
visible in the pass or on its slopes; there is nothing to indicate 
bed rock near the surface, and therefore this pass is believed to be 
the drift-filled valley of the old river. Songo Pond is reported 
to be 4o feet deep, which places its bottom 25 feet below the Andros- 
coggin and gives further evidence of a buried valley. The other 
passes are higher and will therefore not be considered further. 

Profile 3 (Fig. 2) shows the grade of the Sebago River from 
Bethel to the sea. The full line represents the present surface, 
and where the old river course is now buried, its probable grade 
is shown by a broken line. A comparison of the profile of the 
Sebago River with that of the present Androscoggin (Profile r) 
shows that the present river has a series of stretches with very 


FORMER COURSES OF THE ANDROSCOGGIN RIVER 245 


gentle gradient broken by falls, while the Sebago River has a con- 
stant grade without falls, indicating that the present course of the 
Androscoggin is partly at least of recent origin and that the course 
of the Sebago River was for a long time occupied by a large stream. 

There is no obstacle today except glacial drift to prevent the 
Androscoggin from following its old course and a dam 4o feet high 
at Bethel would enable it to do so. 

Sebago Lake is drained by the Presumpscot River, a short 
stream with seven falls and an indirect course. It is at once seen 
to be a postglacial stream without a rock valley, and it can be dis- 
missed as a possible course for the Sebago River. 

On the south side of Sebago Lake is Sebago Lake Station, where 
the Main Central Railroad crosses a very low divide. A rise of 
20 feet in the lake would cause it to overflow here. The topography 
is typical of a terminal moraine; kames and kettles are numerous, 
and there are several kettle ponds. 

This morainal dam marks the preglacial outlet of the Sebago 
Basin and the rock valley which the moraine now fills was undoubt- 
edly the route of Sebago River. 

Between Sebago Lake and the ocean the old valley is lost under 
a heavy mantle of till and extensive marine deposits. There are 
well-developed marine terraces at several levels near the coast. 

The old valley probably reached the present shore in the vicinity 
of Old Orchard, where there is a broad indentation of the coast 
with no rock outcrops over a considerable area. Offshore soundings 
tend to confirm this view. 

_ The valley of Sebago River from Bethel to the ocean has been 
located and described as fully as conditions permit. It presented, 
before being clogged with glacial drift, a natural course for a large 
river and is more probable as the preglacial course of the Andros- 
coggin than the present route. 

The Sebago River, which made this valley, had little in 
common with the present Androscoggin. One branch came from 
the Mount Winthrop Divide to Bethel, where it joined a branch 
from the valley of Sunday River. 

South from Livermore Falls through Lewiston the present 
Androscoggin has a fairly direct north-south course in agreement 


246 IRVING B. CROSBY 


with the general trend of the topography. This region has not 
been studied in detail, but it appears probable that in preglacial 
times there was a river following a course similar to that of the 
present Androscoggin, between Livermore Falls and Lisbon Falls, 
and emptying into Casco Bay. This preglacial stream with the 
Little Androscoggin was the principal river between the Sebago 
and the Kennebec. 

The Androscoggin River east of the mountains now occupies 
parts of two preglacial drainage systems. The main valleys of 
these two systems were clogged with drift in many places and the 
resulting drainage follows a circuitous route and empties into the 
Kennebec with a retrograde course. 


CONCLUSIONS 


This study has shown that the present course of the Andros- 
coggin River is largely postglacial in origin. 

It appears that in preglacial times there was a low divide at 
Mount Winthrop on the northeastward continuation of the main 
ranges of the White Mountains. All the drainage northwest of 
this divide went to the Connecticut River, passing through the 
valley now occupied by the Upper Ammonoosuc. This drainage 
system, including the Androscoggin above Dummer and the stream 
from Mount Winthrop Divide, is here called the “ Mahoosuc” 
system. 

Eastward from this old divide a stream flowed to Bethel where 
it was joined by a branch from the valley of Sunday River; and 
the stream resulting from their union, here called the “Sebago 
River,” flowed southward through the valley now occupied by 
Crooked River, through the basin of Sebago Lake and into the 
ocean near Old Orchard. 

East of the Sebago system was a north-south stream which 
with its tributaries, including the Little Androscoggin, drained 
nearly all the region between the Sebago and Kennebec rivers. 

East of Bethel the drainage lines were greatly disarranged 
by heavy deposits of glacial drift. A single stream may occupy 
portions of several preglacial valleys and have other portions of 
its course of postglacial origin. 


FORMER COURSES OF THE ANDROSCOGGIN RIVER 247 


The capture of the Mahoosuc River by the Sebago River is 
probably not due to glacial disarrangement, but is best explained 
as preglacial piracy by the Sebago River. 

The economic effects of these drainage changes are very impor- 
tant. The falls at Berlin and Gorham are due to piracy; and 
the pirate stream has not yet eroded the valley to grade. The 
falls at Rumford, Livermore, Lewiston, Lisbon, and Brunswick 
are due to the displacement of the river by glacial deposits. 
These falls furnish a large amount of valuable water power and 
there are important mills at all the places named. Prior to the 
rearrangement of the drainage there was probably little if any 
potential water power. When the pirate stream cut through the 
divide it made an easy route through the mountains which is now 
occupied by a railroad. 


NOTES ON THE SAND DUNES OF NORTH- 
WESTERN INDIANA 


GEORGE B. CRESSEY 
Yale University, New Haven, Connecticut 


Of the sand dunes of North America, few present more interest- 
ing phenomena than those bordering the southern shore of Lake 
Michigan. Not only do they illustrate all phases of dune activity 
but they have a significance in ecology and post-glacial history as 
well as rare scenic beauty. ‘They are conspicuously developed in 
northwestern Indiana, their maximum development being confined 
to the 20 miles of shore between Gary and Michigan City. Heights 
of too feet are common and the height of the frontal ridge along 
the lake commonly ranges from 75 to 175 feet. Parts of the dune 
complex are fixed and forested, while in other areas the sand is 
being shifted. The dunes are limited largely to a belt about a 
mile wide along the shore, although in places low dunes extend 
several miles inland. 

Along the western shore of Lake Michigan for 20 miles north of 
Chicago the waves are actively cutting into the thick drift, here 
composed of rather fine materials. This wave-cut cliff may be 
seen on the Highwood topographic map and in places is more than 
80 feet high. A very considerable amount of erosion has taken 
place and some of the débris has been transported southward by 
the alongshore currents. Evidence of this movement is seen in 
the accumulations of sand on the northern sides of piers and the 
southerly deflection of streams’entering the lake where unprotected 
by breakwaters. With the wear and sorting involved in the trans- 
portation, the material which reaches the southern end of the lake is 
almost exclusively quartz sand. Here the topography and prevail- 
ing winds are such that deposition takes place. The shore sands 
become the prey of the winds and the dunes reach their greatest 
development. 

In order to examine the effect of the wind transportation on the 
sand grains, nearly fifty samples were gathered along seven lines 

248 


ON THE SAND DUNES OF NORTHWESTERN INDIANA 249 


running back from the lake and at right angles to the trend of the 
shore. On the assumption that all of the dune sands had been 
blown inland from the present shore, it was anticipated that the 
sand would show increasing signs of wear with increasing distance 
from the shore. The microscope was used to determine the round- 
ing of the grains, and for the gradations in size a tower of sieves 
was found most satisfactory. The customary procedure in the 
latter case was to weigh out 25 grams of sand and pass it through 
sieves of 60, 80, and 100 mesh to the inch, thus giving four grades of 
size, namely the sand retained in each screen and that passing 
through the too mesh sieve. These groups were then weighed and 
converted into percentages. 

The percentage of coarse sand was found to decline back from 
the shore and there was a corresponding increase in the finer grades. 
At a distance varying from one-fourth to one-half mile from the 
beach, however, came a sharp increase in the proportion of coarse 
sand along most of the lines on which samples were taken. The 
change was quite regular, and reference to field notes showed that 
on the lakeward side of the location for each sample which showed 
the unexpected coarseness there was an area of muck or marsh. 
Subsequent field investigations showed a more or less continuous 
belt of this swamp or muck land between the dune belt nearest the 
lake and the next one to the south, the dunes to the south being 
lower and more rounded. Along the southern border of this flat 
there were in places topographic suggestions of an old beach 
ridge and from burrows of animals and holes which were dug, 
very coarse sand was found. In the analyses of the samples from 
the second dune belt, next south of the.swamp belt, there was 
again a progressive decrease in size of grain grading to the south. 
The evidence, both topographic and petrographic, therefore points 
to an old shore line, the dunes immediately to the south having been 
built when the lake shore stood at its position, rather than having 
been blown inland from the present beach. The muck or marsh 
marks the approximate position of a former shore, and was perhaps 
developed behind a barrier beach. 

The South Shore Electric Railway in the dune country runs 
parallel to the lake and about a mile south of it. Most of the 
dunes are north of the railroad but for much of the distance there 


250 GEORGE B. CRESSEY 


is a definite beach ridge, together with low dunes, just south of the 
road. Analyses of sand along section lines through this third 
belt of dunes showed an increase in the size of grain as compared 
with the sand immediately to the north, and on later search shells 
and pebbles were found. ‘The evidence for a former strand line in 
this position is strong. 

South of the third dune belt just referred to, especially in the 
neighborhood of Johnsville, there is another pronounced beach line 
banked with low dunes on the south. The beach here is well 
marked and strewn with beach-worn pebbles. None of these 
shore-line evidences were anticipated but were first suggested by 
variations in the coarseness of the sand. Since the interruption 
in the gradation of size was a rather regular phenomenon and 
scarcely to be accounted for if the sand had been subjected to greater 
wear, search was made for additional evidence of shore lines. In. 
connection with each of the three suspected beach lines, additional 
criteria were found in the form of beach gravels or coarse sands, 
and in the case of the two beaches most distant from Lake Michigan 
the topographic evidence was strong. 

The history of the dunes therefore appears to be closely asso- 
ciated with the development of Lake Michigan. With the north- 
ward retreat of the Late Wisconsin ice sheet a lake was formed in 
the Michigan basin, known as Lake Chicago. Since the present 
northern outlet was blocked by the ice, the water level was higher 
than now, the overflow being to the south through the Illinois 
River, as is well known. With the lowering of this channel and 
the uncovering of other outlets by the retreating ice front, there 
followed successive levels of the water. These halts are known as 
the Glenwood, Calumet, and Tolleston stages of Lake Chicago, the 
level of the water above Lake Michigan being respectively 56, 32, 
and 17 feet. In the absence of topographic maps or precise eleva- 
tions in the area under consideration, determinations of the eleva- 
tions of the old shore lines have not been made. Judging from the 
gradient of the two streams which cross the dunes and by the posi- 
tions of the beaches in adjacent areas, it is thought best to assign the 
four belts of dunes to the various stages in the history of the lake, 
including the present. On this basis, each dune belt in the complex 


ON THE SAND DUNES OF NORTHWESTERN INDIANA 251 


dates back to a distinct period in post-glacial history, and while 
sand from one tract has been blown into the other belts, in places 
obscuring the evidence of a beach, each tract is on the whole distinct. 

The exact locations of the shore lines have not been mapped in 
the region involved, and in some places this will be impossible 
because of the moving dunes; their approximate position, how- 
ever, is as follows: Tolleston Beach passes through the southern 
border of Miller and along the southern margin of Long Lake. 
At Wicliffe it is doubtless north of the station on the electric line 
and beginning with the eastern border of the broad opening through 
the dunes at Wilson it is marked by the southern edge of the marsh 
or muck tract within the main body of the dunes; this continues 
almost to Michigan City. The Calumet Beach is just south of the 
Little Calumet River in Lake County, and as far west as Wilson 
where Lake Chicago had an extensive embayment. Beyond this 
it parallels the South Shore Electric Railway as far as Johnsville, 
being immediately south of it. The Glenwood Beach is south of 
the Michigan Central Railroad in the eastern portion of Porter 
County and after swinging south in a semicircle about the former 
embayment, strikes west south of Hobart, some seven miles from 
Lake Michigan. 

Due to the complex processes which have shaped the sand it is 
difficult to make more than generalizations as to the rounding 
and polishing of the grains, and shifting winds render it impossible 
in some places to ascertain whether the sand at any given point 
has recently arrived or is part of the original material of the dune. 
Microscopic examination of the beach sand shows it dominantly sub- 
angular. It is generally rather transparent and little pitted, the 
corners frequently being sharp with little rounding. Back from 
the beach the characteristic description would be subspherical, 
many of the grains being well rounded and bean-shaped. All 
sand which has been subjected to sand-blast action is pitted and 
frosted, this being quite noticeable in the dune sand. Within 
the older dune belts this gradation from the former shore line land- 
ward is also apparent, but on account of the longer period of 
exposure and consequent aeolian wear of all the sand, the contrast 
between beach and dune sands is here less pronounced. 


PETROLOGICAL ABSTRACTS AND REVIEWS 


ALBERT JOHANNSEN 


Berek, M. “Zur Messung der Doppelbrechung hauptsichlich 
mit Hilfe des Polarisationsmikroskops,” Centralbl. f. Min. 
Geol.-u.-Pal., 1013, 388-00, 427-35, 404-70, 58@_o 2-0 ieee 

The new Berek compensator here described is somewhat similar in 
pattern to the Nikitin quartz compensator. In the present case, how- 
ever, the mineral plate is calcite. The author shows that the dispersion 
of mineral sections nearly at right angles to the optic axis becomes prac- 
tically zero, consequently any inactive, uniaxial mineral is suitable. 

Since rather a thick section is required to give the higher orders of colors, 

quartz cannot be used on account of its rotary polarization. The 

construction of this instrument permits much more accurate measure- 
ments than that of Nikitin. A calcite plate, o.1 mm. in thickness, cut 

at right angles to the optic axis, is attached to a rotatable axis in a 

slider which is to be inserted in the slot in the microscope above the 

objective. Since it is thus between the polarizing prisms, no cap nicol 
is required. The amount of rotation of the plate is read from a graduated 
drum and vernier to 2’. With the thickness of calcite used, it is possible 
to read colors to the fourth order. If higher readings are desired, the 
calcite plate may readily be removed and replaced by one that is thicker. 
The sensitiveness of this compensator is shown to be equal to that of 
Babinet for moderate double refraction, and greater for low colors. 


BerEK, M. “Uber Zirkularpolarisation,’ Fortschr. d. Muin., 
Krist., u. Peir., WV (@or4), 73-114, 221-025) oe 
A full discussion of the phenomenon of rotary polarization. 


Berek, M. “Die astigmatischen Bildfehler der Polarisationspris- 
men,” Centralbl. f. Min. Geol. u. Pol., 1919, 218-24, 247-55, 
275-84. Figs. 5. 

The insertion of the nicol prism between the objective and ocular 
causes astigmatic and focal disturbances. The latter is usually com- 


252 


PETROLOGICAL ABSTRACTS AND REVIEWS 253 


pensated by means of a lens above the analyzer, but the former has 
hitherto been left uncorrected. In this paper the cause of the astig- 

matism is discussed and lenses for its correction are suggested. Two 

photographs of the same slide show the remarkable improvement in 
definition with the corrected nicol. 


Berek, M. “Uber die Berechnung der Polarisationsverhaltnisse im 
Gesichtsfelde der Polarisationsprismen,” Verhandl. d. Deutschen 
Physikal. Gesell., XXI (1919), 338-46. 


Berek, M. ‘Die Scharfentiefe des Mikroskops,” Zeiztschr. f. wis- 
sensch. Mikroskopie, XX XVII (10920), 120-22. Fig. 1. 


Berek, M. “Uber die einfachen und zusammengesetzten charak- 
eteristischen Konstanten der Mikroskopobjektive,” Zeztschr. f. 
wissensch. Mikroskopie, XX XVII (1920), 36-41. Figs. 2. 

A discussion of the use of 250 mm. and A in the determination of 
the enlargement of objectives. 


BERKEY, CHARLES P. “Geological Reconnoissance of Porto 
Rico,” Aun. N.Y. Acad. Sci., XXVI (1915), 1-70. Figs. 20, 
maps and profiles 2. 


This is a report of the New York Academy of Sciences Expedition to 
Porto Rico undertaken, in part, to determine the nature and origin of the 
rock formation and to group them into series suitable for use in subse- 
quent geological work. Two series are described—a younger, consisting 
of Tertiary shales, reef limestones, and recent deposits, and an older 
including tuffs, ashes, shales, conglomerates, limestones, and a great 
variety of probably pre-Tertiary intrusives. Igneous rocks described 
briefly are extrusive basalts and andesites, and intrusive andesite- 
prophyry, granite-porphyry, granite, and diorite. 


Boeke, H. E. “Die Ejisenerze,”’ Die Umschau, XXIII (1919), 
289-92. 
A popular article on the occurrence of iron ore. 


254 PETROLOGICAL ABSTRACTS AND REVIEWS 


Bowen, N. L. ‘Differentiation by Deformation,” Proc. Nat. 
Acad. Sci., VI (1920), 159-62. 
Discusses differentiation according to Darwin’s theory of fractional 
crystallization and the squeezing out of the mother liquor. 


Bowen, N.L. ‘The Sodium-Potassium Nephelites,” Amer. Jour. 
iin IU Gon), mg aa@s Les, 2 
While this is strictly a mineralogical paper, it is of importance to 
petrographers in showing the inaccuracy of calculating nephelite as 
Na,O-Al,O,°6SiO.. It is here shown that the molecules NaAlSiO, 
and KALSiO, are fundamental constituents of it, although they may 
be present in variable amounts. 


Brauns, R. “Skapolithfiihrende Auswiirflinge aus dem Laacher 
Seegebiet,” Neues Jahrb., B.B., XXXIX (1914), 79-125. 
leis 2. 


Describes seventeen different kinds of scapolite rocks from the 
Laacher Sea region. Numerous analyses are given. 


Brauns, R. ‘Der Laacher Trachyt und seine Beziehung zu 
anderen Gesteinen des Laacher Seegebietes,’ Neues Jahrb., 
Babe DxGale ron6)420—502, eellseae 


This is a very complete study of the Laacher Sea trachyte and 
related rocks. Twenty-three new analyses are given, all of which are 
also recast into molecular proportions reduced to 100, and computed 
according to Osann’s system. Nine older analyses are quoted for com- 
parison. A history of volcanic activity in this region is given, and the 
following conclusions as to the relationships of the rocks are reached: 
The white pumice, the Dachsbusch trachyte, and the light Laacher 
trachyte are so closely related chemically that they are thought to have 
been derived from the same magma, their differences in habit being 
due to the action of the inclosed gases. Petrographically they may be 
designated phonolitic trachytes and trachytic phonolites. Further, 
these rocks are closely related chemically to the adjacent noselite- 
phonolites, and it is probable that they came from the same magma, 
their differences being due perhaps to greater differentiation or to 
assimilation of crystalline schists. The dark Laacher trachyte is more 


PETROLOGICAL ABSTRACTS AND REVIEWS 255 


closely related to the tephrites of the region, and it is assumed that the 
trachyte assimilated some of its constituents. The oldest and at the 
same time most basic rocks of the region are the tephritic lavas. These 
are followed by younger and progressively more acid noselite-phonolite, 
white pumice, and finally light trachyte. Among the ejectamenta 
are fragments having the character of dike rocks, some of which are 
more basic than the tephrites, and others more acid than the Dachs- 
busch trachyte. Fragments of plutonic rocks, so far as these have been 
analyzed, show intermediate chemical characters. The rocks are all 
fully described and only lack estimates of the mineral percentages to 
present good word-pictures of their appearance. 


Brauns, R. “Neue skapolithftihrende Auswiirflinge aus dem 
Laacher Seegebiet,’”’ Newes Jahrb., I (1917), 9-44. Pls. 2. 
Nine more scapolite-bearing rocks from the Laacher Sea region are 
described. ‘There are many analyses, and a determination of the refrac- 
tive index of the sulphate scapolite. 


Brauns, R. “Uber aufgewachsene Karlsbader Zwillinge von 
Sanidin vom Laacher See,” Newes Jahrb., I (1917), 45-40. 
Pie a. nee : 


Brauns, R. “Einige bemerkenswerte Auswiirflinge und Ein- 
schliisse aus dem niederrheinischen Vulkangebiet,” Centralbl. 
MoT GEOL Ua Pole Toro. TTA. Hig. 1. 


BRENNER, TH. “Uber Theralit und Ijolit von Umptek auf der 
Halbinsel Kola,” Bull. Com. Géol. Finlande, No. 52, 1920, 
T=20, | JONUSh ie ee 

Here are described the Kola Peninsula theralite, and a new ijolite 
from the Tachtarwun Valley. The descriptions are elaborate and good 
with the exception of the omission of an estimate of the mineral per- 
centages in the theralite. Omissions of this kind lead to such errors as 
the statement that ‘‘es giebt sowohl grobk6rnige, helle als kleinkérnige, 
dunkle Arten.” But theralite is a name given by Rosenbusch to cer- 
tain nephelite, plagioclase plutonic rocks which he thought were repre- 
sented by certain tephrites and basanites described by Wolff from the 


256 PETROLOGICAL ABSTRACTS AND REVIEWS 


Crazy Mountains. As a matter of fact, they were first found some 
years later by Wolff in Costa Rica. While in his first published descrip- 
tion Rosenbusch does not make the predominance of the dark constituent 
a necessary qualification, in his later works he grouped the theralites and 
shonkinites, and said of them: “Die Theralithe und Shonkinite sind 
hypidiomorphkornige Tiefengesteine, welche bei starker Vorherrschaft 
der dunklen Gemengtheile durch die Mineralcombination Nephelin mit 
Kalknatronfeldspath, bezw. Nephelin mit Kalifeldspath, charakterisirt 
sind,’ consequently there can be no light theralites. The analyses of 
various theralites are recast into both the C.I.P.W. and Osann’s system. 
The ijolite is described fully and analyzed. It differs from the usual 
ijolites in the presence of arfvedsonite-hornblende. The modal per- 
centages determined by the Rosiwal method are: nephelite 38.25 
per cent, arfvedsonite-hornblende 31.96 per cent, aegirite-augite 16.24 
per cent, titanite 5.78 per cent, mosandrite 3.81 per cent, magnetite 
2.37 per cent, apatite 1.59 per cent. The rock, consequently, may be 
classified as 3126 (new form, or 3131 old form) of the reviewer’s system. 
The chemical analysis of the rock is computed in the C.I.P.W. system as 
Ivaaros, and is compared with three other ijolites, one a Malignos, one an 
Ivaaros, and one an Jjolos. In an appendix are given analyses of fifty- 
four theralites and related rocks. 


REVIEWS 


The Geology of Hardin County and the Adjoining Part of Pope 
County. By StTuART WELLER with the collaboration of 
CHARLES Butts, L. W. Currier, and R. D. SatrisBury. 
Illinois Geological Survey, Bull. No. 41, 1920. Pp. 377, 
plss 11, figs. 27. 

This report represents the results of three seasons’ field work by the 
author in conjunction with Messrs. Butts and Lee of the U.S. Geological 
Survey and part of two seasons by Mr. Currier in the study of the fluor- 
spar deposits of the county. In addition a chapter on the geography 
of the region is contributed by Mr. Salisbury. 

Hardin County lies in the southeastern part of the state and wholly 
within the Ozark country, an area of considerable relief and in a mature 
stage of dissection. The history of the topography involves four periods 
of uplift, the three cycles of erosion other than the present one, being 
known as the Karber’s Ridge, McFarlan, and Elizabethtown cycles. 

Structurally the area is characterized by rather flat-lying strata 
surrounding an area of doming and intense faulting. The faults are of 
the normal type and form a network along the edges of the dome which 
is roughly circular in outline. They originated through the collapse of 
the dome. Likewise remarkable is the presence within the region of 
igneous intrusives in the form of dikes, sills, and plugs. The origin 
of the dome involves the intrusion of these rocks, the collapse following 
either the gradual spreading out of the lava or its partial withdrawal. 
The structure is remarkable and has but few known parallels. 

‘The rock formations of the region include Devonian, Mississippian, 
and Pennsylvanian sediments, involving a thickness of some 4,000 feet. 
Of special note is the contribution to the knowledge of the Chester group. 

The oldest formation present, known locally as the Devonian lime- 
stone, is correlated with the Lower Devonian, Onondaga, and Hamilton. 
Lithologically it is a unit. The overlying Chattanooga shale contains 
but a meager fauna and from its stratigraphic relation may be corre- 
lated with either the Upper Devonian or Lower Mississippian, or both. 

The Meramec group, as originally defined by Ulrich, included the 
Warsaw, Spergen, and St. Louis limestones. A study of the Ste. Gene- 
vieve faunas convinces Professor Weller that this formation is much 


251) 


258 REVIEWS 


more closely related to the St. Louis group than to the Chester with 
which Ulrich had allied it, chiefly due to the presence of a species of the 
typical Chester genus Coeloconus. On this basis and the recent recogni- 
tion of an unconformity separating the Renault and Ste. Genevieve 
formations, Weller redefines the Meramec group so as to include the 
Ste. Genevieve. 

This report represents the most recent work on the subdivision of 
the Chester and the correlation of these units. Especial attention is 
called to the following corrections of previous correlations of the forma- 
tions of southeastern Illinois and Kentucky with those of Randolph 
County: (1) the Renault and Shelterville formations, the Upper Ohara 
of Ulrich, are correlated with the Renault. Professor Weller considers 
those species of the genus Talarocrinus with the thickened plates form- 
ing bilobed basis as very characteristic of the Lower Chester. This 
fauna is typically developed in the Upper Ohara; (2) the Golconda 
limestone with the Lower Okaw; (3) the Cypress sandstone with the 
Ruma; and (4) the Bethel sandstone with the Yankeetown formation. 
Methods of correlation are discussed in detail. 

The igneous rocks of the region are found chiefly in the vicinity of the 
Ohio River where the more rapid erosion has exposed them. They are 
of three types: fine-grained, dark-colored lamprophyres, medium- 
grained, dark-colored peridotites, and volcanic breccia. 

Economically Hardin County is noted for its fluorite deposits, the 
annual output of the county being over two-thirds that of the country. 
Fluorite occurs as (1) vein deposits, replacing calcite fillings of the 
fault planes; (2) bedded deposits where fluorite has replaced limestone 
bedrock; and (3) superficial residual deposits. Small quantities of 
lead and zinc are recovered incidental to the fluorspar exploitation. 
The origin of the deposits involves (1) doming and faulting in post- 
Pennsylvanian, (2) deposition of calcite in the fault fissures, (3) minerali- 
zation by fluriferous magmatic solutions replacing calcite, and (4) the 
introduction of metallic sulphides. 

The only other large resource of the region is agriculture. There is 
an abundance of limestone suitable for lime, cement, and ground rock 
for fertilizer, as well as an unlimited supply for road metal. In regard 
to the oil and gas possibilities, while the structure is favorable, but little 
is known of the underlying pre-Devonian sediments. 

In the portion of the bulletin devoted to systematic paleontology , 
the work is limited to those common species in need of further illustra- 
tion and description and to those forms important for their bearing upon 


ri 


REVIEWS e256 


the correlation problems. The majority of these are Chester forms, 
together with some Ste. Genevieve and St. Louis species. Of interest is 
a new species of Septopora, S. similis from the lower St. Louis which is 
almost identical in character with the Chester S. subquadrans. This 
occurrence is noteworthy in that the genus Septopora has hitherto been 
considered highly characteristic of the Chester. Seventy-two species 
are described and figured, prominent among which are species of Talaro- 


crinus and Pentremites. 
A. C. McF. 


On the Crinoid Genus Scyphocrinus and Its Bulbous Root Camaro- 
crinus. By FRANK SPRINGER. Memoir Smithsonian Institu- 
tion. tory. Pp. 74) pls. 0, figs. 16. 

For more than a half-century there have been known to paleontolo- 
gists certain bulb-like, supposedly crinoidal or cystoidal, bodies which 
were described from American localities in 1869 by Hall as Camarocrinus. 
Similar structures had been known for some time from the Silurian of 
Bohemia where they had been found by Barrande. He had named 
them Zobolithus without describing them. 

In 1904 Schuschert summarized all the known facts touching the 
occurrence and relations of this form. He found that these structures 
were of widespread occurrence in both Bohemia and America. In the 
former they were confined to a horizon equivalent to the American 
Rochester shale, and in the latter to the Manlius and Helderbergian. In 
Bohemia they were commonly associated with the genus Scyphocrinus 
which was as yet undescribed from America. They were frequently 
found in beds void of any other crinoidal remains and a large majority 
were found in strata with their stalked end-down. He came to the 
conclusion that 
Camarocrinus thus appears to be the float of an unknown crinoid, that was 
held together after the death of the individual by the firmly interlocked double 
walls of the exterior and interior, while the crown and stalk dropped away. 
Under this hypothesis the float drifted with the sea currents, was finally 
filled with water and, the attenuated end being heavier, sank in that position. 


It-is the purpose of this paper to present the results of some later 
studies by Mr. Springer which have resulted in a change in the concep- 
tion as to the functional nature of the so-called Camarocrinus. He 
finds not only that the genus Scyphocrinus does occur in abundance in 
America but that the Camarocrinus bulbs are directly connected at the 
distal end of the stem of crinoids belonging to that genus. Moreover 


260 REVIEWS 


these bulbs, when in their original position, occur with the stalked 
end upward and not downward as previously supposed, the bulb being 
merely a distal specialization of a true root and not a float. 
The bulb-like process may itself be described as a 

rigid, hollow, chambered root, consisting of a large spheroidal bulb with a short 
projecting collar, a stem base with bifurcating roots resting in and forming 
a large part of the floor within the collar; and several internal, laterally opposed 
sacs which abut against the inner side of the bulb wall... . . The collar and 
bulb wall consist of single layers of similar plates derived from rootlet systems 
originating at the ends of the proximal root branches. 


Eight species of Scyphocrinus occurring in America are described and 
are figured together with several species of other genera. A table of 


analysis of the American forms is given. 
A. C. McF. 


The American Species of Orthophragmina and Lepidocyclina. By 
JosrpH A. CusHMAN. Shorter Contributions to General Geol- 
ogy. U.S. Geological Survey, Professional Paper 125 D, 1920. 
IPO AO uel), lS. 7s. 

The genera Orihophragmina and Lepidocyclina belong to the group 
of orbitoid Foraminifera, a group of excellent horizon-markers due to 
their limited stratigraphic range, wide geographic distribution, and 
great abundance in the Tertiary. Hitherto the group has received but 
little attention by American paleontologists. In the present paper the 
author describes all known American species, which form but a small 
percentage of those which he believes will be later described and have 
been described from the European Tertiary. 

Orthophragmina includes those species characterized by the presence 
of rectangular chambers in the equatorial band. The genus is limited, 
so far as known, to the Eocene, and in America, chiefly to the upper 
part. Lepidocyclina differs in that the chambers of the equatorial belt 
are typically hexagonal. It ranges through the Eocene, and Lower and 
Middle Oligocene. 

Sixteen species of Orthophragmina are described, of which two are 
new, and thirty species and varieties of Lepidocyclina of which eleven 
are new. The author includes a table of tentative correlations of the 
Tertiary of Panama by T. W. Vaughan, a key to the species of Lepzdo- 
cyclina and a list of those species which are considered as good index 


forms for the Tertiary of the Coastal Plain. 
| A. €; Mes 


REVIEWS 261 


North American Early Tertiary Bryozoa. By FERDINAND CANU 
and Ray S. BassLer. United States National Museum, Bull. 
No. 106, 1920. Pp. 878, pls. 162, figs. 279. 


This monograph forms a pioneer work in the field of American 
bryozodlogy. Despite the great abundance and splendid preservation 
of bryozoa along the Atlantic and Gulf Coast regions, there has hitherto 
been so little done with them as to leave the field almost virgin. 

Under the term Lower Tertiary the authors have included the 
Midwayan, Wilcoxian, Claibornian, Jacksonian, and Vicksburgian 
formations. ‘There is found here a much richer bryozoan fauna than in 
the succeeding formations. A study of the Miocene, Pliocene, and 
Pleistocene bryozoans has been completed and it is hoped that it will 
soon appear as a monograph supplementing the present work. An 
extensive table showing the geographic and stratigraphic distribution of 
the different species is given. 

Tertiary bryozoans are so intimately related to living forms that a 
thorough knowledge of the taxonomy and anatomy of living species is 
necessary for the proper classification and interpretation of the structures 
found. Accordingly the authors have produced a monograph which is of 
use both to the zodlogist and to the paleontologist. Numerous text 
figures are introduced to make the report of value to the non-specialist. 
A detailed discussion of the structure and classification of living forms is 
given. . 

With the exception of one undetermined species, referred to the order 
Cienostomata, all Tertiary bryozoans belong either to the Cheilostomata 
or Cyclostomata, the latter being about half as numerous as the former. 
These are included in the subclass Gymnolaemata, which includes all 
fossil bryozoans, and is characterized by a circular row of tentacles 
surrounding the mouth. It is in turn incorporated in the class Ectoprocta 
which is distinguished by the presence of the anal orifice outside the 
row of tentacles. 

The Cheilostomata which were probably derived through the Paleozoic 
Trepostomata, differ from the Cyclostomata in that they may be either 
calcareous or chitonous, and the aperture is closed, upon the retraction 
of the polypide, by an operculum. The Ctenostomata are chitonous 
‘or gelatinous forms with the aperture closed by a tooth-like process. 

The studies of the authors have led them to discard for the present 
the former major classification of the Cyclostomata, retaining for con- 
venience only the two larger divisions the Ovicellata and Inovicellata. 
In classifying the Cheilostomata orders were based upon general mor- 


e 


262 REVIEWS 


phology, suborders on the form of aperture and operculum, and families 
on the presence of cardelles, lyrula, ovicells, and radicells. 

With the increased recognition of the value of bryozoans as horizon- 
markers, the student will find this systematic report a reference of 


constant value. 
A. C. McF. 


Louisiana Lignite. By ROBERT GLENK. Department of Conserva- 
tion, Division of Economic Geology, State of Louisiana, 
Bull. No. 8. New Orleans, 1921. 


Although Louisiana lignite has been consistently mapped on all 
fuel maps of late years as a possible source of low-grade fuels, it has not 
until recently begun to receive the attention that it merits. The 
report here reviewed shows a gratifying interest by the state of Louisiana 
in its available resources—an interest that, since the days of Harris’ 
and Veatch’s work on the state survey, appeared to have flagged. 

The lignite of Louisiana is of Eocene age and distributed vertically 
in the Wilcox, Claiborne, and Jackson groups. It varies in thickness 
from a few inches to twenty feet. Interesting paleobotanical collections 
have been made from it. Jn the adjacent states of Texas and Arkansas 
lignite is mined in considerable quantities by stripping the overburden, 
or by the room-and-pillar method where the dip of the lignite bed is 
appreciable; similar methods, though not yet used in Louisiana, might 
well be applied there in the future. 

Figures expressing the lignite reserves of the state are not presented; 
they would be interesting and use‘ul. Analyses of several Louisiana 
lignites yielded the following: 


Percentage 
MOISTURE eet se en ae Pae RenNeES 5-47 
Volatile combustible matter........ 30-75 
Fixed carbonvas 2.40 era pens te ces 5-40 


The highest thermal value obtained was 8,046 B.T.U.’s. The analyses 


presented show Louisiana lignite to be of an intermediate grade. With 


destructive distillation the end products were: 


Percentage 
ENGWEY ChanioOVEle oe aoc soucwe soos 20 —30 
Dati st: G itheme auto Se Te wee eae 4.5- 5 
Carbonizedsresiducsae eee 50 -56 
Gastandsloss qi)... eee ae 18  -20 


The carbonized residue has a heating value equal to good anthracite; 
an average ton of lignite yields 8,o00 cubic feet of gas (mainly hydrogen, 


REVIEWS 263 


methane, and carbon monoxide) with a heating value of 450 B.T.U.’s 
per cubic foot; this is a higher value than producer gas or Mond gas, but 
lower than coal gas or natural gas. 

In general it appears that the chief use to which Louisiana (and 
probably all Gulf Coast) lignite can be put is in the briquetting of the 
carbonized lignite, with the recovery of the tar and ammonia as by- 
products and the utilization of the gas in producing electric power. Such 
a use avoids the difficulties involved in shipping and storing the com- 
bustible fuel, and makes available to the Gulf States, where hydraulic 
power is not always obtainable, a cheap source of electricity. 

The report, although in a large measure a compendium of other works 
on the subject, presents a series of original and very careful fuel analyses, 
and is a most valuable addition to the scanty information on the subject 
of the geology and technology of utilizing Louisiana lignites. 


C. H.'B.,. Jr: 


A Glossary of the Mining and Mineral Indusiry. By ALBERT H. 
Fay. U.S. Bureau of Mines, Bulletin 95. Pg. 754. Wash- 
ington, 1920. 

This glossary constitutes a noteworthy contribution to mining 
literature. It contains about 20,000 terms; these include both technical 
and purely local terms, relating to metal-mining and coal-mining, 
quarrying, and the recovery of petroleum and natural gas, as well as 
many metallurgical terms. It includes also the names of useful and 
important common minerals and rocks, and geological terms. The 
glossary presents, in a single comprehensive volume, essentially all of 
the terms in use in the mineral industries in English-speaking countries, 
together with most of the Spanish terms in use in the United States and 
in Latin America. In addition it includes terms relating to ceramics 
and glass-making, foundry practice, railway and building construction, 
electrical installation and power-plant equipment, and chemical terms 
relating to metallurgical practice. aS Be 


Geology of the Yellow Pine Cinnabar District, Idaho. By E. S. 
Larson and E. C. Livineston. U.S. Geological Survey, 
Bulletin 715 E. Pp. 12. Washington, 1920. 

The Yellow Pine cinnabar district, in Valley County, Idaho, is 
about 70 miles from the town of Cascade. The cinnabar ore bodies 
appear to be in irregular lenses or chimneys of silicification in limestones 
which have been intruded by dikes and irregular bodies of rhyolite 


264 REVIEWS 


t 
porphyry, and the cinnabar, which is the chief sulphide in the ore, lies 
— in part in chalcedonic silica, and in part in friable marble adjoining such 
silica. Some pyrite is present with the cinnabar in most of the deposits. 
Stibnite is present in the district, in association with cinnabar, but was 
not observed in any of the workable deposits. 

Cinnabar was discovered in this district in 1902, and prospecting 
has been active since that time. In 1917 active development on a small 
scale was in progress. E. 5: B: 


Coalin 1918. Part A: Production. - By C. E. LesHer. “ Mineral 
Resources of the United States, 1918,” Part II. Washington: 
U.S. Geological Survey. 118 pp. 1920. 

A noteworthy feature of this report, as contrasted with reports for 
preceding years, is the detailed statistical study, by states and for the 
United States, of the causes of loss of time in the coal-mining industry. 
In a series of instructive diagrams for the United States and for each 
state, the percentage of time lost on account of (a) car shortage, (6) labor 
shortage and strikes, (c) mine disability, (d) lack of market are separately 
shown. These statistical studies are timely because they throw light 
upon the causes of the industrial troubles which have vexed the coal 
industry in recent years. E. 5S. B. 


Some Principles Governing the Production of Oil Wells. By Cart H. 
BEAL and J. O. Lewis. Bureau of Mines, Bull. No. 194, 
Petroleum Technology 61, 1921. 

We are constantly reminded that the resources of petroleum in the 
United States are diminishing at an alarming rate. The fact that more 
oil was produced in the United States during the month of March, 1921, 
than during any previous month suggests that efforts to curtail pro- 
duction are futile. One of the best remedies for this situation is to make 
the most of what we have. In this we have been very successful through 
the Bureau of Mines. ‘ 

Mr. Carl H. Beal and Mr. J. O. Lewis, the authors of this new publica- 
tion, were among the first to make intensive studies of underground con- 
ditions in oil fields, so that the maximum production might be obtained 
from oil wells. The results have been gratifying and have more than 
paid for the cost of investigations. 

The bulletin is a summary of all the known factors controlling the 
production of oil, and contains some of the theoretical aspects of oil 
production. It is of inestimable value to operators and “resident 
geologists.” W. 0. G: 


These microscopes have been developed 
on the basis of latest suggestions from 
petrographists. Their novel and modern 
design will prove a great aid in petro- 
graphical studies. 


Every leading petrographist knows the 
superior optical and mechanical qualities 
of Leitz microscopes and we therefore 
are fully safe if we refrain from enumer- 
ating the many features. 


The variety of models we offer afford 
a suitable equipment for every specific 
requirement. 


The complete assortment of our petro- 
graphical accessories will interest those 
who desire to add to their present 
equipment. 


Microscope “S.M.” with synchronic 
“rotation of the nicols ; 


Some of the prominent features of the new model 
microscopes are: 


_ Improved Objective Clutch Changer which guaranteés the objective to be — 
permanently centered. 


_ Anastigmatic Tube Analyzer which permits a most crisp definition. 


3 _ Special Illuminating Apparatus, whereby a wider range of application is 


afforded. 


Enlarged Microscope Tube which Siceoienioelslies eyepieces of wide diameter, 
whereby the field of view is increased 50% over other types of instruments. 


Ask for Catalog II-B (F) 


\NEW YORK, 


-60 East 10ST. 


Ree! NOT I 
SURPASSED: on- cere 
ms SUPRE 


A LEITZ-QUALITY’ 

€ 2 

tomes (oP TicaTare Pecan AND MECHANICAL 
we WORKMANSHIP _ 


It makes all Writ- 
ing Swift and 
Easy ) 


Remington 


Portable 


A little brother of the big 

Remington. Small but 

strong. Light but efficient. 
f~ Compact but complete. 


Has every feature common to the big 
machines—including automatic ribbon reverse 
and standard keyboard. 


_ Price, with case, $60 


Remington 
Typewriter Company 


_ (Incorporated) 


374 Broadway 
~. New York © 


Branches 
Everywhere 


By DOUGLAS C. RIDGLEY 


Professor of Geography in the Illinois State Normal University 


The author’s aim is to provide an authoritative study of the state, presentir 
compact form the wide range of physical influences which make up the ‘geograp hic 


environment in which men live. 


The natural features and natural resources of the state are treated in some de 
The great occupations of mankind—agriculture, mining, manufacturing, tra 
portation, and trade—are discussed with sufficient fulness to give an adequate 
idea of their development and present importance within the staté. The pop 


By CHARLES Ca COLBY 
Assistant Professor of Geography, the University. 


The fundamentalidea of the back stam 
available to the busy teacher the material on 
subject which is scattered widely through lit 
ture. The course which has been giv 
number of years at the University of ¢ 
has resulted in the collection and organizati 
the material most suitable for the purpose 

The book contains short magazine ar 
and materials from Canadian, Mexican, 
United States government. publications. — 
selection of the material has been governe 
the concept that there is a direct tela 
tween the environment and the economic 
ities of the region. The author aims to su 
the different sections of the country from 
viewpoint of industrial opportunity, 


The Oetecaie of Chicago Pr 
Chicdea 0 


tion census of 1920 for Illinois is treated ina final chapter. The book is cote 


to interpret other regions in comparison with the home state; to all atte vid a 
learn the reasons for the high rank of Illinois in many lines of human endeay 


Abounds with drawings, maps, illustrations, and colored insert maps 
xvitit+ 385 pages, 16mo, cloth; $2.50, postpaid $2.65 


CHICAGO = 


es 


NUMBER 4 


Pan | _ EDITED By 
__ THOMAS C. CHAMBERLIN AND ROLLIN D. S$ 
“if With the Active Collaboration of ' 


LER, Tigereebrate Paleontology ALBERT. Ji OHANNSEN , Petrology 
STIN, Economic Geology ; ROLLIN T. CHAMBERLIN, Dynamic Geology 


ASSOCIATE EDITORS 


JOHN C. BRANNER, Leland Stanford Junior University 
RICHARD A. F. PENROSE, Jr., Philadelphia, Pa. 
WILLIAM H. HOBBS, University of Michigan 

FRANK D. ADAMS, McGill University 

CHARLES K. LEITH, University of Wisconsin 

Ww RTH ‘DAVID, Australia WALLACE W. ATWOOD, Clark University , 
Leland Stanford Junior University WILLIAM H. EMMONS, University of Minnesota 

/ LCOTT, Smithsonian Institution ARTHUR L. DAY, Carnegie Institution 


265 — 


295 


A E. poe 303 


ENCE OF AN APUS IN THE PERMIAN OF OKLAHOMA 
RUDOLF RUEDEMANN 311 


AS ase BREW EBS Sharan ty ce hha ea Ne ee 


oe See any eS = NN Sy sae aie - - 325 


E UNIVERSITY OF CHICAGO PRESS 
CHICAGO, ILLINOIS, U. S. A. 


THE CAMBRIDGE UNIVERSITY PRESS, Lonpon 


THE MISSION BOOK COMPANY, SHANGHAI 


May-June 1922 


EDITED BY wee 
THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY 


With the Active Collaboration of 


STUART WELLER 4 ALBERT JOHANNSE ae ie 
Invertebrate Paleontology : Petrology : 


EDSON S, BASTIN ‘ ROLLIN TE; CHAMBERLIN | 
Economic Geology ; Dynamic Geolog 


The Journal of Geology is published by the University of Chicago at the University of Chicag 
5750 Ellis Avenue, Chicago, Illinois, semi-quarterly, on or about the following dates: February 1, 
May 1, June 15, August 1, September 15, November 1, December 15. {[ The subscription pri 
per year; the price of single copies is 65 cents. Orders for service of less than a half-year will 
at the single-copy rate. {[ Postage is prépaid by the publishers on all orders from the U 
’ Mexico, Cuba, Porto Rico, Panama Canal Zone, Republic of Panama, Dominican Republic, ‘E 
Bolivia, Colombia, Honduras, Nicaragua, Peru, Hawaiian Islands, Philippine Islands, Guam, 
Islands, Shanghai. {| Postage is charged extra as follows: for Canada, 25 cents on annual subse 
(total $4.25); on single copies 3 cents (total 68 cents); for all other countries in the Poste 
50 cents on annual subscriptions (total $4.50), on single copies 6 cents (total 71 cents). Patr 
requested to make all remittances payable to The University of Chicago Press, in Dea or expres 
orders or bank drafts. 

The following are authorized to quote the. prices cidieteds a 

For the British Empire: The Cambridge University Press, Fetter Lane, fation E. c 4. 
subscriptions, including postage, 275. each; single copies, including postage, 4s. 3d. each. __ 

For Japan and Korea: The. Maruzen-Kabushiki-Kaisha, 11 to 16 Nihonbashi Tori San 
Tokyo, Japan. Yearly subscriptions, including postage, Yen 9.04 each; single copies, 
postage, Yen 1.42 each. 

_ For oe The Mission Book Company, 13 North Szechuen Road, Shanghal ‘Yea 


mailed iecer outside of Shanghai, on yearly Leet 50 cents, on single nes 6 cents. a 
Claims for missing numbers should be made within the month following the regular month 
cation. The publishers expect to supply missing numbers free only when losses have been sus 
transit, and when the reserve stock will permit. 
Business correspondence should be addressed to The Gee of Chee Press, Chia 


NAL OF GEOLOGY, the University of Chicago, Chicago, Ill. 

Contributors are requested to write scientific and proper names with sane care and i in 
to follow the form shown in the pages of the Journal. - 

Because of the high cost of printing, this Journal has found it necessary to adlspet the following 
1. If articles exceeding 25 pages in length are accepted for publication in the Journal, the cost 0 
position beyond 25 pages shall be paid by the author or by the institution which he represents, at th 
rates of the University Press. ‘] 2. The cost of illustrative material, in excess of $1.00 per pas 
article, beyond the cost of type which would have occupied the same space, shall be borne by the a 
3: The cost of corrections in proof in excess of 10 per cent of the cost of first composition, 
charged to the author. {[ 4. While the.cost of printing remains as now, the Journal cannot su 
prints without charge, but reprints will be furnished authors at cost, if ordered in advance of P 


Entered as second-class matter, March 20, 1893, at the Post office at Chicago, Ill., under the Act of March . 
- Acceptance for mailing at special rate of postage provided for in Section 1103, Act of October 3, IOT7, 
July ts, hws : 


PRINTED IN THE U.S.A. ‘ 
» , 


Among articles to appear in early 
numbers of the Journal of Geology 
are the following: 


Post-glacial Lakes in the Mackenzie River Basin, Northwest 
Territories, Canada. By A. E. CAMERON. 


The Early Pre-Cambrian Formations of Northern Ontario and 
Northern Manitoba. By E. L. Bruce. 


Dinosaur Tracks in Hamilton County, Texas. By W.E.WRATHER. 


A Scale of Grade and Class Terms for Clastic Sediments. 


_ By CuHester K. WENTWORTH. 
The Pre-Cambrian of Western Patricia. By E. M. Burwasu. 


The Hot Water Supply of the Hot Springs, Arkansas. By Kirk 
BRYAN. 


Problems in Stratigraphy along the Rocky Mountain Trench. 
By FRANCIS PARKER SHEPARD. 


A 
PHYSIOORAPHIC DIA i 


orm 
UNITEDSTAT| | 


PHYSIOGRAPHIC DIAGRAM ( 
THE UNITED STATES _ 


Size 6244 inches, Scale 50 miles to one inch 


By A. K. LOBECK 


Department of Geography, University of Wisconsin 


This diagram is intended to be used asa guide in teaching the physiography 
geology of the United States. The topographic features are expressed with the 
of revealing or suggesting the underground structure, and the general charact 
the underlying rock. 


An introductory drill such as having students draw cross sections through 
Adirondack uplift, the Cincinnati or Nashville Domes, across the Appalachia 
different places, through the Black Hills, Central Texas, the Ozarks, Wisconsii 
many other regions is followed up with the use of other available material, | 
cially state geological maps and the United States Geological Survey topogri 
sheets. 


The student frequently fails to note the significance of a locality with relatic 
the physiographic province or geological region as a whole. This is overcom)y 
requiring the student to locate upon the diagram the position of the topogri! 

quadrangles under consideration. As practically every topographic map publ 
by the Survey was consulted in the preparation of the Diagram, most of the si, fi- 
cant features are recognizable in the Diagram too. Seventeen geological prove 
are discussed briefly on the map. 


S8e 


BSE 


Fe 


SF 


PRICES a 


Student edition (one color) with one-half of U.S. on each side of sheet, size 3nd 


inches. y 
SymayallS COO acarooooce $1.25 Fifty copies, each........- $1.00 © 
Ten copies, each...... I.10 One hundred copies, each.. .90 | 
Two color edition with entire map on same side of sheet, size 62 X44 inches. | 
Sincletcopyaneeneeeeer $2.25 Fifty copies, each........- $1.75 |. 
Ten copies, each...... 2.00 One hundred copies, each.. 1.50 | 
Prices ontwo color edition, mounted on cloth for permanent class room use,* 
62 X44 inches. | 
With Common Rollers at Top and Bottom............+.++++: $4.50 
On Spring Roller and Board with Dust Proof Cover...... Ae 3) 
ee Im! SteeliSprng Roller! Caseuneeneee eee ee ee oe | 
Fj oe ‘ & * 7 
oo .. A J. NYSTROM & COMPAN 
‘pessss. Publishers 2249 Calumet Ave., Chici® 


Goode’s Base Maps 
and Graphs 


Prepared by 


J. PAUL GOODE 


Professor of Geography in the 
University of Chicago 


A series of outline maps for all classes of work: in every grade from 
the common school to the university, in all branches of study, and 
for private use in applied sciences and the various fields of research. 


The desk maps come in three sizes, from 8X 103 to 21X15 inches, 
all perforated for insertion in a standard notebook cover with page 
8X103 inches. The larger sizes fold into this cover. Several wall 
outline maps are issued, 48X 36 inches. 


QOODES SERIES OF BASE MAPS AND GRAPHS. 


WEEE SQ 
WBA 


| 


Miniature of Map No. ror Hc. The homalographic projecticn, interrupted to present the continents 
The entire earth’s surface is shown without distortion of area 


The maps are made from the best drawings, printed with a good quality 
of ink on bond paper, and the presswork is of the highest order. Recent 
issues have been made by the photo-lithographic process, thus preserving 
an absolutely smooth surface of paper on which to work. These maps 
are considered by the many who use them as the best quality of base 
maps on the market. 

Complete list with prices will be sent upon request 


The University of Chicago Press 
5750 Ellis Avenue Chicago, Illinois 


Ward’s Natural Science Establishment 


ROCHESTER, NEW YORK 


Relief Maps 


This map, 29" x 46", scale 32 miles to the inch, with a vertical exaggeration of 20 to 1, costs 
only $30.00. The same map, hand colored to show the geology as given in United States 
Geological Survey professional paper No. 71, is $125.00. Write for circular G-233 describing our 
relief maps. 


Fossils 


We are better prepared than ever before to furnish index fossils and exhibition material. A few 
of our recent acquisitions follow: 


Cambrian material collected by Dr. Wolcot in British Columbia. 

Rare Ordovician material from New York and Bohemia, such as Aeglina, Proetus, Cryptolithus, 
Illaenus, Cystids, etc. 

Rare Silurian forms from Bohemia such as Ascoceras and Ophidioceras, Cromus, Acidaeeiel etc. 

Rare Devonian material from Germany such as Harpes, Cerarges, Goniatites showing sutures, 
Hexacrinus, Cupressocrinus, etc. 

Fine Solenhofen fish, crustaceans and insects, 

Comanchic material from Texas, 

Eocene leaves from Texas and Louisiana, 

Insects in amber. 


Rock Sections 


We have designed and had made a machine for the preparation of microscopic slides of rock 
sections, and believe we can fill your requirements quickly and in a satisfactory manner. Let 
us send you our special announcement circular. 


WARD’S NESSIE Ula 2 ESTABLISHMENG 


ROCHESTER - - NEW YORK 


sia 


VOLUME XXX NUMBER 4 


ilgislite 


JOURNAL OF GEOLOGY 


May-“fune 1922 


ON CONTACT PHENOMENA BETWEEN GNEISS AND 
LIMESTONE IN WESTERN MASSACHUSETTS 


PENTTI ESKOLA 
Geophysical Laboratory, Carnegie Institution of Washington 


CONTENTS 


INTRODUCTION 
BECKET GRANITE GNEISS 
LIMESTONE AND SKARN 


BANDED GNEISS NEAR THE LIMESTONE AND THE VARIATION OF ITS MINERAL 
COMPOSITION 


THE FE:MG PROPORTION IN THE VARIOUS BANDS OF THE GNEISS AND IN ITS 
DIFFERENT Maric MINERALS 


VARIATION OF MINERAL PARAGENESIS IN METAMORPHIC LIMESTONE 
ASSIMILATION OF LIMESTONE BY GRANITIC AND PEGMATITIC MAGMAS 
SUMMARY 


INTRODUCTION 


The eastern part of Berkshire County in western Massachusetts 
is underlain by a vast area of granitic gneiss, known as the Becket 
granite gneiss.* 

1B. K. Emerson, “‘Geology of Massachusetts and Rhode Island,” U.S. Geol. 
Surv., Bull. 597 (1917). 

In older papers by the same author, ‘‘Geology of Old Hampshire County, Mass.,” 
U.S. Geol. Surv., Mon. 29 (1898), and ‘‘The Geology of Eastern Berkshire County, 


Mass.,” U.S. Geol. Surv., Bull. 159 (1809), this granite gneiss is designated in part as 
“Becket conglomerate gneiss” and in part as “‘Tyringham gneiss.” 


265 


266 PENTTI ESKOLA 


This gneiss has been regarded as of pre-Cambrian age and the 
Cambrian and Ordovician rocks that bound it on the west and the 
east have been thought to lie unconformably upon it. 

On its western side, in the Housatonic Valley, the main rock 
is limestone, called Stockbridge limestone, interbedded with quartzite 
and slate, while on the eastern side there is an area of highly 
metamorphic schists. 

Within the area of the Becket gneiss there are several vertical 
layers (up to six hundred feet broad) of limestone, which is older 
than the gneiss. The most important of them, according to 
Emerson, are situated at Coles Brook in Middlefield, north of 
Becket Station in Becket, in Washington and Peru, in Hinsdale, 
and in the valleys of East Lee, Tyringham, and Otis. All the 
limestone formations within the Becket gneiss area are called 
Coles Brook limestone from the type occurrence at Coles Brook in 
the northeastern part of the mass. A striking topographical 
feature of these occurrences of limestone is that they occupy long, 
narrow valleys in the gneiss highland. 

Near the contacts with this limestone the Becket gneiss shows 
two kinds of endomorphic contact phenomena: In many cases 
the boundary type of gneiss grows more basic and is called Lee 
quartz diorite. But in other places, the boundary type of gneiss 
is rich in salic minerals and contains monoclinic pyroxene and 
titanite. It is called by Emerson “‘titanite-diopside diorite aplite,”’ 
and is supposed by him to represent a product of assimilation of 
the limestone by an igneous magma. 

During a trip to western Massachusetts in July, 1921, the 
present writer, together with Dr. N. L. Bowen, had the opportunity 
of studying outcrops and collecting samples of these contact rocks 
around Benson Pond east of Washington Station, on the railroad 
line between Washington and Becket stations, north of Becket 
Station and in the valley of Goose Pond Brook near East Lee. 

As the problem concerning the assimilation of limestone by 
igneous magmas is now being much discussed, this very instructive 
case may be of some interest to petrologists. A brief note of field 
observations and of a microscopic study of the materials collected 
is therefore given in the present paper together with a discussion 
of the results. 


GNEISS AND LIMESTONE CONTACT PHENOMENA 267 


Numerous questions as to why and how the contact rocks have 
obtained their present characters arise from such studies, but to 
many of them no definite answers are yet possible. It often seems 
wiser to direct our efforts toward finding a sound form of the 
question, according to Bacon’s advice: “Bene interrogatum quasi 
dimidium responsi.’”’ These questions, one may hope, will be 
cleared up either by future experimental investigation or by 
accumulating more evidence from natural rocks. 

The questions to be discussed in the present paper are: Why 
is limestone sometimes assimilated by magmas, and sometimes 
not? What are the factors controlling the formation of silicate 
minerals in limestone during metamorphism? What laws control 
the distribution of elements, such as Mg and Fe, which replace each 
other in solid solutions, among different mineral groups, pyroxenes, 
amphiboles, and micas, in silicate rocks? 


I wish to express my best thanks to Dr. N. L. Bowen for pleasant 
companionship during the excursions and for many discussions, 
and especially for his trouble in making the necessary grammatical 
corrections In my writing. 

This work has been carried out with pecuniary assistance from 
the Geophysical Laboratory of the Carnegie Institution of Wash- 
ington and from two funds for the advancement of scientific 
research in my native country, Finland, namely, Alfred Kordelin’s 
General Trust for the Advancement of Progress and Knowledge, 
and Herman Rosenberg’s Travelling Bursaries Trust of the Uni- 
versity of Helsingfors. 


BECKET GRANITE GNEISS 


Emerson! gives the following description of the Becket granite 
gneiss: 


It is a medium to fine grained light-colored biotite (or biotite-muscovite) 
microcline-oligoclase gneiss, with microscopic epidote uniformly blended with 
the scanty biotite, and the microcline grains commonly grouped as if made of 
the crushed fragments of larger porphyritic crystals. There are small areas 
of light-colored porphyritic granite from which the prevalent rock could have 
been produced by crushing. A micrographic texture is common. Over large 


t U.S. Geol. Surv., Bull. 597, p. 154. 


268 PENTTI ESKOLA 


areas the dark constituent is in whole or part magnetite in small octahedra. 
. .. . At its contact with the graphitic rocks the rock is commonly graphitic, 
and against the limestone it contains in many places secondary calcite grains 
and tremolite or actinolite. 


The following chemical analyses of the Becket gneiss and a 
related gneiss have been published: 


I. Gneissoid granite, Alderman’s quarry, Becket, Toscanose-lassenose. 
Analysis by G. Steiger. Specimen in Petrographic Reference Collection of 
U.S.G.S. No. 16409. 

II. Granite gneiss, Hoosac Mountain. Toscanose. Analysis by E. T. 
Allen. Specimen P.R.C. 1718. Collected by J. E. Wolff. Consists of quartz, 
microcline, albite, muscovite, biotite, magnetite, titanite, epidote, apatite, 
zircon. 

III. Composite sample of Becket gneiss from 33 localities in the Sheffield 
quadrangle. Collected by Joseph Barrell and analyzed by R. C. Wells. 


I II iil 
HOB a co mero: cee HAO |) O72 68.56 
AOS Renn toe 75 Qiu 14.97 14.53 
He Oxeetcecancn 1.06 2.61 1.41 
EO irisrdn rye ecale 0.43 2.19 QI 
Ie OG cote oro arcs ©.20 0.54 0.60 
CaOe cece I.30 1.69 2.69 
Nas Olnaeeers cass 4.55 3-92 3.58 
KORE Hee aace 4.01 SoS 3.62 
EEO a aren ©.16 ©.19 0.06 
1B HO y asogd dane 0.72 ene 0.07 
THO o ¢ 0.20 ©. 87) 0.55 
CORRE aheacios 0.88 aaa trace 
PEO i escic tear ©.07 I> Ooi 0.17 ZrO, 0.02 
MTT OV ested eastern Shae akg 0.02 ©.03)9) | 0-03 
1B¥ Oe ata steter eaten etereysta tt crm, ¢ O10) 74O4CC8 || Cue7l? ©.0% 
99.69 |100.26 "99.82 


Through the courtesy of Dr. E. S. Larsen of the United States 
Geological Survey, I had the opportunity of studying the original 
specimen of Analysis I, the Becket gneiss from Alderman’s quarry. 
It is a light gray granite, rich in muscovite. All its minerals are 
clear and unaltered. 

In the plagioclase were determined 6)=1.53450.001; i. = 
1.532+0.002; yp) =1.540+0.002,? corresponding to ADgo. 

tF. W. Clarke, U.S. Geol. Surv., Bull. 591, pp. 33-34. 


2 The composition of the plagioclases, in this study, was ascertained mainly from 
the indices of refraction determined in rock powder by the immersion method, using 


GNEISS AND. LIMESTONE CONTACT PHENOMENA 269 


The muicrocline is finely cross-hatched. The feldspars and the 
quartz form the aplitic mass in which scales of muscovite, biotite, 
and calcite are scattered. ) 

In the muscovite, 6) =1.6150.003, and in the biotite 6,= 
Yn =1.650+0.003. Unfortunately there are not yet sufficient data 
to establish the composition of the micas from the refractive 
indices. ‘The biotite probably has about 80 per cent of the iron 
compound.* 

The mode of this rock was calculated with the following result: 


ENE. a BST a ODIO LLLCe e Ser es er min net ante ae 4.0 
J DSRS. Ogee ee 2B Sirona COlClben nm tenet tater etn oe DEO 
BUORUINEC <2 oo. se hea ce 2k PRN AELCE cA ecient a: On2 
MOMETOCMME. cs Soe ccs see ness 15.0 Tas 
MBEGAL CM cai .5 s\dsc.. « sees 10.0 ee: 3 


This is a rock in which calcite occurs as a primary mineral. 
Emerson remarks (of. cit., p. 153): “Because of its proximity to 
the Coles Brook limestone it contains microscopic grains of calcite.” 

All the analyses of the Becket gneiss show an ordinary granitic 
composition with about equal amounts of potash and soda and 
very little lime. A study of thin sections of a number of specimens 
proved this to be a general rule. Some of them, however, do not 
contain any potash feldspar, but all are rich in mica. 


a monochromatic illuminator and Dr. Merwin’s dispersion diagram for the set of 
liquids (see E. Posnjak and H. E. Merwin, Jour. Amer. Chem. Soc., Oct., 1922). Inthe 
case of feldspars, where birefringence is low and dispersion sufficiently known, this 
method offers the advantage that all the indices of refraction can be determined with 
only one liquid. In the actual work it was found convenient either to determine 8 
only, or to seek grains showing the highest birefringence and determine, in three of 
four grains, a’ and y’, taking the lowest and highest values found. The composition 
was then located on the curves published by F. E. Wright, Am. Jour. Sci. (4) (1913), 
pp. 36,540. The latter is the quickest way and sufficiently accurate to give the com- 
position of the plagioclase within the limits of 2 or 3 mol. per cent. 


t The lowest indices of refraction in biotites recorded are those in yellowish brown 
biotite from Vesuvius, ap=1.5412; Yp=1I-5745, and the highest those in black biotite 
from Somma: ap=1.5795; Yp=1-638(3). The highest value of y is much lower than 
that often found during this work. The maximum value, found in a biotite from 
biotite gneiss, railroad line between Becket and Washington, was yp=1.6600.003. 
Taking this value and yp=1.574 as limits a very rough approximate estimation of 
the composition of biotites may be obtained graphically. 


270 PENTTI ESKOLA 


LIMESTONE AND SKARN*™ 


Emerson characterizes the Coles Brook limestone in the follow- 
ing words: “The Coles Brook is a coarse, highly crystalline, 
magnesian limestone, locally white and pure, generally graphitic 
and greatly changed to a mass of silicates—chondrodite, wollas- 
tonite,3 wernerite, hypersthene, pyroxene, amphibole, titanite, 
adularia, pericline, and others... .. fs 3 

In the following are given the writer’s observations on the 
limestones in western Massachusetts. 

About 3 mile east of Benson Pond, 1 mile east of Washington 
Station, a band of limestone, a few meters broad, occurs in the 
gneiss. It is rendered impure by the presence of greenish white 
mica, reddish brown chondrodite, and clear brown titanite. 

The mica is nearly, though not quite, uniaxial and has By = y= 
1.5706+0.002. It is a magnesian mica, very poor in iron, but 
probably not phlogopite. 

The chondrodite is notably pleochroic, a and 6 reddish brown, 
y pale brown. aAB=31°-a)=1.621+0.003; {8)=1.632+0.001; 
Yv = 1.655 0.003. 

This limestone is a calcite rock and contains no dolomite. 

In apparent connection with the limestone, on its continuation 
along the strike, were found many inclusions of clinopyroxene 
skarn in an aplitic variety of the gneiss which contains varying 
amounts of the same kind of pyroxene (Fig. 1) as the skarn. 

Many of the inclusions are sharply defined, coarsely crystalline 
masses of grayish green clinopyroxene, although narrow veinlets 
always protrude into them from the aplite. These veinlets always 
contain minute grains of red grossularite-andradite which is also con- 
centrated within a narrow zone around the inclusions. It is appar- 

«“Skarn.—An old Swedish mining term for the silicate gangue (amphibole, 
pyroxene, garnet, etc.) of certain iron ore and sulphide deposits of Archaean age, 
particularly those which have replaced limestone and dolomite. The term is used 
in this sense by Fennoscandian geologists, but it has been extended to cover analogous 


products of contact metamorphism in younger formations.” A. Holmes, The Nomen- 
clature of Petrology (London, 1920), p. 211. 


2B. K. Emerson, U.S. Geol. Surv., Bull. 507, p. 21. 


3 This statement about wollastonite does not occur in the other more detailed 
reports and may be erroneous. 


GNEISS AND LIMESTONE CONTACT PHENOMENA 277i 


ently heterogeneous, its index of refraction (n,) varying from 1.775 
0.005 to 1.801+0.005. This means a variation in composition from 
25 to 43 wt. per cent andradite in the mixcrystals, supposing them 
to contain only andradite and grossularite.’ 

The clinopyroxene has 8) =1.696 0.001 and the angle ¢A y= 
43°. Hence its composition should be Dis,He;:.? 


Fic. 1.—Inclusions of clinopyroxene skarn in aplitic gneiss. E. of Benson Pond, 
Washington, Massachusetts. Eight-ninths natural size. 


_Together with the garnet in the veinlets there is quartz, micro- 
cline .and plagioclase. The latter, having o’)=1.532 and y'>= 
1.540, has the composition about Ab,;. 


™W. E. Ford, Amer. Jour. Sci. (4), Vol. XL (1915), pp. 33-49. 


2 The determinations of pyroxenes of the diopside-hedenbergite series by means 
of the indices of refraction were made with a diagram (Fig. 2) based on the optical 
data given by Wiilfing (H. Rosenbusch and E. A. Wiilfing, Mikroskopische Physio- 
grape, I, 2, p. 203) and on the analyses of the same pyroxenes published by G. 
Flink (Zs. Kr., Vol. IL (1895), p. 585). Although the analyses are antiquated, these 
data no doubt give a correct idea of the relations between composition and optical 
properties of clinopyroxenes containing little or no sesquioxydes. Drs. H. E. Merwin 
and H. S. Washington are at present carrying on an investigation of all the pyroxenes 
_ and they no doubt will give a better diagram of this series. As their work is not far 
advanced and it may be some years before their results are published, we give here 
the diagram based on Flink’s and Wiilfing’s data. 


222 PENTTI ESKOLA 


Small amounts of deep green hornblende (8, =1.677-0.002) 
are associated with the pyroxene. 

Considerable quantities of epidote occur near the margins of 
the skarn inclusions. Its refringence varies within single grains, 
being lowest in the centers. 

On the hillside just north of Becket Station there is exposed a 
vertical layer of limestone, surrounded by clinopyroxene gneiss. 


LY, 
ic 


nS 


100 mol.% 80 
DIoPpsiDE 100 
HEDEN BERGITE 


Fic. 2.—Variation of the indices of refraction in the diopside-hedenbergite series. 
The small amounts of ferric oxide present have been calculated as ferrous oxide. 
After E. A. Wiilfing and G. Flink. The indices of pure diopside were determined by 
H. E. Merwin. 

A specimen of the limestone was found to contain, besides calcite, 
much quartz and clinopyroxene of the composition Dis,He;;, 
having 6B, = 1.698+ 0.002. 

One of those narrow valleys within the Becket gneiss area that 
are underlain by the Coles Brook limestone is Goose Pond Brook 
Valley. It joins the broader East Lee Valley near East Lee. 
Limestone occurs in small outcrops in the bed of the brook, and 
little can be said about its mode of occurrence, but it seems to be in 
immediate contact with the clinopyroxene gneiss that is exposed 
close by. 


GNEISS AND LIMESTONE CONTACT PHENOMENA 28 


Among the specimens collected some were very rich in calcite 
with only sparing amounts of dolomite crystals, the latter being 
easily recognizable from their higher indices of refraction (w, = 
1.683 —1.684; the calcite always was found to have w, = 1.660). 

Other specimens collected contain pale brown biotite, having 
Bp = Yn = 1.599 — 1.600, and quariz, as rounded grains. 

A specimen from the wider East Lee Valley, where the Coles 
Brook limestone occurs as a larger mass, represents a dolomite-rock 
containing rounded grains of quartz and microcline, and minute 
crystals of pyrate and scales of brownish, nearly uniaxial mica with 
Bp = Yn = 1.599+ 0.002. | 

It may be mentioned that these characters are the same as 
those in the most common types of the Stockbridge limestone in 
the Housatonic Valley. Analyses' of the Stockbridge indicate 
that there are all variations between dolomite-rocks and calcite- 
rocks represented. We collected specimens of this limestone from 
Glendale quarry. It is a dolomite-rock, with much fine scaly 
brownish mica (8,;=Yn=1.5840.002) and brown tourmaline 
(wp = 1.639 0.001). 


BANDED GNEISS NEAR THE LIMESTONE AND THE VARIATION 
OF ITS MINERAL COMPOSITION 


The gneiss, near the vertical, or almost vertical, layers of 
limestone, always shows a banded structure parallel to the strike 
of the limestone. At the immediate contacts it is clinopyroxene 
gneiss. Farther away there are various bands, some containing 
hornblende, and others with almandite, biotite or magnetite as 
the principal mafic minerals, and among them are bands of the 
clinopyroxene gneiss also. 

The individual bands may be homogeneous or banded in detail 
(Fig. 3), due to the unequal distribution of the minerals. The 
breadth of homogeneous bands varies from less than one meter 
toa hundred meters. Among the pyroxene-bearing bands, however, 
none was found thicker than some ten meters. 

A common feature of all the varieties of gneiss, of widely 
different mineral composition, is the presence of very albitic 
plagioclase (mostly about 90 per cent Ab) and epidote whosé 


*B. K. Emerson, U.S. Geol. Surz., Bull. 159 (18090), pp. 87 and 99. 


274 PENTTI ESKOLA 


amount increases with the quantity of the femic compounds, a fact 
indicating that it is a substitute for anorthite. The epidote occurs 
as clear, individual crystals, often intergrown with quartz in the 
myrmekite fashion. 

Another feature significant in connection with the conditions 
of formation of these banded gneisses is the absence of perthitic 
threads of plagioclase in the microcline, which is clear and finely 
cross-hatched. The texture, on the whole, is aplitic, and all the 
main minerals are equally xenomorphic. 


Fic. 3.—Banded clinopyroxene gneiss. W. of Benson Pond, Washington, 


Massachusetts. Eight-ninths natural size. 


Emerson (op. cit., p. 153) published an analysis of clinopyroxene 
gneiss, called by him “‘titanite-pyroxene diorite aplite,” from 
“east of C. Conwell’s place, South Peru, Mass.’ The analysis, 
made by W. T. Schaller, is quoted below (p. 275). 

I have found these clinopyroxene-bearing rocks to be exceedingly 
variable in composition, and the analysis has happened to be made 
on a rather exceptional type. Most specimens collected by us 
from the same tract and, also, one collected by Professor Emerson 
from the Peru line, ‘‘the same ledge as the rock analyzed” (U.S. 


GNEISS AND LIMESTONE CONTACT PHENOMENA 275 


Geol. Surv., P.R.C. 1715), contain considerable amounts of quartz 
and microcline. They therefore must have a higher percentage of 
silica and potash. Assuming the plagioclase to contain 90 mol. 
per cent Ab and a notable quantity of epidote to be present, the 
mode given in the table was calculated from the rock analysis 
by the writer. The pyroxene was computed as a dliopside- 
hedenbergite with 1o per cent (Mg, Fe)SiO, and 4 per cent Al,O,. 

The clinopyroxene, in the specimen from the Peru line, is pale 
green without any notable pleochroism. It was found to have 
OQ =1.695+0.002, By = 1.706+0.002; Yp=1.724+0.002, and should 


Per Cent Mol. Mode 
SiO, GOA, “At Cror? NOMS. Caw aise cio enon a oe 60.26 
Al.O; 16.26 GO) AAG TAEMES Sogn nooo saboeococ Fo stt 
eon ORO Wen cehs Rotashiteldsparaea sass. 2.78 
FeO 3.52 49 CASIO‘ .5 55000 8.82 
MgO Tre 7/8 44 Expense MgsiO; Renee: 4.40 AOA 
CaO 7.86 140 KeSiO eraser 6.47 
Na.O Wot Il5 (ANE @ eee te dae Seal 

Pee PICO LC Hack n cran ceeds - 1.93 

HzO Set = : (ONO GL Ars tek akeree untae nae QT 
H,O+ OPI She waar MMitanite ee os, pawete sets tec. 3.14 
H,0— One OM teks SEPA ALICE ein ea. Was einen nr ay 1.34 
TiO, 23 TROP AUC OMEN onc sexes ae Re ae 0.09 
P20; 0.58 A PNAS. 5 5 oo ace La Meer aon ae ©.10 
MnO 0.06 I 
ZrO, OO T Sas ex 99-80 
FeS, OMlObe = Aee 

100.27 


therefore contain 60 mol. per cent diopside and 4o per cent heden- 
bergite, or Dig,.He,, while calculation from the analysis gave 
Di,sHe,.. Thus the specimen analyzed contained a pyroxene 
richer iniron. As we shall see, the composition of the clinopyroxene 
in these gneisses varies markedly from place to place. 

The epidote is highly birefracting and has 6,=1.750+0.003, 
nearly the same value as that found in the epidote from Knappen- 
vand, Sulzbachtal, which contains 33 per cent of the ferric epidote 
(16 per cent Fe,O,) and has 8, =1.7540. 

tM. Goldschlag, Tscherm. Min. Petr. Mitt., Vol. XX XVIII (1917), p. 23. The 
data given in this study, the most careful one ever carried out on the epidote group, are 
not yet consistent enough to allow the determination of the amounts of the aluminous 


and the ferric compounds in epidotes more accurately than with a possible error 
of +5 per cent. 


276 PENTTI ESKOLA 


The prominent characters of the mineral composition of the 
various kind of gneiss are stated in the following. The mineral 
constituents of each rock are named in the approximate order of 
abundance. ‘The refractive indices were determined, if not stated 
otherwise, with a maximum error of +0.003 in biotite and horn- 
blende and epidote, and +o.002 in plagioclase and pyroxene. 
The composition of the minerals is given in terms of mol. per cent, 
stating: in plagioclases the percentage of anorthite (An), in 
clinopyroxenes (diopside-hedenbergite) the percentage of heden- 
bergite (He), in hornblende the total Fe as a percentage of Fe+Mg,* 
and in the biotites and epidotes (very roughly) the percentages 
of their iron compounds. 

Aplitic clinopyroxene gneiss around the inclusions of skarn. 
Mainly microcline, subordinate plagioclase (Ans), quartz, clino- 
pyroxene (8p = 1.705, Y> =1.727, He,), epidote, variable, sometimes 
very poor in Fe,’ calcite. 

Banded clinopyroxene gneiss (Fig. 3) in biotite gneiss, W. of 
Benson Pond. Microcline, quartz, plagioclase (a’><1.536; Yo= 
1.544; An,;), clinopyroxene (Hes,),3 titanite, epidote. 


tCalculated from the diagrams given by W. E. Ford, Amer. Jour. Sczi., 
Vol. XXXVII (1914), p. 185. 


. . . + . 
27 measured in one case for different wave-lengths, with a maximum error 


=+0.001: 
a B Y 
fn iiaeeret Te'¥O3", ila Sretaveee 1.707 
DD) Alsaveerersrelne 1.698 I.700 I.702 
(Oe a ataees ey T.3003)> illeteltycteverereysie I .607 


These values are lower than any given in the literature (cf. M. Goldschlag, Joc. cit.). 
This mineral is, however, monoclinic, and not a zoisite. 2V=go° appr. 


3 In it were determined, with a maximum error +0.0o0T: 


a B y, 
Mite perv tesen I.724 1.7305 1.748 
De neta: 1.720 1.726 I.743 
OB Mee omer RG yf) 1.721 1.738 


The birefringence suggested that this clinopyroxene is richer in hedenbergite 
than any other found in this tract. Its refractive indices for various wave-lengths 
are in fair agreement with those of Di,:Hes, as shown on the diagram. Its extinction 


GNEISS AND LIMESTONE CONTACT PHENOMENA 277 


“Pyroxene-titanite aplite’ from the Peru line. Plagioclase 
fe 5305) aa —1.531; 7 >—1.542; Ani), microcline, quartz, 
hornblende (8,=1.676; Fe,,), clinopyroxene (@,=1.695; Bpo= 
mone )y,—1.724; He,), titanite, biotite (6,—7,—1.047; 80 
per cent Fe-comp.), epidote (8,=1.750; 33 per cent Fe-comp.), 
apatite. 

Intrusive layer in limestone N. of Becket Station. Quartz, 
plagioclase (An,;), microcline, clinopyroxene (8,=1.700; He,;), 
hornblende (68,=1.680; Fe,;), grossularite-andradite (n, =1.80; 
4o wt. per cent andradite), calcite, epidote. 

Band in the gneiss, railroad cut W. of Becket Station. Quartz, 
microcline, clinopyroxene (colorless; a)=1.685; (6)=1.690; 
Yo=1.714; yAc=41°; He.;), actinolite (pale green; 6, =1.638; 
Fe,,), graphite, calcite, titanite. 

Clinopyroxene gneiss near limestone in the valley of Goose 
Pond Brook. Clinopyroxene (colorless, 8; =1.681+0.001; He,,), 
amphibole (8, = 1.6330.002; Fe,,), quartz, microcline, plagioclase 
(6p =1.534; An,), epidote (8,=1.730; 18 per cent Fe-comp. or 
9 wt. per cent Fe,O,). Occasional scapolité (w)=1.5720.001; 
€) =1.547+0.001; about Me,,Ma,,) and black tourmaline. 

Dark band in the gneiss South of Benson Pond. Clinopyroxene 
fe9—1.005; He), hornblende (a,=1.672; 8,=1.685; y=1.605; 
Fe,,), plagioclase (zoned, An,, to An,), quartz, microcline, titanite, 
biotite (6, = 7p =1.635; about 65 per cent Fe-comp.). 

Dark band N. of Benson Pond. Plagioclase (zoned, An.; to 
Any), hornblende (8,=1.660; Fe,s), biotite (@;5=Yp 1.642; 75 
per cent Fe-comp.), quartz, epidote, titanite, apatite. 

Dark band in the gneiss N. of Becket Station. Biotite (8,=7.= 
1.020; 50 per cent Fe-comp.), hornblende (8,=1.650; Fez), 
Mieiaclase (85—1.520; a5—1.532; 7 ,=—1.541; An,,), microcline, 
quartz, epidote. The quantity of epidote is considerable. 

“Lee quartz diorite”’ from west end of Goose Pond. Plagioclase 
(zoned, An,. to An»), hornblende (@=1.665 appr.; Fe,), cum- 


angle ~/Ac however, does not accord with the series, being 58° instead of 46° 20’. 
Probably this clinopyroxene contains a certain amount of aegirite, although its very 
- faint pleochroism (a=pure green, y=brownish) indicates that the amount may be 
small.—In another specimen was found a clinopyroxene He¢.(8p=1.721). 


278 PENTTI ESKOLA 


mingtonite, forming the inner parts of the aggregates of amphibole 
(positive, 2V large, 8) =1.650; about 4o mol. per cent FeSiO,), 
- quartz, ilmenite, apatite, zircon. 

Almandite amphibolite, band in gneiss, roadside W. of Benson 
Pond. Plagioclase (a’,>=1.540; y'p>=1.548; An,;), almandite 
(n, =1.806+ 0.001; probable composition" 70 mol. per cent alman- 
dite, 15 mol. per cent pyrope and 15 mol. per cent grossularite), 
hornblende (8,=1.665; Fe) microcline, biotite (@=y=1.648; 
about 80 per cent Fe-compound), ilmenite, apatite. 

Biotite gneiss, railroad line between Washington and Becket 
stations. Quartz, plagioclase (An,) microcline, biotite (6,=Yp= 
1.660, chiefly iron compound), hornblende (a,=1.697; By=1.708; 
Yn =1.709;? 2Va small Fe,,), magnetite, epidote (a) =1.741+ 0.002; 
Bp =1.760+0.005; 2V=75° appr.; about 35 mol. per cent iron 
comp.). 


THE MG:FE PROPORTION IN THE VARIOUS BANDS OF THE GNEISS 
AND IN ITS DIFFERENT MAFIC MINERALS 

A comparison of the Fe:Mg proportions, as determined from 
the refractive indices of the pyroxenes, amphiboles, and biotites 
in the various kinds of gneiss, shows at once that they are controlled 
by definite laws. For the sake of convenience, the values of the 
atomic portions of Fe as a percentage of Fe+M¢g (briefly called 
the Fe-quotient) are tabulated (p. 279). 

In this table, the rocks are arranged in the order of decreasing 
Fe-quotient in their amphiboles and pyroxenes. Now, comparing 
this order with the statements of the approximate quantitative 
proportions of the minerals given in the preceding chapter, we find 
at once that it is, also, at least nearly the order of increasing basicity 
or decreasing acidity, excepting, however, the two last-named 
examples of clinopyroxene gneiss. Their case will be discussed 
below. 


tP. Eskola, ““On the Eclogites of Norway,” Videnskapsselsk. Skri. Mat.-naturv. 
Ki. I, No. 8 (1922), p. 9. 

2The indices match the highest values represented in the amphiboles studied 
by Ford (Amer. Jour. Sci., Vol. XXXVII [1914], p. 181), i.e., those of the hornblende 
from Cornwall, Orange County, New York, which has a mean index 1.71 and con- 
tains 23.35 per cent FeO and 7.41 per cent Fe,O, against 1.90 per cent MgO and 
12.10 per cent Al,O;. 


GNEISS AND LIMESTONE CONTACT PHENOMENA 279 


In other words: the larger the quantity of the mafic minerals, 
the smaller is their Fe-quotient. In the most acid types of gneiss, 
rich in quartz, we find biotite and hornblende, and often clino- 
pyroxene also, extremely rich in iron. In the mafic minerals of 
the “dark bands” the amount of iron in proportion to magnesia 
is much less. 

At the same time as the Fe-quotient decreases, the amount 
of anorthite in the plagioclases increases. This relation is somewhat 
obscured by the epidote, a substitute for the anorthite. 


too FE:(FE+Mce) In 
Rocks ' a OTHER MINERALS 
eae bienties BOWES 

Biotite gneiss, railroad at Becket Station.......|...... GO OS CA ulseens eee rss ese 
Clinopyroxene gneiss, W. of Benson Pond...... Colt eS aealten pone letter mnt a Lat 
Clinopyroxene gneiss, ‘“‘the Peruline”......... 40 51 SO ere hes ye 
Clinopyroxene gneiss, Becket Station.......... 35 Boe sane Grossularite- 
andradite, 40 
per cent andr. 
Dark band in gneiss, S.of Benson Pond........| 31 50 OG aIBEC: oe encase 
Almandite amphibolite, W. of Benson Pond.....|...... 40 80 | Garnet with 
70 per cent 

almandite 
Dark band in gneiss, N. of Benson Pond........|...... 36 IS. | la ieee ee oleae 
Dark band in gneiss, Becket Station...........].. bowel 20) SOMEA ees pe nue! 

Clinopyroxene gneiss, railroad between Becket 

OGM ASMINGLOM «35.0 esis Sans seo heave siede 25 SiG) Wh | eetantenl Mu cheater 
Clinopyroxene gneiss, Goose Pond Brook....... II B/E leg cas ae ell a eater Re 


The percentage of the iron compound in the epidotes does not 
show quite regular relations with that in the mafic minerals. 
This is probably connected with the later origin of the epidote. 

This correspondence between the Fe: Mg ratio and the relative 
proportion of salic and femic minerals is a special case of what is 
inherent in any cognate series of igneous rocks which have been 
derived from a common magma by the processes known as mag- 
matic differentiation. Its occurrence in the banded gneiss proves 
that the band structure is in itself a product of differentiation. 
In other words, the ‘“‘dark bands” are those in which the minerals 
containing solid solutions are richest in the highest melting com- 
pounds or, more exactly, compounds least soluble in the residual 
magmas. ‘The mafic minerals of these most “basic’? members of 


280 PENTTI ESKOLA 


’ 


the series are pyroxene and amphibole. With increasing acidity 
biotite and muscovite appear successively. | 

The occurrence and composition of clinopyroxene does not 
always follow this rule, and for a very obvious reason: its occurrence 
is due to the assimilation of limestone, which is dolomitic, and low 
in iron. Therefore, when it reacts with the highly siliceous gneiss 
magma, there is formed pyroxene and, also, hornblende and biotite, 
low in iron, provided differentiation by crystallization has not 
changed the composition of the magma after assimilation. In 
many cases a differentiation happened, however, and therefore the 
mafic minerals of very acid clinopyroxene gneisses not immediately 
connected with limestone usually show a high Fe-quotient. 

Another interesting question is this: How are magnesia and the 
iron oxides distributed between the different ferromagnesian 
minerals ? 

As appears from the table on page 279, there are definite 
relations between the pyroxenes, amphiboles, and biotites, in this 
respect. Biotite is regularly the richest in iron and clinopyroxene 
the poorest. The figures do not show any constant relations 
between the Fe-quotients, however. This is only what may be 
expected. From theoretical considerations the distribution of 
elements in different isomorphous series in a rock is evidently a very 
complicated process, depending upon the composition of the solu- 
tions from which the minerals crystallized and, also, on the tempera- 
ture and pressure conditions during crystallization. In rocks of 
identical bulk composition these relations may be expected to vary 
with the physical conditions. It may be hoped that, when once 
they are better understood, we shall be able to draw from them 
certain conclusions as to the conditions under which the rocks 
originated. 

The present writer has formerly studied the almandite-pyrope 
garnets of different rocks from this standpoint and has arrived at 
the conclusion that, in rocks of similar composition, those formed 
under highest pressures have garnets richest in the magnesia 
compound.’ 


xP. Eskola, ““The Mineral Facies of Rocks,” Norsk geologisk Tidskrift, Vol. VI 
(1920), p. 172. 


GNEISS AND LIMESTONE CONTACT PHENOMENA 281 


These brief statements are given here in order to call the 
petrologist’s attention to these relations, formerly very little studied. 


VARIATION OF MINERAL PARAGENESIS IN METAMORPHIC LIMESTONE 


Sedimentary limestones generally consist of carbonates of 
calcium and magnesium, and of silica, in the form of quartz. Very 
often they also bear clayey materials, containing alumina, iron 
oxides, potash, etc. When such material is brought from surface 
conditions down to deeper levels where a higher temperature 
prevails there will occur reactions of the general type: 


RCO-ESiO.— RSI. CO” 


The temperature limit above which the right-hand side of the 
foregoing equation represents the stable combination, or the 
transformation point of the chemical system, is raised with pressure. 
For the reaction CaCO,+Si0,=CaSiO,+CO., V. M. Goldschmidt 
calculated the approximate equilibrium curve on the basis of 
Nernst’s affinity theorem.’ According to this curve, the equilib- 
rium temperature increases rapidly with the pressure, being 850° 
at 300 atm. ‘Thereafter the rise of the equilibrium temperature 
with pressure should be nearly linear and so slow that as much as 
15,000 atm. would be needed at gs50°. Considering rocks of 
neighboring occurrences which have originated under similar 
pressures but at different temperatures, as is often the case at 
the contacts of igneous masses, the limestone indicates what parts 
of it have been heated above the reaction point. If the pressure 
be known, the temperature may be estimated in degrees. In the 
case of rocks whose metamorphism has taken place under pressures 
of more than 3,000 atm., or at depths of more than 1o kilometers, 
the equilibrium temperature is but slightly variable with pressure, 
and the mineral composition of silica-bearing limestones is mainly 
an indicator of temperature, or the limestone may be used as a 
geologic thermometer. 

This curve is not claimed to be more than approximate. Experi- 
mental investigation may give considerably different results. 


tV. M. Goldschmidt, “‘Die Gesetze der Gesteinsmetamorphose,” Vid. selsk. Skr. 
Mat.-naturv. Kl., No. 22 (1912). 


282 PENTTI ESKOLA 


Certain facts known at present would seem to indicate that the 
curve may lie at somewhat lower temperatures. Its general 
character, however, agrees well with petrological experience and 
is not likely to be subject to any great changes. 

From the standpoint of the phase-rule, a mixture of calcium 
carbonate and silica is a three-component system and therefore 
can have a maximum of three phases in coexistence under variable 
pressure and temperature conditions, that is, in a divariant system. 
Below the transformation curve the possible three-phase combina- 
tions are either silica, calcite, and carbon dioxide or silica, calcite, 
and wollastonite. Above the curve we may have either wollasto- 
nite, carbon dioxide, and calcite or wollastonite, carbon dioxide, 
and quartz (or any other form of silica). 

Thus wollastonite may be stable and even be formed at tempera- 
tures below the transformation curve, but only if silica meets 
lime in other form than carbonate and no free carbon dioxide is 
present. The occurrence of wollastonite, therefore, is not in itself 
a sufficient evidence that the rock had been heated above the 
transformation temperature. On the other hand the occurrence of 
both quartz and calcite proves conclusively that it never has. 

The temperature of transformation may be considerably 
depressed if the partial pressure of carbon dioxide is kept low as a 
result of its continual removal, e.g., by circulating solutions. This 
case may occur in the formation of metasomatic ore deposits, and 
in fissure veins. The conditions are here complicated also because 
the addition of substances and formation of minerals may have 
continued at different temperatures. As a matter of fact, observa- 
tion usually indicates a definite sequence of formation of minerals 
in such deposits. , 

While, then, many complications are to be expected where 
circulating solutions have been prominent agents, we may still use 
limestone as a geologic thermometer with considerable confidence 
in those cases in which quartz or the silicates occur evenly dis- 
tributed as true rock constituents. 

As we are concerned, in natural rocks, with calcium and magne- 
sium carbonates, the resulting silicates will be those of calcium and 
magnesium, or double compounds of both. It is to be expected 


GNEISS AND LIMESTONE CONTACT PHENOMENA 283 


that magnesium-bearing carbonates will react with the silica at 
lower temperatures than calcium carbonate. Each silicate to be 
formed may have its own equilibrium curve. 

This assumption has been very strongly supported by petro- 
logical investigations which I have carried out in the pre-Cambrian 
limestone of Finland.* In the southeastern part of the pre- 
Cambrian area of Fennoscandia there are extensive formations all 
of whose characters point toward metamorphism at low tempera- 
tures. In the limestones of this area, quartz occurs together 
with dolomite, and fine scaly micaceous minerals (biotites rich in 
magnesia) are usually the only silicates. Sometimes epidote 
occurs, a fact indicating that this hydrated calcium aluminium 
silicate also belongs to the low temperature minerals. 

Northwest of this low temperature area there is a broad zone 
within which the limestones contain, besides mica, amphiboles of 
the tremolite-actinolite series, but no other silicates. Going 
farther northwest diopside-hedenbergite is added to the list of 
minerals of the metamorphic limestones, together with many others, 
such as scapolite, vesuvianite, and grossularite-andradite. Finally 
we find, as local developments near the contacts of igneous masses, 
but never in regional distribution, wollastonite limestone, and 
in this most of the other silicates may occur also. This mode of 
occurrence clearly indicates that wollastonite, among the lime- 
bearing silicates, requires the highest temperature to form. 

We may thus discriminate the following four paragenetic types 
of limestones: 

-z. Quartz limestone, in which quartz is coexistent with the 
dolomite. 

2. Tremolite limestone. Tremolite is usually present; quartz 
occurs in calcite-rock, but is not coexistent with dolomite. 

3. Diopside limestone. Diopside is usually present, quartz 
occurs together with calcite but not with dolomite. 

4. Wollastonite limestone. Wollastonite is present, provided 
the rock contains silica in excess of the amount needed to form the 


t Pentti Eskola, Victor Hackman, Aarne Laitakari ja W. W. Wilkman, Suomen 


_ kalkkikivi. With an English summary by P. E., “Limestones in Finland,” Suomen 


Geologinen Toimisto, Geoteknisié Tiedonantoja, No. 21 (1919). 


284 PENTTI ESKOLA 


magnesium-bearing silicates. Quartz and calcite do not occur in 
contact with each other. Instead either of the combinations 
calcite-wollastonite or wollastonite-quartz occurs. 

Actual rocks have commonly been found in fair accordance 
with these rules of association. As was pointed out above, wol- 
lastonite is in itself stable below the transformation temperature of 
calcite-silica, if there is no free carbon dioxide present. In the 
same way diopside and tremolite are stable below those tempera- 
tures where they may be formed from the carbonates and silica, 
again provided that there is no carbon dioxide present. Now 
there actually is no free carbon dioxide after the formation of the 
silicates, as it is carried away either as gas or in solutions. ‘There- 
fore, although the reactions are reversible, they are not reversed 
during the period of gradual cooling when the rocks are brought 
up toward the earth’s surface through the process of denudation. 
This fact, of course, adds very much to the usefulness of meta- 
morphic limestone as a geologic thermometer. We always read 
the highest temperature to which it has ever been exposed. It is 
a maximum thermometer. 

The formation of different minerals under different conditions 
in limestones presents a special case of the more general rules of 
adjustment of mineral composition in response to the conditions 
existing. The writer has proposed the term mineral facies* to 
designate a group of rocks which have originated under conditions 
so similar that a definite chemical composition has resulted in the 
same set of minerals. Applying this principle to various kinds of 


rocks, we arrive at a natural rock classification in which main ~ 


divisions are the groups called mineral facies. 
The four paragenetic types of limestone may be paralleled with 
the facies system of the silicate rocks in the following way: — 
Under conditions similar to those that give rise to quartz 
limestones, rocks having the bulk composition of gabbros have 
been metamorphosed into chlorite-epidote-albite rocks. This min- 
eral facies has been called the greenschist facies. 


«P, Eskola, ‘The Mineral Facies of Rocks,” Norsk geologisk tidskrift, Vol. V1 
(1920), pp- 143-94. 


GNEISS AND LIMESTONE CONTACT PHENOMENA 285 


Tremolite limestones correspond with transitional forms 
between the greenschist facies and another facies called the amp/ib- 
olite facies, because a rock of gabbroid composition whose minerals 
have formed under the conditions of this facies, appears as a 
plagioclase-hornblende rock, an amphibolite. 

The diopside limestones, so far as known at present, also 
belong to the amphibolite facies, and even some of the wollastonite 
limestones belong here, but the last named cover a wide range of 
temperature and pressure conditions corresponding to several kinds 
of mineral development in the silicate rocks, and a gabbroid 
material may give any one of the mineral combinations: plagioclase 
and clinoenstatite-diopside solid solutions (provisionally referred to 
as the sanidinite facies) or plagioclase, diopside, and hypersthene 
(hornfels facies), or diopside-jadeite solid solutions and pyrope- 
almandite solid solutions (eclogite facies). 

Metamorphic limestone as a geologic thermometer can be 
calibrated against the transformation points of the silica minerals. 
The fact that quartz, and not tridymite, occurs even with wollasto- 
nite, would seem to indicate that the wollastonite-curve passes 
below the transformation point quartz-tridymite. This is, accord- 
ing to Fenner,’ 870°, but it must rise somewhat with pressure. 
As to the transformation point a-$-quartz at 575°, it has been 
established that crystals of quartz in a diopside limestone from 
Parainen in southern Finland, have been originally a-quartz,? 
while the quartz-dolomite rocks of eastern Finland contain clear 
crystals of quartz which is apparently primary 6-quartz. More 
observations on this subject are desirable. 

‘It is to be hoped that the temperature-pressure curves for the 
different silicate-carbonate equilibria may be determined exactly by 
experiment, in the near future. ; 

In the contact zones of the Becket gneiss in western Mas- 
sachusetts three of these paragenetic types of limestone occur, 
namely: quartz limestone, tremolite limestone, and diopside lime- 
stone. The temperature was probably nowhere high enough to 


tC. N. Fenner, Amer. Jour. Sci., Vol. XXXVI (1913), p- 331- 
2 A. Laitakari, Bull. Com. géol. Finl., 54 (1920), p. 40. 


286 PENTTI ESKOLA 


allow wollastonite to be formed,’ and quartz is generally found in 
contact with calcite. 

Quartz limestone is the most widely distributed type in western 
Massachusetts. Most of the Stockbridge limestone belongs to it 
and, also, much of the Coles Brook limestone. Such were the 
specimens from Goose Pond Creek and East Lee described above 
(p. 272). They consist of dolomite and quartz, sometimes with 
microcline. 

During the excursions we did not happen to meet with any 
typical tremolite limestone, but it is apparent from Emerson’s 
descriptions that it is a common type, in the Stockbridge as well 
as in the Coles Brook.? 

Tremolite limestone is noted ‘‘south of Becket’? (Emerson, 
U.S. Geol. Surv., Mon. 29, p. 27), and from Lee (U.S. Geol. Surv., 
Bull. 159, p. 84). Tremolite but no pyroxene in limestone is 
mentioned in the ‘‘ Mineral Lexicon of Eastern Berkshire Co.” 
(last-cited volume) from Windsor, Monterey, and the Cobble in 
Tyringham. 

Through the courtesy of Mr. E. V. Shannon the writer was 
able to study specimens from the collections of the United States 
National Museum of tremolite limestone (Stockbridge) from Lee, 
Massachusetts, and from Canaan, Connecticut, near the Massa- 
chusetts line. The rock from both localities was found to consist 


«Cf. footnote 1 on p. 285. 


2Tt might be objected that it is not justifiable to regard the Stockbridge and the 
Coles Brook limestones as equivalents in their metamorphic development. For, if 
the former is Cambrian, while the latter, with the Becket gneiss intrusive in it, is 
pre-Cambrian, the Stockbridge cannot have been influenced by the heat of this intru- 
sion which undoubtedly has changed the Coles Brook limestone. Now the plain 
fact is that the same three types of mineral paragenesis occur in both of them. The 
Coles Brook limestone, when belonging to the quartz type, is exactly like the most 
common type of Stockbridge limestone. The occurrence of tourmaline in the latter, 
recorded above, points to the vicinity of magmas. Emerson knew this, and he assumed 
that there may be post-Cambrian igneous masses not quite exposed on the present 
land surface. If this is true, then the earlier and the later intrusions have taken place 
during very similar physical conditions. It seems worth while, in future investigation 
of the geology of western Massachusetts, to consider the possibility that the Becket 
gneiss is of post-Cambrian origin, and the Coles Brook limestone only the margin of 
the Stockbridge. In fact our observations at the eastern as well as at the western 
contact of the Becket batholith seemed to indicate that this is the case. 


_-* 


GNEISS AND LIMESTONE CONTACT PHENOMENA 287 


of dolomite, calcite, and tremolite, with some brown mica 
(Bp = Yn = 1.582 0.003), but no quartz. It thus proves to represent 
a true equilibrium under those temperature-pressure conditions 
which give rise to tremolite limestone. 

Diopside limestone occurs so commonly that it is not necessary 
to name more than a few localities, as examples. Emerson! 
describes pyroxene- and actinolite-bearing limestone from Hinsdale 
and East Lee, both being Coles Brook limestone. His “‘ Mineral 
Lexicon of Eastern Berkshire Co.’’ (of. cit.) mentions pyroxene 
from the Stockbridge limestone from New Marlboro and several 
places in Tyringham. 

The writer studied diopside limestone from Becket Station, 
a combination of calcite, quartz and diopside. It is a rock of the 
diopside limestone type. All the dolomite has been exhausted in 
the formation of diopside, while some quartz has been left. 

Theoretical reasoning leads to the conclusion that the quartz- 
carbonate rocks, when they are metamorphosed under medium 
temperature conditions, corresponding to the tremolite and diopside 
limestone types, are likely to become poorer in dolomite. This 
must finally cause a dedolomitization of the limestone. 

The silica need not occur within the limestones themselves, 
but may react at the boundaries. In small bodies of limestone 
bounded by highly siliceous rocks, and especially when submerged 
in magmas of such acid rocks, this may be expected to give rise to 
a perfect dedolomitization. So far as the writer’s experience goes, 
small bodies of limestone inclosed in granites or gneisses really 
consist of calcite-rock, while magnesium-bearing silicates occur 
at their contacts. ‘This rule was confirmed in the Massachusetts 
localities studied, near Benson Pond and near Becket Station. 


ASSIMILATION OF LIMESTONE BY GRANITIC AND PEGMATITIC MAGMAS 


The writer believes that the mode of occurrence of the clino- 
pyroxene gneiss around and near the occurrences of limestone 
included in the gneiss is in itself sufficient evidence that the gneiss 
has received its excessive amount of lime through assimilation of 
the limestone. 


™U.S. Geol. Surv., Bull. 159, pp. 32 and 5st. 


288 PENTTI ESKOLA 


We have found that the clinopyroxene gneiss has, in the majority 
of cases, a composition that is nearly the same as that of the ordinary 
Becket gneiss of the surrounding area, excepting that it has a higher 
amount of lime. A bulk composition may therefore be computed 
from the analysis of the gneiss quoted above (p. 268), by allotting 
CaO enough to form anorthite with all the excess alumina (actually 
present in the micas), to form diopside with all the (Mg,Fe)O 
present, and titanite with all the TiO, present. Such a calculation 
indicates practically no other change in the composition than an 
increase in the percentage of lime from 1.30 to 2.97 per cent. 
The rock would contain 25.2 wt. per cent quartz, while the actual 
Becket gneiss has 27.7 wt. per cent. 

The actual clinopyroxene gneiss in many cases differs from 
this imaginary rock in being poorer in quartz, and having soda in 
excess Over potash or potash over soda. ‘The aplitic gneiss around 
the skarn inclusions near Benson Pond is an example of a highly 
potassic type low in soda and silica, being composed. practically 
of potash feldspar only. The analyzed rock from south Peru, on 
the other hand, is an extremely sodic variety, low in potash and 
silica. In both cases practically all the excess silica is gone, and 
the feldspars and the pyroxene only have been left. 

Where a magnesic clinopyroxene occurs in an acid gneiss it 
proves that the assimilation of limestone had happened nearly 
in place without further development, while the occurrence of 
pyroxene high in the iron compound, as was found in some cases, 
indicates that some differentiation had taken place after assimila- 
tion. Both cases occur in the banded Becket gneiss (cf. p. 277). 

The skarn exhibits a special case. It is supposed to be a 
pneumatolytic contact-product formed in limestone in such a way 
that adjacent magmas have brought in silica and metal compounds 
which have replaced the carbonate. 

It is pretty clear, in the case of the clinopyroxene gneiss in 
question, that it has been formed in part by direct assimilation of 
limestone by the gneiss magma and in part by the assimilation of 
skarn previously formed by contact action. 

The main indication of the banded structure of the gneiss 
seems to be that its intrusion and crystallization has happened 


GNEISS AND LIMESTONE CONTACT PHENOMENA 289 


under the conditions of stress and movements. The effective 
stirring together of the earlier solid materials with the newly 
intruded mass doubtless was one of the chief reasons why assimila- 
tion of limestone on a rather large scale has occurred here. 

Very obviously some amount of assimilation of older silicate 
rocks associated with the limestone has also taken place. Some 
evidences to this effect have been presented by Emerson." In its 
actual characters, however, the banded gneiss is altogether an 
igneous rock and its band structure is not by any means a direct 
result of injection combined with some kind of ultrametamorphism. 
Its whole structure is markedly different from that of such injection 
gneisses and migmatites known to me from many other regions 
and, also, from the eastern boundary zone of the Becket gneiss 
west of Chester. 

We may make some comparisons with similar phenomena in 
other regions. Such clinopyroxene gneisses, or clinopyroxene 
granites, occurring in connection with limestones and evidently 
formed by assimilation of limestones surely do occur in many 
tracts. That they are little noticed may in part be due to their 
inconspicuous aspect and often imperceptible difference in the 
hand specimen from ordinary granites and gneisses. On the other 
hand, it is quite certain that they are not by any means regular 
associates of granites intruding limestones. I shall name some 
examples from my own experience. 

My work in the Orijarvi region in southwestern Finland proved 
the absence rather than the occurrence of such products of assimila- 
tion of limestone, although it is an area where gneissic granites 
are intrusive into limestone-bearing formations. Only in a few 
places was the boundary type of the gneiss granite against limestone 
found to be diopside-bearing,” and only where it intersected lime- 
stone as dikes. None of the granites designated as ‘“‘microcline 
granite,” which are distributed all through southern Finland and 
very often cut limestones or contain inclusions of them, was ever 
found to contain clinopyroxene as a rock mineral, excepting in 
pegmatite dikes (cf. below). Similar relations have also been 


t U.S. Geol. Surv., Bull. 597, p. 154. 
2P. Eskola, Bull. Comm. géol. Finl., No. 40 (1914), p. 61. 


290 PENTTI ESKOLA 


found obtaining in the well-known limestone-bearing area of 
Parainen (Pargas).% 

Another feature with which the non-occurrence of assimilation 
is evidently connected is that this granite which forms veined 
gneisses or migmatites with all kinds of siliceous rocks has never 
been intruded into limestones to form intimate mixtures. There- 
fore the writer made the following general statement? “The 
resistance offered by limestones against granitization is very 
remarkable. Even in the midst of a migmatite area, where all 
siliceous rocks have been thoroughly mixed or assimilated with the 
granite magma, the limestones are generally quite free from granitic 
injections, and are intersected only by rectilinear dikes.’’ More 
evidence of assimilation of limestone by granite magma was 
brought forth during an investigation of certain areas in Trans- 
baikalia.s 

The peninsula of Sviatoy Noss, on the east coast of Lake 
Baikal, is made up of crystalline schists, among which are numerous 
masses of limestones, and of huge batholiths and smaller bodies 
of granites intrusive into the schists. The granites, when meeting 
limestone either as dikes or as large masses, show no endomorphic 
change whatever at the contacts, while in certain intrusive masses 
there is a very remarkable change in the whole rock bodies, the 
rocks (called sviatonossite) being composed of alkali feldspars 
with aegirite-augite and andradite. The high percentage of lime 
present has clearly been derived from the limestone, but the process 
of assimilation has taken place in such a way that andradite- 
clinopyroxene skarn has first been formed from the limestone and 
the skarn has subsequently been absorbed. This is evident from 
the common occurrence of fragments of skarn in the rock showing 
every degree of assimilation. Finally, all the minerals have crystal- 
lized out from the magma and the rock is now an ordinary unaltered 
igneous rock. 


t Laitakari, op. cit. 

210p. Git., P. 30. 

3 P. Eskola, “On the Igneous Rocks of Sviatoy Noss in Transbaikalia,” Oversikt 
av Finska Vet.-Soc. Férhandlingar, Vol. LXII, A, No. 1 (1921). 


GNEISS AND LIMESTONE CONTACT PHENOMENA 291 


On the island of Gursk6, on the west coast of Norway,' the 
writer studied a large mass of crystalline limestone surrounded by 
an intrusive gneiss. At the boundary of the limestone there occurs 
a zone in the gneiss consisting of a banded clinopyroxene-oligoclase 
rock. This rock undoubtedly has obtained some of its lime by 
assimilation of the limestone, and its characters are, in so far as 
regards the structure and probable mode of intrusion and consolida- | 
tion during vehement movements, closely similar to those of the 
western Massachusetts rocks. 

In all cases mentioned, acid magmas have been enriched in 
lime derived from limestones and have crystallized as such without 
any considerable change of composition through differentiation. 
We shall not discuss here at all those processes by which alkaline 
rocks are supposed to form from igneous magmas which have as- 
similated carbonates.” 

Diopside pegmatites in limestone regions.—While rocks, in 
occurrence and origin similar to the diopside gneisses described, 
are uncommon, there are in most limestone areas cut by granites, 
very numerous dikes of pegmatite containing diopside, titanite, 
and other lime-bearing minerals. Such pegmatites occur so com- 
monly in the western Massachusetts area’ and elsewhere, that we 
do not need to give any further examples. The large quantity of 
titanite present in many of these dikes is remarkable. 

_ The writer’s field evidence from southern Finland goes to prove 
that the assimilation of lime has taken place within the pegmatite 
fissures themselves and not in the granitic parent magmas of the 
pegmatite. This is well illustrated by the frequently observed 
feature that a pegmatite, cutting through limestone and other 
rocks as well, has developed much diopside and titanite only while 
intersecting the limestone, but is an ordinary mica pegmatite 
outside of the limestone.* 


tP. Eskola, “On the Eclogites of Norway,” Vid. selsk. Skri. Mat.-naturv. KI. 1, 
No. 8 (1922), p. 24. 


? As the writer has suggested, the Sviatoy Noss rocks show an incipient stage of 
development toward “alkalinity.” 


3B. K. Emerson, U.S. Geol. Surv., Bull. 597, p. 10. 
4 Cf. A. Laitakari, op. cit., p. 7. 


292 PENTTI ESKOLA 


The pegmatites, like other igneous rocks, show a varying 
behavior toward the limestones: The writer has seen, in several 
limestone areas, swarms of pegmatite dikes some of which carry 
clinopyroxene and titanite whereas others do not and could find 
no differences to which this might be ascribed. 

The common occurrence of phenomena of assimilation in 
pegmatites interests us because it proves decisively that this 
process does not require any high temperatures. The pegmatites 
crystallize out from residual solutions which still exist in liquid 
form after the main igneous bodies have become solid. The 
common occurrence of 6-quartz as well as a-quartz proves that 
the temperature has frequently been below their transformation 
point at 575° during the crystallization of pegmatites. 

Now, what has been the temperature of consolidation of the 
diopside-bearing varieties of the Becket gneiss? If we could 
estimate this, we would know that the assimilation took place at 
slightly higher temperatures than that of the beginning of crystalli- 
zation. This must have happened between the three-phase points 
quartz-calcite-wollastonite and quartz-dolomite-diopside invariant 
under the existing pressure. The higher of these points apparently 
lies below the inversion-point a-quartz-tridymite, but at present 
we cannot state it more closely. 

The other minerals of the gneiss indicate that the temperature 
had decreased very much before crystallization was complete. 
The rock contains epidote as individual large grains associated 
with albite, the epidote probably not being entirely of secondary 
origin. The potash feldspar does not contain any threads of 
albite (perthite), a fact that might indicate that the temperature 
was so low when the microcline crystallized that no considerable 
amount of albite could be taken into solid solution to separate later 
and form perthite. Only the lowest-temperature pegmatites 
contain such homogeneous potash feldspar. 

Turning finally to the question why limestone is assimilated 
in some cases and in others not at all, it seems that this may be 
largely dependent upon mechanical conditions. Assimilation is 
promoted, if the intrusion is connected with folding and the intrusive 
magmas are agitated and mixed with the crushed materials from 


GNEISS AND LIMESTONE CONTACT PHENOMENA 2903 


the country-rock. ‘This is in accordance with the experience that 
assimilation usually occurs where the intrusive rock shows banded 
structure and a high degree of protoclastic crushing and shearing, 
while it does not occur in masses which have been intruded and 
crystallized under quieter conditions and have come in contact 
with the country-rock along smooth surfaces only. 

It does not seem, however, that the whole question can be 
covered by this explanation only. ‘There are no doubt differences 
in the physicochemical conditions, in the concentration of volatile 
compounds, etc., which cause assimilation to occur in one case and 
not in another. Such relations may be understood better as soon as 
more experimental evidence has been gathered about the solubility 
of carbonates in magmas. 


SUMMARY 


Within the area of the igneous Becket granite gneiss in western 
Massachusetts there occur several tilted up layers of crystalline 
limestone, called Coles Brook limestone, older than the gneiss and 
metamorphosed by its contact influence. In the vicinity of the 
limestone the gneiss contains considerable quantities of lime- 
bearing silicates, especially of clinopyroxene (diopside-hedenbergite) 
and titanite, apparently the result of assimilation of limestone by 
the gneiss magma. 

The gneiss is markedly banded, with alternating darker and 
lighter bands. It was found, by determining the refractive indices 
of the chief mafic minerals, biotite, clino-amphibole, and clino- 
pyroxene, that the amount of their magnesia compounds in propor- 
tion to their ferrous compounds increases with the total quantity 
of the mafic constituents. At the same time the amount of anor- 
thite in the plagioclase increases. Thus the dark bands behave 
like the earliest separated rocks in a differentiation series. Some 
differentiation by crystallization really seems to have taken place 
after the assimilation. In certain places, however, and especially 
at the immediate contacts against the limestone, the actual compo- 
sition of the gneiss appears to be a direct result of assimilation and 
no correspondence between Fe:Mg ratio and “basicity” exists.— 
The distribution of magnesia and ferrous oxide among the different 


204. PENTTI ESKOLA 


mafic minerals was found to show a certain regularity, the mica 
always being richest and the clinopyroxene lowest in the iron 
compound. The variation of the “Fe-quotient”’ is believed to offer 
an important characteristic of crystalline rocks, though at present 
little understood. 

When silica-bearing limestones are subjected to metamorphism 
there occur reactions between the carbonates and silica, and 
silicates of lime and magnesia are formed. The temperature of 
reaction varies with pressure and is different for different minerals 
formed, as pointed out by V. M. Goldschmidt. The writer’s 
earlier investigations have established that, among the common 
accessory silicates in limestones, wollastonite requires the highest 
temperatures to form, and diopside and tremolite, successively 
lower. At still lower temperatures silica, in the form of quartz, 
remains uncombined. Thus we may distinguish the following 
types of metamorphic limestone: wollastonite limestone, diopside 
limestone, tremolite limestone, and quartz limestone. ‘These types 
may be used, under certain conditions, as a geologic thermometer, 
and it is hoped that the equilibrium curves of the different silicates 
with the carbonates may soon be determined experimentally. 

The limestones of western Massachusetts were found to repre- 
sent all the above-named types excepting the wollastonite limestone. 
Their mode of occurrence harmonizes with the writer’s earlier 
experience, diopside limestone occurring at the immediate contacts 
of the gneiss and tremolite limestone and quartz limestone 
successively farther away. 

A review of the writer’s experience from limestone-bearing 
regions where intrusive granites occur seems to prove that such 
phenomena of assimilation of limestone as those observed in 
western Massachusetts are not at all of regular occurrence. Prefer- 
ably they seem to occur in those regions where gneiss magmas 
have been intruded in connection with mountain folding, thus 
being dependent on the mechanical conditions in all probability. 
It appears, also, that assimilation does not require very high 
temperatures, being a very common phenomenon in granite 
pegmatite cutting limestones. 


Pee RiiiCisM OF THE ~“FAUNAL RELATIONSHIPS OF 
THE MEGANOS GROUP” BY BRUCE L. CLARK 


ROY E. DICKERSON 
Manila, Philippine Islands 


In a recent paper’ published in the Journal of Geology Dr. Bruce 
L. Clark revises the Eocene scale of California by introducing a 
new division, the Meganos group, by cutting off the lower portion 
of the strata which had previously been referred to the lower 
Tejon. Clark’s essential basis for division is stratigraphy first 
recognized in the area north of Mount Diablo, Contra Costa 
County. After his recognition of an unconformity in this area, 
subsequent faunal work led him to assert a marked faunal break 
as well although he recognized that the fauna obtained from these 
beds “‘appears to be more closely related to the Tejon than to the 
Martinez.”’ In this paper Clark deals with the general correlation 
of the middle and upper Eocene Sections of the Pacific Coast and 
he tentatively correlates his Meganos group with the Wilcox of 
the Gulf Coast. | 

Owing to his absence from the United States, the writer is 
unable to discuss this paper in detail, but there are certain general 
conclusions of Dr. Clark’s to which he wishes to record a firm 
dissent. 

Une evaluation of an unconformity is frequently a difficult 
matter and in many cases only a close study of the faunas from 
above and below the line of unconformity will enable the paleon- 
tologist to determine the relative value of the time break recorded 
in the rocks. Now the recognition of the existence of unconformi- 
ties within the Tejon Group is not new, and largely upon this 
account the writer has consistently clung to the term group in 
describing the Tejon as a stratigraphic and faunal unit. In 
making a study of the Tejon Group, its probable future division 

t Bruce L. Clark, “‘The Stratigraphic and Faunal Relationships of the Meganos 
Group, Middle Eocene of California,” Jour. Geol., Vol. XXIX (1921), pp. 125-65. 


295 


296 ROY E. DICKERSON 


“. = 
— 


’ aysun b 


ont 
sony 


= 


aba 
Wasik: 


x 
= 
. 


2) 


ne 


“a 
| 
ry 


rel 


WE 


Fic. 1.—Outline map of California showing (1) Mount Diablo region; (2) region 
north of Coalinga; (3) south end of San Joaquin Valley; (4) Camulos quadrangle; 
(5) Table Mountain, in the vicinity of Oroville. 


“FAUNAL RELATIONSHIPS OF THE MEGANOS GROUP” 297 


into formations was constantly kept in mind. In the special case 
in point the writer regards the unconformity north of Mount 
Diablo as being such a one as would separate the Tejon group into 
formations, while Dr. Clark would make two groups. Dr. Clark 
admits that this unconformity was not stratigraphically recognized 
in the fine Tejon-Eocene section only a few miles distant on the 
south side of Mount Diablo. 


Detailed mapping has failed to show any marked difference in dip and 
strike between the Meganos and the Tejon in this southern area, such as occurs 
to the north of the mountain (Mount Diablo). At a few localities there is an 
apparent difference in dip between the beds of the two horizons; this, however , 
could not be verified with certainity, the division being recognized by a sharp 
change in lithology, and by faunal evidence [p. 141]. 


The footnote on this same page is likewise significant: 


In the former paper [‘‘Meganos Group, a Newly Recognized Division in 
the Eocene of California,” Bull. Geol. Soc. America, Vol. XXXIX (1918), 
pp. 281-96] referred to above, the writer stated that in this section there is a 
marked difference in strike between the Meganos beds and those of the Tejon, 
and the difference was taken as one of the evidences of unconformity between 
the beds of these two horizons. Later work, however, has shown that this 
apparent difference in strike is, in part at least, the result of faulting. Also it 
was stated that to the east of this area the Meganos disappeared due to this 
unconformity. At that time the writer had not recognized that the so-called 
Tejon beds to the east, as described by Dickerson, were in part Meganos. 


_Concerning the presence of unconformities in other parts of 
California, Clark refers to his studies made in the vicinity of 
Coalinga and Simi Hills, Ventura County, as follows: 


- The results of this work show conclusively that beds of both Meganos 
and Tejon age are present in all of these areas, and that there is in each an 
unconformity separating the strata of those two series... . As seen 
between . . . . Domengine Creek and Cantua Creek (Coalinga Quadrangle), 
the upper beds of the Meganos consist of a white sandstone which was mapped 
by Anderson and Pack as a part of the Tejon. The contact between the 
Meganos and Tejon comes in between this sandstone and somewhat similar 
sandstones of the Tejon. It is, as a rule, marked by a conglomerate and is 
irregular at numerous localities. The sandstones below the contact, due to 
the unconformity, thicken and thin very noticeably along the strike. Also, 
‘at a number of localities the lower sandstones show a dip and strike appreciably 
different from those of the Tejon beds above. While these differences amount 


* Evidently series is used loosely. 


298 ROY E. DICKERSON 


at the most to only a few degrees, it is sufficient to cause the lower sandstone 
layers to be cut off obliquely, and on the cliff sections they are seen to abut 
against the basal beds of the Tejon [pp. 143-45]. 


The writer cannot see how “the sandstones below the contact, 
due to the unconformity thicken and thin very noticeably along the - 
strike’? but rather thinks that the beds were deposited near shore 
and that the sandstone lenses into shale or the shale gradually 
grades into sandstone. If memory is correct, Anderson and Pack’s 
mapping indicates that several comparatively thick members 
exhibit this same phenomenon on a great scale. Anderson and 
Pack, F. M. Anderson, Clarke Gester, and J. A. Taff had good 
opportunities to study this section and as far as I recall none of 
them recorded any notable dip differences. ‘The writer did not do 
extended field work in this section, but from what was observed, 
he believes that careful search will bring to light several erosional 
unconformities in the region. That is, the Tejon group in this 
region was deposited under strictly littoral conditions, and from 
time to time comparatively slight emergences of the Eocene con- 
tinental strand line are recorded by these erosional unconformities. 
Dr. Clark states that the lithology of the sandstones above and 
below this unconformity are essentially similar. This similarity is 
so close that a series of hand specimens from above would not be 
separable from a series from below if the two were mixed. Deposi- 
tional conditions over the present sites of Simi Hills region and 
vicinity of Grapevine Canyon are characteristically littoral as 
indicated by both the fauna and the lithology. 

The fauna of only fifteen species listed by Clark from San 
Emigdio Canyon on page 149 is entirely too meager upon which 
to base definite broad conclusions. Of these, eight are new species, 
one is only generically determined, two are doubtfully referred to 
described Eocene species. However, it is quite possible that the 
Turbinolia Zone of the Tejon Group (Meganos Group of Clark) 
may be present here. 

Dr. Clark assigns the Siphonalia sutterensis Zone of Dickerson 
to his Meganos Group largely upon faunal grounds, since Turritella 
merriami, Ancilla (Oliverato) californica, and a few other forms 
are found at Oroville, Marysville Buttes, in the vicinity of Mount 


“FAUNAL RELATIONSHIPS OF THE MEGANOS GROUP” 299 


Diablo, Camulos Quadrangle, and the Coalinga region. On pages 
130 and 131 Clark states that 

After discussing the various Eocene sections, reasons will be given for 
correlating the beds referred to the Meganos group in these different areas in 
the Coast ranges with one another and with the marine Ione formation in the 
Sierra Nevada foothills, as mapped and described by Lindgren and Turner 
not, however, including the type section of the Tone.t 
Why not discuss the type Ione? It is well described and at one 
locality has yielded a fair but determinable marine fauna contain- 
ing Turritella merriami and other typical Tejon species. Also the 
strata of the type Ione clearly intergrade with the Tertiary Aurif- 
erous gravels of the Sierra Nevada foothills. Now the type Ione 
may be traced southward and connected with the Marine Eocene 
strata a half-mile south of Merced Falls where specimens of Venerz- 
cardia planicosta merriami may be collected in abundance. Farther 
south, the stratigraphy of the Ione clearly demonstrates deposition 
by a sea transgressing from the west. North of the type section of 
the Ione, Lindgren and Turner have traced these beds through to 
Oroville. Dr. Clark in his historical review quotes from Dicker- 
son’s “Note on the Faunal Zones of the Tejon Group” as follows: 

A study of the relationship between zone 3, Mount Diablo region, and the 
Siphonalia sutterensis zone and their geographic position suggest that the 
uppermost strata of the Marysville Buttes and Oroville were deposited by a 
transgressing sea, and that only in favored places along the western borders 
of the Sierra have the latest Eocene sediments been preserved from erosion. 
Lava caps such as that of the older Basalt of South Table Mountain have 
preserved these youngest Tejon sediments which have heretofore been regarded 
as Ione. 
This quotation creates the impression in the reader’s mind that 
Dickerson’s concept of the Tejon-Ione relations was purely theo- 
retical, whereas such is not the case. The stratigraphy of the Ione 
at Bear Creek 20 miles south of Merced Falls, at Merced Falls, at 
Ione, at Oroville, all clearly indicate deposition by a transgressing 
sea in close proximity to an old Eocene shore. Into this Eocene 
sea the streams of the low mountainous Eocene upland poured their 
golden sands. The reader is referred to Dickerson’s paper, “Stratig- 
raphy and Fauna of the Tejon Eocene of California,” for a full 


t Dickerson’s italics. 


300 ROY E. DICKERSON 


discussion of these essential matters. Clark has evidently missed 
the significance of this evidence as he states on page 162 of his 
paper that 


Dickerson attempted to establish the stratigraphic sequence of his upper 
faunal zone in relation to that of the typical Tejon indirectly, not having the 
two faunas in the same section. His idea that the Siphonalia sutterensis 
fauna is younger than that of the typical Tejon appears to have been founded 
principally upon what he considered evidence for different stages of evolution 
of certain pelcypods, such as Venericardia planicosta merriami Dickerson and 
Cardium marysvillensis Dickerson. He believed that the variety merriami 
was derived from the variety horniz. Later stratigraphic work has shown that 
these species occur in a sequence the reverse of that which Dickerson originally 
supposed, the Venericardia planicosta merriami coming in beds older than those 
containing the variety hornii. "The same is true of the other species, which were 
derived from typical Tejon species. 


It is true, however, that the problem of Ione-Tejon relations was 
attacked with faunal weapons as well. Clark states that ‘Later 
stratigraphic work has shown that these species occur in a sequence 
the reverse of that which Dickerson originally supposed... . . 
In the historical review on page 129, Dr. Clark reviews a short 
paper, by Arnold and Hannibal, and includes the following quota- 
tion from it: 


The writers have shown that in Oregon and Washington the Eocene may 
be divided into three faunal divisions, the Chehalis, Olequa, and Arago or 
Ione formations. The Chehalis formation is characterized especially by 
Venericardia hornii Gabb, Meretrix californica, Pecten (Chlamys) landesi or 
Venericardia hornii Gabb and a tropical flora, and the Arago or Ione formation 
by Turritella merriami Dickerson, a form of V. hornii with obsolete ribs (var. 
avagonia A. and H.), and a tropical flora. 

The Arago or Ione beds represent a horizon younger than any Tejon 
recognized in the Tejon or Puget Basin. The Arago or Ione beds occurring 
as they do in basins distinct from those in which the Tejon series is developed, 
and being formed at a different period, must be treated as a distinct division of 
the Eocene. 


t Arnold and Hannibal use formation as an equivalent for faunal zone or horizon 
and loosely use formation, group, series. The form referred to as V. hornii var. aragonia 
A. and H. was not described by them but was collected at the type locality of V. 
planicosta merriami, on Little River, Roseburg Quadrangle, Oregon. Arnold and 
Hannibal classed these beds as Arago (or Ione). 


“FAUNAL RELATIONSHIPS OF THE MEGANOS GROUP” 301 


In connection with this statement Clark refers to Weaver’s 
stratigraphic studies wherein Weaver shows that the Olequa and 
Chehalis of Arnold and Hannibal were reversed. Weaver’s care- 
ful work cleared up this succession but apparently does not invali- 
date Arnold and Hannibal’s assignment of the Arago as the upper- 
most formation of the Eocene. Arnold and Hannibal regard the 
type locality of Venericardia planicosta merriamt, on Little River, 
Roseburg Quadrangle, Oregon, as being in the uppermost portion of 
the Eocene, their Arago formation, and in this general conclusion the 
writer isin agreement. The evidence yielded by evolutionary forms 
of Venericardia planicosta have not been sufficiently studied by 
Dr. Clark. The first of these forms is V. planicosta venturaense 
Waring and was described from the Martinez (Lower Eocene) of 
the Simi Hills, Ventura County, California. Waring’s type was 
considerably eroded but a mature specimen collected from near the 
type locality by the writer shows strong V-shaped ribs marked by 
very prominent nodes. Now these characters are conspicuous only 
in the youthful stages of V. planicosta hornii Gabb, the nodose 
character disappearing rapidly as the specimen matures. In very 
youthful specimens of V. planicosta merriami the same characters 
appear but these forms upon reaching maturity are marked by 
nearly complete obsolescence of ribs as well. 

Clark, in the writer’s opinion, overemphasizes the presence of 
new species in the Eocene and permits this to color his views. It 
is the writer’s experience that in California, where unusually good 
preservation is found, many new species will be discovered. We 
must not lose sight of the fact that the pelecypod and gastropod 
fauna of the Tejon group is probably not much more than half- 
described. And on this account we must not create new horizons, 
based largely on such evidence. Again, let us not forget that the 
Tejon group is largely composed of inshore or strictly littoral sedi- 
ments, and that the lignite seams occurring commonly throughout 
California, Oregon, and Washington, generally indicate deposition 
in lakes or lagoons bordering the Eocene shore. ‘Thus during the 
deposition of the upper Tejon north of Mount Diablo three different 
carbonaceous beds were laid down, and there is evidence to show 
that the sea was at least temporarily withdrawn while these lignitic 


302 ROY E. DICKERSON 


strata were being formed. In other words, minor unconformities 
are here present. Sixty miles east of Mount Diablo there is a ten- 
foot seam of coal at Ione in the type section. Now if Clark is 
right in correlating the Ione with his Meganos group, then the 
Meganos group is again broken by an unconformity, since several 
hundred feet of Eocene sediments underlie the Ione coal seam 
which rests unconformably upon them. 

Along the foothills of the Sierra Nevada in the undisturbed, 
nearly horizontal Eocene beds many interesting data are yet to be 
secured, as here the old Eocene shore is traceable and the streams 
of the old Eocene peneplain are still preserved beneath thick lava 
for the inspection of some untiring geologist interested in recon- 
structing the past of this wonderful land. 


THE AGE OF THE-DOMES AND ANTICLINES IN THE 
LOST SOLDIER-FERRIS DISTRICT, WYOMING? 


A, 18 JveNIU Sl 
United States Geological Survey, Washington, D.C. 


INTRODUCTION 


The relative age of the major and minor folding in Wyoming is 
a somewhat mooted question among geologists, especially among 
those who are interested in the problems of oil and gas accumulation. 
In a recent paper Ball? reached the conclusion that nearly all the 
minor folds of Wyoming were formed during the period of formation 
of the major uplifts. That exceptions to this general rule may 
exist is frankly admitted by Ball, and he cites the Simpsons Ridge 
fold as the one example with which he is familiar of an uplift in 
which the minor folding is clearly younger than most of the major 
folding. Although agreeing in general with Ball’s conclusion, the 
writer believes that exceptions to the rule are more numerous than 
Ball suggests. He believes further that a most noteworthy excep- 
tion to the rule is to be found in the folds of the Lost Soldier— 
Ferris oil and gas district of south-central Wyoming, the age 
relations of which are here discussed. 


GEOGRAPHIC AND GEOLOGIC RELATIONS 


The Rawlins uplift, in south-central Wyoming, is about fifty 
miles in length and twenty miles in width and trends in a northerly 
direction (see map). It is not large enough to be classed among 


« Published with the permission of the Director of the United States Geological 
Survey. The information presented in this paper was obtained in the summer of 
1920, during an examination made to obtain data for classifying the public land in 
the oil and gas fields of the Lost Soldier—Ferris district. The results of this examina- 
tion are being prepared for publication by the United States Geological Survey. 


2Max W. Ball, “The Relative Ages of Major and Minor Folding and Oil 
Accumulation in Wyoming,” Amer. Assoc. Petroleum Geologists, Bull. 5 (1921), No. 1, 
PP- 49-63. 


Io8} 


304 A. E. FATEH 


the major uplifts in Wyoming on account of its size, but the fact 
that its central axis is elevated so high that the pre-Cambrian 
rocks are now exposed at the surface is sufficient, as pointed out 
by Ball,’ to justify ranking it as a major uplift. Its alignment 
makes it a part of that series of Rocky Mountain flexures char- 
acterized by northerly lines of folding and faulting. It is to be 
noted especially that the Rawlins uplift is one of a group that forms 
the northernmost member of this northerly series, beyond which 
the Rocky Mountain folds abruptly change in direction to a 
transverse series with east-west trend, of which the Sweetwater 
uplift, described below, is one. 

From the general horizontal position of the Wasatch beds on 
the west flank of the Rawlins uplift, it seems certain that the 
development of this uplift was complete, or practically complete, 
by the beginning of Wasatch time. 

The central pre-Cambrian granite core of the Rawlins uplift 
is faulted along its west side, and toward the north this fault zone 
turns northeastward and crosses the axis of the fold. The portion 
of the uplift north of this fault is on the down-dropped side, and 
this lower-lying north end of the uplift is occupied by the oil and 
gas fields of the Lost Soldier—Ferris district. The oil and gas 
accumulations of this district are controlled by minor folds, and 
it is these minor folds that constitute the subject of this paper. 

North of the Rawlins uplift is the Sweetwater uplift, a major 
fold about one hundred miles long and forty miles wide that trends 
nearly due east. The Granite Mountains, which occupy the 
central part of this uplift, represent the higher peaks of the much 
dissected pre-Cambrian crystalline rock core, whose valleys and 
lower-lying parts are now filled and covered by nearly horizontal 
Tertiary sediments. These sediments form a nearly flat plain, 
above which the Granite Mountains rise like islands in a sea. On 
the south margin of the Sweetwater uplift, immediately adjacent 
to the Lost Soldier—Ferris district, are the Ferris and Seminoe 
mountains. The north side of the Ferris Mountains consists of 
pre-Cambrian crystalline rocks, adjacent to which, in a sharply 
upturned attitude, lie the Paleozoic and Mesozoic sediments that 


™ Max W. Ball, op. cit., p. 51. 


¢ 


JourNAL oF GEOLOGY, Vou. XXX, No. 4 


R.QTW. 26 


Plate TIT 


lve 
Ly 


INDEX MAP 


Mesaverde and younger 
formations. Within the 


Sweetwater Uplift: flatlying 


Steele and Nio- 
brara formations 


White R. and Wind R. formations 


Grorocic Map or THE RAWLNcs anp SWEETWATER UPLIFTs, WYOMING 


PLANATION 


Formations from base 
of Niobrara to base 
of Cambrian 


7 


Pre-Cam 


‘Map compiled trom U:S- Geological Survey pu 
by Co. Hares, 


b 


x 
NY 


25 See Se 2 = T 
—————— | 
| | T.3SN. 
T.34N 
TS3N 
T:32N 
T3IN 
TON 
T:29N 
T.28N 
/; T27N 
ie SOT eg 
GNM GEES Pale 
LINCOM TI Zi Z 
: QL VA i: 
“piss O gE ea R, 
SHER YL “muy Wey , OU) T.25N 
Ly yy T.2AN 
SOS T25N 
1 
4 ‘y i Yy T2IN 
RORY GY; Y/ b 
LY Wize 


ae Anticlinal folds with domes Synclinal folds 


E.E.Smith, aid A:C.Ve 


THE AGE OF THE DOMES AND ANTICLINES 305 


form the south side of the mountains. In some places high on 
the flanks of the mountains the sedimentary beds are vertical or 
even slightly overturned. 

To the southeast of the Ferris Mountains this sharp folding is 
replaced by faulting with thrust from the north, so that the crystal- 
line rocks of the Seminoe Mountains now lie against the Cretaceous 
rocks of the Lost Soldier—Ferris district. The fault plane along 
which this overthrusting took place probably was almost vertical, 
and the horizontal displacement therefore slight, but even so the 
adjacent sedimentary beds of the Lost Soldier—Ferris district are 
somewhat overturned. It seems very probable that this over- 
thrusting continues eastward as far as the Freezeout Hills. West- 
ward from the Ferris Mountains the folding changes into faulting, 
but in this direction the amount of overthrusting is not so great as 
to the southeast. 

The Sweetwater uplift, like the Rawlins uplift, is considered to 
have been developed mainly in pre-Wasatch time, but it differs 
from the Rawlins uplift in that the main deformation was followed 
by later movements of considerable magnitude. The importance 
and relations of these later movements are discussed below. 


AGE OF THE FOLDS 


The points to be brought out are two: (1) The folds of the 
Lost Soldier—Ferris district were not produced by the same forces 
that formed the Rawlins uplift, on which they were superimposed, 
but were produced by the forces that formed the Ferris and Seminoe 
mountains, to the north and northeast. (2) The Ferris and 
Seminoe mountains are considerably younger than the Sweetwater 
uplift, on whose south margin they rise, and probably represent a 
relatively late readjustment along this margin. If these points 
are established it follows that the folds of the Lost Soldier—Ferris 
district are younger than the Rawlins uplift and also younger than 
the main development of the Sweetwater uplift. 

1. With the geographic setting of the Rawlins and Sweetwater 
uplifts in mind (PI. III), attention should be directed to the align- 
ment of the minor folds in the Lost Soldier—Ferris district. The 
downfold that structurally separates the Rawlins and Sweetwater 


306 A. E. FATH 


uplifts is the sharply flexed Camp Creek syncline, which lies only 
a few miles south of the Ferris and Seminoe mountains. It par- 
allels the direction of the fault line along the southwest side of the 
Seminoe Mountains. North of this syncline there are no minor 
folds on the south slope of the Sweetwater uplift, but instead the 
abrupt upfold and thrust of the uplift itself, which is represented 
topographically by the Ferris and Seminoe mountains. South of 
the Camp Creek sycline the Rawlins uplift is represented by a 
long, comparatively gentle slope up to the granite axis of the 
upfold in the Rawlins Hills. This gentle north and northeast 
slope is interrupted by minor folds, including a long upfold that 
extends from the Wertz dome at the west and is accentuated in 
its eastward course by the Mahoney, Ferris, and G.P. domes. 
As the surface rocks are concealed by dune sand and alluvial wash 
the exact course of this upfold is not well known except at the high 
points mentioned. The axis of this upfold is parallel not to that 
of the Rawlins uplift but to that of the Camp Creek syncline, 
which in turn lies parallel to the Ferris-Seminoe line of deformation. 
From this close parallelism of structure it would appear that this 
minor upfold is the result either of the same forces that produced 
the Ferris and Seminoe mountains or of similarly directed forces. 
Certainly it could not have been the result of the forces that formed 
the Rawlins uplift. The north flanks of the Wertz and Mahoney 
domes and the northeast flanks of the Ferris and G.P. domes are 
steeper than the opposite flanks, a relation which also helps to 
indicate that the forces which formed them probably came from 
the north and northeast. ‘To the north of the Wertz dome is the 
small Bunker Hill dome, which lis parallel to both the Wertz 
dome and the Camp Creek syncline. The course of the synclinal 
fold lying immediately south of the Wertz-Mahoney—Ferris—G.P. 
upfold is not clearly defined, although the syncline that separates 
the Mahoney dome from the Sherrard dome, and the Wertz dome 
from the Lost Soldier dome, is_probably the westward extension of 
the Table Hills syncline. It was this downfold which divided the 
north end of the Rawlins uplift and formed the Lost Soldier dome 
as a distinct feature from the Wertz-Mahoney-—Ferris—G.P. upfold, 
to the east. By considering the Table Hills syncline to have a. 
westward extension, as above mentioned, its course also shows a 


THE AGE OF THE DOMES AND ANTICLINES 307 


transverse direction to that of the Rawlins uplift and a parallelism 
to the Ferris-Seminoe line of deformation. The O’Brien Springs 
anticline, still farther south, shows a similar trend. 

From the considerations above set forth it seems reasonable to 
suppose that the minor transverse folding in the Lost Soldier—Ferris 
district on the Rawlins uplift was produced by compressive forces 
that came from the direction of the Sweetwater uplift, to the north; 
and from the parallelism between these folds and the Ferris- 
Seminoe line of deformation, it seems reasonable to suppose further 
that these minor folds represent the more distant effects of the same 
forces that formed the Ferris and Seminoe mountains. 

2. Ii the truth of the preceding arguments is conceded, there 
remains, to complete the purpose of this paper, proof that the 
minor folds of the Lost Soldier—Ferris district are decidedly younger 
than the major Rawlins uplift and the main development of the 
Sweetwater uplift. It has been shown that the axes of the minor 
folds are transverse to those of the Rawlins uplift, a relation which 
in itself implies, although it does not necessarily prove, that the 
minor folds are more recent than the uplift itself. The best proof, 
however, is to be found in showing that the Ferris and Seminoe 
mountains, which are of the same age as the minor folds of the 
Lost Soldier—Ferris district, are considerably younger than the 
main development of the Rawlins and Sweetwater uplifts. 

The Ferris and Seminoe mountains may be spoken of as a 
marginal rim to the Sweetwater uplift. The central core of this 
uplift, now represented by deeply eroded granite, is a structurally 
much more highly elevated part of the uplift than the Ferris and 
Seminoe mountains at its margin. It is possible that at the time 
of the main deformation the central core of this major uplift was 
elevated to several times the height of the marginal rim; but the 
particular fact to be noted is that the Ferris and Seminoe mountains 
now attain a maximum altitude of 10,025 feet (as determined by 
triangulation and vertical angles), a considerably higher altitude 
than that of the peaks of the Granite Mountains, which now form 
the core of the uplift, and whose highest point, Hayden Peak, is 
_ reported by Hares‘ to have an altitude of only 8,040 feet. This 


tC. J. Hares, ‘“Anticlines in Central Wyoming,” U.S. Geol. Survey Bull. 641 
(1916), p. 234. 


308 A. E. FATH 


difference of nearly 2,000 feet in altitude cannot be due to difference 


in erosion alone. Moreover, the Granite Mountains now show 
smooth surfaces, whereas the Ferris and Seminoe mountains are 
rough and rugged, a difference which indicates that the Ferris and 
Seminoe mountains are in a more youthful stage of erosion. 

From these differences in altitude and in character of topog- 
raphy, therefore, it appears clear that the present Ferris and 
Seminoe mountains are younger than the Granite Mountains. 
This conclusion further indicates that the Ferris and Seminoe 
mountains are probably the result of a relatively late readjustment 
along the south margin of the Sweetwater uplift, and, from the 
nature of the structural features in this region, that this later 
deformation was a slight overthrust in which the Sweetwater area 
moved southward toward the Rawlins uplift. 

How much later in geologic time than the main development 


of the Sweetwater and Rawlins uplifts this readjustment took place ~ 


is not readily ascertainable. From unpublished field information 
pertaining to T. 28 N., Rs. 90 to 95 W., along the south edge of 
the Sweetwater uplift, gathered by Hares just beyond the south 
margin of the area covered in his report on “‘Anticlines in Central 
Wyoming," and by Smith along the north margin of the area 
covered in his report on “The Eastern Part of the Great Divide 
Coal Field, Wyoming,’” it is clear that there has been considerable 
readjustment in these townships since pre-Wasatch time, when 
the major upfolds were first developed. The significant evidence 
along this marginal zone is to be found in the nearly flat-lying 


=. 


Tertiary formations (Wasatch?) in the Green and Crooks moun- 


tains, at altitudes of 1,000.feet or more above the nearly flat-lying 
Tertiary beds (Wind River and White River formations) of the 


basin that now marks the Sweetwater uplift. It is also an interest- — 


ing fact that. the White River (Oligocene) formation of the Sweet- 
water basin—the “‘sea’”’ in which the Granite Mountains stand out 
as “‘islands’’—seems to be limited on the south by the fault zone 
that marks the boundary between the pre-Cambrian rocks of the 
Sweetwater uplift and the Paleozoic and Mesozoic formations in 
tC. J. Hares, op. cit., pp. 233-79. 
2. E. Smith, U.S. Geol. Survey Bull. 341, pp. 220-42. 


THE AGE OF THE DOMES AND ANTICLINES 309 


T. 28 N., Rs. 89 to 95 W.* This greater altitude of the nearly 
flat-lying Tertiary beds south of the Sweetwater uplift and the 
possibility that the White River and Wind River formations lie 
only on the uplift itself are indicative of a readjustment movement 
here in post-Wasatch time and possibly as late as White River 
time. Ball? mentions the fact that the Hanna formation (Wasatch ?) 
is vertical and overturned in the Freezeout Hills. 

There has even been some post-Pleistocene movement in the 
region, for gravel-covered terraces in some places are reported to 
have slopes opposite to those of the present drainage. These 
comparatively recent movements were probably of small magnitude 
and are cited principally to indicate that deformation did not 
necessarily cease here in the Tertiary period, but that warping has 
occurred in the Quaternary. It is possible that deformation has 
taken place even within recent times. 

Inasmuch, therefore, as mountain-forming deformation has 
occurred here in post-Wasatch time and possibly as late as White 
River (Oligocene) time, or even later, it would seem that the 
Ferris and Seminoe mountains should be regarded as the local 
results of this late movement of readjustment along the south 
margin of the Sweetwater uplift. Although the exact time when 
these mountains were formed has not been ascertained, it seems 
nevertheless that they must be considerably younger than the 
main deformative movements which produced the Sweetwater and 
Rawlins uplifts. If it is regarded as established that the Ferris 
and Seminoe mountains are due to a late and possibly final spasm 
of mountain-forming movements in this general region, it should 
probably be conceded that the minor domes and anticlines of the 
Lost Soldier—Ferris district are also due to the same movements. 
Credit must be given to Ball* for his recognition of a marked 
difference between the deformation features of the Ferris and 
Seminoe mountains and those of the Wyoming mountains of the 


* No positive statement to this effect can be made, for the relation was noted 
over only a small area by the writer, and the evidence gleaned from the unpublished 
work of Hares and Smith above cited, though apparently supporting the inference, 
does not fully confirm it. 


2 (Os Cition Da 55x 
3K. C. Heald, personal communication. 4 Op. cit., p. 59. 


310 A. E. FATH 


usual type, but although he realized that this difference existed, 
he did not consider its significance in relation to the age of minor 
folds of the Lost Soldier—Ferris district. 


ECONOMIC CONSIDERATION 


The minor folds of the Lost Soldier—Ferris district are shown 
above to be distinctly younger than the Rawlins uplift, on which ~ 
they are superimposed, and the question arises as to the relation 
of this difference in age to the accumulation of the oil and gas in 
the Lost Soldier, Wertz, Mahoney, Ferris, and G.P. fields. As 
the Rawlins uplift was in existence in pre-Wasatch time, it is 
obvious that any oil and gas that prior to the minor folding of the 
Lost Soldier—Ferris district had been formed or had migrated a 
short distance above the margin of the uplift, where the catchment 
areas of the present fields are situated, must surely have been lost 
through the eroded edges of the formations on the top of this major 
uplift. The oil and gas of the present fields and of possible 
unproved pools in this region must have been formed largely by 
the dynamo-chemical action of the later deformational forces that 
caused the minor flexing. Some oil and gas of earlier distillation 
may have been already present within the catchment areas of the 
existing fields, but this oil and gas must have been only a fraction 
of that which was formed within the area embraced by the Rawlins 
uplift. The oil and gas now available in this district for the use 
of man must represent, therefore, merely a remnant of the total 
quantity derived from the mother-material which the rocks of 
this region originally contained. 


ON THE OCCURRENCE OF AN APUS IN THE PERMIAN 
OF OKLAHOMA 


RUDOLF RUEDEMANN 
New York State Museum, Albany, N.Y. 


Professor J. W. Beede last year sent two specimens of a crus- 
tacean from the Permian of Oklahoma to the New York State 
Museum for investigation. They were from a suite collected by 
Dr. Thomas T. Jackson in a thin sandstone bed of the Enid forma- 
tion, exposed “‘on the top of a hill four or five miles north of Elkeno, 
_ Oklahoma.” 

The specimens’ proved, on inspection, of exceptional interest 
for the reason that they not only exhibit an outline of the carapace 
as seen in Apus but even the impressions of the “‘shell-glands” 
or excretory organs in a form as it is today known in Apus, and its 
close relative Lepidurus. We have, therefore, no hesitation in 
considering this Permian form a true Apus, and propose for it the 
name A pus beedei sp. nov. 


A pus beedet sp. nov. 


-Description.—Carapace small (the larger specimen 15 mm. X 
1™4mm.); broadly elliptic, nearly circular in outline, with a small 
posterior emargination; shieldlike and sloping from the subcentral 
apex abruptly forward, more gently backward, and _ probably 
originally fairly steeply toward the lateral margins. A transverse 
cervical fold is situated about one-fifth of the length of the carapace 
posterior to the anterior border. Immediately posterior to this 
are the large but slightly curved “‘shell-glands” or excretory 
organs; beginning at either side of the median line they extend 
obuiquely backward about midway of the lateral faces of the 


t Professor Beede had stated in his letter that he had more specimens than the two 
sent but none showing features not seen in the ones inclosed. On inquiry, we learn 
that this further material is not available at present, but in view of Professor Beede’s 
statement and the great interest of the material, we venture to publish this notice 
without having had access to the less favorably preserved specimens. 


311 


Bia RUDOLF RUEDEMANN 


carapace to about one-fifth of the length of the carapace from 
the posterior border. The median line is marked behind the 
cervical fold by a distinct depression extending about one-third 
of the distance toward the posterior border. 

Horizon and locality——Permian (Enid formation), near Elkeno, 
Okla. 

Note on identification with A pus.—The writer is well aware that 
considering the enormous stretch of time from the Permian to 
the present time, and further the fact that only the carapace of the 


Fic. 1 Fic. 2 


Fic. 1.—A pus beedei sp. nov. The two specimens seen obliquely from above and 
the front. Natural size. The type-specimen is on the right. It shows the trans- 
versal wrinkle in the place of the cervical fold, portions of the frontal part of the 
carapace and the median depression. 


Fic. 2.—A pus beedei sp. nov. Lateral view of the type-specimen. 3. Shows 
the shell-gland and posterior emargination. 


Permian form is now at hand, there is a great possibility that if 
the body of the animal were preserved, differences of generic 
importance might be quite apparent, e.g., such as now separate 
Lepidurus from Apus and which consist in the different develop- 
ment of the post-anal plate. It could, therefore, be urged that 
this Permian form should be made a séparate genus on theoretic 
grounds. In such an attempt it would, however, be found that 
the carapace of the older form is not distinguishable from that of 
the recent type and the new genus would have no diagnosis, but 
only its great age to stand on. It is, however, to be understood 


ON THE OCCURRENCE OF AN APUS Bie 


that the term Apus expresses here the hard parts only and the entire 
group (Apus and Lepidurus) and that it is merely intended to 
point out the persistence of the Apus-type of carapace. The 
persistence of the exact form of the carapace in the group indicates 
that no evolutionary development profound enough to affect the 
carapace took place in all this time. 

Note on preservation of material.—The specimens are casts of 
the interior surface of the carapace. It is for this reason that the 
impressions of the shell-glands are so well preserved in some of the 
fossils, for it must be remembered that these organs are situated 
between the two layers that form the posterior portion of the 


Fic. 3 A Fic. 4 
Fic. 3—A pus beedei. Flat projection of outline of carapace. 


Fic. 4.—Lateral view of Apus aequalis Packard, 2; from Colorado and in 
N.Y. State Museum. 


carapace, and that these glands open where the inner layer termi- 
nates at the cervical fold. The frontal portion of the carapace is 
poorly preserved, partly no doubt on account of the greater thin- 
ness of the carapace there, which consists of but one layer, and 
partly owing to the lateral compression of the tests during entomb- 
ment. There is, however, enough left of the frontal outline 
(especially in the smaller specimen) to leave no doubt that it was 
rounded and unbroken originally. It would seem that the shield 
was sufficiently sloping toward the lateral margins to come fre- 
quently to rest on the side and then suffer lateral compression. 
The amount of lateral slope is greater also in the recent Apus than 
the usual dorsal views of the creatures would suggest, as shown in 
the lateral view drawn from nature by the writer and reproduced in 
Figure 4. A restoration of the dorsal view of the carapace of 
Apus beedei is given in Figure 3. It is obtained by plotting the 


314 RUDOLF RUEDEMANN 


larger specimen, in natural size, on the horizontal plane. An 
oblique view from above of the same specimen is given in Figure 1 
and a lateral view in Figure 2. The former presence of a cervical 
fold in the Permian form, corresponding to that in recent species of 
Apus (Fig. 4, where it is shown in the profile behind the eye), is 
indicated by a trans- 

cs verse wrinkle, especially 

distinct in the larger 
specimen, but also seen 
ta in thesmaller one. This 
\ wrinkle has resulted 
\ from the yielding of this 

Avie transverse fold during 
‘ Te? A ' lateral compression. It 
sh.gl We : has thereby become 
pressed downward and 
changed into a deep 
transverse wrinkle, in 
front of which the cara- 
pace has split and been 
drawn inward. The 
smaller specimen retains 
the frontal portion in 
more perfect forma 
\ though also much com- 
| pressed. Owing to this 
Fic. 5.—Apus cancriformis Schiffer. Dorsal folding-in of the frontal 


aspect. From Parker and Haswell (after Bronn’s part, the perforations of 


Thierreich). d.o., dorsal organ; FE, paired eye; the carapace for the eyes 
e, median eye; si. gl., shell-gland. : 


are not observable. 

The shell-gland appears as an elongate elliptic body consisting 
of two concentric furrows and one median and another outer 
longitudinal one. There are thus altogether six urinary tubes 
counted on the transverse line, just as in the recent Apus (Fig. 5). 
All these are very distinct on both sides of the larger specimen and 
also well recognizable on the smaller one. They bend inward and 


- 


ON THE OCCURRENCE OF AN APUS 315 


downward at the cervical fold, where as in the recent form they 
open (at the underside of the body). 

The median line of the carapace is marked in its anterior third 
behind the cervical fold by a deep depression, fading out backward, 
which corresponds to the carina seen in Apus cancriformis and 


-some of its congeners. 


The posterior emargination is distinct in both specimens though 
not appearing in the photographs. The original of Figure 2 is 
just sufficiently compressed obliquely to have the emargination 
transferred to the other side. 

General bearing of discovery.—Apus has long been famous in 
paleontologic literature as a primitive phyllopod that on account 
of its great number of simple appendages and other primitive 
features has served well as a model for comparison with extinct 
crustaceans, especially with the trilobites; and again in the case 
of the wonderful middle Cambrian branchiopods discovered by 
Walcott in British Columbia. It is equally famous among zodlo- 
gists for its strange life-cycle as well as its archaic characters. 
Notwithstanding its frequent citation in paleontologic literature, a 
true Apus has only once been found in fossil state. This is the 
A pus antiquus Schimper from the Buntsandstein of the Vogesian 
mountains. This find carried the range of Apus back to the 
Triassic, and the occurrence in Oklahoma extends it now to the 
Permian. 

Apus is thereby made one of the few persistent types that 
have existed from Paleozoic to recent time. Like Limulus it is 
a “‘living fossil.”” Connected with this amazing persistence is 
undoubtedly the strange life-cycle of this creature, as the writer 
will elaborate more fully in another paper. Apus, as typically 
represented by A pus cancriformis, appears only at long intervals, 
usually only after decades of years, during which the eggs were 
buried in the dry mud of roads, ditches, and desiccated pools and 
exposed to heat and cold. It is therefore an extremely rare animal 
and the writer still remembers with thrill, how when still a school- 
boy he one day espied a large specimen crawling on the muddy 
bottom in the water of a swamp along which he was botanizing, 


316 RUDOLF RUEDEMANN 


rushed in, grabbed the strange creature, and ran back at top 
speed to the high-school professor of biology who promptly took it 
away declaring it to be the first he had ever seen alive! It is 
told of Goethe that he had one brought to him while walking once 
near Jena and that he became so excited over the weird animal 
that he offered a thaler for the second, a guilder for the third 
specimen, and so forth, but no other was found. Apus cancri- 
formis grows within two weeks to full size which sometimes is nearly 
five inches, lays an enormous number of eggs and dies. Another 
strange fact connected with this animal is that it produces the 
eggs parthenogenetically. It was fully a hundred years after 
the description of the species that any males were found. The 
seventy pairs of appendages (among them fifty-two pairs of 
abdominal feet with their gill-appendages) which give it its archaic 
appearance are also connected with its life in very temporary 
pools that lack the vegetation which oxygenates the water. 

All these facts show that the animal is adapted to most peculiar, 
and it would seem also most precarious, conditions. It has been 
therefore thought by some, as Salter and Packard,’ to be, together 
with the rest of the Branchiopoda, highly specialized and com- 
paratively modern. Its fossil record, however, contradicts this 
conclusion, and from evidence, which the writer’ in a former paper 
on arrested evolution has brought forward in regard to the factors 
of persistence in animals, it follows that Apus belongs to that 
class of persistent forms, that, at the height of their once vigorous 
development, were able to penetrate into fields where weaker forms 
could not follow them and it is precisely in these outskirts of the 
arena of organic struggle that these types, after they were overtaken 
by more rapidly developing forms, have managed to persist. This 
would suggest that Apus had followed its present mode of life 
through many periods, and the fossil evidence does not contradict 
this suggestion; for the single Apus found in the Buntsandstein 
may well have drifted in, together with the numerous remains of 
landplants (Voltzia, etc.) found in that Triassic terrane. Regard- 
ing the character of the Oklahoma deposit, from which our speci- 

1 Weigand, 1013. 

2 See Schuchert, 1897, p. 675. 3 Ruedemann, 1918. 


ON THE OCCURRENCE OF AN APUS’ 317 


mens came, Professor Beede writes me under date of December 14, 
1921, that it might be brackish water or, possibly, even fresh. 
Leaving out the Apus dubitus Prestwich, described from the 
Coal Measures of England, which, according to Beecher," ‘“‘seems 
to be an abdominal segment or plate of some eurypterid,”’ there is 
evidence of a still older form, at least closely related to Apus, even 
in the Lower Cambrian. This is the well-known Protocaris marshi 
Walcott from the Waucoban (Georgian) of Georgia, Vermont. Its 
similarity to Apus was recognized by Walcott? and has since been 
commented upon by Clarke? and Bernard,‘ the latter author even 
proposing to call the form A pus marshii.s This Cambrian relative 
has also been found in but a single example and therefore was hardly 
a common marine form, as the tribolites collected in the same beds. 
Therefore, even this might have drifted in from the fresh water. 
As to the remaining fossils that by some have been compared 
with Apodidae, we refer to Pompeckj’s excellent summary in the 
chapter ‘“‘Crustacea”’ in the Handworterbuch der Naturwissen- 
schaften.© The remarkable crustacean-fauna discovered by Wal- 
cott in Burgess-Pass contains both Notostraca (Burgessia and 
Naraoia) and Anostraca (as Opabinia, Leancholia, Yohoia), but 
not any forms that are directly referable to Apus, though some 
may, according to Pompeckj, be ancestral to later apodids, as 
Naraoia to the later Carboniferous Dipeltis, which behind the 
parabolic carapace possesses two large thoracic segments or rather 
shields.’ The position of certain finds is considered doubtful, 
as that of a carapace, similar to Apus, described as Lynceztes ornatus 
Goldberg from the Carboniferous of Saarbriicken and now currently 
referred to the Cladocera. There are further to be mentioned the 
problematic, laterally compressed carapaces of Ribeiria Sharpe 
and Ribeirella Schubert and Waagen, which occur in the Ordovician 
and Silurian of Bohemia, Portugal, England,and North America, and 
which were placed by Schubert and Waagen with the Apodidae 


t See Schuchert, 1897, p- 675. 3 1893, Pp. 799. 

21884, p. 50 41894, Pp. 413. 

5Schuchert (1897, p. 674), however, points to the subquadrangular shield and 
frontal emargination as distinguishing characters. 


6 Pompeckj, 1912, p. 789. 7 Schuchert, op. cit. 


318 RUDOLF RUEDEMANN 


and considered as possible ancestors of Apus. Pompeckj holds 
that neither this nor the leptostracan nature of these fossils can 
be proved. We believe that the two muscle-impressions which 
Ribeiria shows on the dorsal median-line’ indicate that in this form 
the carapace was attached to the body in nearly its entire length 
along the dorsal line while in Apus it is free from the cervical fold 
backward, i.e., in its greater portion. This would already indicate 
a very different organism, even if the deep transversal sulcus in 
front of the apex of the casts of Ribeiria could in some way be 
compared with the cervical fold in the carapace of Apus. 


BIBLIOGRAPHY 


1863. SALTER, J. W. “On Peltocaris, a New Genus of Silurian Crustacea,” 
Quar. Jour. Geol. Soc. Lond., Vol. XIX, p. 87. 

1884. Watcorr, C. D. ‘On the Cambrian Faunas of North America; 
Preliminary Studies,” Bull. U.S. Geol. Surv. No. ro. 

1893. CLARKE, J. M. ‘On the Structure of the Carapace in the Devonian 
Crustacean Rhinocaris,” Amer. Nat., p. 799. . 

1894. BERNARD, H. M. ‘The Systematic Position of the Trilobites,” Quar. 
Jour. Geol. Soc. London, Vol. L (1894), p. 411. 

1897. ScHUCHERT, C. ‘On the Fossil Phyllopod Genera, Dipeltis and Proto- 
caris, of the Family Apodidae,” Proc. U.S. Nat. Museum, Vol. XIX, 
Da O7r- 

ro10. PARKER, T. J., and HASweLL, W. A. A Text-book of Zoélogy, Vol. I, 

. London, roto. 

1912. Pomprcxy, J. F. Crustacea, in Handwirterbuch der Naturwissenschaften. 
Jen, Wolk UG ios 772 

1913. WEIGAND, BRUNO. Mitteilung tiber das Auftreten der Limnadia Her- 
manni Ad. Bret. bei Strassburg im September 1912. Mitt. d. Philomath. 
Gesellsch. Elsass-Lothr. Band 4, Heft 5, 20. Jahrg. (1912). 

1918. RUEDEMANN, R. ‘The Paleontology of Arrested Evolution,” New 
York State Mus. Bull. 196, p. 107. 


t Ribeiria compressa Whitfield, Bull. Amer. Mus. Nat. Hist. N.Y., Vol. I, No. 8 
(7886), Pl. 33, Fig. 3. 


PETROLOGICAL ABSTRACTS AND ReEvIEWws 


CAMPBELL, ROBERT. ‘Rocks from Gough Island, South Atlantic,” 
Trans. Roy. Soc. Edinburgh, Vol. L (1914), Part II. PI. 1. 


Describes various igneous rocks collected by the Scottish National 
Antarctic Expedition, 1902-4. 


Ciapp, CHARLES H. “Geology of the Igneous Rocks of Essex 
County, Massachusetts,” U.S. Geol. Surv., Bull. 704 (Wash- 
MMStON O21) Ppsa132) pls: £o, maps 2: 

In this report is given a summary of previous work done in this 
interesting region of alkalic rocks, and considerable new or unpublished 
data are added. Two groups of rocks are recognized, an older sub- 
alkaline group consisting of granites, granodiorites, quartz-diorites, 
gabbro-diorites, and gabbros, and a younger alkaline group consisting of 
alkali-granites, alkali-syenites, and nephelite-syenites, with some diorite, 
diabase, and gabbro. There are many varieties of dike rocks, quartz- 
porphyries, paisanites, sdlvsbergites, tinguaites, diabases, camptonites, 
vogesites, kersantites, minettes, fourchites, quartz-keratophyres, etc. 
The petrographic descriptions are fairly complete, many analyses are 
given, some of the modes are determined, and the structural relations 
are described. Of the essexite of Salem Neck, the author says (pp.124-25): 
“Tt is not a differentiate of the alkaline or nephelite syenite but is 
a contact-metamorphosed gabbro or gabbro-diorite of the Salem type or 
in some places a metamorphosed olivine-bearing diabase. The schistose 
varieties and the more siliceous varieties of essexite, such as those con- 
taining microperthite, considerable nephelite, and large fawn-colored 
augites, are true hybrid rocks.”’ 


CocKFIELD, W. E. ‘‘Sixtymile and Ladue Rivers Area, Yukon,”’ 
Canadian Geol. Surv., 1921. Pp. 60, pls. 6. 

This report, largely stratigraphic and economic, contains general 
descriptions of various schists, amphibolites, granite-gneisses, pegma- 
_ tites, andesites, diabases, rhyolites and granite-porphyries, and various 
ash beds and sediments. 


319 


320 PETROLOGICAL ABSTRACTS AND REVIEWS 


Coxttins, W. H. ‘The Age of the Killarney Granite,” Canadian 
Geol. Surv., Museum Bull. 22 (1916). Pp. 12, pl. 1, map 1. 
The Huronian formations along the coast of Lake Huron have been 
greatly folded and faulted, and intruded by granite batholiths. Both 
disturbance and granite invasion were completed long before Ordovician 
time. 


Cotony, R. J. ‘‘Petrographic Study of Portland Cement,” School 
of Mines Quart., Vol. XXXVI (1914), pp. I-21. Figs. 14, 
bibliog. 14 items. 

Gives petrographic criteria by which the fitness of cement and 
concrete may be judged. 


Daty, Recinatp A. ‘‘Petrography of the Pacific Islands,” Bull. 
Geol. Soc. Amer., Vol. XXVII (1916), pp. 325-44. Bibliog. 
58 items. 

Concludes that underneath the Pacific Ocean the only primary 
magma is basalte, that pyroxene andesite and picrite are direct differ- 
entiates from it, and that the alkaline rocks may possibly be due to the 
solution of small proportions of limestone. ‘There is a nine-page alpha- 
betic list of the various islands with the different rock types occurring 
on each, and a list giving the number of islands from which the various 
rocks have been reported. 


Daty, Recinatp A. ‘The Geology of Pigeon Point, Minnesota,” 
Amer. Jour. Sci., Vol. XLIII (1917), pp. 423-48. Figs. 5. 
Believes that the “red rock” originated through both assimilation 
and differentiation rather than through the differentiation of a wholly 
primary magma, but says a final decision concerning its origin must for 
the present be delayed. 


Daty, Recinatp A. “ Metamorphism and Its Phases,”’ Bull. Geol. 
Soc. Amer., Vol. XXVIII (1917), pp. 375-418. 
Proposes to classify metamorphic processes as follows: 
A. Regional metamorphism. 
1. Static metamorphism. 
a) Stato-hydral or hydrometamorphism (low temperature). 
6) Stato-thermal or load metamorphism (high temperature). 


PETROLOGICAL ABSTRACTS AND REVIEWS 321 


2. Dynamic metamorphism. 
a) Dynamo-hydral or slaty (?) metamorphism (low temperature). 
b) Dynamo-thermal or friction (?) metamorphism (high temper- 

ature) 
3. Dynamo-static metamorphism. 
B. Local metamorphism. ; 
1. Contact metamorphism. 
2. Load-contact metamorphism. 


DRESSER, JOHN A. ‘“Granitic Segregations in the Serpentine 
Series of Quebec,” Trans. Roy. Soc. Canada, Vol. XIV (1921), 
Pp. 7-13. 

Granitic dikes are limited to the peridotite-serpentine area of this 
region, and are believed to be local fillings of contraction cracks in the 
cooling peridotite by part of the still liquid acidic magmatic residues. 
Larger irregular masses are believed to be residual segregations from the 
magma of the peridotite and to have been formed im situ by differen- 
tiation. 


Du Torr, AtEx. L. ‘“‘The Karoo Dolerites of South Africa: a 
Study in Hypabyssal Injection,” Trans. Geol. Soc. South 
Africa, Vol. XXIII (1920), pp. 1-42. Figs. 5. 

The uneroded remnants of strata invaded by these South African 
“dolerites” (British usage) cover fully 220,000 square miles, and fully 
half as much more has been removed. The rocks are composed of 
labradorite and augite, with or without olivine, and are of ophitic tex- 
ture. They are almost exclusively confined to beds of the Karoo system, 
and date from the Rhaetic or Lias, Middle Jurassic at latest. The 
intrusions form intersecting dikes and a series of sheets, one above the 
other. The latter are practically horizontal and vary in thickness from 
100 to 3000 feet, although in some cases still thicker sheets are found. 
Thus the curved sheets of the Queenstown district attain a thickness of 
fully 1,500 feet and cover some 4o square miles, and the Ingeli mass is 
Over 3,000 feet in thickness. Metamorphic action is practically confined 
to baking of the sandstones and shales into quartzite and hornstone. 
With the exception of a few isolated and small-scale cases, no signs of 
assimilation of the strata by the invading magma were found. The 
sheets, with the exception of decrease in size of grain at the margins, are 
generally uniform in texture. In a few localities, enstatite partially or 


322 PETROLOGICAL ABSTRACTS AND REVIEWS 


wholly replaces augite, and the rock has a gabbroidal texture; in others 
there are more acid phases, and the rock passes through quartz-dolerites 
and diorites to granophyres and granites. All such departures are 
thought to be due to post-injection into the cooling body, or pre-injection 
into the intercrustal reservoir. ‘The intrusive sills belong to one period 
of injection, and all phases were completed within a relatively short 
time. Disbelieving that the intrusion could have taken place at one 
time, owing to the fact that the slabs of sediment maintained their 
orientation, Du Toit thinks that they were injected progressively from 
the summit downward to the base, a mode of injection which he describes 
as decensional /it-par-lit stoping. 


EGGLESTON, J. W. ‘‘Eruptive Rocks at Cuttingsville, Vermont,” 
Amer. Jour. Sci., Vol. XLV (1918), pp. 377-410. Figs. 5. 


The eruptive body at Cuttingsville, Vermont, is thought to be a 
composite stock, all of the rocks presumably coming from a single magma. 
Essexite was the earliest intrusion, and nordmarkite the last. Horn- 
blende-biotite-syenite, pulaskite, foyaite, and sodalite-nephelite-syenite 
probably came between these two in about the order given. The order 
of intrusion of the dikes is likewise from basic to acid. Essexite- 
porphyries were earliest, some of them perhaps apophyses from the 
essexite body. Aplite came after nordmarkite, and between these there 
were syenite-porphyries and pulaskite-porphyries. There are also dikes 
of tinguaite and of camptonite. The descriptions are not always clear. 
Thus under essexite it is said that the plagioclase ranges from Ab,An, 
to Ab,An,, but it is not stated whether this is zonal, whether two different 
plagioclases occur in the same section (!), or whether the plagioclase 
differs in different parts of the area. A system of classification unknown 
to the reviewer is used, for the statement is made that “the abundant 
plagioclase allies it [the so-called syenite] to monzonite, but the ratio of 
dark silicates to feldspar warrants the designation of the rock as a 
hornblende-biotite-syenite.” When it is stated that the “pyroxene is 
next to the feldspars in abundance,” one imagines that it must be present 
in approximately the same quantity, yet the amount of feldspar is given 
as go per cent, sodalite as 3 per cent, and nephelite 3 per cent, leaving 
4 per cent to be divided among pyroxene, biotite, apatite, magnetite, 
titanite, and possibly pyrite. The norm shows 2.35 per cent of corun- 
dum, but no feldspathoid. With five good rock analyses and an analysis 
of the hornblende, more carefully modal percentages would have been 
very instructive. 


PETROLOGICAL ABSTRACTS AND REVIEWS 323 


Erret, W. “Uber das Vorkommen von Zinkblende im Basalt des 
Biihls bei Cassel,” Centralbl. f. Min., etc., 1920, pp. 273-85. 
Figs. 6. 

Describes the occurrence of zinc blende in basalt. At an unknown 
depth, the erupting basalt broke through dikes of blende with a little 
pyrite and much quartz. Included fragments of the dike rock were 
assimilated and recrystallized. 


ErreEL, W. “Bemerkungen zu einer Untersuchung von Lewkonja 
tiber die von Hornstein im Basalt des Biihls bei Kassel gefun- 
denen Eisenknollen,’ Senckenbergiana, Vol. II (1920), pp. 
1365338 igs. 2. 

Native iron, from the basalt of the Biihl, near Cassel, is described. 


EirEL, W. ‘“Bemerkungen zu chemischen Untersuchungen des 
Herrn F. Flade iiber das Eisenvorkommen im Biihl bei Cassel,” 
Senckenbergiana, Vol. II (1920), pp. 158-63. 

Gives an early (1909 2?) chemical analysis of the Bihl native iron. 


E1rEL, W. ‘Studien iiber die Genesis der Einschliisse des Biihl- 
basaltes,” Abhandl. d. Senckenbergischen Naturforsch. Gesellsch., 
Vol. XXXVII (1920), pp. 139-76. Figs. 20. 


Among the separate studies in this paper are the following: On 
pseudomorphs of pyrrhotite after pyrite in the Bihl basalt; on the 
origin of the magnetite inclusions in the Biihl basalt; a comparative 
study of the native iron from Ovifak and Biihl; on the genetic relation- 
ship of the native iron to the inclusions of pyrrhotite and magnetite; 
the relationship between the strata penetrated by the Bihl basalt and 
the inclusions in the latter; experimental studies on the formation of 
pyrrhotite from pyrite at high temperatures; on the occurrence of a 
sillimanite-graphite rock with pseudomorphs of spinel, rutile, and 
enstatite after garnet as an inclusion in the Biihl basalt; and the gas 
reactions in the Biihl basalt and their rdle in the origin of the native iron. 


EsKoLA, PENTTI. ‘‘The Mineral Facies of Rocks,” Norsk, geol. 
tidskrift, Vol. VI (1920), pp. 143-94. 
Metamorphic rocks, by recrystallization, may arrive at a state of 
chemical equilibrium. The term “metamorphic facies” is here used to 
designate a group of rocks characterized by a definite set of minerals 


324 PETROLOGICAL ABSTRACTS AND REVIEWS 


which were at perfect equilibrium with each other under the conditions 
of their formation. Igneous rocks, likewise, may reach a state of 
equilibrium, and igneous and metamorphic facies may be spoken of 
together as mineral facies. Mineral facies, therefore, comprise all rocks 
which have originated under conditions of temperature and pressure so 
similar that from a definite chemical composition there results the same 
set of minerals, regardless whether formed by primary crystallization 
or by metamorphism. 


EskoLa, Pentti. ‘‘On the Igneous Rocks of Sviatoy Noss in 
Transbaikalia,” Overskit av Finska Vetensk.-Soc. Forhand., 
Vol. LXIII (1920-21), pp. 1-99. Figs. 9. 


This is an abstract from the reports of the Moscowian expedition to 
Transbaikalia in search of radium in 1914. Two great complexes of 
rock were found, a granite-granodiorite mass and the injected crystalline 
schists. The first mass is here and there intersected by lamprophyric, 
aplitic, and pegmatitic dikes, and there are larger aplitic masses. As 
the boundary between the granite and the migmatite is approached, the 
aplitic portions increase, until near the border there is an almost uninter- 
rupted zone of light gray, aplitic granite. Still farther west the inclusions 
of crystalline schist become more numerous and the rock grades into 
migmatite with only occasional veins of granite. Before the intrusion of 
the granite, the crystalline schists were invaded by diorites and gabbros 
which are now metamorphosed and foliated, and cut by the later aplites 
and pegmatites. 

The main igneous complex is called granodiorite, following Iddings, 
although the alkali feldspar and lime soda feldspar are in nearly equal 
amounts and therefore, as Eskola clearly recognizes, more rationally 
quartz-monzonite. The complete mode is not given. The aplitic 
border has a calculated mode which probably closely approximates the 
true composition. According to the reviewer’s classification it is 227’ 
(new form), consequently a granodiorite-aplite, the plagioclase being 
almost twice as abundant as the microcline. Two types of “‘sviatonos- 
site” are described, one aplitic, but otherwise the two are of practically 
the same composition. They contain 25-31 per cent potash feldspar, 
45-48 per cent oligoclase to andesine, 9-10 per cent andradite, and 8-17 
per cent aegirite-augite and accessories. The author says they are 
andradite-syenites, but from the proportions of potash feldspar and 
plagioclase the reviewer would call them andradite-syenodiorites, or, if 
monzonites are considered in the classification, andradite-monzonites. 
They fall just over the line in Family 11” (2211’), very near 11’. 


REVIEWS 


The Copper Deposits of Ray and Miami, Arizona. By FREDERICK 
LESLIE Ransome. U.S. Geological Survey, Professional 
Paper 115. Washington: Government Printing Office, 1o19. 
pEoO. pls 54, index. 


The region described is about 70 miles southeast of the center of 
Arizona, in the mountainous district separating the plateaus of north- 
eastern Arizona from the desert plains of the south. The Ray district 
lies in Pinal County; Miami is 18 miles north-northeast of Ray in Gila 
County, and 4 miles west of Globe, the geology of which was described 
by Ransome in Professional Paper 12. 

A brief description of the earlier mining operations is followed by a 
bibliography and commentary. ‘The stratigraphy of the region is then 
considered. ‘The area is crossed by four ranges of hills and mountains, 
trending roughly northwest and southeast and separated by broad 
valleys which are partly structural in origin. The rock sequence com- 
prises limestone and clastic sediments of pre-Cambrian, Cambrian, 
Devonian, and Carboniferous age, followed by intrusives of Mesozoic 
and Tertiary time, above which lie Tertiary and Quaternary clastics 
and lava flows. ‘The succession does not differ markedly from that of 
the Globe district as presented in the work already referred to. The 
pre-Cambrian rocks are partly sedimentary, partly meta-igneous, as 
shown by chemical analysis and petrographic study. New formations 
are distinguished in the Cambrian, thanks to the better exposures in 
this area. ‘These and the succeeding rocks are described in chronologic 
order, without separating the igneous rocks, as is the common custom. 

Notable igneous intrusions occurred during Mesozoic times, the 
magmas ranging from basaltic to granitic. Andesitic extrusions of 
Cretaceous age are noted; no definite progressive differentiation is 
observed. The early Tertiary, too, was marked by acid and inter- 
mediate extrusions, and these were succeeded in turn by the deposition 
of conglomerates, sandstones, and flows of which the Gila conglomerate 
(provisionally assigned to the Pleistocene) is especially noteworthy; this 
formation is thought to be of alluvial origin and has a thickness of 2,500 
feet; it offers attractive problems to the student of sedimentation. 


325 


326 REVIEWS 


A table shows the correlations of the rocks with the Grand Canyon 
section. 

Structurally the region forms a part of the Great Basin Province. 
The faulting, which is especially prominent in the southerly ranges, is 
of the mosaic type, the displacement being normal for the most part, 
though thrusts are known, possibly attributable to the crowding inci- 
dental to block faulting. Excellent plates illustrate the topographic 
effects of homoclinal structure. Folding is, on the whole, negligible, 
though some of the field relations might be explained on this basis. The 
block faulting is supposed to be an expression of larger movements of 
the same sort, and the great valleys between the four ranges, the Globe 
Hills, the Pina-Mescal Mountains, the Dripping Springs Range, and the 
Tortillas, are all thought to have a tectonic rather than a purely erosive 
origin. 

The ores of Ray and Miami are of the familiar disseminated type, 
the ore bodies being large and of tabular form. The metallization did 
not follow regular or systematic zones of fissuring, but networks of small 
fractures resulting from widespread disturbance of the rocks. The ore 
bodies are undulating, flat-lying masses of irregular horizontal outline 
and variable thickness, and they mostly lack sharp boundaries. The 
ore body constituting the east part of the Miami-Inspiration zone has a 
total length of 5,500 feet and a maximum width of 1,600 feet. The 
average thickness of the Ray ore body is estimated at 120 feet and its 
maximum at 400 feet. The average thickness of the Miami ore zone 
has not been estimated, but is somewhat greater. 

The shapes of the ore bodies have been determined largely by exten- 
sive drilling, and graphs in which the copper assays of drill-hole samples 
from various depths are plotted, are used effectively to bring out the 
demarkation between the three principal zones, (1) the leached zone 
nearest the surface, (2) the zone of sulphide enrichment, and (3) the 
unenriched protore. 

By far the greater part of the ore in both districts is mineralized 
Pinal schist; a relatively small amount is mineralized granite porphyry 
or quartz monzonite porphyry. Drilling at Ray has disclosed important 
bodies of undeveloped ore in diabase. 

The principal metallic minerals of the protore are pyrite and chal- 
copyrite. The protore at Miami appears richer than that at Ray, the 
average of 126 assays under the main Miami ore body being 1.18 per 
cent. At Ray the copper tenor of the protore usually lies between 
0.3 and o.7 per cent. 


REVIEWS B27 


The principal metallic mineral of the sulphide ore is chalcocite, 
and the average tenor of the ore thus far mined by the Ray Consolidated, 
1.7 per cent copper. The average tenor of the ore sent to the Inspiration 
mill in 1916 was about 1.55 per cent copper. 

The oxidized material is of two kinds. ‘The first carries chrysocolla, 
azurite, and malachite, and is in part rich enough to ship to the smelter. 
Such material is relatively little iron-stained and is formed by the 
encroachment of oxidation in nearly pure chalcocite ore. The second 
type of oxidized material is leached of copper; it is mof stained blue or 
green, but is mostly reddish from iron oxide. It is formed by the oxida- 
tion of material carrying considerable pyrite. No criteria were developed 
for recognizing capping over rich ore bodies beyond the observation that 
very red capping is likely to overlie protore or thin, highly pyritic ore. 
Assays of this sort of capping show from a trace to 0.2 per cent copper. 
The average thickness of the oxidized zone at Ray is 250 feet (range 
45 to 600 feet). 

The lower limit of oxidation and the position of the ground-water 
level at the time mining began were far from coincident, the divergence 
being particularly striking at Miami. Drill holes show that enriched 
ore lies partly above and partly below that water level. This suggests 
that enrichment was related, in important measure, to an earlier topog- 
raphy. This conclusion is supported by evidence which shows that 
much faulting has occurred subsequent to enrichment bringing oxidized 
and leached material in some places in lateral contact with rich ore on 
the same level. Rounded fragments of crushed chalcocite ore and 
fragments of oxidized material are found in the gouges of a number of 
these faults. The Pleistocene (?) Gila conglomerate appears to have 
been deposited subsequent to most of the enrichment, for the ores lie 
deepest in fault blocks capped by this conglomerate. 

The deposition of the protores is attributed to thermal solutions 
coming from deep-lying portions of the granite-porphyry magmas, 
probably in early Tertiary times. 

E. S. B. anp C. H. B. 


Gypsum Deposits of the United States. By R. W. Stone. U.S. 
Geological Survey, Bull. No. 697, 1920. 

This bulletin was prepared to take the place of Bulletin No. 223 
published in 1904, on the same subject. Since that time (to 1918) the 
' production of gypsum has increased more than 300 per cent. In 1918 
the production was valued at $11,000,000. 


328 REVIEWS 


Gypsum deposits are classified according to origin, mode of occur- 
rence, relation to the earth’s surface, etc. 

According to mode of occurrence there are the following classes: 
(1) Interbedded deposits alternating with shales, limestones, and sand- 
stones, laid down in seas or lakes. (2) Efflorescent deposits due to 
evaporation of water which has come to the earth’s surface through 
gypsum deposits. The result is gypsite, a finely crystalline form of 
gypsum. (3) Periodic lake deposits due to deposition of gypsum from 
the waters of intermittent lakes. These are granular and crystalline, 
and vary much in size. (4) Gypsum veins due to re-working of deposits 
of the bedded type by ground water, and redeposition in veins. The 
deposits are crystalline, occurring as satinspar or selenite. (5) Gypsum 
dunes due to disintegration of massive gypsum or efflorescent deposits, 
and their transport by the wind. (6) Isoltate crystals and flakes due to 
formation of sulphuric acid from pyrite and the reaction of the latter 
on limestone. 

According to origin the deposits are classified as: (1) surface-water 
deposits; (2) ground-water deposits. 

In age gypsum deposits vary from Silurian to Quaternary. In 1918 
gypsum was produced in eighteen states and in Alaska. New York 
was the largest producer, Iowa second, and Michigan third. 

In New York gypsum has been mined more than one hundred years, 
during which time it is estimated that about ten million tons have been 
produced. The deposits are in the Salina (Silurian) formation. The 
gypsum is in a series of lenses and was deposited by evaporation from an 
inland sea. The future of the industry in this region is difficult to pre- 
dict, since the gypsum beds run under younger rocks to the south and the 
distance they can be profitably mined is uncertain. 

In Iowa there are two areas of gypsum, the area around Fort Dodge 
being the more important. Here the gypsum occurs in bedded deposits 
which overlie the Mississippian and Pennsylvanian unconformably. 
They are thought possibly to be Permian. These beds have been 
worked since 1872. At Centerville gypsum was discovered a few years 
ago at a depth of 500 feet, but is not yet developed. At this place the 
gypsum is in limestone of Mississippian age. 

In Michigan the deposits are in the Michigan formation of the 
Mississippian system, and in the Bass Islands formation, and Salina 
formation of the Silurian. The gypsum is of the massive rock variety, 
and constitutes lenses in shale and limestone. The deposits are almost 
inexhaustible. The chief development is in the central part of the 
southern peninsula at Grand Rapids, and north of Saginaw. 


REVIEWS 329 


In the west production of gypsum is not very great, but a number of 
states, especially Oklahoma, Wyoming, Utah, Nevada, California, 
and New Mex‘co have large reserves. The Wyoming deposits are 
found in the Embar (Permian), Chugwater (Permian or Triassic), and 
Spearfish (Triassic) formations. The deposits are in the Red Beds. 

In California the most promising deposits are south of San Francisco 
Bay, associated with Tertiary and Pleistocene formations. The most 
valuable are of the gypsite variety. Many of them lie on the side of 
knobs or ridges. The deposits are due to the ground water being 
drawn to the surface and evaporating, leaving behind its load of gypsum 
acquired from the underlying rocks. There are two other forms of 
deposits in the California districts, intermittent lake deposits and inter- 
bedded deposits. 

In New Mexico there are large deposits of gypsum which have been 
developed but little. They occur as bedded deposits in the Manzana 
group of the Pennsylvanian and in the Wingate Sandstone of the Jurassic. 
In the top of the latter there is a bed too feet thick. Gypsum also 
occurs as surface crusts due to evaporation at the surface, and as dunes, 
the material being derived from such crusts. The dunes are in the. 
Tulurosa Desert and cover an area of 270 square miles. The gypsum 
from the dunes is used to a slight extent. rP.S 


Geology of Webster County and a Portion of Mingo District, Randolph 
County, South of Valley Fork of Elk River. By Davin B. 
BEcER. West Virginia Geological Survey, 1920. Pp. 682, 
pls. 35, figs. 24, maps 2. 


Webster County lies in the Cumberland Plateau, the westernmost 
division of the Appalachian province. The topography is characterized 
by deep valleys cut into an old peneplain, the relief varying from 500 to 
1,000 feet. The structure is a gently southeastward dipping monocline 
with some minor folds. The stratigraphic range includes beds of the 
Upper Mississippian and Pennsylvanian. The Allegheny, and the 
Kanawha and New River groups of the Pottsville, form the greater part 
of the surface outcrops, the Monongahela having been entirely stripped 
off and the Conemaugh mostly. Mississippian formations, represented 
by the Mauch Chunk shales, of continental origin, and the Green Brier 
limestone, outcrop only in some of the deeper valleys of the county and 
- in the included portion of Randolph County. The latter is remarkable 
for its profusion of marine forms. Devonian beds are known only from 
deep-well records. 


330 REVIEWS 


The area under consideration lies to the southeast of the main proved 
oil and gas belt of the state. Six deep tests showed no oil and but four 
of them gas. It is to be noted, however, that not one of the six was 
drilled on a favorable structure. 

Webster County, although possessing an immense amount of good 
coal, has but little commercial mining and no coke production. There 
are nineteen coal beds that appear to be workable commercially. The 
author estimates that the total recoverable tonnage present is about 
5,144,000,000 tons. 

Other minor resources include the following: (1) unutilized water 
power; (2) iron ore (possibly); (3) clay, not extensively utilized; (4) 
limestone the Hinton member of the Mauch Chunk and the Green Brier 
limestone; and (5) sandstone, suitable for building purposes in both the 
Pennsylvanian and Mauch Chunk. 

The portion of the report devoted to paleontology includes some 
notes by W. Armstrong on invertebrate fossils from the Pottsville 
series. The following contributions are made: 

1. “The Maximum Size of West Virginia Derbyas as Influenced by 
Sediments.” The author concludes that the largest specimens of each 
species are to be found in the light-colored shales and the purer argil- | 
laceous limestones, the smallest in the fine black sediments. 

2. ‘‘An Example of Shell Regeneration in Derbya crassa.” This is 
an instance of abnormal shell growth repairing a probable break in the 
shell during the life of the animal. 

3. “Notes on the Correlation of Certain Fossiliferous Members of 
the Pottsville Series.”’ A discussion of the present uncertain status of 
the question is given. Some faunal lists are included. 

4. “Fossiliferous Shale Beds in the Row’esburg Section.”’ 

5. “Invertebrate Fossils Collected from the Pottsville Series of 
Webster County.” In general the Pottsville of West Virginia shows 
three faunal types, a normal marine type, a restricted marine type, and 
a fresh-water type. Thirty-two species are listed, of which twenty-three 


are described and a number figured. 
A.C. Meck: 


Bulletin No. 36, Illinois State Geological Survey. Yearbook for 1916, 
consisting of administrative report, and economic and geologic 
papers. Pp. 188, figs. 7, pls. 16, tables 33. 

The report consists of four papers, the first of which is the adminis- 

trative report of F. W. De Wolf, state geologist, for 1916. In Part II 

by N. O. Barrett, on the mineral resources of the state in 1916, the 


REVIEWS 331 


author finds that while agriculture is the leading source of wealth, 


mineral industries are gradually gaining in importance. In 1016, 
Illinois ranked behind Pennsylvania and West Virginia only, in the total 
value of mineral production. It ranks first in the country in the pro- 
duction of fluorspar, sand and gravel, and tripoli, third in brick and tile, 
and coal, and fourth in petroleum, limestone, and clay products. An 
extensive bibliography is included. 

Part III, “Clay Deposits Near Mountain Glen, Union County, 
Illinois,” by Stuart St. Clair, describes the occurrence of a deposit of 
clay believed to be superior to the foreign product for the manufacture 
of graphite crucibles and glass pots. It is a bedded deposit, underlain 
by sand, and overlain by sand, gravel, and in some places by an iron- 
cemented conglomerate, the whole being covered by loess to varying 
depths. The base never has been determined by drilling. The sedi- 
mentary origin is evident. The area where it is was part of a great 
Cretaceous-Tertiary embayment. The existence of this deposit has 
been known for many years, but its development was delayed until the 
cutting-off of the importation of German refractory clays by the war. 
Four clay pits are now being worked. 

Part IV, “The Structure of the La Salle Anticline,” by Gilbert H. 
Cady. The La Salle Anticline is an asymmetric fold, extending south 
from La Salle to the oil fields of Crawford and Lawrence counties. It is 
bordered on either side by synclinal troughs, that on the east forming the 
northern part of the Indiana coal basin, and that on the west forming the 
larger and main portion of the Illinois coal fields. There are numerous 
minor structures. The stratigraphic section includes beds from the St. 
Peter sandstone through the Pennsylvanian. Several unconformities 
are described. These are at the base of the Chester and the Pennsylvan- 
ian, between the St. Peter sandstone and Platteville dolomite, between 
the Lower Magnesian limestone and the St. Peter sandstone, and several 
within the Pennsylvanian. The author suggests that there may be 
some possible relation between the anticline and the distribution of the 
areas of dolomization in the Platteville formation. 

Two structural contour maps of the area are given. The key beds 
used in mapping were the top of the St. Peter sandstone and No. 2 
coal of the Pennsylvanian. NG aUNCE 


The White River Badlands. By CitropHus C. O’Harra. South 
Dakota School of Mines, Bull. No. 13. Rapid City, 1920. 


This is a useful volume on the badlands of South Dakota. It out- 
lines the development of knowledge concerning the region, its geology, 


332 REVIEWS 


and paleontology. The volume is abundantly illustrated and both the 
formations and the fossils afford excellent material for this purpose. A 


full bibliography enhances the value of the volume. 
R: D.,38 


Mineral Resources of Michigan for to1t4 and Prior Years. Pre- 
pared under the direction of R. C. ALLEN. With a treatise on 
Michigan copper deposits by R. E. Hore. Michigan Geo- 
logical and Biological Survey, Publication No. 19, 1915. 


Mineral Resources of Michigan for 1917 and Prior Years. Prepared 
under the direction of R. C. AtteEN. Michigan Geological 
and Biological Survey, Publication No. 27, 1918. 


These volumes were not received until late in 1920. The note- 
worthy feature (besides the statistics on the copper and iron industries, 
as well as on the non-metallic minerals) is the presence in the 1914 
number of a 150-page treatise on the Michigan copper deposits, by 
R. E. Hore. This article serves as an excellent summary of existing 
knowledge on these deposits, as well as giving the author’s ideas on the 
subject. Hore believes the native copper is essentially a primary replace- 
ment deposit from solutions (probably carrying the copper as the 
chloride) which accompanied and followed the extrusion of the lavas. 
A feature of the work is the presence of some thirty photomicrographs 


of polished sectons. 
DJ. 


Field Methods in Petroleum Geology. By G. H. Cox, C. L. DAKE, 
and G. A. Murtenpurc. First edition, pp. Xiv+305. 
McGraw-Hill Book Company, Inc., 1921. $4.00. 


Petroleum geologists, particularly those who are lacking in field 
experience, will welcome this book. It treats chiefly of the recognition 
of structural features favorable for the accumulation of petroleum, 
and of map-making and the instruments used in making maps. It includes 
the solution of geologic problems and the making of a geologic report. 
Problems of a “resident geologist” are not included. Graphic solutions 
of geologic problems are also omitted. It is assumed that the reader 
has a knowledge of the fundamental principles of geology and mathe- 
matics, including trigonometry. 

Chapter I contains a description of the large variety of instruments 
used by geologists, and Chapter II outlines instrumental methods in 


~~ 


REVIEWS 338 


general use. Chapters III and IV include a discussion of the surface feat- 
ures which lead to the identification of strata and structural conditions; 
the methods of obtaining and recording geologic data; also the actual 
field procedure from the selection of the field party to the preparation 
of the final reconnaissance or detailed report. 

The statement is made (p. 129) that ‘‘the field work of a petroleum 
geologist is . . . . made up largely of a search for anticlines and terraces, 
and of mapping such areas.” This statement would have been more 
nearly correct a few years ago. 

The book contains a glossary of about four hundred words, such as: 
Algonkian, Carboniferous, Cenozoic, Contours, Dip Slope, Orientation 
Rod, Stadia, Sedimentation, Volcanic Ash. ‘There is also an appendix 
containing tables of natural functions, reductions of stadia observations 
for rod readings of 100, stadia tables for obtaining differences of eleva- 
tions, gradienter table (Stebbinger drum) for determining distances, and 
a number of other tables, including barometric corrections. 

A limp leather binding and pocket size make the book convenient 


for field use. 
W. O. G. 


Lithologic Subsurface Correlation in the ‘Bend Series” of North 
Central Texas. By Marcus I. GotpmMan. U.S. Geological 
Survey, Professional Paper 129 A, 1921, Govt. Printing Office, 
Washington. Pp. 22, pl. 1, fig. 1. 


Since the early work of Hatch, the micro-petrology of sediments 
remained a rather neglected field to which the physiographer has only 
turned now and then in the exceptional instances when there was a 
question whether a certain sand was wind- or water-laid, a field almost 
wholly ignored by the stratigrapher. Now the subsurface lithologic 
correlations in oil fields have assumed economic importance, however, 
interest in the long-neglected subject is revived, and geologists are glad 
to learn of the establishment by the U.S. Geological Survey of a labora- 
tory devoted to the study of sediments. The paper here reviewed 
represents an invaluable addit’on to the technology of petrographic 
correlation from well logs and well samples. 

The problem presented was the correlation of sediments thought to 
be the equivalents of the Smithwick and Marble Falls beds and of a 
part of the Strawn formation of Pennsylvanian age, as well as of the 
Lower Bend Series (Mississippian) in north central Texas. The method 
employed was much like that outlined by Trager (Econ. Geol., XV, 1920); 


334 REVIEWS 


the more exact scheme of analysis described by Trager, however, was 
not resorted to. Depths from the surface are plotted as ordinates in 
ten-foot units and the composition in percentage of argillaceous, arena- 
ceous, flinty, and calcareous sediments, as determined from the log 
samples, are the abscissae; the diagram thus constructed is essentially 
a graphic columnar section. Correlation is based on the relative increase 
or decrease of one type of sediment, rather than on the absolute per- 
centage composition of the bed in question. A mere glance at the plate 
given serves best to illustrate the method used. Flinty sediments are 
taken to be equivalents of calcareous deposits, since the conditions under 
which they originally formed are similar. 

Such ratios established a generalized sequence for the counties of 
Comanche and Palo Pinto (and probably also Eastland and Stephens, 
the intervening counties). This sequence follows: 


( Strawn formation 
True ‘‘ Upper Smithwick” shale 
““Smithwick lime” = True Smithwick 
: “‘Lower Smithwick shale” of the Ranger shale 
Pennsylvanian } rela 

“Black lime” of the Ranger field | 

A succession of limestones and sandy = True Marble 

| limestones, hitherto unnamed ) Falls limestone 


Micsen J Lower Bend limestone 
ISSISSIPP141\ Lower Bend shale 


Unconformity 
Ordovician—Ellenburger limestone 


The entire Pennsylvanian and Mississippian (including the lower part 
of the Strawn) has an average thickness of 1,100 feet. 

No marked unconformities are reported above that which separates 
the Ordovician and Mississippian, though eleven disconformities of 
varying importance, not indicated above, are recognized. Several 
are intraformational. In the case of the disconformities, glauconite and 
phosphate nodules in some cases mark the plane of separation. These 
glauconite grains are coarser and less rounded than those of the thick 
greensands of other horizons and resemble more closely those of the 
New Jersey Cretaceous. The occurrence of sulphides at the horizons 
marking ‘ akinetic’”’ surfaces of maximum base-leveling is another note- 
worthy feature. 

Many other interesting facts are discussed, and points of consider- 
able theoretical significance are brought forward—such as the source of 


REVIEWS 335 


the oil in the north central Texas region, the northeastern source of the 
Bend sediments, and the very great trustworthiness of lithologic corre- 
lation in regions where faunas change slowly and sediments accumulate 
rapidly. 

The one disappointment felt by the reviewer was due to the lack of 
explicitness in dealing with the technology of examinations. For 
novices a careful outline of the things to be sought for in making litho- 


logic subsurface correlations would be helpful. 
Cia Ba PR: 


Report on Mining Operations in the Province of Quebec, 1919. 
Province of Quebec, Canada, Bureau of Mines, 1920. Pp. 160. 


The phenomenal growth in the mineral industries of Quebec is 
evidenced by the increase in the value of her annual mineral output 
from two and a half million dollars in 1900 to nearly twenty-one million 
dollars in 1919. 

While metals contribute to some degree to this output, Quebec’s 
most important resources are non-metallic—asbestos and building- 
materials dominating. 

Asbestos is to be credited with over half of the total mineral output 
by value in 1919, the mines of Quebec constituting the world’s principal 
source of this mineral. The United States is very directly interested in 
this Canadian industry because about 89 per cent of the output comes 
to the United States, mainly in an unmanufactured state, and is there 
fabricated for use in the United States and for shipment abroad. Some 
3 per cent of the Canadian output is exported directly to England and the 
remainder to various other countries of Europe and to Japan. 

-The magnesite industry of Quebec, which came into prominence 
with the cutting off of the German and Austrian imports during the war, 


declined in 19109 to less than half the 1918 tonnage. 
E. S. B. 


Deposits of Iron Ore near Stanford, Montana. By L. G. WESTGATE. 
U.S. Geological Survey, Bull. 715-F, 1920. Pp. 85-92. 

This report describes several bodies of low phosphorus-hematite ore 
in the northern part of the Little Belt Mountains. The deposits are as 
yet undeveloped. Tonnages at two of the best showings are roughly 
estimated at one and one and one-third million tons respectively. 


336 REVIEWS 


The main facts of the occurrence and character of the ore and the associated 
rocks are as follows: 

1. The iron ore occurs in tabular bodies at the contact of the porphyry 
and the Madison limestone. The ore bodies range in width from 5 to 60 feet, 
and average about 20 feet. 

2. The ore is the result of the replacement of the limestone, as shown by 
its much more uneven contact surface against the limestone and by the reten- 
tion here and there in the ore body of the banding of the limestone and of 
parts of the limestone itself. 

3. Where the contact is inclined the hematite is more commonly found 
where the limestone is the footwall. 

4. The ore is a compact gray or reddish-gray hematite. It contains in 
places enough magnetite to make it react to the magnet. It is not to any 
large degree limonitic at the surface. At the one point where any considerable 
depth has been reached (125 feet, on the Snowbird claim) the ore contains a 
little pyrite and chalcopyrite. 

The limestone at the contact with the porphyry is usually altered to a 
yellowish, finely crystalline marble. No contact silicates were seen except a 
small amount of wollastonite in the rock taken from the tunnel on the Snowbird 
claim [pp. 90 and or]. 

Ews. B: 


Gypsum in t919. By R. W. Stonr. Mineral Resources of the 
United States, 1919. Part II, pp. 99-113. 


The gypsum industry in rg19 showed a slight recovery from the low 


level of production touched in 1918. ‘The report gives the usual statis- 


tical data, the only unusual feature being a discussion by Dr. William 
Crocker, professor of plant physiology at the University of Chicago, of 
“ Agricultural Gypsum and Its Uses.” 

Eighty years ago land plaster was one of the most used of fertilizers, and 
there are indications that it will again come into general use. 

There are four main uses of this substance in agriculture: As a source of 
sulphur for alfalfa, red clover, or other crops of high sulphur requirement, and 
for combination with ground-rock phosphate as a substitute for acid phosphate; 
as a preserver of manure; asa soil stimulant; and as an amendment for black 


alkali [p. roo]. ae 


te Peiographival Microscopes 


oNEW. MODELS—WITH ENLARGED FIELD OF VIEW 


These microscopes have been developed 
on the basis of latest suggestions from — 
petrographists. Their novel and modern 
design will prove a ere ae in petro- 
graphical studies. 


Every leading petrographist knows the 
superior optical and mechanical qualities 
of Leitz microscopes and we therefore 
are fully safe if we refrain from enumer- 
ating the many features. 


The variety of models we offer afford 
a suitable equipment Le every specific 
Pequicmens 


% 


The coniplete assortment of our petro-_ 

Sai graphical accessories will interest those 
mre 8 =8§©§©6who desire to add to their present 

: Microscope ‘“S.M.”’ with synchronic equipment. 

rotation of the nicols 


Some of the prominent features of the new model 
Microscopes are: 3 


Be astoved Oban Clutch Changer which ee the objective to be 
permanently centered. 


_ Anastigmatic Tube Analyzer which permits a most crisp definition. 


3. | - Special Iluminating Apparatus, whereby a wider range of application is. 
afforded. 


a Enlarged Microscope Tube which accommodates eyepieces of wide dines 
_ whereby the field of view is increased 50% over other types of instruments. 


Ask for Catalog II-B (F) 


ee, |S NOLS 
SURPASSED - on: EQUALLED 
It!s SUPREME 


60 East /Ot¢St.. 


Fits in case only four inches high and can 
be carried and used anywhere. Has the 
standard keyboard, with four rows of keys 
and. no shifting for figures. Has the auto- 

- maticribbon reverse. Has every feature you 
associate with typewriting at its best. 


Monthly Htaseraseee Magazine 
EDWARD F. BIGELOW, Editor 


Gee 


The extraordinary demand for the Remington 
Portable—which is taxing to the utmost our 
factory capacity—is proof of its leadership, 
in quality and popularity. 

Price, with-case $60 


Subscription, $1.50 per year; single cop ‘Is 
three months’ trial, 25 cents 


. 


Please write for particulars” 


SOUND BEACH _CONNECTICU: 


_ Remington Typewriter Company 
(Incorporated) 


374 Broadway New York 


Branches and Dealers Everywhere 


: Cambringe Gniversity Dress “i 


The Study of Geological Maps. By Gerreune L. Ettes, D.Sc., Fellow of New: 
ham College, Cambridge. Crown 4to. With 7 plates, 64 figures, and a map. Pr 
application. Cambridge Geological Series. 


A Manual of Seismology. By Cuartrs Davison, Sc.D. With 100 ‘lustra 
Demy 8vo. Price on application. 


Petrology for Students: an Introduction to the Study oF Rocks u 


the Microscope. By Atrrep Harker, M.A., F.R.S., F.G. S. Fifth edition, Te 
Crown 8vo. $3.75. 


A Treatise on Crystallography. By W. J. Lewss, M. A. Deny 8vo. $5. 00. 


The Principles of Stratigraphical Geology. By J. E. Mar, Sc. D., RR. S. es 
Crown 8vo. $2.50. 


$2.50. 


Published: by the Cambridge University Press (Engle 
The Macmillan Co., Agents in the United States — 
64 and 66 Fifth Avenue, New York City 


NUMBER 5° 


GEOLOGY 


ee 


A ‘Sem ARTERY, 
EDITED BY 


R, Invertebrate Paleontology é : : ALBERT J OHANNSEN , Petrology 
ROLLIN T. CHAMBERLIN, Dynamic Geology 


"ASSOCIATE EDITORS 


EIKIE, Great Britain pelts JOHN C. BRANNER, Leland Stanford Junior University 
RICHARD A. F. PENROSE, Jr., Philadelphia, Pa. 
WILLIAM H. HOBBS, University of Michigan 
FRANK D. ADAMS, McGill University 
CHARLES K. LEITH University of Wisconsin 
WALLACE W. ATWOOD, Clark University 
WILLIAM H. EMMONS, University of Minnesota 

. ARTHUR L. DAY, Carnegie Institution 


= = = = - - - A. E. CAMERON 337 
UR TRACKS IN HAMILTON COUNTY, TEXAS - - - W.E.-Waammer 354 


- STRATIGRAPHY ALONG THE ROCKY MOUNTAIN TRENCH 
¢ Francis PARKER SHEPARD 361 


. 


oe ee _K. WENTWORTH 377 
3 2 is = E. M. BurwAsH 393 
Ria an = Lye Seg cea rite 401 


- - - = - - - - AIL 


7 UNIVERSITY OF CHICAGO PRESS 
+ CHICAGO, ILLINOIS, U. S.A. 


THE CAMBRIDGE UNIVERSITY PRESS, Lonpon 
“MARUZEN-KABUSHIKI. KAISHA, Toxyo, Osaka, Kyoto, Fukuoka, SENDAI 
: THE MISSION BOOK COMPANY, SHANGHAI _ 2 


ee at s 
Fe 5. = d Boy , en ae ree pave fr ¥ oy 
July-August 1922 0 els si ieee eee a 


EDITED BY . ny 
THOMAS C. CHAMBERLIN AND ROLLIN ai: SALISBURY ; 
9 Ay With the Active Collaborationof = _ 3 4 
STUART WELLER a ALBERT JOHANNS 
Invertebrate Paleontology : ; Fae 

EDSON S. BASTIN ere ROLLIN T. CHAMBE 


Economic Geology . : ae 


The Journal of Geology is published by the University of Chicago at the Universi 
5750 Ellis Avenue, Chicago, Illinois, semi-quarterly, on or about the following dates: I 
May 1, June 15, August 1, September 15, November 1, December 15. {J The subscr 
per year; the price of single copies is 65 cents. Orders for service of less than a half-year 
at the single-copy rate. {| Postage is prepaid by the publishers on all orders from the Un 

“Mexico,“Cuba, Porto Rico, Panama Canal Zone, Republic of Panama, Dominican Repub ic, E 
Bolivia, Colombia, Honduras, Nicaragua, Peru, Hawaiian Islands, Philippine Islands, Guat 
Islands, Shanghai. { Postage is charged extra as follows: for Canada, 25 cents on annual si 
(total $4.25); on single copies 3 cents (total 68 cents); for all other countries in th 
so cents on annual subscriptions (total $4. 50), on single copies 6 cents (total 71 cents). aj 
requested to make all remittances payable to The University of Chicago Press in postal o 
orders or bank drafts. ; Sot ae 

The following are authorized to quote the prices indicated: Ps Ae ee 

For the British Empire: The Cambridge University Press, Fetter Lane, London, E.! 
subscriptions, including postage, £1 5s. each; single copies, including postage, 4S. € bh: 

For Japan and Korea: The Maruzen-Kabushiki-Kaisha, 11 to 16 Nihonbashi Tori Ss 
Tokyo, Japan. Yearly subscriptions, including postage, Yen 9.45 each; single copie 
‘postage, Yen 1.50 each. x peor 

For China: The Mission Book Company, 13 North Szechuen Road, Shanghai. — 
scriptions $4.00; single copies 65 cents, or their equivalents in Chinese money. © Po : 

mailed direct outside of Shanghai, on yearly subscriptions 50 cents, on single copies 6 cer t 

Claims for missing numbers should be made within the month following the regular month 
cation. ‘The publishers expect to supply missing numbers free only when losses have been 
transit, and when the reserve stock will permit. = 

Business correspondence should be addressed to The University of Chicago Press, Chi 

Communications for the editors and manuscripts should be addressed to the Editors xe) 
NAL oF Geotoecy, the University of Chicago, Chicago, Ill. Oe 4S ee 

Contributors are requested to write scientific and proper names with particular care ani 
to follow the form shown in the pages of the Journal. san ee 

Because of the high cost of printing, this Journal has found it necessary to adopt the 1 
q 1. If articles exceeding 25 pages in length are accepted for publication in the Journal, ‘the c 
position beyond 25 pages shall be paid by the author or by the institution which he represents. at 
rates of the University Press. 2. The cost of illustrative material, in excess of $1.00 per p 
article, beyond the cost of type which would have occupied the same space, shall be borne 
{ 3. The cost of corrections in proof in excess of ro per cent of the cost of first com 
charged to the author. § 4. While the cost of printing remains as now, the Journal can 
prints without charge, but reprints will be furnished authors at cost, if ordered in advan 


Entered as second-class matter, March 20, 1893, at the Post office at Chicago, Ill., under the Act of M aI 
Acceptance for mailing at special rate of postage provided for in Section 1103, Act of October 3 


July 15, 1018. 
: PRINTED IN THE U.S.A. 


° 


Among articles to appear in early 
numbers of the Journal of Geology 
are the following: 


The Early Pre-Cambrian Formations of Northern Ontario and 
Northern Manitoba. By E. L. BRuce. 


The Hot Water Supply of the Hot Springs, Arkansas. By Kirk 
BRYAN. 


The Time of Glacial Loess Accumulation in Its Relation to the 
Climatic Implications of the Great Loess Deposits: Did the 
Loess Chiefly Accumulate during Glacial Retreat? By 
STEPHEN 5S. VISHER. 


The Problems of the Anorthosites and Other Monomineral Igneous 
~ Rocks. By F. LoEwINson-LEsSINe. — 


A Devonian Outlier near the Crest of the Ozark Uplift. By Jostau 
Brince and B. E. Cuartes. 


The Nephelite Syenite and Nephelite Porphyry of Beemervill, New 
Jersey. By M. AuroussEau and Henry S. WASHINGTON. 


The Lava Field of the Parana Basin, South America. By CHARLES 
LAURENCE BAKER. 


The Behavior of Inclusions in Igneous Magmas. By N. L. BowEN. 


Ward’s Natural Science Establishmil | 


ROCHESTER, NEW YORK q 


Relief Maps 


This map, 29"x 46", scale 32 miles to the inch, with a vertical exaggeration of 20 to 1, costs 
only $30.00. The same map, hand colored to show the geology as given in United States 
Geological Survey professional paper No. 71, is $125.00. Write for circular G-233 describing our 
relief maps. 


Fossils 


We are better prepared than ever before to furnish index fossils and exhibition material. A few 
of our recent acquisitions follow: 
Cambrian material collected by Dr. Wolcot in British Columbia. 
Rare Ordovician material from New York and Bohemia, such as Aeglina, Proetus, Cryptolithus, 
Illaenus, Cystids, etc. 
Rare Silurian forms from Bohemia such as Ascoceras and Ophidioceras, Cromus, Acidaspis, etc. 
Rare Devonian material from Germany such as Harpes, Cerarges, Goniatites showing sutures, 
Hexacrinus, Cupressocrinus, etc. 
Fine Solenhofen fish, crustaceans and insects, 
Comanchic material from Texas, 
Eocene leaves from Texas and Louisiana, 
Insects in amber. : 


Rock Sections 


We have designed and had made a machine for the preparation of microscopic slides of rock 
sections, and believe we can fill your requirements quickly and in a satisfactory manner. Let 
us send you our special announcement circular. 


i cnn ee EE EEEEIE asa EINE an SSRIS REESE 


WARD’S NATURAL SCIENCE ESTABLISHMENT 


ROCHESTER - - - NEW YORK 


YOUR MAP NEEDS 


can be selected quickly and with assurance of genuine satisfaction from the 
most comprehensive assortment of effective geography ‘‘teaching tools’? ever 
offered. Note below a summary of that part of 


**The Johnston-Nystrom Line’’ 


applying particularly to geography and geology teaching. 


POLITICAL GEOGRAPHY 
Political Maps—Ten Series comprising 136 
different maps. 


Blackboard Outline Maps in two sizes. 
All continents, U.S., World, and many 
states. 


Desk Outline Maps (for pupils) two sizes. 
Wall Outline Maps. 

State Maps. 

Globes—Political, in four sizes. 
Globes—Slated, in four sizes. 
Atlases—Political and Commercial. 


COMMERCIAL GEOGRAPHY 


Finch Series for United States—34 maps 
on 10 plates, 50’ 38”. Shows Products, 
Industries, etc. 


World Maps (Trade and Industry). 
Atlases. 


PHYSICAL GEOGRAPHY 


Physical Maps—Five Series consisting of 
82 different maps. 


Rainfall Maps, U.S., World and Continents. 


PHYSICAL GEOGRAPHY (Continued) 
Vegetation Maps, Continents and World. 
Pressure and Winds Map. 

Major Natural Regions Map. 
Thermal Regions Map. 
Structure Map. 
Timber Distribution Maps. 
Desk Outline Relief Maps. 
U. S. Outline Map with Topography. 
Graph Charts (slated cloth) two sizes. 
Globes—Physical, 15” diameter. 
Globes—Political. 
Globes—Slated. 
Tellurians. 
Atlases. 

PICTURES 
Land Forms, 24 Illustrations. 
Physical Phenomena, 16 pictures. 
Plants and Trees of Commerce, 30 pictures. 
Types of Nations, rz pictures. 


Geographical Types and Scenes: 
series, 70 pictures. 


two 


Tell us what subject you are interested in and we will gladly give you information 
about any item in our line that may meet your needs. 


USE THIS CONVENIENT COUPON TODAY 


I am interested in the subject checked below and request literature describing your 


maps, etc., applying to the same. 
O Political Geography 

- O Commercial Geography 
O Physical Geography 


O Other Items as follows: 


Mail A, J. NYSTROM & CO. 


(Go22) 
OO Pictures 
O No. Azzr Geography Catalog 


2249 Calumet Ave, 
CHICAGO 


THE UNITED STATES. 


Size 62X44 inches, Scale 50 miles to one inch 


By A. K. LOBECK 


Department of Geography, University of Wisconsin 


This diagram is intended to be used asa guide in teaching the physiography an| 
geology of the United States. The topographic features are expressed with the idk 
of revealing or suggesting the underground structure, and the general character | 

the underlying rock. [ 


An introductory drill such as having students draw cross sections through tk 
Adirondack uplift, the Cincinnati or Nashville Domes, across the Appalachians i 
different places, through the Black Hills, Central Texas, the Ozarks, Wisconsin an} 
many other regions is followed up with the use of other available material, esp j 


cially state geological maps and the United States Geological Survey topograph 
sheets. 


The student frequently fails to note the significance of a locality with relation { 
the physiographic province or geological region as a whole. This is overcome hb 
requiring the student to locate upon the diagram the position of the topograph| 
quadrangles under consideration. As practically every topographic map publishe 
by the Survey was consulted in the preparation of the Diagram, most of the signif| 
cant features are recognizable in the Diagram too. Seventeen geological proyine: 
are discussed briefly on the map. { 


PRICES 


Student edition (one color) with one-half of U.S. on each side of sheet, size 1X4 
inches. f 


Single copy .......... $1.25 Fifty copies, each......... $1.00 
Ten copies, each...... I.10 One hundred copies, each.. .90 
Two color edition with entire map on same side of sheet, size 62 X44 inches. 
Singlercopysneeneeeee $2.25 Fifty copies, each......... $1.75 
Ten copies, each...... 2.00 One hundred copies, each.. 1.50 


Prices on two color edition, mounted on cloth for permanent class room use, Si2/ 
62X44 inches. 


With Common Rollers at Top and Bottom.................+-- $4.50 
On Spring Roller and Board with Dust Proof Cover........... 7.75 
In’Steel/Spring) Roller @ases 4.0.0. ee eee 8.75 


A. J. NYSTROM & COMPANY 


Publishers 2249 Calumet Ave., Chicag¢ 


Goode’s Base Maps 
and Graphs 


Prepared by 


J. PAUL GOODE 


Professor of Geography in the 
University of Chicago 


A series of outline maps for all classes of work: in every grade from 
the common school to the university, in all branches of study, and 
for private use in applied sciences and the various fields of research. 


The desk maps come in three sizes, from 8X 10% to 21X15 inches, 
all perforated for insertion in a standard notebook cover with page 
8X 10% inches. The larger sizes fold into this cover. Several wall 
outline maps are issued, 48 X 36 inches. 


QOODE'S SERIES OF BASE MAPS AND GRAPHS. THE WORLD ON THE HOMALOGRAPHIC PROJECTION (INTERRUPTED) FOR THE CONTINENTS. NO. f01 HC 


TSS 
S 


ee Se ee 


Y Z 


Miniature of Map No. ror He. The homalographic projection, interrupted to present the continents 
The entire earth’s surface is shown without distortion of area 


The maps are made from the best drawings, printed with a good quality 
of ink on bond paper, and the presswork is of the highest order. Recent 
issues have been made by the photo-lithographic process, thus preserving 
an absolutely smooth surface of paper on which to work. ‘These maps 
-are considered by the many who use them as the best quality of base 
maps on the market. 

Complete list with prices will be sent upon request 


The University of Chicago Press 
5750 Ellis Avenue Chicago, Illinois 


Important Books for Scientists 


A Laboratory Manual for Comparative Vertebrate Anatomy. By Lippre H. 
Hyman. Cloth, $2.50, postpaid $2.62. 


A Laboratory Manual for Elementary Zoélogy. By Linpir H. Hyman. Cloth, 
$1.50, postpaid $1.60. 


A Laboratory Outline of Embryology. With special reference to the Chick and 
the Pig. (Third edition, revised.) By FRANK R. Lititz AND Cart R, 
Moore. Paper, 50 cents, postpaid 53 cents. 


A Laboratory Guide in Bacteriology. By Paut G. Hetmnemann. Cloth, $12.75, 


postpaid $1.85. 

The Anatomy of Woody Plants. By E. C. Jerrrey. Cloth, $4.75, postpaid 
$4.87. 

Plant Genetics. By Jonn M. CouLTer and MERLE C. CouttEr. Cloth, $1.50, 
postpaid $1.60. 


Morphology of Gymnosperms. (Revised edition.) By JoHn M. CouLTer and 
CHARLES J. CHAMBERLAIN. Cloth, $5.00, postpaid $5.12. 


Methods in Plant Histology. By CHARLES J. CHAMBERLAIN. Cloth, $3.25, 


postpaid $3.37. 
Artificial Parthenogenesis and Fertilization. By Jacques Lors. Cloth, $2.50, 
postpaid $2.62. 


The Control of Hunger in Health and Disease. By ANTON J. Cartson. Cloth, 
$2.50, postpaid $2.60. 


Essentials for the Microscopical Determination of Rock-Forming Minerals and © 


Rocks. By ALBERT JOHANNSEN. Cloth, $2.00, postpaid $2.15. 


Readings in Evolution, Genetics, and Eugenics. By Horatio Hackett NEw- 
MAN. Cloth, $3.50, postpaid $3.62. ; 


Heredity and Eugenics. By WititAm E. Castle, JoHN M. CouLTER, CHARLES 


B. DAVENPORT, EpwarD M. East, and Witi1am L. Tower. Cloth, $3.00, 


postpaid $3.12. 


The Teaching of General Science. By W. L. EIKENBERRY. Cloth, $2.00, post- 


paid $2.10. 
The University of Chicago Science Series. 


The University of Chicago Nature-Study Series. 


THE UNIVERSITY OF CHICAGO PRESS 
CHICAGO - , : : : - ILLINOIS 


Cambridge Cutversity Press 


The Study of Geological Maps. By Gerrruve L. Exzes, D.Sc., Fellow of Newn- 
ham College, Cambridge. Crown 4to. With 7 plates, 64 figures, and a map. Price on 
application. Cambridge Geological Series. 


A Manual of Seismology. By Cartes Davison, Sc.D. With 100 illustrations. 
Demy 8vo. Price on application. 


Petrology for Students: an Introduction to the Study of Rocks under 
the Microscope. By Atrrep Harker, M.A., F.R.S., F.G.S. Fifth edition, revised. 
Crown 8vo. $3.75. 


A Treatise on Crystallography. By W. J. Lewis, M.A. Demy 8vo. $5.00. 


The Principles of Stratigraphical Geology. By J. E. Marg, Sc.D., F.R.S., F.G.S. 
Crown 8vo. $2.50. 


An Introduction to Geology. By J. E. Marr, Sc.D., F.R.S., F.G.S. Crown 8vo. 
$2.50. 


Published by the Cambridge University Press (England) 
The Macmillan Co., Agents in the United States 
64 and 66 Fifth Avenue, New York City 


The Geography of Illinois 


By DOUGLAS C. RIDGLEY 


Professor of Geography in the Illinois State Normal University 


The author’s aim is to provide an authoritative study of the state, presenting in 
compact form the wide range of physical influences which make up the geographic 
environment in which men live. 


The natural features and natural resources of the state are treated in some detail. 
The great occupations of mankind—agriculture, mining, manufacturing, trans- 
portation, and trade—are discussed with sufficient fulness to give an adequate 
idea of their development and present importance within thestate. ‘The popula- 
tion census of 1920 for Illinois is treated ina final chapter. The book is designed 
to be of interest to the busy citizen who wishes to know his state as a unit in its 
present-day activities; to teachers and pupils who would know Illinois well enough 
to interpret other regions in comparison with the home state; to all who wish to 
learn the reasons for the high rank of Illinois in many lines of human endeavor. 


Abounds with drawings, maps, illustrations, and colored insert maps 
Cloth; $2.50, postpaid $2.60 


THE UNIVERSITY OF CHICAGO PRESS 


CHICAGO - a 


- ILLINOIS 


Essentials for the Microscopical Determination of 


Rock-Forming Minerals and Rocks 


By ALBERT JOHANNSEN 


Professor of Petrology 
The University of Chicago 


A laboratory manual designed to provide students of petrology and geologists 
interested in petrology with information on methods of determining microscopically 
the points of separation between similar minerals and their modes of occurrence. 


This valuable volume contains, in addition to a folding alphabetical list so 
placed that reference may be made to it without turning pages, the following 
important features: ‘‘A Quantitative Mineralogical Classification of Igneous 
Rocks’’; “ASummary of Petrographic Methods”’; ‘“Modes of Occurrence of Various 
Minerals”; ‘Rules for Computing Rocks from Their Modes”; and “Names Pro- 
posed for Various Families.” 

Cloth; $2.00, postpaid $2.10 


THE UNIVERSITY OF CHICAGO PRESS 
CHICAGO” - - ILLINOIS 


Second Edition: Revised and Enlarged 


Source Book for the Economic 
Geography of North America 


By CHARLES C. COLBY 


Associate Professor of Geography 
in the University of Chicago 


The fundamental idea of this book is to make available to the busy teacher the 


material on this subject which is scattered widely through literature. The course which — 


has been given for a number of years at the University of Chicago has resulted in the 
collection and organization of the material most suitable for the purpose. 


The book contains short magazine articles and materials from Canadian, Mexican, 
and United States government publications. The selection of the material has been 
governed by the concept that there is a direct relation between the environment and 
the economic activities of the region. The author aims to survey the different sections 
of the country from the viewpoint of industrial opportunity. 


With chapter and maps on Alaska Cloth, $3.00, postpaid $3.10 


THE ee ae et PRESS 
CHICAGO - . ILLINOIS 


VOLUME XXX NUMBER 5 


Wels 


HOURNAL OF GEOLOGY 


“‘fuly-August I 922 


POST-GLACIAL LAKES IN THE MACKENZIE RIVER 
BASIN, NORTH WEST TERRITORIES, CANADA 


A. E. CAMERON 
University of Alberta, Edmonton, Alberta, Canada 


It is a well-known fact that lakes or ponds of impounded water 
would form whenever the retreating continental ice sheet receded 
down a stream valley. It is therefore evident that the run-off 
from the eastward slopes of the Rocky Mountain cordillera would 
be impounded behind the retreating ice front of the Keewatin 
glacier as it receded, and the lakes so formed would expand laterally 
along the margin of the ice sheet, and the water would collect both 
from the inflowing streams and the melting ice until it rose above 
the lowest point in the stream valley walls, when it would spill 
over and form a new river course, possibly at considerable variance 
to the pre-established drainage lines. Thus we would expect to 
find evidence of ice-dammed lakes of greater or less magnitude 
throughout large areas in the northern portions of the great plains 
area of the Dominion of Canada. It is not the object of this 
paper to deal with the possible extent of such lakes throughout the 
northern regions, but rather to consider one or two specific stages 
in the lake expansions as they were apparently developed in the 
_lower Peace and Athabaska river valleys, Athabaska Lake and 
Great Slave Lake. 

337 


338 A. E. CAMERON 


The area to be discussed is shown in the accompanying map 
(Fig. 1). It lies between latitudes 56° north and 63° north, longi- 
tudes 107° west and 120° west, and it comprises the lower portions 
of the valleys of Peace and Athabaska rivers, Athabaska Lake and 
Great Slave Lake. 

Tyrrell,* in his report on Athabaska Lake and Churchill River, 
makes frequent reference to the great post-glacial lakes of this 
region. He says: 


North of the watershed between Churchill and Stone rivers, most of the 
akes appear to have stood at a higher level than they do at present, in the 
time immediately subsequent to the retirement of the great ice-sheet. The 
natural inference is that they lay between the face of the waning ice-sheet and 
higher land over which the water flowed to form the great rivers of the glacial 
period. 


He proposed the addition of the prefix ‘‘hyper-” to the name of 
the present lake or river to designate the former high-level lake that 
occupied its basin or valley. 

Tyrrell’s report covers the area east of Athabaska River and 
south of Athabaska Lake, and is therefore in part included in the 
region under discussion. Certain parts of his reports are of prime 
importance in any consideration of the question. Particular refer- 
ence must be made to his Hyper-Black Lake, Hyper-Athabaska 
Lake, and Hyper-Churchill Lake. Black Lake lies directly east of 
the eastern extremity of Athabaska Lake and is connected with 
Athabaska Lake by a broad, trenchlike valley at present occupied 
by Stone River. ‘Tyrrell says: 


Hyper-Black Lake stood 125 feet above the present level of Black Lake, 
and extended for a long distance up Cree and Stone rivers. Hyper-Athabasca 
Lake rose above the present level of Lake Athabasca, as is shown by the 
beautiful raised beaches on Beaver-lodge Island, and the wide sandy plains 
seen by Mr. Dowling on William River; but whether it at any time was 
confluent with Hyper-Black Lake was not determined..... Hyper- 
Churchill Lake lay in the present valley of Churchill River, and, when at its 
greatest height, seems to have extended southward as far as the sand-hills 
around Clearwater Lake on the Green Lake trail. 


«J. B. Tyrrell, ‘Athabasca Lake and Churchill River,” Gos. C. Annual Report, 
New Series, Vol. VII, “‘D,” 1895. 


2. + if eon 


POST-GLACIAL LAKES IN CANADA 339 


From a perusal of-Tyrrell’s report it is possible to outline at 
least one main drainage channel of the great post-glacial lakes that 
is of particular importance to our problem. 


Fic. 1.—Map of portion of Mackenzie River basin, showing lower portions of 
Peace and Athabaska river valleys and basins of Athabaska Lake and Great Slave 
Lake. Scale, 1 inch=117 miles. 


This channel appears to connect with the Clearwater River 
valley at Methy portage, following it southeast through Buffalo 
Lake, Isle la Crosse Lake, Beaver River, Green Lake, Big River, and 
‘Shell Creek to the valley of the Saskatchewan. It is reported as a 
broad, trenchlike valley with the watershed between the Saskatche- 


340 A. E. CAMERON 


wan and Churchill rivers lying in a sandy plain at the bottom of the 
deep valley. As will be shown later, a portion of this valley prob- 
ably at one time was the principal drainage channel of great post- 
glacial lakes lying in the Athabaska and Peace river valleys, 
though possibly Hyper-Churchill Lake intervened to take the 
waters from Hyper-Athabaska Lake before discharging them into 
the Saskatchewan River. 

A second channel is suggested in his account of the east end of 
Athabaska Lake and the valley of Stone River below Black Lake. 
This appears as a short, narrow channel connecting the waters of 
Hyper-Athabaska Lake with those of Hyper-Black Lake. It 
would seem very probable that during a certain period of the lake 
formations these two bodies of water were confluent by way of this 
channel, and that the outlet of the chain of lakes lay possibly by 
way of Mudjatick River Valley to Hyper-Churchill Lake or by 
another channel farther east. 

Just at the eastern end of Athabaska Lake Tyrrell reports two 
distinct sets of glacial striae: an earlier one tending south, 65° west, 
parallel to other striae seen almost everywhere along the shore and 
doubtless made by the ice sheet from the northeast; and a later one, 
tending south, 35° west, probably made by a local glacier descending 
from the high land to the north after the greater ice sheet had 
withdrawn. 


A portion of the moraine of this later local glacier may be seen as a great 
stretch of huge broken masses of rock, forming a prominent point, and covering 
the shore for a considerable distance beyond it. .... Athabasca Lake is here 
five miles wide, and lies in a long narrow valley with a steep sandstone escarp- 
ment between 400 and 500 feet high on its south side. The later glacier from 
the north flowed into the valley at this point, and probably reached across to 
the south side, completely filling it and damming up the water from the east 
to the height of the sandstone plain on the south, which is at about the level 
of the high benches previously described on the banks of Cree River and 
along the west shore of Black Lake. The occurrence of an ice dam across the 
valley accounts fully for the former existence of a large lake in the present 
basin of Black Lake. Without the ice dam, or some other dam of which no 
evidence can be found, the water of Black Lake could not have stood much 
above its present level in glacial or post-glacial times, for the great valley of 
Athabasca Lake, which extends eastward to Black Lake, dates back to a period 
long before the glacial epoch. 


POST-GLACIAL LAKES IN CANADA 341 


The formation of this dam is important in the consideration of 
the drainage of the great lakes formed in the Athabaska and 
Peace river valleys to the west. ‘Tyrrell does not seem to appreciate 
the fact that for a long period the northward drainage of the 
Athabaska Lake Valley was blocked, and that consequently the 
waters of Hyper-Athabaska Lake must have stood high and the 
natural outlet of these waters would, at one time at least, be by 
way of Black Lake and the Stone River Valley; and, therefore, 
the damming would probably have more to do with the closing of a 
possible outlet of Hyper-Athabaska Lake than the formation of a 
Hyper-Black Lake. As has been already suggested, the writer 
believes this damming separated the confluent waters of Hyper- 
Athabaska and Hyper-Black lakes, allowing the rapid drainage 
of the smaller Hyper-Black Lake eastward, while the water still 
remained high in Hyper-Athabaska Lake. 


PEACE RIVER VALLEY 


If one stands on the plain level above the town of Peace River, 
an extended view both up and down the valley of the Peace is 
available. The strikingly flat character of the plain level is appar- 
ent, and, if one could extend his vision, he would be struck by the 
similarity of elevation between the level.on which he stands and that 
of the high lands lying to the north and the east. The elevation 
of the plain at Peace River is 2,250 feet above sea-level. That of 
the Watt Mountains is about 2,700 feet, and that of the Eagle 
Mountains is about the same. Caribou Mountain Plateau is 
slightly higher. The summit on the twenty-ninth base line is 
3,225 feet, and to the northward the elevations probably average 
Over 3,500 feet. Buffalo Head Hills and the Birch Hills average 
about 2,700 feet, and the plain level at the end of steel on the 
Alberta and Great Waterways, sixteen miles from McMurray, is 
2,500 feet. This similarity of elevation is conspicuous, and can 
only point to the fact that these present outlying plateaux were 
in recent geological times connected and formed a continuous plain 
of fairly uniform relief. 

At Peace River, the valley of the Peace is both broad and deep, 
and yet it is evidently well filled by the river it contains. Here 


342 A. E. CAMERON 


is no river flowing sluggishly in a valley much too large for itself, 
but a mighty stream which, still in its youth, has carved and is 
still carving a valley in just proportion to its size. 

On descending the river below the town, the valley walls con- 
tract to gorgelike proportions, and the apparent crest of the valley 
gradually lowers. Ascending to the crest at a point some 50 miles 
below Peace River, one would find himself on another plain level 
at a lower elevation than that above Peace River town. 

The plain is narrowed to a width of about 50 miles, but is 
distinctly flat. It is bordered on the east by the shoulder of Buffalo 
Head Hills, and on the west by a southerly extension of the Watt 
Mountain Plateau. The elevation of this plain is about 1,600 feet. 

Descending the river still farther to a point near Battle River 
and there scaling the immediate valley walls, another plain level 
confronts the eye on reaching the top, this time at an elevation of 
1,100 feet. East or west from the river this plain stretches some 
thirty to forty miles before rising gradually to a comparatively 
narrow bench land at an elevation of about 1,600 feet which must ~ 
be crossed before ascent can be made to the old, original plain level. 

About 50 miles above Vermilion the Peace River swings rapidly 
eastward and, the valley walls receding, it enters a widely extended 
plain area at an elevation of about 800 feet which, at Vermilion, 
stands scarcely more than 30 feet above water level. This plain 
level is practically continuous with the basin of Athabaska Lake 
and extends northward around the foot of Caribou Mountain 
Plateau to the basin of the Great Slave Lake. 


WABISKAW AND ATHABASKA RIVERS 


Strikingly similar features to those outlined above are to be 
found in the valleys of the Wabiskaw and Athabaska rivers to 
the east. These are well brought out in the profiles of the base, 
lines and meridians of the Dominion Topographical Surveys as 
shown in the accompanying sketches (Fig. 2). 

The 1,600-foot level shows as the broad plain south of Hay 
River on the sixth meridian; distinctly on the shoulder of Caribou 
Mountains on the twenty-ninth base line; on both slopes of 
Birch Hills on the twenty-seventh base line; on both sides of the 


POST-GLACIAL LAKES IN CANADA 343 


valleys of the Peace and Wabiskaw rivers on the twenty-sixth; 
on the Peace, Athabaska, and Wabiskaw valleys on the twenty-fifth; 
slightly on the Peace and Athabaska rivers on the twenty-fourth 
base line and just faintly on the east side of the Athabaska River 
on the twenty-second base line. 

The 1,100-foot level shows on the northern half of the broad 
plain of Hay River on the sixth meridian; throughout the greater 


eS wo, 
Watt Mt Sad 
2 
7500] 
jo00 
500 
verfical Scale (Feet) 


6° MERIDIAN 
Horizontal Scale pe Ce SE GED fes) 


Sir | Pane wa 36 4g 


Cariboo mts. 


—— ee ee 


2q% BASE LINE | 


25+ BASE LINE 


ci 
Bw. 


3 
Qe 


Wabiskaw Lakes = = | 


220d BASE LINE 


"Fic. 2.—Profiles of base lines and meridians of the Dominion Topographical 
Surveys, Department of Interior. 


part of the twenty-seventh base line; somewhat on the Peace River 
and very pronouncedly on the Athabaska River on the twenty- 
sixth base line; and on the Athabaska River only on the twenty- 
fifth base line. 

The 800-foot level shows between the fourth and fifth meridians 
on the twenty-ninth base line; and in two places on the twenty- 
_ seventh base line—the valley of Athabaska River, and near the 
fifth meridian. 


344 A. E. CAMERON 


HAY AND BUFFALO RIVERS 


A trail leads overland from Vermilion, on Peace River, to Hay 
River. For about 80 miles northwest from Vermilion it traverses 
a flat plain at 
an elevation of 
about 800 feet 
above sea-level. 
It then crosses 
a low divide, the 
summit being at 
1,500 feet, and 
enters the long, 


Fic. 3.—View of Alexandra Falls, Hay River, N.W.T., ae p lain 
looking upstream. (Photo by A. E. Cameron, 1916.) area in, thie 


center of which 
flows Hay River. The elevation of this plain at its upper end is 
about 1,100 feet. Hay River 
follows this plain in a north- 
easterly direction for 150 miles, 
dropping scarcely more than 
two feet to the mile, until, 
within 50 miles of its mouth, it 
falls abruptly over an escarp- 
ment into the basin of Great 
Slave Lake, forming Alexandra 
Falls (Figs. 3 and 4). 


SLAVE RIVER 


At Fort Smith the Slave 
River cuts through a poorly 
developed escarpment some 125 
feet high. The crest of the 
escarpment shows distinct evi- 
dence of the shore-line condi- 
tions in the form of well- 
developed sand dunes and a 
flat horizon to the north and 


Fic. 4.—View of Alexandra Falls, Hay 
; River, N.W.T.,showing close view of escarp- 
east. West of Fort Smith at ment. (Photo by A. E. Cameron, 1916.) 


POST-GLACIAL LAKES IN CANADA 345 


Salt River an abandoned lake basin is distinctly shown in the 
salt plains lying at the foot of an escarpment facing east (Figs. 5 
and 6). The plains are for the most part void of vegetation and 
form a broad flat, water-soaked in the wet season and salt-incrusted 
in the dry. Shore-line conditions are shown by narrow spruce- and 


Fic. 5.—View of the salt plains on Salt River, west of Fort Smith, N.W.T. 
(Photo by A. E. Cameron, 1920.) 


Fic. 6.—Another view of the salt plains on Salt River, west of Fort Smith, 
N.W.T. (Photo by R. T. Hollies, 1920.) 


poplar-clad points jutting out from the irregular face of the escarp- 
ment into the clay flats, bordered frequently by bowlder pavements 
or shingle beaches. Low off-shore islands showing water-deposited 
material are also noticeable (Fig. 7). The Salt River escarpment 


346 A. E. CAMERON 


follows north down the west side of Little Buffalo River for about 
30 miles and then gradually disappears to the west. 


GREAT SLAVE LAKE 


Great Slave Lake lies at an elevation of about 500 feet above 
sea level. The south shore of the lake near Pine Point, west of 


= 


ee he gt li eatin 


Fic. 7.—Small island in salt plains, west of Fort Smith. Salt spring deposits 
in foreground. (Photo by A. E. Cameron, 1920.) 


Fic. 8.—Wave-cut limestone bluff at elevation of 200 feet above Great Slave 
Lake. (Photo by R. T. Hollies, 1920.) 


Resolution, rises rather steeply for a few miles, showing storm- 
built beaches of limestone shingle, to a limestone bluff at an ele- 
vation of about 700 feet (Fig. 8). Inland lies a rolling country 
which rises gradually to the south. Each roll shows limestone 
shingle beaches and bowlder pavements, and not a few are topped 
by sand dunes. | 


POST-GLACIAL LAKES IN CANADA 347 


Each roll may be traced eastward to where it pinches out into 
the hollows between. The trend of these rolls conforms closely 
with the direction of ice movement as established by glacial striae at 
various points along the lake shore. Each roll can only represent 
a point, carved out by the glacier when it was excavating the basin, 
around which the waters of the lake lapped at some stage in its 
development. 


Fic. 9.—View of shingle beach, at elevations up to 150 feet above present lake 
level, Great Slave Lake, N.W.T. (Photo by A. E. Cameron, 1916.) 


Fic. 10.—View of shingle beach, at elevations up to 150 feet above present lake 
level, Great Slave Lake, N.W.T. (Photo by R. T. Hollies, 1920.) 


Elsewhere about the lake shores undoubted lake beaches and 
wavecut cliffs are observable at various elevations above the 
present lake level. They are very numerous and excellently well 
developed, and are noticeable wherever high land exists in the 
vicinity of the present shore line (Figs. 9 and to). 


TERMINAL MORAINES 


Tn the valleys of Hay and Buffalo rivers occur terminal moraines, 
marking positions of the ice front during stages of halt or of slight 


348 A. E. CAMERON 


re-advance of the ice sheet during the general retreat from the 
region. ‘The moraines consist of low ranges of irregularly shaped 
hills, somewhat higher in elevation than the adjacent country, and 
trend in a general direction at right angles to the movement of 
the glacier, as shown by glacial striae. The best-developed moraine 
occurs immediately north of Buffalo Lake. The morainic hills 
here have an elevation of 200 to 300 feet above the surrounding 
country, and form a dam behind which the waters draining from the 
north slopes of Caribou Mountains are ponded, forming the large 
shallow body of water known as Buffalo Lake. This moraine 
extends north to within a few miles of Great Slave Lake. Two 
moraines were noted in the valley of Hay River; one tending to 
connect the Watt Mountain Plateau with that of Eagle Mountains; 
and the other, the Caribou Mountains with Eagle Mountains. 
The low ridge of glacial drift now existing between Watt and 
Caribou mountains and forming the present watershed between 
Peace and Hay rivers appears to be an interlobate moraine formed 
between two lobes of the waning glacier. 


PROBABLE LAKE EXPANSIONS 


From this somewhat scanty evidence an attempt may be made 
to outline the various stages of lake formations developed as the 
continental ice sheet retreated from the region. 

At least three definite glacial lobes are apparent in the area. 
One extended up the valley of Hay River; a second swung west, 
south of the Caribou Mountains, and probably sent tongues up 
the valleys of the Peace and Wabiskaw rivers; while the third 
lay in the basin of Athabaska Lake with its tongue pointing up 
the valley of Athabaska River. 

The first stage to be considered (Fig. 11), is when the water 
level stood at about 1,600 feet. ‘The Hay River lobe extended up 
the valley to a point south of the sixtieth parallel. The edge of 
the other lobes is not determinable from the information at hand, 
but it would appear that the Peace River lobe extended well up 
toward Vermilion and probably sent a tongue south up the valley 
of the Wabiskaw to a point close to the twenty-sixth base line; 
while the Athabaska lobe must have extended at least as far south 
as this same line. The similarity of elevation of the lake benches 


POST-GLACIAL LAKES IN CANADA 349 


in all the valleys can only point to the fact that the waters were in 
conjunction with one another. Those in the Hay and Peace river 
valleys undoubtedly were connected through straits between 
Caribou Mountains and Watt Mountains. The other connections 
were apparently by marginal channels along the ice front. Drain- 
age northward by way of the Mackenzie River was blocked, and it 


Fic. 11.—Outline map showing probable position of Keewatin ice sheet and 
lake expansions when the water stood at the 1,600 foot level. 


seems likely that the lake in the Athabaska Valley was somewhat 
lower in level than the others, and may have been drained eastward 
toward Hudson’s Bay. The more probable outlet, however, was 
by way of the Clearwater River and Methy portage to the Churchill 
River Valley. The summit of Methy portage is 1,735 feet above 
sea-level—scarcely more than 200 feet higher than the lake level 
at that stage. This difference could easily be accounted for by 


350 A. E. CAMERON 


differential elevation of the land following the retreat of the ice 
front. 

In the second stage (Fig. 12) the water level stood at about 
1,100 feet. The Hay River ice had retreated to the morainal 
ridge north of Buffalo Lake and the Peace and Athabaska lobes 
had similarly retreated. Drainage down the Mackenzie was 


Fic. 12.—Outline map showing probable position of Keewatin ice sheet and lake 
expansions when the water stood at about the 1,100 foot level. 


still blocked, the lake basins were all continuous, and outlet must 
have been by way of Fond du Lac on Athabaska Lake, in the long, 
extended arm of that lake, and thus eastward, probably by way 
of Churchill River. 

The Mackenzie River outlet must have been cleared soon after, 
but was apparently blocked farther north, for we find the water 
level standing at 800 feet and developing beaches on all sides of 


POST-GLACIAL LAKES IN CANADA 351 


the present shore line. At the 800-foot elevation (Fig. 13) the 
ice had receded clear of Athabaska Lake and practically clear of 
Great Slave Lake, though a small tongue remnant of the Great 
Slave Lake lobe may have occupied the eastern extension of the 
lake. ‘The water level was well below the Fond du Lac outlet on 
Athabaska Lake, but the Mackenzie Valley was open and drainage 


‘ 
, 
1 
‘4 


6 
=2, ee Ss 
qe . 
Z in 
Q BSRESscss= 
‘ 
ay 
‘ } 
‘ ‘ 
/ " 4 
r ' - 
s 
, 
f 
‘ ‘ ‘ 
‘ AN 
A 1 
‘ 


Fic. 13.—Outline map showing probable position of Keewatin ice sheet and lake 
expansions when the water stood at about the 800 foot level. 


should have been that way. The two main basins are continuous 
by means of a narrow strait across the low escarpment at Smith. 
The Great Slave Lake basin was expanded to take in the present 
basin of Buffalo Lake, though probably low morainal islands 
marked the position of the morainal ridge north of Buffalo Lake. 

On the retreat of the ice, isostatic readjustment of the land 
areas took place, with a raising of the land in a series of differential 


352 A. E. CAMERON 


elevations to the north and east, possibly on successive hinge lines. 
The area south of McMurray was probably differentially raised, 
causing elevation of the Methy portage outlet, and, later, move- | 
ments developed in the neighborhood of Fort Smith, raising a 
land barrier there about 125 feet high, and causing a separation 
of the two basins (Fig. 14). We thus find two large bodies of 


Fic. 14.—Outline map showing probable lake expansion when water stood at the 
700 foot level. 


water existing in the separated basins: Athabaska Lake, standing 
at about 750 feet, and Great Slave Lake at some 600 feet. 

The two basins have acted very differently since their separa- 
tion. Isostatic readjustment north of Athabaska Lake has tended 
to close the outlet, keeping the waters ponded back, and drainage 
has been accomplished by Slave River cutting its channel as fast as 
the ground rose. As a consequence the water level has dropped 


POST-GLACIAL LAKES IN CANADA 353 


only slightly since that time. The level of Athabaska Lake is 
now 699 feet above sea-level at average water conditions. 

Northward, however, conditions are very different. The 
outlet of the basin lies at its western corner, and differential eleva- 
tion to the north and east has only had the effect of spilling more 
water down the Mackenzie River, with the result that the basin has 
been rapidly drained. The rapidity with which the water level 
fell on Great Slave Lake is excellently well shown in the storm or 
seasonal beaches found on the forelands, particularly along the 
north shore. At one place on Windy point over 100 such beaches 
are observable in an elevation of some 250 feet, and the crest of the 
hill is marked by a horseshoe beach developed as the ground rose 
above the water level. The beaches are of fairly uniform depth, 
and throughout most of the rise occur as a series of very regular 
waves, indicating that elevation has been at a constant, uniform 
rate. That movement is still going on is apparent from the beach 
lines built up in recent times. 

In conclusion, the writer can only admit that these are wide 
generalities based on very insufficient data, and undoubtedly 
will bear much correction. He feels sure, however, that the 
outlines suggested are at least approximately correct. Much work 
has yet to be done, and detail work on the problem would yield 
many points of scientific interest. One main problem that presents 
itself is the study of the Great Slave Lake shore lines with a view 
to determining the time period since the last glacial age. Dr. A. P. 
Coleman produced’ much interesting information on this subject 
by a detail study of raised beaches, delta deposits, and lake bars 
on Lake Ontario. The seasonal beaches of Great Slave Lake, 
studied in detail, would give similar information. The Alexandra 
Falls on Hay River have receded 6.5 miles since their original 
development—a distance almost equal to that of Niagara Falls. 
Slave River has completely filled up the long, southward-extending 
arm of Great Slave Lake, and is still building rapidly into the 
present lake. The rate at which alluviation is taking place could 
be readily determined, and an excellent time record seems available. | 


December, 1921 


tA. P. Coleman, Proceedings of the International Geological Congress, Toronto, 
Canada, 1913. 


DINOSAUR TRACKS IN HAMILTON COUNTY, TEXAS 


W. E. WRATHER 
Dallas, Texas 


Attention has been called by E. W. Shuler to the occurrence of 
dinosaur tracks in the Glen Rose limestone of Lower Cretaceous 
age, near the town of Glen Rose, Somervell County, Texas. 
Another interesting occurrence of similar tracks has recently come 
to light in the extreme southern portion of Hamilton County, 
Texas, about sixty miles south-southwest of the first- mentioned 
locality (Fig. 1). 

The tracks in Hamilton County are also in limestone belonging 
to the Glen Rose formation. They are exposed in the bed of 
Cottonwood Creek (Fig. 2), a small headward tributary of Lam- 
pasas River, and are confined to a single stratum of rather soft, 
compact, yellowish limestone about a foot thick, which for a 
distance of probably 800 feet makes the bed of the creek. In the 
Glen Rose locality the tracks were evidently made by one indi- 
vidual moving continuously in the same direction; but in Hamilton 
County they were made by a number of individuals, and the tracks 
point in every direction of the compass, this spot seemingly having 
been a favorite haunt of dinosaurs of every size and presume 
also, of every age. 

The examination upon which these notes are based was made 
hurriedly and without adequate means properly to clear the creek 
bed of the accumulated débris. Normally this portion of the bed 
of Cottonwood Creek is covered with several feet of water, but the 
unusually dry summer season of the past year offered a favorable 
opportunity to examine the footprints, since the creek was free 
from running water; but it developed that cattle which frequent 
the water holes along the creek had worked down shale and gravel 
from the banks in such quantities that almost the entire surface 


* Amer. Jour. Sci., Vol. XLIV (October, 1917), pp. 294-98. 
354 


DINOSAUR TRACKS IN HAMILTON COUNTY, TEXAS 355 


- of the rock was covered with mud or sun-baked soil to a depth 
varying from several inches to as many feet. It was therefore 
impossible to clean off the rock over more than limited areas in 
the short time available. 


Fic. 1.—Sketch map of a portion of Central Texas showing outcrop of Glen Rose 
Formation. (After R. T. Hill.) 1, location of Glen Rose tracks. 2, location of 
Hamilton County tracks. 


A spot was selected at random, and the soil was removed from a 
space about four feet square. Three well-defined tracks were exposed 
within this space, and four others were less clearly recognizable 
(Fig. 3). At six other spots the soil and gravel were removed 


356 W. E. WRATHER 


from patches two to four feet square, and in each instance one or 
more tracks were found. Some of the tracks were shallow, due 
to the abrasion of gravel swept over the bed rock by the swift water 
of freshets, and others were rendered indistinct by superimposed 
tracks. The creek bed, which is floored with the stratum carrying 
the tracks, varies in width from four to twelve feet for a distance 
of about 800 feet. On 
the basis of the writer’s 
observations, he is con- 
vinced that the estimate 
made by residents of the 
neighborhood, placing 
the number of tracks 
at considerably over a 
hundred, is not likely - 
to be extravagant. 
This number could 
quite likely be consider- 
ably increased by clear- 
ing away the slumped 
material along the foot 
of the caving banks. 
Reliable parties who 
have seen the locality 
when the whole expanse 
of rock in the creek had 
been swept clear of its 
covering state that it ex- 
hibits a maze of tracks 
for fully two-thirds of the foregoing distance. Near the lower or 
southern end of the rock exposure, where erosion has cut through 
the level-lying stratum bearing the tracks, a marginal expanse of 
the limestone on either side of the channel shows that the tracks are 
infrequent or entirely absent, but northward they are known to be 
present until the rock disappears under the stream bed. 

The footprints examined varied in length from eight to twenty 
inches. A plaster cast made by Mr. C. B. James, of Hamilton, 


Bets: 


Fic. 2.—Cottonwood Creek along bed of which 
dinosaur tracks are found. (Photo by C. B. James.) 


DINOSAUR TRACKS IN HAMILTON COUNTY, TEXAS 357 


shows the dimensions indicated in Figure 5. This particular track 
was covered with deep mud at the time of the writer’s visit, but 
enough mud and water was baled out to determine that it was only 
one of three large tracks within a space of scarcely more than a 
square yard. 

Near the southern end of the locality, where the tracks are 
infrequent, a single, large, well-formed track three or four inches 
deep and about fifteen inches long was found. Judging the prob- 
able direction of movement from the orientation of the track, 


_ Fic. 3.—Dinosaur tracks in the bed of Cottonwood Creek, Hamilton County, 
Texas. (The tracks were lightly dusted with white powder to secure definition in the 
photograph.) 


it appeared that other tracks of the same individual should be 
present on the uncovered rock. ‘Two saucer-shaped depressions 
were located in the direction of motion, at intervals of about four 
feet, which were apparently vestiges of footprints, though quite 
indistinct. The stride of the animal seems therefore to have been 
about the same as that measured by Shuler at Glen Rose, and the 
size of the tracks in the two localities corresponds quite closely. 
This was the only instance in which the stride of a given individual 
could be measured, though undoubtedly other instances could be 
found if the rock were adequately cleared of débris. 

A correlation of the geologic horizons at which the tracks are 
found in Somervell and Hamilton counties would be particularly 


358 W. E. WRATHER 


interesting, but at this time such a correlation cannot be made with 
any degree of accuracy. Shuler placed the Glen Rose tracks in 
the middle third of the Glen Rose formation which at that locality 
has an approximate total thickness of 315 feet:' In Hamilton 
County the interstream divides are capped with basal Edwards 
limestone containing an abundance of chert. The dinosaur 
tracks are about 200 feet below the lowest chert bed. Immediately 
beneath the Edwards limestone are the soft, chalky beds of the 


Fic. 4.—Dinosaur track of which dimensions are shown in Figure 5. (Track 
dusted with white powder.) (Photo by C. B. James.) 


Walnut formation, eroded to form a wide, shallow valley, and in 
the midst of this valley Cottonwood Creek has cut down slightly 
into the Glen Rose beds. The upper limit of the Glen Rose could 
not readily be determined in the immediate locality, owing to poor 
outcrops, but it is tentatively placed about fifty feet above the 
horizon of the tracks. This tentative correlation is not of much 
assistance, however, as the total thickness of the Glen Rose is no- 
where exposed nearby, and there are no reliable data upon which 
to postulate thickness in this vicinity. The Glen Rose formation 


~R. T. Hill, Twenty-first Annual Report U.S. Geol. Surv., Part VII, p. 153. 


—— 


DINOSAUR TRACKS IN HAMILTON COUNTY, TEXAS 359 


is probably thicker here than in Somervell County, and northward 
it thins out and almost completely disappears northwest of Fort 
Worth, where the Paluxy and Trinity sands, ordinarily separated 
by the Glen Rose, merge into one thick formation known as Antlers 
sand. 


22320 2 


of 


3 to4 aweep. 


(7d 


Fic. 5.—Diagram showing dimensions of track in Figure 4 


It is interesting to note that R. T. Hill in 1886 found dinosaur 
bones in the upper strata of the Basement sands of the Lower 
Cretaceous near Lambert, Parker County," and that he assigned 
the name “Dinosaur sands” to this horizon.? It is reasonably 
certain that the beds in which the bones were found are approxi- 
mately the equivalent of those in which the tracks occur in Somer- 
vell and Hamilton counties. 

The writer expresses no opinion as to whether the tracks in the 
two localities were made by dinosaurs of the same species. The 


t Twenty-first Ann. Rept. U.S. Geol. Surv., Part VII, p. 192. 
2 Amer. Jour. Sci., third series, Vol. XX XIII (April, 1887), p. 208. 


360 W. E. WRATHER 


general shape of the tracks seems to be slightly different. After 
examining one of the Somervell County tracks, now in the museum 
of Southern Methodist University at Dallas, it is noticeable that 
the heel prints shown in Figures 3 and 4 are proportionately longer, 
narrower, and usually much better developed than in the Glen 
Rose tracks. The shape of the heel is shown quite clearly in 
Figure 4. This difference may be due to the fact that the animal, 
when walking, did not press down heavily on the heel, but carried 
the weight thrown forward on the toes. ‘The Glen Rose tracks were 
quite certainly made by an animal in motion, while those in Hamil- 
ton County, with the better development of the heel, may have 
been made in more of a resting position. It would be interesting 
to make a comparison of a number of the Hamilton County tracks, 
to see whether this difference is characteristic of all the tracks 
found there. 

Dinosaur tracks, exclusive of those found in the Texas Cre- 
taceous, have usually been preserved in sandstone which clearly 
indicates littoral deposition. The tracks were evidently made by 
animals walking along a wet, sandy beach or in very shallow 
water. Shuler adequately discussed this problem in the paper 
referred to above, and the writer concurs in the conclusions there 
set forth. The tracks seem to have been made in a soft or plastic 
ooze which was probably covered by several feet of water. This 
“lime mud” was probably deposited in broad, shallow, quiet seas, 
relatively free from currents. There is no noticeable amount of 
sand in the immediately associated strata. Blue-clay shales carrying 
selenite crystals occur for at least fifteen feet above and three or 
four feet below the limestone, and in the overlying shales are thin 
lenses of coquina bearing a typical Glen Rose fauna. 

Mr. C. B. James, of Hamilton, who called the writer’s attention 
to the locality, sent a brief description of the Hamilton County 
tracks, accompanied by photographs, to the Smithsonian Institu- 
tion, and in a reply C. W. Gilmore wrote: 

These are in all probability the footprints of one of the large three-toed 
dinosaurs. Similar footprints have been reported to the authorities of the 
Institution from near Glen Rose, Texas. The fossil remains of an animal 


known as Trachodon, have been found in Cretaceous rocks of Texas, which 
are of sufficient size to have made such tracks as those depicted. 


PROBLEMS IN STRATIGRAPHY ALONG THE ROCKY 
MOUNTAIN TRENCH 


FRANCIS PARKER SHEPARD 
University of Chicago 


In an investigation of the structure of the Rocky Mountain 
trench, from Gateway, Montana, to Golden, British Columbia,’ 
the writer came in contact with some of the stratigraphic problems 
of the Canadian Cordillera. For determination of fossils and for 
helpful criticism he wishes to acknowledge his indebtedness to 
Dr. Stuart Weller. 


STRATIGRAPHIC SERIES ALONG THE ROCKY MOUNTAIN TRENCH 


From the fossils collected at various localities and the sections 
which have been previously made of the trench, it was possible 
to map with some degree of accuracy a large part of the zones 
flanking this great valley (Fig. 1). There are two principal 
series of rocks represented. One of these is dominantly clastic, 
the other dominantly limestone. The former contains many meta- 
morphic varieties of shales, sandstones, and conglomerates, but 
the metamorphism is not extreme. Limestone is not lacking in the 
clastic series, but most of it is in thin bands. (For convenience 
this series will be termed the ‘‘clastic series.’’) 

_ The age of these formations is somewhat problematic. In those 
places where it has been studied hitherto, it has been generally 
considered pre-Cambrian, but fossils found recently in one locality 
have shown part of it at least to be as young as Lower Cambrian.’ 
That the series is older than Upper Cambrian is indicated by the 
following points: (1) In a number of places it was found underlying 
Upper Cambrian formations, and nowhere has it been observed over- 
lying formations containing Paleozoic fossils. (2) It is in general 
metamorphosed more than formations of the limestone series. 


™F. P. Shepard, Jour. Geol., Vol. XXX (1922), p. 130. 
28. J. Schofield, Science, Vol. LIV (1921), p. 666. 


301 


= 


Con Du PWN H 


Tie 


ea Pleistocene 


Mississippian 


Fe Devonian 


Upper Ordovician 
RS Lower Ordovician 


i Upper Cambrian 
ES Pre-Cambrian & Cambrian 


. Beavermouth 

. Golden 

. Harrogate 

. Brisco 

. Sinclair 

. Lake Windermere 
. Columbia Lake 
. Canal Flats 

. Fort Steele 

. Cranbrook 

. Bull River 

. Elko 


th 
—— 


=> 


2, oldel 
=) TREN 


SYSTEM 


PURCELL 
RANGE 


Gateway 


Scale 1”=20 mi. 


Frc. 1.—South portion of the Rocky Mountain Trench 


ROCKY MOUNTAT 


———— 


—= 


! 


~— mm ti aLarseabialh 


STRATIGRAPHY ALONG THE ROCKY MOUNTAIN TRENCH 363 


(3) Fossils were found only in one place, while in the other series 
they were abundant. No attempt has been made to divide or 
correlate the different members of the clastic series, because of the 
great difficulty and the extremely detailed work necessary for 
classification of unfossiliferous formations in a folded region. A 
part of it is probably pre-Cambrian. 

In the limestone series fossils of the following ages have been 
found: Upper Cambrian, Lower Ordovician, lower Richmond, upper 
Richmond, Upper Devonian, and Mississippian. Also the Middle 
Cambrian horizon at Elko might be considered as part of the 
“limestone series.” Other horizons, including portions of the 
Silurian, are possibly present. Disconformities are common, 
showing that there was considerable oscillation of the areas of 
active deposition. Angular unconformities have not been found 
but may occur. 

Stratigraphic sections were made at several localities along the 
trench and some of the more representative have been combined 
into a generalized section (Fig. 2). This section consists chiefly 
of formations found in the vicinity of Sinclair Springs. 

Upper Cambrian and Lower Ordovician.—No stratigraphic break 
has been observed between the Cambrian and Ordovician in the 
northern portion of the trench visited. Fossils were collected 
which have been identified by Dr. Walcott as belonging at the 
top of the Upper Cambrian, and above this is a horizon containing 
a fauna identified by Dr. Raymond as corresponding to the base 
of the Ordovician in Europe.t Walcott identified the following 
species: 


Dicellomus sp. Crepicephalus? fragments of 
Agnostus sp. cephalon 
Agnostus sp. Ptychaspis cf. striata Whitfield 


The species identified by Dr. Raymond included: 


Lingulella moosensis Cyphaspis brevimarginata 
L. allani Hemigyraspis mcconnelli 
Obolus molisonensis H. carbonensis 

Eorthis desmopleura Megalaspis shepardi sp. nov. 
E. sp. (ind.) Dalmanella hamburgensis 


1P. E. Raymond, Am. Jour. Sci., fifth series, Vol. III, p. 204. 


2,500 


2,200 


400 


95° 


1,600-++ 


1,000= 


I,000= 


FRANCIS PARKER SHEPARD 


Massive limestone Mississipian 


Thin-bedded limestone and shale 


Quartzite 


Upper Devonian 


Massive limestone 


Calcareous shales : 
) 


Massive limestone crystalline Silurian or Devonian 
Bowlder bed 
Fairly massive gray limestone 
Massive limestone, gray and black 

Upper Richmond 
Sandstone and quartzite 
Massive limestone Lower Richmond 


Thin-bedded limestone and shales Lowest Ordovician 


Shales and thin-bedded limestone 
Upper Cambrian 
Massive pink weathered limestone 


Thin-bedded li 
‘hin-bedded limestone Uppenor Nidal 


Cambrian 
Massive gray and black limestone 


The Clastic Series 


Conglomerates, sandstones Cambrian and 
2 : Pre-Cambrian 


Shales, and schists 


Fic. 2.—Generalized section along the Rocky Mountain trench 


STRATIGRAPHY ALONG THE ROCKY MOUNTAIN TRENCH 365 


Syntrophia nundina 
Raphistoma nasoni 
Arthrorachis sp. 


Symphysurus cleorus 
S. elongatus 


ind. Menocephalus sp. ind. 
Hystricurus tuberculatus 


According to Raymond this is a typical Ceratopyge fauna, com- 


parable to that at the base of the Ordovician in Europe. 


It has 


been found in several other localities in the west directly overlying 
the Upper Cambrian, and Dr. Raymond thinks that it implies the 
absence of the Ozarkian series from the west. 

Upper Ordovician and Silurian.—Comparing the section made 
by Allen’ along the main line of the Canadian Pacific with that of 
the trench, there are seen to be some resemblances but several 
important differences. 


SECTION ALONG THE CANADIAN PACIFIC 


(After Allen) 


SYSTEM 


Upper Cambrian..... 


Middle Cambrian... . 


Lower Cambrian..... 


FORMATION 


Lower Banff limestone (partly 


Lower Banff shale 
Devonian) 


Sawback limestone (Devonian ?) 


Intermediate limestone 
(thickness, 1,128 m.) 


Halysites beds 


fGraptolite shale 
Goodsir shale 


Ottertail limestone 
Chancellor shales 
{Sherbrooke limestones 
Paget limestones 
Bosworth limestones 


Eldon limestones 
, Stephen limestone-shale 
(Cathedral limestones 


Mt. Whyte sand- 


stone shale Rocky 
St. Piran quartzite + Moun- 
Lake Louise shale tains 


| and sandstone 


THICKNESS 

Feet Meters 
I, 200 366 
T,500 457 
1,800 548 
1,850 563 
I,700 518 
6,040 1,841 
1,725 526 
4,500 1,372 
1,375 419 

360 a Ke) 
1,855 565 
2 PDS) 831 

640 196 
1,595 486 


ry. A. Allen, Can. Geol. Surv. Guidebook No. 8, Part 2, p. 120. 


366 FRANCIS PARKER SHEPARD 


In Allen’s section the Halysites limestone is placed in the Silu- 
rian. This formation, however, contains the same fauna that is 
found in the formation considered upper Richmond in the accom- 
panying section (Fig. 2). The fauna contains the corals which are 
generally typical of the Silurian, namely Halysites catentulatus, Fa- 
vosites, but these are also known to occur in the west in formations 
which are generally considered to be Ordovician,’ as for example in 
the upper portion of the Big Horn limestone. The fauna also in- 
cludes characteristic Upper Ordovician (Richmond) brachipods 
among which a form of Rhynchotrema capax is most common. 
Whether Silurian strata are actually present in the section is open to 
question. In two localities, overlying the upper Richmond and under- 
lying formations of Devonian age, there are several thousand feet of 
limestone and shale in which no fossils were found. In one locality 
the upper Richmond is overlain by a bowlder bed which strongly 
suggests tillite,? and as Silurian tillites have been reported in 
southeastern Alaska? at a similar horizon, it is possible that this 
formation is of the same age. 

Devonian.—The Devonian seas probably invaded the Rocky 
Mountain trench from two directions. An invasion from the 
south by an arm of the Jefferson seas, which covered most of the 
Rocky Mountain region of the United States, is believed to have 
occupied the southern portion of the trench. Formations of this 
age have been found by Daly at the international boundary line* 
and again farther north near Elko and Canal Flats by Schofield. 
The fossil species collected from this formation, which are also 
known to be present in the Jefferson limestone are Spirifer englemani 
and S. utahensis, both of which are guide fossils for the Jefferson. 

Highly fossiliferous Devonian strata were also observed a 
mile east of Harrogate in the Beaverfoot Range. The fauna is 
unlike that of the Jefferson limestone, but is related to that of the 
Devonian formations off the MacKenzie River district to the north, 


1N. H. Darton, Bull. Geol. Soc. of Am., Vol. XV, p. 399. 

2F, P. Shepard, Jour. Geol., Vol. XXX (1922), p. 89. 

3E. Kirk, Am. Jour. Sci., fourth sec. (September, 1918), p. 511. 
4R. A. Daly, Geol. Surv. Can. Mem. 38, p. 115. 

5S. J. Schofield, Geol. Surv. Can. Mem. 76, p. 53. 


STRATIGRAPHY ALONG THE ROCKY MOUNTAIN TRENCH 367 


an occurrence which suggests the extension of an arm of the Mac- 
Kenzie Devonian sea into the northern part of the trench. The 
fossil forms in this fauna which are also known in the MacKenzie 
Devonian include the following species: 


Martinia meristoides Spirifer tulia 

Reticularia fimbriata Atrypa reticularis 
Schizophoria macfarlini Astraespongia hamiltonensis 
S. striatula Heliophyllum halli 


a faunal group which suggests late Middle or early Upper Devonian 
age. 

Whether this invasion from the MacKenzie basin was con- 
temporaneous with that from the south, with the two submergences 
separated by a land barrier in the Lake Windermere district, is 
not established, but the invasion from the north was probably 
somewhat later than that from the south. 

Mississippian.—The Mississippian is known to occur only in 
the southern portion of the trench in the vicinity of Wardner and 
Bull River, where it was discovered by Schofield Dr. Raymond 
identified the following species: 


Camarophoria explanata Cleiothyridna crassicardonalis 
Camaroechia cf. C. metallica Spirifer cf. S. centronatus 
Composita madisonensis Productella cooperensis 


RELATION OF PRE-CAMBRIAN TO CAMBRIAN 


The stratigraphic relation of the Cambrian sediments to the 
pre-Cambrian formations in the Canadian Cordillera has been 
the subject of considerable discussion. Daly considers the two 
series to be essentially conformable, while Walcott and Schofield 
think them unconformable. 

The situation along the main line of the Canadian Pacific is 
well summed up by Daly in his report of the “Reconnaissance 
between Kamloops and Golden.’? The thick lower Cambrian 
fossil-bearing members have at their base a conglomerate (the 
Fairview formation). Below this the Hector Shale is considered 


tS. J. Schofield, Geol. Surv. Can. Mem. 76, p. 57. 
2R. A. Daly, Geol. Surv. Can. Mem. 68, pp. 87-93. 


368 FRANCIS PARKER SHEPARD 


to be pre-Cambrian. Walcott' believes that between these two 
formations there is an unconformity, shown by the following: 
(1) The overlying conglomerates contain fragments of the Hector 
shale. (2) Slight hollows in the Hector shale are filled with thin 
sandstone lenses. (3) The underlying Beltian (Proterozoic) 
formation varies in character at the contact from place to place. 
(4) The sediments seem to change from brackish water in the 
Beltian, to marine in the Cambrian. (5) Allen’s discoveries of 
the contact between the Hector shale and the Fairview formation 
showed apparent conformity in one place, while in two others 
discordance of dip of 4°. 

‘Daly makes the following reply to these arguments: (1) The 
fragments of the Hector shale in the overlying Fairview conglom- 
erate are all angular and do not show a significant time break. 
(2) The Fairview is nearly identical with formations occurring 
at various horizons in the pre-Cambrian. (3) The sandstone 
beds in the upper surface of the Hector are merely lenses which 
are common elsewhere in the Beltian. (4) The discordance of dip 
seen between the Hector and the Fairview can be explained by 
irregularities in uplift as such are seen in continuous sedimentary 
series in many localities. (5) The Selkirk series is divided on 
lithological grounds into Cambrian and pre-Cambrian with the 
dividing plane in the Ross quartzite, where there is no sign of 
unconformity. 

Daly’s first four points appear to be sound. Walcott’s argu- 
ments for an unconformity lack proof that a time break is repre- 
sented. There might well have been a change in the conditions 
of sedimentation without unconformity. In the natural order of 
clastic sedimentation it is to be expected that there will be some 
stirring of older sediments by changes in the activity of the cur- 
rents. In the “clastic series” along the west side of the trench 
many examples of such disturbances are found. It seems that 
more evidence is necessary here to warrant the conclusion of an 
unconformity of importance between the two systems. Minor 
disconformities are of course present in great numbers in most 
thick sedimentary series, especially if there are clastic sediments. 


tC. D, Walcott, Smithsonian Misc. Coll., Vol. LVII, p. 343. 


STRATIGRAPHY ALONG THE ROCKY MOUNTAIN TRENCH 3269 


In regard to Daly’s fifth point concerning the Selkirk series, 
recent evidence shows that the age of the series is very uncertain.* 
As stated above, Daly places the dividing line between the Cam- 
brian and the pre-Cambrian in the Ross quartzite, because of 
lithological resemblances of the sequence of formations in the 
Canadian Rockies and the Selkirks, but the discovery of upper 
Paleozoic fossils in the Laurie formation, which underlies the Ross 
quartzite formation by 12,650 feet and is separated from it by the 
Nakimu limestone and the Cougar quartzite, disproves this con- 
clusion. 

A recent observation further complicates the situation.? The 
Cougar quartzite was found underlying the Upper Cambrian 


Devonian Upper Richmond 


Py SS 


Shaleand Quartzite Massive gray Massive gray Quartz- Shaleand 
thin-bedded limestone limestone ite thin-bedded 
limestone limestone 


Fic. 3.—Section at Harrogate 


shales and limestones on the west side of the trench at Golden, 
and, while a detailed study of the contact was not made, it appears 
probable that there was at least no angular unconformity separating 
the two formations. Therefore one discovery places the Cougar 
formation above upper Paleozoic fossils and the other places it 
below the Upper Cambrian. It seems probable that the series in 
the Purcell Range is not of the same age as that in the Selkirks, a 
conclusion which seems more likely when it is considered that 
the section in the Purcell Range exhibits only a part of two of the 
formations which are present also in the Selkirks (Fig. 3). 


tL. D. Burling, Bull. Geol. Soc. Am., Vol. X XIX, p. 146. 
2F, P. Shepard, Jour. Geol., Vol. XXX (1922), p. 146. 


370 FRANCIS PARKER SHEPARD 


In the southern part of the trench Daly and Schofield are not 
in agreement as to the dividing line between the Cambrian and 
the pre-Cambrian. In Daly’s ‘‘Forty-ninth Parallel Survey” the 
division between the Cambrian and the Beltian was placed within 
the ‘‘clastic series” between the Altyn formation below and the 
Hefty formation above.t This report was based on reconnais- 
sance work in the field, and the division of the unfossiliferous 
formations was necessarily quite arbitrary. Schofield’s more 
detailed work in the Cranbrook area? made changes in Daly’s 
classification and produced more evidence concerning the age of 
the unfossiliferous series. He found Middle Cambrian fossils 
in the Burton formation near Elko, and below this, while there are 
no angular unconformities, there are marked signs of disconformity. 
There is a thin basal conglomerate at the base of the Middle Cam- 
brian, and the surface of the Roosville formation beneath is some- 
what weathered. From these facts Schofield concluded that there 
is an unconformity at the top of the Lower Cambrian, and placed 
the Roosville in the pre-Cambrian. Since the Roosville is the 
highest member of the “clastic series’? (which is represented), this 
places all of that series in the pre-Cambrian. Recently, however, 
Colonel Pollin discovered a remarkably fine trilobite fauna in a 
formation which is probably lower than the Roosville. According 
to Walcott the fossils indicate a horizon at the top of the Lower 
Cambrian. Therefore the extent of the disconformity near Elko 
is minimized, and incidently Daly’s original dividing line between 
the Cambrian and the Beltian is more nearly correct than that 
of Schofield. ! 

However, Schofield has again attempted to find an unconform- 
able relation between the Cambrian and the pre-Cambrian.% 
His evidence for this includes: (1) The thickness of sediments of 
the Siyeh formation between the Purcell lava and the basal con- 
glomerate of the Lower Cambrian, varies from a few feet to 300 
feet. (2) The lithological and metamorphic contrast above and 


rR. A. Daly, Geol. Surv. Can. Mem. 38, p- 79. 
2S. J. Schofield, Geol. Surv. Can. Mem. 76, pp. 41-52. 
3S. J. Schofield, Science, Vol. LIV, p. 666. 


STRATIGRAPHY ALONG THE ROCKY MOUNTAIN TRENCH 371 


below is marked. (3) Basal conglomerates of the Cambrian have 
rounded fragments of the underlying argillites. 

This evidence does not necessarily place the underlying Siyeh 
formation in the pre-Cambrian. That the Olenellus horizon is not 
nearly at the base of the Cambrian is shown by the great thickness 
of Lower Cambrian below the Olenellus horizon in the Waucobian 
of California. Whether the length of the time bridged in this 
disconformity is sufficient to place the underlying formations 
in the Beltian, is open to doubt for the following reasons: (1) The 
degree of metamorphism varies greatly in members of the same 
age in the Purcell series. (2) The difference in the thickness of 
the Siyeh formation is not important, because of its clastic nature. 
(3) The occurrence of the conglomerate does not prove an uncon- 
formity, as conglomerates are common in the “‘clastic series.” 
(4) Since there are great abundance and all varieties of argillites in 
the “‘clastic series” the fragments of argillites in the Cambrian 
conglomerates may have come from argillites other than those 
directly underlying the conglomerate. In general it may be said 
that minor unconformities are to be expected in a series of this 
sort, and their significance is not great. Since the contact of the 
Cambrian on the pre-Cambrian has been especially well studied, 
it is not surprising that minor unconformities should be found at 
some horizon not far below the lowest horizon that is known to be 
fossiliferous. 

From observations on the relation of the “clastic series” to 
the ‘‘limestone series’’ farther north in the trench, new but rather 
incomplete evidence was found of the relation between the older 
series and the Upper Cambrian limestones (the base of these lime- 
stones may be Middle Cambrian). At Premier Lake, below what 
are probably the Upper Cambrian limestones, there is a series of 
argillaceous quartzites and shales. While no accurate measure- 
ment of the series was made, it was estimaetd to be at least 5,000 
feet thick. As the series was examined in several places and 
was found to be quite similar in character in all of them, the condi- 
tion of sedimentation is thought not to have varied in any important 
degree. ‘This series is probably, at least in part, of the same age as 
the Purcell and Galton series farther south. The Purcell series 


372 FRANCIS PARKER SHEPARD 


is only 15 miles to the south in the Fort Steele region, and appears 
to be traceable this far north. The clastic nature of the deposits 
as against the limey deposits of the Upper Cambrian (and perhaps 
Middle Cambrian) indicates that there was variation in the charac- 
ter of the sedimentation in the Cambrian. In the vicinity of 
Parsons the Upper Cambrian is somewhat argillaceous, but the 
base of the Ordivician is mostly limestone. 

If the Beltian is actually present at the base of the ‘‘clastic 
series,’ it seems as though the greatest change during the Beltian 
and Cambrian times was a change from semiterrestrial and near 
shore deposits to deposits in clear broad seas. Probably this 
change is of more importance than any minor unconformities that 
may exist below the lowest rocks which are known to contain 


fossils. 
CORRELATION OF UNFOSSILIFEROUS FORMATIONS 


Correlation of the unfossiliferous formations over wide areas by 
lithological comparison has been resorted to frequently in the 
Cordillera. Dawson, McConnell, and others who made the 
pioneer surveys of the region relied especially on this method. 
Daly, as has been shown above, also correlated formations of 
widely separated portions on the same basis. Mistakes made in 
one of these attempts is illustrated by the recent findings concern- 
ing the Purcell series (p. 369). 

The most detailed work on the unfossiliferous formations is 
that by Schofield in the Cranbrook area. Here the members of 
the Purcell series are mapped over an area of about 2,500 square 
miles. The characteristics of the different formations are such 
that they are likely to vary considerably, and to grade from one into 
another rather readily, so that with the great complication of the 
fault system in the region, a classification is extremely difficult. 

From observations based chiefly on the fossiliferous formations, 
it appears that classification of the unfossiliferous formations is 
especially complicated by the extreme variability of the lithology, 
and of the metamorphic character of the formations along the 
trench. 

Lithological variations.—The lithological character of the differ- 
ent horizons in the Paleozoic was examined at several places along 


STRATIGRAPHY ALONG THE ROCKY MOUNTAIN TRENCH 


Feet 


Mastive gray 
imestone 


Thin bedded 
gray limestone 


[180 | Brown shale | 


Massive black 
limestone 


Gray and white 
bands of limestone 
6 inches thick 


Black shales 


Fairly massive 
gray limestone 


Upper Richmond 
fossils 


Section at 
Fairmont Springs 


Feet 


x 


‘ 
Lower Ordovician \ 
‘\ 


Sinclair Springs 


Calcareous 
shales 


White crystalline 
limestone 


Upper Richmond 
fossils 
Gray massive 
limestone 


Upper Richmond 
fossils 


Black massive 
limestone 


Section at 
Sinclair Springs 


Feet 


\ 


Feet 


Graptolite 


Goodsir shales 


Lower Ordovician 
East of Golden 
(after Allen) 


Devonian 
fossils 


Massive gray 
limestone 


Upper Richmond 


fossils 


Quartzite and 
sandstone 


Thin bedded 
limestone 
and 
shale 


Section at 
Harrogate 


373 


374 FRANCIS PARKER SHEPARD 


the east side of the trench. The change in the character of the 
beds containing the same faunal succession was the more notable 
because the formations were dominantly calcareous. A few sample 
sections (p. 373) will illustrate this point. 

The lithology of a section east of Harrogate (Fig. 4) would 
indicate that there is an anticline or syncline forming the summit 
of the Beaverfoot Range, because of the apparent repetition of 


SelKiak Purcell Mountans Rocky 
TRoontains Moontain 
¢ TRench 


Le 


= Ordovician and Upper Cambrian 

Sir Donald quartzite (Lower Cambrian) 

= Ross quartzite (Lower Cambrian and Beltian) 
= Cougar quartzite (Beltian) 


fa) = 
{R] 


Section along the main line of the Canadian Pacific Railway (after Daly) 


Sel Kiak : 
Movntans Recell Mountains 


= Ross Quartzite (Upper Cambrian ?) 
Cougar Quartzite (Cambrian ?) 


= Ordovician and Upper Cambrian 
TR] 
rom 


Same section revised 


Fic. 4 


the same formations in reverse order on the two sides. However, 
fossils proved that this reversed succession was a mere coincidence, 
and, as shown in the section, one side of the supposed anticline is 
Devonian and the other Ordovician. An occurrence of something 
of this sort in the Cranbrook area, or wherever else fossils were 
lacking, could very easily introduce error into the classification. 
Metamorphic variations—Metamorphic variations are espe- 
cially important in the Purcell Range on the west side of the 
trench. This is perhaps especially true in the Cranbrook area, 
where there are more igneous intrusions than farther north, but 


STRATIGRAPHY ALONG THE ROCKY MOUNTAIN TRENCH 375 


even where the intrusions are absent metamorphism is found to 
vary greatly. In the Selkirk Range along the main line of the 
Canadian Pacific Railway, where there are no intrusives, the 
upper Paleozoic Laurie formation contains the same type of semi- 
metamorphic beds as are found farther east in the lower portions of 
the Cambrian. Many of the upper Paleozoic rocks of the Arrow 
Lake region are more metamorphosed than the pre-Cambrian 
(or early Cambrian) beds to the east in the Cranbrook area. This 
last case is probably explained by contact metamorphism. 

A quartz grit having well-rounded quartz pebbles and a cal- 
careous cement was found in several places along the west side of 
the trench in the vicinity of Lake Windermere. This formation 
was associated with argillaceous rocks which varied from shales to 
schists in the different localities, although it most likely represents 
the same horizon because of its very distinctive characteristics, 
which were not found in any other formation. 

At Sinclair Springs there is a bed of highly crystalline limestone 
which is higher in the stratigraphic series than another limestone 
which is not crystalline. So many similar cases were found that 
it seems as though the anamorphic influences along the trench 
were very irregular. The cause of this irregularity is varied. 
In the southern Purcells, the intrusions are chiefly responsible. 
Farther north the intrusives rarely appear at the surface, but 
their presence below the surface is shown by the hydrothermal 
alteration of the rocks in many places along the west side of the 
trench. In the northern portion, however, the most important 
anamorphic results are connected with diastrophic movements. 
In the vicinity of large faults the alteration is often more pro- 
nounced than elsewhere. The Purcell Range was probably deformed 
at two periods’ so that the formations on that side would tend to be 
differently metamorphosed from those of corresponding age on the 
other side. At present the exposed pre-Cambrian and later forma- 
tions are found at the same general level, but formerly the pre- 
Cambrian was more deeply buried. As deformation went on, the 
pre-Cambrian was brought gradually toward the level of the 
_ younger formations by folding and faulting. After the two reached 


1S. J. Schofield, Geol. Surv. Can. Mem. 76, p. 101. 


376 FRANCIS PARKER SHEPARD 


the same level, metamorphism would tend to be somewhat equal 
in effects on each. Before that time the more deeply buried pre- 
Cambrian would tend to be more severely metamorphosed unless 
it was below the zone in which the most effective compression was 
being concentrated. Further confusion in regard to differential 
metamorphism may be produced by the cutting of great valleys in 
the rising mountain ranges. Along the lines of these valleys the 
overburden is less, and therefore there would be less severe meta- 
morphism. Thus the amount of metamorphism of a local series of 
rocks in this general region can only be used to tell its relative age 
in exceptional cases. 

The foregoing consideration of the varying character of the 
lithology and metamorphism is intended to show how extremely 
difficult it is to correlate unfossiliferous formations in a mountainous 
region. Attempts at such correlations are of course suggestive, 
but they should be duly qualified. 


A SCALE OF GRADE AND CLASS TERMS FOR 
CLASTIC SEDIMENTS? 


CHESTER K. WENTWORTH 
State University of Iowa 


CONTENTS 
INTRODUCTION 


THE GRADE TERMS 
Fragment Terms 
Aggregate Terms 
Rock Terms 


THE CLass TERMS 
INTRODUCTION 

In no other science does the problem of terminology present 
so many difficulties as in geology. With the growth of knowledge 
in any field of investigation, men devise new terms or redefine old 
ones in the attempt to convey more precise and definite ideas. 
In all the branches of science much confusion has followed the 
redefinition of old terms because of the indiscriminate use of the 
terms both in the old and the new senses. But in geology, dif- 
- ficulties of this kind are peculiarly great. 

Because geology is a field science and has followed in the 
footsteps of exploration, it has acquired terms from all parts of the 
world. Many of the names for the less common special features 
have come from the dialect or colloquial speech of that part of the 
world where they are best developed. With the use of these 
terms of geologists of other regions, much irregularity of usage 
and hence much confusion has arisen. 

Since 1917, the writer had been engaged in the study of abrasion 
and shaping of cobbles and pebbles by the action of running water. 
In the course of this study the loose usage of cobble, pebble, and 
related terms (in which his own practice was no exception) has 
impressed him with the need of greater uniformity of usage and 


t Published by permission of the Director of the United States Geological Survey. 
Say 


378 CHESTER K. WENTWORTH 


more careful definition of such terms. With this need in mind, 
he sent to about sixty of his colleagues of the United States Geologi- 
cal Survey a questionnaire asking them to give the limiting dimen- 
sions in their conception or usage of the terms bowlder, cobble, 
pebble, sand grain, and clay particle. Replies were received from 
about thirty of the men. These were studied and compared and 
the composite results presented in preliminary unpublished form 
which was distributed to more than one hundred geologists through- 
out the country in the hope of receiving additional comment and 
criticism. A small number of very helpful replies were received 
and utilized in modifying, to some extent, the size limits and the 
terms used. 

Early in 1921, mimeographed copies of this modified scheme 
of terms were sent to about a dozen geologists in this country and 
England who were known as workers in the field of sediments and 
sedimentary rocks, and deemed competent to criticize the usages 
proposed. They were asked to reply to specific questions in regard 
to the terms which had been subject to the most criticism and to 
comment in general upon the plan. The replies from this smaller 
group were most gratifying, since nearly every geologist addressed 
sent a reply which the writer found useful in the preparation of 
the classification here presented. 

In addition to the studies mentioned above, the writer com- 
menced in 1920 the collection of definitions of sedimentary rock 
terms. These definitions are taken verbatim from textbooks, 
dictionaries, encyclopedias, and glossaries. They are typewritten 
on cards with the proper references and filed under the name of 
the term defined. Many of the definitions collected are from 
sources seventy-five to one hundred years old and represent the 
former usage of certain terms as understood by the compiler. 
The definitions collected in this way vary greatly in value and none 
is to be regarded as of absolute authority. They constitute, how- 
ever, part of the data of the problem. 

As will appear from the foregoing, the writer has compiled the 
present scheme of classification in part from a specific study of the 
terms here presented and in part from the results of a general 
consideration of terms in the field of sedimentary rocks. He is 


GRADE AND CLASS TERMS FOR CLASTIC SEDIMENTS 379 


indebted to a large number of geologists who have helped him by 
spoken and written criticism. Space will permit acknowledgment of 
gratitude only to Dr. M. I. Goldman and to Dr. J. B. Woodworth, 
whose interest and frank criticism have been especially helpful in 
the preparation of this paper. 


THE GRADE TERMS 


It is the writer’s purpose here to suggest terms which are specific 
as regards size of piece and, at least for the larger pieces, as regards 
shape of piece. The terms of this scheme apply to rounded 
materials in so far as materials of the size in question become 
rounded by transportation. Strict uniformity in this regard will 
not fit the sediments as they occur in nature. Bowlders, cobbles, 
and pebbles are rounded rock fragments, whereas most clay 
particles are angular, yet geologists will recognize that they all 
belong to a natural series. Likewise, bowlders and clay particles 
are not commonly of the same mineral composition but in spite 
of this fact they are the two extremes of the series of transported 
rock fragments. By an excessive multiplication of terms it would 
be possible to make a classification in which each term was specific 
as to size of particle, shape of particle, lithologic character, and 
other characteristics. Such a scheme would be highly artificial in 
many of its categories and seems to the writer impracticable in the 
present state of knowledge. 

The present scheme of grade terms is, accordingly, just what 
its name implies—a series of names for clastic fragments of different 
sizes. They apply only to rounded fragments except in the case of 
fine sands, silts, and clays in which even prolonged transportation 
does not always round the pieces. The names applied to the 
different grades carry no lithologic, mineralogic, or chemical 
significance so far as the present scheme is concerned. Sands are 
dominantly quartzose, whereas clays are largely made up of kaolin, 
but this fact is incidental and not necessary in the use of the 
terms. 

FRAGMENT TERMS 


Bowldey—This term is in common use in English-speaking 
countries for rounded and smoothed masses of rock larger than 


380 CHESTER K. WENTWORTH 


cobbles’ resulting: from abrasion in transportation. Angular 
masses of rock of the same size are commonly called blocks or 
slabs. The word bowlder is related to the English word bellow; 
compare Swedish bullra, to ratile or roar. Equivalent terms in 
several other languages carry similar ideas of rumbling or rolling 
in their derivation. 

Cobble-—Cobble or cobblestone is used generally, both by 
geologists, and in common speech, for a rounded stone smaller 
than a bowlder’and larger than a pebble. The term is a diminutive 
of the word cob, meaning a rounded hump or knob, and related to 
the German Kopf, for head. 3 

Pebble.—This term is a very ancient one which is used commonly 
for rounded, transported rock fragments smaller than cobbles. 
In the past it was more commonly used than it is at present for 
rounded stones up to the size of bowlders. The tendency now is 
to use the term cobble in an intermediate sense, as stated above. 
Pebble is from the Anglo-Saxon papol, which meant something 
small and round, perhaps akin to the Latin papula, a pustule. 

Granule2—The term granule is here proposed by the writer 
as a term for rounded rock fragments larger than very coarse sand 
grains but smaller than pebbles. Rounded pieces too small to be 
called pebbles have still been too large to be called sand grains 
in the practice of most geologists. Granule is from the Latin 
granulum, diminutive of granum, grain, meaning a /ziile grain, a 
pellet. In spite of apparent infelicity of meaning (little grain), 
this term was chosen as best adapted for this grade of material. 
The term grit grain was considered for use in this sense, but was 
thought less satisfactory. Grit is used in another sense, as for 
fine sandstone of angular grain. It seemed undesirable to include 
these grains either with small pebbles or with coarse sand grains, 
and it is hoped that the term granule may fill an apparent gap in 
the series of terms heretofore used. 

Sand grain.—The several terms made up by the use of adjectives 
qualifying sand grain are self-explanatory. 

« For explanation of the basis on which the sizes limiting the several grades were 
chosen, see the text which follows. 

2 This term was suggested to the writer by Dr. Herbert A. Baker, of England. 


GRADE AND CLASS TERMS FOR .CLASTIC SEDIMENTS 381 


Silt particle —The term silt particle is here applied to individual 
particles smaller than very fine sand grains but larger than clay 
particles. ‘The term silt from which it was derived was objected 
to by some geologists on grounds that are stated under the heading 
of silt. These grounds were not sustained even by a minor part 
of the data available to the writer and the term is here used as the 
most satisfactory one. 

TABLE I 


THe GRADE TERMS 


The Pieces The Aggregate The Indurated Rock 

Bowlder Bowlder gravel Bowlder conglomerate 
256mm. 

Cobble Cobble gravel Cobble conglomerate 
64 mm. = 

Pebble Pebble gravel Pebble conglomerate 
Sg SS 

Granule Granule gravel Granule conglomerate 
2 a.,§ ——— 

Very coarse sand grain Very coarse sand Very coarse sandstone 
TRE, ee 

Coarse sand grain Coarse sand Coarse sandstone 
Fn 71010 | 

Medium sand grain Medium sand Medium sandstone 
r/4mm. 

Fine sand grain Fine sand Fine sandstone 
(iL. $$ 

Very fine sand grain Very fine sand Very fine sandstone 
TG) LO, 

Silt particle Silt Siltstone 
2/256 a, ———  ————— 

Clay particle Clay Claystone 


- Clay particle—After consideration of several other terms’ for 
the materials finer than silt, the term clay was finally adopted 
as most likely to meet with general approval. Clay particle is 
therefore used for the individual pieces. 

The size limits.—In fixing the limiting sizes of the several 
grades of the scheme shown in the table, the writer has been 
governed by two considerations. First, there is a growing accept- 
ance among geologists and engineers of a series of sieves for the 
classification of natural clastic materials in which the openings of 
consecutive size stand to one another in the ratio 2 or 1/ 2 starting 
with 1 mm. as the standard. | 


382 CHESTER K. WENTWORTH 


It has long been recognized that the differences between two 
consecutive screen size openings should be greater for the large 
sizes than for the small. This principle is followed in the selection 
of such limits as 1, 2, 5, 10, 20 millimeters, making the limits fall 
on convenient whole numbers in the decimal notation. This 
series, however, is a crude approach to a geometrical series in which 
each value bears a constant ratio to the preceding one. A geo- 
metrical series is the ideal for such a purpose, since a change of 
1’ is of the same significance and importance in the size of 10” 
cobbles as a change of ;1,” in the size of 1’’ pebbles. Only by the 
use of logarithmic or some similar graphical scheme of representa- 
tion can the size composition data be shown adequately for great 
size ranges. ‘The use of a geometrical series makes the successive 
grades fall into equal units on the graph—an arrangement much 
easier to read and interpret than any other known to the writer. 
The most convenient ratio for the construction of such a series is 
the ratio 2, and the most convenient and logical starting-point, 
1mm. A large number of mechanical analyses of sediments made 
with screens and by microscopic measurement conforming to such 
a series have been made.’ If a more minute subdivision is needed, 
the ratio // 2 can be used, giving twice the number of grades, or in 
exceptional cases 1/2. These extra subdivisions fit in with and 
form further subdivisions of the fundamental series of the powers 
of 2. Conformity to this geometrical series is the first consideration 
which has guided the writer in fixing the limits between the several 
grade terms. 

The second consideration has been the desire to make each of 
the limits as close as possible to the common practice of the majority 
of geologists. Figure 1 shows the composite opinions of twenty- 
eight geologists of the United States Geological Survey, as reported 
by them in response to a questionnaire on the sizes limiting several 
of the terms. The table below shows a number of different schemes 
of classification which have been published. There is a close agree- 
ment between some of those shown, but, with the exception of that 
of Udden, all lack, in the sizes of successive grades, the uniformity 


tJ. A. Udden, ‘‘Mechanical Composition of Clastic Sediments,” Bull. Geol. Soc. 
Amer., Vol. XXV (1914), pp. 655-744. 


GRADE AND CLASS TERMS FOR CLASTIC SEDIMENTS 383 


of ratio of the geometrical series which seems to the writer to be 
essential to any thorough quantitative study of the mechanical 
composition of sediments. 

Using the data shown in Figure 1 and Table II, the writer 
has selected the limits conforming to the power series of 2 
which most closely conform also to the concensus of the opinions 


Fic. 1 


of authorities quoted there. The names of the several grades 
thus established were then chosen as described above. 


AGGREGATE TERMS 


The assignment of definite limiting size values to the terms for 
the aggregates as named below, as well as the rocks named in 
another part of this paper, will be objected to on the ground that 
aggregates are not made up in nature of one grade or even of a few 
grades, and that therefore the names are inapplicable. The author 


——"ulul ghoz /I—— 
Avo OUT 
——uut ¢zo1 /t 
ArlD Avo WNIpayy 
—— “uur z18/1 
Avo as1B09D 
— “uur 982 /1 —|— "ari gS /1 eID 
VIS ouy AI9A, 
—— “WU gz1/I “WUr I0° 
eA | 
IS —— wu 9/1 
PCSROL BSG WS 
——— wu z£ /T 
}IIS as1e0D “tau So: 
“UIUT O/T pUIC OU qIIS asxe0D 
puvs sug A1aA puvs ouy Ala ‘wn I 
“wu g/T “WU g/I 
pus ouLy pues our Pes SNA 
wu 7/1 “un +/1 “wu Sz" 
purs winipeyy purs wntpayy pues umIpsyy 
“Uru z /T “WU @ /T ——uut §-——___ 
pues asiv0f pues asiz09 pues asiv0p 
“UIUI I “WUT I “WU I 
pues as1v0d AIO, [PAvIs OUT pues asiv0d AI9 A 
“WU Z “UI @ — “WILL @ 
pPAvis anueIDy JeAvis WINIpaPy 
“mu “WU Py 
[PAvId asrv0D PPAeID 
“WU g 
[PAIS 9SIvOD AIDA 
PPAris aqqoq “UIU QT 
SIap[Moq |]ews A193, 
“UU ZE. 
SIOp[Mog [jews - 
— wu $9 ———_ | ___‘urut }g —_—_ 
Siap[Moq wnIpsypy 
pPAvIZ aIGqGoD “UIUT QZ 
SIop[Moq ad1v'T 
“UU OSz “UU QSz 
[PARIS Iop[Mog 
asodoig 919 
Ronwsguss | ffueppavef | dsoxea-y-H 


UOKROgISse[D 


: “phL-SSq “dd 
AXX ‘TOA “42m Y “209 “1005 “TINE ,(SIVEUMIPAS INS¥ID Jo VON SodwO jLoUEYDayL,, ‘ 


4 


‘ausoppsaurpy snf yonqaype sana 
“Aueduloy W Ialag “4 “W ‘HIOK MaNy ‘(€1OI 
“ID 2B JopedeyD “y ‘staeg ‘V9 ‘d ‘suaraun ja yanjov sutsD 
‘oyu pueurpsiey ‘J1v5}4N1S ‘aseyny 932 
*gge ‘d “(£1 


azis Ae[D 


“tur Soo" _ 
amopy yor suygiednsg 


“Mur 10° 


Moy yoy 


“urur So" 
pues sugisdns 
“UU T° 


pues oury 

Apodion Sie —— 

pues wmnipsyy 
“uur S- 

pues asiv09 

“WU I 


4 (@) Ae ysnp 4sourg 
——"wur Soo" —— uu. Soo" 
AUS TEEIEC UL 
—— ‘wut 10° —— ‘wUuI 10° 
US 
HIS isnq 
“uur So" ——_— “urut Sor —— “urur So" 
pues ouy A1a A pues ouy Ai A 
“WU I" “WUT T° pues ourg 
puvs our pues sung 
“Uru Sz" “UIUL Sz" “uu Sz" 
pues wmnIpoyr pues uintpeyy pues uinipoyy 
“Wut S$: “wut S* “WU S* 
purs asie0Z pues asiz0od puvs asiv0d 
——— “Wu I “WUT I “WU I 
[PAvID OUT PPAvIs oUTT 
ae ni pues asiv0o AoA 
“mu £ 
JPArIn 
[PAvID 
S[IoS Jo nvain 
rae a brid “sf ||9920 


#597815 POU) 


pues asivoo Aa A. 

“WU Sz 

[PAvIs OULT 
suru § ———_ 


[PAvIs osIvOD 


“UU Sz. 
[PAeId:9SIM09 AIDA 
“uu oS 
s914q°9 

—— ‘tur o$1 


SIop[Mog 


§neqein “MV 


“rogt “d ‘(z06r) 


ae ee 


pues auyg-auLy 
‘uu Vo" 


pues aug AraA 


———. ‘UIUI gz* 
pues oul 

“uu Sv* ——— 
pues uMIpeyy 


“UIUL Og* ———] 


pues 9SIvO) 


“Ul © 


[PAvIs oul 


frenogy “f 


oSi * 


ing “ang “oop “s"A ‘ota *S “fb 
“SLI 10y aUMJOA ‘18S ‘d 


“SOUAOM SIP ASSQAK) dap YODU sapuDy sap Sunuyaiazag ‘1310 | 

) YFossynys fo saigruiug ‘Meqeiry “MW “YW 4 

u-snos Spuof ap asKjpun.p sgag “jatnoyy, *{ t 

*(g001) az3070a5 uayrsyypsd sap yongayaT ‘xoeUTIOy peru0y { 

61) Aydossyoys fo sajgroutag ‘neqeis “\\ “WY &q pajond) x 
a eee 


azis Ae[d 
ysnp 4souryT 
“wu Soo* 
Inof yo1 augaedns 
——— ‘uIUI Io" “Wur Io" 
ysnq NOP YoY 
“ulur So" “urur So" 
pues sugisdng pues sugisdns 
“UU I" “WU I 
puvs oun 
Haare pues our 
“UU Sz" 
QUES) Wan AL purs umrpoyy 
“uur S° ————"ulur $*-——___. 
pues asiv0D pues asiv09 
“UU I “WUE I 
Pues asIvOD AIOA 
"UIUL Z [eAvis 
aul 
———“uiur $ 
PAPI PPAvID 
asIvoD 
UWOISSIUIUIOD, 
dpPeyIey “W jonpenby 


~AWID YIOR MONT 


I ATaVL 


GRADE AND CLASS TERMS FOR CLASTIC SEDIMENTS 385 


recognizes this difficulty and the grade terms here defined are 
applicable strictly and without modification to but few natural 
sediments. They are proposed as the foundation of, and in the 
course of development of, the several class terms and adjectives 
proposed in the last part of this paper. Photographs of grade 
aggregates from 64 mm. to 3'5 mm. are shown in the accompanying 
Figures 2 and 3. 

Gravel—Among some geologists the term gravel has been used 
only for material composed of small pebbles and granules, but more 
commonly and especially in America and in reference to glacial 
gravels, the term has been used to include material containing 
great bowlders up to a meter or more in diameter, and has recently 
been so defined by J. S. Flett" who considers the term to be the 
equivalent of conglomerate as applied to the consolidated rocks. 
It is in this latter and prevailing sense that the term is here used 
and combined with qualifying words in the terms bowlder gravel, 
cobble gravel, pebble gravel, and granule gravel. 

Sand.—The term sand is in common use among all English- 
speaking geologists for mineral grains smaller than 1 or 2 mm. 
and larger than silt. By some writers, sand is applied only to 
rounded mineral grains, but others use the term more generally. 
Sand is from the Anglo-Saxon word of the same spelling and 
meaning. 

- Silt—The term silt is considered by some geologists to apply 
properly to deposits containing organic matter in addition to the 
mineral particles. The writer was unable to find any considerable 
support of this view by either past or present authorities, and has 
here used the term for the grades designated in the table on page 
384. The word silt is probably akin to a number of Germanic 
roots meaning to sift or filter, compare German sezhen, to strain. 

Clay.—Aiter consideration of a number of alternative terms, 
the term clay has been selected as most likely to be acceptable to 
geologists for the finest clastic sediments. A few geologists 
objected to the term on the ground that it implied plasticity or 
that it referred to a definite chemical composition. It is the view 


tJ. S. Flett, Encyclopaedia Britannica, 11th ed. (1911), Vol. XII, p. 382, and 
Vol. VI, p. 913- 


CHESTER K. WENTWORTH 


386 


16-8 mm 


32-16 mm 


Pebble gravel 


1Z€ 


Natural S 


2 


Fic. 


GRADE AND CLASS TERMS FOR CLASTIC SEDIMENTS 387 


of the writer and of many other geologists that nearly all clastic 
materials of this grade consist largely of the hydrous aluminum 
silicates which make up the clay of the chemist and also that the 
material is always more or less plastic. There is, therefore, in his 
opinion a common ground for the geologist and chemist without 
an insistence on the use of the term clay for the pure chemical 
compounds kaolin or other minerals of this group. 


ROCK TERMS 


Conglomerate-—There are certain terms which are used with 
greater uniformity and less abuse than others. One of these is 
conglomerate. This term is very widely applied to rocks which 
are the consolidated equivalents of gravels. It is applied just as 
is the term gravel to rocks which vary widely in the sizes of their 
constituent particles. It seems desirable, therefore, to apply to it 
modifying adjectives as has been done with gravel, making the terms 
bowlder-conglomerate, cobble-conglomerate, pebble-conglomerate, 
and granule-conglomerate. The term granule-conglomerate is 
preferred to the term grit because grit has been used in England 
for both coarse- and fine-grained and angular-grained sandstones. 
The use of the term grit in the present sense seems therefore 
inadvisable. 

Sandstone.—This term is generally used by geologists and no 
great change in its usage is here proposed. It is qualified by the 
adjectives very coarse, coarse, medium, fine, and very fine, and 
used for the consolidated equivalents of the various grades of sand. 
The term grit is not used for the coarser grade for the reasons 
stated in considering the term granule-conglomerate. 

Silistone —After a consideration of a number of terms, siltstone 
was adopted by the writer as most acceptable for the consolidated 
equivalent of a silt. Shale, as proposed by some geologists, was 
considered objectionable because, in the usage of a majority of 
geologists at present, as well as etymologically, it is a structural 
term referring to the shelly structure of the rock rather than to 
the size of its grains. 

Claystone.—The term argillite has already been used in another 
sense and the term shale is objectionable for the reasons given 


CHESTER K. WENTWORTH 


388 


pues osivo) 


puvs sug Aro, 


uu “1 § 


pues asivod Aa A 


wu I-72 


= OLY 


azig [einqey 


pues our 


aAvId a[NUBID 


wu Z—P 


“pues uInipayy 


Jeavis aiqqog 


wu y—g 


GRADE AND CLASS TERMS FOR CLASTIC SEDIMENTS 389 


above. The term claystone is here proposed as most satisfactory 
for the indurated equivalent of clay, as defined above. 


THE CLASS TERMS 


The grade terms defined above for aggregates and for rocks are 
applicable, as the writer has pointed out in another part of this 
paper, to very few sediments of natural origin. Most such sedi- 
ments are composed of particles of several or many grades and the 
names suggested above with definite numerical limits cannot 
properly be applied to them. This difficulty has been long recog- 
nized by students of sedimentary rocks and several schemes for 
meeting the difficulty have been proposed. Notable among these, 
because of the large amount of data published in accordance with 
it, is that of the United States Bureau of Soils which is given below. 


CLASSIFICATION OF SOIL MATERIAL 
UNITED STATES BUREAU OF SOILS* 


Soils containing 20% silt and clay: 


Woarsesand. 2275.2... more than 25% very coarse sand and coarse sand and 
less than 50% any other grade 
SPDING cs ea a Peo more than 25% very coarse sand, coarse and medium 
sand, and less than 50% fine sand 
Rimersangd cs. ye). ¢ more than 50% fine sand, or less than 25% very 
coarse sand, coarse and medium sand 
- Very fine sand...... more than 50% very fine sand 


Soils containing 20-50% silt and clay: 
‘Saiavahye Woyhnnve Gea pioae more than 25% very coarse sand, coarse and medium 
sand 
Fine sandy loam....more than 50% fine sand or less than 25% very 
coarse sand, coarse and medium sand 


Smunhyr CERes gaeos ue less than 20% silt 

Soils containing more than 50% silt and clay: 
Brava eset thee s ye ons) «32 less than 20% clay, less than 50% silt 
Silit lomians 54 Goede nes less than 20% clay, more than 50% silt 
Clayeloamie a5 20-30% clay, less than 50% silt 
Silty clay loam. ....20-30% clay, more than 50% silt 
CIES Rs ae eee more than 30% clay 


*C. C. Fletcher and H. Bryan, “Modification of the Method of Mechanical Soil Analysis,” U.S. 
Dept. of Agric., Bur. Soils, Bull. No. 84 (1012). 


390 CHESTER K. WENTWORTH 


The scheme of class terms presented below was devised by the 
writer after a somewhat extensive consideration of several more 
complicated classifications. These were characterized by more 
extensive use of the adjectives coarse and fine and by the use of 
many combinations and permutations of several grade terms. 

These were all rejected in favor of the simpler plan here 
presented because of the seeming futility of attempting to make 
detailed quantitative discriminations by use of a verbal scheme. 
Such discrimination can be satisfactorily made only by graphic 
or tabular methods and the writer believes the simpler classification 
will be most generally acceptable and therefore most useful in the 
study of sediments. 

The table shows only class terms for aggregates; the proper 
rock terms will be chosen and used in an analogous manner. 


CLass TERMS FOR SEDIMENTS 


Specifications: by Grade Terms for Classes 


Gravel >80% Gravel 

Gravel >sand > 10% Others< 10% Sandy gravel 
Sand > gravel >10% Others< 10% Gravelly sand 
Sand >80% Sand 

Sand >silt >10% Others< 10% Silty sand 
Silt >sand >10% Others< 10% Sandy silt 
Silt >80% Silt 

Silt >clay >10% Others< 10% Clayey silt 
Clay >silt >10% Others< 10% Silty clay 
Clay >80% Clay 


Certain materials of sedimentary origin but highly variable 
composition, notably glacial till, will not fall into any of the divi- 
sions of the table above. No attempt has been made to make an 
all-inclusive scheme. Till is known by its extreme range of 
mechanical composition and by definition, there‘ore, will not fit 
into so simple a classification. The test of any classification is in 
its applicability to natural objects. The terms here proposed 
were tested by reference to a large number of analyses made by 
Udden.!' By inspection of about fifty of these chosen at random 
it was found that only the analyses of till did not accord with the 
specification given for one of the terms, and that each of the other 


ty. A. Udden, Joc. cit. 


GRADE AND CLASS TERMS FOR CLASTIC SEDIMENTS 301 


sediments tested was assigned to a suitable name differing only in 
added exactness from that used in the original description. In 
the table below are given nine of these analyses showing the 


TABLE OF MECHANICAL ANALYSES BY J. A. UDDEN* SHOWING THE SEDIMENT NAMES 
APPROPRIATE TO THE CLASSIFICATION OF THIS PAPER 


i 1 | 
Diameter in Millimeters ang t Seatin eoael Gaver eee silty ‘Sena 
O4=92 o ob oilers o's Meroe eka ln ed ater etanee Bs oat leone hear ese (ale | Acc ey  et ceo | St aera 
BI=LO o 0.6 oie Ane R RETRO cane [erie Tears AGB ACV N slirhu deen. 525 ill jlrato reneoiceal cuca Re 
1O=8 6 6 ore eee ree Bo3 ieee) Mlevata stale ee 5 ain Kota ieee oro eu ceall Saicral ecoeoars 
SA. 0 06S oie aeRO EE eee 65.9 20.0 OSE 4 ll esravaed irae deal rome ack cnet eA 
A=® 5 § 9: 001 OG NRO RCE 26.0 14.4 7.4 ree 52 
DT 59.610 te Ena a eee 3.0 II.4 3.6 24.4 58 
T= [Dag ty Be eRe ee 08 18.2 ToD 49.2 TB 
TT / D=Tt fA este een RR oD 5.6 30.8 8.3 I7.9 
Tt JARS ato cade ae meer 57] 2.6 Ba) ok 14.8 48.7 
it /G=U /iO pA oleae ae 53 o2 Tee? I.0 8.4 
I f HO BD ob des boat NOREEN ATSRG ON ere a | eee 2 4 13.2 
Fa poral Oar ee PALS ey eire alin wr Sune ettuees sills maim igre lecens-e sieuk ase neetiace o it BoD 
Tf JOA=T I TBS 5:6 ero Oo BOO ePOPEDNA Ose neCel| Oh cecerrce a aor teT| PERERA are ee | ea eee Pera Wf 
Fg AS ml £215 amet ate ner spon eS ctere tere ace lie ena tawal Guemce ellis ewavavayareene all eaaecalaiercweres .I 


* J. A. Udden, Joc. cit. 


{ The numbers are those used by Udden in describing these sediments. 
by the author of this paper according to the terminology here presented. 


The names are applied 


| { 
Jinaugies aa Wi blimctons crore | oa | ones ll sie des 
DAT 5 oo 009.6 6 DLE GD BIND 6 PERRO LS lS: Gto Gisuesal | sanes Eiepeceeteie| Gre tan een steanc | a umeticcey Senameta| (Cina aaa are 
TET 2 4.5 Anca See eee em a Oe 52 ia ETS es at | aerate 
Tt JOT [Adis S ola oe ae CRIN TOE CCCI ae -9 tr. Eisen ANP iieacon nes arene 
T JT Bia ose oe Cea ee Re IER orc ne II.9 tr. 4 23 
T/S=H [TO scecee ere pause eee tee Bie cock teers 9.8 Boi B53 2.0 
if / IO /BBa sia oetcl ees eaae eRe te e 28.7 ED AL 302 5.6 
i / QO OT iatars au ctee eeteecee TERCRE Eee eee eae 40.6 aa 2 28.2 1952 
HY AO ay fe? Stereracee is casan ea a ciel Sane 5.9 24.6 Bil oI 26.4 
1 / TAGT/ ARO ache covommctoeee aloe ree eon 1.5 9.6 TE 19.8 
Tt / 2EOH [Rit s Sa Goce 3 Oe De CER 52 BES 6.4 17.8 
i / SLOT f TOBY cic ace ool pe en nenGO.cuoeie oat | eRe ne ee 9 2.4 IO.0 
Tf LOBARAe / Noy eye epipolar ne eee eae eee JP 8 Ds & 


composition, the number in Udden’s table, and the name according 
to the present classification. 

It will be pointed out that in order to name a sediment, one 
must first make a mechanical analysis. This is true to the degree 


392 CHESTER K. WENTWORTH 


that it is true that microscopic study precedes the final naming of 
an igneous rock. However, as in the case of igneous rocks, field 
names based on simple megascopic inspection will be extensively 
used before detailed studies are made. It is believed that field 
names can be more accurately and expressively used if the principles 
and essential facts of a quantitative classification such as that 
here proposed are duly recognized. 


THE PRE-CAMBRIAN OF WESTERN PATRICIA’ 


E. M. BURWASH 
University of Manitoba, Winnipeg, Canada 


The purpose of this paper is briefly to summarize the results 
of two seasons’ field work in western Patricia. Patricia lies in the 
extreme northwestern part of the province of Ontario, north of 
the Winnipeg, English, and Albany rivers, and their expansions, 
Lac Seul and Lake St. Joseph. The areas studied by the writer 
lie in the southwestern part of the district, and in a sense intervene 
between the classic areas studied by Lawson and others around 
the Lake of the Woods and Rainy Lake and a number of areas in 
the province of Manitoba, more recently studied by Bruce, Alcock, 
Moore, and others. Reconnaisance work had already been done 
in the Lac Seul region and northward along the Trout Lake, Wen- 
nasaga, Cat Lake, and other rivers by Dowling, A. W. G. Wilson, 
Camsell, and others, and the work of E. S. Moore in the region 
southeast of Lake Winnipeg had extended in a few places across 
the boundary of Manitoba into Patricia. The writer’s work was 
done under the direction of the Department of Mines of Ontario 
in the Lac Seul region in 1919 and along the Ontario-Manitoba 
boundary from the Winnipeg River northward in 1921. No great 
originality or finality is claimed for the results as the work in 
both cases was confined to narrow strips of country immediately 
adjacent to the survey-lines or rivers which were traversed, and 
was not of such a nature as to allow of much areal mapping, 
which would afford a more complete knowledge of the relation- 
ships of the various formations than is possible by reconnaisance 
methods. There has not been time for microscopic study of the 
rocks in the boundary region, but their general character is evident 
and in most cases they have been determined already with accuracy 
by Moore, so that data are now available for at least a tentative 


= Read at the Toronto meeting of the American Association for the Advancement 
of Science. 


393 


304 E. M. BURWASH 


discussion of the relations of the two areas with one another and 
with the surrounding regions. 

Lawson’s general statement of the geological sequence of the 
Lake Superior region is as follows: 


Keweenawan 
Algonkian Unconformity 
Animikie 
Eparchaean Interval 


Algoman granites, gneisses, etc. 
Irruptive contact 

Upper Huronian 
Unconformity 

Lower Huronian 
Unconformity 

Laurentian granites, gneisses, etc. 
Irruptive contact 


Keewatin Gata 
Coutchiching Manan 


Archean 


The Algonkian formations need not concern us for the purposes 
of this paper, although dikes exist in some of the areas considered 
which may be of Keweenawan age. 

The region around Lac Seul and northward as interpreted by 
the writer from data collected in 1919 includes the following geo- 
logical sequence :* 

4. Younger porphyritic granite (Birch Lake River). 

3. Older red and gray, sometimes gneissoid granites. 

2. Upper sedimentary and schist series, including a (basal ?) 
conglomerate, hornblende schist and other schists and possibly 
limestones and quartzites.” 

Lower volcanic and schist series, including more or less altered 
volcanic rocks of rhyolitic, andesitic, and basaltic types, frequently 
ellipsoidal in structure, hornblende schists, in part ferruginous, 
and some jaspylite. 

1. The lowest member of this succession is correlated with some 
confidence with the Keewatin of Lawson on account of its litho- 


* But see the Report of the Department of Mines of Ontario, Vol. XXIX, Part 1 
(1920), p. 181, for an alternative classification. 


2 The two last were not seen by the writer. See C. B. Dowling, Geol. Surv. Can., 
Vol. VII, Part F. 


THE PRE-CAMBRIAN OF WESTERN PATRICIA 395 


logical resemblance, and because it is the oldest rock-formation 
present and antedates all of the granites with which it is in contact. 
Tentatively, however, the local name ‘‘Lac Seul Series’? may be 
applied to it. 

2. The series above the volcanics appears to be of predomi- 
nantly sedimentary origin, with a basal conglomerate which has a 


SSS Do gZ Ea 


Pre- Granite Granite Animikle Polaeozole Mesozole 
Meeweenawin 


aN X% Areas discussed by Bruce and Alceck. 


Ath kom SS 
jopapus: ie 


Fic. 1.—Map indicating pre-Cambrian areas mentioned in this paper 


rather wide extent. It was observed especially at Slate and Birch 
lakes and was also reported by Dowling at Red Lake. In all three 
localities it is of practically identical lithological composition. 
Following Moore’s example in the case of a similarly related series 
in eastern Manitoba, it was assigned to the Lower Huronian.* 


t See Ont. Dept. Mines Report, 1920, p. 181, and compare Geol. Surv. Can. Sum- 
mary Report, 1912, pp. 262 ff. 


306 E. M. BURWASH 


However, it affords no evidence of this so far as examined either 
by lying unconformably on post-Keewatin granites or by the 
presence of granite pebbles in the conglomerate. Therefore, its 
assignment to the Lower Huronian would imply either (1) that the 
granites of this region are all of post-Lower Huronian age, or (2) 
that the denudation of the great erosion interval which followed, 
or accompanied and followed, the intrusion of the Laurentian 
batholiths, was not here enough to uncover the batholiths before 
the Lower Huronian sedimentation began. If batholithic intrusion 
took place here in Laurentian time, the batholiths were not uncov- 
ered until much later perhaps between the Upper and Lower 
Huronian or even during the Eparachaean interval. Of these two 
alternatives, the second appears to the writer to be the less probable, 
judging by comparison with other regions not far distant. Neither 
alternative would demand the complete absence of Laurentian 
granite pebbles, since these might have been transported from a 
distance, and the quartz pebbles, which occur and which are well 
rounded in contrast to the angular chert-pebbles of Keewatin origin, 
probably are the residuals of such traveled material. For this series 
the name of “Birch Lake Series”’ is suggested. 

3. The granite bodies which were observed in intrusive relations 
with the Lac Seul (Keewatin) series only are in two cases pink to 
grayish pink incolor. In some parts, usually near contacts or where 
inclusions of older rocks are abundant, they display a gneissoid 
texture; in the central parts of the exposed areas and where rela- 
tively free from inclusions they are more massive. They are also 
in parts porphyritic. The rocks with which they are found in 
intrusive contact are frequently cut by dikes of pegmatite, while 
aplitic dikes as well as pegmatite occur within the mass of the 
granite itself. The rock is biotite granite with some hornblende, 
micropegmatite, and microcline, but no muscovite, tourmaline, or 
titanite. Another type of granite observed in intrusive contact 
with both the lower (Lac Seul or Keewatin) and upper (Birch Lake) 
schist series, is of gray color and is practically a binary granite with 
microcline as the chief feldspar. It also contains a little tourmaline, 
some oligoclase or andesine, and a little zircon and garnet. In the 
pegmatite dikes which accompany this granite tourmaline and 


THE PRE-CAMBRIAN OF WESTERN PATRICIA 307 


muscovite are largely developed. This type of granite occurs on 
the Wennasaga River from a point about ten miles above Slate 
Lake to Hailstone Lake, and also on Pakwash and Little Shallow 
lakes, where it has a syenitic phase with very little quartz. The 
same intrusion may be represented by dikes on a small island near 
the outlet of Lac Seul, which are composed of mica diorite. 

Granites and gneisses of the red type were observed to the south 
of South Cove, Lac Seul, on the lower Wennasaga River, in the 
Cat River basin from Big Portage Lake to Springpole Lake, and 
on the lower Woman and Trout Lake rivers. 

The relations in age between these two types of granite were 
not observable on the route followed by the writer. Both are 
post-Keewatin. The gray type is later also than the upper schist 
series, and both may be. If one is Laurentian and the other later, 
say Algoman, the Algoman is the gray granite. 

There is, in addition, a coarsely porphyritic red granite intrusive 
in the red granite-gneiss of the Birch Lake River. It was not seen 
in contact with the schists, either upper or lower. This porphyritic 
granite is distinguished from the others by the presence of acces- 
sory titanite. 

We pass now to the area studied during the summer of 1921 
which lies along the Ontario-Manitoba boundary, a north and 
south meridian line, and is on the average about seventy-five miles 
(from sixty to one hundred) west of the field already described. 
Here we find as the oldest constitutents of the complex two series 
of rocks quite similar lithologically, and in their stratigraphic 
relation to each other, to the two series of schists already 
described. 

t. The lower is a predominantly volcanic series which includes* 
greenstone (massive diabase and ellipsoidal lavas), quartz-porphyry, 
rhyolite, trachyte-felsite, green and gray schists due to alteration 
of these, and in one place a bed of white quartzite, partly iron- 
stained which occurs on the south side of the Winnipeg River a 
little west of the boundary, and has a total thickness of about thirty 
feet. To these rocks the name ‘‘Rice Lake Series”’ was given by 
Moore. 

t. S. Moore, Geol. Surv. Can. Summary Report, 1912, pp. 262 ff. 


308 E. M. BURWASH 


2. Lying above this is a predominantly sedimentary series whose 
most important rock is an extensive and thick conglomerate com- 
posed of the materials of the lower series with the exception of 
two constituents, a granite which does not appear within the area 
examined by Moore, and cherty fragments from iron-formations 
which were also not found in the lower series, although similar rocks 
are found in the upper series itself. These cherty fragments may 
have come from portions of the lower (Rice Lake) series since 
removed by erosion, or from parts farther east where they are 
known to occur in the lower series. There is a possibility that a 
gray gneissoid granite which may be seen on the Winnipeg River 
intruding the Rice Lake series may account for the granitic pebbles 
in the conglomerate. This gneissoid granite was not found in intru- 
sive contact with the upper (Wanipigow) series and is older than the 
red and grayish-yellow or white binary pegmatitic granites which 
occupy most of the region, since these granites and the pegmatites 
derived from them intrude the older gray granite, and the red 
granite contains inclusions of it which, in many observed cases, 
have themselves included xenoliths of the Rice Lake (Keewatin) 
series. But while this is conclusive as to the relative ages of the 
various granites and the volcanic series, it is not conclusive as to 
the relative ages of the gray granite and the upper or Wanipigow 
series. The Wanipigow may then be older than any of the local 
granites and the pebbles found in its conglomerates may have been 
transported from a distance. 

Comparing the Wanipigow conglomerate with that found in 
the upper or Birch Lake series of the Lac Seul region, it is true of 
both that they consist largely of pebbles derived from the volcanic 
rocks which immediately underlie them. Both also contain frag- 
ments of chert and jasper, which can be accounted for by the 
presence of iron-formation in the lower series of the eastern field, but 
are absent in that of the western area. In both, too, are pebbles of 
quartz. In the Lac Seul area these are hard to account for. They 
are well rounded and consist of a ‘‘sugary”’ white quartzite, partly 
stained to a yellowish color by iron oxide, which shows under the 
microscope ‘‘a granular texture of somewhat interlocking grains, 
no secondary enlargement of grains, slight amounts of bleached 


THE PRE-CAMBRIAN OF WESTERN PATRICIA 309 


biotite and brown iron oxide, very occasional grains of apatite and 
magnetite and flecks of kaolin.”” The quartz pebbles in the con- 
glomerate on the interprovincial boundary are of a rather more 
glassy and massive appearance and must undoubtedly be described 
as vein-quartz. They cannot be accounted for as coming from the 
quartz veins which are now to be seen throughout the area, as these 
are of later age and owe their origin to the subsequent granite 
intrusions. But there is a bed of quartzite, as already described 
in the Rice Lake rocks, which outcrops on the Winnipeg River a 
little west of the Ontario-Manitoba boundary, and to this the 
pebbles of the conglomerate may in part be attributed, while 
others are no doubt derived from distant sources. 

Taking everything into account there seem to be fairly good 
reasons for correlating the conglomerate horizons of the two areas 
and also the volcanics which underlie them, which may be taken 
as in all probability Keewatin. The conglomerate would then 
agree in age with the great erosion interval which has been found 
elsewhere after the Keewatin. In the region to the south and 
east, however, the extrusion of the volcanics, partly at least under 
water, was followed by mountain-building accompanied by granitic 
intrusion and denudation. In the area which is now under review 
there is no evidence of angular unconformity between the volcanics 
and the sedimentaries. The sedimentaries may have been largely 
subaerial in origin and in that case the history indicated is simply 
(x) emergence, (2) erosion of low-lying land to a surface of low 
relief, and (3) deposition of coarse materials perhaps on a pied- 
mont plain. Some of these materials are of local origin while 
others (the quartzes, the cherts in part, and the granites) have been 
transported for greater distances. 

A recent paper by Drs. Alcock and Bruce‘ sums up very well the 
results obtained from a study of a number of areas in northern Mani- 
toba together with that studied by Moore to the southeast of Lake 
Winnipeg and the Star Lake area on the Ontario boundary farther 
south—in all ten localities. Their general conclusion is that the his- 
torical succession begins in most instances, as here, with very ancient 
volcanic extrusions. This was preceded in some cases by sedimen- 


t Bull. G.S.A., Vol. XXXII, pp. 267-92. 


400 E. M. BURWASH 


tation, as in the Rainy River region, ‘‘in others, sediments were 
interbedded with the volcanics, and in other areas these early pe- 
riods of sedimentation continued long after the extrusion of lavas 
had ceased.” In some cases a conglomerate exists which may mark 
an unconformity between the two divisions of this pre-granite series, 
as in the Lac Seul area; in others conglomerates occur at several 
horizons in the upper sediments, but some of them are not of great 
lateral extent, while graywackes and arkoses indicate little trans- 
portation or sorting and possibly subaerial deposition. In some 
places cross-bedded deposits are interpreted as ancient deltas. 

The whole would agree with the hypothesis that while post- 
Keewatin mountain-building occurred in the Rainy Lake field and 
others farther east, the country to the north and west, although 
elevated above sea-level, remained in the condition of a piedmont 
costal plain and, not being folded, was not intruded by granites of 
Laurentian age.t At the same time it was covered with the ill- 
sorted products of mountain erosion, largely through the action 
of torrential streams. At a later time, however, this region was 
subjected to folding, batholithic intrusion, and subsequent deep 
erosion. This was followed in some localities by one or more 
periods of deposition with intervening folding and erosion, after 
which long erosion had produced peneplanation before the ad- 
vance of the Ordovician sea. 

This would agree very well with the facts observed by the writer, 
except that the later sediments following the first folding (Upper 
and Lower Missi, Churchill quartzites, etc.), do not occur in the 
region on which this paper is based. 

«An exception to this is made in the case of the Athapapuskow area in north- 


western Manitoba which would appear to limit the area in which the Laurentian may 
be absent on the northwest. 


PETROLOGICAL ABSTRACTS AND REVIEWS 


CHARLES H. BEHRE, JR. 


Fercuson, J. B., and Merwin, H. E. “The Melting Points of 
Christobalite and Tridymite,”’ Amer. Jour. Sci., XLVI (1918), 
AUA=2O, WSs Ds 

A description of experimental methods employed in testing the melting 
points of these two minerals. A high-temperature furnace, built on the cas- 
cade principle, and capable of maintaining a temperature of 1700° for several 
hours, is described. By means of this furnace quartz has been inverted directly 
through dry heat alone. The melting point of tridymite has been determined 

(for the first time) as 1670°10° C.; that of christobalite has been redeter- 

mined; it is1710°+10° C. Christobalite is stable at higher temperatures than 

tridymite as determined earlier by Fenner. 


Frercuson, J. B., and Merwin, H. E. “Wollastonite (CaO.SiO,) 
and Related Solid Solutions in the Ternary System Lime- 
Magnesia-Silica,” Amer. Jour. Sci.,.XLVIII (1919), 165-89, 
figs. 8. 


A further experimental study of the liquidus curves for the system of these 
three oxides, though not completely quantitative, yet confirms the earlier work 
with the same system, demonstrates the existence of solid solutions of diopside 
up to the amount of 16 per cent in pseudowollastonite and diopside, and of 
Akermanite in wollastonite and pseudowollastonite. A new compound— 
5 CaO.2MgO.6SiO,—is reported. The paper contains also a fairly complete 
discussion of results obtained in the curves derived, and photographs of the 
solidus-liquidus concentration-temperature models. 


FERGUSON, J. B., and Merwin, H. E. ‘The Ternary System CaO. 
Mg0O.SiO,,” Amer. Jour. Sci., XLVIII (1919), 82-123, figs. 19. 


This is the most complicated system of any that can be constructed of three 
of the four oxides, CaO, MgO, AI.O;, and SiO.. Férsterite, diopside, enstatite, 
tridymite, cristobalite, lime, magnesia, pseudowollastonite, and two compounds 
of the three oxides were obtained. Solid solutions of five types were also 
recognized in crystalline form; these included clino-enstatite-diopside solutions, 
pseudowollastonite solutions of varying composition, wollastonite solutions, 
solutions of an unnamed compound of the three oxides, and menticellite solu- 


401 


402 PETROLOGICAL ABSTRACTS AND REVIEWS 


tions. Scholler’s 4kermanite could not be prepared, but 5CaO.2MgO.6SiO, 
and 2CaO.Mg0O.2SiO. were recognized for the first time. The tridymite- 
cristobalite inversion temperature is essentially 1470° C., but measured with 
difficulty because of the great sluggishness of the inversion. 


Fercuson, J. B. “The Oxidation of Lava by Steam,” Jour. 
Washington Acad. Sci., VX (1919), 539-46. 


Demonstrates experimentally that the formerly generally supposed oxidiz- 
ing power of steam in lavas is not effective. Hence volcanoes may give off 
lava bearing ferrous iron and yet discharge large amounts of steam. 


FLORKE, WILHELM: Uber die kunstliche Verwitterung von. Silikat- 


gesteinen unter dem Enfluss von schwefliger Sdure. Giessen, 
HOWE JED. 22 
This is a study of the action of sulphur dioxide on coarse-grainea igneous 
rocks. Experiments were carried on in the presence of water. All types of 
rock, varying from granite to gabbro, were attacked by the SO, vapors. The 
ferro-magnesian minerals were most corroded, hence the basic rocks also suf- 
fered most. Soluble sulphates of aluminum, iron, manganese, calcium, 
magnesium, sodium, and potassium were produced, as well as insoluble hydrox- 
ides of aluminum and iron. 


Foye, WILBUR GARLAND: “ Geological Observations in Fiji,’”’ Proc. 
Amer. Acad. Arts and Sci., No. 1, LIV (1918), 1-145, bibli- 
ography, figs. 40, pls. 1; in two parts. 

A summary of the geology and petrology of the Fiji Islands is here presented. 
The group consists of two larger islands, Viti Levu and Vanua Levu, having 
central cores of deeply eroded plutonic rocks and possibly representing an earlier 
continental mass, and of several smaller fringing land masses. Four periods 
of vulcanism are recognized, the first being rhyolitic, the last basaltic, and the 
two intermediate ones andesitic. On Viti Levu a series of Miocene (?) sedi- 
ments is much folded and overlain on the coastal plains by gently dipping post- 
Tertiary beds. On Vanua Levu are Pleistocene or recent coastal plain lime- 
stones and volcanics. Here subsidence has been generally accompanied by 
reef formation. An extended discussion of the coral reef question is given and 
a modification of Darwin’s hypothesis thought to be favored in some measure. 

The history may be summarized as follows: 


g. Subsidence. 

8. Basaltic intrusions and extrusions. 

7. Uplift and erosion of (6). 

6. Subsidence, with the deposition of about 150 feet of limestones and marls. 


PETROLOGICAL ABSTRACTS AND REVIEWS 403 


. Second period of volcanic eruption, with subsequent erosion. 
. Uplift and erosion of (3). 

. Subsidence and deposition of limes, sands, and clays. 

. Andesitic lavas extruded, later eroded. 

. Plutonic intrusives, eroded to late maturity. 


HNHW PN 


The first part of the paper describes the general geology of the islands; the 
foregoing is a brief summary. The second part discusses the petrography of 
the region studied. The important sediments are chiefly calcareous, but some 
conglomerate is known—triver gravel, with bowlders of all sorts, and a calcar- 
eous conglomerate with over 50 per cent igneous bowlders but much shell-waste 
in the cement. The igneous rocks described are chiefly andesites and basalts, 
of which the latter preponderate in the descriptions. 

‘Some of the extrusions were submarine, marked generally by the presence 
of hornblende; this character almost appears to be diagnostic of the subaqueous 
origin of such flows as bear it. Another interesting feature is the persistent 
association of hypersthene and hornblende. 

The first andesitic period is recognized only on the two large islands. The 
flows of the second andesitic period are widespread and may be essentially 
synchronous in origin with some basalts and rhyolites. The lavas of the 
basaltic period, again, are more limited in their distribution. In general, 
differentiation in Fiji progressed from acid to basic. Volcanic vents have been 
singularly persistent, the basaltic flows seldom or never forming cones of their 
own. ‘Ten rock analyses are presented, but no modes are given and only a few 
of the rocks are classified according to the norm system. 


GoipscHmipT, V. M. Die Gesetze der Gesteinsmetamorphose, mit 
Beispielen aus der Geologie des Siidlichen Norwegens. Christi- 
amiaenor2.,| Pp. ro) ie. ©. 

A concise application of the laws of thermodynamics to metamorphism. 
The “mineralogical phase rule” states that in a stable combination as many 
minerals may be formed as there are variables in the system. Thus in a mix- 
ture of silica, magnesia, and alumina only three minerals may be developed. 
The conditioning agents in contact metamorphism are pressure and tempera- 
ture. By plotting the pressure horizontally and the temperature vertically, 
various fields are obtained which are either gradational or sharply marked off 
from each other. The melting temperatures of the more refractory minerals 
limit the phase-curve above. The effect of pressure is well illustrated by 
grossularite, which requires a higher pressure to maintain its identity at high 
temperatures than at low ones; its compact atomic structure is also favorable 
to stability at high pressures. Separate curves for each system may be pre- 
pared by applying the third principle of thermodynamics, Nernst’s theorem; 
the approximate formula used in calculating the curve is from the same inves- 
tigator. Such a curve applied to the system CaO.SiO..CO; yields a change 


404 PETROLOGICAL ABSTRACTS AND REVIEWS 


from calcium carbonate and quartz at lower temperatures (mean about goo° C.) 
to wollastonite and carbon dioxide. ‘This and similar rules are exemplified by 
the crystalline schists of the Christiania region. 


GotpscumipTt, V. M. Das Devongebiet am Réragen bei Réres. 
With a paleontological supplement: Natuuorst, A. G. Die 
Pflanzenreste der Roragen-Ablagerung. Christiania, 1913. Pp. 
27, pls. 5, maps 2, figs. 3. 

This is a report on the areal and structural geology of an area measuring 
about 10 by 7 kilometers in south-central Norway, bounded on the north by the 
Lake of Betnen, on the south by the Lake of Feragen. The sedimentaries 
shown here are Cambrian, Ordovician, and Devonian, the last composing by far 
the greater part. Granite, augen-gneiss, saussuritized gabbro, and basic peri- 
dotite-serpentines are the igneous rocks. Chromite has been mined in the 
last-named rock. The Devonian beds overlie unconformably the eruptives 
and older sedimentaries; they consist of conglomerates, sandstones, and slates. 
An especially interesting member of the Devonian sequence is a 200-meter- 
thick serpentine conglomerate, of which the angular bowlders resemble breccia- 
fragments of serpentine; at its base it goes over locally into white magnesite. 
The total thickness of the Devonian is about 400-500 meters. 

Faulting has affected the pre-Caledonian sediments, and this was followed 
by a period of intrusion from which date the eruptive rocks mentioned above. 
The pre-Devonian beds dip toward the northwest. 

Devonian beds, however, have a general southeasterly dip. These sedi- 
ments were probably deposited in a basin or basins of limited extent, formed 
and filled during the middle Devonian. 

Illustrative plates and a brief description of the Devonian floras of the 
region accompany the report. 


Goipscumipt, V. M. “Uber einen Fall von Natronzufuhr bei Kon- 
taktmetamorphose,” Neues Jahrb., XX XIX (t914), 193-224. 


oll, i. 


A study of contact metamorphism in sandstones of Norway. The unmeta- 
morphosed sandstones are somewhat arkosic (microperthite, albite, orthoclase, 
and microcline are the feldspars) and flakes of muscovite and chlorite are also 
present; the matrix is calcareous or argillaceous. These sandstones are 
Devonian or Silurian in age. When metamorphosed they normally become 
biotite-rich quartzite or even hornfels. Wollastonite is commonly developed. 
In other instances secondary potash feldspars, sometimes intercrystallized 
with plagioclase, appear in the matrix. Besides quartz, a pyroxene (heden- 
bergite) is very common. This description characterizes the normal type of 
metamorphism. 


PETROLOGICAL ABSTRACTS AND REVIEWS 405 


An entirely distinct type of metamorphism, common along the borders of 
pegmatite dikes, is conspicuous in the presence of aegirite and riebeckite, as 
well as of long, needle-like titanite crystals. The latter, as well as the sode- 
rich pyroxene and amphibole, are attributed to an increase in soda-content, 
with a corresponding decrease in calcium oxide. The formation of aegirite 
ana riebeckite takes place especially in the absence of alumina. 


Gotpscumipt, V. M. Geologisch—Petrographische Studien in Hoch- 
gebirge des Stidlichen Norwegens: III, Die Kalksilikatgneise 
und Kalksiltkatglimmerschiefer des Trondhjem-Gebiets, Christi- 
ania, 1915. Pp. 37, pls. (and maps) 2, table r. 

This is a summary of investigations in the zone of the Gula gneisses and 
slates of the Trondhjem district, Norway. The Gula group is probably of 
Silurian age. The rocks are highly metamorphosed, and the particular facies 
here studied are those rich in calcium silicates. Several zones of the schists are 
recognized, varying in their chlorite and biotite content. The results of extreme 
metamorphism may be summarized as follows: (a) Highly calcareous rocks 
lose their biotite content, this being replaced by plagioclase, amphibole (or 
pyroxene), and potash feldspar; or, (0) If poor in calcium oxide, the rocks are 
converted into a biotite-plagioclase rich slate, with varying amounts of quartz. 
In the former case the rock texture and constituents vary markedly with varia- 
tions in degree of metamorphism. ‘The latter type may be readily shown to 
be of sedimentary origin by means of microscopic study, corroborated by 
chemical analysis. Methods given in the text for calculating the theoretical 
constitution of these meta-sedimentaries are most instructive. 


GotpscumMipT, V. M. Geologisch-Petrographische Studien im Hoch- 
gebirge des Siidlichen Norwegens: IV, Ubersicht der Erupti- 
_gesteine im Kaledonischen Gebirge zwischen Stavanger und Trond- 
jem. Christiania, 1916. Pp. 140, pls. (and maps) 7, table 1, 
figs. 2. 


Reports the petrologic findings in an area about 600 kilometers long lying 
between Stavanger and Meraker, in southern Norway. This area bears igne- 
ous rocks of all sorts, from acid to basic. It is impossible to treat these as 
though they were all products of a single magmatic province; more probably 
they belong to three separate groups, as follows: (1) Green lavas, tuffs, and 
intrusives—the differentiation products being chiefly basic; (2) Anorthosite- 
charnokite rocks, with numerous differentiation products; and (3) Tonalite- 
granodiorite magmas, with many differentiation products. Some few of the 
rocks, indeed, do not even fit into such a classification. 


406 PETROLOGICAL ABSTRACTS AND REVIEWS 


In the first group are lavas, tuffs, agglomerates, and dike-rocks. Pillow- 
lavas are common. ‘The intrusives of this group are gabbros and olivine-rich 
rocks. These are often saussuritized, uralitized, or converted into serpentine. 
Analyses of members of this class, as well as of the two others, are presented. 
The green lavas probably range in time from the upper part of the Ordovician 
into the Silurian; they are apparently homologues of the greenstones of Eng- 
land and Wales, which, however, were extruded at a slightly earlier date. 

The anorthosite-charnokite rocks vary greatly in character. Some are 
typical gabbros and norites, others bear a little potash feldspar. Uralitiza- 
tion is common here. 

Masses of pyroxenites and peridotites also belong to this period; these 
generally appear as lenses enclosed in basic or intermediate masses of more 
recently consolidated magmas of the same group. Another type of rock found 
here is labradorite-rich, hypersthene-bearing ‘‘labradorfels” (=anorthosite). 
All the above rocks of this segregation group are basic. ‘There are also norites 
and mangerites; hypersthene-syenites represent a gradation toward the more 
alkalic end of the series; and finally acid rocks are represented by granites— 
hypersthene-granite (birkremite), commonly showing droplet-like plagioclase 
within the potash feldspar, augite-granite, diopside-granite, aegirite-granite, 
amphibole-granite, and biotite-granite, the latter two forming the commonest 
acid rocks of the group. Pegmatites and dike-rocks are common, and the latter 
range from acid to basic (peridotites). The sequence in the anorthosite-charno- 
kite group is known with certainty to be (1) pyroxenite-peridotite, gabbro and 
norite, (2) norite, mangerite, and labradorite, (3) pyroxene-syenite, monzo- 
nite, and granite. 

These anorthosite-charnokite rocks appear to have played a noteworthy 
part in the diastrophism of the region. They are assigned to early Caledonian 
age. 

The third group studied comprises pyroxenites, peridotites, and gabbros; 
intermediate rocks, too, such as diorites, hypersthene-mica-diorites, and a 
peculiar type newly called “opdalite,” which appears to be essentially a quartz- 
diorite. The acid phase is represented by “‘trondhjemite,” (new name) which 
Kjerulf calls an oligoclase-granite, and which might as well be characterized 
as a granodiorite. Associated with these rocks are some dike-rocks, which 
incline generally toward the acidic; thus trondhjemite-porphyries, -aplites, and 
-pegmatites are recognized. The porphyritic phases occur already as borders 
of the larger intrusive masses. 

A noteworthy fact is the sharp boundary of these aplites against the schistose 
basic country rock. This is explained by applying the principles of differentia- 
tion through fractional crystallization. A granite which has differentiated and 
achieved stability through having the basic constituents of the mother magma 
crystallize out, cannot resorb similar basic components; a magma can only 
assimilate the country rock when the mineral: constituents of the latter have 
not been previously lost by differentiation in the mother magma. 


—— 


PETROLOGICAL ABSTRACTS AND REVIEWS 407 


Most of the rocks of the trondhjemite character are laccoliths or laccolithic. 
They all appear to be post-Ordovician; they may be described in general as of 
Caledonian (Upper Silurian) age. 

Many rocks do not definitely fit into any of the foregoing three groups. 
These include pyroxenites, gabbroid types, basic and normal diorites, tonalite, 
granites, and their aschistic and diaschistic differentiates. This group may, 
if the sequence be as given above, indicate progressive differentiation. They 
occur south of the Trondhjemfjord. Granites are found on the west coast, too, 
and here are definitely assignable to Caledonian age, but cannot be readily 
placed in a magmatic province. Some augengneisses are also assigned to the 
same period. 

As to the consanguinity of the three main groups recognized above, nothing 
definite can as yet be stated; surely the relationship is not as close between the 
separate groups as within each group. Each group, however, illustrates beauti- 
fully the differentiation-parallelism of separate magmas. Whether or not they 
are related might perhaps be answered by calculating the general composition 
of each group from a study of the rock-components. Data for such a calcula- 
tion is, in the opinion of Doctor Goldschmidt, not yet sufficiently complete to 
undertake any generalizations. ‘This is unfortunate, for here is a large geo- 
graphic province which might furnish the best of examples for a study of 
magma-relationships and petrographic provinces in time if not in space. 

Rocks similar to the green lavas occur in Scotland, and others like the 
Bergen-Jotun-granites (Group 2, above) are recognized from Sweden. The 
third group (called by Goldschmidt the opdalite-trondhjemite group) is 
undoubtedly represented in southern Scotland by the Galloway granite of 
Peach, Horne, and Teall, the age of which is less than the Ludlow and greater 
than the Old Red. 

The writer seeks to draw analogies between the petrographic character of 
the rocks in the Stavanger-Trondhjem district and those elsewhere, especially 
in the Alps. 

Twenty-two analyses of typical representatives of the several groups are 


appended, and many more, sometimes recalculated, are incorporated in the 
text. - 


GotpscumipT, V. M. Geologish-Petrographische Studien im Hoch- 
gebirge des Siidlichen Norwegens: V, Die Injektionsmetamor- 
phose 1m Stavanger-Gebiete. Christiania, 1921. Pp. 142, pls. 
15, map (with separate legend) 1, with numerous tabulated 
analyses. 


The rocks in this district are of several periods. At the base of the section 
are pre-Cambrian rocks, unconformable below Cambro-Silurian deposits. The 
_ latter consist of phyllites of Cambrian and lower Ordovician age and of green 
slates of the upper Ordovician and Silurian. These are followed by eruptives 


408 PETROLOGICAL ABSTRACTS AND REVIEWS 


of Caledonian age, which are partly green lavas, partly members of the opdalite- 
trondhjemite group discussed in a previous paper. Locally on the borders of 
the district are conglomerates, arkoses, and sandstones essentially of Downton- 
ian age (Upper Silurian). Still younger dikes of diabase traverse the other 
rocks, and Quaternary sands and clays, partly glacial, are the last deposits to 
be laid down. 

The pre-Cambrian rocks are chiefly granitic, with lesser amounts of amphib- 
olites. Structurally they represent a major syncline, the axis of which 
parallels the trend of the Caledonides, i.e., runs northeast-southwest. The 
phyllite division of the Cambro-Silurian is definitely assigned to the Cambro- 
Ordovician on the basis of fossils occurring outside the area; it begins with a 
basal conglomerate, and passes into phyllites and calcareous strata, with local 
increase in sandy content. ‘This phyllite division is at least 300 meters thick. 

The green saltes are assigned to the Silurian also on the basis of fossil find- 
ings. These “slates” bear sandstones and conglomerates (rarely) and consid- 
erable amounts of limestone and dolomite (which may have been derived 
metasomatically from limestones), but metamorphosed clayey deposits are 
predominant among the sediments. Eruptive rocks are highly metamorphosed, 
so that they cannot always be assigned with certainty to volcanic origin. 

Sedimentary rocks are known from the Caledonian; they consist of con- 
glomerates, typical sandstones, arkoses, and similar types, derived largely 
from the pre-Caledonian and Caledonian effusives. Hence also these actually 
follow the Caledonian opdalite-trondhjemites and green lavas. A description 
of the first of these two Silurian types is to be found in the preceding review; 
both basic and acid rocks are known, the latter including adamellite, granite, 
and quartz-pegmatite. 

The Cambro-Silurian rocks were deposited on a peneplained surface. This 
period of sedimentation was followed by one of intrusion—that of the opdalite- 
trondhjemite group, either synchronously with or slightly preceding the 
Caledonian mountain-building, during which sediments and intrusives alike 
were folded and fractured, the magmas perhaps being deformed even before 
complete congealing. Finally in late Caledonian times there was extensive 
faulting of all the rocks of the Stavanger District. 

A study of the opdalite-trondhjemite masses of Stavanger demonsirates 
that the larger intrusives are limited to the boundary between the phyllites 
and the overlying green slates. These actually represent one type of structure, 
whereas an entirely different type is indicated by flat-lying masses of intrusives 
that seem to have been more severely displaced during cooling. In the Stavan- 
ger District the two types of structures, one representing igneous rocks cooled 
in place, the other those cooled after some dynamic action, lie close together; 
of these the distorted and ‘‘dynamo-transported”’ masses are by far the more 
common. 

After this general discussion of the geology of the region, the writer proceeds 
to a description ot the metamorphism that has affected the rocks, with special 


PETROLOGICAL ABSTRACTS AND REVIEWS 409 


regard for the injection-metamorphism that borders the opdalite-trondhjemite 
type of rocks. The clayey sediments have undergone regional metamorphism 
ranging in its products from the mere development of good cleavage, through 
the development of garnet-mica-schists, and finally the growth in these of basic 
plagioclase and of pyroxenes. 

Most distantly from the borders of the acid masses the phyllite becomes 
altered to garnet; farther in toward the intrusive this is replaced by biotite, 
thus presenting an order that is the reverse of that more commonly observed. 
There is also a progressive increase in sodium content as the intrusions are 
approached. Closer yet microperthite makes its appearance. These rocks 
are in turn altered to injected mica-schists and gneissic mica-schists; the 
““injected”’ schists are not so in the sense that a liquid magma has been forced 
inward along the lines of schistosity; injection, as here used, implies rather a 
metasomatism, with a relatively small addition of extraneous material. 

The upper Silurian green slates, too, form significant “injection schists.” 
Even the basic differentiates in the opdalite-trondhjemite series have under- 
gone some contact alterations, because affected by the later acidic intrusions. 

Various types of metamorphic rock are described in detail; these include 
quartz-muscovite-chlorite-phyllite, quartz-muscovite-chlorite-garnet-phyllite, 
quartz-muscovite-biotite-garnet-phyllite, chloritoid-slates, and albite-porphyro- 
blastic schists. Transitions between these and “‘augengneisses”’ are also 
known. When acid material is added by injection, the albite-porphyroblastic 
rocks are altered to injection-(injected)gneisses. These are partly bedding- 
plane injection types, with marked parallelism in the texture of the crystalline 
injected rock, partly vein-injected gneisses, and augen-gneisses (which are 
attributed also to injection of clay slates by the more acid rock). 

Much of the alteration is due especially to the addition of quartz, water, 
and soda. These may have been brought in as soluble sodium silicate. 

The metamorphism here described is compared with that elsewhere in 
Europe and also with that in the Christiania District, which has been studied 
in detail. In the latter instance it is interesting to note the far greater regional 
distribution of the contact phases, attributed by Goldschmidt to the greater 
water content of the effective magmas. In the Stavanger District probably 
the most conspicuous thing is the extensive contact metamorphism of siliceous 
sediments. Into the same class of metamorphism, due essentially to chemical 
differences between the country rock and the “magma milk,” falls that de- 
scribed by Brauns under the name of ‘“‘pyrometamorphosis.”’ 


Gorpon, SAMUEL W. ‘Ordovician Basalts and Quartz Diorites 
in Lebanon County, Pennsylvania,” Proc. Acad. Nat. Scz., 
Philadelphia, Nov., 1920. Pp. 354-57, figs. 5. 

Includes the first report of Paleozoic volcanic rocks in the form of a basalt 
flow, from Lebanon County, Pennsylvania. Intrusive into the Martinsburg 


410 PETROLOGICAL ABSTRACTS AND REVIEWS 


shale are sills and dikes of quartz diabase, to the northwest of which lie the 
interbedded basaltic flows. -The quartz-diabase is a fine-grained dark greenish 
rock becoming very fine-grained at the contact; it may bear phenocrysts of 
augite, attaining a length of a centimeter; the texture is diabasic or ophitic, 
the labradorite being quite zoisitized. 

The basalt is brecciated or tuffaceous and amygdaloidal, hence the flows 
were probably subaqueous; under the microscope the basalt breccia shows as 
a perlitic, dark greenish glass. 


Guitp, F. N. ‘‘A Microscopic Study of the Silver Ores and Their 
Associated Minerals,” Econ. Geol., XII (1917), 297-353, pls. 11. 


This paper represents a summary of rather extensive work done by Mr. 
Guild on silver ores from many localities, both here and abroad. The charac- 
teristic early minerals of silver deposits are pyrite, sphalerite, and arsenopyrite. 
These are chiefly hypogene in origin, though the first may also occur as a 
secondary mineral. They are commonly followed by the deposition of argen- 
tiferous galena and tetrahedrite; the order of deposition is arsenopyrite, pyrite, 
sphalerite, tetrahedrite, chalcopyrite, and galena. The later silver minerals 
that are discussed include stromeyerite, which exhibits exceedingly close 
“‘pseudo-eutectic” intergrowths, that may actually be eutectic in nature, 
pyrargyrite and proustite, stephanite, polybasite, argentite, and native silver. 
The latter certainly forms in some cases as a result of the breaking down of 
complex silver minerals. Other sources of silver are also briefly considered— 
cerargyrite and the rarer minerals huntlinite (silver arsenide), dyscrasite (silver 
antimonide), brogniardite (a complex sulphide of silver, lead, and antimony), 
and schirmerite (bismuth-silver-lead sulphide). Variously associated with the 
silver are chalcopyrite, bornite, chalcocite, and covellite. Commonly associ- 
ated cobalt and nickel minerals are smaltite, niccolite, and breithauptite. 

The predominant gangue minerals found with silver are carbonates—notably 
in the lead-silver type of ores; this indicates a tendency for such ore-minerals 
to come from neutral or alkaline solution, as carbonates could not be developed 
in acid solutions. The gangue minerals mentioned are calcite, dolomite, 
siderite, rhodochrosite and barite. 

Excellent hints as to methods for distinguishing silver minerals micro- 
chemically are presented, a subject so far neglected only too often. Good 
manuals for the ore-geologist are indeed rare, and good manuals dealing with 
the microscopy of silver minerals are wholly lacking. 

The paper contains interesting discussions of graphic intergrowth, of 
“enrichment”? of silver ores, and of the composition of some important silver 
minerals. A table presents the investigator’s ideas regarding the sequence of 
deposition of double salts of silver and other metals. 


REVIEWS 


Proceedings of the Coal Mining Institute of America. Thirty-third 
Annual Meeting, Pittsburgh, Pennsylvania, December, 109109. 
IEDs Wiser. 

In an article on the “Future Development of Fuels,’’ Henry Koeis- 
inger, fuel engineer, United States Bureau of Mines, discusses those 
modes of coal consumption that utilize most fully the energy latent in 
the fuel, namely by-product production of coke, the use of powdered 
coal, and the use of so-called “colloidal fuel,’’ a mixture of powdered 
coal and fuel oil. Electrification of railroads and better combustion of 
coal in the commoner methods of utilization come in for discussion. 

Pages 34 to 43 are devoted to an excellently illustrated article by 
Reinhardt Thiessen of the United States Bureau of Mines, on the “ Con- 
stitution of Coal through a Microscope.” 

Pages 89 to ror are devoted to a paper by H. C. Ray, professor of 
ore dressing at the University of Pittsburgh, on “Modern Practices in 


the Washing of Coal,” illustrated with nine text figures. 
Ii, Se 18%. 


Report on Some Sources of Helium in the British Empire. By 
J. C. McLennan and Associates. Canada Department of 
Mines, Mines Branch, Bulletin 31, 1920. Pp. 72. 


Shortly after the commencement of the war, it became evident that, 
if helium were available in sufficient quantities to replace hydrogen in 
naval or military airships, the losses in life and equipment arising from 
the use of hydrogen would be enormously lessened. Helium, as is 
known, is most suitable as a filling for airship envelopes, in that it is 
non-inflammable and non-explosive, and, if desired, the engines may be 
placed within the envelope. By its use it is also possible to secure 
additional buoyancy by heating the gas (electrically or otherwise), and 
this fact might possibly lead to considerable modifications in the tech- 
nique of airship maneuvers and navigation. The loss of gas from diffu- 
sion through the envelope is also less with helium than with hydrogen, 
but, on the other hand, the lifting power of helium is about 1o per cent 
less than that of hydrogen. 


AIT 


412 _ REVIEWS 


Early in 1915 Dr. J. C. McLennan, head of the department of 
physics in Ontario University, was requested by the Board of Invention 
and Research, London, England, to investigate the helium content in 
various natural gas supplies within the British Empire. As a result of 
these investigations it has been shown that the largest source of supply 
of helium at present known within the empire is located in Canada. 
Commercial methods of separating helium from the other components 
of natural gas were developed as a result of the preliminary investiga- 
tions, and considerable progress was made toward the development of 
methods for the production of helium on a commercial scale, as a result 
of which it was shown that helium could be produced at a cost of some- 
what less than $0.25 per cubic foot at normal pressures and temperatures. 

As a result of a large number of analyses, which are given in this 
report, it was found that the richest natural gases in Canada contained 
about 0.33 per cent of helium—a percentage believed to be sufficient 
for profitable commercial extraction, but considerably lower than the 
percentages which characterize the gases from a number of wells in 
Kansas, where the helium content ranges from 1.5 to 2 per cent. Analy- 


ses are given of natural gases from Ontario, Alberta, and British 


Columbia. 

As a supplemental investigation, a study was made of the radio- 
activity of a number of these gases. It was found that when the gases 
escape from the well they usually contain the emanations of radium 
and thorium. The thorium emanations are very short-lived, but the 
decadence in the radium emanations is much slower. Measurements of 
the radioactivity of a number of gases were made, the method used 
involving the deduction of the amount of radium emanation from 
measurements of the increase in electrical conductivity which the 


presence of radium emanations imparts to the gas. 
E. SB 


Potash Recovery at Cement Plants. By ALFRED W. G. WILSON. 
Canada Department of Mines, Mines Branch, Bulletin 29, 


NOOO, Ie, Bu, IS. Ue. 

Considerable interest was awakened during the war by the cessation 
of potash imports from Germany in all possible substitute sources in the 
United States and Canada. The report under review is the outcome of 
investigations started then by the Canadian War Trade Board. 

The wisdom of peace-time development of our slender resources of 
potash-bearing brines may be challenged. Much might be said in 


—s 


REVIEWS 413 


favor of allowing such reserves to remain in the ground pending the 
next national emergency. But the recovery of potash as a by-product 
of cement manufacture and from iron-blast-furnace dust is utilization 
of raw material and of heat energy that otherwise would be wasted and 
should command the support of all conservationists. Such processes 
have the added advantage of minimizing the dust nuisance around 
cement plants, which is a danger to the health of employees and is detri- 
mental to crops in agricultural districts. It has been estimated that in 
the United States, if all cement plants were equipped to recover potash 
salts, potash equivalent to over one-fourth of our normal imports of 
German potash salts could be recovered. 

The report under review outlines briefly the principles underlying 
the recovery processes and describes in outline the equipments at all the 
plants in the United States and Canada where potash recovery has thus 
far been practiced. At the temperatures of 1,400 to 1,500° C. obtained 
in cement kilns, the potash-bearing silicates are in part decomposed, 
potash reuniting usually with the sulphate radical to form potassium 
sulphate, or to a lesser extent with the chloride radical to form potassium 
chloride—salts which are soluble and which pass into the stacks with 
the gases, where they can be recovered by spraying the gases or precipi- 
tating the dust by the Cottrell electrical process. The quantity of 
potash salts produced varies from 2 to 7 pounds per barrel of Portland 
cement. 

Installation of potash-recovery equipment does not involve any 
changes in the processes of cement manufacture, nor does it affect the 
grade of the cement produced. 


The report closes with a complete bibliography. 
dis Sp 1p 


Timiskaming County, Quebec. By M. E. Witson. Canadian 
Geological Survey, Memoir 103, Ottawa, 1918. Pp. 197, pls. 
XVI, figs. 6, map. 

This is a concise, detailed report of the results and conclusions of 
geological field work carried on for a number of years in Timiskaming 
County, together with a theoretical discussion of some of the more 
important problems of the area. 

Timiskaming County has an area of approximately 20,000 square miles 
and lies on the east side of the boundary line between Ontario and Quebec 
and east and northeast of Lake Timiskaming. The National Trans- 


414 REVIEWS 


continental Railway crosses the northern part of the county and this 
part can be easily reached. 

This area lies within the Laurentian Plateau and is characterized by 
a remarkable uniform relief but the surface in detail is irregular. The 
minimum elevation is almost everywhere more than 800 feet, and the 
maximum less than 2,o00 feet; thus the range in elevation is seldom 
greater than 1,200 feet. The northern part of the county is drained by 
the Bell and Harricanaw rivers flowing into James Bay, and the south- 
ern part by the head waters of the Ottawa River flowing into the St. 
Lawrence River. | 

A very widespread base-level represented by the erosion surface 
between the basal complex and the Cobalt series or other late pre- 
Cambrian formations is still preserved. So also is the peneplain devel- 
oped before the advance of the early Paleozoic seas. Since these early 
Paleozic rocks are approximately horizontal, all the movements of this 
great area since that time have been due to regional uplift, warping, and 
faulting. Of these, faulting has been the most important. 

Another very noticeable physiographic feature is the large number of 
linear valleys, which, when classified according to general direction, fall 
into three groups striking approximately northwest, north, and north- 
east. The evidence in favor and against the faulting hypothesis of 
origin for these valleys is given. The faulting hypothesis most easily 
accounts for these valleys, and is assumed to be the most reasonable 
explanation, although very little field evidence of faulting along the 
valleys has been found. These valleys are post-Silurian and pre-Glacial. 

The rocks occurring in Timiskaming County fall into the following 
four divisions: (1) the basal complex, (2) the Cobalt series and associated 
intrusives, (3) the Ordovician and Silurian sediments, and (4) Quaternary 
Glacial and post-Glacial gravels, sands, and clays. Detailed lithologic 
descriptions are given of the many different rock types represented, 
with notes on the structure and origin of many of the metamorphic 
types. Based on lithology and origin, the rocks of this region may be 
grouped into three parallel belts running in a northeasterly direction. 
Sediments belonging to the Grenville series characterize the south belt; 
banded gneisses mostly of igneous origin, the central belt; and volcanics 
and clastic sediments, the Pontiac series, the north belt. In the southern 
part of Timiskaming County inclusions of Grenville sediments are found 
along the southern edge of the banded gneiss belt. Because of the many 
difficulties and uncertainties of long-distance correlation of very intensely 
metamorphosed pre-Cambrian formations outlined in chapter IV, no 


a 


REVIEWS 415 


attempt is made to correlate the Grenville sediments south of the belt of 
igneous rocks, with the Pontiac series north of this belt, although the 
three possible theoretical relationships are given. 

Chapter VI deals with “Special Problems of the Timiskaming 
Region,”’ a brief summary of the literature together with evidences and 
conclusions reached from a detailed study of this particular area. Con- 
clusions are given concerning such problems as the “Origin of Pillow Struc- 
ture,” “Origin of Ferruginous Dolomite,” “Origin of Banded Gneisses,”’ 
“Origin of the Cobalt Series”; a general discussion of the “Clay Belt 
of Northern Ontario and Quebec,”’ and the “Origin, Extent, and Dura- 
tion of Lake Barlow and Lake Ojibway.”’ 

In chapter VII the gold, silver, and molybdenite prospects of the area 
are described. When this report was written none of these prospects 
was developed to the state of producing mines. The great mantle of 
post-Glacial lake clays which cover a large part of the county makes 
prospecting difficult. 

In a brief review of this Memoir it is impossible to give an adequate 
summary of the many important problems of pre-Cambrian geology 
discussed. This Memoir will be found useful to anyone interested in 
the problems of pre-Cambrian geology of Quebec and Ontario. 

J. FW: 


The Paleozoic Rocks of the Canton Quadrangle. By G. H. CHADWICK. 
New York State Museum Bulletin, Nos. 217, 218. Albany, 
NEVE EO LOn wep woo, piss 112). ness 3) maps Tr. 

Across the northern third of this quandrangle the Paleozoic rocks 
form the country rock, while in the southern two-thirds they outcrop as 
outliers among the pre-Cambrian rocks. These Paleozoic rocks are 
Upper Cambrian and Early Ordovician in age. A section from the base 
up is as follows: Potsdam sandstone and conglomerate (Cambrian) 
o-150 feet, followed by a possible unconformity, then deposition of 
white sandstone too (?) feet, Theresa dolomite and sandstone 50 feet, 
and Heuvelton white sandstone ro—25 feet in thickness, all of which are 
Saratogan or Ozarkian in age. The Heuvelton is followed by a dis- 
conformity and after this period of erosion the Ordovician (Beekmantown) 
represented by the Bucks Bridge mixed beds of dolomite and sandstone 
50-75 feet in thickness, followed by an unconformity and above this 
unconformity the basal 30 feet of the Ogdensburg dolomite. These 
various formations are described in detail. The Heuvelton and all beds 
above it are fossiliferous, but the fossils are poorly preserved. 


416 REVIEWS 


The Potsdam sandstone is thought to represent the residuum of 
insoluble material from deep and thorough weathering of pre-Cambrian 
gneiss and quartzite. ‘The total absence of fossils, the undercut erosion 
in the Grenville quartzite at some of the contacts, the fine and even 
nature of the sand itself right up to the contacts suggest a wind-blown 
origin. ‘The position, the tillite-like nature, and many other features of 
some of the basal beds of the Potsdam suggest glacial deposits but no 
striated pebbles were found. The late pre-Cambrian and Cambrian 
history of this region is thought to have been as follows: (z) peneplana- 
tion, then (2) uplift and further deep weathering under a moist and 
warm climate, next possibly (3) glaciation followed by (4) arid, cold, 
desert conditions, and finally (5) slow submergence and encroaching of 
sea from the northeast. The late Cambrian and early Ordovician record 
is one of intermittent submergence and elevation. From the history of 
the region to the north it is almost certain that Chazy, Black River, and 
Trenton seas covered this area but their deposits have all been removed 
by erosion with the possible exception of a local thin bed of Trenton 
limestone. By the end of the Mesozoic the general region had been 
reduced to a peneplain. This peneplained surface is recorded by the 
high land with a nearly even sky-line. This peneplain was elevated, 
dissected by Tertiary rivers, and before Pleistocene glaciation, wide flat 
areas were developed which record a late Tertiary peneplain. The 
author calls attention to the close resemblance between the Mesozoic, 
Tertiary, and recent history of the region and the late pre-Cambrian 


and Cambrian history as outlined above. 
Jc EAWe 


The Pre-Cambrian Rocks of the Canton Quadrangle. By JAMES C. 
Martin. New York State Museum Bulletin, No. 185. 
Albany, N.Y. 1916. | Bp. 112, pls) 20, figs: go amemome 

The Canton Quadrangle lies in northwestern New York State about 

12 miles east of Ogdensburg on the St. Lawrence River and 30 miles east 

of the Thousand Islands region. ‘The geology is very similar to that of 

the Adirondack Mountains while the physiography is that of a line of 
foothills which mark the approach of the rugged interior to the scutheast. 

The oldest rocks, Grenville sediments, are crystalline limestone, 
garnet gneiss, quartzite, quartz-schist, and various other siliceous and 
pyritous gneisses. Owing to intense post-Grenville deformation no 
good continuous sections of the sediments remain, but the probable 
thickness is somewhere about two or three miles. The various rock 


REVIEWS 417 


types of the series are intimately interbanded and in many cases one type 
gradually grades into the other. The dark-colored bands and spots 
rich in silicate minerals, in the crystalline limestone, are thought to 
represent recrystallized sedimentary matter and not to be due to the 
introduction of silica, iron, or alumina from the gabbroic or granitic 
magmas. Although some of the metamorphic amphibolite is of igneous 
origin and younger than the Grenville sediments, most of the field 
evidence points to a sedimentary or contact metamorphic origin for the 
garnet gneiss and a large part of the amphibolite. The different varieties 
of garnet gneiss and amphibolite associated with the igneous contacts 
probably are due to the different phases of intensity in the contact action. 

The post-Grenville pre-Cambrian rocks are gabbro-amphibolite and 
granite gneiss intrusives. Two noncomformable contacts and their 
mineralogical composition indicate that certain of the amphibolites are 
igneous. The abundant inclusions of Grenville sediments in these 
amphibolites indicate their post-Grenville age. These gabbro- 
amphibolites are intruded by a gray to pinkish red massive to gneissoid 
granite gneiss. In almost all cases, owing to the severe regional defor- 
mation, the original nature of the contacts between the various types of 
pre-Cambrian rocks of this region is obliterated and it is very hard to 
work out field relationships. Inclusions of gabbro-amphibolites in 
pegmatitic phases of the granite gneiss prove a later age for the granite. 
Also during the intrusion of the granite magma, the amphibolite inclu- 
sions broke under deformation while granite flowed, which shows that 
the amphibolite antedates the intrusion of the magma. However, in 
only a very few cases is there evidence that these amphibolite inclusions 
are igneous and not metamorphic Grenville sediments. The great 
abundance of these inclusions, and the apparent incompetency of granite 
contact alteration to account for so much, suggests that much of it was 
formed by the gabbro-amphibolite intrusions before the invasion of the 
granite magma. 

The rocks are almost universally foliated and over the whole area 
the foliation dips toward the west or northwest. A huge sigmoid fold 
in the southern part of the area is described and diagrammed in detail. 
Many minor folds are described. 

This is an interesting description and application of a number of 
important principles of pre-Cambrian geology, especially the methods of 
separating the sedimentary and contact metamorphic amphibolite from 
the igneous gabbro-amphibolite, and the determination of the younger 


age for the granite as compared with the gabbro-amphibolite. 
ease We 


418 REVIEWS 


Genesis of the Zinc Ores of Edwards District, St. Lawrence County, 
N.Y. By C. H. Smyra, Jr. New York State Museum 
Bulletin, No. 201. Albany, N.Y., 1917. Pp. 39, pls. 12. 


This report presents the results of laboratory studies of ore from the 
zinc deposits of the Edwards district, which is about 15 miles south of 
the Canton area described in Bulletin 185, reviewed above, and the 
geology of the two areas is very similar. The ore minerals are sphalerite, 
pyrite, and galena, and occur as fillings of narrow cracks or as replace- 
ments along shear zones in the Grenville crystalline limestone. The 
deposits are of the high-temperature type, and the sequence of mineral 
deposition is as follows: (1) diopside, tremolite, (2) pyrite, (3) sphalerite, 
(4) galena, (5) talc, (6) serpentine. ‘This sequence of mineral deposition 
indicates changing conditions from intense-contact, metamorphic con- 
ditions to those normal for the depths involved and without outside 
agencies. At all stages during this transition from intense to moderate 
conditions calcite was subject to repeated solution and recrystallization. 

The granite magmas which intrude the Grenville sediments are 
considered the source of the ore minerals and during the cooling of the 
magma, gases and solutions were given off which carried the sulphur 
and metals into the country rocks where they were precipitated. The 
wall rock of the sphalerite is always crystalline limestone while the 
typical pyrite ore is always in schist or gneiss. Also the pyrite ore is 
always rich in graphite while the zinc ore contains but little graphite. 
The typical pyrite ore contains but little sphalerite but the zinc ore 
contains censiderable pyrite. 

These deposits are compared with zinc deposits of other regions and 
especially with the contact zinc deposits of the Christiania district 
described by Goldschmidt. Excellent photomicrographs show clearly 
the various mineral] relations described in detail in the text. This is a 
valuable contribution to the subject of contact metamorphic ore 


deposits. J. FW 


New Edition of Coal, Oil, Gas, Limestone, and Iron Ore Map. West 
Virginia Geological Survey. 

Thoroughly revised, showing oil and gas pools, many anticinal lines 
not heretofore shown, and also booklet giving the names and post-office 
addresses of all the principal coal-mining operators in West Virginia up 
to July 1, 1921. Scale, 8 miles to the inch. Price, folded in strong 
envelope and delivered by mail, $1.00. Remittances to West Virginia 
Geological Survey, Box 848, Morgantown, West Virginia. 


REVIEWS 419 


Detailed Report on Nicholas County. By Davip B. REGER. West 
Virginia Geological Survey, No. 31. 1921. 847 pages+xx 
pages of introductory matter; illustrated with 34 half-tone 
plates and 22 zinc etchings in the text, accompanied by a 
separate case of topographic and geologic maps. 


Nicholas County contains the New River Coal Group, as also the 
Kanawha Group and the lower members of the Allegheny Series in its 
northern portion. This report contains a chapter on the “ Paleontology 
of Nicholas County”? and a short description of the chert deposits of 
West Virginia by Dr. W. Armstrong Price. Price, including case of 
maps, delivery charges paid by the Survey, $3.00. Extra copies of 
topographic map, 75 cents; of the geologic map, $1.00. Remittances 
to West Virginia Geological Survey, Box 848, Morgantown, West 
Virginia. 


Geology and Mineral Deposits of a Part of Amherst Township, 
Quebec. By M. E. Witson. Memoir 113, Canadian Geologi- 
cal Survey, Ottawa, 1919. Pp. 54, figs. 3, pls. VII, maps 2. 

This district is thirty miles north of the Ottawa River and almost 
equidistant from Montreal and Ottawa, and lies within the dissected 
southern border of the Laurentian Plateau. ‘The presence of extensive 
deposits of kaolin near the southern part of Amherst Township is of 
considerable geological interest, because kaolin is commonly thought 
to be the product of surface-weathering and in Canada, for the most 
part, the deposits formed by surface-weathering have been removed by 
Pleistocene continental glaciation. 

The oldest rocks of this region belong to the Grenville sedimentary 
series and consist of quartzite, garnet gneiss, and crystalline limestone. 
These sediments are intruded by the Buckingham series of basic igneous 
rocks (gabbro, pyroxene diorite, and pyroxene syenite). Both the fore- 
going series are intruded by batholithic masses of granite-syenite gneiss. 
Glacial drift and marine Champlain clay partially fill the depressions 
between the rock ridges. 

The kaolin and graphite deposits of the district are described in 
detail. The kaolin occurs in an extensive zone of fracturing and faulting 
in the Grenville quartzite and garnet gneiss and has been brought in by 
solutions from either above or below and deposited along open fracture 
planes or by the replacement of the quartzite wall rock. Crystals of 
tourmaline, a mineral formed at high temperatures, in the kaolin and 
the nearby outcrops of granite-gneiss suggest a deep-seated origin for 


420 REVIEWS 


the kaolin. ‘The presence of oxidized and kaolinized garnet gneiss at a 
depth of 85 feet is equally suggestive of the derivation of the kaolin 
from a superficial source. A summary statement of these two hypotheses 
to explain the origin of kaolin deposits is given, but the writer has no 
definite basis for deciding between them. ‘The shattered zone of quart- 
zite in which the kaolin occurs has a known width of 1,000 feet and a 
length of 7,000 feet. This kaolinitic quartzite rock can be easily crushed 
and is suitable for making silica brick of the ganister type or the kaolin 
can be washed from the crushed material and the quartz used for silica 
sand. 

North of the kaolin locality a number of graphite deposits have 
been opened along the contacts of pegmatite, pyroxene granite, syenite, 
and Grenville limestone. The ore consists of aggregates of orthoclase, 
wollastonite, diopside, scapolite, and graphite. These relations and 
associations of minerals indicate that this material was formed by the 
interaction of emanations from the igneous intrusions and the limestone. 


J. F. W. 


Map of the North Pacific. By W. E. Jonnson. U.S. Coast and 
Geodetic Survey, Map No. 3080, North Pacific Ocean; scale 
1:20,000,000; dimensions 14 by 41 inches. Price 25 cents. 


A new base map of the North Pacific Ocean on the transverse 
polyconic projection has been prepared by W. E. Johnson, Cartographer, 
of the U.S. Coast and Geodetic Survey of the Department of Commerce, 
and is now available for distribution. This system of projection was 
devised by Ferdinand Hassler, who was the organizer and first Superin- 
tendent of the U.S. Coast and Geodetic Survey. This projection was 
computed and constructed by C. H. Deetz, Cartographer, U.S. Coast 
and Geodetic Survey. 


The Mogollon District, New Mexico. By HENRY G. FERGUSON. 
Bulletin 715-L, United States Geological Survey, Government 
Printing Office, Washington, D.C., 1921. Pp. 34, pls. 6, figs. 2. 

The Mogollon (Mo-go-yohn) or Cooney district les in southwestern 

New Mexico, about fourteen miles from the Arizona line. The district 

was discovered in 1875, when James Cooney found rich silver-copper 

ores there. Since then the mines have yielded about $16,000,000 worth 

of ores (estimated to 1917). 

The topography is generally very rugged, especially on the eastern 
edge of the district, which lies along the Mogollon Range. The rocks are 


REVIEWS ALAA 


dominantly Tertiary lavas, with interbedded sandstones. Faulting is 
extensive, and the fault fissures have been mineralized. In Quaternary 
times erosion was great and thick gravels accumulated; the gravel depo- 
sition was followed by more faulting, the fault plane paralleling the 
present front of the range, with downthrow on the west. Renewed 
erosion followed this last orogeny. The Tertiary formations, chiefly 
igneous, reach 8,000 feet in thickness. The flows are chiefly andesites 
and rhyolites, though there are a few basaltic lavas. Of the total 
thickness mentioned, 6,400 feet are pyroclastics and flows; the remainder 
are stream-laid sediments. Erosional unconformities within the se- 
quence are common. Intrusives are represented by dikes of rhyolite 
and basalt. 

Structurally the area is affected by complex normal faults; two 
periods of faulting are recognized—one before, the other after, the deposi- 
tion of the rival gravels. The former was far more widespread, and 
nearly all the fault fissures of this period are now the site of fissure veins 
trending roughly north-south or northwest-southeast; one of these— 
the most prominent—can be traced for a length of seven miles, and the 
fault involves a displacement of a thousand feet orso. ‘The first faulting 
resulted in increase erosion and the development of mature topography; 
on the west slope of the region there was at this time a broad valley, 
which later became filled with gravel to a depth of several hundred feet. 
Then came the great fault that now defines the front of the range; this 
renewed erosion and initiated the present cycle with its higher flat 
benches and lower steep canyons. 

The ore deposits of the region are all in veins closely connected with 
the faults. ‘The richest are the small ones lying along the fault fissures 
that are distributive from the great seven-mile fault (Queen fault) 
mentioned above. Locally the mineralization shifts from the main 
fissure to minor fissures in the wall, indicating that mineralization was 
not contemporaneous with the faulting. The mineral content of the 
veins includes quartz, calcite, a little adularia, and flourite (locally 
plentiful). Argentite, with small amounts of associated pyrite, chal- 
copyrite, bornite, galena, and sphalerite are the metaliferous primary 
minerals; in a few veins copper minerals—bornite, chalcopyrite, chalco- 
cite, and tetrahedrite—predominate. Quartz and calcite are the chief 
gangue constituents. Oxidation is shallow and irregular. Sulphide 
enrichment appears to have been somewhat effective. The oxidized ore 
bears cerargyrite, native silver and gold, basic copper carbonates, 
limonite, copper pitch ore, cuprite, and manganese oxides. Chalcocite, 


422 REVIEWS 


covellite, some argentite, pyrite, and native silver are probably due to 
sulphide enrichment, but a part of the argentite and pyrite and all of the 
bornite, chalcopyrite, tetrahedrite, and galena are apparently primary. 

Quartz was the first mineral to be deposited and it was followed at 
once or even accompanied by the greater part of the sulphides. The 
later stages of vein-filling were marked by coarsely crystalline calcite, 
largely manganiferous and barren of sulphides. Mineralization of the 
wall rock is not prominent; in andesites, pyrite has penetrated the rock 
for short distances from the veins, or the rock is cut by veinlets of quartz, 
calcite, and copper sulphides. The dark silicates in the andesites are 
often almost completely altered, and calcitization is common. The 
rhyolitic country rock near the veins shows some secondary silicification, 
and the feldspars are frequently replaced by quartz with an apparent 
decrease of volume. 

Although the oxidation zone is generally shallow, it may extend to 
the 500- or 700-foot levels. Argentite appears here and there as filling 
in minute cracks, apparently later than the veins, and pyrite clearly of 
later origin is seen in fissures in the quartz. No sulph-arsenic or sulph- 
antimony salts of silver, such as commonly characterize many sulphide 
enrichment zones, are found, but chalcocite and a small amount of 
covellite do attest the effectiveness of some enriching action. Probably 
sulphide enrichment is appreciably active only where unusually favorable 
conditions, such as later faulting, were present. The ground water-level 
lies very deep and the mines are dry, except along the Queen fault. 


The ore shoots are 300 to 600 feet in drift length and about as long ~ 


parallel to the dip, with widths averaging 5 to 15 feet. ‘They tend to 
show flat bottoms, which is thought by the investigator to be suggestive 
of enrichment. 

The peculiar localization of the productive veins within a small 
district not over one square mile in extent is attributed to the faulting, 
which only produced the throw necessary to bring the ore within reach 
of the surface over a very small area; greater or lesser throws either 
resulted in too deep an erosion or in the exposure of the barren low- 
temperature manganiferous-calcite veins. | 

The important mines are the Cooney, Little Fanney,and Last Chance; 
the last being the largest. In the past twenty-five years the Last Chance 
mine has yielded about $7,500,000, worth of ore. The metals obtained 
are gold and especially silver, and a high-grade ore lies close to $20 a 


ton in value. 
C. H. Baars 


oe 


a 


REVIEWS 423 


Tungstenin 1918. By FRANKL.HEss. From ‘ Mineral Resources 
of the United States,’ (United States Geological Survey), 1918. 
Part I, pp. 973-1026; Government Printing Office, Washing- 
ton, D.C. Pp. 53; with summary of recent publications. 


The years 1916, 1917, and 1918 recorded the largest production of 
tungsten ores which this country has known, about 5,000 or 6,000 short 
tons for each of these years as compared with a maximum around 2,000 
tons in previous years. Slightly over 50 per cent of the 1918 output 
was scheelite from California and Nevada, and nearly 40 per cent was 
ferberite from Colorado. Preliminary figures for tg19 and 1920 issued 
by the United States Geological Survey show a slump in production to 
around 200 to 300 tons, figures lower than any recorded since 1903 and 
due to the depression in the steel industry which is the principal tungsten 
consumer. 

No fully satisfactory substitute for tungsten in tool-steel has been 
developed. To quote Mr. Hess: ‘‘It is said that in England and France 
molybdenum has been used to replace about half of the tungsten in some 
high-speed tool-steels, but it is apparently not a preferred metal, being 
used only when it is difficult to obtain tungsten.” 

In California and in Nevada scheelite is found mainly in contact 
metamorphic deposits in limestone near granitic intrusives. In South 
Dakota the production was mainly wolframite from replacement deposits 
in dolomite near Lead and Deadwood. ‘The chief competitor for the 
American market is China, where the discovery of great, easily worked 
placers in the southern provinces has been the most striking event of 
recent years in tungsten geology. 

A brief summary of the international situation shows that Great 
Britain controlled and imported, in 1918, 33.7 per cent of the world’s 
output, and thus owned by far the greater part of the 1918 production. 

Another interesting fact in regard to tungsten minerals is their 
rather general limitation in economically important deposits (with the 
exception of those on the Iberian Peninsula) to areas closely contiguous 
to the Pacific Ocean—a good illustration of the provincial character of 
metallic distribution, which may perhaps be correlated with the petro- 
graphic provinces as pointed out by Alfred Harker. 

The paper presents an excellent and much-needed summary of the 
tungsten situation, treated both from the geological and economic point 
of view, and closes with a summary of recent literature on tungsten. 


IDg Sp 135 Nwpo) (Ce Lele 1B [fe 


424 REVIEWS 


Geology of the Matachewan District, Northern Ontario. By H. C. 
Cooke, Memoir 115, Canada Department of Mines, Geological 
Survey, Ottawa, 1919. Pp. 60 (including index), with map, 
figs. 5. 

The area described lies in the district of Timiskaming and includes 
about 430 square miles. ‘Topographically two elements are recognizable 
—first, pre-glacial erosion, chiefly of the Cobalt series, which frequently 
outlines structural features in the pre-Cambrian rocks, and second, drift 
features, changing with the character of glacial erosion and deposition. 
The bed rocks are entirely pre-Cambrian, ranging from Keewatin (?) to 
Keweenawan. The Keewatin (?) rocks are volcanics—extrusives, for 
the most part, but with smal] masses of peridotite intrusive into them. 
The peridotites may prove to be of commercial importance, since 
asbestos of good quality has been found in them; the rock is highly 
metamorphosed—kaolinized, or altered to talc, or sericitized. The 
rhyolites are slightly less quartzose than those of northern Quebec, and 
the basalts of the Keewatin frequently show pillow structure. All these 
rocks are highly altered and closely folded and faulted, the folding 
probably following the deposition of the overlying Kiask series which 
are dominantly metasedimentaries of many types. From their character, 
it is thought that the Kiask sediments were laid down rapidly, without 
much weathering, on an uneven surface. 

Kiask sedimentation was succeeded by granitic intrusions and later 
by a period of basic intrusion, marked by diabase dikes. The overlying 
Cobalt series is divided into the Gowganda formation (basal conglom- 
erates and very coarse clastics) and the Lorraine quartzite, following 
Collins. Faulting and some gently folding have been developed here also. 

Deposits of asbestos and small deposits of barite, fluorite, and 
hematite have been found. The asbestos occurs as veinlets in small 
masses of serpentinized peridotite. The barite, fluorite, and hematite 
occur in veins. By far the most important mineral however is gold, 
which has been known in this district since 1917. ‘The gold is closely 
associated with intrusive granite porphyry; solutions thought to have 
come from the granite porphyry magma have mineralized the volcanic 
country rock with the deposition of auriferous pyrite. The gold is in 
narrow veins of quartz intersecting the granite porphyry or in lenticular 
ore bodies in the tuff and schist, varying in size up to 75 feet, with their 
long dimensions parallel to the bedding planes of the tuff and schists. 

A geologic map makes the work complete, but the absence upon it 
of topographic contours is regrettable. C.H.B., Je. 


Vee! | 


: MICROSCOPE “K, M” 


“for routine savertieationg 


-It is provided with the same features 
as our large research model, viz.: 


Wide tube and eyepieces of enlarged 

diameters which renders a field of 

view 50% larger than those of other 
Model - manufacture. Improved illumina- 
KM? ting systems. 


~ Its liberal size permits the accommoda- 
tion of all existing accessories, including 
“Federow” Universal-Rotating-Stage. 


We offer an extensive line of petro- 
graphical microscopes and accessories 
and are able to meet all requirements 
for research or elementary investiga- 
tions. 


The microscope “S:.M,” for syn- 
chronic rotation of both nicols is our 
latest achievement. 


z Microscopes are recognized throughout the world as leading 
1 are invariably selected by petrographists who appreciate quality. 


Z/LEITZ-QUALITY \\ 


te! NOT 
seein oe a easay 


ITis 


E.LEITZ 
NEW YORK / 


60 East 10ST. 


A Joy and Benefit 
; and Lovers of N 


Fits in case only four inches high and can 
be carried and used anywhere. Has the 
standard keyboard, with four rows of keys 
and no shifting for figures. Has the auto- 
matic ribbon reverse. Has every feature you 
associate with typewriting at its best. 


The extraordinary demand for the Remington 
Portable—which is taxing to the utmost our 
factory capacity—is proof of its leadership, 
in quality and popularity. 


' Price, with case, $60 


Remington Typewriter Company 
: (Incorporated) 


374 Broadway New York 


Branches and Dealers Everywhere 


SOUND BEACH — 


THE ORIGIN OF THE E A 


. _ Professor Emeritus in ae Dobueenen oF Gaotands ke eyes ion 


The University of Chicago 
Ww vA 


One of ‘The University of Chicago Science Series” . =a 
FROM THE PREFACE 


into ae we have let the incidents that led the i inquiry on from one. ‘s 
another fall in with the steps of the inquiry itself. It is in keeping with 1 
purposes of this series of booklets that the motives which set researche 80 
should have their place with the quests that arose from them. .... T 1a 
story of the birth of the earth will come only after a time, when the ves ae ‘ 
creation have been more keenly discerned and more faithfully rendered han 

possible now.” ioe 


Cloth, $1.75 net, postpaid $1.85 


THE UNIVERSITY a CHICAGO 
~CHICAGO ——- 


i 


a 


eee 


ee 8: EDITED BY 
: THOMAS Cc. CHAMBERLIN AND *ROLLIN Be Ss 
<= ‘ oes With the Active Collaboration of 


S < ; 
me Tevertebrate Paleontology : “ALBERT JOHANNSEN, Petrology 
Economic Geology ie aa ae ROLLIN T: "CHAMBERLIN, Dynamic Geology 


_ ASSOCIATE EDITORS 


RICHARD A. F. PENROSE, Jr., Philadelphia, Pa. 
WILLIAM H. HOBBS, aerate of Michigan 
_ FRANK D. ADAMS, McGill University 
CHARLES K. LEITH, University of Wisconsin 
~ WALLACE W:. ATWOOD, Clark University 
_ WILLIAM H. EMMONS, University of Minnesota 
_ ARTHUR L. DAY, Carnegie Institution 


_ AUGUST-SEPTEMBER 1922 
R SUPPLY OF THE HOT. SPRINGS, ARKANSAS - - Kirk BRYAN 


OUTLIER NEAR THE CREST OF THE OZARK UPLIFT 
Sage Jostan BRIDGE AND B. E. CHARLES 


; PRE- CAMBRIAN FORMATIONS OF NORTHERN ONTARIO AND 
- =) ch ibe BRUCE 


ACCUMULATE DURING GLACIAL RETREAT? STEPHEN SARGENT VISHER 
EDITORIAL—ROLLIN DeSALISBURY, 0 = aegis 2b CE 
a ABSTRACTS FUND RE VIE Wisr n= eos te a ee ig 


e UNIVERSITY OF CHICAGO Bivies 
CHICAGO, ILLINOIS, U.S. A. 


“THE CAMBRIDGE UNIVERSITY PRESS, Lonpon 
E MARUZEN-KABUSHIKL-KAISHA, Tokyo, Osaka; Kyoto, FUKUOKA, SENDAI 
sa tone MISSION BOOK COMPANY, SHANGHAI ‘ 


Zeggeanin | ia a rn 


ns IMPLICATIONS OF ‘THE GREAT LOESS “DEPOSITS: DID THEY 


NUMBER 6 


GEOLOGY 


*JOHN C. BRANNER, Leland Stanford Junior University 


425. 


“450 


459 


472 
480 
482 


501 


August-September 1922 ee ce Bee ne ie 


EDITED BY | 
‘THOMAS C. CHAMBERLIN AND ROLLIN D. SALISBURY 


, : With the Active Collaboration of < 
STUART WELLER gig ALBERT JOHANNSEN — 


ave 


Invertebrate Paleontology ‘ . Petrology 
EDSON S. BASTIN Se hse ROLLIN T. CHAMBERLIN 


Economic Geology Degree gol 


The Journal of Geology i iy Paniened by ie Tee ‘ae Chicago at the Univeesey e Chicago 
5750 Ellis Avenue, Chicago, Illinois, semi-quarterly, on or about the following dates: February T,. r¢ 
May 1, June 15, August 1, September 15, November 1, December 15. {] The subscription pric is. 
per year; the price of besle copies is 65 cents. Orders for service of less than a half-year will be ch 
at the single-copy rate. Postage is prepaid by the publishers on all orders from the United j 
Mexico, Cuba, Porto Rico, Panama Canal Zone, Republic of Panama, Dominican Republic, El Sal 
Bolivia, Colombia, Honduras, Nicaragua, Peru, Hawaiian Islands, Philippine Islands, Guam, Sa: 
Islands, Shanghai. Postage is charged extra as follows: for Canada, 25 cents on annual subse 
(total $4.25); on single copies 3 cents (total 68 cents); for. all other countries in the Postal Uni 
50 cents on annual subscriptions (total $4.50), on single copies 6 cents (total 71 cents). { Patrons a 
requested to make all remittances payable to The Daten of Chicago Press i in Bee geek or bees mon 
orders or bank drafts. ; 4 
The following are authorized to quote the prices aricaten 
For the British Empire: The Cambridge University Press, Fetter Trane Latin E: CH i: 
subscriptions, including postage, £1 5s. each; single copies, including postage, 4s. ‘each. a 
For Japan and Korea: The Maruzen-Kabushiki-Kaisha, 11 to 16 Nihonbashi Tori Sanchor i 
Tokyo, Japan. Yearly subscriptions, including postage, Yen 9.45 each; as fODIS, includi dit 
postage, Yen 1.50 each. s 

For China: The Mission Book Gonuoany: 13 North Szechuen Road, Shanghai. Yearly 
scriptions $4.00; single copies 65 cents, or their equivalents in Chinese money. Postage ex 
- mailed direct outside of Shanghai, on yearly subscriptions 50 cents, on single copies 6 cents. 
Claims for missing numbers should be made within the month following the regular month of “pub 
cation. The publishers expect to supply missing numbers free only when losses have Dee a 
transit, and when the reserve stock will permit. 
Business correspondence should be addressed to The University of Chicken Daeee’ Chicages a 
Communications for the editors and manuscripts should be addressed to the Editors of Tue Jo 
NAL OF GEOLOGY, the University of Chicago, Chicago, Ill. : a 
Contributors are requested to write scientific and proper names with particular care and i in 
to follow the form shown in the pages of the Journal. : 
Because of the high cost of printing, this Journal has found it necessary to adopt. ‘the following 1 
{1 x. If articles exceeding 25 pages in length are accepted for publication in ‘the Journal, the cost 
position beyond 25 pages shall be paid by the author or by the institution which he represents, at the 
rates of the University Press. 2. The cost of illustrative material, in excess of $1.00 per page 
article, beyond the cost of type which would have occupied the same space, shall be borne by the 
{ 3. The cost of corrections in proof in excess of ro per cent of the cost of first composition, 
charged to the author. { 4. While the cost of printing remains as now, the Journal cannot su] 
prints without charge, but reprints will be eS es at cost, if ordered in’ Bes of publ 


N 


Entered as second-class matter, March 20, 1893, at the Post office at Chicago, Il., Aare the Act of March oe 1870. 
i a cceDEAaee for mailing at special rate of postage provided for in Bees 1103, Act of October ay form au 
July 15, tor ; : 


PRINTED IN THE U.S.A. ay “as 


Se 


Among articles to appear in 
early numbers of the Journal 
of Geology are the following: 


The Lava Field of the Parana Basin, South 


America. By CHartes Lawrence BAKER. 


The Physical Chemistry of the Crystallization 
and Magmatic Differentiation of Igneous 
Rocks. V and VI. By J. H. L. Voer. 


Origin of the Triassic Trough of Connecticut. 
By W. G. Foye. 


The Nephelite Syenite and Nephelite Porphyry 
of Beemerville, New Jersey. By M. Aurous- 


sEAU and Henry S. WASHINGTON. 


The Problems of the Anorthosites and Other 
Monomineral Igneous Rocks. By F. 


LoEWINSON-LESSING. 


Ward’s Natural Science Establishment — 


ROCHESTER, NEW YORK 


Relief Maps 


This map, 29" x 46", scale 32 miles to the inch, with a vertical exaggeration of 20 to 1, costs 
only $30.00. The same map, hand colored to show the geology as given in United States 
Geological Survey professional paper No. 71, is $150.00. Write for circular G-233 describing our 
relief maps. 


Fossils 


We are better prepared than ever before to furnish index fossils and exhibition material. A few 
of our recent acquisitions follow: 
Cambrian material collected by Dr. Wolcot in British Columbia. 
Rare Ordovician material from New York and Bohemia, such as Aeglina, Proetus, Cryptolithus, 
Illaenus, Cystids, etc. 
Rare Silurian forms from Bohemia such as Ascoceras and Ophidioceras, Cromus, Acidaspis, etc. 
Rare Devonian material from Germany such as Harpes, Cerarges, Goniatites showing sutures, 
Hexacrinus, Cupressocrinus, etc. 
Fine Solenhofen fish, crustaceans and insects, 
Comanchic material from Texas, 
Eocene leaves from Texas and Louisiana, 
Insects in amber. 


Collections of Rocks 


Our rock collections, varying in price from $1.00 to $200.00, are arranged for Genetic, 
Stratigraphic, Petrologic or Petrographic study. Write for circular No. 220, listing these and 
also our Mineral collections. 


WARD’S NATURAL SCIENCE ESTABLISHMENT 


ROCHESTER - - - NEW YORK 


Important Books for Scientists 


A Laboratory Manual for Comparative Vertebrate Anatomy. By Linsie H. 
Hyman. Cloth, $2.50, postpaid $2.62. 


A Laboratory Manual for Elementary Zodlogy. By Linpte H. Hyman. Cloth, 
$1.50, postpaid $1.60. 


A Laboratory Outline of Embryology. With special reference to the Chick and 
the Pig. (Third edition, revised.) By FRANK R. LILLIE AND Cari R. 
Moore. Paper, 50 cents, postpaid 53 cents. 


A Laboratory Guide in Bacteriology. By Paut G. HEINEMANN. Cloth, $1.75, 
postpaid $1.85. 


The Anatomy of Woody Plants. By E. C. Jerrrey. Cloth, $4.75, postpaid 
$4.87. 


Plant Genetics. By JoHN M. CouLTER and MERLE C. CouLTER. Cloth, $1.50, 
postpaid $1.60. 


Morphology of Gymnosperms. (Revised edition.) By JoHn M. Courter and 
CHARLES J. CHAMBERLAIN. Cloth, $5.00, postpaid $5.12. 


Methods in Plant Histology. By CHARLES J. CHAMBERLAIN. Cloth, $3.25, 
postpaid $3.37. 

Artificial Parthenogenesis and Fertilization. By Jacques Logs. Cloth, $2.50, 
postpaid $2.62. 


The Control of Hunger in Health and Disease. By ANToNn J. CArtson. Cloth, 
_ $2.50, postpaid $2.60. 


Essentials for the Microscopical Determination of Rock-Forming Minerals and 
Rocks. By ALBERT JOHANNSEN. Cloth, $2.00, postpaid $2.15. 


Readings in Evolution, Genetics, and Eugenics. By Horatio Hacketr NEw- 
MAN. Cloth, $3.50, postpaid $3.62. 


Heredity and Eugenics. By Witiiam E. Caste, JoHun M. COULTER, CHARLES 
B. DAVENPORT, Epwarp M. Fast, and Wittiam L. Tower. Cloth, $3.00, 
postpaid $3.12. 


The Teaching of General Science. By W. L. EIKENBERRY. Cloth, $2.00, post- 
paid $2.10. 


The University of Chicago Science Series. 


The University of Chicago Nature-Study Series. 


THE UNIVERSITY OF CHICAGO PRESS 
CHICAGO : : - : : - ILLINOIS 


Second Edition: Revised and Enlarged 


Source Book for the Economic 
Geography of North America 


By CHARLES C. COLBY 


Associate Professor of Geography 
in the University of Chicago 


The fundamental idea of this book is to make available to the busy teacher the 
material on this subject which is scattered widely through literature. The course which 


has been given for a number of years at the University of Chicago has resulted in the 
collection and organization of the material most suitable for the purpose. 


The book contains short magazine articles and materials from Canadian, Mexican, 
and United States government publications. The selection of the material has been 
governed by the concept that there is a direct relation between the environment and 
the economic activities of the region. The author aims to survey the different sections 
of the country from the viewpoint of industrial opportunity. 


With chapter and maps on Alaska Cloth, $3.00, postpaid $3.10 


THE UNIVERSITY OF CHICAGO PRESS 
CHICAGO : ; : - - ILLINOIS 


Cambridge Gntversity Press 


The Study of Geological Maps. By Gerrrune L. Extes, D.Sc., Fellow of Newn- 
ham College, Cambridge. Crown 4to. With 7 plates, 64 figures, and a map. Price on 
application. Cambridge Geological Series. 


A Manual of Seismology. By Cuartes Davison, Sc.D. With 100 illustrations. 
Demy 8vo. Price on application. 


Petrology for Students: an Introduction to the Study of Rocks under 
the Microscope. By Atrrep Harker, M.A., F.R.S., F.G.S. Fifth edition, revised. 
Crown 8vo. $3.75. 


A Treatise on Crystallography. By W. J. Lewis, M.A. Demy 8vo. $5.00. 


The Principles of Stratigraphical Geology. By J. E. Marr, Sc.D., F.RS., F.GS. 
Crown 8vo. $2.50. 


An Introduction to Geology. By J. E. Marr, Sc.D., F.R.S., F.G.S. Crown 8vo. 
$2.50. 


Published by the Cambridge University Press (England) 
The Macmillan Co., Agents in the United States 
64 and 66 Fifth Avenue, New York City 


YOUR MAP NEEDS 


can be selected quickly and with assurance of genuine satisfaction from the 
most comprehensive assortment of effective geology and geography ‘‘teaching 
tools’? ever offered. Note below a summary of that part of 


“‘The Johnston-Nystrom Line’’ 


applying particularly to geography and geology teaching. 


POLITICAL GEOGRAPHY 
Political Maps—Ten Series comprising 136 
different maps. 


Blackboard Outline Maps in two sizes. 
All continents, U.S., World, and many 
states. 


Desk Outline Maps (for pupils) two sizes. 
Wall Outline Maps. 

State Maps. 

Globes—Political, in four sizes. 
Globes—Slated, in four sizes. 
Atlases—Political and Commercial. 


COMMERCIAL GEOGRAPHY 


Finch Series for United States—34 maps 
_on to plates, 50’’X 38”. Shows Products, 
Industries, etc. 


World Maps (Trade and Industry). 
Atlases. 


PHYSICAL GEOGRAPHY 


Physical Maps—Five Series consisting of 
82 different maps. 


Rainfall Maps, U.S., World and Continents. 


PHYSICAL GEOGRAPHY (Continued) 
Vegetation Maps, Continents and World. 
Pressure and Winds Map. 

Major Natural Regions Map. 
Thermal Regions Map. 
Structure Map. 
Timber Distribution Maps. 
Desk Outline Relief Maps. 
U. S. Outline Map with Topography. 
Graph Charts (Gleted eer) two sizes. 
Globes—Physical, 15” diameter. 
Globes—Political. 
Globes—Slated. 
Tellurians. 
Atlases. 

PICTURES 
Land Forms, 24 Illustrations. 
Physical Phenomena, 16 pictures, 
Plants and Trees of Commerce, 30 pictures. 
Types of Nations, 11 pictures. 


Geographical Types and Scenes: 
series, 70 pictures. 


two 


Tell us what subject you are interested in and we will gladly give you information 
about any item in our line that may meet your needs. 


USE THIS CONVENIENT COUPON TODAY 


I am interested in the subject checked below and request literature describing ee 


maps, etc., applying to the same. 
O Political Geography 

Li Commercial Geography 

1) Physical Geography 


O Other Items as follows: 


~ A. JI. NYSTROM & CO. 


Go22) 


1 Pictures 
O No. AG21 Geography Catalog 


2249 Calumet Ave. 
CHICAGO 


a 4 
PHYSIOORAPHIC DIAGRAM 
OF THE ; 


' 
UNITED STATES | 


AM Lopece 


enivanerry OF Shestaty 
woe 


SS at n 
AJ.Nrstrom 6 Co. 


muses = CCN 


PHYSIOGRAPHIC DIAGRAM OF 
THE UNITED STATES j 


Size 6244 inches, Scale 50 miles to one inch 


By A. K. LOBECK 


Department of Geography, University of Wisconsin 


This diagram is intended to be used asa guide in teaching the physiography an 
geology of the United States. The topographic features are expressed with the ide 
of revealing or suggesting the underground structure, and the general character ¢ 
the underlying rock. 


An introductory drill such as having students draw cross sections through tt! 
Adirondack uplift, the Cincinnati or Nashville Domes, across the Appalachians 1 
different places, through the Black Hills, Central Texas, the Ozarks, Wisconsin an| 
many other regions is followed up with the use of other available material, espt’ 
cay state geological maps and the United States Geological Survey topograph 
sheets. i | 


The student frequently fails to note the significance of a locality with relation 1 
the physiographic province or geological region as a whole. This is overcome b 
requiring the student to locate upon the diagram the position of the topograph’ 
quadrangles under consideration. As practically every topographic map publishe 
by the Survey was consulted in the preparation of the Diagram, most of the signif! 
cant features are recognizable inthe Diagram too. Seventeen geological provinct’ 


are discussed briefly on the map. 
PRICES 


Student edition (one color) with one-half of U.S. on each side of sheet, size 31X?; 


| 


| 
4 


inches. 4f 
Singleicopysacen ses oee $1.25 Fifty copies, each......... $1.00 
Ten copies, each...... I.10 One hundred copies, each.. .90 
Two color edition with entire map on same side of sheet, size 62 X44 inches. 
Sing lelcopyaaaeeeeeere $2.25 Fifty copies, each......... $1.75 } 
Ten copies, each...... 2.00 One hundred copies, each.. 1.50 | 


Prices on two color edition, mounted on cloth for permanent class room usé, Sli 
6244 inches. } 
With Common Rollers at Top and Bottom..................- $4.50 
On Spring Roller and Board with Dust Proof Cover........... 7.75 
In ‘Steel'Spring Roller ‘Cases... go. oc oa) oie eee 8.75 | 


A. J. NYSTROM & COMPANY 


Publishers 2249 Calumet Ave., Chicag' 


Goode’s Base Maps 
and Graphs 


Prepared by 


J. PAUL GOODE 


Professor of Geography in the 
University of Chicago 


A series of outline maps for all classes of work: in every grade from 
the common school to the university, in all branches of study, and 
for private use in applied sciences and the various fields of research. 


The desk maps come in three sizes, from 8X 103 to 21X15 inches, 
all perforated for insertion in a standard notebook cover with page 
8104 inches. The larger sizes fold into this cover. Several wall 
outline maps are issued, 48 X 36 inches. 


SS 


ig ae 


ai els — 


BX ae aS 
[] cian ae 
eee 


TAMERS 
bee cae 


incon 
EEaRE et cae 
aa 


ney aaa 

AS can, 
ee WEEE 
SY WZ 


N We S ee 


= 


“| 


Miniature of Map No. ror Hc. The homalographic projection, interrupted to aaa me Roninans 
The entire earth’s surface is shown without distortion of area 


The maps are made from the best drawings, printed with a good quality 
of ink on bond paper, and the presswork is of the highest order. Recent 
issues have been made by the photo-lithographic process, thus preserving 
an absolutely smooth surface of paper on which to work. These maps 
are considered by the many who use them as the best quality of base 
maps on the market. 

Complete list with prices will be sent upon request 


The University of Chicago Press 
5750 Ellis Avenue Chicago, Illinois 


Leitz Petrographical Microscopes 


NEW MODELS—WITH ENLARGED FIELD OF VIEW 


Microscope ‘‘S.M.” with synchronic 
rotation of the nicols 


These microscopes have been developed 
on the basis of latest suggestions from 
petrographists. Their novel and modern 
design will prove a great aid in petro- 
graphical studies. 


Every leading petrographist knows the 
superior optical and mechanical qualities 
of Leitz microscopes and we therefore 
are fully safe if we refrain from enumer- 
ating the many features. 


The variety of models we offer afford 
a suitable equipment for every specific 
requirement. 


The complete assortment of our petro- 
graphical accessories will interest those 
who desire to add to their present 
equipment. 


Some of the prominent features of the new model 
microscopes are: 


1. Improved Objective Clutch Changer which guarantees the objective to be 


permanently centered. 


2. Anastigmatic Tube Analyzer which permits a most crisp definition. 


3. Special Illuminating Apparatus, whereby a wider range of application is 


afforded. 


4. Enlarged Microscope Tube which accommodates eyepieces of wide diameter, 
whereby the field of view is increased 50% over other types of instruments. 


Ask for Catalog II-B (F) 


New!S NOTES 
SURPASSED “or EQUALLED 
ms SUPREME 


60 EAST lO! St. 


VOLUME XXX NUMBER 6 


[alle 


HOURNAL OF GEOLOGY 


September-October I 922 


THE HOT WATER SUPPLY OF THE HOT 
SPRINGS, ARKANSAS? 


KIRK BRYAN 
U.S. Geological Survey, Washington, D.C. 


CONTENTS 
INTRODUCTION 


ACKNOWLEDGMENTS 

CHARACTER OF THE Hot WATERS 

GEOLOGIC SETTING OF THE HoT SPRINGS 

GEOLOGY OF THE SPRINGS AREA 

RELATION OF THE SPRINGS TO THE WESTERN BoDY OF SANDSTONE 
THEORIES OF ORIGIN OF THE HoT WATER 

JUVENILE vs. METEORIC WATER 

ANALYSIS OF THE MERITS OF THE HYPOTHESES 

UsE AND DEVELOPMENT OF THE Hor SPRINGS WATER 


INTRODUCTION 

An investigation of the geology of the hot springs in the Hot 
Springs National Park was recently undertaken for the purpose of 
determining whether the supply of hot water can be increased. 
Some new facts were obtained, earlier work critically examined, 
and recommendations made. ‘The problem is intimately related 
to the ultimate origin of the water. Thus an investigation begun 
solely for economic reasons led to a consideration of one of the most 


t Published by permission of the Director of the United States Geological Survey, 
425 


426 KIRK BRYAN 


intricate and uncertain realms of geologic theory. Is the hot 
water of meteoric, juvenile, or mixed origin? On the answer 
to this question depends, in a measure, the future of Hot Springs, 
Arkansas. 

ACKNOWLEDGMENTS 


To the officials of the National Park Service, through whose 
co-operation the work was undertaken, the writer is indebted for 
the opportunity of making this study. The local officers of the 
Park Service and Colonel John R. Fordyce extended numerous 
courtesies in Hot Springs. Messrs. H. G. Ferguson, O. E. Meinzer, 
Clyde P. Ross, and W. D. Collins have read the manuscript and 
offered valuable suggestions. Mr. H. D. Miser has generously 
allowed the use of much unpublished material from his extensive 
researches in the geology of Arkansas. 


CHARACTER OF THE HOT WATERS 


The waters of forty-six springs have been analyzed by Haywood.* 
The mineral contents vary from 170 to 310 parts per million, and 
in only a few of the springs do the mineral contents fall below 270 
parts or rise above 290 parts. Silica is an important constituent 
ranging from 32.5 to 52.3 parts per million but being usually be- 
tween 44 and 47 parts. Calcium (Ca) ranges between 26 and 50 
parts per million while the bi-carbonate radicle (HCO) ranges 
between 94 and 172 parts per million. The excess of carbon 
dioxide is satisfied by small amounts of magnesium, potassium, and 
sodium. ‘The sulphate radicle ranges from 6 to 28 parts per million, 
and chloride from 2.36 to 3.33. Their salts therefore form only 
a small part of the total solids. Small quantities of manganese, 
traces of phosphorus, of combined nitrogen, iron, and aluminum are 
present. Boron, iodine, and bromium are reported as small quan- 
tities or in traces. 

The waters of two cold springs which are located at the pavilion 
north of the Arlington Hotel have a mineral content of 36.4 and 
43.7 parts per million. ‘The water is similar to the hot water except 


1J. K. Haywood, Report of an Analysis of the Waters of the Hot Springs, etc. 
Sen. Doc. 282, 57th Congress, rst Sess. (1902), pp. 1-78. 


HOT WATER SUPPLY OF THE HOT SPRINGS 427 


for lower mineralization and greater proportionate content of silica 
and magnesium. 

The contrast in mineral content between the hot water and the 
two cold springs mentioned above seems to be general in the region, 
and in a later paragraph the available temperature measurements 
are discussed. In the table below the water of one of the hot 
springs is compared with four other springs in Garland County. Big 
Chalybeate, Mountain Valley, Blanco, and Dripping springs are 


ANALYSES OF SPRING WATERS IN GARLAND COUNTY, ARKANSAS 
(Parts per million) 


Constituents A B @ D E F G H I J K 
Silicay(Si@2) ee nese. 45.6 3.8 | 16 22 14 Tah || Ly Aa 6.5 Boi | 1 I5 
Iron (Be)ijc oss. c cess i : 7.4 .2 53 moe iif \ .2 |f AG ll Ons 
Aluminum (Al)...... here eiltl peevsieyecs 1.2 3 Gta Nl. Dose | shes sinh), PEMse cea 35 
Calcium (Ca)........ 46.9 | 70 78 83 76 I.9 3.8 Bok Rep | Bos I.4 
Magnesium (Mg)....| 5.1 Arta |er2) 8.4 2.9 I.4 I.4 aii iebetcsaners 2.4 a5 
Sodium) (Nahe. e An, I.4 6.5 3.9 2.1 2.1 2.2 1.5 |f 3} a 2.4 
Potassium (K)....... 1.6 Bott 35 Boe 5 9 I Bull\ I 9 
Bi-carbonate radicle 

(HICO) Warren cise 168.1 |260 284 288 228 ViA.ae | pone |) ats) 8 2.4 9 
Sulphate radicle (SO,).| 7.8 9.4 | 25 8.2 | 12 2.5 2.3 BEM Eeeelewazaye 21 16 
Chloride radicle (Cl)..| 2.5 2 4.4 7 6.2 1.8 2 2.7 Ack) || S05} 4.4 

Total solids (cal- 

culated)....... 198.5*|229.7 |283 278.7 1226.8 | 42 4I 28.3 | 17.6 | 40.5 | 52.4 


* Total solids determined. 


EXPLANATION OF TABLE 


A. Big Iron Spring, No. 15 of Hot Springs group. Smallamounts of nitrogenous material; POs, .o5; BOs, 
1.29; Brand I, trace; Ba and Sr, trace; Li, trace; gases, nitrogen 8.8, oxygen, 3.79, carbon dioxide 
(free) 6.92, cubic centimeters per liter at o° C. and 760 mm. pressure. Analyst, J. K. Haywood, 57th 
Congress, Sen. Doc. No. 282, p. 46. 

B. Big Chalybeate Spring; NW. 4, SE. 4, Sec. 22,T.2S., R.19 W. Analyst, A. E. Menke. Reported 
~ by us C. Branner, “Mineral Water of Arkansas,” Arkansas Geol. Survey Ann. Rept. 1891, Vol. I (1892), 
p. 28. 

. Mountain Valley Spring, Sec. 19, T. r S., R. 19 W., 12 miles north of Hot Springs. Reported by 
Branner, zbid., p. 69. 

. Blanco Spring, NE. 3, Sec. 1, T. 2.S., R. 21 W. Reported by Branner, zbid., p. 30. 

. Dripping Springs, one of Grandma Gone s springs, 6 miles northeast of Hot Springs. Reported by 
Branner, ibzd., p. 48. 

. Liver Spring. Cold spring in pavilion north of Arlington Hotel. Small amounts of nitrogenous 
matter; PO,, trace; BOs, trace; Brand I, trace; Li, trace; gases, nitrogen 14.36, oxygen 6.24, carbon 
dioxide (free) 21.8, cubic centimeters per liter at 0° C. and 760 mm. pressure. Analyst, J. K. Haywood, 
op. cit., D. 75. 

5 Kidney Spring. Cold spring in pavilion north of Arlington Hotel. Small amounts of nitrogenous 
matter; PO,, BO, Br, I, and Li, traces; gases, nitrogen, 15.3, oxygen 5.3, carbon dioxide (free) 28.5. 
Analyst, J. K. Haywood, op. cit., p. 70. 

. Happy Hollow Spring, 600 yards north of Arlington Hotel. Reported a Beene op. cit., D. 52. 

. Same as above. Analyst, R. B. Riggs. Reported by Branner, op. cit., 

. Red Chalybeate Spring, one of Sea Chase’s springs. NE. 4 iD NE. oe Bee. Gi) ANE BF Shy IS te) Whe 
Reported by Branner, op. cit., 

. Happy Hollow Chalybeate, ee (feet west of Happy Hollow Spring. Reported by Branner, of. cit., 
Pp. 54. 


i) st jellel © 


A oe 


428 KIRK BRYAN 


all relatively strong springs, though none of them are considered 
to be “hot” springs. The two cold springs in the pavilion, Liver 
and Kidney, are small seeps and their waters are similar in type 
and total content with Happy Hollow, Happy Hollow Chalybeate, 
and Red Chalybeate springs. ‘The last-mentioned spring was not 
seen but the two Happy Hollow springs and the springs in the 
pavilion have their origin in the storage of rain water in soil, talus, 
and the upper fractured part of the underlying rocks. There 
seems then to be a notable difference between the shallow meteoric 
waters and the waters of larger springs. 

In 1904 Boltwood' determined the radioactivity of samples from 
forty-four springs. He found no evidence of radium salts in the 
water and attributes the radioactivity to the presence of radium 
emanation, a gas. The intensity of radioactivity varies from 0.5 
to 265.8, a numerical expression for the equivalent uranium repre- 
sented by the radium emanation (gX107* U). ‘There are, there- 
fore, great differences in the radioactivity of the springs, but 
their average intensity is 24.9. The cold springs north of the 
Arlington have activities of 17.4 and 106.8. The spring having an 
activity of 106.8 is exceeded by only one of the hot springs, and the 
other has an activity not far from the average of the hot waters. 
Boltwood says: ‘‘As a general summary it can be stated that it 
has been found impossible to establish any connection between 
the temperature, flow, location, or chemical composition of the 
water of the springs and the observed differences in the radioactive 
properties.”” 

Previous observations on temperature have been reviewed by 
Weed: 

In 1804 Dunbar and Hunter recorded a temperature of 1oo° F., for the 
larger spring and 154° F. for another spring... . . The comparison of the 
old records with those recently made shows that the highest temperature 
known today is 147° F. as against 154° in 1804, and 150° by Glasgow and 148° 
by Owen in 1860. In a number of springs there is a decline of 2° since the 
latter date. Such a slight difference might, however, be due to differences 
in the manner or place of taking temperatures, or the instruments used in the 


t Bertram B. Boltwood, Aun. Rept. Secy. of Interior, 1904; also Amer. Jour. of Sci., 
4th Ser., Vol. XX (1905), pp. 128-32. 
2 Iaith (5 BEA 


HOT WATER SUPPLY OF THE HOT SPRINGS 429 


earlier years may not have been accurate. In one instance, that of Alum 
Spring, there is a marked decrease in temperature. ... . In 1804 this had a 
temperature of 132°. In 1859... . 133° . and today it is but 114.8°.t 


Haywood? gives temperature measurements for forty-four 
hot springs which range from 95.4° to 147° F. For thirty-nine 
springs he gives two measurements each, separated by about two 
months’ time. In fourteen of these thirty-nine springs there is a 
decrease in temperature, in eighteen there is an increase, and in 
seven there is no change during this interval of about two months. 
The average difference between readings is 1.5°F. The maximum 
decrease is 6.3° F. and the maximum increase 6.4° F. With such 
unsystematic discrepancies in the measurements of a competent 
observer with good instruments, no conclusion can be reached as to 
a general decrease in temperature or to the character of the probable 
variations in temperature. 

The Hot Springs are usually considered to be the only hot springs 
of the region. ‘Three warm springs, however, are known from the 
vicinity of Caddo Gap, about 50 miles west of Hot Springs. Data 
concerning these springs collected in 1915 by H. D. Miser are given 
in the following table: 


SPRINGS NEAR CADDO GAP 


Temperature 
Name Location Geological Formation (Degrees 
Fahrenheit) 
Springs in bed of Caddo 
River at Caddo Gap.....| NE. 4, Sec. 19 T. 4S., | Upper part Arkan- 
; R. 24 W. sas novaculite 
INGmEn, COC MUTN ER) 566.5 A allawe OMe aia ee on elas cme lame dbo oo ceeeceg era 04 
SOUCMBO PE MIM Cem tena eer dock Caters sil tulatmall sat apatites Ree Mente Oy ge 96.8 
Spring on Little Missouri 
TRIRVEIR ec gee Greene eo tanere eaten IN Geo eris leony fal baw eta )ary Mee lfeheeta as eta a Pre. cine At tate 74.3 
R. 27 W. 
Spring on Redland Moun- 
(TUT id ee SW. i, Sec. 12, T. 5 S., | Arkansas novacu- 77.0 
IR, AD We lite 


tW. H. Weed, “‘Notes on Certain Hot Springs of the Southern United States,”’ 
U.S. Geol. Survey, Water-Supply Paper 145 (1905), pp. 204-5. See p. 439 of this article 
for references to Weed’s authorities. 


2 OD Clin \ODo HOB 


430 KIRK BRYAN 


Certain springs near Hot Springs conform to and others are 
above the mean annual temperature of the air at Hot Springs 
which, based on the thirty-year record of the United States Weather 
Bureau, is 60.5° F. To this temperature the water of ‘‘ordinary”’ 
springs should closely approximate. Springs above the normal 
temperature are probably common as shown by the table below: 


TEMPERATURE OF GARLAND COUNTY SPRINGS 


Degrees 
Fahrenheit 
BiguGhalybeate? «cae ol ee. see 78.9 
Grandma Chase’s Springs:* 
1D hay oy opooy atop oy Ahoy AC eNen ate alae lens HMM cet id had con 59.2 
Redi@halybeate Sprig emacs tee eee 62.8 
Happy Hollow, Chalybeates quan ae ee 64.6 
(Not Happy Hollow Spring) 
Potash Sulphur Springs:* 
WieSteS DEINE cations tinct aia ete aie pare 64 —71.6 
South: Springsciaacahoacraebyc ec eee 70.2—72 
Hast Spring. ceca sae, 7 ae eee 68 -69.8 
Springs in Paviliont north of Arlington Hotel: 
Dhiver Springs. shee eee cecea ary cease een eae 46.4 
Kidney Spring. sou aa fa ase ane ne aera 55-4 


* J. C. Branner, op. cit., pp. 28, 48, 50, 54, and 77-81. 
7 J. K. Haywood, op. cit., pp. 75 and 76. 


GEOLOGIC SETTING OF THE HOT SPRINGS 


Most of the following discussion of the general geology of the 
region is condensed from a paper by Miser’ and from the manu- 
script of a geologic folio by Purdue and Miser,” to be published by 
the United States Geological Survey. The geologic map, Figure 1, 
is largely a redrawing of the map in this folio. The Hot Springs 
are situated in that part of Arkansas known as the Ouachita Moun- 
tains. These mountains are composed of numerous nearly east 
to west ridges and several intermontane basins. Some of these 
mountains are simple ridges, but others are small ranges. The 


1H. D. Miser, “Llanoria, the Paleozoic Land Area in Louisiana and Eastern Texas,” 
Amer. Jour. of Sci., 5th Ser., Vol. II (1921), pp. 62-89. 

2 Purdue and Miser, ‘“Hot Springs and Vicinity Quadrangle Geol. Atlas of U.S.,” 
U.S. Geol. Survey, folio, in preparation. 


HOT WATER SUPPLY OF THE HOT SPRINGS 431 


Hot Springs are located in the southern edge of one of these ranges 
called the Zigzag Mountains, and on the northern border of a low- 
land called the Mazern intermontane basin. 

The rocks of the Ouachita Mountains are nearly all of sedimen- 
tary origin, but at Magnet Cove and Potash Sulphur Springs there 
are small areas of igneous rocks and at numerous localities near by 
there are small dikes. 


“ 
oe 


S 
See 


a —Be aan Se 
pres ter LNG 
MME ogat tise hiv: 


EXPLANATI 
SEDIMENT 


a! 


BS 


| 
N 
IRN? 


Stanley shale Hot Springs sandstone Arkansas novaculite Missouri Mtn.shale 
IGNEOUS 


WA ISS 


Polk Cr. shale Bigfork chert Dikes and sills 
° 1 2 3 MILES 


LA 


>O 


Fic. 1.—Geologic map of vicinity of Hot Springs, Arkansas, after Purdue 
and Miser. 


The sedimentary rocks are indurated and hard, but are only 
slightly affected by metamorphism. The maximum thickness 
of the rock beds exposed in the Ouachita Mountains is 37,000 feet, 
but only a fraction of this total is exposed in the vicinity of the 
Springs. 


432 KIRK BRYAN 


The rocks exposed near the Hot Springs consist of the following, 
though both older and younger are known in the Ouachita Moun- 
tains: 


GENERAL SECTION OF ROCKS NEAR HOT SPRINGS, ARKANSAS 


Geologic Age Name of Formation and Description Thickness, Feet 
Stanley Shale; black, fissile, clay shale, and hard 
; compact sandstones a). aa 3500-5 
Carboniferous 


Hot Springs Sandstone; hard quartzitic lami- 


Oiississippian) nated gray standstone with heavy bedded con- 
glomeératemt the bases... 55.0.5 4. eee 200 
Unconformity 
Devonian Arkansas Novaculite; upper half mainly thin- 
bedded novaculite and black shale; lower half 
MASSIVE NOVACUIILE.-.yers. ees ae ae eee 500+ 
Unconformity(?) 
Silurian Missouri Mountain Shale; clay shale generally 
dark greenish drab to black but red in many 
DIACES Cts esses een ae eee 150 
Unconformity( ?) 
Polk Creek Shale; black graphitic shale in which 
Oninnoan praptolites|are abundant. a. 40 4: aoe 200 
Bigfork Chert; thin-bedded gray to black chert 
much shattered and black shale............ 700 


The rocks mentioned above were deposited one above the 
other in great sheets. Since their deposition they have been sub- 
jected to intense lateral compression which besides lifting the area 
has produced folds of a general east and west trend. Near the 
springs these folds have a northeast and southwest trend and the 
edges of the strata now appear at the surface, and on the map form 
great looping curves. The major folds consist of numerous smaller 
folds only a few miles in length, overlapping each other lengthwise. 
It is with these smaller folds that the springs are associated. 

The structures which have the most to do with theories of the 
origin of the spring waters are the anticlinal fold whose limbs 
inclose the valley between West, Indian, and Sugarloaf mountains, 
the synclinal fold of North Mountain, and the anticlinal fold of 


HOP WALER SUPPLY OF THE HOT SPRINGS 433 


Hot Springs Mountain. The character of these folds is brought 
out in Figures 1 and 2. 


GEOLOGY OF THE SPRINGS AREA 


The hot water rises in an area of about 20 acres that lies along 
the east side of Hot Springs Creek, at the southwest base of Hot 
Springs Mountain. One spring lies west of the creek. Five 
are said to have risen in the bed, though only one of these can now 
be found. The spring area is marked by a deposit of calcareous 
tufa (travertine) from a few inches to eight feet thick over the 
older rocks. To the tufa the springs are daily making additions, 
though the present structures for collecting the waters have reduced 
the rate of formation of the tufa. 


“A 13) - \ 
EO EN » 
NW EG: 2 Bios Wh SE 
SP EON ON 
H BS Ee Wa Hot Springs Mtn. 
6 EO WNEEE a 
1500’ oY pro Meo~ 
f/f 7% 427 


1000 Stoney shale,’ Le 
Sea level Y 


Fic. 2.—Geologic cross-section from Sugarloaf Mountain to Hot Springs 
Mountain (line H-H’, Fig. 1), after Purdue with modifications. 


_. The grounds and springs were carefully mapped by Captain 
R. R. Stevens, U.S.A., in 1890, and he mapped the tufa, hard-rock 
outcrops, and springs. Figure 3 reproduces his boundaries for 
the tufa and for rock outcrops, except that corrections in the rock 
outcrops have been made at critical points during this investigation. 
On this map (Fig. 3), the boundaries of the geologic formations have 
been traced. Landscape gardening, roads, walks, and buildings 
all tend to conceal outcrops and in a number of places, as stated 
below, the location of geologic boundaries is uncertain. 

The Hot Springs sandstone outcrops on Fountain Street, in 
Happy Hollow, where it is nearly vertical. From this point it 
extends along the foot of Hot Springs Mountain southwesterly 


434 KIRK BRYAN 


HOT SPRINGS 
ARKANSAS 


EXPLANATION 
Outerop of Hot Springs sandstone 
wf Ovtcrop of Arkansas novaculita 
SY Ovtcrop of Stankey shale 
Ca) Outcrop of Tof 


——— fissures parallel to the fault 
© §6<Springs and shallow wells 
0 Drilled well 
SCALE 
2 WO 2p0 3p0 490 SO FEET 


i! 


---5, Soe ees 
H's | “He> 

t 

{i Oey 


Fic. 3—Map of the spring area, base and outcrops from Captain R. R. Stevens, 
U.S.A., 1890. 


HOT WATER SUPPLY OF THE HOT SPRINGS 435 


to the Arlington Hotel and then swings in a broader north and 
south belt to Reserve Avenue. In this broader belt it has a dip 
of about 30° and the contact with the overlying Stanley shale is 
well displayed in the basement of the Maurice, Fordyce, and 
“new” bathhouses. The lower contact of this body of Hot 
Springs sandstone extends along the hillside above the springs 
from the vicinity of the nurses’ dormitory in the hospital grounds 
northwest. In general this contact can be located within 25 feet. 

South of Reserve Avenue there are no outcrops of the sandstone, 
nor are there any in the western part of the hospital grounds. 


ARMY AND NAVY HOSPITAL, TANK 
EL, 780° 


5 d 
: ZEN Yip 
A a LLG YY YU 
5 LL MMM 
CENTRAL 5 LILLE ES /Abkcnsas /if, sis 
cu g00 i LL LILLIE | pny 
g Vy, 


nee 
Z g 2 ZA a 
| BPA LLL. 
GLE LLL ZA LOL FA Zi HORIZONTAL AND VERTICAL SCALE 
: 2 100 200 * 300 FEET 


Fic. 4. Geologic cross-section on line AB, Figure 3, through Fordyce well 


Since the sandstone is hard and commonly produces outcrops it 
is assumed that the sandstone does not extend southward any 
appreciable distance beyond the park line. 

This western body of sandstone is the northwestern limb of the 
Hot Springs Mountain anticline which in its extension along Bath- 
house Row forms the nose of the plunging structure. Figure 4 
is the cross-section of this part of the mountain on the line AB 
which shows the relation of this body of sandstone to the other 
rocks. 

A second body of Hot Springs sandstone begins on the southeast 
flank of Hot Springs Mountain and extends southwesterly within 
the hospital grounds. This eastern body of sandstone dips north- 


436 KIRK BRYAN 


westerly and is underlain by the younger Stanley shale and overlain 
by the older Arkansas novaculite. Obviously this body is the 
overturned southeastern limb of the Hot Springs Mountain anti- 
cline. Figure 5 is a cross-section of this part of the mountain 
and shows the relation of this body of sandstone to the other rocks. 

The relation of these two bodies of sandstone once a continu- 
ous layer is somewhat uncertain because of the lack of outcrops 
in the western part of the hospital grounds. If the two bodies are 
continuous, an extremely close fold is necessary to bend the bed 


Cc RESERVOIR 
EL.780 


—PROPOSED WELL 


RESERVE AVE. 


HORIZONTAL AND VERTICAL SCALE 


@ 100 200 GOO FEET 


Fic. 5.—Geologic cross-section on line CD, Figure 3, through proposed cold- 
water well. 


from a position dipping 30° west at the superintendent’s office 
to 65° northwest near the reservoirs on the hospital grounds, goo 
feet away. It seems likely that instead of bending, the beds broke 
along a thrust fault and this interpretation is shown in Figures 1 
and 3. This postulated fault would have a plane which dips to the 
northwest and a trend about 30° east of north. 


RELATION OF THE SPRINGS TO THE WESTERN 
BODY OF SANDSTONE 


As shown in Figure 3, all the hot springs shown on maps since 
1890 or now in existence emerge from the outcrop of the western 
body of sandstone or from the immediately adjacent Stanley 
shale. Similarly all the bodies of tufa which indicate the position 


HOT WATER SUPPLY OF THE HOT SPRINGS 437 


of springs active or formerly active lie on the sandstone or on the 
immediately adjacent Stanley shale. 

The Hot Springs sandstone of this body is somewhat harder 
than normal, is a darker color, and is much fractured. These 
fractures are commonly sealed with quartz and calcite, both appar- 
ently deposited by the hot waters. In a number of instances, 
notably in the Maurice and “‘new”’ bathhouses, where excavation 
has exposed the sandstone, the hot water can be seen emerging 
from the cracks and fissures of the rock. 

The fractured sandstone is then the conduit for the hot water 
which is prevented from breaking out in the lower depression 
(the creek bed) by the Stanley shale through which it maintains only 
a few openings. Similarly the water closes the cracks and joints 
of the sandstone by deposition and consequently all the springs 
do not break out at the contact of the sandstone and Stanley shale 
but many of them emerge higher up the hill. 

On the hillside, back of the bathhouses, the outcrops of sand- 
stone are each extended in a northeasterly direction. Each outcrop 
in addition is marked by strong, nearly vertical jointing in this 
direction. The maps show also that the spring openings are 
arranged in lines of which the most marked is that belt of springs 
from the Egg Spring, No. 1, to the Maurice Bathhouse, which 
includes the strongest springs of the group. Four such lines of 
springs are marked on the map. These lines are approximately 
parallel to the thrust fault, postulated between the east and west 
bodies of the Hot Springs sandstone. A fifth line may be drawn 
parallel to the contact of the sandstone with the shale, but this 
line includes many springs situated on other lines. Obviously, if 
the sandstone were uniformly permeable and the shale uniformly 
impermeable, all the springs would lie on the contact. 

The strong jointing in the sandstone, and the distribution and 
elongation of outcrops indicate a fracturing of the sandstone in a 
direction north northeast parallel to the Hot Springs anticline. 
Weed? noted the line of springs extending from Egg Spring, No. 1, 
to the Maurice Bathhouse and suggested that this line was a 
“fault fissure.” It seems more likely that this line and the three 


1 Op. cit., p. 201. 


438 KIRK BRYAN 


other lines of springs simply mark the position of more open joints 
like the joints visible in the outcrops but that all the joints are 
parallel to and related in origin to a thrust fault which lies to the 
south of them. Doubtless similar cracks might have been formed 
by simple folding. However, the very absence of outcrops in the 
western part of the hospital grounds seems to be an argument in 
favor of faulting which would shatter and comminute the rocks along 
the line of the fault and thus make them more susceptible to 
erosion. Folding without faulting, on the other hand, would givea 
a double thickness of sandstone which would be almost sure to 
outcrop, and this same folding would presumably so shatter the 
sandstone as to make the line of the fold the locus of springs. 

Whatever the ultimate origin of the water, it emerges through 
the cracks and joints of the Hot Springs sandstone and mainly 
along the strong jointing parallel to and probably related in origin 
to the postulated thrust fault. The bearing of these relations 
on development of the springs is obvious. 


THEORIES OF ORIGIN OF THE HOT WATER 


In 1804 William Dunbar' and Doctor Hunter visited the 
springs. They observed that the mountain was “principally 
siliceous, some part of it being of the hardest flint, others a free- 
stone extremely compact and solid and of various colors. The 
base of the hill, and for a considerable extent, is composed of a 
blackish blue shistus, which divides into perpendicular lamina, 
like blue slate.’”’ They make extensive comments on the tufa 
deposited by the water. They estimated the flow of all the springs 
at 165 gallons per minute or 237,600 gallons daily. They suggested 
chemical reactions as the cause of the heat of the water, having 
found no evidence of volcanic action in the vicinity. 7 

In 1806 a writer relates that he saw a volcanic outburst and 
streams of molten rock near Hot Springs. He is generally dis- 
believed by later writers.” 


Thomas Jefferson, Message of the President of the U.S. Communicating Dis- 
coveries Made in Exploring the Missouri, Red River, and Washita by Captains Lewis and 
Clark, Doctor Sibley, and Mr. Dunbar, etc. A. & G. Way, printers, Washington, 1806. 


2 New Vork Medical Repository, Vol. III, No. 1 (1806), pp. 47-50. 


HOT WATER SUPPLY OF THE HOT SPRINGS 439” 


In 1860, David Dale Owen,’ state geologist of Arkansas, pub- 
lished an account of the springs with analyses and observations on 
temperatures. He rejects all chemical theories of origin of the 
heat. 


On the contrary, I attribute the cause of it to the zmternal heat of the earth, 
I do not mean to say that the waters come in actual contact with fire, but 
rather that the waters are completely permeated with highly heated vapors 
and gases which emanate from sources deeper seated than the water itself. 


Owen believed that the novaculite was a sand rock which had 
been changed by the ‘‘permeation”’ of heated alkaline waters and 
considered the hot springs merely the dying phase of this extensive 
movement of water. He gives, however, no mechanism or conduit 
for these waters. 

In 1892 J. C. Branner,? state geologist, discusses the origin of 
the heat and attributes the heat of the water to “coming in con- 
tact with the masses of hot rocks, the cool edges of which may or 
may not be exposed at the surface.”’ 

In 1902 Walter Harvey Weed published a geological sketch 
of the hot springs. Weed noted that the principal springs are 
arranged along a line running NNE., parallel to the axis of the 
fold forming Hot Springs Mountain. He thought this a fault 
fissure. Fissuring in connection with faulting seems confirmed 
-(p. 438). Weed considered that the purity of the waters, par- 
ticularly their low content of silica, and the included gas which 
appears to be dissolved air, all point to a meteoric origin of the 
water, i.e., that the water is derived from rain and differs from 
ordinary spring water only in being heated. He believed this 
heat to be derived from still uncooled igneous rock intruded into 
the sediments below the springs. The upper parts of similar bodies 

* David Dale Owen, Second Report of a Geological Reconnaissance of the Middle 
and Southern Counties of Arkansas, etc., pp. 18-27, Philadelphia, 1860. 


2 J. C. Branner, ‘‘Mineral Waters of Arkansas,” Arkansas Geol. Survey, Ann. Rept., 
1891, Vol. I (1892), pp. 8-23. 

3J. K. Haywood, Report of Analysis of the Waters of the Hot Springs, etc.; and 
Walter Harvey Weed, Geological Sketch of Hot Springs, Arkansas. Senate Doc. 282, 
57th Congress, 1st Sess. 1902. Also with modifications U.S. Geol. Survey, Water- 
Supply Paper 145 (1905), pp. 189-206, and separate by Interior Department, 1912. 


440 KIRK BRYAN 


are exposed at Magnet Cove, Potash Sulphur Springs, and as 
dikes in the vicinity of the Hot Springs. 

In 1910, Purdue’ published an elaborate paper on the origin of 
the hot water. He follows Weed in believing that the water has a 
meteoric origin and that in its passage through the ground derives 
heat from uncooled masses of igneous rock. He goes a step farther 
and outlines the structural conditions for the collection and trans- 
mission of the water. He believes that the water falls as rain in 
the anticlinal valley between North and Sugarloaf Mountains, 
where it is absorbed by the Bigfork Chert. ‘The considerable 
thickness of this chert, its much fractured nature, and the thin 
layers of which it is composed all combine to make it a water 
bearing formation of unusual importance.” 

The water having been collected in this formation is confined 
by the impermeable overlying Polk Creek and Missouri Mountain 
shales. ‘Thus confined the water is conducted beneath the syncline 
of North Mountain, where it most probably comes in contact with 
some uncooled mass of igneous rock. Purdue ‘suggests, but 
rejects the hypothesis that the water is expelled by the cooling 
of such an igneous mass. 

Lindgren,’ in 1919, accepts Purdue’s views and considers that 
the springs have “clearly derived their saline constituents from the 
surrounding sedimentary rocks.” 

Another hypothesis should be advanced. On this hypothesis 
the water is of deep-seated origin derived from a covered mass of 
igneous rock intruded into the sediments, but not showing at the 
surface, which discharges water expelled from its molten interior 
by the gradual crystallization of its mass, or the water is derived 
from a deeper less definite but similar mass and rises to the upper 
crust through a deep, probably fault, fissure. Such water is com- 
monly called juvenile, ic, new water coming to the surface for 
the first time. 

tA. H. Purdue, The Collecting Area of the Waters of the Hot Springs, Hot 
Springs, Arkansas,” Jour. of Geol., Vol. XVIII, No. 3 (1910), pp. 278-85, 3 figs. Also 


Indiana Acad. Sci. Proc. 1909, pp. 269-75, 3 figs. 1910. 
2 Waldemar Lindgren, Mineral Deposits, p. 90. New York, 19109. 


HOT WATER SUPPLY OF THE HOT SPRINGS 441 


JUVENILE VS. METEORIC WATER 


One of the great triumphs of modern geology has been to estab- 
lish that the majority of metalliferous ore bodies, including most 
quartz veins, are deposited by ascending aqueous solutions which 
are derived from and excluded from crystallizing igneous bodies. 

Granite rocks have been traced into pegmatite veins; pegma- 
tite veins into metalliferous quartz veins; metalliferous quartz 
veins into quartz veins without impurities. It has thus been 
shown that aqueous vapors and gases gradually cooling and purging 
themselves of many substances rise through the crust and approach 
the surface in a purer and purer state. There is no theoretical 
objection to cold water with a minimum of mineral matter being 
attributed to a juvenile origin from an underlying crystallizing 
igneous mass, except the difficulty of proof. The majority of 
geologists do not hestitate to ascribe ore deposits to deposition from 
juvenile water, yet they hesitate to ascribe a juvenile origin to 
water emerging at the surface. It is well then to examine the 
criteria on which a discrimination between these two classes of 
water can be based. 

Springs of small volume, and large variation in flow and temper- 
ature, can usually be referred to a meteoric origin. There are, 
however, many difficulties in determining the precise geological 
structure which gives rise to a particular spring. The requisite 
structures necessary for such a spring are: (1) an intake area, 
(2) a reservoir, and (3) a conduit to the surface. Under different 
geologic conditions the three requisites assume a multitude of forms 
and vary in size according to the hydraulic conditions. In a previ- 
ous publication’ twenty-four named varieties, divided into five 
groups, are described and illustrated. The field geologist, know- 
ing the many possible structures, may have difficulty in deciding 
on the right one for any particular spring because of lack of evidence. 
Deep weathering of the rocks and a mat of vegetation and vegetable 
mold are usual at springs and tend to destroy, locally at least, the 


t Kirk Bryan, ‘“‘Classification of Springs,’ Jour. of Geol., Vol. XXVII (1919), 
PP. 521-61, 26 figs. 


442 KIRK BRYAN 


evidence of structure. However, a group of springs of common 
origin can usually be identified with the geologic structure to which 
they are due. 

Springs of relatively large volume with little variation in flow 
or temperature present, especially if they are hot springs, difficult 
problems. Certain hot springs are undoubtedly of meteoric 
origin and depend for temperature on the descent of meteoric 
water from the surface into the crust and its rise, without great 
loss in temperature, to the surface. Such springs are due to the 
fracture, usually by faulting of the cover of a definite artesian 
structure, but unfortunately no adequate description of such a 
spring has yet been published. Buckhorn, Indian, and Willow 
springs in Antelope Valley, California, which served as examples 
of the fracture artesian’ type of spring, are not thermal. Nearby 
flowing wells, having water of similar chemical composition, are 
from 200 to 4oo feet deep. ‘The artesian circulation in this valley, 
does not go to great enough depths to yield hot water. 

Many springs of steady flow and high. temperatures arise in 
localities where it is impossible to postulate a structure which will 
receive the water at the surface, carry it to depths, and return it to the 
surface. Waring? found that of ninety-eight groups of hot springs in 
California, thirty-eight rise from granite or granitic rocks; of 155 
carbonate springs, some of which are above the normal temperature, 
thirty-two occur in granite or granitic rocks. In such rocks the 
hot waters must arise from below through deep fissures. In Cali- 
fornia there is a notable association of the springs with faults, to 
which the fissures may be attributed. From these deep fractures 
in the crust, juvenile water from underlying magmas or incipient 
magmas may arise or there may be admixtures of meteoric and 
even connate waters which have or may have a circulation due to 
obscure or unknown forces. Certainly it seems simpler to assume 
that the water is juvenile. Certain springs, such as those near 
the Fish Springs Range, Utah,3 are associated with faults of large 


t Kirk Bryan, op. cit., pp. 553-55- 

2 Gerald A. Waring, “Springs of California,” U.S. Geol. Survey, Waisr-Supele 
Paper 338 (1915), p- 154. 

3 Kirk Bryan, op. cit., pp. 533-35. 


HOMWADEK SUPPLY OF THE HOT) SPRINGS 443 


throw and recent age. ‘This association seems a definite indication 
of juvenile origin. Hot springs in volcanic regions are probably 
in part of mixed origin. The presence of uncooled or even molten 
rock near the surface makes easy the heating of meteoric water 
and its return to the surface. Doubtless some springs in volcanic 
regions have a wholly meteoric origin, as Hague’ has proposed for 
the geysers of the Yellowstone. Yet beneath volcanoes magmas 
are crystallizing and expelling water. It is inconceivable that all 
of this water is absorbed in chemical reactions or in the interstices 
of the rocks below the surface. Some of it, with its contained 
minerals and gases, must reach the surface. 

Springs with steady flow and without great variations in 
temperature or quantity, especially if they are hot, must then arise 
from some deep artesian circulation or be of juvenile origin. ‘The 
artesian circulation should be susceptible of proof on structural 
grounds. In the absence of such proof the indication of juvenile 
origin is very strong. 

Elaborate investigations of the chemical characteristics of 
water, with the object of discovering its origin, have so far proved 
disappointing. Sodium chloride and sodium carbonate waters 
from granitic rocks carry a strong presumption of juvenile origin 
since the ordinary springs of such regions have water of the calcium 
carbonate type.? But unusual substances such as boron and 
fluorine have been found in spring waters of such diverse types as 
to be without critical value. 


ANALYSIS OF THE MERITS OF THE HYPOTHESES 


The question of the ultimate origin of the water in the Hot 
Springs of Arkansas is not only of intense theoretical interest, but 
has practical bearings. If the water is juvenile there is presumably 
a constant supply, diminishing very gradually through the centuries 
in quantity and temperature. When all the water is conserved 
by adequate structures it is probable that no more can be obtained. 
If, on the other hand, the water has a meteoric origin, it is variable 

t Arnold Hague, “Origin of the Thermal Waters in the Yellowstone National 
Park,” Bull. Geol. Soc., Vol. XXII (1911), pp. 101-22. 

2W. Lindgren, op. cit., p. 64. 


444 KIRK BRYAN 


in quantity, fluctuating with the seasons or with the groups of 
years having heavy or light rainfall. Also if the intake area is 
adequate, heavy drafts on the springs as by pumping should reduce 
the quantity in the reservoir and increase the absorption of rainfall 
in the intake area. Such increase in the volume of water flowing 
through the system may decrease the temperature of the water, 
an important consideration from the standpoint of use. A meteoric 
origin implies an intake area and this area must be found and pro- 
tected from pollution. 

It should be confessed that the present state of the science of 
geology is so imperfect that a definite conclusion as to the ultimate 
origin of the water in the Hot Springs cannot now be reached. 
As pointed out by Weed, the absence of unusual substances in the 
waters, low mineralization, and a gaseous content of oxygen and 
nitrogen in the proper ratio to form air are facts which do not 
show any unusual or non-meteoric origin for the water. On the 
other hand juvenile waters by deposition might purge themselves 
of all unusual substances, though if they originally contained 
sodium chloride there are difficulties in accounting for the loss of 
this stable compound. The radioactivity of the water, or rather 
the fact that it contains radium emanation (a gas) as determined 
by Boltwood in 1904,’ is not of critical significance, for the amount 
of radioactivity is not unusual in wells and springs. 

The previous temperature determinations are analyzed by 
Weed who rightly considers that they do not indicate either 
decrease in the heat or fluctuation in heat. Similar conclusions are 
reached from previous measurements of the water. But the 
existing measurements of these factors are neither adequate nor 
systematic. Fluctuations in temperature and volume are easily 
determinable if they exist. The critical value of such measure- 
ments is so great that it is to be hoped that they will be made for a 
sufficient period to provide adequate data. 

The meteoric hypothesis calls for a structure to carry the water 
from the intake area at the surface to depths and then return it to 


: Bertram B. Boltwood, ‘‘Annual Report of Secretary of Interior, 1904,” Amer. 
Jour. of Sci., 4th Ser., Vol. XX (1905), p. 168. 


HOT WATER SUPPLY OF THE HOT SPRINGS 445 


the surface. As postulated by Purdue,’ the water falls on the Big- 
fork Chert in the anticlinal valley between West and Sugarloaf 
mountains, is absorbed in the chert and passes below the syncline of 
North Mountain and arises in the anticline of Hot Springs Moun- 
tain. The course of the water is shown in Figure 2. The contact 
of the Bigfork Chert with overlying beds along the southwestern 
base of Sugarloaf Mountain (see the geologic map, Fig. 1), is about 
850 feet above sea-level. Other parts of the chert outcrop as low 
as 650 feet, this being the elevation at the point nearest the hot 
springs. The hot springs break out at elevations between 600 
and 694 feet. Parts of the intake area are thus at the same level 
as the springs and below some of them. Assuming the greatest 
difference 200 feet, it is doubtful if 200 feet of head or 80 pounds of 
pressure per square inch is sufficient to force the water through 
the channel assumed by this hypothesis. Even if this head is 
sufficient, it is remarkable that the water comes across the strike 
under the North Mountain syncline, when it could swing north- 
eastward and around North Mountain into the Hot Springs 
anticline without notable change in level. This path is possible 
because the North Mountain syncline plunges southwestward 
and near the ‘“‘Gorge’’ of West Branch brings the Bigfork chert 
near the surface. From this analysis it appears that the postulated 
structure is a very special hypothesis of dubious validity. 

Similarly the hypothesis calls for an uncooled mass of igneous 
tock below the North Mountain syncline to supply heat, for the 
internal heat of the earth would not raise the temperature the 
required amount unless the water descended to 5,000 feet and then 
came to the surface without loss of temperature. Obviously the 
postulation of such an igneous body is a special hypothesis, par- 
ticularly when apparently the water could easily avoid the plug by 
a change in route as shown before. The occurrence of igneous 
masses in the neighborhood as at Magnet Cove and Potash Sulphur 
Springs adds probability to this hypothesis, but these intrusions are 
thought to be of Cretaceous age, a time so remote that it is stretch- 
ing credulity to believe that rocks of this age are still uncooled at 
moderate depths. 

t A. H. Purdue, op. cit. 


446 KIRK BRYAN 


The emergence of the water through the Polk Creek and 
Missouri Mountain shales into the Hot Springs sandstone requires 
no special hypothesis, if the thrust fault with associated jointing 
and fissuring which seem to be indicated by field relations is 
granted. 

By reference to the table of temperatures, page 430, it will be seen 
that Big Chalybeate Spring has a temperature above normal. In 
chemical composition the water is of the calcium carbonate type, 
and of about the same mineralization as the Hot Springs waters, 
differing mainly in having less silica, (see table, page 427.) Ithasa 
strong and according to local observers a steady flow, which 
measured by H. D. Mitchell’ was found to be 186 gallons per 
minute. This spring lies 53 miles northeast of Hot Springs on the 
northwestern flank of the mountain which is the extension of the 
North Mountain syncline. The spring apparently arises from 
the Polk Creek shale in the flat valley of a tributary of the West 
Branch of Gulpha Creek. If the water is derived from rainfall on 
the Bigfork, which saturating the chert arises through a fracture 
in the shale, it is difficult to account for the abnormal temperature, 
18° above the mean annual air temperature, unless an uncooled 
igneous plug is postulated for this spring also. The contact of 
the shale with the chert on the west is less than one-tenth of a mile 
away and less than 20 feet above the spring. The difference in 
head seems insufficient to force the water to travel to depths and 
return. On the other hand, southwest of the spring three-fourths of 
a mile is another anticlinal area of Bigfork chert which has its 
contact with the shale at elevations between 620 and 820 feet. 
It might be postulated that water from this area would flow north- 
west under the syncline and emerge at Big Chalybeate Spring. 
This hypothesis has the advantage that a depth of 1,000 to 2,000 
feet would be attained, and this depth would doubtless be sufficient 
to account for the temperature of the water. The spring, however, 
has an elevation between 620 and 640 feet. For this postulate, 
also, there appears to be a lack of hydraulic head. The origin of 
Big Chalybeate Spring is then as much an unsolved problem as the 
origin of the Hot Springs, but because the waters are both thermal 

«J. C. Branner, op. cit., p. 20. 


HOT WATER SUPPLY OF THE HOT SPRINGS 447 


and similar in chemical composition, it seems likely that they have 
a common origin. Special hypotheses are invalidated by this 
probability. 

The meteoric hypothesis then suffers from two main defects: 
(1) A possible lack of head to force the water through the postu- 
lated structure; (2) the very special association of uncooled rock 
with the structure. 

The hypothesis of juvenile origin for the waters when examined 
is perhaps more satisfactory, but suffers from conspicuous defects. 

This hypothesis may take two forms: (1) That there is a 
buried mass of uncooled igneous rock which is discharging water 
due to cooling and crystallization; (2) that a fracture or fissure 
extends from the springs into the deep interior of the earth, similar 
in character to the great fault fractures and through this fracture 
deep-seated waters, juvenile or of mixed origin, rise to the surface. 

A mass of uncooled igneous rock discharging juvenile water is 
a hypothesis of the same special character as the uncooled body 
postulated under the meteoric hypothesis. It is no more unreason- 
able to assume that it is still crystallizing and discharging water, 
than that it is not crystallizing, but is still hot enough to heat the © 
water by contact. Moreover, the nearby igneous bodies are of 
Cretaceous age and there is no other evidence of igneous activity 
in the general region. 

That a deep fracture or fissure exists is also a special hypothesis 
but only special in that it provides that the water rising in this 
fissure is warm at the surface. A source of heated water is every- 
where present in the deeper crust and in regions of disturbance 
there is a rise in the geotherms and in some instances at least 
invasion of batholithic bodies. Deep fissures in this general region, 
though their position is now almost wholly concealed by erosion, 
must have occurred as late as the Pleistocene epoch during which 
time the principal uplift is thought to have occurred. Under a 
hypothesis of recent faulting the position of the springs at the 
nose of Hot Springs Mountain anticline is purely accidental 
except that the rising waters have taken advantage of the pre- 
existing fracture by overthrust of the Hot Springs sandstone and 
the two underlying shales. 


448 KIRK BRYAN 


Faulting, except in connection with folding, throughout the 
general region around Hot Springs, is difficult to establish. How 
ever, in the coastal plain region of central and southern Arkansas 
and adjacent states, Late Tertiary and post-Tertiary faulting have 
probably taken place. According to Stephenson,* small faults 
are recognized in connection with the Preston anticline. In the 
Monroe Gas Field? a post-Eocene fault with a total displacement 
of 150 feet has been mapped. ‘The theory of origin of salt domes 
advanced by Harris’ rests on the postulate that faulting on an 
extensive scale has taken place throughout eastern Texas, Louisiana, 
and southern Arkansas. Quarternary faulting with a displace- 
ment of at least 1,000 feet has been shown for the Jennings oil 
field.4 A post-Tertiary fault with a throw of 2 feet to the south 
was noted by Professor H. A. Wheeler’ and Colonel John R. 
Fordyce 3 miles north of Stephens, Arkansas, a town 75 miles 
south of Hot Springs. This fault appears to be very recent. 

The recorded evidence of recent faulting is thus incomplete, 
but the known uplift of Pleistocene time must have offered favor- 
able conditions for faulting however difficult the matter of proof 
may be. While no recent faulting has been discovered at or near 
Hot Springs, the foregoing facts indicate that such faulting is not 
improbable and may yet be found. 

The hypothesis of juvenile origin thus also rests on an insecure 
foundation since it postulates either a special igneous mass or a 
special fault fissure. For neither of these is there other evidence. 


USE AND DEVELOPMENT OF THE HOT SPRINGS WATER 


The quantity of the Hot Springs water used for bathing and drink- 
ing fluctuates from year to year but has gradually increased. The 


«L,. W. Stephenson, ‘“‘Contribution to the Geology of Northeastern Texas and 
Southern Oklahoma,” U.S. Geol. Survey, Contrib. to General Geol., Prof. Paper 120 
(1919), pp. 129 ff. 

2H. W. Bell and R. A. Cattell, Louisiana Dept. Conservation, Bull. No. 7, 1921. 

3G. D. Harris, Bull. Louisiana Geol. Survey No. 7 (1908), pp. 75 ff., also Econ. 
Geology, Vol. IV (1909), pp. 12-34, and U.S. Geol. Survey Bull. 429 (1910), pp. 6-10, 
IAL ar. 

4G. D. Harris, ‘Oil and Gas in Louisiana,” U.S. Geol. Survey Bull. 429 (1910), 
pp. 56-61. 

5 Letter, October 22, 1921. 


HOT WATER SUPPLY OF THE HOT SPRINGS 449 


private investment in bathhouses and hotels is large and the 
government investment in the Free Bathhouse and in the Army 
and Navy Hospital is considerable. The volume of business can 
be measured by the number of baths given which reached a maxi- 
mum of 1,194,872 in 1911. A careful engineering study will doubt- 
less lead to improvements in the present system of distributing 
the water. Even though economies may increase the capacity of the 
resort to handle patients and visitors, large future growth of Hot 
Springs as a health and pleasure resort depends on an increased 
supply of hot water. The practical means by which additional 
hot water may be obtained are not here discussed, but it is obvious 
that development can be attempted intelligently only with accurate 
knowledge of the origin of the water. 

The Fordyce Well, shown on Figures 3 and 4, is 6 inches in 
diameter and 673 feet deep. It penetrates the Stanley shale and 
extends into the Hot Springs sandstone. The well has a flow of 
hot water amounting to 50,000 gallons daily. Since the well 
appears not to have decreased the flow of any existing spring or 
well, it must be supplied with water which had previously reached 
the surface In minor seeps and concealed springs. Other wells in 
the same geologic position near the contact between the shale and 
sandstone will save seepage, but will probably dry up the hillside 
springs. The resulting concentration of flow will be convenient 
for a single unified distributing system and the substitution of 
artificial openings for the natural openings or hot springs will be 
an inconsiderable sentimental loss. 

. Whether water not to be considered as salvage can be developed 
by such shallow wells depends on the ultimate origin and the 
mechanism of flow of the hot water. Each stage in attempted 
development of new water will raise anew these fundamental 
questions. 


A DEVONIAN OUTLIER NEAR THE CREST OF 
THE OZARK UPLIFT 


JOSIAH BRIDGE anv B. E. CHARLES 
Missouri School of Mines, Rolla, Missouri 


INTRODUCTION 


The Paleozoic history of the Ozark Uplift constitutes an 
extremely interesting stratigraphic study, and one upon which 
comparatively little has been written. The earlier stages of this 
history, up to the close of the Canadian epoch, are fairly well 
known, for rocks belonging to this and the preceding epochs are 
well exposed throughout the uplift. Of the later history but little 
is known, for so thoroughly have the subsequent periods of erosion 
performed their work, that only small and widely separated outliers . 
of the younger formations remain on the higher parts of the uplit. 
Around the borders of this dome are belts of younger formations, 
more or less completely encircling it, and it has always been a 
question as to how far these formations once extended toward the 
crest of the structure and which of them, if any, completely covered 
it. Upon this question no two sets of paleogeographic maps agree. 
One of the chief reasons for this disagreement is lack of information 
concerning the region, for, as a whole, there has been little detailed 
stratigraphic work done in the region of the Ozark Uplift. 

In the area around Rolla (110 miles southwest of St. Louis and 
well toward the crest of the uplift) the existence of outliers of 
Pennsylvanian and Mississippian age has long been known. The 
former consist of stratified deposits of sandstone and shale, and 
cover far greater areas than are shown on existing maps.* The 
Mississippian outliers are, for the most part, small areas covered 
with fossiliferous, residual bowlders,? which have apparently weath- 

t The new geological map of Missouri, now in press, shows many of the larger 
Pennsylvanian outliers. 

2 Josiah Bridge, ‘‘A Study of the Faunas of the Residual Mississippian of Phelps 
County (Central Ozark Region), Missouri,” Jour. Geol., Vol. XXV (1917), pp. 558-75: 


450 


A DEVONIAN OUTLIER OF THE OZARK UPLIFT 451 


ered out of the basal conglomerates of the Pennsylvanian. In one 
or two instances, however, there are good reasons for believing 
that there are small areas of Mississippian rocks which are still in 
place, and which rest unconformably upon the Jefferson City dolo- 
mite (Beekmantown), as do also most of the Pennsylvanian outliers. 

A few months ago the junior author discovered a small outlier 
containing a fauna which is characteristically Middle Devonian 
in age. This fauna when identified proved to be of sufficient 
diagnostic value to place the age of the outlier within very narrow 


Fic. t.—Sketch map showing general location of Devonian outlier. The shaded 
_ portion represents the area shown in detail in Figure 2. 


limits. Inasmuch as Devonian has never before been reported 
from this area, the finding of this outlier gives a little more light 
on the history of the uplift, and the purpose of this paper is to 
describe this occurrence and its fauna, and to add its bit of evidence 
to the history of the region. 


LOCATION AND GENERAL DESCRIPTION 


The Devonian outlier lies about 14 miles northwest of Rolla, in 
the NE. + NW. i sec. 3, T. 37 N., R. 8 W. (Fig. 1). Itison the east 
side of the road and lies fifty feet below the crest of the ridge 


452 JOSIAH BRIDGE AND B. E. CHARLES 


(Fig. 2). It occupies a little knoll between two gullies, and when 
first found consisted of a number of knobs of quartzite projecting 
from the hill and having a rough alignment. The base was con- 
cealed, but just above the top of the outlier were a number of 
Mississippian bowlders, and on the north, west, and south at higher 
and lower levels are outcrops of the Jefferson City formation™ 
(Fig. 2). 

The summit of the ridge is capped by a thin stratum of Penn- 
sylvanian sandstone, and bowlders of Pennsylvanian float are found 
in abundance at lower levels. The Pennsylvanian rests uncon- 
formably upon the Jefferson City formation on the east side of the 
hill. On the west side there appears to be a thin stratum of Missis- 
sippian between them, but the exposures are very poor, and the 
exact relationships are difficult to determine. The areal distribu- 
tion of these rocks is shown on the map? (Fig. 2). 

An excavation was made along the side of these quartzite 
masses in an attempt to expose the lower contact. This was not 
entirely successful, but the excavation showed that there was a 
continuous ledge of quartzite at least thirty feet long and from 
three to six feet in thickness, increasing in thickness toward the 
east. Beyond the limits of the excavation there are a few other 
knobs of quartzite, probably continuous with the part just 
described. This makes the total length of the outlier about fifty 
feet, and its greatest width slightly less. It is entirely confined 
to the little nose between the two gullies, and none has been found 
on neighboring hillsides. Bedding is indistinct, but there are 
indications of an eastward dip of about 15°. This is regarded as 


t The Jefferson City beds exposed on this hillside are among the youngest known 
in this area. They consist of yellow earthy dolomites (cotton rock) interbedded with 
chert layers, which are often abundantly fossiliferous, a condition not observed in the 
lower beds of the same formation in the Rolla area. The fossils consist entirely of 
one or two species of Hormotoma, closely allied to H. artemesia. It is quite probable 
that these upper beds belong to the Cotter formation, which overlies the Jefferson 
City in the southern portion of the uplift, but which has not been recognized in this 
area. This cannot be definitely stated, however, until the Cotter and its contained 
fauna are more completely described. 

2 The writers are greatly indebted to Major C. E. Cooke, professor of topographical 
engineering in the department of vocational education, and his students, Messrs. 
Kimball and Hazlewood for making the topographic base for this map. 


A DEVONIAN OUTLIER OF THE OZARK UPLIFT 453 


/ 


4 


ee ies 
Seale GOOo | 
Contour Interval S 


GB Fe nnsylra nlan Devonian 
BS Mesosippien |_| Hiern Cit 


X fossil Locality. 


Fic. 2.—Geologic map showing the relationship of the Devonian to the older 
and younger rocks. 


Ase JOSIAH BRIDGE AND B. E. CHARLES 


initial dip, and the entire mass appears to rest against an old east- 
ward facing slope developed on the Jefferson City dolomite in a 
post-Beekmantown pre-Onondaga erosion interval (Fig. 3). 


Fic. 3.—Structure section on the line A—4 of Figure 2 


The Jefferson City formation has been 
found outcropping to within ten feet of the 
summit of the hill on the east side and to 
within fifteen or twenty feet of the summit 
on the west side. Overlying the Jefferson 
City beds on the west side is a layer of 
large bowlders containing a typical Burling- 
ton fauna. This layer is continuous for 
several hundred feet, and the bowlders are 
quite large, and it seems altogether prob- 
able that they are in place, or that they 
have not been moved far from their original 
location. On the east side of the hill this 
layer is not as prominent, but bowlders of 
Mississippian and Pennsylvanian float are 
abundant all over the hillside, and some 
of them rest directly upon the Devonian. 


LITHOLOGIC CHARACTERISTICS 


Lithologically the Devonian rock is a 
hard, dense quartzite breaking with a 
splintery fracture. In color it ranges from 
white through gray to bluish and almost 
black. The lighter shades predominate. 
Thin sections show great numbers of well- 
rounded quartz nuclei with strong evi- 
dences of secondary growth. In -some 
sections the grains show as angular inter- 
locking crystals because of this secondary 
growth; in other sections, large second- 


arily enlarged grains are separated from each other by a finely 
crystalline ground mass of the same material. The rock contains 
numerous cavities, most of which have been formed by the leaching 
out of large fossils, and on the surfaces of these cavities are to be 
found many small but perfect crystals of quartz and of limonite 


A DEVONIAN OUTLIER OF THE OZARK UPLIFT 455 


pseudomorphous after pyrite. The quartzite is somewhat frac- 
tured, and these fractures are filled with a soft, yellow, non- 
fossiliferous, somewhat conglomeratic sandstone. 

From the structure and secondary growth it seems evident that 
the rock was originally a calcareous sandstone, laid down by an 
advancing sea against an old land mass. The calcareous matter 
has been completely leached out, and much of it replaced by silica. 
The soft, yellow sandstone is of later age, either basal Mississippian 
or Pennsylvanian. 

PALEONTOLOGY 

Fossils are abundant in the quartzite. They are not evenly 
distributed, but are most abundant at the base. At first glance, 
parts of the stratum appear to be barren, but careful search of 
almost any fragment will reveal fossils. The fossils all occur as 
external and internal molds, and in most cases the preservation 
is excellent. Corals and Mollusca dominate the fauna. Other 
forms are not common, though individuals of a given species may 
be very abundant. 

The following table gives a list of the species which have been 
obained from this quartzite, and also their occurrence at other 
localities. Column 1 shows species occurring in the Grand Tower 
formation in southern Illinois; column 2, species occurring in the 
Jeffersonville beds at Louisville, Kentucky; column 3, species 
occurring in Michigan; column 4, species occurring in Ohio; 
and column 5, species occurring in New York.* 


t Faunal lists: 


1. Illinois: S. Weller, ‘“‘Correlation of the Devonian Faunas in Southern Illinois,” 
Jour. Geol., Vol. V (1897), pp. 625-35. 

T. E. Savage, ‘‘The Grand Tower (Onondaga) Formation of Illinois and Its 
Relation to the Jeffersonville Beds of Indiana,” Trans. Ill. Acad. Sci., Vol. III (1910). 

2. Indiana, Kentucky: E. M. Kindle, ““The Devonian Fossils and Stratigraphy 
of Indiana,” Ind. Depi. of Geol. and Nat. Res., Twenty-fifth Ann. Rept. (1900), pp. 529- 

8. 

ay H. Nettleroth, Kentucky Fossil Shells. Monograph, Kentucky Geological 
Survey, 1889. 

W. J. Davis, Kentucky Fossil Corals. Monograph, Kentucky Geological Survey, 
1885. 

3. Michigan: C. Rominger, Geol. Surv. Mich., Vol. III, 1876. 

4. Ohio: F. B. Meek, Pal. Ohio, Vol. I, 1873. 

H. A. Nicholson, zbid., Vol. II, 1875. 

5. New York: James Hall, Paleontology of New York, Vol. IV, 1867, and Vol. V, 
Part II, 1879. 

General: Grabau and Shimer, North American Index Fossils, 1910. 


456 JOSIAH BRIDGE AND B. E. CHARLES 


TABLE I 


List OF SPECIES IDENTIFIED FROM THE DEVONIAN QUARTZITE AT ROLLA, Missouri 


I z 3 s 
qilinoss | ere eee ear Ohio | New 
Coelenterata: 
Zaphreniis eicantea Wesueutsnen pee ee seen x K x 28 
LOT ARAS POON GG, WWE. ooo nooo osoollone nor Ps x fi | Aa ee 
LG PUT ENLUS SP 3. s0 oo ene ye ooo an be os ores Oh ge Elias eS | ee ose 
Acerumlerza rugosa (i: and El.)sn sees eee x XK Xo 
Amplexus yandellt (E. and H.).............- x x x eee 
Favosites winchella, Rominger...............|....-. x Gee x 
Favosites emmonsi, Rominger............... x X Da EE eralcida ot 0 
Havosttes basaliacus | Goldtuss;.45 44a a eee ee os ene ee eee x 5s 
Favosites turbinatus (2) Billimgs.............|...... x aK x 5 
Havosiues lamitaris, Romingers. 2224-452 ssee\eos os: x x) |_| oe x 
AUOSUES CLAUSHS AROMIN Sela. cee sere ere | tee x 9 || ane x 
Cladoporavabrosaa(Bullings) sane ee | eee XK x, |e x 
Molluscoidea: 
Cystodictya gilberti (?) Meek............... x EL 9 ees <1 Loe 
Stropheodonta demissa (Conrad)............ x x Pali tae 5 c's x 
Rhipidomella vanuxemt, Hall................ XK x K x x 
Cenironella glansfagea, Hall....2............ K xg K i x 
Hunella (2) SB. cei oc ec esate a losin o| ee ee ee 
Syeeruiar HalariCoNS, IGE. ocanccccnoncodocoslsaccce x) | oer x XE 
SHOP (Graze (2))5 abl, oe accccacvescanse x x) a eeeee x x 
Marino Gaeee (2) (BHUIEE))s60c050cenecanasloseaccloevascllooocon See 
INAECLeOS Pin GnGONci as Elalle ae eee XE cil Beoeeaye 3 x 
Mollusca: 
Actinodesma occidentale (?), Hall............ x KX |scanscc's all Oe | eee 
Conocardium cuneus (Conrad).............. x VAT i \seneeeene x x 
Conocardium ohioense, Meek...............|...... DI Tee be 6 x. No see 
Bellenopion pel ops sia! lass serene K ee x x 
Bellerophon mewbernyi.Nieekr sass. yyeeieeee| peel eee | eee x ? 
Hormotoma: maia, Hall... 2.6 i cccdies ecco wien 05 |'c ences she ols os en ee 
Wigocenasiconici 7 (lal) ae ee xe x eae x x 
Toxonemarouustum: (7). lalla ane ceere een tenes tee rien eee x) o\oceaee 
Cyclonema crenulata, Meek.................\.....- Deed Ieee c | cee 
Cillonemanvellatn lan all eee tee |e ene XE x Xivileecereeae 
Callonema humile, Meek.) 0) 55.225 sees eel onan: Xe bg act 
Callonemaxconuskandlenn ser eer ee eee PRIA G| |S. cdes oic|lo us 20 0 
Tentaculites scalariformis, Hall............. x Ky ce eee x x 
Coleolus tenuicinctum, Hall.................. Ne ae all eee H H 
Orihocer ds Spin Obie ho ed sva wo ties aed eas + WAR ee OO S| ee | ee 
Gomphoceras Sp (2) isi jek oa 8h bok es iced s Ae ale ee ee ee 3 
Lultleoceras: nereus\ Wallis, ic. ames <1 sealed redio| aan oe |e |e Xe 
Arthropoda: 
IERUCOPS CHSC, BEM. bscoddoonnboodavanne x X | Wicca | eee b¢ 
IEVOGOS QUaHB, WEEMWos. ocohaoonsenossoouenne x Galland omalloce-s => 
Dalmanites calypso, Wall’ .................. x Ke enna aes x x 


*H indicates that species occurs in the Hamilton at that particular locality. 


A DEVONIAN OUTLIER OF THE OZARK UPLIFT 457 


CORRELATION 


From the foregoing table it is evident that this fauna is of 
Onondaga age, and that it is the partial equivalent of the Grand 
Tower formation of southern Illinois and southeastern Missouri. 
Its affinities are entirely with the eastern Devonian and have 
no resemblance to the later Devonian faunas of Iowa and north 
central Missouri. Of the thirty-seven identified forms, seventeen 
occur in the Grand Tower formation of southern Illinois, thirty 
in the Jeffersonville beds of southern Indiana, thirteen in the Onon- 
daga group of Michigan, twenty-two in the Onondaga of Ohio and 
eighteen in the Onondaga of New York. These figures are not 
exact and are probably too low, for the faunal lists from the various 
regions are incomplete and represent compilations in most cases. 
The most complete lists are those from the Grand Tower formation 
and from the Jeffersonville beds. Weller’ and Savage? have shown 
the relationship of the Grand Tower fauna to the Onondaga of the 
eastern United States. 

Savage? has also shown that the Jeffersonville beds are the 
equivalent of the upper portion of the Grand Tower of Illinois. 
Since more than 80 per cent of the forms occurring at Rolla are 
also found in the Jeffersonville beds, it seems certain that this 
outlier belongs to the upper portion of the Grand Tower formation. 
In Ste. Genevieve County, Missouri, Weller* assigns over two 
hundred feet of strata to the Grand Tower formation. Certain 
horizons in this formation are reported to be full of corals, but 
until the faunal lists for this formation are completed, a closer 
correlation cannot be made. 


CONCLUSIONS 


The presence of a Grand Tower outlier at Rolla indicates a 
much greater submergence of the Ozark uplift during Onondaga 
time than has commonly been supposed. The nearest outcrops of 
the Grand Tower formation are at least 100 miles to the east. The 

tS. Weller, loc. cit. 

2T. E. Savage, Joc. cit. 3 Ibid. 


4Stuart Weller, unpublished manuscript on Ste. Genevieve County, Missouri, 
Missouri Bureau of Geology and Mines. 


458 JOSIAH BRIDGE AND B. E. CHARLES 


St. Francois Mountains, the structural center of the uplift, are 
directly between the two exposures, and it is not believed that 
these were covered in Devonian time. In view of the thinning 
out of the Devonian to the north, it may be assumed that the Onon- 
daga sea extended westward along the southern border of the 
St. Francois Mountains, and that it may have covered much of the 
southern portion of the uplift. Further field work may reveal other 
outliers, which will enable the boundaries of this sea to be traced 
more definitely. 


THE EARLY PRE-CAMBRIAN FORMATIONS OF 
NORTHERN ONTARIO AND NORTHERN 
MANITOBA 


BL. BRUCE 
Queen’s University, Kingston, Ontario, Canada 


The classification of rocks earlier than the Cambrian is one of 
the difficult problems of geology. Correlation from point to point 
rests largely upon lithology and upon the succession of lithological 
units, and difficulties arise both from original similarity of forma- 
tions of quite different age, and from similarity through the develop- - 
ments of the same metamorphic minerals in rocks originally of 
quite different character. Correlation is especially difficult in 
formations of early pre-Cambrian time, since these have undergone 
much longer periods of deformation than have the later ones. 
Varying successions of these early rocks have been worked out in 
some detail in different parts of the Canadian shield, and many 
attempts have been made to formulate a generalized succession 
that will fit all areas. None of these attempts has been successful, 
since the determination of the age of a formation by its likeness 
to a certain formation as described in an accepted classification, 
is liable to lead to quite erroneous conclusions. A review of some 
areas in western Ontario and northern Manitoba will show the 
diversity in various sections and will, it is believed, make possible 
certain generalizations. 


RAINY LAKE DISTRICT 


The first attempt to subdivide the early complex was made 
by Lawson! in the Rainy River district. He recognized two 
formations earlier than the first granite intrusion. The lower of 
these is his Couichiching series, which he believes to be the oldest 
formation in the area. The Coutchiching rocks consist of mica 


1 Geol. Survey of Canada, Ann. Rept., New Series, Vol. IiI, Part I (1887-88). 
459 


Ab BH IREE 

cy E SO" [Pom 921) 
Sal yee Nal 

Oe eg ee RT 


rans i 
Se 


pee 
bee 


ybeop es y 


a 2 ety F, 
J 4S if 


BE BROCE, 


8 oye] vuy 

sino Fenlay Sh L aye] ssolg 

oun wa 9 “ry OYSNAE NA 

ee s eqoyuey, “MN 

5‘ v uOosICIN dxe'] 

g Tqniqy 

r z aurdnsi0g 

/ I ayey Aurey 

SLOTALSIG 
Ss i 
+ nh 
aa 


PRE-CAMBRIAN FORMATIONS—ONTARIO AND MANITOBA 461 


schists, garnetiferous mica schists, hornblende schists, and phyllite. 
In some of the mica schist, quartz makes up three-fifths of the 
rock mass, biotite one-fifth, with zoisite the other important 
mineral present. It seems evident from Lawson’s descriptions of 
the field occurrences and from the results of petrographical examina- 
tions and chemical analyses as given in his report, that the rocks 
are sediments, possibly of somewhat abnormal types. They are 
entirely clastic; no limestones have been found in association with 
them. A striking and peculiar circumstance is the lack of coarse 
sediments or conglomeratic beds in this supposedly thick series. It 
seems possible that these rocks were formed along the seaward margin 
of a delta which supplied large quantities of fairly fine débris, 
the site of deposition for the Coutchiching being so far from shore 
that no gravels were supplied, but not far enough out to allow 
the formation of limestones, assuming that conditions were suitable 
for the deposition of lime rocks at that early period. 

The Keewatin series consists of: ‘‘(1) fine-grained greenstones 
showing frequently ellipsoidal or amygdaloidal structures or both; 
(2) coarser-textured greenstones showing neither ellipsoidal nor 
amygdaloidal structures; (3) greenstone schists of varying degrees 
of schistosity; (4) rather massive chlorite schists; (5) evenly 
fissile chlorite schists; (6) irregularly cleaved chlorite schists; 
(7) black glistening hornblende schists usually on the periphery 
of the Keewatin belts where they come in contact with granitic 
intrusions; (8) gray felsite sometimes amygdaloidal; (9) sericitic 
schists; (10) various stratified grayish-green schists, probably ash 
beds; (11) agglomerates; (12) gray siliceous slates and schist; 
(73) banded cherts; (14) mica schists; (15) limestone.” 

This group is clearly made up for the most part of lava flows 
and their derivatives with minor amounts of sediments inter- 
banded with the igneous rocks. In summary, then, it may be 
said that the Coutchiching consists chiefly of sediments, with 
possibly some beds, such as the hornblende schists, of igneous 
origin; the Keewatin series is chiefly igneous, with minor sedi- 
mentary beds intercalated with the lava flows. There seems 

t Lawson, Memoir go, Geol. Survey of Canada, p. 28. 

2 [bid., Pp. 35- 


462 E. L. BRUCE 


to have been no erosion period between the formation of the two 
series. 

Lawson considered the Coutchiching to be a distinct series 
underneath the Keewatin, but the international committee to 
revise the classification of pre-Cambrian formations did not agree. 
In some of the areas examined rocks mapped as Coutchiching in 
the original work were found to belong to the Seine series, which 
lies with a great unconformity above the Keewatin. Mistakes of 
this kind are quite to be expected in determining somewhat similar 
series under the difficult conditions of the original mapping. Law- 
son, in his later publication, Memoir 40, Geological Survey of 
Canada, admits these mistakes, but still maintains that there is a 
great sedimentary series below the Keewatin. Some competent 
observers who have visited the area agree with him." The point 
has also been raised that the Coutchiching may not be the oldest 
formation, but may be similar to the interbedded sediments—(z2) in 
the Keewatin—and that beneath the Coutchiching again there 
may be still older lava flows. If so, it is argued, the Coutchiching 
series may quite logically be included as part of the Keewatin. 


PORCUPINE DISTRICT 


The earliest rocks in the Porcupine district of northern Ontario 
consist of pillow lavas with schists derived from them, “carbonate” 
rocks of doubtful origin, iron formation, and some fragmental 
rocks of doubtful character.? A group of undoubtedly sedimentary 
rocks is correlated by Burrows with the Temiskaming series. 
They consist of ‘‘conglomerate, interbanded slate and greywacké, 
and quartzite.” The relations of these sediments to the Keewatin 
group are shown by the following quotations: 


A contact of the sedimentary rocks with the volcanic rocks can be seen 
immediately south of the open pit at the Dome mine. Fragments of the 
volcanic series are abundant in the sedimentary series, and it is likely that 
the conglomerate has been deposited on the surface of the volcanic series of the 
Keewatin. .... However, one half a mile north west of the north end of 


« Private Communications, F. J. Alcock and T. L. Tanton. 


2A. G. Burrows, ‘“‘The Porcupine Gold Area,” Third Ann. Rept. Ont. Bureau of 
Mines, Vol. XXIV, Part III (1915). 


ine ing emusic 


PRE-CAMBRIAN FORMATIONS—ONTARIO AND MANITOBA 463 


Porcupine lake in lot 11 in the fourth concession of Whitney there is a contact 
of the sedimentary series with pillow lava in which the relationship suggests 
an igneous contact, that is that the pillow lava is later than the sedimentary 
rock. It is therefore probable that some of the pillow lavas mapped with the 
Keewatin are later in age than the Temiskaming series. 


While there is much evidence pointing to a separate sedimentary series 
of rocks the possibility of some of what has been called Keewatin being 
contemporaneous with the Temiskaming or of some of the sediments being 
of Grenville age must be considered. Toward the southwest from the open 
pit at the Dome mine there is a narrow band of conglomerate which has been 
mapped at Temiskaming. Much of. the material in this band immediately 
north of the readily recognized pillow lava and amygdaloidal rock resembles 
volcanic fragmental or agglomerate. There is no break, however, between 
the apparent volcanic fragmental and the interbedded slate and greywacke 
which occur along the south margin of the open pit and which can be followed 
northward for a mile. If the rock above mentioned is a volcanic fragmental, 
and not a true conglomerate deposited on an eroded surface, then there is 
reason for considering the pillow lavas, fragmental rocks, slates, greywacké 
and conglomerate as belonging to one series. For lithological reasons it seems 
preferable to consider the large area of sediments as a separate series. 


It is apparent that Burrows recognized the possibility of an 
interbedded series of sediments and volcanics, but that he decided 
to correlate the sedimentary part of the series with the lithologically 
similar Temiskaming rather than accept an alternative hypothesis 
of a great continuous series made up of lava flows, volcanic frag- 
mentals, and true sediments. The evidence for an erosional 
unconformity between the volcanics and sediments does not 
seem to be conclusive. 


ABITIBI DISTRICT 


The district south of Lake Abitibi has recently received con- 
siderable attention, and an interesting series of rocks has been 
found.t. Lava flows which are classed as Keewatin are exceptionally 
well developed, and the scoriaceous surfaces of successive flows 
make it possible to determine the top and bottom of the forma- 
tions. Sediments are associated with these lavas. The following 
is a quotation from the report mentioned: 


tC. W. Knight, A. G. Burrows, P. E. Hopkins, and A. L. Parsons, “Abitibi Night 
Hawk Gold Area,” Eighteenth Ann. Rept. Ont. Bureau of Mines, Part II (1919). 


4604 . E. L. BRUCE 


In our map sheet there is an interesting series comprised of highly altered 
sediments which are closely associated with the Keewatin. ... . One belt has 
an apparent thickness of one and one half miles and a length of rr miles. . . . . 

The rocks in the three large areas mentioned in the preceding paragraph 
consist of slate, greywacké, quartzite, and a little conglomerate, all of which 
have been altered to schists. Both the cleavage and bedding of the sedi- 
ments have nearly vertical dips, but there are usually small angles between 
their strikes. A little chert is also present. Conglomerate schist was seen 
in four localities and, in each case, near the outer edge of the sediments. .. . . 
The pebbles which are somewhat flattened, consist of quartz porphyry and 
greenstones, suggesting an unconformity between the sediments, on the one 
hand, and the greenstones and quartz porphyry on the other. However, the 
only good contacts which were seen between the sediments and the greenstones 
were on lot 7 in the second concession of Coulson township and these might 
suggest that the sediments were interbedded with the pillow lavas of the 
Keewatin. It may be added that in this locality the banded cherts which 
appear to be a part of the main group of sediments are older than the pillow 
lavas. In view of these apparently conflicting observations it is seen that 
the relationship between the Keewatin lavas and these old sedimentary rocks 
has not been definitely worked out. Possibly the conglomerates may be of 
interformational origin or may belong to the Temiskaming series. 


It seems clear that there are two possibilities. Either the 
lava flows here called Keewatin are in reality much younger than 
those usually classed as Keewatin, younger in fact than the Temis- 
kaming series, or there is a great interbanded series of lava flows, 
tuffs, and sediments. The presence of conglomerate in such a 
series is quite to be expected. 


LAKE NIPIGON DISTRICT 


In the Kowkash area east of Lake Nipigon, Hopkins™ has 
found a series of rocks which he calls the Marshall Lake series. 
This group consists of quartz-mica schists, garnet, and staurolite 
schists. Hopkins says: 


The chemical composition, microscopic evidence, and frequent occurrence 
of alternating coarse and fine bands in these quartzose rocks suggest that they 
are clastics or volcanic fragmental rocks deposited in water. Since they are 
interbanded with ellipsoidal lavas on Cross lake and contain some iron forma- 
tion they are apparently closely associated with the Keewatin. 


t Ann. Rept. Ont. Bureau of Mines, Vol. XXVI (1917), p. 206. 


PRE-CAMBRIAN FORMATIONS—ONTARIO AND MANITOBA 465 


No evidence is given for considering these rocks as volcanic 
fragmental types, and judging from the description they seem to 
be normal fine-grained clastic sediments. 

Along the Canadian Northern Railway east of Lake Nipigon,* 
Burrows found a complex of igneous and clastic rocks, all of which 
he grouped tentatively as Keewatin. Concerning them he says: 


The age relationship between the mica and quartzose schists of sedimentary 
origin and the pillow lavas and other igneous rocks is not known. For the 
most part the sedimentary rocks stand so nearly in a vertical attitude that 
their relationship cannot be determined. It seems advisable to group all 
these rocks with the Keewatin until information is available to show that the 
sedimentary rocks may possibly be older than the lavas.? 


Both Burrows and Hopkins classify certain other conglomerate 
rocks as Temiskamian, but in neither area are the conglomerates 
found definitely unconformable with the lavas. The correlation 
is a lithological one in both cases, the later age being assumed from 
the presence of pebbles of jasper, greenstone, and granite in the 
conglomerate beds. 


PRE-CAMBRIAN SECTIONS IN MANITOBA 


Various sections in northern Manitoba have been examined in 
some detail, and a strip of country extending almost across the 
province has been mapped. Beginning at the Saskatchewan 
boundary, where the pre-Cambrian basement emerges from beneath 
the Paleozoics, a series of three map sheets extends eastward to 
the Hudson’s Bay Railway. Northeast of Lake Winnipeg two 
areas—the Cross Lake district and the Knee-Oxford Lake district— 
have been studied. 

In the most westerly section the oldest rocks are ellipsoidal 
greenstone and derived schists.3 Supposedly later than these is a 
thick series known as the Kisseynew gneiss, a garnetiferous, quartz- 
biotite gneiss apparently sedimentary in origin. There is also a 
group of slates, quartzites, and conglomerates. The latter are 
quite evidently the result of torrential deposition probably in 


t Ann. Rept. Ont. Bureau of Mines, Vol. XXVI (1017), p. 232. 
2 Tbid., p. 230. 
3 Mem. 105, Geol. Survey of Canada. 


466 E. L. BRUCE 


river flood plains. This group, the lower and upper Missi forma- 
tions, are, however, on structural evidence, thought to be sepa- 
rated by a mountain-making and erosional interval from the 
volcanic rocks and are therefore not considered in this discussion 
of the early pre-Cambrian formations. 

In the Wekusko (Herb) Lake district, approximately seventy- 
five miles east of the Saskatchewan-Manitoba boundary, ellip- 
soidal lavas occur which are apparently the continuation, so 
far as lavas can be continuous, of the area just described. They 
are lithologically similar and outcrop practically continuously 
across the interval. Quartz-biotite gneisses lithologically similar 
to the Kisseynew gneiss are associated with them. Staurolite 
schist, conglomerate, slate, and acid volcanic flows also occur. 
All these are interbanded with the basic ellipsoidal flows. 

Alcock’s summary’ is very definite with regard to the relations 
and character of these early rocks: 

The pre-granite complex is interpreted, therefore, as representing a 
series of interbanded sediments and volcanic rocks of varying composition. 
Though the sedimentary division contains members which have pebbles of 
granite, quartz, and volcanic rocks, no evidence was found that these pebbles 
were derived from any rocks now exposed in the area, nor was any evidence 
found, aside from the presence of these bowlders and pebbles, which would 
suggest that the members containing these fragments represent a younger 
series infolded with the complex and separated from it by an erosional uncon- 
formity. The whole group is regarded as a series of flows and contemporaneous 
sediments. The absence of limestone, the dominance of clastic sediments, 
the irregularity of the beds, the great thicknesses locally, the recurrence of 


conglomeratic horizons, point to a continental rather than to a marine origin 
for the series. 


At Cross Lake, an expansion of the Nelson River below Lake 
Winnipeg, a series of sedimentary rocks consists of para-gneiss, 
arkose, and conglomerate; some of the gneisses are garnetiferous. 
Greenstone, which in places is ellipsoidal, occurs in the area. The 
sediments are interbanded with the lavas. They are interpreted 
as continental, probably fluviatile deposits.s 


tF. J. Alcock, Memoir 119, Geol. Survey of Canada. 
2 Tbid., p. 24. 
3F. J. Alcock, Summary Report, Geol. Survey of Canada, Part D (1919). 


PRE-CAMBRIAN FORMATIONS—ONTARIO AND MANITOBA 467 


Eastward across the divide on the headwaters of the Hayes 
River, early pre-Cambrian rocks are exposed at Knee Lake and 
Oxford Lake.t A lower, dominantly sedimentary part, consists of 
rusty weathering garnetiferous biotite gneiss, impure quartzite, 
slate, conglomerate, tuffaceous rocks, and some interbedded flows. 
The thickness is probably several thousand feet. Above the domi- 
nantly sedimentary group are flows of ellipsoidal weathering lavas 
together with a few bands of iron formation. These groups are 
apparently merely parts of a great continuous series. The sedi- 
ments are in great part typical continental deposits. 


COMPARISON OF THE FORMATIONS 


From the descriptions of the various areas quoted it is clear 
that there are two distinct types of rocks in the early formations: 
(z) volcanic flows now altered to greenstone and chlorite schist, 
and (2) sedimentary rocks consisting largely of gneiss but also in 
places including slate, quartzite, and minor amounts of conglomer- 
ate. ‘The gneisses retain evidences of bedding although in many 
occurrences metamorphism has destroyed some of the original 
texture. Analyses of specimens of these old gneisses are comparable 
to analyses of typical sediments. The slates commonly show the 
original bedding as color variations at slight angles to the fissility. 
Some arkosic rocks still retain the cross bedding and ripple- 
markings of the original sands and conglomerate, even though 
the matrix may be thoroughly schistose, with complete recrystalli- 
zation of the constituent minerals, and are still recognizable as 
water-laid clastic rocks. 

The peculiarities of all the sedimentary formations of this 
eatly period are the comparatively small amount of conglomerate 
and the complete lack of limestone. The sediments found in 
Manitoba are continental deposits probably formed under deltaic 
or piedmont conditions. The lack of any large amount of coarse 
material is evidence that no high land masses existed near the 
site of deposition, but the ripple-marking and cross-bedding of 
some of the rocks indicate shallow-water conditions during the 
formation of some of the beds. The early sediments in other areas 


‘HE. L. Bruce, Summary Report, Geol. Survey of Canada, Part D (1910). 


468 E. L. BRUCE 


are similar to those in Manitoba, and it may be assumed that 
they were formed in much the same way. 

The volcanic rocks are lithologically similar throughout the 
whole region discussed, and this striking similarity quite naturally 
has led to the correlation of these rocks wherever they occur. 
The flows are ellipsoidal or massive greenstone of medium basicity. 
Many of them are now altered to schists. Along with these, 
minor thicknesses of acidic flows and tuffaceous beds occur. In 
many districts thin sedimentary beds are found with the igneous 
rocks. Banded iron formation is very commonly associated with 
the flows of basic composition. 

Although the rock types are comparable, the age relations are 
variable. In some districts the sedimentary rocks lie above the 
volcanics. In others the two are interbanded, and in others the 
greater thickness lies below the igneous rocks. No erosion interval 
has been recognized; the separation into igneous and sedimentary 
divisions made in some localities, is purely arbitrary, and implies 
simply that the divisions are dominantly clastic or dominantly 
igneous. For in most occurrences there are sedimentary beds 
among the igneous rocks and flows intercalated with the sediments. 


CLASSIFICATION 


None of the general classifications of pre-Cambrian formations 
is applicable to this early complex. The classification™ accepted by 
the International Committee places the Keewatin as the lowest 
formation and does not recognize the presence of great thicknesses 
of clastic sediments below the great unconformity at the base of the 
Huronian, nor Lawson’s Coutchiching as a great series below the 
igneous flows. There are, however, not only in the Rainy River 
district, but in other districts, thick sedimentary formations 
below the oldest lava flows recognized in those districts. It is 
possible, as suggested, that other older lava flows exist beneath 
the sediments, but if so, the application of the term Keewatin, if it 
is to be retained, must be extended to include a large amount of 
sedimentary rocks. 


t Jour. of Geol., Vol. XIII (1905), pp. 89-104- 


PRE-CAMBRIAN FORMATIONS—ONTARIO AND MANITOBA 469, 


On the other hand, the original classification suggested by 
Lawson, in which the sedimentary Coutchiching is the oldest 
formation, cannot be applied to those successions in which the 
sediments are interbanded with lava flows or even lie above rocks 
which are lithologically similar to the Keewatin. 

From theoretical considerations it seems unlikely that any of 
these formations can be used to correlate successions in different 
districts. Commonly the basic flows have been used in this way. 
Since they are lavas, it is impossible that any one eruption could 
have extended to any great distance, and hence correlation on the 
basis of lithology of separate flows must be most uncertain. Nor 
is this affected by the possibility that many of the flows are sub- 
aqueous, as there is no evidence that the bodies of water beneath 
which the flows may have been extruded were large or continuous. 
In fact, in some instances the interbedding of ellipsoidal flows and 
shallow water or terrestrial sediments is evidence that the bodies 
of water were limited in area and of brief duration. Correlation 
by means of the sedimentary beds is even less reliable. The con- 
glomerate, slate, and gneiss of this early period are believed to be 
almost entirely of terrestrial or shallow-water origin. No bed 
formed in this way could be expected to have great lateral extent, 
and no determination of age can be made on the-ground of its 
similarity to rocks in other districts. 

Since no erosion break has been recognized in any of the suc- 
cessions yet worked out, and since there is this very marked differ- 
ence in the relations of sedimentary and igneous rocks in various 
areas, it is plain that no course is possible, at present, except the 
interpretation of the early part of the pre-Cambrian as a period 
of volcanic activity in which eruptions of lava alternated with 
deposition of ordinary clastic sediments. These periods of erup- 
tion were recurrent, but not necessarily contemporaneous even in 
neighboring districts. Hence the succession of volcanic and 
sedimentary rocks is naturally not the same in any two districts. 
The result is a great thickness of lavas, tuffs, and sediments, all of 
which belong to one great period in the earth’s history. It is 
manifestly impossible to apply to rocks of such origin either of the 
terms Keewatin or Coutchiching and, if the view set forth here be 


470 E. L. BRUCE 


accepted, it seems necessary to restrict those terms to the original 
area in which they were applied. If it seems convenient in any 
other area to divide the rocks of this early period, local names 
should be applied to the divisions without implying any wide 
regional correlation. Detailed examination may later make clear 
the time relations of events; and if so, correlations can then be 
made. If that ever becomes possible, it seems more than likely 
that instead of rocks of similar lithology being found to belong 
to the same period, it will be found that flows in some districts are 
contemporaneous with sediments in others, or that a period of 
deposition of sediments in one section corresponds to a period of 
local erosion in a neighboring section. 

Although at present sufficient work has not been done to make 
correlation possible, it is interesting to note the distribution of 
the sediments in relation to the igneous rocks. In western Mani- 
toba the sediments lie above the flows. In north central Manitoba 
flows and sediments are interbedded. In eastern Manitoba and 
western Ontario the great mass of sediments lies beneath the 
volcanics. In eastern Ontario the two are again interbedded. 
These relations can be explained by the presence of a great area 
of continental deposition extending southward from an old land 
mass in central Canada in the very earliest times. Over this area, 
terrestrial and shallow-water deposits were laid down on river 
plains, piedmont fans, or deltas along whose margin sediments 
were interbedded with subaqueous lava flows. Still farther out 
no sediments at all were deposited until a later readjustment of 
land and water shifted the zone of sedimentation to areas where, 
previously, only igneous rocks were forming. At the same time 
the central area became the site of igneous activity and the extrusion 
of lavas over the clastic sediments already laid down. At present 
this can be considered only a suggestion resting upon slight field 
evidence. 


SUMMARY 


The points raised in the preceding discussion can be settled 
only by more detailed work, but the following conclusions seem to 
be warranted from present knowledge: 


PRE-CAMBRIAN FORMATIONS—ONTARIO AND MANITOBA 471 


1. In the early part of the pre-Cambrian, periods of volcanic 
activity alternated with periods of normal sedimentation. The 
resulting rocks form one great series. 

2. At present local terms only should be used in subdividing - 
this complex. The terms now in use in the generalized classifica- 
tions are inapplicable, and should be restricted to the areas in 
which they were used originally. 

3. The sedimentary record of the early pre-Cambrian seems 
to be largely one of continental rather than marine conditions. 
Some of the deposits were undoubtedly deltaic, others were likely 
piedmont, lacustrine, fluviatile, and even basin deposits. It 
points to the conclusion that even from the very earliest pre- 
Cambrian, the Canadian shield was a positive element. 


THE TIME OF GLACIAL LOESS ACCUMULATION IN ITS 
RELATION TO THE CLIMATIC IMPLICATIONS OF 
THE GREAT LOESS DEPOSITS: DID THEY CHIEFLY 
ACCUMULATE DURING GLACIAL RETREAT? 


STEPHEN SARGENT VISHER 
Indiana University, Bloomington, Indiana 


Although deposits similar in several respects to glacial loess 
are forming today near the borders of certain deserts and along 
the bluffs of some great rivers, the widespread, thick loess deposits 
which are associated with some drift sheets imply peculiar cli- 
matic conditions, for no deserts are now close to these ancient 
deposits, and parts of them are far from great rivers. There have 
been many discussions of the probable origin of loess, and thus, 
indirectly, of its climatic implications. Much has been learned, 
among other things that different deposits accumulated under differ- 
ent conditions. But one question appears not to have been satis- 
factorily settled, that is, At what time, in respect to glaciation, did 
the greater part of the accumulation take place? Several American 
and European students have thought that the great loess deposits 
date from interglacial times. On the other hand, Penck has 
concluded that the loess was formed shortly before the commence- 
ment of the glacial epochs; while many American geologists have 
held that most of the loess accumulated while the ice sheets were 
at approximately their maximum size. Chamberlin and Salisbury,’ 
McGee, and others lean toward this view. 

There is evidence in support of each of these hypotheses, but it 
seems well to reconsider the possibilities that a large share of the 
great deposits associated with glaciation were formed at the one other 
possible glacial time, namely immediately following the retreat of 
the ice. Recent evidence affords light not available to the workers 

tChamberlin and Salisbury, Geology, Vol. III (1906), pp. 405-12, a comprehensive 
discussion of the characteristics and distribution of the American loess, with references 
to McGee, Shimek, and others. 

472 


THE TIME OF GLACIAL LOESS ACCUMULATION 473 


years ago when the origin of the loess was under heated discussion. 
Furthermore, the emphasis in the past has been on the agencies 
of deposition rather than on the time of deposition. Indeed the 
latter question does not seem to have received much consideration, 
in spite of its importance in the interpreting of the climatic condi- 
tions during a part of the past. 

These four hypotheses as to the time of origin of loess imply 
differences in its climatic relations. If loess was chiefly formed 
during typical interglacial epochs, or toward the close of such epochs, 
profound general aridity must seemingly have prevailed in order 
to kill the vegetation and thus enable the wind to pick up sufficient 
dust. If the loess was chiefly formed during times of extreme 
glaciation when the glaciers were supplying large quantities of fine 
material to out-flowing streams, less aridity would be required, but 
seemingly there must have been sharp contrasts between wet 
seasons in summer when the snow was melting and dry seasons in 
winter. Alternate floods and droughts would thus affect broad 
areas along the streams. Hence arises the hypothesis that the wind 
obtained the loess from the flood plains of streams at times of 
maximum glaciation. If the loess was chiefly formed during the 
rapid retreat of the ice, alternate summer floods and winter droughts 
would still prevail, but much material could also be obtained by the 
winds, not only from flood plains, but also from the deposits exposed 
by the melting of the ice and not yet covered by vegetation. 

In support of the hypothesis of the interglacial origin of loess, 
Shimek and others state that the glacial drift which lies beneath 
the loess commonly gives evidence that some time elapsed between 
the disappearance of the ice and the deposition of the loess. For 
example, most of the locally abundant shells of snails in the loess 
are not of the sort now found in colder regions, but resemble those 
found in the drier regions. It is probable, Shimek concludes, that 
if they represented a glacial epoch all would be dwarfed by the 
cold as are the snails of far northern regions. The gravel pavements, 
discussed below, are pointed out by Shimek as strong evidence of 
erosion between the retreat of the ice and the deposition of the loess. 

Turning to the second hypothesis, namely that the loess accumu- 
lated near the close of the interglacial epoch rather than in the 


474 STEPHEN SARGENT VISHER 


midst of it, we may follow Penck.t The mammalian fossils seem 
to him to prove that the loess was formed while boreal animals 
occupied the region, for they include remains of the hairy mammoth, 
woolly rhinoceros, and reindeer. On the other hand, the typical 
interglacial beds not far away yield remains of species characteristic 
of milder climates, such as the elephant, the smaller rhinoceros, 
and the deer. In connection with these facts it should be noted 
that occasional remains of tundra vegetation and of trees are found 
beneath the loess, while in the loess itself certain steppe animals, 
such as the common gopher, or spermaphyl, are found. Penck 
interprets this as indicating a progressive desiccation culminating 
just before the oncoming of the next ice sheet. 

The evidence in favor of the hypothesis that the loess was 
formed during the maximum extension of the ice chiefly concerns 
its relation to the ice sheets and to the streams which flowed from 
the melting ice. If the great American deposits of loess do not 
represent the outwash from the Iowan ice, there is little else that 
does, and presumably there must have been outwash. Also the 
distribution of loess along the margins of streams suggests that 
much of the material came from the flood plains of overloaded 
streams flowing from the melting ice. Furthermore, in many 
places at least, the drift just beneath the loess presents little or 
no evidence of having been weathered or leached before the loess 
was laid down. Chamberlin found that many tests showed it to 
contain about as much calcareous material as the loess itself. 
This suggests that it was laid down at about the same time as the 
underlying drift, not notably afterward.? Likewise, although 
Shimek has emphasized the fact that most of the snails do not show 
depauperization, McGee reports that depauperization is evident 
among those found near the glacial margins, and that shells are 
very rare there. Both of these conditions suggest that much of 
the loess accumulated under glacial conditions. 

Thus although there are some points in favor of the hypothesis 
that the loess originated (z) in strictly interglacial times, (2) at 


= Penck’s conclusions are given in full in W. B. Wright, The Quarternary Ice Age, 
London, 1913. 


2'T. C. Chamberlin, personal communication. 


THE TIME OF GLACIAL LOESS ACCUMULATION 475 


the end of the interglacial epochs, and (3) at times of full glaciation, 
each hypothesis is much weakened by evidence that supports the 
others. The evidence of boreal animals seems to disprove the 
hypothesis that the loess was formed in the middle of a mild inter- 
glacial epoch. On the other hand, Penck’s hypothesis as to loess 
at the end of interglacial times fails to account for certain character- 
istics of the lowest part of the loess deposits and of the underlying 
drift. Instead of normal valleys and consequent prompt drainage, 
such as ought to have developed before the end of a long interglacial 
epoch, the surface on which the loess lies shows many undrained 
depressions. Some of these can be seen in exposed banks, while 
many more are inferred from the presence of shells of pond snails 
here and there in the overlying loess. The pond snails presumably 
lived in shallow pools occupying depressions in the uneven surface 
left by the ice. Another reason for questioning whether the loess 
was formed chiefly at the end of an interglacial epoch is that this 
hypothesis does not provide a reasonable origin for the material 
which composes the glacial loess deposits of important loess- 
covered regions. Near the Alps, where the loess deposits are 
small and where glaciers probably persisted in the interglacial 
epochs and thus supplied flood material in large quantities, this 
shortcoming perhaps does not appear important. In the broad 
Upper Mississippi basin, however, and also in the Black Earth 
region of Russia there would seem to be, during an interglacial 
epoch, no way to get the large body of material composing the loess, 
except by assuming the existence of great deserts to windward. 
But there is little or no evidence of such deserts where they could 
be effective. The mineralogical character of the loess of Iowan 
age proves that the material came from granitic rocks, such as 
formed a large part of the drift. The nearest extensive outcrops 
of granite are in the southwestern part of the United States, nearly 
a thousand miles from Iowa and Illinois. But the loess is thickest 
near the ice margins and thins toward the southwest and in other 
directions, whereas if its source was the southwestern desert its 
maximum thickness would probably be near the margin of the 
desert. Furthermore the similarity in calcareous content of the 
loess and the underlying drift, reported by Chamberlin, points 


476 STEPHEN SARGENT VISHER 


against Penck’s hypothesis, for if the loess did not accumulate 
until near the close of an interglacial epoch, it is probable that the 
calcareous matter would have been largely leached from the upper 
layers of the underlying drift. The interglacial epochs are now 
known to have been sufficiently long for much weathering to take 
place. Thus considerable evidence seems inconsistent with the 
hypothesis that the loess was formed chiefly toward the close of an 
interglacial epoch. 

There is much less evidence against the hypothesis that the great 
loess deposits accumulated chiefly during the maximum extension 
of the ice. Indeed it remains as a worthy working hypothesis. 
However, the question may be raised as to whether or not flood 
plains of streams would provide adequate supplies of materials for 
such widespread, rather uniform deposits as those of Russia and 
of Iowa and Illinois. A further question comes to mind: Would 
the type of vegetation which would probably occur along the ice 
front at its maximum extension be that of which the loess gives 
evidence? Indeed it seems probable that when the ice advanced, 
its front lay close to areas where the vegetation was not much thinner 
than that which today prevails under similar climatic conditions. 
If the average temperature of glacial maxima was only about 6°C. 
lower than that of today, as many authorities consider likely, the 
conditions just beyond the ice front when it was in the loess region 
from southern Indiana to Nebraska would probably have been 
like those now prevailing in Canada from New Brunswick to 
Winnipeg. The vegetation there is quite different from the grassy 
vegetation of which evidence is found in the loess. The roots and 
stalks of such grassy vegetation are generally agreed to have helped 
produce the columnar structure which enables the loess to stand 
with almost vertical surface. Thus it seems appropriate to add a 
supplementary hypothesis to suggest that certain phenomena would 
be readily explainable in case the chief accumulation was during 
glacial retreat, rather than at the time of maximum extension of ice. 

We are now ready to consider the probability that loess accumu- 
lated mainly during the retreat of the ice. Such a retreat exposed 
a zone of drift to the out-blowing glacial winds. Most glacial 
hypotheses, such as that of uplift, or depleted carbon dioxide, 


THE TIME OF GLACIAL LOESS ACCUMULATION 477 


call for a gradual retreat of the ice scarcely faster than the vegetation 
could advance into the abandoned area. Under Huntington’s 
solar-cyclonic hypothesis,” on the other hand, the climatic changes 
may have been sudden and hence the retreat of the ice may have 
been much more rapid than the advance of vegetation. Now 
wind-blown materials are derived from places where vegetation 
is scanty. Scanty vegetation on good soil, it is true, is usually 
due to aridity, but may also result because the time since the soil 
was exposed has not been long enough so that it may be covered 
with vegetation. Sand bars, mud flats, and flood plains are common 
examples. Moreover, violent winds and low temperatures may 
prevent the spread of vegetation. Thus it appears that unless ‘the 
retreat of the ice were as slow as the advance of vegetation, a barren 
area of more or less width must have bordered the retreating ice 
and formed an ideal source of loess. 

Several other lines of evidence seemingly support the conclusion 
that the loess was chiefly formed during the retreat of the ice. 
For example, Shimek, who has made almost a life-long study of 
the Iowan loess, emphasizes the fact that there is often an accumu- 
lation of stones and pebbles at its base. This suggests that the 
underlying till was eroded before the loess was deposited upon it. 
The first reaction of most students is to assume that of course 
this was due to running water. That is possible in many cases, 
but by no means in all. So widespread a sheet of gravel could not 
be deposited by streams without destroying the irregular basins 
and hollows of which we have seen evidence where the loess lies 
on glacial deposits. On the other hand, the wind is competent to 
produce a similar gravel pavement without destroying the old 
topography. ‘“‘Desert pavements” are a notable feature in most 
deserts. The commonest winds are outward near the edge of an 
ice sheet, as Hobbs has made us realize.2 They often attain a 
velocity of eighty miles an hour in Antarctica and Greenland. 


tEllsworth Huntington, Earth and Sun, Yale Press, New Haven, 1922; and 
Huntington and Visher, Climatic Changes, Their Nature and Causes, Yale Press, 1922. 


2W. H. Hobbs, Characteristics of Existing Glaciers, 1911; ‘‘The Role of the 


Glacial Anti-Cyclone in the Air Circulation of the Globe,”’ Proceed. Am. Phil. Soc., 
Vol. LIV (10915), pp. 185-225. 


478 STEPHEN SARGENT VISHER 


Such winds, however, usually decline rapidly in velocity only a few 
score miles from the ice. Thus their effect would be to produce 
rapid erosion of the freshly bared surface near the retreating ice. 
The pebbles would be left behind as a pavement, while sand and 
then loess would be deposited farther from the ice where the winds 
were weaker and where vegetation was beginning to take root. 
Such a decrease in wind velocity may explain the occasional vertical 
gradation from gravel through sand to coarse loess and then to 
normal fine loess. As the ice sheet retreated, the wind in any given 
place would gradually become less violent. As the ice continued 
to retreat, the area where loess was deposited would follow at a 
distance, and thus each part of the gravel pavement would in turn 
be covered with loess. 

The hypothesis that loess is deposited while the ice is retreating 
is in accord with many other lines of evidence. For example, it 
accords with the boreal character of the mammal remains as de- 
scribed above and of the depauperated snail fauna found in the zone 
nearest the ancient ice sheets. Again, the advance of vegetation 
into the barren zone along the front of the ice would be delayed by 
the strong out-blowing winds. The common pioneer plants depend 
largely on the wind for the distribution of their seeds, but the 
glacial winds would carry them away from the ice rather than toward 
it. The glacial winds discourage the advance of vegetation in 
another way, for they are drying winds, as are almost all winds 
blowing from a colder to a warmer region. Such winds, however, 
would interfere less with the northward spread of grasses propagated 
by root shoots and by abundant seeds than it would interfere 
with the spread of trees. The fact that remains of trees sometimes 
occur at the bottom of the loess probably means that the deposition 
of loess extended into the forests which almost certainly persisted 
not far from the ice at its maximum advance. This seems more 
likely than that a period of severe aridity before the coming of the 
glacier killed the trees and made a widespread steppe or desert. 
Penck’s chief argument in favor of the formation of loess before the 
advance of the ice rather than after appears to be that since loess 
is lacking upon the youngest drift sheet in Europe it must have 
been formed before rather than after the last or Wiirm advance of 


THE TIME OF GLACIAL LOESS ACCUMULATION 479 


the ice. This argument is not convincing for two special reasons: 
First, on the corresponding (Wisconsin) drift sheet in America 
loess is present—in small quantities to be sure, but unmistakably 
present. Second, there is no reason to assume that conditions 
were identical at each advance and retreat of the ice. Indeed, 
the fact that in Europe, as in the United States, nearly all the loess 
was formed at one time, and only a little is associated with the 
other ice advances, points clearly against Penck’s fundamental 
assumption that the accumulation of loess was due to the approach 
of a cold climate. The relative abundance of loess associated with 
the Iowan ice sheet would be explained by the present hypothesis 
if ice retreated more rapidly for a time than did any of the later ice 
sheets. 

Thus the hypothesis that the loess accumulated chiefly during 
the retreat of the ice sheets appears to have enough support to 
merit consideration by students of loess. 


SER RE RS sR ES TS 
MEMORIAL EDITORIAL 


ROLLIN D. SALISBURY 
August 17, 1858—August 15, 1922 


It is with deep sorrow that the Journal of Geology records the 
death of its active managing editor, Dean Rollin D. Salisbury. 
After a severe illness of two and a half months, he passed away 
on the evening of August 15, within two days of his sixty-fourth 
birthday. For the past four years he had been the responsible 
editor of the Journal, while from its founding in January, 1893, 
he participated actively in the general responsibilities of its editorial 
management and had special charge of contributions relating to 
the physiographic aspects of geology. This special service in the 
dissemination of the literature of the science of the earth thus ran 
through a period of almost thirty years. The more than 1,300 
standard articles, 2,500 abstracts and reviews, 150 editorials and 
shorter notices, embracing more than 24,000 pages of printed 
matter, which received all phases of editorial care from the reading 
of manuscript to the approval of the final proof, attest at once 
the importance and the burden of this work. Professor Salisbury 
himself prepared 82 contributions. 

The scientific investigations of Dr. Salisbury will be reviewed 
in a later article more fully than is possible here. His field work was 
begun under the auspices of the United States Geological Survey as 
early as 1881 and continued until 1910. It embraced extensive 
studies on the glacial and other Pleistocene formations of the north- 
ern states and the lower Mississippi Valley. In connection with 
this, he made a report on Crowley’s Ridge to the Geological Survey 
of Arkansas. From 1891 to 1910 he was geologist in charge of 
the Pleistocene Division of the Geological Survey of New Jersey, 
where his work on the older drift and on fluvial deposits formed 
near the neutral zone between marine and upland horizons was 
notable for its keen insight and acute discrimination. He made 
important contributions to the Geological Survey of Illinois and 

480 


MEMORIAL EDITORIAL 481 


in 1919 was appointed to the Board of Commissioners in charge 
of the Survey. Besides these official services he made independent 
investigations in several lines. He was geologist of the Peary 
Relief Expedition to Northern Greenland in 1895, in connection 
with which he studied existing glaciers under the unparalleled 
advantages presented in very high latitudes. 

Dr. Salisbury was a very lucid writer. The reports of his 
researches and the texts of the several works he prepared for the 
general reader and for students put into the easy possession of others 
what he saw so clearly himself. The printed results of his studies 
in field and office will long stand as a lasting memorial to Professor 
Salisbury’s industry and clarity of vision. 

Large and important as were these contributions, Dr. Salis- 
bury’s greatest service to science lay in his singular success in 
stimulating and training young talent not only for the teaching of 
science but for research. This distinguished service began at 
Beloit College, 1883-91, was continued at the University of Wis- 
consin, 1891-92, and was transferred to the University of Chicago 
at its opening, where he took part in founding the Department of 
Geology thirty years ago. For nearly twenty years he was active 
executive of the Department and for the last four years bore its 
full responsibilities. In connection with this geological service 
he developed the Department of Geography and served as its head 
from 1913 to 1918, when he was made head of the Department of 
Geology, and the Department of Geography was transferred to 
one who, first as a student under him, and then as a colleague, had 
grown to marked efficiency. From 1899 onward Dr. Salisbury 
was dean of the Ogden (Graduate) School of Science of the Uni- 
versity of Chicago. In these varied relations he came into touch 
with thousands of young minds and gave them effective impulses 
toward sound scholarship and the higher life. The ultimate effects 
of this work are beyond estimation. Through the growing efficiency 
and the rising power of the young talent thus inspired by his 
leadership, Dean Salisbury’s greatest service to science and to 


humanity has only fairly begun. 
PCAC: 


PETROLOGICAL ABSTRACTS AND REVIEWS 
ALBERT JOHANNSEN 


Harker, ALFRED. Petrology for Students. Cambridge, 1919, 5th 
ed. Pp. 300, figs. 100. 


A comparison of the fifth with the fourth edition of Harker’s book shows 
the chief change to be that practically all of the American examples have been 
“improved out.” With the exception of one chapter where nephelite- and 
leucite-syenites are now separated from the syenites, the two editions could be 
used in the same class, for though the pagination is different, the different para- 
graphs can be easily located. The change of the chapter heading “ Diabase”’ 
to “‘Dolerite’’ will not be considered an improvement in this country. Both 
terms are bad, in that each has two meanings; the former being used for ophitic 
dike-rocks of the composition of gabbros as well as for Paleozoic basalts, the 
latter, originally for coarse-grained basalts, now according to British usage for 
the rocks we call diabase. Harker still clings to the classification of diorite as a 
hornblende-bearing rock (p. 63) while gabbros are defined as “‘characterized by 
pyroxenes in place of hornblende”’ although he recognizes the modern tendency 
in classifications when he says: ‘“‘ The distinction between the hornblende- and 
augite-bearing types is rather an artificial one. It was established before the 
strong tendency of atgite to pass over into hornblende was thoroughly appreci- 
ated.”’ Also, ‘‘The family as so defined cannot be regarded as a natural one.” 
The feldspar of diorite is given as ‘‘andesine or labradorite or exceptionally a 
more basic variety,’’ while in gabbros it is given as “labradorite, with exception- 
ally one more acid and occasionally orthoclase.” 


HirscHwatp, J. Letisdize fiir die praktische Beurteilung, zweck- 
massige Auswahl und Bearbeitung natiirlicher Bausteine. Berlin, 
LOLS. ps soy uss. Lo: 

This book was primarily written for stone-workers and aims at giving the 
most important points needed in the selection and judging of building-stone. 
There is a short classification of rocks giving the desirable and undesirable 
characteristics of each, a chapter on the examination of stone quarries with 
examples of written forms for describing quarries, and chapters on the essen- 
tials of stone-testing, advantageous modes of working and using certain rocks 
with reference to schistosity, cleavage, etc., various uses of different kinds of 


stone, etc. 
482 


PETROLOGICAL ABSTRACTS AND REVIEWS 483 


Hoimes, ARTHUR. The Nomenclature of Petrology. London, 1920. 
Pp. 284. 

This little book is intended as an English substitute for Loewinson-Lessing’s 
Lexique Pétrographique, which is now about twenty years old. It is a very 
convenient and useful volume, giving in brief form, definitions of most of the 
common terms used in petrology. The only fault that might be found with 
it is that it is too brief, and that the references are, in many cases, not to the 
work in which the term was originally given, but usually to later British authors. 
In other cases references are hard to find; for example, the reference for the 
source of umptekite is under “‘chibinite,” and maenaite under “‘grorudite,” but 
this can hardly be called satisfactory, especially since no hint is given as to the 
terms under which they may be found. Other references, such as ‘‘Syenodio- 
rite, Evans, 1916,” etc., are too incomplete to be traced. On the whole, how- 
ever, the book is very good, and it is likely to prove useful to students by giving 
them the means for quickly finding unfamiliar terms. 


Hormes, Artuur. Petrographic Methods and Calculations. Lon- 
dons 192m Pp: 515, figs. 82, pls. 4. 

Quite different from previously published books on petrographic methods 
is this one by Holmes, in that very little space is devoted to optical methods. 
The author says in his preface, ‘“‘I have tried to produce a more evenly balanced 
treatise which should penetrate the petrological domain, and not merely skirt 
its border land.” Chapter i deals with Petrology, its scope, aims, and applica- 
tion, chapter ii with specific gravity, chapter iii (40 pp.), separation of minerals 
by heavy liquids, magnet, etc., chapter iv (52 pp.), optical examination of min- 

-erals, including not only optical properties but tables for the determination of 
minerals, chapter v (71 pp.), the examination of detrital sediments, chapter vi 
(19 pp.), the preparation of thin sections, chapter vii (45 pp.), microchemical 

-and staining methods, chapter viii (38 pp.), the examination of thin sections, 
including a discussion of the Rosiwal and similar methods, genesis of minerals, 
saturated and unsaturated minerals, etc. Incidentally it is stated that “‘Jo- 
hannsen has suggested the use of a planimeter for measuring areas of particular 
minerals on a drawing of a thin section made with the aid of a camera lucida. 
The procedure is more complicated and tedious than the ordinary linear method 
now in general use.”” But in the method mentioned no camera lucida drawing 
is made and the time required is less than a fourth of that of the Rosiwal. 
Chapter ix deals with rock textures, chapter x with chemical analyses and 
their interpretation and includes a discussion of the C.I.P.W. system, and 
chapter xi with graphical representation of chemical analyses. The book 
thus brings together much of the miscellaneous data so often needed by the 
petrographer. It will therefore be examined with interest and profit by all 
advanced workers in the science. 


484 PETROLOGICAL ABSTRACTS AND REVIEWS 


Homer, W. ‘‘Systematische Petrographie auf genetischer Grund- 
lage,’ Vol. I, Das System. Gebriider Borntraeger, Berlin, 
HOMO Jeane, iis i OS E. 

Hommel proposes a double-barreled system of classification, in that he 
represents a rock by a formula composed of two parts. The first, which he 
calls his ‘‘ Molekularformel,” is derived directly from the chemical analysis and 
represents the molecular proportions of the various oxides. The second, or 
“‘Konstitutions formel,” represents the percentages of certain minerals cal- 
culated from the chemical analysis. The percentages of the computed minerals, 
however, resemble those of the normative minerals of the C.I.P.W. system. 
since they generally do not represent the amounts actually present in the rock. 
As in the C.I.P.W. system, the feldspars are calculated from the amounts of 
KO, Na.O, and CaO present in the analysis. Orthoclase is considered pure 
K.O, AI.O;, 6Si0., and all the soda is calculated as part of the plagioclase. As 
a matter of fact, a considerable amount of the soda may occur in the orthoclase 
or microcline as lamellar intergrowth, either visible or invisible. Where all 
is calculated in the plagioclase, the resulting mineral is too near the soda end. 
Potash in muscovite is disregarded, and the only lime or alumina used for the 
dark minerals is that which occurs in excess of the requirement of the theoretical 
feldspars. 

So far as the first part of the formula is concerned, no objection can be made 
to it, since it is confessedly chemical. It has the advantage of simplicity over 
the C.I.P.W. system, and can be shown in a diagram. Whether it is better 
than Osann’s system is a question. Certainly Osann’s latest modification 
(Der chemische Faktor in einer natiirlichen Klassifikation der Eruptivgesteine, 
Heidelberg, 1919. Review to follow in this Journal) offers a rapid means for 
finding rocks of similar composition. Since only Part I of Hommel’s book is 
printed, however, it is possible that similar groups of rocks may appear later. 

Hommel’s system can best be illustrated by an example. The kern granite 
of the Brocken in molecules, reduced to 100, gives SiO, 80.2, TiO: —, Al,O; 8.6, 
Fe,O; 0.5, FeO 1.5, MgO 0.4, CaO 1.4, Na,O 3.3, K.0 4.1. Hommel’s formula 
for this rock is: 

80.2 3 4.1 F e mc 2.6=+31 P | 0,01. [2 Be 


I.2 2 


The first part is the molecular formula and is obtained as follows: The figures 
80.2 represent the amount of the silica molecules (plus TiO, and P.O;) as com- 
puted above. The large figures immediately after the sign 2 represent the 
K,O molecules; the Na,O mol. and the CaO mol. in feldspar being represented 
by the figures above and below the sign. The CaO figures necessarily cannot 
be greater than the amount of Al,O; remaining after enough has been used to 
satisfy the soda and potash. Excess CaO is added to the mafic minerals as c. 
The mafic constituents are represented by e for Fe.O,, f for FeO4+-MnO, m for 


———— 


PETROLOGICAL ABSTRACTS AND REVIEWS 485 


MgO, and c for the remaining CaO. If all of the CaO was used for feldspar, 
the excess of Al.O; is represented by ¢t among the mafic constituents. If there 
is insufficient Al,O, to satisfy the alkalies, the remaining Na.O is indicated by 
m and it also is placed among the ferromagnesian constituents. In the above 
example, K,0+Na,0=4.1+3.3=7.4, corresponding to 7.4 ALO, leaving 1.2 
AIO; to unite with CaO for the anorthite radical. The remaining 0.2 CaO is 
represented by c among the mafic constituents. The ferromagnesian con- 
stituents are given in the order of decreasing abundance, and their sum is 
represented by the figure following (here 2.6). The relative abundance is 
indicated in the symbol, thus 


fm indicates an FeO/MgO ratio below 5 FeO/4 MgO. 

f= 14 f, therefore fm shows a ratio between 4FeO: 3MgO and 3FeO: 2MgO. 
f=2 f, or fm indicates the ratio of 2/r. 

F=3f, and Fm=3 FeO to 1 MgO. 

F=af, and Fm=4/1. 


F=s5f, and Fm=s/t. 


f= Gi, 
F,=8f, 
F;=10, 


F,=12f, etc., etc. 


Since the remaining constituents in the order of decreasing abundance in the 
example given are FeO=1.5, Fe.0,;=c.5, MgO=0.4, and CaO=o.2 (the ratios 


being therefore as 7 : 2.5 : 2 : 1 approximately), they are represented by Fie: 


m:c. (These symbols are complicated. Why not use the ratio directly as 
subscripts; thus here F;.; . €o.5 . Mo.4 . Co-2, or even the simplified similar ratios 
Peo Gag o WOH 6 ©))g 

The rock may be plotted on rectangular co-ordinates as shown by the figure. 
The horizontal line (S axis) represents the silica content. The vertical line 
through so per cent is the 2 axis and is the measure of the feldspar content. 
It is drawn at double the scale of the S axis because for each molecule of alkali 
or lime, a molecule of AI,O, is also automatically included. Further, there are 
two diagonal lines at 45° through the 50 per cent and too per cent points on the 
S line; the former is called the quartz normal, the latter the F axis. 

The Brocken granite is thus plotted by measuring from the 80.2 point on the 
S axis downward 4.1 for K,0. This K point is indicated by an x. A further 
distance of 3.3 for Na,O is laid off on the same line and the point is marked by a 
circle. This is called the Locus of the rock (Ort des Gesteins). Finally 1.2 is 


486 PETROLOGICAL ABSTRACTS AND REVIEWS 


plotted on the extension of the line for CaO, and this is also marked by an x. 
This is the 2 point. The horizontal distance of the 2 point from the F axis 
(in units as measured on S) gives the amount of the ferromagnesian constituents. 
The amount of free (theoretical) quartz is given by the horizontal distance of 
the Locus of the rock from the “‘Quartz normal.”” (As measured on the S scale, 
the reading is to be multiplied by 2.) In the case of the Brocken granite this 
distance is 15.4, therefore the theoretical amount of free quartz is 30.8. 

Since the quartz normal and the F axis are both at 45°, the readings may be 
made along the downward extension of the line passing through the locus of the 
rock, instead of along the horizontal. 


The second part of the formula, the “Konstitutionsformel,’’ represents 
theoretical minerals. The method for computing the free quartz follows from 
the diagram. Twice the sum of the alkalies is subtracted from the silica value 
less fifty (the diagonal line starting at the 50 point), and the result is multiplied 
by two. 

Free quartz=2[(SiO.—50)—2(K.0+Na.O)], where SiO., K.O, etc., are 

‘molecular values. The Brocken granite therefore gives 2[(80.2—50)— 
2(4.1+3.3)|=+30.8 or approximately +31. The plus sign gives the position 
of the rock above the quartz normal in the diagram, and shows free quartz. 
Below the line is shown by a minus sign. This indicates nephelite, olivine, 
etc. The letter P (plutonic) indicates the texture. O stands for orthoclase. 
Its value is obtained by multiplying the K,O mol. by 8, since the formula is 
K,0.AL0;.6SiO.. (K,0=4.1, O=8X4.1=32.8 or ca.33). The plagioclase 
indicated by the molecular formula is 3.3 Na.O, 1.2 CaO, or basic oligoclase, 
Ab¢sAn;. The molecular value of the albite is eight times the Na.O, but the 
anorthite, with the formula CaO.Al,0;.2SiO. is only four times the CaO. 
(3.3 8+1.2X4=31.2). The symbol for basic oligoclase is Ol’. Beta indi- 
cates biotite, and 5 per cent represents the remainder left for this mineral. 
It amounts, of course, from the formula RO.SiO, to twice the F value. 

With insufficient Al,O;, the symbol A is used in the formula instead of 2, 
and the aegirite molecule takes the place of the anorthite. Thus A 4 means 


PETROLOGICAL ABSTRACTS AND REVIEWS 487 


4(K,0.AL0,). 6(Na20.ALO;). 2(Na.O. Fe.0;). These rocks may contain free 
quartz (or nephelite). Where there is quartz it is to be noted that only four 
molecules of SiO, are required to one of Na.O, consequently there will be more 
quartz than in the case where all the soda goes into feldspar. The formula for 
the quartz, in this case, consequently, reads 2[(SiO.mol.— 50) — {2(K.O mol.+ 
Na.O mol.)-++aeg. mol }]. 

With insufficient silica, so that a negative result is obtained by the quartz 
formula, it is evident that olivine or nephelite (or leucite) is present. The 
formula here is divided instead of multiplied by 2, and we have nephelite 
_ SiO.—s50—[2(K.0+Na,0)+aeg. mol. 

2 
feldspar and feldspathoid only). 

Examples of the nephelite rocks and rocks with insufficient alumina would 
make this abstract too long. 

The third section of the book consists of a natural classification of rocks on 
genetic principles. A detailed description is impossible here and the reader is 
referred to the original work which is full of suggestions. Briefly, rocks are 
divided into five groups: (1) the orthogene class of rocks includes those 
formed by slow cooling in the interior of the earth; (2) the paragene class 
originated in a relatively rapid cooling of the magma during the period of 
crystallization of olivine, consequently that mineral could not settle out but 
is represented by its recrystallizations as diopside, etc.; (3) the hypogene 
class includes extrusive rocks formed under conditions of rapid cooling, relief 
from pressure, and movement; (4) peratogene rocks are those altered by 
metamorphism and include the crystalline schists; (5) diagene rocks include 
the sediments. The first four groups are subdivided into ten zones each, each 
zone representing a distinct temperature-interval through which the magma 
passed on cooling, and marked by the separation of typical and chemically well- 
defined minerals. There are thus the zones of (1) chromite, (2) olivine, (3) 
enstatite, (4) diopside, (5a) labradorite, (5b) nephelite, (6) labradorite-ande- 
sine, (7) andesine, (8a) andesine-oligoclase, (8b) oligoclase, (9) orthoclase, 
(10) quartz. 

The book is one which will repay very careful study in every part. While 
one might wish that the “‘ Konstitutionsformel” represented actual rather than 
theoretical minerals, one can easily understand how the former might make 
difficulties in certain cases. Undoubtedly this book is one of the most impor- 
tant works on petrography that has appeared in recent years. 


(the alkali molecules are those in’ 


Ippincs, J. P.,and Mortery, E.W. “A Contribution to the Petrog- 
raphy of the South Sea Islands,” Proc. Nat. Acad. Sci., IV 
(1918), 110-17. 

This is a preliminary statement of the geological structure and character 


of seven islands of the South Seas, including Tahiti, Moorea, and the Society 
Group. Thirty new chemical analyses are here published for the first time. 


488 PETROLOGICAL ABSTRACTS AND REVIEWS 


The rocks of Tahiti are almost wholly olivine basalts, some with nephelite or 
analcite. Trachytic and phonolitic lavas occur on five of the seven islands. A 
nephelite-latite, consisting of alkalic feldspar and andesine with some nephelite 
and sodalite and abundant hornblende, much titanite, and few augites and 
micas is given the new name of fautirite, from the valley, Tautira, Taiarapu, in 
which it occurs. The mode is not given but the norm contains Or 31.1, Ab 28.3, 
An 13.3, and Ne 10.8. 


Jeuu, T.J., and Camppett, Ropert. “The Highland Border 
Rocks of the Aberfoyle District,” Trans. Roy. Soc. Edinburgh, 
LIT (ror), 175-212, mes) 10, pls. 6: 

A series of grits, shales, limestones, cherty shales, graphitic shales, and 
various igneous rocks, in places highly altered, extending from Stonehaven to 
Arran, is grouped under the term ‘‘The Highland Border Rocks.” They are 
here arranged in two divisions. The sediments of the Lower Series are either 
Upper Cambrian or transitional between Cambrian and Ordovician, and are 
thought to have been deposited in clear water near the verge of sedimentation. 
The lavas indicate submarine eruptions. The Upper Series, which belong 
higher in the Ordovician, include limestones, hornblende- and chlorite-schists, 
and igneous rocks. 


JoHANNSEN, ALBERT. Essentials for the Microscopical Determina- 
tion of Rock-forming Minerals and Rocks in Thin Sections. 
Chicago, 1922. Quarto, pp. vi + 54, figs. 24, pls. 6, and a 
folding table. 


This laboratory manual contains practically all of the data originally pub- 
lished in the writer’s Determination of Rock-forming Minerals, and in addition 
gives modes of occurrence and many more points of separation between similar 
minerals. Only a few very rare species, such as johnstrupite, mosandrite, 
laavenite, etc., have been omitted, but by uniting the tables which contained 
minerals having birefringences greater or less than quartz, and refractive indices 
greater or less than Canada balsam, much repetition has been avoided, and the 
number of pages has been materially reduced. Orthorhombic minerals have 
been united with the other biaxial minerals, since sections which cut all of the 
crystallographic axes in this system show inclined extinction, but the extinc- 
tion angles are given in the descriptions. The separation lines between the 
various plagioclase feldspars have been changed from those given in the former 
book to 5, 273, 50, 724, and 95 per cent anorthite. Albite and anorthite have 
been limited to a variation of only 5 per cent since these names are also applied 
to the pure end members, and compound names, such as oligoclase-albite, 
labradorite-bytownite, etc., have been omitted. The section on the deter- 
mination of the feldspars has been but little reduced, but that on optical 


PETROLOGICAL ABSTRACTS AND REVIEWS 489 


methods has been condensed as much as possible, since this data is given else- 
where. The alphabetical list of minerals has been much extended, and is now 
placed at the back of the book in such a position that it may be consulted 
without turning pages. Finally, a short section on the classification of igneous 
rocks according to the author’s system has been added. As given here, this 
classification is essentially the same as that published in the Journal of Geology 
several years ago, except that the whole of the monzonitic series, with the 
exception of monzonite and quartz-monzonite, has been omitted. 


Kaiser, Ericu. “Der Elaolithsyenitlakkolith der Serra de Mon- 
chique im stidlichen Portugal,” Neues Jahrb., B. B., XX XTX 
(1914), 225-67, pls. 2, profiles 2, map 1, figs. 6. 

Discusses the structure of the Serra de Monchique in southern Portugal, 
with special reference to the nephelite-syenite central mass. He considers this 
a laccolith that was intruded during the period of folding. The main foyaitic 
mass with border pulaskites, and various dike-rocks are briefly described. 


Kaiser, Ericu. ‘Uber ein Demonstrationsmikroskop fiir den 
mineralogischen und petrographischen Unterricht,” Zezischr. 
f. Kryst., LIT (1914), 397-403. 

Describes a microscope for demonstration purposes. It has a large, rotat- 
able stage with ten openings, so that that number of thin sections can be placed 
on it at one time, and the specimens observed one after the other. A glass 
plate prevents the disturbance of the sections by students. Since the prepara- 


tions cannot be turned there is provided a means for rotating the tube of the 
" microscope. 


Kaiser, ErtcuH. ‘‘Studien wahrend des Krieges in Stidwestafrika,”’ 
Zeitschr. d. deutsch. geol. Gesell., LX XII (1920), 50-76. 


The three studies included in these papers are: (1) Assimilationsercshein- 
ungen an den Elaeolithsyeniten des Granitberg in siidlichen Namib,” (2) “Zur 
Kenntnis der Hohlformen, Eindeckungen, Ausfiillungen und Aufschiittungen 
der Trockengebiete,” and (3) ‘‘Kalkkrusten.”’ In the first paper the nephe- 
lite-syenite stock of Granitberg, in southern Namib, is described. Besides the 
normal rock there are certain other types which cannot be explained by normal 
differentiation. The contact of the main mass is irregular and blurred, owing 
to the penetration of the country rock by numerous veins and apophyses of 
the igneous rock, thus giving the impression of an irregular breccia penetrated 
by the nephelite-syenite. But it is not a breccia in the ordinary sense since 
the fracturing was caused by the intrusion itself. Both intrusive and intruded 


490 PETROLOGICAL ABSTRACTS AND REVIEWS 


rock are altered, the character of the resultant depending upon the nature of 
the country rock. Against sandstone and arkoses there is a decrease in the 
feldspathoids and a transition to syenite and even to quartz-bearing and 
nephelite-free alkali granites. Where the igneous rock is in contact with the 
dolomitic Cambrian rocks there is an increase in the amount of nephelite and 
the grain is coarse, in some cases giving nephelite-syenite-pegmatite phases 
with nephelites several centimeters in size. Such changes in the character of 
the rock cannot be explained by differentiation alone. Kaiser thinks that 
assimilation of the country rock is clearly indicated, although he supposes that 
there was also subsequent differentiation. He warns, however, against using 
such local occurrences either for or against the assimilation theory of magmas 
in general. The mechanics of the intrusion, he thinks, agree with Daly’s 
theory of magmatic stoping. He believes the magma reached its present 
position not by intrusion between strata, but by forming a place for itself. 


Kaiser, Ericu. “Bericht iiber geologische Studien wahrend des 
Krieges in Siidwestafrika,” Abhandl. d. Giessener Hochschul- 
gesell., II (1920). Pp. 57, pls. 6, figs. 4 

A general geological description of Namib, an interesting petrographic des- 
cription of which is given in greater detail in the preceding paper. The scope 
of the work is indicated in the chapter headings, some of which are crystalline 
schists, Cambrian, eruptive rocks, Tertiary, alteration of the Tertiary land 
surface, underground water, mineral occurrences, etc. 


Kato, TAKEO. ‘Microscopic Secondary Sulphide Enrichment in 
the ‘Kuromono’ Ore from the Kosaka Mine, in the Province 
of Rikuchii, Japan, Jour. Geol. Soc. Tokyo, XXV (1918), 1-7, 
pls. 1. 


Kato, Takeo. ‘A Contribution to the Knowledge of the Meso- 
zoic Igneous Rocks Developed Around the Tsushima Basin, 
Japan, Jour. Geol. Soc. Tokyo, XXVII (1920), 1-22, 23-38, 
pls. 4, figs. 2. 


A porphyrite complex with associated tuff beds, late Jurassic in age, was 
invaded by a series of igneous rocks in the following order: (1) effusive quartz- 
porphyry; (2) intrusive quartz-masanite, masano-tonalite, etc., with melano- 
cratic marginal facies; (3) tsingtauites and masano-tsingtauites; (4) dikes of 
masano-hornblendite; (5) aplitic and pegmatitic rocks. The rocks of (z) and 
(2) were probably derived from the same magma basin; (2), (3), (4), and (5) 
are clearly related. Sporadic corroded quartz crystals in some portions of the 


PETROLOGICAL ABSTRACTS AND REVIEWS 491 


porphyrite as well as those in many lamprophyres are thought to have sunk 
from the overlying acid magma layers, in which they were beginning to form, 
into the basic lower layers from which the basic rocks were derived. 


Kato, Takeo. ‘A Contribution to the Knowledge of the Cas- 
siterite Veins of Pneumato-Hydatogenetic or Hydrothermal 
Origin, Jour. Col. Sci. Imp. Univ. Tokyo, XLIII (1920), Art. 5. 
Epyoo, map t, pls. 6, figs. rr. 

Largely economic, although a considerable number of pages are devoted to 
igneous rocks. Among these are diorite, gabbroid-diorite, diorite-mylonite, 
and a rock to which the new name ‘‘akenobeite” is given. Dike-rocks are 
hornblende-hypersthene-andesite, garnetiferous-felsite-porphyry, felsite, por- 
phyrites, and diabase, and there is extrusive rhyolite. The rocks are 
described in some detail but neither chemical nor modal analyses are given. 
The akenobeite is a quartz-monzonite-pegmatite or -aplite, according to Kato. 
It consists of tabular crystals of feldspar in haphazard orientation with the 
interstices filled with an aggregate of quartz. The feldspar is orthoclase and 
oligoclase, the latter always in excess. How much in excess is not stated, so it 
is a question whether the rock might not be called a granodiorite-aplite. Biotite 
is very subordinate and occurs in minute flakes in the feldspars or attached to 
its borders. In the system of the reviewer it belongs to Class 1, Order 2, 
Family 7 (new form), and probably in 7’. The amount of quartz is not stated, 
but from the two photographs given it appears to amount to 15 or 20 per cent. 


KiemM, G. “Die Granitporphyre und Alsbachite des Oden- 
waldes,’ Notizbl. d. Vereins f. Erdk., etc., z. Darmstadt, IV 
Holeewriiasicy conas Pp: To-5o; pis. 2, fg. 1. 

‘In the crystalline schists of the Odenwald there are various dikes of anortho- 
clase-granite-porphyry and of alsbachite. Of the former rock there are twelve 
chemical analyses, eight of them apparently new, and of the latter, three, of 
which two are new. Five new analyses of dike granites are also given. All the 
analyses are re-computed into Osann’s system. 


Kiemm, G. ‘“Bemerkungen tiber die im Gabbro des Franken- 
steins gangartig aufsetzenden Gesteine und tiber seine Ein- 
schliisse von Korundfels,”’ Notizbl. d. Vereins f. Erdk., etc., z. 
Darmstadt, IV Folge, Hf. 35, 1914, 5-9. 


Describes certain corundum-bearing rocks occurring in the Frankenstein 
gabbro. The author disagrees with Kalkowsky, who regards them as diff- 


492 PETROLOGICAL ABSTRACTS AND REVIEWS 


erentiation products of the gabbro magma, but considers them metamorphosed 
inclusions of sedimentary rocks which were resorbed by the gabbro and later 
recrystallized. 


Kiem, G. “Uber die angebliche Umwandlung von Andalusit in 
Disthen in den Hornfelsen des Schiirrkopfes bei Gaggenau in 
Baden,” Zeitschr. d. Deutsch. Geol. Gesell., LXVIII (1916), 
86-92. 

Eisele regarded the disthene in the hornfels from Schiirrkopf in Baden as 
due to alteration of andalusite by orogenic pressures which were post-contact- 


metamorphism; Klemm regards the two minerals as contemporaneous 
products. 


Kiemm, G. “Zur Errinerung an Richard Lepsius,” Notizbl. d. 
Vereins f. Erdk., etc., z. Darmstadt, V Folge Hf. 1, 1915, 5-22. 
With portrait and bibliography. 


Kiemm, G. “Die korundfiihrenden Hornfelse und die Schmirgel- 
gesteine von Laudenau und Klein-Gumpen bei Reichelsheim 
im Odenwald und ihre Nebengesteine,” Notizbl. d. Vereins f. 
Erdk., etc., z. Darmstadt, V Folge, Hf. 1, 1915, 23-41, pl. 1. 


This is a more detailed description of the corundum-bearing rocks of the 
Odenwald, mentioned in the third preceding abstract. Ten new analyses 
are given. 


Kiem, G. ‘Uber den ‘Variolit von Asbach,’”’ Notizbl. d. Vereins 
f. Erdk., etc., z. Darmstadt, V Folge, Hf. 2, 1916 (1917), pl. 1. 


The ‘“‘variolite’”’ from Asbach, called diabase by Chelius, is here determined 
as malchite. The varioles, representing fillings of vesicles, vary in size from 
hazel-nuts to walnuts, occasionally to the size of eggs. They consist of quartz 
feldspar, iron ore, and epidote. 


Kiemm, G. Blatt Neunkirchen. Erlaéuteru. z. Geol. Karte Hes- 
sen. 2d ed. Darmstadt, 1918: Pp. 81, pl. 1. 


In this general geological report on the quadrangle lying between lat. 49°48’ 
and 49°32’, in Hessen, there are short descriptions of many igneous rocks, and 
analyses of 4 hornfelses, 2 amphibolites, 9 gabbros, 3 hypersthene-gabbros, 3 
serpentines, 4 diorites, 14 granites, 5 granite-porphyries, 2 granophyres, 9 mal- 
chites, and 6 odinites. 


PETROLOGICAL ABSTRACTS AND REVIEWS 493 


Kiemm, G. ‘Uber die Enstehung der ‘Felsenmeere’ des Fels- 
berges und anderer Orte im Odenwalde,” Notizbl. d. Vereins f. 
Erdk., etc., z. Darmstadt, V Folge, Hf. 3, 1917 (1918), 3-11, 
pls. 2. 


Considers these particular rock-streams as due to removal of interstitial 
weathered material. The blocks lie practically in the positions where they were 
originally formed. 


Kiemm, G. “Der Granatfels von Gadernheim im Odenwalde und 
sine Nebengesteine,”’ Notizbl. d. Vereins f. Erdk., etc., z. 
Darmstadt, V Folge, Hf. 4, 1919, 3-32. 


The flaser granite of Gadernheim shows an original contact at the south and 
east against metamorphosed sediments. This metamorphism was produced 
by gabbro which altered the sediments to garnetfels, cordierite-hornfels, biotite- 
hornfels, amphibolite, graphite-schist, and graphite-quartzite. The gabbro 
itself was altered by endogene contact action to hornblende-gabbro with a 
dioritic selvage. 


KOLDERUP, CARL FRED. ‘‘Egersund,” Norges Geol. Undersék, No. 
71, 1914. Pp. 60, pls. 4, map 1. 

A geologic report on a quadrangle in the southwestern part of Norway, 
about 60 km. SSE. of Stavanger. The principal part of the area is composed 
of labradoritites and pyroxene-poor norites. The composition of one of the 
latter is given (p. 18) as labradorite 92 per cent, pyroxene 7 per cent. and 
ilmenite 1 percent. There is a smaller area of mangerites and norite-mangerites. 
Among the dike-rocks are mangerite, birkremite, quartz, ilmenite, granite- 
pegmatite, and diabase. 


KOLDERUP, CARL FRED. “Fjeldbygningen i stréket mellem s¢r- 
fjordem og Sammangerfjorden i Bergensfeltet,”’ Bergens 
Museums Aarbok, 1914-15, No. 8. Pp. 257, figs. 91, colored pls. 
2, maps 3. 

The district of the Bergen arches is characterized by a peculiar arrangement 
of the various formations in curves. The rocks are dynamo-metamorphosed 
and are in part sedimentary and in part igneous. Phyllites are the most com- 
mon sediments, and consist of quartz and muscovite with various accessories. 
In the phyllite zone occur serpentines, soapstones, green-schists, saussurite- 
diabases, saussurite-gabbros, labradoritite, norite, mangerite, and birkremite. 
The serpentines are considered metamorphosed igneous rocks, probably peri- 
dotites or pikrites. The green-schists were probably originally basic volcanic 


494. PETROLOGICAL ABSTRACTS AND REVIEWS 


rocks and tuffs. In the western part of the area are Middle Silurian lime- 
stones, metamorphosed to marble. Younger than the marble for it contains 
fragments of this rock, is the polymict conglomerate. It is probably a volcanic 
conglomerate whose cement originally was a tuff. The “gray granite” of 
Reuss is considered a pressed granite. 

The structure of the area indicates that the metamorphism must have taken 
place in the upper part of the earth’s crust where one-sided pressure, low tem- 
perature, and a considerable saturation with water were factors. The rocks 
of the inner arch are more strongly compressed than those of the outer, but 
the materials are of approximately the same kind. 


Koro, Bunpjiro. ‘The Great Eruption of Sakura-jima in 1914,” 
Jour. Col. Sci., Imp. Univ. Tokyo, XX XVIII (1916). Pp. 237, 
pls. 23, map 1, figs. 46. 

Sixty-eight pages of this elaborate report are devoted to petrography. The 
descriptions are accompanied by 61 photomicrographs on 8 photogravure 
plates, which are beautifully clear and of considerable help. The author states 
that “‘the characterization of the rocks is merely of preliminary qualitative 
nature.” It is to be hoped that quantitative determinations may be published 
later. The older lavas of the volcano are hypersthene-andesites, and are dull 
black or light brown porous and light in weight; the later lavas are pitch black, 
slaggy, and heavy, and are hypersthene-bearing pyroxene-andesites with spor- 
adic olivine. The lavas of 1914 are also olivine-bearing hypersthene-andesites 
and are like those of 1779. Among the inclusions in the lava are fragments 
of earlier segregations, among them a lava composed principally of anorthite 
(printed anorthosite by mistake on p. 191) with interstitial orthoclase (?). 

‘This rock is called microtinite but deserves a new name, since microtinite as 
described by Lacroix was neither a pure nor nearly pure anorthite rock. No 
quantitative data as to the relative amounts of the two feldspars are given. 
The name ceramicite is given to porcelain-like ejectamenta which contain cor- 
dierite as the characteristic component. The other constituents are basic 
plagioclase and colorless glass, with subordinate hypersthene. Quartz is said 
to be totally wanting. With 23 beautiful photogravure plates, one may over- 
look the muddy half-tones which disfigure the text. 


K6zu, S. (1) “The Dispersion Phenomena and the Influence of 
Temperature on the Optic Axial Angle of Sanidine from the 
Eifel; (2) ‘The Dispersion Phenomena of Some Monoclinic 
Feldspars,”’ Mineralog. Mag., XVII (1916), 237-52, 253-73: 

Many valuable determinations on the optic axial angles and dispersion in 


feldspars, which would be more valuable if chemical analyses of the materials 
used were given. 


PETROLOGICAL ABSTRACTS AND REVIEWS 495 


Lacroix, A. “Les roches basiques non volcaniques de Madagas- 
car,” Comptes Rendus, CLIX (1914), 417-21. 


Anabohiisite is a new name given to a “‘pyroxenite”’ with olivine, hypers- 
thene, hornblende, and 30 per cent ilmenite and magnetite. It forms the 
periphery of the gabbro massif in Anabohitsy. The amount of olivine is not 
stated. Why not a peridotite? An analysis is given. 


Lacroix, A. ‘Sur un type nouveau de roche granitique alcaline, 
renfermant une eucolite,” Comptes Rendus, CLX (1015), 
253-57- 

A rock composed of much quartz, alcalic feldspar (microperthitic soda- 
orthoclase or anorthoclase with albite, in other cases albite alone), aegirite, 
riebeckite, and eucolite, occurring either fine-, medium-, or coarse-grained, is 
given the name faszbitikite. It is compared with rockallite. A chemical but 
no modal analysis is given. 


Lacrorx, A. “Sur quelques roches volcaniques mélanocrates des 
Possessions frangaises de l’océan Indien et du Pacifique,” 
Comptes Rendus, CLXIII (1916), 177-83. 


Describes and gives analyses of two rocks which “‘are more calcic than felds- 
pathic picrites, generally a little more ferruginous, and a little richer in silica. 
They are separated from the picrites in that their pyroxene is in excess over 
olivine.’ Further the quantity of light-colored constituents is often a little 
greater.’ No mineral percentages are given, consequently it is difficult to 
place it. Possibly a labradorite (?) bearing picrite. To it is given the new 
name ankaramuite. 


Lacrorx, A. “La constitution des roches volcaniques de |’Ex- 
treme Nord de Madagascar et de Nosy bé; les ankaratrites de 


Madagascar en général,” Comptes Rendus, CLXIII (1916), 
Di OAS, 

Melanocratic nephelite-basalts with considerable feldspar and phenocrysts 
of olivine, titaniferous augite in the form of microlites, ilmenite, often perof- 
skite, and biotite are given the new name avkaratrite. The nephelite never 

forms more than ro to 15 per cent of the rock and is in many cases accompanied 
by melilite. With simply a description for a definition, its relationship to 
other rocks is indeterminable. 

A granular rock formed of augite and nephelite with a little olivine, biotite, 
apatite, and some orthoclase is called fasinite. Rosenbusch’s definition of 
bekinkinite would cover this rock, but Lacroix says the rock from: Bekinkiny 


496 PETROLOGICAL ABSTRACTS AND REVIEWS 


generally contains plagioclase, is always richer in hornblende, and has part of 
the nephelite changed to analcite. The unfortunate usage of giving locality 
names to new rock types is ever a cause for confusion. Should the original 
definition be considered the standard, or must a rock possess every accessory 
of the original rock to be of the same type, that is, should the words of the 
definition or the rock from a certain locality be the standard? If the latter, 
then practically every outcrop should have a new name, for slight variations 
will always be found. 


Lacrorx, A. ‘‘Les syénites 4 riebeckite d’Alter Pedroso (Portu- 
gal), leurs formes mésocrates (lusitanites) et leur transforma- 
tion en leptynites et en gneiss,” Comptes Rendus, CLXIII 
(1916), 279-84. 

Under /usitanite, after the name of the country where found, there is des- 


cribed a mesocratic riebeckite-aegirite-syenite. Where new names are based 
upon variations in the mineral percentages, modal percentages should be given. 


Lacrorx, A. “Les laves a haiiyne d’Auvergne et leurs enclaves 
homoeogénes,” Comptes Rendus, CLXIV (1917), 581-87. 


Ordanchite is applied to certain hauynite-tephrites from la Banne d’Or- 
danche. Phenocrysts of plagioclase (‘‘oscillent entre le labrador et |’andésine”’), 
blue hauynite, more or less corroded hornblende, and augite are visible to the 
unaided eye. Microscopically there are seen in addition titaniferous magnetite, 
titanite, apatite, and rarely olivine. Tahitite, after its occurrence on Tahiti, 
is a lava resembling ordanchite but more alcalic. Lacroix considers it a micro- 
litic form of nephelite-monzonite. It contains phenocrysts of hauynite in a 
vitreous groundmass of microlites of augite, titaniferous magnetite, hauynite, 
and perhaps a little orthoclase and leucite. Mareugite is a name given to a 
medium-grained rock from Mareuges, Auvergne, which contains 60 per cent 
light minerals (bytownite and hauynite). The dark minerals present are not 
mentioned, but in an associated rock, called type two, they are hornblende, 
augite, titaniferous magnetite, and a little blue hauynite, and therefore they 
apparently also occur in the mareugite. > 


Lacrorx, A. “Les laves leucitiques de la Somma,” Comptes Ren- 
CUS AC IDOVa TO) Ast SO. 

Vesuvites are the leucite-tephrites from Vesuvius. Four apparently new 
analyses are given. Nine analyses of ‘“‘ottajanites”’ are also given. They 
contain more plagioclase and less leucite than the vesuvianites, but since no 
proportions are given the dividing lines cannot be stated. Terms such as more, 
less, little, and much are of no value for comparison. 


. PETROLOGICAL ABSTRACTS AND REVIEWS 497 


Lacroix, A. “Les roches grenues d’un magma leucitique étudiées 
a l’aide des blocs holocristallins de la Somma,’’ Comptes Ren- 
dus, CLXV (1917), 205-11. 


Ottajanites are leucite-tephrites having the chemical but not the mineralog- 
ical composition of sommaites. They are microlitic and contain leucite and 
plagioclase. Vesuvites are the leucite-tephrites of Vesuvius. These are 
described in more detail in a later publication (see next abstract). Puglianites 
are coarse-grained rocks composed of automorphic augite in a groundmass of 
leucite and anorthite. Certain varieties contain a little biotite, hornblende, 
and orthoclase. Sebastianites are like the puglianites chemically but are 
different mineralogically. They are composed of more or less automorphic 
anorthite, with a little augite, apatite, and biotite. Leucite is absent, all 
the potash being in the biotite. Chemical analyses but no modal percentages 
are given. 


Lacrorx, A. ‘Les formes grenues du magma leucitique du volcan 
laziale,”’ Comptes Rendus, CLXV (1917), 1029-34. 


Braccianite is applied to certain leucitites from the Lake of Bracciano but 
why given a new name is not clear, except to distinguish them from the Capo 
di Bove leucitites. 


Lacrorx, A. ‘“‘Dacites et dacitoides, 4 propos des laves de la Mar- 
tinique,” Comptes Rendus, CLXVIII (1919), 297-302. 

According to Lacroix, petrographers classify as dacite only those quartz- 
bearing andesites which have phenocrysts of quartz; those having quartz in 
the groundmasses or hidden in glassy groundmasses, and therefore only shown 
chemically, he says are called andesites. The latter rocks he would call 
dacitoides. He would extend the term to cover, not only the extrusive equiv- 
alents of the quartz-diorites, but also quartz-gabbros, in that he would have 
oligoclase-, andesine-, and labradorite-dacitoides, to the further confusion of 
present nomenclature. (H. S. Washington, Amer. Jour. Sci., L [1920], 456, 
objects to these terms and uses Rosenbusch’s term hyalodacite.) 


Lacrorx, A. ‘La constitution minéralogique et chimique des laves 
des volcans du Tibesti,’’ Comptes Rendus, CLXIX (1919), 
169-75. 

In this paper basanitoides is proposed for basanites in which the nephelite 


is not crystallized but remains “potential in the glass.” Chemically they are 
basanites. 


498 PETROLOGICAL ABSTRACTS AND REVIEWS i 


LAITAKARI, AARNE. ‘“Einige Albitepidotgesteine von Siidfinn- 
land,” Bull. Comm. Geol. Finlande, No. 51,1918. Pp. 13, figs. 5. 


The term helsinkite is proposed for a haphazard-textured dike-rock, consist- 
ing of albite and epidote, and in some cases quartz (quartz-helsinkite). There 
is, in some cases, a little microcline, biotite, apatite, and iron ore. The amount 
of epidote varies from 15 to 35 per cent. In the specimen analyzed there is 
about 31 per cent epidote and 67 per cent feldspar, of which microcline forms 
less than 5 per cent (see review of Makinen), consequently it belongs, in the 
reviewer’s system, to 2112 (new form). Both albite and epidote are thought to 
be primary magmatic minerals, the CaO having formed epidote rather than 
anorthite in plagioclase on account of greater water content and lower temper- 
ature of crystallization than the surrounding granite. 


LAITAKARI, AARNE. “Uber die Petrographie und Mineralogie der 
Kalksteinlagerstatten von Parainen (Pargas),’’ Bull. Comm. 
Geol. Finlande, No. 54, 1921. Pp. 113, figs. 40, pls. 3. 


The island of Ald, in South Finland, consists principally of migmatite, a 
mixture of granite and gneiss. Within this rock are long, narrow lenses of 
limestone, calcareous gneiss, and amphibolite, which represent remnants of 
strata which were infolded in the igneous rock. Cutting all these rocks there 
are granitic and basic dikes. A short introduction of 5 pages, descriptions 
of rocks 28 pages, descriptions of minerals 57 pages, contact action 7 pages, 
mineral paragenesis and metamorphism of the limestone 5 pages, and a bibliog- 
raphy of 7 pages comprise the report. 


LarSEN, Esper S. ‘The Microscopic Determination of the Non- 
opaque Minerals, Bull. 679, U.S. Geol. Survey, Washington, 
TOD Te e204. 

In this very valuable bulletin, Mr. Larsen gives data for determining by 
their refractive indices all of the transparent minerals recognized by Dana in 
his System of Mineralogy as well as a considerable number not given by him. 
Thus, if the optical properties are known, there is afforded a rapid means for 
determining the minerals by immersion in liquid media of known indices. 
Furthermore, about 125 pages give the results of new measurements of optical 
constants of about 500 species for which data was not previously available. 
The bulletin is invaluable and should be in the hands of every petrographer. 


LarsEN, ESPER S., and Hunter, J. F. ‘“Melilite and Other 
Minerals from Gunnison County, Colorado,” Jour. Washington 
Acad. Sci., IV (1914), 473-79. 

While this paper is principally mineralogic, here is first proposed the new 
name uncompahgrite, from the Uncompahgre quadrangle, Colorado, where the 


” PETROLOGICAL ABSTRACTS AND REVIEWS 499 


rock was found. It is a coarse-grained rock made up of about two-thirds 
or more melilite, with considerable pyroxene, magnetite, perofskite, and apa- 
tite, while in places biotite, calcite, and other minerals occur. In texture it 
is commonly coarse, but varies. Cleavage pieces of melilite a foot across and 
mottled by inclusions of other constituents are not uncommon. The finest 
grained rock is hypautomorphic-granular with crystals averaging 1 mm. 

across. In the new system of the reviewer, the rock belongs to 2125. q 


LARSEN, Esper S., and MANSFIELD, GEORGE R. “‘ Nepheline 
Basalt in the Fort Hall Indian Reservation, Idaho,” Jour. 
Washington Acad. Sci., V (1915), 463-68. 


The mode of this nephelite-basalt, as given by Mr. Larsen, is nephelite 20, 
diopside 39, forsterite 26, biotite 8, magnetite 3, ilmenite 1, and apatite 1 per 
cent. In the reviewer’s system it belongs in 3125 (new form). 


LrenMann, E. “Die Ermittlung der Brechungsexponenten der 
Mineralien im Diinnschliff durch Vergleich mit Canadabalsam 
und Kollolith,” Centralbl. f. Min., etc., (1921), 102-12. 


The refractive index of Canada balsam in certain slides was found to be 
lower than the recently accepted values, being less than 1.5243. A series of 
expetiments with Kollolith showed that the refractive index depended upon the 
original solvent and upon the temperature of heating. With hard kollolith 
and a temperature of 150° to 156°, the resulting mount was found to have prac- 
tically a constant value of 1.5335, but with kollolith dissolved in xylol exposed 
to air, or under cover-glasses, the values varied from 1.519 to 1.5343, though 
the first averaged values between 1.5231 to 1.5343 and the latter 1.5257 to 
1.5298, depending upon the time of exposure to the air, the higher values being 
obtained after about six weeks. 


Lynes, HUBERT, AND SmiTH, W. CAMPBELL. ‘‘Preliminary Note 
- on the Rocks of Darfur.” Geol. Mag., LVIII (1921), 206-15. 


The author describes rocks from the volcano of Dereiba and from various 
other points between there and El Obeid, Africa. None from this area has 
previously been described. Troctolite, granite, graphic granite, sandstones, 
gneisses, and quartz-monzonite occur on the road. The rocks of the volcano 
are quartz-bearing soda-trachyte, soda-trachyte, quartz-bearing riebeckite- 
trachyte, andesine-bearing kenyte, and mugearite. 


MAxINEN, EEro. ‘‘Oversikt av de Prekambriska Bildningarna i 
Mellersta Osterbotten i Finland,” Bull. Comm. Geol. Finlande, 
PNG tp OVO ep. (52) map 1, tes. 25: 
The pre-Bothnian rocks of the district described are represented by two 
small areas of ortho-gneisses. ‘The Bothnian complex consists of plagioclase- 


500 PETROLOGICAL ABSTRACTS AND REVIEWS ‘ 


gneiss, mica-schist, conglomerate-gneiss, quartzite, and limestone, with some 
leptites. All but the latter are of sedimentary origin. Even the plagioclase- 
gneiss, which in chemical composition approaches certain igneous rocks, is 
sedimentary. The rocks have been formed, to a considerable extent, of 
unweathered volcanic material (volcanic ashes, as well as mechanically dis- 
integrated materials from tuffs and lava beds) partly assorted and deposited in 
water. Associated with these rocks are certain intrusives and extrusives, 
mainly plagioclase-porphyrites, uralite-porphyrites, and amphibolites. The 
post-Bothnian rocks form a comagmatic series, chiefly abyssal but in part 
hypabyssal in character. The rocks are granite, syenite, granodiorite, quartz- 
diorite, diorite, gabbro, diabase, hornblendite, and peridotite. In the south- 
western part of the area there were apparently two periods of intrusion from a 
magma differentiated im situ. In the parish of Haapavesi to Kivijarvi, there 
is no evidence of two periods of intrusion but the rocks, from granite to grano- 
diorite and gabbro, grade into each other. The northern and eastern part of 
the area is an area of migmatites. Here the magma has been differentiated to 
a certain extent, but the different rocks have not been separated to form large 
homogeneous masses. The older rocks have undergone fairly complete assimi- 
lation by the magma. In the reviewer’s modified classification the ortho-gneiss 
is 227’ (a granodiorite-gneiss), the plagioclase-porphyrite is 228 (quartz-diorite 
porphyry or tonalite-porphyry), four granodiorites are all 227’ (granodiorites 
in the broad sense, but quartz-monzonites if this group is included in the 
classification), twelve microcline-quartz-diorites are 227’ (monzotonalites nar- 
row, or granodiorites in the broad sense), three tonalites are 228 (tonalite), a 
tonalite-gneiss is 227’ (monzotonalite-gneiss), three quartz-diorites are 228 
(tonalite), a quartz-diorite is 328 (mela-tonalite), a diorite is 227’ (monzo- 
tonalite), two biotite-granites are 227’’ (adamellites in the narrow, or grano- 
diorites in the broad sense), an olivine-gabbro is 2312, and two soda-syenites are 
2111’ (an albite-monzodiorite in the narrow, or albite-syenodiorite in the broad 
sense. The latter is the rock to which Laitakari compares his albite-epidote 
rock which he calls helsinkite. Laitakari’s rock is given above in the reviews 
as 2112, from definition, but he says it may in some cases have a little micro- 
cline. The present rocks contain enough microcline,one 11 and one 7 per cent, 
to throw them into the next family.) 


REVIEWS 


Preliminary Report on the Deposits of Manganese Ore in the Batesville 
District, Arkansas. By Hucu D. Miser. Bulletin 715-G, 
United States Geological Survey, Government Printing Office, 
Washington, D.C., 1920. Pp. 93-124 (32), pls. 3, figs. 4, 
bibliography, tables, and analyses. 

The Batesville manganese district is in the southern part of the 
Ozark region, in Independence, Sharp, Izard, and Stone counties, in 
north-central Arkansas. The deposits have been worked intermittently 
since 1849. ‘They lie in a region of rough topography but of no great 
relief. In the manganese-bearing areas the following formations are 
exposed: 


Age Formation 


IMASSISSIPPlAN ayes. es 8s, Boone shert 


Cason shale 

Fernvale limestone 
Kimmswick limestone 
Plattin limestone 
Joachim limestone 
St. Peter sandstone 


Ordovician 


The Cherty Fernvale limestone is the principal source of manganese 
ore. Its weathering leaves cherty nodules and sticky residual clay, 
varying in color from yellow to red. The overlying thin Cason shale 
bears phosphate, here as pebbles almost an inch in diameter, there as 
shell fragments or grains; such phosphate deposits have, however, not 
been extensively worked. The only fossils in the formation are flattened 
“buttons” of supposedly algal origin, composed of calcium or manganese 
carbonates; the “buttons”? may occur in great quantity in the residual 
clay, as at the Cason mine. 

Structurally the beds are almost flat-lying; a general doming of the 
region has given them a gentle dip southward, upon which are super- 
imposed several minor flexures. There are seven small normal faults, 
with a throw not exceeding 400 feet. An uncomformity separates Mis- 
sissippian and Ordovician beds and four others occur in the Ordovician 
sequence. 


501 


502 REVIEWS 


The manganese minerals are psilomelane, hausmannite, braunite, 
manganite, pyrolusite; and wad. Psilomelane is most abundant. 
Tron oxides (limonite and hematite), and ferruginous manganese ores 
(hematite and limonite with psilomelane), barite, quartz, calcite, 
arsenopyrite, and sulphides of the heavy metals are also associated . 
with the ores. The workable manganese and ferruginous manganese 
deposits occur under the following conditions: (1) replacement de- 
posits in the Cason shale and ics residual clay; (2) replacement 
deposits in the Fernvale limestone; (3) residual deposits from the 
Fernvale limestone; (4) replacement deposits in clays; (5) transported 
stream-gravel deposits. Of these, (3) is most important as a source of 
manganese ores, and (1) has furnished more ferruginous manganese. The 
replacement deposits in the Cason shale and its residuum occur in 
irregular masses, ‘‘buttons,”’ or horizontal seams and beds; ‘‘buttons” 
of red iton oxide are also found under similar conditions. The occurrence 
of manganese-bearing calcite suggests that all the manganese oxides 
were derived from the carbonate. A similar origin for the manganese 
replacements in the Fernvale limestone is well demonstrated by deep 
cuttings in the district: here cores of carbonate are found surrounded 
by envelopes of oxides of manganese and ferruginous manganese. 

The largest yield of manganese ore, the largest reserves, and a 
considerable part of the low-grade ore output comes from the residual 
deposits of the Fernvale or lower limestones, into the deeper portions of 
which the nodules were settled by gravity or were washed by streams. 
The decomposition of the limestone has formed manganese-bearing 
surface hollows and channels; elsewhere manganese-bearing caves and 
sinks have developed. Slumping and sinking of the soft, plastic clay 
(and of the overlying Cason shale) have greatly disturbed the ore bodies. 
These residual deposits consist of psilomelane, hausmannite, wad, and, 
subordinately, braunite. The hard oxides may occur as bowlders that 
weigh as much as 22 tons; generally the coarser the masses, the more free 
the ore from iron and the higher its grade. 

Some small deposits of manganese oxides have been formed by the 
introduction of manganese into the clays through the action of ground 
waters; such manganese probably came from the Cason shale or Fernvale 
limestone. ‘These ores are low grade and diluted though the presence 
of silica, iron, and alumina. 

Finally some manganese is obtained from alluvial cones and gravel 
bars, the deposits being composed of compact masses of manganese oxides, 
and of small pebbles of oxide of iron. Many of the largest concentrations 


REVIEWS 503 


lie in synclines, for reasons which the writer hopes to demonstrate in a 
later paper. 

Most of the high-grade ores bear from 45 to 52 per cent of manganese, 
but many otherwise good deposits bear too much phosphorous to justify 
exploiting them. “Hand-picked” ore is the highest grade. The ore is 
not used for chemical purposes, but is employed in various high- 
manganese iron and steel products; it is also of importance in making 
brown, gray, and speckled bricks, when mixed with clay. 

The manganese reserves probably amount to 250,000 tons of 40 
per cent manganese. The deposits are covered to the south by younger 
formations, and could probably not be extensively worked in that 
direction anyway, since concentration and oxidation have not been 


extensive under the heavy capping. 
©. Ho Ba JR: 


Magnesite Deposits of Grenville District, Argenteutl County, Quebec. 
By M. E. Witson. Memoir 98, Canadian Geological Survey, 
Ottawa, 1917. Pp. 88, figs. 2, pls. 11, maps 3. 

This district is bordered by the Ottawa River on the south and is 
about halfway between Ottawa and Montreal. The magnesite deposits 
are about ten miles north of the Ottawa River. 

Chapter i gives information of general interest about magnesite, 
its uses, foreign source of supply, other Canadian magnesite deposits, 
and the history of magnesite mining in Grenville district. 

Chapter ii is a brief statement of the geology of the district. The 
oldest rocks belong to the Grenville sedimentary series and are intruded 
by pyroxene-rich gabbro, diorite,and syenite belonging to the Buckingham 
series. These two series are intruded by batholithic masses of granite- 
syenite gneiss. All these rocks are intensely metamorphosed and are Early 
pre-Cambrian in age. These Early pre-Cambrian rocks are intruded by 
diabase dikes and a stocklike mass of granite-syenite probably of Late pre- 
Cambrian age. The Paleozoic is represented by the Potsdam, Beekman- 
toan and Chazy formations named in ascending order. Glacial bowlder 
clay and gravel and Champlain marine clay form an irregular mantle 
over the bed-rock surface. 

Chapter iii gives a description of the magnesite deposits and their 
origin. The magnesite is associated with serpentine, dolomite and other 
minerals in the metamorphosed Grenville sediments and close to outcrops 
of the pyroxenic rocks of the Buckingham series. The deposits are 
lens-shaped and the material is banded, the banding being due to 


504 REVIEWS 


differences in color of the magnesite or from variations of amounts of 
serpentine and other minerals present. The strike of the bands and 
lenses is parallel to that of the Grenville sediments. The deposits 
have been intensely faulted and crumpled and probably the lenticular 
structure is the result of deformation. The mode of occurrence of the 
following minerals associated with the deposits is described: magnesite, 
serpentine, dolomite, diopside, phlogopite, quartz, talc, pyrite, sphalerite, 
magnetite, and graphite. 

The three methods of origin for magnesite deposits are: (1) deposits 
formed by the decomposition of serpentine, (2) sedimentary deposits, 
(3) deposits formed by the replacement of limestone. The Grenville 
deposits are thought to have been formed by the replacement of lime- 
stone. Silication of limestone to diopside and phlogopite is very 
common along the contacts of limestone and igneous rocks in this region 
and the igneous rocks are very close to these particular deposits. The 
writer summarizes the method of origin: ‘‘The probable order of events 
by which the magnesite deposits of the Grenville district were formed 
was as follows: (1) silication of the limestone to diopside and the forma- 
tion of phlogopite in places, (2) formation of serpentine in places, (3) 
replacement of limestone by dolomite, (4) replacement of dolomite by 
magnesite, and (5) the alteration of diopside to serpentine.” 

Chapter iv is a detailed description of the properties and gives 
tabulated descriptions of many magnesite samples with the percentage 
of CaO. While dolomite and magnesite are very intimately inter- 
mingled, yet by 1916 development work had proved the presence of 
686,900 tons of magnesite containing less than 12 per cent CaO and 
483,700 tons containing over 12 per cent CaO. 

Map 1680 issued in 1919 shows in detail the geology of a portion of 


the township surrounding the deposits. 
J. F.. W. 


Pleistocene Marine Submergence of the Hudson, Champlain and 
St. Lawrence Valleys. By HERMAN L. FatrcHiItp. New York 
State Museum Bulletins, Nos. 209, 210, Albany, N.Y., 19109. 
Bp 70; plsa2ge 

This is the closing paper by Professor Fairchild on the glacial] and 
post-glacial waters of New York State and in it he discusses the proof 
and extent of the marine submergence following the retreat of Wisconsin 
glacial ice from northern New York State. The stratified clay and 


REVIEWS 505 


sand in these valleys, the cobble bars, the wave-cut terraces, the deltas, 
and many other evidences of high-level standing water, with no known 
barriers to hold this water in, is strong evidence that the land in this 
region once stood below sea-level. This marine shore line has been 
uplifted and tilted and is now less than 1oo feet above sea-level a short 
distance north of New York City and 740 feet above sea-level at Covey 
Hill on the International boundary, a distance of about 350 miles. 
Diagrams are given to show the profile of this tilted marine shore line 
and also the shore line of Lake Iroquois, the last of the glacial lakes to 
occupy the Ontario basin. In the St. Lawrence-Ontario basin the 
Iroquois plane is 290 feet above the marine plane and thus when one is 
found in the field it is easy to locate the position of the other. Also 
knowing the present elevation of these two planes and the total amount of 
uplift of the region, the amount of either glacial or post-glacial uplift 
can easily be determined. From numerous measurements and calcula- 
tions of this sort it appears that northern New York State was not raised 
as a rigid body but by a progressive wave movement, as the southern 
side of Iroquois basin received one-half its total uplift during Iroquois 
time while the northern end of the same basin received very little 
uplift until after Iroquois time. The uplift of the land seems to 
have been wavelike and to have followed the margin of the retreating 
ice front. a 

Detailed descriptions of shore features in the various sheets along 
the Hudson, Champlain, and St. Lawrence valleys and the Ontario 
basin are given. The shore features at Covey Hill, the point of junction 
of the Champlain and St. Lawrence valleys and of the Champlain marine 
waters and the Lake Iroquois waters, are described in detail. Some of 
these shore features are at present somewhat above what the level of the 
water should have been at these particular localities. Many complica- 
tions probably enter the Pleistocene history as there may have been 
many up-and-down land movements and the present height of the 
summit plane above the sea must represent only the excess of land 
uplift over the rise of the ocean surface and the arithmetical sum of 
all the up-and-down movements. 

A large number of photographs are inserted to show summit shore 
features and at the end of the report a classified bibliography is given. 
This report summarizes in a very thorough and clear manner Professor 
Fairchild’s interpretation of the various glacial and post-glacial deposits 
and physiographic features of this region. 

Va 18 ie 


506 REVIEWS 


Iron Depositing Bacteria and Their Geological Relations. By 
EpmunpD Crcit Harper. United States Geological Survey, 
Professional Paper 113, 1919, Government Printing Office, 
Washington. Pp. 85, pls. 12, figs. 13. 


Geology, probably more than any other science, occupies an apical 
position in the pyramid of the natural sciences; its function is not so 
much to enunciate the more fundamental theories not based on the 
concepts of other branches of knowledge, as to weld together the contri- 
butions of its sister-sciences and apply them to its own purposes. From 
this point of view a work such as that under review is especially illumi- 
nating; it demonstrates conclusively the hitherto only partially appreci- 
ated breadth of scope of the ‘‘science of rocks and minerals.”’ 

The paper begins with a careful description of the living iron- 
depositing bacteria, both those of the higher and those of the lower type. 
Not only the more common forms, such as Leptothrix, Galionella, and 
Spirophyllum are mentioned, but a brief review of the classification, 
morphology, and physiology of essentially all relevant forms known to 
date is given. The iron-bearing algae also are named. From the early 
work of Cohn and Zopf, who thought iron accumulation in bacterial 
sheaths essentially a mechanical process, through that of James Campbell 
Brown, who considered the deposition of ferric hydroxide to be merely 
incidental to the extraction of the organic constituents in the water 
affected, to the studies of Lieske, which demonstrated conclusively that 
the carbonate radicle of ferrous carbonate is extracted by the bacteria, 
leaving the insoluble hydroxide, the increasing importance of bacterial 
metabolism has come to be recognized. It is probable, in fact, that 
some bacteria require ferrous bicarbonate; that others can use it inter- 
changeably with other soluble iron compounds; while still others, 
notably the lower groups, can use only the organic salts of iron. 

Mr. Harder himself prepared and studied cultures of various forms. 
Crenothrix was obtained from city water of Madison, Wisconsin, which 
contained large amounts of magnesium and calcium carbonates. Cul- 
tures of Leptothrix ochracea were grown from the water of a chalybeate 
spring near Madison, the water bearing much ferrous iron in solution, 
probably as the bicarbonate; this form was also brought up from a low 
level of a Cuyuna District mine, where ferric hydroxide is being pre- 
cipitated in large amounts. Cultures of Galionella ferruginea from the 
Federal Mine of Wisconsin and from the Kennedy Mine of central 
Minnesota are also reported; in both of these localities a brown, gelat- 
inous scum occurs on the walls of the tunnels and in little pools in the 
drift-floors. Spirophyllum ferrugineum appears in the waters of the 


REVIEWS 507 


Wisconsin zinc mines; it was also found in other mine-waters, probably 
being carried downward by surface waters that descended through the 
soil and rock for many hundreds of feet. The form is especially 
abundant in the water of Vermilion Lake, Soudan, Minnesota. All 
these bacteria precipitated ferric hydroxide. 

Besides these there are other iron bacteria which precipitate ferric 
hydroxide or ferrous sulphide. The former group is especially wide- 
spread, and though its members do not require iron in solution for their 
development, still it is thrown down quite rapidly as a waste product. 
The precipitating action of such forms was studied by means of weak 
solutions of the slightly acid ferrous ammonium sulphate. 

Whereas some of the sulphide-precipitating bacteria owe this ability 
to the action of the hydrogen sulphide liberated by them on the ferric 
salt in solution, others precipitate the sulphide because of their reducing 
action on the sulphate. When water bearing ferric ammonium citrate 
was inoculated with hay and soil infusions, there resulted a precipitation 
of ferric hydroxide not observed in the sterile solution. This ensues even 
under anerobic conditions. ‘The organisms that induced it were grown on 
plates of Heyden Naehrstoff agar with ferric ammonium citrate as indi- 
cator and various types of bacteria were recognized. Slopes of Heyden 
agar to which no citrate was added showed practically no growth. 
Other organic salts of iron were also used more or less successfully, and 
similar experiments showed that no precipitations resulted from the 
solutions bearing salts of manganese in place of iron. 

On the other hand, no precipitation that could be definitely assigned 
to organic processes could be obtained from inorganic iron salts such as the 
bivalent carbonate or sulphate or ferric chloride, the precipitation that 
did result in these cases being better attributed to the purely inorganic 
oxidation and (or) hydrolysis. These results do not agree with those of 
Mumford (Chem. Soc. Jour., Vol. 103, 1913). 

A review of the earlier work of Beijerinck, Van Delden, and Fred 
on the formation of hydrogen sulphide by bacteria may be summarized 
as follows: sulphates are formed abundantly by sulphur bacteria from 
hydrogen sulphide; these sulphates are then reduced by other bacteria 
to yield sulphides and hydrogen sulphide; if ferrous or manganese 
sulphides are formed they are precipitated. These observations have 
been ably and fully discussed in other papers, notably in their bearing on 
the origin of the Sicilian sulphur deposits. 

The relations of these facts to geologic processes are amplified by 
the writer. He points out the solubility of iron carbonate; soluble 
organic compounds, chiefly humides, of iron may also be formed; it is 


508 REVIEWS 


these that lend the dark color to streams flowing through regions of 
abundant vegetation. Finally, a smaller amount of iron may be carried 
as sulphate. 

Tron deposits occur in the form of hydroxides and oxides, carbonates, 
and silicates. Iron is readily precipitated as the hydroxide, when a solu- 
tion containing an excess of carbon dioxide is induced, through changes in 
pressure or temperature, to give up its gas with a concomitant saturation 
with oxygen; if the solution merely loses its carbon dioxide without 
undue oxidation, ferrous carbonate tends to be precipitated. These 
several iron precipitates generally accumulate in bogs, lakes, or lagoons. 
Elsewhere rapidly flowing waters may (rarely) form iron deposits. 

Most deposits of bedded hematite, bog ore, and brown iron ore were 
laid down originally as hydroxides, chiefly through biologic, but also in 
part through chemical, agencies. Iron sulphide, too, may be formed 
in either way, but ferrous carbonates and silicates are not definitely 
traceable to bacterial action. Iron phosphates and basic ferric sulphates 
are chemical precipitates. Ferrous silicates tend to be deposited where 
alkaline silicates are abundant in regions of precipitation of ferrous 
carbonates—formed as indicated above. Probably gc per cent of all 
the iron ores being worked today are of the sedimentary type. A list 
and description of those thought to be originally laid down as ferric 
hydroxide includes the Clinton ores, the Wabana ores, the Lake Superior 
hematite-chamosites, the itabirites of Minas Garaes, (Brazil) the 
hematite-magnetites of the Dharwar terrain of India, and the ores of 
the Norwegian Lias. Bog iron ores, too, were probably deposited in the 
same form and are now widely distributed, being especially abundant in 
the eastern part of Canada and the United States, in Sweden, and in the 
glaciated sections of Europe and Asia; ferric hydroxide is also present 
in large amounts in the red mud off the coast of Brazil. 

Another type of sedimentary iron ore is that originally deposited 
as the carbonate; this is represented by the “black band” ores in Ohio, 
Pennsylvania, and West Virginia, the odlitic siderites of eastern England, 
and the siderites of the Lake Superior region. 

The silicates present a third type of sedimentary iron ore. The 
forms in which the iron occurs include glauconite, bavalite, thueringite, 
bertherine, and chamosite; of these glauconite is the most widespread. 
Greenalite was probably the original constituent of the Mesabi ores, 
while the chamosite ores are predominant in the Bohemian Brdagebirge. 

Ferric and ferrous sulphides represent the fourth type of iron deposits. 
Pyrite is important in the Huelva region of southern Spain, in the 


REVIEWS 509 


Carpathian and Harz Mountains, and in Westphalia; the sedimentary 
origin of some of these beds, however, is still in doubt; more typically 
sedimentary are some of the odlitic pyrite beds of Wabana. Ferrous 
sulphide is common in the limans of the Black Sea. The general expla- 
nation offered for these deposits is that of Doss, who believed that the iron 
is carried as a carbonate, upon which bacteria may act directly or 
indirectly to yield colloidal ferrous sulphide or ferric hydroxide, which 
in turn is converted into ferrous sulphide. 

Various facts of importance to students of sedimentation are brought 
out in a discussion of the origin of the separate iron deposits; for instance, 
the experiments by Spring and Ruff on the conditions favoring and 
opposing the derivation of limonite from ferric hydroxide. It is shown, 
also, that primary deposits of ferric hydroxide may be readily altered 
to the carbonate, especially in the presence of large amounts of organic 
matter. 

From a discussion of the inorganic causes for the precipitation of 
iron compounds as sediments, the writer returns to a consideration of 
organic causes. Here, especially in connection with the precipitation 
of ferric hydroxides, the line is not readily drawn between oxidation, 
taken as the purely inorganic process, and bacterial action; the latter, 
however, is surely of much importance, whether actually preponderant 
or not. The conditions under which iron-bearing bacteria are active 
vary between wide limits. Temperature, if too high or too low, may be 
inhibitory, as may also be a reduction in the amount of organic matter 
present. 

To present in detail all the significant facts of this interesting paper, 
would necessitate a review as long as the original publication; it must 
be read to be fully appreciated. It displays both the clear thought and 
the technical skill of Mr. Harder in original work and his ability to join 
his observations to those of other investigators of the subject. All in 
all, it is a most admirable contribution to experimental geology. 


C.H.B., Jr. 


Helium-Bearing NaturalGas. By G.SHERBURNE ROGERS. United 
States Geological Survey, Professional Paper 121. Washing- 
ton, 1921. 

This report comes from the press many months after the accidental 
death of its author in South America. It reflects so zealous a spirit of 
scientific research in the public service that it stirs anew the sorrow of 


510 REVIEWS 


his former associates at the passing of a geologist of such brilliancy and 
promise. The report perpetuates one of Mr. Rogers’ contributions to 
the war needs of his country, namely the investigation of the occurrences 
of helium-bearing natural gas in the United States with a view to its 
utilization in dirigibles and military balloons. 

Helium differs from hydrogen, the gas commonly used in balloons, 
in being non-inflammable; on the other hand it is about twice as heavy as 
hydrogen though still so much lighter than air that its lifting-power in 
air is 93 per cent that of hydrogen. Its advantages are so great that 
its use, both commercial and military, in airships is likely to be limited 
solely by the supply of helium available and the expense at which it can 
be produced. 

The main body of this report is devoted to a description of the field 
occurrence of natural gases containing helium within the limits of the 
United States, but a brief discussion of possible sources of helium in 
other countries is included, and the report summarizes also, in very 
concrete and admirable fashion, the history of helium, its properties, 
and its relations to the radioactive elements, as well as the methods 
which have been applied for separating it from the commoner constituents 
of natural gases. The field work upon which the investigation rests 
consisted chiefly in collecting samples of gas for analysis and in gathering 
data regarding depth, geologic position, rock pressure, and volume of 
the gas sampled. This work involved careful examination of the 
geological structure in one or two areas, though a number of helium-rich 
districts had been previously studied by Survey geologists, and data on 
others were furnished by oil companies. 

As a result of the investigations it is shown that although most of 
the natural gas produced in the eastern and central parts of the United 
States contains at least a trace of helium, gas containing more than 4 
per cent is known to occur only in two areas, one in northern Texas and 
the other in southern Kansas and northern Oklahoma. The helium-rich 
gas of the Kansas-Oklahoma area is confined to strata of Middle and 
Upper Pennsylvanian age, though gas carrying almost 3 per cent of 
helium occurs in the Lower Pennsylvanian. The Mississippian and 
Permian gases in that locality are poor in helium. Conditions in the 
Texas area are almost identical. In Ohio gas carrying 3 per cent of 
helium occurs in the Lower Mississippian and in the Clinton of the 
Silurian. Nearly all samples of Cretaceous gas from various parts of 
the United States show only traces of helium and most samples of 
Tertiary gas contain none. 


REVIEWS 511 


Although traces of helium occur in most natural gas, noteworthy 
proportions have been found only in gas rich in nitrogen. The per- 
centage in helium, moreover, seems to depend, in a measure, on the 
percentage of nitrogen, though there is not direct proportionality 
between the two. Some of the Kansas gases contain about 85 per cent 
of nitrogen and 2 per cent of helium. 

After a discussion of the occurrence of helium in minerals and rocks, 
in mine gases and in the gases of mineral springs, volcanoes and fumeroles, 
the theories that have been advanced in explanation of the origin of the 
helium in natural gases are discussed. While recognizing that the origin 
of the helium is still a matter of great uncertainty, the writer is inclined 
to favor the view that the helium is derived from deposits of uranium 
and thorium, probably disseminated through the strata not far beneath 
the horizons at which the helium gas occurs. 

E. S. BASTIN 


The Economic Aspects of Geology. By C. K. LerrH. New York: 
Henry Holt and Company, 1921. 


Because each year modern industry is becoming more technical, 
industrial progress has come to stride pace after pace with the develop- 
ment of science. Geology has shared with other sciences in the tighten- 
ing of the bonds between science and industry. It is no longer necessary, 
as it was a generation ago, for the geologist to be continually bringing 
before the public the practical potentialities of his science; certain 
industries are now even snatching the half-fledged geologists from their 
academic nests. ‘The increasing industrial importance of geology lays a 
new responsibility upon those engaged in the training of economic 
geologists, and Professor Leith’s book, an embodiment of lectures given 
at the University of Wisconsin, drives home to the geology student in 
vigorous fashion not only the fundamental facts of useful mineral 
occurrence but also the réle which each of these minerals plays in the 
economic life of the nation and of the world. ‘The book is, in fact, an 
outgrowth of its author’s war-time experiences during which he dis- 
tinguished himself as a leader in the first real inventory of her mineral 
resources that the United States had ever taken. 

The work is adapted to the use of students having an elementary 
knowledge of geology, such, for example, as is commonly acquired in the 
first year’s work in college geology. While adapted for the use of 


512 REVIEWS 


students, most mining engineers and teachers of geology will find that 
it gives them not only a new array of facts, but a broadened viewpoint. 

After introductory material, a chapter is devoted to the common 
elements, minerals and rocks of the earth and their origin, and another 
chapter to a simple exposition of the processes by which mineral deposits 
are formed. 

In the consideration of individual mineral resources, the conventional 
classification as metallic and non-metallic is abandoned in favor of one 
that is more expressive of their rdle in industrial life. Early in the 
scheme come the fertilizer group of minerals, nitrates, potash, phosphates, 
pyrite, and sulphur. Then come the fundamental fuels, coal, oil, and 
gas. In the same chapter with iron and steel are treated the ferro- 
alloying minerals and the minerals fluorite, magnesite, and silica, that 
find their main use as fluxes or refractories in the iron and steel industry. 
Other major base metals are followed by the precious metals, and then 
by minor metals. The commodity chapters are concluded with one 
devoted to miscellaneous non-metallic minerals. 

A feature of the treatment of individual commodities that makes 
for clarity is the treatment of ‘‘Economic Features” and “Geologic 
Features”’ under separate heads. 

Conservation of mineral resources, particularly coal, comes in for 
separate treatment in a chapter near the end of the volume. Especially 
noteworthy are several topics that have received scant treatment, if 
any, in previous books on economic geology; namely, exploration and 
development of mineral deposits, the réJe of the geologist in their valua- 
tion and taxation, and a brief discussion of the legal aspects of geology. 

The book concludes with chapters summarizing the international 
aspects of mineral resources, on “‘ Geology at the Front” and “Geology 
Behind the Front.” 

The field occupied by this book is covered by no other, and few books 
are likely to fill as useful a place in geologic education. 

Epson S. BASTIN 


-MASTERKEY TO ALL LANCUACES 


, ech or hoe is Boras Morris PIERCE, 16 languages, $1.94 postpaid, each 


“Arabic, ‘Chinese, Danish, Dutch, English, Scorch English, French, German, Italian, 
anjabi, Polish, Portuguese, Russian, Spanish, Swedish. 


HES Leh t 


Price 96 (Is 6) a ume 1915-1 Prix 60¢ 


~ ¢ TEXTES PHONIQUES 


| 
lauusiigue pense | 
| 
| 


La Maison 


que 


Pierre a batie 


‘Trasiscnr? bu TEXTE NoMiOUE OniCIMAL FAX * 


"Professor of Romance Languages in Harvard University 


Somes. Matalog poutireh 


Lancoaces PustisiNc Company 
143 Warr A7ra Sracer, New-Your 


> 


TS, by Ropert Morris Pierce: four, each measuring Goaner: size) 24 by 15 inches, and 
‘respectively: Chart of the Universal Alphagam, Chart of the Organs of Speech, Chart of 
h ee. oe) of the Spanish vee folded and bound in paper.6$ ee Ag inches, 


f ae languages, edited by ROBERT Morris PIERCE, 30 cents eee each language: 
nal Albanian, Old Armenian, Bearnese (Arrens), Beumcte (Garlin), Bohemian, Bragancian, 
hui, Burgundian, Cantonese, Catalan, Charentese, Cumbrian, Dauphinese, Dubian (Bourmois), 
ontarlier), Dutch, Emilian (Badi), Heian (Bolowaa). Emilian (Novellara), Tuscan 
(Carrara), Tuscan Emilian (Gombitelli), Tuscan Emilian (Sassalbo), English, American 
lish, British English, Scotch English, Middle English (Orm), Middle English (southern dialect, 

13 ‘ury), Middle English (East Midland, time of Chaucer), Old English, Finnish, French, West 
Galician, German, wogere Suess Aeginian Modern Greek, Cycladic Moder. Cees Lesbic 


es ara: Kanauri, Kui, Kurdish, Pahelt Tanase Latin, Laz, Tigeion, 
|, Lorrainese, Marchesian, Mingrelian, Molisian, Nomnan, Norwegian, Pahari, Picard, 
Se, Provengal, Old Provencal, Rhetian, Rumanian, Sanskrit, Scotch, Spanish, Ween \ 
osnian Judeo-Spanish, Bulgarian Judeo-Spanish, Mexican Seige Tridentan, Umbrian 

0) Umbrian (Rieti), Venetian (Grado), Venetian (Murano), Venetian (Triest), Walloon — 
Halipon (Stavelot), W Welsh, Yorkshire. 


8 West 40th Street, New York City. 


A Living Monument to 


AGASSIZ 


A Joy ey) Benefit to Students 
and Lovers of Nature 


Please co-operate in the great work of 


THE AGASSIZ ASSOCIATION, Inc. 


Forty-seven years of effective service 
Science, Education and Humanity 


The Guide to Nature 


Monthly Illustrated Magazine 
EDWARD F. BIGELOW, Editor 


. Subscription, $1.50 per year; single copy, 15 cents; 


three months’ irtal, 25 cents ceo eye aos post ° 


in quality and pons 


Please write for particulars 


ARCADIA 
SOUND BEACH CONNECTICUT 


(incorporated) a 
374 Broadway - 
Branches and Dealers E: ; 


Rock-Forming Minerals aid Rocks : 
By ALBERT JOHANNSEN = : 


Professor of Petrology on 
The University of Chicago 


A laboratory manual designed to provide students of petrology and geologi 
interested in petrology with information on methods of determining microscopicall 
the points of separation between similar minerals and their modes of occurret 

This valuable volume contains, in addition to a folding alphabetical lis ‘sl 
placed that reference may be made to it without turning pages, the followin 
important. features: “A Quantitative Mineralogical Classification of Igneou: 
Rocks’’; “A Summary of Petrographic Methods’; “Modes of Occurrence of Various ” 
Minerals” ; “Rules for Computing Rocks from Their Moen and “Names TO- | 
posed for Various Families.”’ a 

Cloth; $2.00, postpaid $2.10 


CHICAGO 3 


_ SUPPLEMENT TO NUMBER 6 


3EOLOGY 


ott “ 
gi 


‘With the Active Collaboration of 


WELLER, Invertebrate Paleontology ALBERT JOHANNSEN, Petrology 
BASTIN, Economic Geology 4 ROLLIN T. CHAMBERLIN, Dynamic Geology 


ASSOCIATE EDITORS 


: GEIKIE, Great Britain " *JOHN C. BRANNER, Leland Stanford Junior University | 
RICHARD A. F. PENROSE, Jr., Philadelphia, Pa. 
WILLIAM H. HOBBS, University of Michigan 
FRANK D. ADAMS, McGill University 

Nee sale CHARLES K. LEITH, University of Wisconsin 
ORTH :DAVID, Australia WALLACE W. ATWOOD, Clark University - 
TLLIS, Leland Stanford Junior University WILLIAM H. EMMONS, University of Minnesota 

ALCOTT, Smithsonian Institution ARTHUR L. DAY, Carnegie Institution 


‘ 


SUPPLEMENT ere | 
_ AUGUST- SEPTEMBER ‘1922 
oF IN CLUSIONS IN IGNEOUS MAGMAS - - - - N.L. Bowen 
ease! = - - - - - - - - - Briain sins ed 
Bot emen st wn ier caren JA ee = = 2 = 5 5 - 514 
: Eis: OF - ener PRS pre HON AUR Gh aaay nce dere Nee aiken: CI 
xcTS BETWEEN ‘‘INCLUSIONS” _AND Lrquis: IN INVESTIGATED SYSTEMS - = §23 
TREN EDRs GERRGR Sle gsc eer ale Beemer ty ph OL wre PR Sa Ne I hy oe 
m r = Tees, % Te Sh 
= = = ; i ek pate 
3 = z 5 a ; eesyskce 
Bi a a ay ee he ee tae are SS 
F Graxime eee ON TECListons'o OF SEDIMENTARY Onion eeu ee - - 560 
- - - = = ap Ars = HBOO 
a ae ES i ae eae 


CHICAGO, ILLINOIS, U.S. A. 


ie é THE CAMBRIDGE UNIVERSITY PRESS, Lonpon 
4 THE MARUZEN-KABUSHIKI-KAISHA, Toxyo, Osaka, Kyoto, Fukuoka, SENDAI 
i THE MISSION BOOK COMPANY, SHANGHAI 


August-September 1922 


THE JOURNAL C foi, 


Supplement. 'to August: ienteminee 1922, ; 


a 


ary 


EDITED BY ; A ie 


STUART WELLER 
Invertebrate Paiieleee 


EDSON S. BASTIN aa . 5 ROLLIN T. . CHAME 7 
+ Economic Geology A Dif ey A 
* Deceased. be 


The Journal of Geology is published by the University of Chicago at the Uni 

5750 Ellis Avenue, Chicago, Illinois, semi-quarterly, on or about the following dates: ry 

May 1, Juhe 15, August 1, September rs, November 1, December 15. J The subs ription 

per year; the price of as copies is 65 cents. Orders for service of less than a half- 

at the single-copy rate. { Postage is prepaid by the publishers on all orders. fro 

Mexico, Cuba, Porto Rico, Panama Canal Zone, Republic of Panama, Dominican 
Bolivia, Colombia, Honduras, Nicaragua, Peru, Hawaiian Islands, Philippine Isl: ads 

‘Islands, Shanghai. {| Postage is charged extra as follows: for Canada, 25 cents on an 
(total $4.25); on single copies 3 cents (total 68 cents); for - all other countries in 


requested to make all remittances payable to The ye of Chicago Press i in n postal (Or 
orders or bank drafts. © he. Mere e 
The following are authorized to quote the prices indicated 264)" St yaaa re 

For the British Empire: The Cambridge University Press, Fetter Lane, London, 
subscriptions, including postage, £1 ss. each; single copies, including postage, 4s. _ each. 
For Japan and Korea: The Maruzen-Kabushiki-Kaisha, 11 to 16 Nihonbashi To 
Tokyo, Japan. Yearly subscriptions, including postage, Yen 9.45 each; single co} es, 
postage, Yen 1.50 each. : Hoe 
For China: The Mission Book Company, 13 North ascohten Road, ‘Shanghai ( 
scriptions $4.00; single copies 65 cents, or their equivalents in Chinese money. — Po age 
mailed direct outside of Shanghai, on yearly subscriptions 50 cents, on single copies 6 
Claims for missing numbers should be made within the month following the regular mo 
cation. The publishers expect to supply missing numbers free only when losses have been 
transit, and when the reserve stock will permit. att 
Business correspondence should be addressed to The University of Chicago Press cl 


NAL oF GEoLocy, the University of Chicago, Chicago, Ill. 
Contributors are requested to write scientific and proper names with cartes care an 

to follow the form shown in the pages of the Journal. $7: 

Because of the high cost of printing, this Journal has found it necessary to adopt the f ilo 


"rates of the University Press. {[ 2. The cost of illustrative material, in excess of * ° 
article, beyond the cost of type which would have occupied the same space, shall be borne, 
{| 3. The cost of corrections in proof in excess of 10 per cent of the cost of first compo 


Entered as second-class matter, March 20, 1893, at the Post office at Chicago, Ill., iiaee the Hee 
Tal ances Diane for mailing at special rate of postage provided for in Section I103, Act of Octobe 
'y I5, 191 


PRINTED IN THE U.S.A. 


VOLUME XXX SUPPLEMENT TO NUMBER 6 


Weve 


JOURNAL OF GEOLOGY 


August-September 1922 


SUPPLEMENT 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 


N. L. BOWEN 
Geophysical Laboratory, Carnegie Institution of Washington 


INTRODUCTION 


Many igneous rocks contain inclusions of foreign material and 
not infrequently these inclusions show evidence of having been 
attacked by the magma, some to a moderate extent and others to 
such an extent that only traces of the inclusion remain. To some 
petrologists these inclusions are but the remnant of a great host, 
most of which has been completely incorporated in the magma, and 
to such incorporation or assimilation of foreign matter they would 
assign the principal variations of igneous rocks. ‘The variations are 
not usually regarded by these petrologists as the result of assimila- 
tion alone but of assimilation followed by the differentiation of the 
syntectic magma which is supposed to have special powers of differ- 
entiation not possessed by the original magma. Other petrologists 
believe that magmas cannot be expected to have the energy content 
necessary for the solution of a significant amount of foreign material; 
that the amount of solution actually observed at and near contacts 
is an approximate measure of the total and that the variations of 
igneous rocks are quite independent of these slight additions, being 
due to spontaneous powers of differentiation possessed by original, 
uncontaminated magmas. 

513 


514 N. L. BOWEN 

In the present paper it is proposed to discuss the behavior of 
inclusions with special reference to these questions. By application 
of the principle of the reaction series, as developed in a former 
paper,’ it is hoped to effect a certain amount of reconciliation of 
these extreme views. 


HEAT EFFECTS OF SOLUTION 


Since one of the important questions involved is that relating 
to the heat effects resulting from solution it is desirable to consider 
the information available on 
these effects. The ordinary 
equilibrium diagram, commonly 
regarded as a freezing-point dia- 
gram, is at the same time a solu- 
bility diagram. It gives the 
change of solubility of any phase 
with temperature. But the 
change of solubility with temper- 
ature depends mainly? on the 
heat effect involved in solution 
and the equilibrium diagram 
contains complete information 


A 


1480 


80 


fo) 40 60 
ANORTHITE MOL PERCENT ALBITE 


Fic. 1.—Equilibrium diagram of the 
plagioclase feldspars. The circles indicate 
determined points. ACB and ADB are 
calculated curves assuming no heat of 
mixing. 


on this heat effect. Unfortu- 
nately the information may be 
very difficult of extraction; in 
the present state of knowledge, 
often impossible. One solubility 


diagram, that of the plagioclase 
feldspars, has proved particularly simple in this respect. This dia- 
gram is shown in Figure 1, the curves being calculated on the basis 
of a latent heat of 104.2 cal. per gram for anorthite and 48.5 cal. per 
gram for albite. The determined points are given by the small circles 
and their correspondence with the calculated curves is very remark- 
able. Since the curves were calculated on the basis of constant 
latent heats (solution heats) this correspondence simply means that 


tN. L. Bowen, Jour. Geol., Vol. XXX (1922), pp. 177-98. 


2 The volume change is involved also but is relatively unimportant. 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 515 


there are no mixing-heat effects and that the heat of solution of any 
plagioclase in liquid plagioclase is simply the latent heat involved in 
the change from solid to liquid. 

In Figure 2 is plotted the equilibrium diagram for anorthite and 
diopside. Now we know the latent heat of anorthite from the cal- 
culated results of Figure 1 and we may calculate, using this value, a 


| 


an fon =n 
20 4O 60 EZ) ae) 
Cangs 6G mal percent CAA, St, O, 


Fic. 2.—Equilibrium diagram of diopside and anorthite. Determined curves in 
full limes. Broken curves calculated on the assumption of no mixing heats. 


curve of freezing-point depression for anorthite according to the 
equation for ideal concentrated solutions, viz. : 


pices lnx 
Q 
where 7 is the melting temperature (saturation temperature) of 
anorthite in a solution of mol fraction « in anorthite; Q is the 
latent heat of melting per mol of anorthite (=29,000 cal.) and 
T, its melting-point (=1550°). As a result of this calculation we 
obtain the right-hand dotted curve. It will be noted that the 
determined curve corresponds with the calculated dotted curve in 
the upper portion (as far as about 80 per cent anorthite) and then 
falls below. In other words, we could have calculated the latent 


™See N. L. Bowen, Melting phenomena of the plagioclase feldspars. Amer. 
Jour. Sci., Vol. XXXV (1913), Pp. 590. 


516 N. L. BOWEN 


heat of anorthite from a point on the upper portion of the deter- 
mined curve. If now we do this for the diopside curve, that is, cal- 
culate the latent heat of diopside from a point on the upper portion 
of its solubility curve, we find a latent heat of 23,420 cal. per mol 
or 108 cal. per gram.* Calculating the further course of the 
curve with the use of this value we obtain the dotted curve. This 
calculated curve also lies above the determined curve at points dis- 
tant from pure diopside. ‘There is one factor which can cause such 
a deviation of the freezing-point curve from the theoretical curve of 
the above equation, viz., a heat of mixing of the liquids, and Van 
Laar has developed an equation which enables one to calculate the 
differential heats of mixing involved. The equation is 


a(1—x)? 


“eie@—ar 


T=T, 


where a and 7 are coefficients from the Van der Waals equation of 
state for binary mixtures. The numerator gives the number of 
times the solution heat, at the concentration x and temperature T, 
is greater than the melting heat Q, and the difference between this 
and Q is the differential heat of mixing. From this equation we 
find the following values of the differential heats of mixing per 
mol (q). 
TABLE I 


Moat DIFFERENTIAL HEATS oF MIxING IN ANORTHITE-DIOPSIDE MIxTURES 


: Pe ay (Coa nerer nical: 0.8 0.5 0.4 0.35 
Eom amonthite Nigh (calss) eee 100 600 1100 1250 
Haridionside e He te ttardasts 0.7 0.65. ° lv nee eee | Eee 


(GalS|) hee 120 340. ie Seb ho a 


Now these differential heats of mixing are the heats of mixing of one 
mol of liquid with a very large amount of solution of the various 
concentrations referred to. These in themselves have no particular 
7A direct determination by W. P. White gave 10615 cal. Amer. Jour. Sci., 
Vol. XXVIII (1909), p. 486. 
2Z. physik. Chem., Vol. VIIL (1891), p. 188. 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 517 


interest from the present point of view but from them the so-called 
integral heats can be calculated by a graphical method. Rooze- 
boom? shows that if the curve of integral heats of mixing is plotted 
against mol fractions of the components then the intercept on 
the heat axis of the tangent to the curve gives the differential heat 
of mixing for the composition represented by the point. Thus in 
Figure 3 if the curve ABC represents the heats of mixing of diopside 
liquid and anorthite liquid in various proportions to form one mol 
of mixture, then the differential heat of mixing of one mol of 


<O 


40 60 30 
Ca/19 Slo Qs mol per cent CAL Sts Og 


Frc. 3.—Curve of integral mixing heats of diopside and anorthite liquids (ABC). 
Showing graphical method of determining integral mixing heats from differential mixing 
heats. 


anorthite ina large amount of liquid of the mol fraction x is 
given by the intercept on CE of the tangent at x. Also the differ- 
ential heat of solution of diopside in this same mixture is given by 
the intercept on AD of the same tangent. Now in our particular 
case we have determined the differential heats by calculation from 
the freezing-point curves and we wish to know the integral heat. 
This is the reverse of the above problem and while not as straightfor- 
ward can nevertheless be solved in the same way. We have, 
fortunately, for the composition of the eutectic, the differential 
heats of mixing for both diopside and anorthite which completely 
fixes the tangent at the composition of the eutectic. Figure 3, 
which has been referred to for purposes of illustration, is also the 
actual figure for diopside and anorthite determined graphically from 


« Die Heterogenen Gleichgewichte, Zweiter Heft, Erster Teil (1904), pp. 287-90. 


518 N. L. BOWEN 


the calculated differential heats of Table I. Thus FG is the tan- 
gent at the composition of the eutectic joining the value of the dif- 
ferential heat of mixing for diopside liquid in the eutectic with the 
differential heat of mixing for anorthite liquid in the eutectic liquid. 
This fixes immediately that the integral heat of formation of one 
mol of the eutectic liquid from its liquid components is 650 cal. 
Knowing that the curve of integral heats is tangent to this line at 
x=0.65, we may readily draw a complete curve (the envelope of 
the family of tangents) that satisfies the known values of the differ- 
ential heats. When thisis done we get the curve ABC. Thiscurve, 
then, gives us directly the amount of heat evolved when liquid 
anorthite and liquid diopside are mixed in any proportion. Thus 
when + mol diopside liquid is mixed with ? mol anorthite liquid 
500 calories are evolved. The maximum amount of heat per mole- 
cule of mixture (720 cal.) is evolved when about .047 mol of anorth- 
ite liquid is mixed with 0.53 mol of diopside. This happens to be 
about equal weights, so that the maximum amount of heat is evolved 
when about equal weights are mixed and is equal to about 3 cal. 
per gram of mixture. This heat is sufficient to heat the mixture 
about 10°. 

If now we turn to the equilibrium diagram of diopside and albite 
and calculate a curve of freezing-point depression for diopside, using 
the value of the latent heat found from the anorthite diagram, we 
find again that the calculated curve coincides with the observed 
curve in its upper portion and then deviates from it (Fig. 4), but in 
this case in the opposite direction to that found for the other dia- 
gram. Again, this deviation can be interpreted as due to a heat of 
mixing of the liquid but now of the opposite sign. If we apply the 
Van Laar equation we can calculate the differential heats and from 
these determine graphically as before the integral heats of mixing. 
Thus we get Figure 5. 

We find, then, that albite liquid and diopside liquid mix with 
absorption of heat, the maximum absorption (790 cal.) taking place 
when 0.48 mol albite is mixed with 0.52 mol diopside. ‘This also 
is about equal to 3 cal. per gram of mixture. 

Assuming that the theoretical basis of our calculations is to be 
relied upon we have proved that anorthite and diopside, both 


THE BEHAVIOR OF INCLUSIONS IN-IGNEOUS MAGMAS 519 


characteristic molecules of basic rocks, mix in the liquid state with 
evolution of heat, whereas albite, one of the most characteristic 
molecules of acid rocks, mixes with diopside in the liquid state with 
absorption of heat and with anorthite without significant heat effect. 


4500 


1200. 


—— 


20 40 60 30 
Ca/1q5t 20, W770/ Per Cent WaAl St; Oy 


Fic. 4.—Equilibrium diagram of diopside and albite. Determined curves in 
full lines. Broken curve calculated on the assumption of no mixing heats. 


These results are quite the opposite of what is often assumed to be 
the case. Many statements are to be found in the literature to the 
effect that acid and basic rock material will mix with evolution of 
heat. There is nothing in the results of studies of phase equilibrium 


> iy 
S 8 
COVOLICS 


g 


— 2 4 = 


20 — 


40 60 80 
Callg5StG, mol per Cent NGAI St; Gg 


Fic. 5.—Curve of integral mixing heats of diopside and albite 


to warrant such a statement; indeed, there are, as we have seen 
above, good reasons for doubting it. 

Possibly in rare cases, where more fundamental reactions are 
involved, greater heats of mixing are available, but the mixing of © 
two liquids has, in itself, little importance in petrogenesis. Our 
particular interest lies in the heat effect on mixing solid rock-matter 


520 N. L. BOWEN 


with liquid magma. In-that connection, however, the heat of 
mixing of liquids is not without significance, for the heat of solution 
of a solid is to be regarded as the resultant of two heats, the heat of 
melting of the solid and the heat of mixing of the liquids. Now the 
heats of mixing that we have calculated above are really very insig- 
nificant as compared with the heats of melting of the solids, and the 
heats of solution of the solids are, therefore, very nearly equal to 
the latent heats of melting. For the solution of solid anorthite in 
diopside a little less than the latent heat is required, for the solution 
of solid albite in diopside a little more than the latent heat is re- 
quired. The differences are noteworthy in connection with any 
theory that postulates an evolution of heat when an acid rock is 
immersed in basic magma, but for the purposes of the present 
inquiry which seeks to find merely the order of magnitude of the 
heat effect when solid rock is added to magma, we may state that 
the heat of solution of solid anorthite, albite, or diopside in any 
liquid mixture of them is substantially equal to the latent heat of 
melting. Moreover it is probably true of silicates in general that the 
solution of the solid is attended by a large absorption of heat, 
though not many determinations lend themselves to interpretation 
in the same way as the above. Thus when one attempts similar 
calculations from the equilibrium of anorthite with nephelite and of 
anorthite with silica it is found necessary to assume molecular associ- 
ation in nephelite and silica, and, while this is not surprising since 
both occur in more than one crystal form, it nevertheless so compli- 
cates the case that interpretation becomes impossible. One thing is 
certain, namely, that all the solubility curves of silicates yet deter- 
mined show a marked increase of solubility with temperature, which 
means a strong absorption of heat upon solution. No example of 
retrograde solubility is known. In conclusion, then, we may state 
that the solution of a silicate in a magma is usually accompanied by 
a large absorption of heat, probably of the order of magnitude of the 
heat of melting. 
THE QUESTION OF SUPERHEAT 


Having thus arrived at a general conception of the heat re- 
quired for solution we may consider the question of the heat avail- 
able. One aspect of this is concerned with the superheat of the 


, 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 521 


magma, that is, the excess of temperature above that at which 
crystallization begins. Somewhere within the earth there is material 
(whether a rigid liquid or a crystalline solid we need not here con- 
sider) which is capable of becoming fluent liquid upon release of 
pressure. At great depths there may be excessively hot material 
capable of giving excessively hot liquid, but this is hardly concerned 
with matter that ever comes to the light of day as an igneous rock. 
There must be a transition zone representing a passage from stable 
crustal material incapable of becoming magma, through a zone 
capable of giving magma with some suspended crystalline matter, 
to a zone capable of giving a completely molten magma. It is from 
the zone giving rise to magma with some suspended crystals, rather 
than from the zone giving a simple liquid, that we must expect the 
great upwellings of magma to come, because of the advanced posi- 
tion of the partly crystalline material with respect to an action pro- 
ducing such upwelling and because it may be readily mobile when 
the amount of suspended crystals is small. As the magma rises in 
the crust several factors may combine to reduce the amount of 
crystals. Among these factors is the increased solubility of the 
crystals resulting from the lowered pressure, for the solution of sili- 
cates usually takes place with increase of volume. Besides this 
there may be an actual addition of heat as a result of the Joule- 
Thomson effect? and of possible exothermic reactions ensuant upon 
reduced pressure. It is probable that, in the usual case, the magma 
still retains some crystals even when it rises to shallow depths within 
the crust for, though the intensity of the above actions is thereby 
increased, it is also passing into colder and colder surroundings and 
must lose a corresponding amount of heat. It is possible that in 
some cases a magma may have lost all its crystals by the time it has 
risen to shallow depths and perhaps may have a temperature above 
the saturation temperature under the conditions there prevailing. 
However, even if we make the liberal assumption that it is super- 
heated 100° (say its temperature is r100° and the saturation tem- 
perature tooo°) and that this condition is acquired when it has 
reached a level 5 km. below the surface, it should be noted that 
the mere immersion of blocks of country rock amounting to about 


tL. H. Adams, Bull. Geol. Soc. Amer., Vol. XXXIII (1922), p. 144. 


522 N. L. BOWEN 


To per cent of the mass of magma would be sufficient to wipe out this 
superheat, for the original temperature of blocks even at this quite 
considerable depth would be only 200°. _ 

This, too, is quite apart from the amount of heat that must be 
lost by the magma in heating up a large amount of wall rock other 
than the immersed fragments. If we consider greater depths, where 
the surroundings would be at a higher temperature, it must be 
remembered that we are thereby limiting the amount of superheat 
that may have been acquired as a result of those processes attendant 
upon the rise of the magma and proportional in amount to the extent 
of rise. 

It is plain that there can seldom be more than a very small 
amount of superheat left after the heating up of immersed frag- 
ments. Occasionally, perhaps, such small amounts may be avail- 
able for the direct solution of fragments but it would require only 
an excessively small amount of solution to use up this heat, a fact 
that becomes apparent when the enormous discrepancy between the 
specific heats of silicate liquids (0.2-0.3 cal.) and the solution heats 
of silicates (50-125 cal.) is realized. 

We have thus deduced, from general considerations, that mag- 
mas cannot be expected to have much superheat and in particular 
that they cannot retain a significant amount after the immersion of 
foreign fragments in such quantity that an important effect upon the 
composition of the magma would ensue if solution did occur. This 
deduction need not be regarded as contradictory to the observed 
fact that on very rare occasions inclusions of foreign rock have been 
found converted into glass by magmas and that the measured tem- 
peratures of lavas at certain active volcanoes are sometimes high 
enough to indicate a condition probably significantly above that 
of beginning of crystallization. It is becoming increasingly appar- 
ent that in central volcanoes there are sources of heat, probably 
from exothermic gas reactions, that are capable of producing the 
temperatures observed, but these must be regarded as locally con- 
centrated where the gases have their vent. Moreover, inclusions 
converted to glass are found almost exclusively in such extrusive 
rocks. Even their rare appearance in intrusive masses need not 
militate against the general conclusion reached, for, admitting a 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 523 


certain amount of superheat, the first few inclusions added would 
receive the full benefit of it and might be fused, but the magma would 
suffer a correspondingly marked loss of heat. In other words, the 
net result of the addition of an amount sufficient to produce a signifi- 
cant change of composition, say ro per cent, would be the same even 
though they were added slowly and the first additions were drasti- 
cally affected. Not only do magmas commonly fail to convert inclu- 
sions into liquid but they also fail to effect such changes as the 
transformation of quartz into tridymite and of wollastonite into 
pseudo-wollastonite, which fact must be regarded as indisputable 
evidence of the prevailing low temperatures. 

There seems no reason, therefore, to doubt that direct solution 
of foreign material in superheated magmas cannot be a factor of 
importance in petrogenesis. However, the importance of superheat 
has been greatly exaggerated both by those who adhere to the view 
that magmas dissolve large quantities of foreign matter and by 
those who deny it. We shall find in the sequel that even saturated 
magmas may produce very marked effects in the way of incorpora- 
tion of foreign material. 


EQUILIBRIUM EFFECTS BETWEEN ‘“‘INCLUSIONS”’ AND LIQUIDS 
IN INVESTIGATED SYSTEMS 


In the following discussion of the effects of liquids upon inclu- 
sions, whose composition is embraced within investigated systems, 
attention will be confined, unless otherwise stated, to saturated 
liquids. ‘This is done because we have found in the foregoing dis- 
cussion that the superheated condition has little importance in 
nature. But lest it be thought that this is going too far in the way 
of eliminating superheat we have made another assumption that 
should amply compensate, namely, that the inclusions are already 
heated to the temperature of the liquid before immersion in it. 
It is proposed to discuss the magnitude of the heat effects 
involved when reactions go on between solid inclusions and liquid 
in systems that have been experimentally investigated and where 
the equilibrium relations at various temperatures and the approxi- 
mate heat effects are known. It is not expected that the numerical 
results so obtained will have any direct applicability to natural 


524 N. L. BOWEN 


magmas, but it is hoped that some principles of general significance 
may be thereby brought to light. 

As a beginning we shall refer to equilibrium in the plagioclase 
feldspars which is given in Figure 1. In the figure we note that 
liquid of composition Ab,An, begins to crystallize with the separa- 
tion of crystals of composition about Ab,An,. As the cooling pro- 
ceeds, if perfect equilibrium obtains, the crystals will be made over 
by the liquid so that the composition of the crystals changes along 
the curve ACB. It is plain, then, that if one had a mass of liquid 
Ab,An, at 1450°, that is just saturated, and added to this mass some 
crystalline material of the composition Ab,An, (approx.) (already 
heated to 1450°) which we shall now call foreign inclusions, the 
liquid would, if perfect equilibrium were attainable, make over these 
inclusions as the temperature fell so that their composition followed 
the curve ACB. How much of this work the liquid will be able to 
accomplish in any individual case will depend on such factors as the 
size of the inclusions, their permeability to the liquid, and the rate 
of cooling, but the tendency is very plain. We can thus have a 
liquid exerting a marked influence upon inclusions even though 
these are precisely of the composition in equilibrium with the liquid, 
provided the solid is a member of a solid solution series. It should 
be noted, however, that this action is without effect on the course 
of the liquid. The composition of the liquid follows the curve ADB 
whether the inclusions are present or not. The career of the liquid 
may, however, be brought sooner to a close as a result of reaction 
with the inclusions, that is, the liquid may be entirely used up at a 
somewhat higher temperature. 

We may now examine the case of adding to a plagioclase liquid 
some solid plagioclase more calcic than that with which the liquid is 
in equilibrium. To 50 g. liquid Ab,An, at 1450° (just saturated) 
let us add 50 g. solid Ab,Any, already heated to 1450°. Equilibrium 
will be established, if the temperature is kept constant, only when 
the solid is completely changed to that with which the liquid is in 
equilibrium, Ab,An, (approx.). Since the total composition is 
represented by the point K we can easily determine the proportions 
of liquid and crystals. They will be 17 per cent liquid of composi- 
tion Ab,An, and 83 per cent plagioclase of composition Ab,An, 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 525 


(approx.). We have now 66 g. An and 17 g. Ab in the crystalline 
state. We had formerly only 45 g. anorthite and 5 g. ablite. Equi- 
librium will therefore be established with evolution of heat to the 
amount of 21 X104.2+12X48.5 cal. In order to keep the tempera- 
ture constant, then, we should have to abstract 2770 cal. If, on the 
other hand, no heat were abstracted the temperature would rise 
somewhat and equilibrium would be established at a slightly higher 
temperature. For the particular case we have assumed we may 
readily calculate that equilibrium would be established at about 
1465°, when the mass would consist of about 60 per cent crystals of 
composition Ab,An;. Even when the reaction takes place adia- 
batically, there is an increase in the proportion of crystals. The 
reaction is in no sense a solution of foreign material. Rather by a 
making over of the foreign material it becomes no longer foreign 
but identical with the crystalline matter with which the liquid is in 
equilibrium. Moreover, if the originally foreign matter is more 
calcic than the crystals with which the liquid is in equilibrium the 
reaction is an exothermic one. How much of this reaction would 
take place in any individual case cannot be predicted. It will 
depend upon rate of cooling and other factors that readily suggest 
themselves, but there is plainly a tendency toward such a reaction 
and the reaction is exothermic. 

Let us now examine somewhat more minutely into the cause of 
this exothermic reaction and we shall find that it is due to a general 
principle and is not dependent upon the particular properties of the 
plagioclase series discussed. During the crystallization of a plagio- 
clase mixture a small decrement of temperature results in the reac- 
tion: 
plagioclase+-liquid=a little more plagioclase of somewhat more sodic com- 

position. 

Since this is an equilibrium reaction taking place with falling temper- 
ature, it must be exothermic. When we add the plagioclase Ab,An, 
to liquid Ab,An,we merely integrate this reaction over the tempera- 
ture range 1500°—-1450° and the composition range Ab,An,—Ab,An,. 
Thus we could start with a liquid of composition Ab,An,, permit it 
to crystallize until at 1500° the crystals would be of the composition 
Ab,Any, filter off the crystals, permit the liquid to cool to 1450°, 


526 N. L. BOWEN 


when it would attain the composition Ab,An,, and then add the for- 
eign crystals Ab,An, that we filtered off. It is plain that we would 
have available by the making over of these crystals at 1450° all the 
heat that would have been evolved between 1500° and 1450° by the 
continuous process of making over of these crystals had they been 
left in contact with the liquid. Though we have used the plagio- 
clase series as an illustration it is clear that the exothermic reaction 
taking place at 1450° is merely a deferred result of the principle of 
Le Chatelier which states that an equilibrium reaction proceeds, 
with falling temperature, in the direction resulting in evolution of 
heat. It is a perfectly general property of any solid solution series 
that if, at any temperature, crystals which are at equilibrium with 
liquid at a higher temperature are added to saturated liquid, the 
reaction which ensues between liquid and crystals is exothermic. 

The case of the addition of an inclusion of composition nearer 
the low temperature (more sodic) end of the solid solution series may 
now be examined. A liquid of composition Ab,An, is just saturated 
at 1490°. We cannot add to it a more sodic solid inclusion at the 
same temperature because any more sodic inclusion will be liquid 
at this temperature. But if we add inclusions of Ab,An, at such a 
temperature that they are solid, say 1200°, it is plain that the actual 
temperature of the liquid is adequate to melt these inclusions; the 
only question is the source of the quantity of heat. ‘The liquid must, 
of course, be cooled off in supplying the heat required to heat up 
the inclusions, but, since the liquid is saturated, it cannot be cooled 
without some crystallization taking place. However, it will take a 
very small amount of crystallization to supply the heat necessary to 
heat up a considerable amount of inclusions. The enormous dis- 
crepancy between the specific heats of silicates and their solution 
heats is plainly of double significance in connection with these prob- 
lems. Not only can the heat necessary to heat up the inclusions be 
supplied by crystallization of some of the liquid, but so also can the 
heat required to melt the inclusions. To accomplish this it will 
require, however, the formation of crystals approximately equal in 
amount to the amount of inclusions melted. If we added 20 per 
cent of inclusions of Ab,An, at 1200°, to a liquid of composition 
Ab,An, at 1490° we would obtain (assuming that these thermal 


4 
i: 
i 
: 
| 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 527 


adjustments took place very rapidly compared with concentration 
adjustments) a mass of liquid about (Ab,An,) containing in it about 
20 per cent of crystals of the kind in equilibrium with it (about 
Ab,An,) and containing also the inclusions converted to liquid, the 
whole at a temperature of about 1470°. There is no objection, 
therefore, to the conversion of an inclusion into liquid if the inclusion 
is sufficiently contrasted with the magma in composition in the 
proper direction; nor is. such melting of an inclusion to be reneided 
as evidence of superheat in the magma. 

This condition, in which liquid inclusions are contained in the 
magma, is of course a temporary one and it is rather the end result 
that has particular significance in petrogenesis. Final adjustment 
of concentration takes place by formation of a single homogeneous 
liquid and adjustment of the composition of the crystals to that in 
equilibrium with this liquid. This necessitates a further drop in 
temperature to about 1460° where the whole mass consists of the 
liquid Ab,;An,,; containing somewhat less than 20 per cent crystals 
of the composition about Ab,An,. ‘Thus the net result of the addi- 
tion of inclusions of composition more sodic than the liquid is that 
the inclusions become a part of the liquid and at the same time calcic 
crystals are formed in amount slightly less than the amount of inclu- 
sions added, the heat needed in order to make the inclusions part of 
the liquid being supplied by the formation of crystals. In the sense 
that they become a part of the liquid the inclusions are dissolved, 
but the process is not simply the formation of a liquid whose com- 
position is the sum of that of the magma and the inclusions. 

Let us now observe how these effects are carried over into more 
complex systems. Figure 6 is the equilibrium diagram of the system 
diopside: anorthite: albite for which we have discussed the heat 
quantities on an earlier page and found that the solution heat of any 
solid phase can be regarded as substantially equal to its latent heat 
of melting. A liquid of composition A is, at 1250°, just saturated 
with plagioclase of composition Ab,An, approximately. If foreign 
inclusions consisting of the plagioclase Ab,An, were added to this 
liquid we would have an effect strictly analogous to that described 
for simple plagioclase mixtures. The liquid would tend to make the 
inclusions over into Ab,An, which takes place with evolution of heat. 


528 N. L. BOWEN 


If the rate of withdrawal of heat were very slow this reaction might 
result in an actual rise of temperature and the establishment of 
equilibrium at the higher temperature where the plagioclase crystals 
would be slightly more calcic than Ab,An,. If the liquid A had 
come into being as a result of the partial crystallization of another 
liquid, and if it carried plagioclase crystals suspended in it that 
showed zoning, the addition of the foreign inclusions mentioned 
might therefore result in a reversal of the zoning. 


DIOPSIDE 
R 


é » 
ALBITE ANORTH'TE 


Fic. 6.—Equilibrium diagram of diopside, anorthite, and albite 


The ‘‘attack”’ of the liquid upon the inclusions would be facili- 
tated at the margins and along any channels where the inclusion 
happened to be more readily penetrated. The replacement of a 
small unit of Ab,An, by Ab,An, means a large local increase of 
volume so that the action is bound to have a disintegrating effect 
upon an inclusion suspended in liquid. Thus the material of the 
inclusion, as it is gradually made over, tends to become strewn about 
in the surrounding liquid but there is no actual solution, no increase 
in the amount of liquid—indeed, there is a diminution. Such should 
be the behavior of an inclusion richer in calcic plagioclase than the 
crystals with which the liquid was in equilibrium. At a later point 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 529 


in the paper it will be shown that the observations of many petrog- 
raphers upon actual rock inclusions strongly suggest just such an 
action. 

If we turn now to inclusions of plagioclase less calcic than the 
crystals with which the liquid is in equilibrium we find that quite a 
different condition obtains. We may begin with the liquid A, 
again at 1250°, and add to it foreign inclusions of Ab,An, already 
heated to 1250°. It will be noted that we are adding plagioclase of 
the same composition as that in the liquid since liquid A is a mix- 
ture of Ab,An, and diopside. Suppose that the reaction takes place 
at first between a thin layer of liquid adjacent to the inclusion and 
an equal weight of the peripheral part of the inclusion. The com- 
position of this reacting mass is then to be represented by the point 
Y. Suppose further that the reacting mass is a very small portion 
of the total so that no significant drop of temperature occurs, for in 
this case the reaction absorbs heat. Equilibrium will be established 
in this reacting layer sensibly at 1250° when the composition of the 
crystalline outer crust of the inclusion is Ab,An, (Z) and that of the 
adjacent thin layer of liquid is MV, which is a mixture of Ab,An, and 
diopside. If the general cooling of the liquid were proceeding 
rapidly so that this condition were ‘frozen-in’? we would have a 
central core of unaltered inclusion of composition Ab,An,, an 
altered crust of the inclusion of composition Ab,An,, a reaction rim 
(formerly liquid) about this, consisting of a mixture of diopside and 
Ab,An,, all surrounded by the main mass consisting of a mixture of 
diopside and Ab,An,. ‘The reaction rim is not intermediate in com- 
position between the inclusion and the main mass, nor yet between 
the altered crust of the inclusion and the main mass. This is a 
commonly observed feature of reaction rims and has been referred 
to “selective diffusion,” an explanation which is correct but rather 
misleading. ‘The effect is really due to equilibrium effects between 
the inclusion and the liquid and such diffusion as occurs is selective 
only because equilibrium selects certain constituents to become part 
of the liquid and others to be fixed in the solid phase. In the case 
we have described, more sodic plagioclase is emphasized in the new 
liquid formed and more calcic plagioclase is emphasized in the solid 
product. We shall find it to be a general result of reactions of this 


530 N. L. BOWEN 


kind that the liquid should be enriched in the constituents toward 
the low temperature end of.a solid solution series and the solid in 
those toward the high temperature end. 

We have seen, then, that even when we add inclusions of a plagio- 
clase of the same composition as the plagioclase existing as liquid 
in the magma, quite marked reaction effects may be found. 

Let us now examine the same example but make the cooling of 
the liquid very slow. In other words, we shall discuss the end result 
of the action described above, which gives a reaction rim only as a 
temporary condition. In this case we shall imagine that the liquid 
formed about the inclusion becomes a part of the main mass of 
liquid as a result of diffusion and convection, and also that the solid 
products of the reaction become distributed through the liquid upon 
disintegration of the inclusion, due to the formation of local pockets 
of liquid and to the volume changes taking place in the change of 
composition of the solid phase. ‘Thus the inclusion completely dis- 
appears as a distinct entity. Let us imagine that the amount of the 
inclusions was to per cent of the total and determine what the effect 
on the mass as a whole will be. The bulk composition of the mass 
is represented by the point K and if the temperature were main- 
tained at 1250° the inclusions, formerly Ab,An,, would be changed 
to crystals of a composition close to Ab,An, and the liquid would 
acquire the composition P. The crystals would amount to only 
about 6 per cent of the mass and to maintain the temperature con- 
stant would require addition of heat. If the only heat available 
were that of the mass itself its temperature would be lowered and a 
slightly greater amount of crystals formed. The net result would 
be, however, a marked change in the composition of the inclusions, 
a moderate decrease in the actual amount of solid material with 
corresponding increase in the amount of liquid. The liquid, too, 
becomes more albitic, that is, is pushed onward upon its normal 
crystallization course. 

If, instead of inclusions of Ab,An,, more sodic inclusions were 
added, say Ab,An,, there would be a somewhat greater increase in 
the amount of liquid and a markedly greater enrichment of the 
liquid in more albitic plagioclase. Inclusions as rich in albite as 
Ab,.An, might be completely melted by the liquid before becoming 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 531 


a part of the general liquid, their melting being accomplished by 
precipitation of more basic plagioclase. 

Summing up the result of adding to a liquid various members of 
a solid solution series with which it is saturated we find that if the 
added inclusion is nearer the high temperature end of the series than 
the crystals with which the liquid is saturated the reaction is such as 
to decrease the amount of liquid and is exothermic. If the inclusion 
is nearer the low-temperature end of the series than the crystals 
with which the liquid is saturated, the reaction is such as to increase 
slightly the amount of liquid and is endothermic. The liquid, too, 
is enriched in this case in the constituents of the low-temperature 
end of the crystallization series. Even inclusions consisting of the 
precise crystals with which the liquid is in equilibrium must react 
with the liquid as the temperature falls. 

Whatever the composition of the inclusions, then, the liquid may 
show very marked effects upon them even though it is saturated, 
and consequently such effects would not constitute evidence of 
superheat. A little consideration will show, too, that these reac- 
tions have no significant effect upon the course of the liquid. Re- 
garding the progress of the liquid as resulting from fractional 
crystallization, the liquid will in all cases run along the boundary 
curve (ED) as the temperature falls. The point it will eventually 
reach on this curve will depend upon the perfection of fractionation 
which in turn depends upon the rate of cooling. Inclusions of the 
more calcic kind tend to limit the career of the liquid by using it up, 
but at the same time furnish heat that tends to slow up the rate of 
cooling. Inclusions of the more sodic kind tend to push the liquid 
onward upon its course of crystallization, but hasten the cooling. 
The original unaffected liquid has all the differentiation potentialities 
that the liquid has after entering into the reactions mentioned. 


REACTION SERIES 


In the discussion relative to liquids of the diopside: anorthite: 
albite system nothing has been said regarding the effect of adding 
solid diopside to liquids saturated with diopside for the reason that 
there is no effect. The solid diopside simply remains as such on 
account of its being a pure compound of definite composition. But 


532 N. L. BOWEN 


rock-forming minerals are seldom so simple. Quartz is the only 
important example of a rock mineral of definite composition, prac- 
tically all others being of a variable nature, that is, solid solutions. 
The rock-forming pyroxenes do not fall behind in this respect and the 
addition of pyroxene to a natural magma saturated with pyroxene 
would in general be attended by reaction phenomena similar to those 
we have described for the plagioclases. The precise nature of the 
reactions in the case of the pyroxenes cannot be stated except for 
the clino-enstatite-diopside solid solution series. 

In another paper the writer has discussed solid solution series and 
offered reasons for calling them continuous reaction series. The 
importance of the reaction relation between liquid and crystals was 
there discussed in its bearing on crystallization. Here we have seen 
its importance in connection with the behavior of inclusions. 

As was pointed out in the paper referred to, there is another 
type of reaction relation between liquid and crystals that is exhib- 
ited in the reaction pair and the discontinuous reaction series.? 

The existence of such series is again of great significance in 
connection with the behavior of inclusions. An important reaction 
pair are the olivine, forsterite, and the pyroxene, clino-enstatite. 
Their relation is exhibited in its simplest form in the binary system, 
forsterite-silica, of which the equilibrium diagram is shown in 
Figure 7. The effect of adding crystals of the first member of a 
reaction pair to a liquid saturated with the second member is well 
illustrated by this system. A liquid of composition (/) is saturated 
with clino-enstatite but lies on the unsaturated side of the metastable 
prolongation of the forsterite liquidus. It is therefore unsaturated 
with forsterite and we may imagine that around each added forster- 
ite crystal a small quantity of liquid of composition (VV) may form. 
This condition is, however, metastable and from this liquid clino- 
enstatite would immediately be precipitated with formation of the 
liquid (WZ). Through constant repetition of this formation from 
forsterite of an infinitesimal quantity of the metastable liquid, 
with immediate precipitation of clino-enstatite, the forsterite is 


*N. L. Bowen, Amer. Jour. Sci., Vol. XXXVIII (1914), pp. 228-37. 
2“The Reaction Principle in Petrogenesis,” Jour. Geol., Vol: XXX (1922), pp. 
177-98. 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 533 


converted into clino-enstatite. Effectively, then, the liquid (/) is 
supersaturated with forsterite. It cannot dissolve forsterite but 
can only convert it into clino-enstatite, the phase with which it is 
saturated. This principle is capable of general application and 
we may state that a liquid saturated with any member of a discon- 
tinuous reaction series is effectively supersaturated with all higher 
members of the series; it cannot dissolve them but can only con- 
vert them into the phase with which it is saturated. 

In connection with the specific case we have discussed, it should 
be noted that the amount of clino-enstatite formed is not simply the 


4600 


MQ, aLig. 


1950 


4700 /1g35¢0,, 20 MgSih; 40 60 60 Ste 
Sn 


Fic. 7.—Lower temperature portion of the equilibrium diagram, forsterite-silica 


chemical equivalent of the forsterite changed. ‘The chemical equiv- 
alent would be somewhat less than one and one-half times the for- 
sterite, whereas the amount of clino-enstatite formed is much greater, 
being in fact about five times the amount of forsterite. In other 
words, the action is not simply an addition of silica to the forsterite, 
with consequent impoverishment of the liquid in silica, for the liquid 
cannot have silica subtracted from it without passing under the 
clino-enstatite saturation curve, that is, without precipitating clino- 
enstatite. In order to convert the forsterite inclusions into clinoen- 
statite, the liquid must precipitate a large amount of clino-enstatite 
from its own substance, and the action uses up a large amount 
of the liquid. The liquid left has, however, in this binary case, the 
same composition as the initial liquid, if the temperature is kept 


534 N. L. BOWEN 


constant. The liquid is not at all desilicated, even though it has 
caused the conversion of the inclusions into a more siliceous phase. 
By precipitation of the phase with which it is saturated, it has 
adjusted its composition in such a way as to remain on the same 
saturation curve. This again is a principle that can be applied in 
general to the reaction between a liquid and inclusions belonging 
at an earlier stage of the reaction series than the phase with which 
the liquid is saturated. 

Further considerations relative to this reaction pair will be 
developed in connection with their behavior in the more complex 
liquids containing anorthite as worked out by Andersen* and shown 
in Figure 8. From this figure we may very readily predict what 
would happen to ‘“‘inclusions”’ of the various solid phases immersed 
in liquid. Let us take first a liquid just saturated with forsterite 
(say M at 1450°). Ordinarily crystals of forsterite would separate 
first; they would then react with liquid to form pyroxene; pyroxene 
would continue to separate for a time and would then be joined by 
silica; finally, at the ternary eutectic 1222°, anorthite would join 
these, and the product would consist of pyroxene, silica, and anorth- 
ite. If to the original liquid some ‘‘foreign inclusions” of forsterite 
were added the liquid would, on cooling, attempt to make the 
forsterite over into pyroxene, but now there would not be enough 
liquid to accomplish this entirely, and it would be used up at 1260° . 
while there was still some forsterite left; so that the solidified mass 
would now consist of pyroxene, forsterite, and anorthite. Thus we 
see that even though it is precisely the substance with which the 
liquid is in equilibrium, the addition of forsterite has a considerable 
effect upon the crystalline product formed. The exact effect is a 
tendency to limit the scope of the products to the early members of 
the crystallization sequence. In this case, silica, a later member of 
the sequence, does not appear. It is plain, too, that the effect does 
not depend on the particular properties of this reaction pair and 
that we may conclude that it would be true of any discontinuous 
reaction series. We have already seen that the same effect is found 
in the continuous reaction series (solid solution series). We may 

«The system anorthite-forsterite-silica, Amer. Jour. Sci., Vol. XXXIX (1915), 
Pp: 440. 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 535 


state it as a general law therefore that a saturated liquid in any sys- 
tem dominated by reaction series will not remain indifferent to 
inclusions even of the exact composition with which it is in equilib- 
rium, and that the effect of the addition of such inclusions is a 
tendency to limit the scope of the crystalline products to adjacent 
members of the reaction series involved. 

Let us now examine what happens when inclusions of an early 
member of a discontinuous reaction series (or reaction pair) are added 


sO 
2 


Mc,SO, Cra SiO, 


Fic. 8.—Equilibrium diagram of the system anorthite-forsterite-silica. (After 
Andersen.) 


to a liquid saturated with a later member. The liquid F is just 
saturated with clino-enstatite at 1450°. If it is cooled, clino- 
enstatite will separate first, it will later be joined by silica and then 
by anorthite, when the whole mass will solidify at 1222°. But if 
inclusions of forsterite are added to this liquid it will instantly start 
to react with these inclusions and make them into the mineral with 
which it is saturated, viz., clino-enstatite. If there is perfect oppor- 
tunity for reaction we may readily predict what the end result will 
be for varying amounts of added inclusions. 


530 N. L. BOWEN 


Suppose first that the amount of inclusions was such as to give a 
total composition represented by the point K. The liquidwould 
then be completely used up by the reaction while some inclusions 
yet remained, so that the finally solidified mass would consist of 
forsterite, clino-enstatite, and anorthite. If, on the other hand, the 
amount of inclusions was such as to give a total composition repre- 
sented by the point JZ, all the inclusions would be changed to clino- 
enstatite while some liquid yet remained, and this liquid would then 
pass onward to the deposition of silica and anorthite in the ordinary 
way. Thus the addition of inclusions of this kind, if in sufficient 
quantity, restricts the career of the liquid and confines the crystalline 
products to adjacent members of the reaction series. 

It is perhaps not necessary to add that if the liquid did not react 
with the inclusions no effect on the course of the liquid would ensue. 

Just as was the case in the binary system, the liquid is not desili- 
cated by this addition of silica to the forsterite inclusion, but 
by precipitating an appropriate amount of the phase with which it 
is saturated (clino-enstatite), it maintains its position on the same 
(clino-enstatite) saturation surface. 

Nothing has been said of the heat effect of this reaction and it has 
been tacitly assumed that the temperature remains constant. The 
reaction is, in fact, exothermic, as can be readily shown by applying 
the same reasoning as was applied to the similar case in a continuous 
reaction series. Ifno heat were abstracted from the system it would 
heat itself up and equilibrium would be established at a slightly 
higher temperature with a somewhat more magnesian liquid than 
the initial liquid. But the formation of this somewhat more mag- 
nesian liquid is not properly to be taken as an indication that the net 
result of the process is a direct solution of some of the inclusions. A 
direct solution of inclusions would mean a decrease in total solids 
and an increase in liquid, whereas the reaction referred to results na 
diminution in the amount of liquid and a corresponding increase in 
the amount of solids even when this heating effect takes place. If 
heat is being taken from the system this process would act as a 
deterrent upon the rate of cooling. 

All of these effects we have found to be true of analogous inclu- 
sions in the case of the continuous reaction series. 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 537 


There is yet to be examined the example in which a late member 
of a discontinuous reaction series is added to a liquid saturated with 
an early member. ‘To the liquid P at 1500°, where it is just satu- 
rated with forsterite, inclusions of clino-enstatite are added in an 
amount sufficient to give a total composition Q (about 20 per cent). 
If the temperature were kept constant equilibrium would be estab- 
lished when the mass consisted of 4 per cent forsterite and 96 per 
cent liquid, that is, the inclusions have been changed into the phase 
with which the liquid is saturated and there has been an increase in 
the amount of liquid. In order to effect this change heat would have 
to be added to the system. If the system is self-contained, that is, 
if the only heat available is the heat of the system itself, a cooling 
would result and this would necessitate the crystallization of a fur- 
ther amount of forsterite until the necessary heat was supplied by 
this crystallization and the cooling of the mass. The net result 
would depend entirely on the relative heats of solution of forsterite 
and clino-enstatite in the liquid. These are probably of the same 
order of magnitude, so that equilibrium would be established at about 
1475 when the mass consisted of 10 per cent forsterite and 90 per 
cent liquid approximately. The net result, then, has been the con- 
version of the inclusions added (clino-enstatite) into the phase with 
which the liquid is saturated (forsterite), an enrichment of the 
liquid in the material added, with, at the same time, a pushing 
onward of the liquid along its usual course of crystallization. Or it 
could be stated that the inclusions pass into solution by precipitat- 
ing their heat equivalent of the phase with which the liquid is satu- 
rated.t This, then, is the result of adding inclusions which belong 
to a discontinuous reaction series and are later in that series than 
the phase with which the liquid is saturated. It is sensibly the same 
result as that obtained in the corresponding case in the continuous 
reaction series. 


EFFECTS OF MAGMA UPON INCLUSIONS OF IGNEOUS ORIGIN 


In the paper in which the conception of the reaction series was 
developed it was shown that these series are very prominent in rocks. 
An attempt was made there to arrange the minerals of rocks as 


t This is only approximately true, for equilibrium is always established at a some- 
what lower temperature and the cooling of the mass supplies a little of the heat. 


538 N. L. BOWEN 


reaction series and it was found that there are, toward the basic end 
of rock series, two fairly distinct reaction series that finally merge 
into one in the more acid rocks. This was expressed diagrammati- 
cally as follows. 


TABLE II 
olivines ey 
be 
\ 2 calcic plagioclases 
bo | 
Mg pyroxenes a) i 


\ calci-alkalic plagioclases 
Mg-Ca pyroxenes 
alkali-calcic plagioclases 


amphiboles J 
alkalic plagioclase 
biotites 


potash feldspar 


muscovite 


| 


Quartz 


On the basis of the principles that we have found to govern the 
behavior of inclusions belonging to reaction series we may deduce 
with considerable confidence the effects of liquid upon inclusions in 
this more complex series. 

It may be stated immediately that any magma will tend to make 
inclusions over into the phase or phases with which it is saturated, 
in so far as the composition of the inclusions will permit. It may 
be stated also that any magma saturated with a certain member of 
a reaction series is effectively supersaturated with all higher members 
of that reaction series. It cannot, in any sense, dissolve inclusions 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 539 


of such higher members but can only react with them to convert 
them into that member of the reaction series with which it is satu- 
rated, often by passing through other members of the series as inter- 
mediate steps. The material used to effect these changes cannot 
be regarded as simply subtracted from the liquid for the liquid is not 
free to become impoverished in any random substance. In general 
impoverishment in any substance will cause the liquid to pass within 
a region of saturation and induce the precipitation of some of the 
phases with which the liquid is saturated. 

Let us take, for example, a magma saturated with biotite, say, a 
granitic magma. This magma is effectively supersaturated with 
olivine, pyroxene and amphibole and cannot dissolve them in spite 
of the marked contrast of composition, which is often supposed to be 
an aid to the solution of inclusions. But the magma can and will 
react with these minerals and convert them into biotite, usually by 
steps. The subtraction of material necessary to produce biotite 
will cause the precipitation of the minerals with which the magma 
is saturated until either the liquid or the inclusions are used up or the 
reaction is brought to an end on account of mechanical obstruction. 

Similarly, granitic magma saturated with an acidic plagioclase 
cannot dissolve basic plagioclase but can only react with it and con- 
vert it into more acid plagioclase. 

These remarks are tantamount to the statement that saturated 

‘granitic magma cannot dissolve inclusions of more basic rocks. The 

magma will, however, react with the inclusions and effect changes 
in them which give them a mineral constitution similar to that of 
the granite. These changes will often be accompanied by disin- 
tegration of the inclusions and the strewing about of the products 
which may be indistinguishable from the ordinary constituents of the 
granite. The inclusions may thus become completely incorporated 
though not in any sense dissolved. It is this action of magmas upon 
inclusions that makes particularly difficult the problem of distin- 
suishing xenolith from autolith, i.e., accidental inclusion from cog- 
nate inclusion. 

Whatever origin one may assign to a granitic magma—let it be 
formed by differentiation of more basic magma, by differentiation of 
syntectic magma or by palingenesis of sediments—there seems no 


54° N. L. BOWEN 


escape from the conclusion that it will normally be saturated. The 
normal effect of granite upon more basic inclusions should therefore 
be such as has been outlined above. ‘Thus we find that Fenner, in 
describing the action of granitic magma on basic bands in an injec- 
tion gneiss, says, ‘‘In other places the dark minerals appear to have 
been taken up or digested by the magma and to have crystallized 
out again in large blades. Even in the latter case it is not always 
certain that perfect solution has been effected at any one time. The 
process may have been rather in the nature of a chemical reaction 
with the original minerals or the solution and redeposition of a por- 
tion of the material at a time,’ leaving the general relations undis- 
turbed. This possibility is suggested by the fact that frequently 
even the coarser micaceous blades or aggregates of dark minerals 
show evidence of parallelism and this would be difficult to account 
for under the supposition that solution was so perfect that the 
original structure was completely wiped out.’’” Here, apparently, we 
have a good example of the transformation, by reaction rather than 
solution, of the dark bands into mica-rich material, mica being the 
dark mineral with which the magma is saturated. ‘The change of 
serpentine into biotite as observed by Gordon at the borders of 
granitic pegmatites is precisely the action to be expected.’ 

V. M. Goldschmidt describes the strewing about of the minerals 
of a basic hornfels in magmas of the Christiania region. He says, 
“This strewing about hardly has its origin in a solution of the miner- 
als and their later separation. Had solution occurred the grains 
would not have retained their original forms and they would have 
differed in composition from the minerals of the hornfels.” Near 
the border of an apophysis of the nordmarkite, grains of diopside 
from the hornfels are surrounded by a rim of aegirite. In the center 
of the apophysis the aegirite has no core of diopside.‘ | 


t Note discussion on p. 532. 

2C. N. Fenner, ‘The Mode of Formation of Certain Gneisses in the Highlands 
of New Jersey,” Jour. Geol., Vol. XXII (1914), pp. 602-3. 

3‘Desilicated Granite Pegmatites,’’ Proc. Acad. Nat. Sci. Phila., Part I (1921), 
p. 169. 

4V. M. Goldschmidt, ‘Die Kontaktmetamorphose im Kristianiagebiet,” Vid. 
Selsk. Skr. I. Mat. Naturv. Klasse (1911), No. 1, pp. 107-8. 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 5A1 


In the foregoing discussion granitic magma has been taken 
merely as an example to which the principles developed may be 
applied. Asa further example it may be pointed out that saturated 
dioritic magma cannot dissolve inclusions of gabbro, peridotite or 
pyroxenite but, given the opportunity, it will react with those 
inclusions and convert them into the hornblende and the plagioclase 
with which it is saturated, at the same time precipitating a further 
amount of this hornblende and plagioclase from its own substance. 

These are but examples of the application of the principle that a 
saturated magma cannot dissolve inclusions of material farther 
back in the reaction series (in general more basic) than the crystals 
with which it is saturated. At the same time the magma can attack 
these inclusions, reacting with them in such a manner as to convert 
them into the crystals with which it is saturated. 

The dioritic magma we have considered will not remain indiffer- 
ent to inclusions even of the exact composition of the crystals with 
which it is in equilibrium, for as the temperature falls it will modify 
the composition of these inclusions just as it modifies the composi- 
tion of its own crystals. Indeed this case may be regarded as a 
special case of that just discussed, for, as the temperature falls, the 
composition of the liquid changes, and the inclusions then pass into 
the class of those considered above. 

We now come to the case of inclusions of material later in the 
reaction series than the crystals with which the liquid is saturated. 
It should be noted that this includes masses of rock of the same com- 
position as the liquid itself, for example its own chilled border phase. 
Saturated basaltic magma can react with inclusions of igneous 
rocks later in the reaction series (in general more acid) in such a way 
that the inclusions become part of the liquid, crystals of the phases 
with which the basalt is saturated being precipitated at the same 
time. If these crystals are removed by gravity or otherwise the 
action on the inclusions may continue, the liquid changing in com- 
position toward the composition of the inclusions and precipitating 
later and later members of the reaction series until finally it is satu- 
rated with precisely the crystalline phases contained in the inclu- 
sions. If granitic inclusions, say, were available at the upper 
contact of a mass of basaltic magma, they would be attacked by the 


542 N. L. BOWEN 


magma in the manner noted and, in a lower layer, accumulation of 
the precipitated products of the reaction would take place. These 
would be the early crystals formed in basaltic magma. The upper 
liquid is thus gradually changed in composition and the crystals 
precipitated from it are successively later and later members of the 
reaction series. Attack upon the inclusions continues until finally 
the upper liquid becomes granitic. All of this depends on a rate of 
cooling slow enough for free crystal settling to occur. But if the 
cooling is sufficiently slow for crystal settling all of these results 
could accrue from the simple differentiation of the basaltic magma. 
Indeed the principles developed show that the inclusions can become 
part of the liquid only when they have a composition toward which 
the composition of the liquid can vary by spontaneous differentia- 
tion. 


tiate might be greatly augmented by this action. It may safely be 
assumed therefore that in many individual cases considerable quan- 
titative importance in the production of a granitic differentiate of 
basic magma is to be assigned to the action noted. It is a sort of 
solution of granitic inclusions though not a simple, direct solution 
and is in no sense essential to the production of a granitic differen- 
tiate. 

Daly is of the opinion that many granites are secondary, that is, 
are formed by solution of granite in basaltic magma and subsequent 
differentiation. It is seen from the above that theoretical consider- 
ations support belief in a process which, in its results at least, is 
practically that advocated by Daly. The process itself he considers 
to be rather a simple solution of granite in superheated basaltic 
magma. We have seen that no superheat is necessary to produce 
solution by a sort of reactive process. Moreover, we have seen that 
the incorporated granitic material is to be regarded rather as a 
contribution to the normal granitic differentiate. There appears, 
however, to be no reason to doubt that, at times, this contribution 
might equal or possibly even exceed in amount the granitic differ- 
entiate capable of formation from the uncontaminated magma. 


«Indeed, Daly derives in this manner all granites except a supposed original 
granitic shell of the earth (Igneous Rocks and Their Origin, p. 323). 


Nevertheless it is apparent that the amount of granitic differen- 


——— a 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 543 


The limiting factors are principally mechanical rather than thermal 
or chemical and are very difficult to evaluate. It should be noted, 
in particular, that the combination which is most favorable for sig- 
nificant effects in the way of reactive solution, viz., decidedly acid 
inclusions and decidedly basic magma, is unfavorable in another 
respect. The inclusions will be lighter than the magma and will not 
tend to sink in it, whereas it is the sinking of inclusions through the 
magma which favors particularly notable reaction effects since it 
continually brings new magma into contact with the inclusions. 

As an example of the effect of basic magma on more acid 
igneous inclusions basaltic magma and granitic inclusions have 
been taken. Between such extremes the more marked effects 
should be obtained, but it cannot be doubled that any basic magma 
can dissolve, by the same reactive process, inclusions of a rock later 
in the reaction (crystallization) series. Direct melting of granitic 
inclusions to masses of liquid by basaltic magma is not ordinarily 
to be expected because the solid granite does not retain the volatile 
components that aid in lowering the melting temperature of granitic 
magma below that of basaltic magma. This lack is no bar to the 
reactive solution process described, though it may limit it some- 
what. 


EFFECTS OF MAGMA ON INCLUSIONS OF SEDIMENTARY ORIGIN 


The general problem of the effects of magma upon inclusions 
of sedimentary origin is much more difficult than the similar prob- 
lem in connection with igneous inclusions. Sedimentary rocks have 
their compositions determined by processes wholly independent of 
igneous action and do not correspond in composition with the prod- 
ucts precipitated from magmas at any stage of their career, that 
is, cannot be placed definitely in the reaction series. However, 
certain minerals that can be formed in magmas do occur in the sedi- 
mentary rocks and often the composition of a sediment is such that 
by mere heating it can be transformed into an aggregate made up 
exclusively or almost exclusively of igneous rock minerals. Again 
sediments exhibit extremes of composition, being very rich in cal- 
cium carbonate, aluminum silicate or silica itself, and these present 
a special problem. Yet it is perhaps not generally realized how 


544 N. L. BOWEN 


much even of these extreme sediments might be incorporated in an 
igneous rock without changing its mineralogy. The fact is an 
obvious deduction from the equilibrium diagram of any investigated 
three component system and it is equally true of a more complex 
system. Figure 9 shows the solid phases formed immediately upon 
complete consolidation of any mixture of CaO, Al,O, and SiO... A 
mixture of composition (A) consists, upon complete consolidation, of 
anorthite, wollastonite, and gehlenite, one third of each. One may 


$10, 


Av 205.5103 


LENG s = ————— 
Ca0 3Ca0.AL20, 50 00.3A.20, Ca0.Ac,0, 30a0.5A.,0, Ac,0, 


Fic. 9.—Diagram of the system CaO-Al,0,-SiO. showing phases formed upon 
complete consolidation. (After Rankin and Wright.) 


add to this mixture any amount of CaO up to about 15 per cent of 
itself, without changing the mineral composition of the consolidated 
product. Similar amounts of either Al,O, or SiO, might be added, 
the only change in all cases being in the relative amounts of the 
minerals, not in the kind of minerals. A certain amount of change 
in the order of separation of the minerals would be effected but the 
temperature of final consolidation, the composition of the final 
liquid and the possible differentiates that might be formed by frac- 
tional crystallization would in all cases remain as before. Only 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 545 


when amounts are added in excess of those mentioned will a 
new crystalline phase appear and new fractionation possibilities 
enter. 

There are, however, certain mixtures in the system that will 
immediately present a new phase upon addition of the slightest 
amount of CaO, ALO,, or SiO,. Such is the mixture (B) which 
consists on consolidation of anorthite and wollastonite, one half of 
each. The addition of either CaO or Al,O, will bring in the new 
phase, gehlenite, and of SiO, the new phase, tridymite, and each will 
change the course of crystallization. This is because the composi- 
tion chosen is a limiting case in the ternary system and is in reality 
of only two components; and the addition of, say, lime carries it 
out of the two component system. It will be noted that the mix- 
ture considered contains the three oxides, though of only two com- 
ponents, and on consolidation only two solid phases are formed. It 
might seem on first thought that this corresponds with the case of 
the natural magmas, for these usually form solid phases fewer in 
number than the oxides present. However, there is another factor, 
namely, solid solution, that may give rise to this peculiarity and 
we shall find that it is to solid solution that the limited number of 
phases formed from magmas is usually to be referred. 

Let us now examine a ternary system of oxides that has been 
completely investigated and in which the factor of solid solution 
enters. Such is the system CaO:MgO:SiO, studied by Ferguson 
and Merwin.’ A liquid of composition A, Figure 10, forms on com- 
plete consolidation just two solid phases, the olivine, forsterite, and 
clinopyroxene of composition between diopside and MgsSiO,;. To 
this liquid any amount of calcite up to 12 per cent of its weight 
could be added without changing the mineralogy of the consolidated 
product. This would still consist of olivine and clinopyroxene but 
the pyroxene would be richer in CaO, that is, closer to diopside. 
Only an amount of calcite in excess of 12 per cent would bring in 
another phase, akermanite, together with the diopside and olivine. 
It is easy to see from an inspection of the figure that addition of 
dolomite would likewise have no effect on the kind of phases crystal- 
lizing unless more than about 16 per cent were added. 

t Amer. Jour. Sci., Vol. XLVIII (1919), p. r09. 


546 N. L. BOWEN 


When we have spoken of adding calcite and dolomite to the mix- 
tures mentioned we have imagined that the lime and magnesia 
have been taken into solution as they might be in a laboratory fur- 
nace where the furnace supplied the requisite quantity of heat. 
This quantity would be very large, for the conversion of carbonate 
into silicate is an endothermic reaction and its conversion into sili- 
cate in solution is undoubtedly still more strongly endothermic. ~ 
Now if the liquid were originally a saturated liquid and calcite or 


5x0, 
1710 


Crastobalte, 


iW 400 
2Q0-SGO-251 
eee: 


a) 70 
2570 — Bounaores : —Jsotherms determined; —--Isotherms hy extrapolation; © Compounds ; 2800 


+++,ZZZ Solid solutions. 
Fic. 10.—Equilibrium diagram of the system CaO-MgO-SiO,. (After Ferguson 
and Merwin.) 


dolomite were added without any provision for additional heat 
other than that already contained in the liquid, the lime and mag- 
nesia would not be simply dissolved. Instead, precipitation of 
olivine (forsterite) would occur, no matter whether calcite or dolo- 
mite was added, and as the forsterite separated the liquid would 
attack the calcite, converting it into lime silicates. Since this is an 
endothermic reaction the heat for it can be supplied only by crystal- 
lization of something from the liquid which will be, of course, the 
phase with which it is saturated, viz., forsterite. There is another 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 547 


reason why forsterite is precipitated for the formation of lime sili- 
cates requires some silica but the liquid cannot be desilicated since 
it lies on the forsterite saturation surface. Therefore the liquid 
changes its composition by moving on the forsterite surface toward 
lower temperature. It does not move directly away from forsterite, 
however, but curves somewhat in the direction of the composition 
of the inclusions. If the reaction with the inclusions is strongly 
endothermic, as it is in this case, the liquid would move down to 
the forsterite-pyroxene boundary curve where separation of pyrox- 
ene would occur and the liquid would be entirely used up without 
ever getting over to such compositions that akermanite would 
separate from it. However, some part of the inclusions might have 
been converted to akermanite. Thus, however large a supply of 
inclusions were available, even in excess of the 12 per cent mentioned 
above, the liquid might never get over to such compositions that — 
any new phase is precipitated from it, but would precipitate only 
forsterite and pyroxene as in normal crystallization, with, however, 
a certain increase in the lime content of the pyroxene. 
These considerations lead us to the conclusion that the liquid 
we have chosen, even if it has a moderate amount of superheat and 
therefore is capable of directly dissolving a little lime and magnesia 
from calcite or dolomite, does not suffer a change in the kind of 
solid phases capable of forming from it. There results only a modi- 
fication of the composition of the phase of variable composition 

(pyroxene). And, as a consequence of the heat effect of the solu- 
tion of inclusions, saturation shortly ensues and thereafter further 
action upon the inclusions is accomplished only with concomitant 
precipitation of the phase or phases with which the liquid is satu- 
rated, whereby the liquid is constrained to follow a general course 
not significantly different from the one it would follow were no 
inclusions present. 

The results obtained in the foregoing enable us to draw certain 
conclusions as to the effects of natural magmas upon inclusions of 
various sedimentary rocks. One point that does not seem to be 
realized is that when a sedimentary inclusion becomes immersed in 
a magma nothing is added that the magma does not already contain. 
Both belong to the same polycomponent system embracing all the 


548 N. L. BOWEN 


rock-forming oxides. Obviously the effects of all possible sedi- 
ments cannot be examined, but our purpose will be served if we take 
the most extreme departure from igneous composition. As repre- 
sentative of this condition, for quartzite we may imagine the addi- 
tion of pure quartz; for limestone, of pure calcium carbonate; and 
for shale, of pure kaolin. Any actual sediment would usually con- 
tain all of these, together with other constituents that lessen its 
departure from igneous composition. 

Let us take a magma of basaltic composition which, on crystal- 
lization with comparatively rapid cooling, would form mainly plagio- 
clase, and clinopyroxene, with some olivine, a little ore mineral and 
possibly some orthopyroxene. All of these are minerals of variable 
composition; some of them, in particular the pyroxene, vary with 
respect to several components and to this is to be attributed the 
fact that the number of solid phases formed is less than the number 
of oxides present. This fact permits particularly wide adjustments 
in the composition of the solid phases without the appearance of 
new ones. Such basaltic magma, with a little superheat, could 
directly dissolve a moderate amount of sediments, yet even if these 
were of extreme composition the magma would crystallize with the 
production of the same solid phases as those mentioned above if 
crystallized under the same conditions. 

Normally only saturated magma would be available and the 
superheated magma mentioned above would rapidly become satu- 
rated as a result of solution of inclusions. For the case of such 
saturated magma it may be stated as a first principle that the sedi- 
ment would, in so far as its composition permitted, tend to be con- 
verted into the phases with which the magma is saturated. And 
the material necessary for such changes in the sediment would not 
be merely subtracted from the liquid but adjustments of the com- 
position of the liquid would occur through separation of further 
amounts of the phases with which the liquid is saturated. 

The precise changes in the composition of the solid phases formed 
cannot be represented graphically on account of the number of 
components involved, but equations can be written that afford a 
generalized conception of the possible adjustments for the addition 
of calcite, silica, and kaolin respectively. 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 549 


The phases capable of formation from the original unchanged 


Magma are: 
Phase Mineral molecules represented 
Olivine Mg,si0,+ Fe.Si0, 
Magnetite FeO. FeO; 
Plagioclase CaAL,Si,0Os+ etc. 
Pyroxene CaMgsi.0c+ MgSiO,+ALO,*+ CaFeSi.06+- 


FeSiO,+ Fe.0,;* 


* Often written as existing in the Tschermak molecule, for which there is no good reason. See 
Washington and Merwin, Amer. Jour. Sci., Vol. III (1922), p. 121. 


Upon addition of CaO the following principal adjustments in the 
proportions of these mineral molecules may occur without the 
appearance of new phases: 

CaO- 2FeSiO,+ Fe.0,= CaFeSi,0¢-+ FeO. Fe.0,; 

CaO+3MgsiO,;=CaMgsSi,0¢+ Mg.Si0,; 

CaO+AlL0;+-4MgsiO,= CaALSi,0s+ 2Mg,si0,. 


Upon addition of SiO,: 


SiO.-+ Mg,SiO,= 2Mgsi0,; 
Si0.+ AlLO,+ CaMgSi,Os= CaALSi.Os+ MgSi0,. 


Upon addition of kaolin, which we may regard as Al,SiO;+Si0., 
with SiO, having the same effect as above: 


ALSiO,-+ CaMgSi,0s= CaALSi,0:-+Megsi0,. 


The results may be put in words by stating that addition of lime 
tends to increase the amount of magnetite and olivine, to make the 
pyroxene more nearly a pure diopside-hedenbergite and to increase 
the anorthite content of the plagioclase. The addition of silica 
tends to decrease the amount of olivine and to increase the mag- 
nesian content of the pyroxene and the anorthite content of the 
plagioclase. The addition of kaolin tends to increase the amount of 
magnesia in the pyroxene and of the anorthite in the plagioclase. 
In the case of superheated magma the added material might be 
directly dissolved and upon solidification the adjustments noted 
would appear in the crystalline phases. In the case of saturated 
magma the phases noted would be developed by reaction with the 
added material and at the same time a further amount of them 
would be precipitated from the liquid. In neither case would the 
course of crystallization be fundamentally changed since crystalliza- 


550 N. L. BOWEN 


tion produces only the same solid phases slightly modified in com- 
position. If consolidation took place under conditions permitting 
fractionation by settling of crystals no fundamentally new differ- 
entiation potentialities would be introduced by the solution or reac- 
tion with foreign material that has been discussed. The magma 
thus modified could give a diorite-granodiorite-granite sequence, 
say, only if the original magma could also have done so under the 
same conditions. It is a question whether, in the case of unsatu- 
rated magma, it can be safely assumed that the degree of superheat 
may be such that an amount of material can be dissolved in excess 
of that which can be taken care of by the adjustments in composi- 
tion of existent phases. If this is possible new phases will appear 
and the course of crystallization and differentiation might be funda- 
mentally modified. In the case of added CaO, for example, the new 
phase melilite might appear and the differentiates formed might be 
fundamentally different; might be, say, certain alkaline types as 
Daly has postulated. 


EFFECTS OF BASALTIC MAGMA ON ALUMINOUS SEDIMENTS 


With this possible exception in the event of excessive superheat, 
the statement should hold that reaction with foreign material can 
produce no new differentiation potentialities in the magma. Yet it 
is certain that some lines of differentiation may be emphasized by 
such agency, that is, that certain types of differentiate should be 
quantitatively of greater importance. This we have found to be 
true of reaction of magma with previously solidified igneous material. 
Thus when basaltic magma reacts with granitic material the tend- 
ency is to increase the amount of granitic differentiate capable of 
forming from the basalt. The effect of reaction with aluminous 
sediments is of sufficient importance to justify further discussion at 
this point. The result has been shown to be the emphasizing of 
more magnesian pyroxene and of anorthite. Now under certain 
conditions, not well understood, magnesian pryoxene separates from 
magmas as a distinct phase, an orthopyroxene, and the addition of 
aluminous sediments should emphasize this tendency. ‘The forma- 
tion of norite and of pyroxenites characterized by orthopyroxene as 
differentiates from basaltic magma, may therefore be facilitated by 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 551 


reaction of such magma with aluminous sediments. A suggestion 
along similar lines has already been made by Evans." 

An example of the reaction of basic magma with aluminous sedi- 
ments, in which the relations discussed seem particularly clear, is 
afforded by the so-called Cortlandt series.2 The early work of 
Williams and the later work of Rogers on this series treats in some 
detail the features bearing upon the question here at issue. The 
Cortlandt series is intrusive into the Manhattan schist (locally into 


TABLE III 
I Il 

Si0,. 57-94 40.16 
JENN AO}; a dein chrfio. hence Bae mer eg 21.70 29.50 
IDSA O Fis atts ten Giclees me Ree eee none 1S 19.66 
Ee OM ete payee eee ee a: 5-90 5.80 
IVES OPIS R seat ere timer muna ett, 2.49 trace 
CaO ees sea ean kot i lee 58 85 
TNIV OD shaun esi ie ee A cso en 1.74 1.46 
Re Oe vse Hewat rn ga) lal 4.68 | 1.36 
H.O+ eee ehce eenstwer aistioveheitsteyarisas = AotW Wodosasonveace 
TEL OS ly eects of Mae eee eee DOr al Pha emramnan oes 
INO ale Boe eRe ete eee iTMOLE A haelhenes Stat eva ekese ate 
Ming ORs irarici Se ana ait TG AWA | aes Seto Biche 
Ss dici.c1 0° 6 SO lore 2 Sante Pree | eee oR PA .82 

100. 26 99.61 


I. Manhattan schist. Composite analysis of five specimens be- 
yond border of the Cortlandt series. Analyst, G. S. Rogers. 
If. Manhattan schist on contact of Cortlandt series. Analyst, F. 
L. Nason in G. H. Williams, Amer. Jour. Sci. (3), Vol. XXXVI (1888), 
Dp. 259. 
other formations also) and the interaction between schist and 
magma is in places rather well displayed. The Manhattan schist 
is a metamorphosed sedimentary rock of the nature of a shale. Its 
composition, as given by a composite analysis of five specimens, is 
shown in Table III under I. No doubt it varies considerably from 
this average and is sometimes more aluminous, but it is always far 
from the composition of kaolin, which composition we have used 
tJ. W. Evans in discussion of paper by C. E. Tilley, Quar. Jour. Geol. Soc., Vol. 
LXXVII, p. 133. 


2G. H. Williams, Amer. Jour. Sci. (3), Vol. XXXI (1886), p. 26; (3), Vol. XX XIII 
(1887), pp. 135, 191; (3), Vol. XXXV (1888), p. 438; (3), Vol. XXXV (1888), p. 254. 


3G. S. Rogers, Ann. N.Y. Acad. Sci.,-Vol. XXI (1911), pp. 11-86. 


552 N. L. BOWEN 


in discussing the effects of basic magma on aluminous sediments. 
This affords an opportunity of discussing the behavior of an actual 
example of aluminous sediment. The ordinary Manhattan schist 
is made up principally of the minerals quartz, biotite, muscovite, 
orthoclase and plagioclase. ‘These are all minerals of ordinary 
igneous rocks, particularly of more “acid” types, and correspond to 
a rather low temperature equilibrium. A glance at the analysis 
shows, however, that the composition is far from that of an ordinary 
igneous rock, which means that the minerals are of somewhat differ- 
ent composition and are present in different proportions. Now we 
have found in our discussion of the reaction of.any saturated magma 
upon igneous inclusions that if the inclusions belong to a later stage 
in the reaction series they may become a part of the liquid by caus- 
ing the precipitation of the phases with which the liquid is saturated. 
Average Manhattan schist, since it consists of the minerals of an 
acidic igneous rock, may be regarded as consisting in part of material 
belonging to a later stage in the reaction series than basaltic magma, 
but since it does not correspond exactly with any such igneous mass 
it must be regarded as having a certain amount of surplus material 
in addition. If we imagine saturated basaltic magma reacting with 
inclusions or wall rock of schist we may expect the action to be 
selective. Such substances as may become a part of the liquid 
would be removed from inclusions or wall rock with corresponding 
enrichment in what has been called surplus material. The sub- 
stances removed would be principally silica, alumina, alkalis and 
to a minor extent other oxides, all in the proportions in which they 
enter into some “acid” igneous rock. Our knowledge of the exact 
proportions may be thus indefinite and yet sufficient for a. general 
solution of the problem. Comparison of the analysis of the Man- 
hattan schist with those of acid igneous rocks gives us a good con- 
ception of what the surplus material will be. It will plainly be 
rich in alumina and iron. A certain stage of the reaction between 
magma and inclusions or wall rock should exhibit a mass rich in 
these oxides. This stage is abundantly represented in both wall 
rock and inclusions by richness in sillimanite, staurolite and other 
aluminous and ferrous minerals. Chemically it is shown by analysis 
II in Table III which represents wall rock at the margin of the 
intrusive. 


——— ee 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 553 


The so-called surplus material does not remain indefinitely as 
such, but by the time we have obtained inclusions very rich in sil- 
limanite a turning point in the process is reached. Hitherto certain 
constituents of the schist have become a part of the liquid in virtue 
of the precipitation of various phases with which the liquid (magma) 
was saturated. Now reactions between magma and inclusions be- 
come of such a nature that the precipitation of phases with which 
the magma is saturated is the sole process, these phases being appro- 
priately modified by the inclusions. There is now no addition to 
the liquid. The exact modification of the phases that is produced 
by sillimanite has already been discussed and equations representing 
the changes have been written. ‘The net result is an increase in 
the amount of anorthite and magnesian pyroxene at the expense of 
lime-bearing pyroxene, and this tends to promote the separation of 
magnesian pyroxene as a distinct phase, orthopyroxene. Thus the 
tendency of the magma to give a noritic differentiate is increased, 
as well as the likelihood of formation of a pyroxenite containing 
orthopyroxene. ‘These expectations are well matched by the Cort- 
landt series. 

Apart from possible later differentiation this production of 
norite and related types is the end result of the action of basic 
(basaltic) magma on sillimanite-rich inclusions. We may with 
profit examine the details of the action, that is, the processes going 
on within and immediately around the sillimanite-rich inclusions. 
immersed in a magma rich in plagioclase and pyroxene. This exam- 
ination serves to throw some light on the detailed mineralogy of the 
inclusions, and in particular on the separation of free alumina, as 
corundum. 

If a mass of sillimanite were added to some anorthite just above 
its melting point, it can be readily seen from examination of Figure 
11 that some of the sillimanite would be converted into corundum. 
This is because the line joining sillimanite and anorthite passes 
through the corundum field. We may imagine, for example, that 
the bulk composition of a layer immediately surrounding the silli- 
manite is 50 per cent sillimanite and 50 per cent anorthite. The 
mixture represented by this layer, at 1550°, would consist of sillaman- 
ite, corundum, and liquid. [If the original anorthite liquid had an 


554 N. L. BOWEN 


excess of silica amounting to about to per cent, no corundum would 
be formed from sillimanite, because the join of sillimanite with such 
a composition misses the corundum field. Plainly the freeing of 
corundum from sillimanite depends in these liquids on the liquid 
being “‘basic,” that is, having not more than a moderate excess of 
silica over the feldspar composition. Free silica associated with 


609 


Cristodalite 


cao 


3C40.Al203 5CB0,Alz05 
70 1455 


Al20q 


€a0,Al203 
1600 2050 


3C30.5Al205 
1720 


Fic. 11 —Equilibrium diagram of the system CaO-AlLO,-SiO,. (After Rankin 
and Wright.) ; 


the sillimanite does not have a comparable effect in restricting the 
formation of corundum; indeed, the silica would require to be 
nearly equal in amount to the sillimanite in order to neutralize the 
tendency to form corundum. 

The system MgO:Al,0,:SiO, given in Figure 12 shows the same 
general condition. If a mass of sillimanite were immersed in 
molten MgSiO, and Mg,SiO, in equal parts, a layer around the 
inclusion, which might have the bulk composition 75 per cent silli- 
manite, 25 per cent the above mixture, would consist at higher 
temperatures of corundum and liquid; at lower temperatures, of 


ee 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 555 


sillimanite, spinel, and liquid; and at still lower temperatures, of 
sillimanite, spinel, and cordierite. 

It should be realized that these conditions we have pictured as 
occurring about sillimanite inclusions are transient states. We may 
deduce from an equilibrium diagram the condition of a certain layer 
about an inclusion, but the system as a whole is not in equilibrium 


~L 


MgO MgO-Al203 Alp! 
2ECIo) ons ES 


Fic. 12.—Equilibrium diagram of the system MgO-Al,0,-SiO.. (After Rankin 
and Merwin.) 


since there are composition gradients about these inclusions. 
Nevertheless this changing state might become fixed and be revealed 
as a result of complete consolidation of the mass. 

The actual magma by which the schist inclusions were attacked 
may be regarded as showing the combined effects of the anorthite 
liquid and the magnesian liquid of the experimental systems. It is 
true the magma did not correspond definitely with any such mix- 
ture of anorthite and magnesian silicate, but it was closely related, 
consisting mainly of plagioclase and magnesian silicates, and its 
effect on sillimanite-rich inclusions might reasonably be similar. 


550 N. L. BOWEN 


The effect was indeed very similar, for the sillimanite-rich inclusions 
are found to be changed by the magma to masses rich in corundum, 
spinel, and cordierite, with other related minerals. It is inclusions 
that have been thus affected that constitute the emery deposits. 
They represent the reaction between the magma and inclusions 
arrested midway. No doubt many inclusions completely disap- 
peared, becoming an integral part of the igneous rock in virtue of 
reactions involving adjustment of composition of existing phases. 
(See p. 549.) The feldspathic emery and the noritic emery may, 
from this point of view, be regarded as inclusions approaching their 
final disappearance. 

These transient states in which inclusions may be very rich in 
certain substances are no doubt of some importance in differentia- 
tion. Localized masses in process of reaction with the general mass 
may move, say in response to gravity, and their accumulation may 
give rise to bodies rich in minerals formed during the reactions. 
All up and down the Appalachian Mountain system of Eastern 
North America there are intrusive masses showing ultrabasic dil- 
ferentiates with dunite as the extreme and with associated perido- 
tites and pyroxenites frequently rich in rhombic pyroxene, saxonite, 
websterite and enstatolite itself. These are usually, perhaps always, 
intrusive into slates and mica schists that were originally aluminous 
sediments, the Farnham slates of Quebec, the Savoy and Rowe 
schists of New England, the Manhattan of New York, the Wissa- 
hickon of Pennsylvania and Maryland, and the Carolina gneiss of 
the Southern States. These have perhaps had an important influ- 
ence in emphasizing differentiates of the types mentioned above. 
There are, moreover, corundum deposits, either as emery or in 
purer forms, in frequent association with these ultra-basic rocks 
and in some cases the origin and accumulation of corundum may be 
referred to processes outlined above. Gordon has demonstrated a 
different origin for some of them," but the action here described 
seems unquestionable for the Cortlandt emery and the gangue 
minerals of some of the Carolina deposits strongly suggest a similar 
origin. ‘These gangue minerals are basic plagioclase, sillimanite and 


«Formation by reaction between pegmatite and serpentine. S. G. Gordon, 
Proc. Acad. Natural Sci. Philadelphia, Part I (1921), p. 160. 


——— 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 557 


cyanite. That such minerals together with corundum could form 
from basic magnesian rocks by simple differentiation is very doubt- 
ful. 

The progressively selective nature of the reaction between 
magma and inclusions that is shown by these Cortlandt examples 
and the theoretical considerations discussed in connection therewith 
demonstrate that a former statement of mine that “the formation 
of an obviously hybrid rock should, apparently, be the normal result 
of assimilation’ is erroneous and that Daly’s objections thereto are 
justified. Daly’s objections are based on considerations other than 
those raised above and refer rather to the changes that may occur in 
a superheated magma during the slow diffusion into it of xenolithic 
material. We may recall here that the reactions herein described 
are those that are to be expected between saturated magma and 
schist inclusions. ‘The fact that the reactions are selective in a way 
that matches the expectation may presumably be regarded as evi- 
dence that the magma was saturated. 

Recognition of the probable saturated condition of the magma 
is of importance because it shows that there could not have been 
formed, at any time, a liquid whose composition was simply the 
sum of that of the original magma and the inclusions. Even that 
portion of the foreign matter which becomes a part of the liquid does 
so only by precipitating phases with which the magma is saturated 
and must itself be of a composition toward which the liquid may go 
spontaneously by fractional crystallization. So in the case of the 
Cortlandt series various diorites, some syenite and a considerable 
amount of granite (first recognized by Berkey as a part of the series)‘ 
were formed by differentiation as they would have been under the 
same conditions without reaction with slate material, though pre- 
sumably the amount of “acid”’ differentiates was augmented by its 
addition. 

Another suggestion concerning the incorporation of slaty rocks 
in basic magmas may be made at this point. In the neighboring 


t See Rankin and Merwin, “The Ternary System MgO-AlL0,-SiO.,” Amer. Jour. 
Sct., Vol. XLV (1918), p. 325. 

2,N. L. Bowen, Jour. Geol., Supplement to Vol. XXIII (1915s), p. 85. 

3R. A. Daly, Jour. Geol., Vol. XXIII (1918), p. 126. 

4C. P. Berkey, Science, Vol. XXVIII (1908), p. 575. 


558 N. L. BOWEN 


state of Connecticut there occurs a group of rocks showing a striking 
resemblance to the Cortlandt series.‘ With these are associated 
nickeliferous sulphide deposits of magmatic origin. This associa- 
tion is highly characteristic, examples of norite and related rocks 
with nickeliferous sulphides being too familiar to require special 
enumeration. It seems possible that the incorporation of slaty 
rocks may be of importance not only in connection with the forma- 
tion of norite but also of the sulphide deposits, for slaty rocks are 
very commonly pyritic. At first thought it might seem that the 
addition of pyrite to a basic magma would not account for the 
separation from it of the sulphides commonly found in such nickel 
deposits. However, if pyrite, say as globules of immiscible liquid, 
remained long immersed in a basic magma it would be subject to the 
same kind of modification of composition as is any other foreign 
matter. In short, it would have its composition changed to such 
sulphides as are particularly insoluble? in such magma. Thus if 
iron, nickel, copper sulphides could separate from such a magma 
by normal processes, were they present in sufficient amount, then 
immersed pyrite would be changed to just such iron, nickel, copper 
sulphides. It seems worthy of consideration, therefore, that the 
sulphide deposits associated with norites may often be reconstituted 
deposits having their ultimate origin in the incorporation of pyritic 
slates. Indeed at Sohland and Schweidrich on the Saxony-Bohemia 
border typical ores of this kind are intimately associated with 
masses representing what we have termed the transient state of slate 
inclusions, that is, with masses rich in spinel, sillimanite and 
corundum. 

In connection with the formation of norite and perhaps of sul- 
phide deposits, through the influence of absorbed slates, it should be 
realized that the action is probably an emphasis upon normal pro- 
cesses. It may very well be, however, that if there were no argil- 
laceous sediments norites would be of much rarer occurrence than 
they are. 


*W. H. Hobbs, Festschrift v. H. Rosenbusch., pp. 25-48. Stuttgart, 1906. 

2 Ernest Howe, Economic Geol., Vol. X (1915), P- 330- 

3 Probably as liquid, i.e., immiscible. 

4 Beyschlag, Vogt, Krusch, Truscott, Ore Deposits, Vol. I, p. 299, London, 1914. 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS. 559 


THE ACTION OF BASIC MAGMAS ON SILICEOUS SEDIMENTS 


A number of examples of argillaceous quartzites are known that 
have been invaded by basaltic magma and in which it is believed by 
some investigators that absorption of the siliceous sediments by 
the magma has occurred. ‘Two carefully studied examples are the 
Pigeon Point sill of Minnesota‘ and the Moyie sills of British Colum- 
bia.2 Such quartzites may again be regarded as consisting of 
material belonging to a later stage of the reaction series than basaltic 
- magma, together with a certain surplus. The former, corresponding 
in composition with some acid igneous rock, should be capable of 
becoming part of the liquid magma by precipitating its heat equiva- 
lent of the phases with which the magma is saturated. There is no 
theoretical objection, therefore, to the belief that a certain amount 
of the inclusions could be incorporated in this manner even though 
the magma is saturated. It should be borne in mind, however, that 
_ the material that can thus become a part of the liquid must be of a 
composition toward which the magma could change spontaneously 
by fractional crystallization. Once incorporated, it requires, to 
produce the acid differentiates that were there formed, the same con- 
ditions of crystallization as would have produced an acid differen- 
tiate from the uncontaminated magma. In allof these examples the 
normal course of differentiation is the primary consideration. The 
extent to which incorporated material contributed to the bulk of 
the acid differentiate may not have been important in these small 
bodies even though there is plain evidence of incorporation. 

At both localities mentioned evidence of some incorporation is 
unquestionable. About inclusions of the sedimentary rocks reaction 
rims of a granitic nature have been formed. We have seen on page 
529 that the reaction should emphasize, in the liquid around the 
inclusion, material belonging to a later stage of the reaction series 
(i.e., toward which the magma can crystallize), and this should not 
be intermediate in composition between magma and inclusion. Cor- 
responding with this deduction we find that the rims about the xeno- 

tW. S. Bayley, U.S. Geol. Survey, Bull. tog. R. A. Daly, Amer. Jour. Sci., 
Vol. XLIII (1917), p. 423. 


2R. A. Daly, Geol. Survey Can., Mem. 38, p. 226. S.J. Schofield, Geol. Survey 
Can., Museum Bull. No. 2. 


560 N. L. BOWEN 


liths are not simply melted xenolith but essentially normal igneous 
material of a late stage of the reaction series. ‘The reaction-rim 
stage is a temporary one except in so far as it may be preserved 
about some inclusions by exhaustion of.the magma. Others dis- 
appear entirely by diffusion of the rim material into the magma and 
distribution (with possible precipitating effects) of any surplus 
material. Thus the liquid is pushed onward in the reaction series, 
not only through addition of the rim material, but also because this 
necessitates some precipitation of the early-formed minerals from 
the basic magma. Further fractional crystallization may therefore 
give differentiates identical with, or closely related to, the reaction- 
rim material, but normal differentiation might have given it also. 

Inclusions of some acid sediments, richly charged with volatile 
substances (mainly water) and immersed in a basic magma under 
conditions permitting their retention of this volatile matter, might, 
theoretically, be converted to liquid in toto. Actually, few inclu- 
sions show any such effect, from which fact it may be assumed that 
the conditions mentioned are seldom realized. That the reaction 
effect discussed is the important one quantitatively seéms unques- 
tionable. 


EFFECTS OF GRANITIC MAGMA ON INCLUSIONS OF 
SEDIMENTARY ORIGIN 


In discussing the reaction of magmas with inclusions we have, 
in the case of basaltic magma, made some reference to the super- 
heated condition. Daly points to basaltic magma as the heat 
bringer, and has presented evidence that such magma enters into 
igneous-rock economy on a different basis from all other magmas.* 
If this be true, and his reasons seem to me convincing, basaltic 
magma is the one magma that may, presumably, be assumed to have 
superheat on some occasions. All other magmas, whether they may 
be formed by differentiation of basaltic magma or by differentiation 
of syntectic magma must usually be saturated, unless it be that 
locally, at volcanic vents, a special source of heat is available. This 
possibility has little quantitative petrogenic significance and it is 
perhaps a realization of the commonly saturated condition of other 


tR. A. Daly, Igneous Rocks and Their Origin, p. 458. 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 561 


magmas that has led Daly to adopt basaltic magma as his solvent. 
Thus all of Daly’s syntectics are of basaltic magma with various 
types of foreign matter. We have already seen, however, that the 
saturated condition is no bar to a reaction between magma and 
inclusions. ‘This fact is as true of intermediate magmas as of any 
others and the same principles apply to them. 

If the foreign material belongs to an earlier stage of the reaction 
series the tendency is to make it over into those phases with which 
the magma is saturated and to precipitate a further amount of these 
phases from the magma itself. If the foreign material belongs to a 
later stage of the reaction series it tends to become a part of the 
liquid by precipitating phases with which the magma is saturated. 

Sediments do not belong in the reaction series at all and certain 
types of sediments contain material belonging in both the above 
classes and both effects may be obtained. Our chief purpose here is 
to consider principles, and it seems unnecessary, therefore, to dis- 
cuss individually the action of various intermediate magmas on 
various foreign inclusions. 

It is perhaps desirable, however, to discuss the action of granitic 
magma on sediments, its action on igneous inclusions having already 
been described. For the reason mentioned above, only saturated 
granitic magma will be considered. Quartzites and slaty rocks offer 
no special problem. They are readily transformed into phases with 
which the granite is saturated, an action that any magma will 
accomplish in so far as the composition of the sediment permits. 
The conversion of inclusions of such rocks by granitic magma into 
masses of quartz, feldspars, and micas, in varying proportions, 
should therefore be the result. A certain amount of mechanical dis- 
integration might cause the strewing about of these products in such 
a way as to make them an integral part of the mass but there should 
be no solution. Intermediate steps might see the formation of such 
minerals as sillimanite, garnet, and others characteristic of contact 
rocks, but these should be temporary or should survive only because 
of exhaustion of the liquid. 

The kind of effect that sillimanite produces by reaction with 
basic magmas, namely, the precipitation of orthopyroxene and basic 
plagioclase is not to be expected in graniticmagma. Rather should 


562 N. L. BOWEN 


we expect precipitation of the micas in acid magmas, and formation 
of orthopyroxene in such magmas is to be referred to other causes 
than that proposed by Evans' and here adopted for basic rocks. 

When we turn to the case of carbonate rocks we find that the 
reaction with granitic magma is of a different nature. It is often 
observed that wall rock and inclusions of carbonates are altered to 
silicate minerals.*? It has been assumed by some investigators, 
therefore, that silica has been subtracted from the granitic liquid 
and that this may occur to such an extent that some of the feldspar 
molecules are transformed into the less siliceous, feldspathoid mole- 
cules with consequent formation of alkaline rocks. This assumed 
action is said by certain writers to be in agreement with Daly’s 
theory of the origin of alkaline rocks. We have seen above, how- 
ever, that Daly assumes that superheated basaltic magma is the 
starting point for all his syntectic magmas and that alkaline rocks 
are differentiates of some of these syntectics, principally those 
formed with carbonate rocks. We have pointed out on a preceding 
page that, if adequately superheated basalt were available, it might 
form, by solution of carbonates, a melilite basalt and, given the 
latter, alkaline differentiates seem not impossible and so indeed 
some alkaline rocks may be formed. Not all alkaline rocks can be 
so explained, for much nephelite syenite shows intimate genetic 
relations with granites and on Daly’s general theory the original 
basaltic magma would require to be silicated by solution of acid 
material to form the granite and desilicated by solution of carbon- 
ates to form the nephelite syenite. That the differentiates should 
show evidence of both seems out of the question. The solution 
of foreign matter must result in either desilication or silication 
according to the preponderance of one or the other type of foreign 
matter and subsequent differentiates should be in conformance 
with one or the other but not both. 

This brings us back to the question whether a nephelite syenite, 
intimately associated with a granite, could have been formed as a 
result of desilication of the granite by carbonate inclusions, that is, 


1J. W. Evans, Quar. Jour. Geol. Soc., Vol. XLVIILI (1921), p. 133- 
2 For example in the large-scale production of amphibolites in the Haliburton- 
Bancroft area. Adams and Barlow, Can. Geol. Surv., Mem. No. 6 (1910). 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 563 


back to the question of the effect of granitic magma on carbonate 
inclusions. 

We have noted the silication of the inclusions and the consequent 
supposed desilication of the liquid. In discussing investigated sys- 
tems we have already found, however, that saturated liquids cannot 
have ingredients subtracted from them at random without causing 
precipitation of other ingredients so that the effect on the liquid is 
not mere impoverishment in the ingredient subtracted. This may 
perhaps be made clear by a simple example. If we had a solution 
of salt at — 20° C, and any hypothetical substance was placed in the 
solution that withdrew the salt from it, the result would not be sim- 
ply the leaving behind of liquid water. The reason is simply 
that liquid water cannot exist at —20° C and the actual result 
would be that, as each small amount of salt was removed, a small 
amount of ice would form and, when all the salt was withdrawn, all 
of the water would have become ice. For the maintenance of 
liquidity, the salt and the water are necessary each to the other. 
And so it must be with a saturated granitic solution. Remove the 
silica from it and other substances must be precipitated. Now the 
reaction of granitic magma with inclusions of carbonate rock is not 
a simple addition of silica to the latter but usually other substances 
are added as well, these being such as to convert the inclusions 
ultimately into diopsidic pyroxene or hornblende. The reason for 
the formation of these phases is that they belong at an earlier stage 
of the reaction series than the biotite with which granitic magma 
is normally saturated. The subtraction of the substances necessary 
to produce these minerals must, for reasons outlined above, cause 
concomitant precipitation of the other phases normally formed from 
granitic magma, principally feldspar. Thus the action described 
must bring about an exhaustion of the liquid by causing precipita- 
tion. ‘There seems to be no reason for believing that it could first 
exhaust the free silica, leaving a feldspar-rich liquid, then, upon 
further action, cause removal of some of the silica from the feldspar 
liquid leaving a liquid containing feldspathoid molecules. 

A reaction of the kind described, that is, a using up of some silica 
to form diopside with the consequent precipitation of feldspar and 
quartz, would seem to me to be the mode of formation of the 


564 N. L. BOWEN 


diopside-bearing variety of the Beckett gneiss, of which Eskola has 
written a description and interpretation.t I visited the localities 
with Dr. Eskola, and in my opinion the transformation of the solid 
dolomite into solid diopside with its effect upon the granitic liquid 
was the dominant action in the production of the types of gneisses 
there found, and of their banding, rather than actual solution of 
the dolomite or skarn in the granite and subsequent differentiation of 
the syntectic liquid. It must be admitted, however, that the solu- 
bility of CaCO, in magmas is probably greater than that of CaO and 
that under conditions permitting the retention of CO, an amount of 
limestone might be dissolved greater than that suggested by the 
reaction effects already discussed. ‘The usual free conversion of 
limestone into silicates indicates, however, that it is not commonly 
so dissolved. 

The formation of basic silicates, without the production of feld- 
spathoids, seems to be the ordinary result of the action of granitic 
magma on limestone inclusions. ‘Thus, in the granitic portions of 
the Bushveldt laccolith altered limestone inclusions are surrounded 
by a halo of dioritic material, but not by alkaline rock.? Other 
examples might be given; in fact, the ordinary effects of granite on 
limestone seem to be those we have deduced for a saturated granitic 
magma. 

In one locality the alkaline facies of the Bushveldt complex is, 
it is true, intimately associated with a mass of limestone, and as a 
result of a study of this locality Shand has concluded that there is 
some connection between the production of the feldspathoids and 
the desilicating action of the magma. Apparently Shand does not 
believe that the entire production of nephelite is due to this action 
but rather that a nephelite syenite magma becomes ijolite by desili- 
cation. This is a quite different matter from the production of the 
original nephelite syenite magma by such desilication. No theoret- 
ical objection can be raised against the belief that interaction with 
limestone could reduce the amount of feldspar and increase the 


«P. Eskola, Jour. Geol., Vol. XXX (1922), pp. 265-94. 
2 Oral communication. Professor Brouwer. 
3S. J. Shand, Trans. Geol. Soc..South Africa, Vol. XXII (1921), pp. 144-46. 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 565 


amount of feldspathoid in a magma already capable of precipitating 
both of these. Such adjustment of the relative amounts of minerals 
we have found to be a common effect of inclusions. 

It is probable that alkaline rocks are ordinarily produced by 
crystallization differentiation from subalkaline magmas. A possible 
method in the case of a leucite-bearing rock has been demonstrated 
by Morey and Bowen, who show that a liquid of the composition of 
orthoclase or even one with a moderate excess of silica over the 
amount necessary to form orthoclase will precipitate leucite as the 
first-formed crystals. If an orthoclase-rich liquid came into being 
by fractional crystallization of a more basic magma it might, under 
the appropriate conditions, show the above effect. The fact that 
the excess silica must be no more than a small amount should be 
noted, for this fact renders it possible that limestone may, in spite of 
the many objections that have been raised above, have some influ- 
ence in promoting the formation of alkaline rocks. The influence is, 
however, an emphasizing of a normal tendency rather than a funda- 
mental necessity. ‘This we have found to be a general rule in con- 
nection with the effects of inclusions. If the differentiation of the 
magma which gave rise to the orthoclase-rich liquid took place in 
the presence of a supply of limestone inclusions this would tend to 
reduce to a minimum any excess silica that might otherwise be asso- 
ciated with the orthoclase. Thus the normal tendency of the ortho- 
clase to break down into leucite under the proper conditions would 
be free to assert itself. _Wemay therefore accept the possibility that 
reaction with limestone may emphasize the tendency toward the 
formation of an alkaline differentiate, though it is not essential to it. 
Other factors, such as the failure of olivine to form at an early stage 
in the magma’s history, or the free resorption of such olivine as does 
form, may also assure a low excess of silica at a late stage with likeli- 
hood of the separation of feldspathoids. The tendency of feld- 
spathoid to separate under these conditions has been demonstrated 
as yet only for leucite, but the frequent intimate association of 
leucitic and nephelitic rocks renders it probable that the factors 
- governing the formation of the nephelitic rocks are not unrelated. 
* Amer. Jour. Sci., Vol. IV (1922), pp. 1-21. 


566 N. L. BOWEN 


DEDUCTIONS TO BE COMPARED WITH OBSERVED RESULTS 


Throughout the foregoing study of the reactions between inclu- 
sions and magma, attention has been directed mainly to its theo- 
retical aspects, that is, to deducing from equilibrium considerations 
what reactions should occur, together with the effects of these upon 
the further crystallization of the magma. All of these deductions 
can be put to the test by observation of what has actually occurred, 
in particular by a study of the reaction rims formed about inclusions. 
It should not be expected that each inclusion will tell the whole 
story, but a general study of inclusions should do so. Not all the 
differentiates that might later form from the hybrid mass need be 
shown by the reaction rims, but certainly there should be formed 
some whose relationship to these possible later differentiates is 
established by their frequent association in many areas. 

In some instances examples have been cited which appear to 
show that the expected reactions do occur. Such are the formation 
of granitic reaction rims by the action of basaltic magma on acidic 
rocks, the making of basic inclusions into biotite-rich masses by 
granitic magma, and others. The formation of alkaline rocks by 
the action of ordinary magmas on limestones is, at present, incapable 
of support on the above grounds. No example is known where 
inclusions of limestone, contained in an ordinary rock, are sur- 
rounded by reaction rims of feldspathoid-bearing rock. It is true 
that limestones and alkaline rocks are often intimately associated, 
but there is no assurance that the magma was not already an 
alkaline magma before it acquired this association. As we have 
already pointed out, this appears to be the conclusion that Shand 
reaches concerning the Sekukuniland occurrence, though he favors 
also the conception that the limestone emphasized its alkaline 
nature. . 

In the Fen area of Norway, one of the newer areas to which the 
limestone-syntectic hypothesis has been applied, there is a very strik- 
ing association of alkaline rocks and carbonate rocks.‘ However, 
nothing there displayed demonstrates a change of subalkaline 
magma to alkaline magma through the influence of the carbonate 


«Cf. W. C. Brégger, ‘Die Eruptivgesteine des Kristianiagebietes IV,” Vid. Selsk. 
Skr. I. Mat. Naturv. Klasse (1920). 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 567 


rock. No support in the way of reaction rims of the appropriate 
kind has yet been found for the limestone-syntectic hypothesis. 
Some such support is desirable before the hypothesis can be ac- 
cepted, even though there is reason to believe, as pointed out above, 
that the presence of limestone might emphasize the normal ten- 
dency of magmas to give an alkaline differentiate. 


SUMMARY 


The question whether magmas can dissolve large quantities of 
foreign inclusions is one that has been much debated by petrologists. 
Some have claimed great powers for magmas in this respect and in 
addition have assigned a dominant réle in the production of differ- 
entiation to such solution of foreign matter. Others have insisted 
that magmas have not the necessary heat content to enable them to 
give significant effects of this kind. A study of some simple equi- 
librium diagrams, with the object of determining the heat effects 
connected with solution, gives every reason for believing that the 
effect is a large absorption of heat, usually of the order of magnitude 
of the latent heat of melting. For simple solution, then, it is 
unquestionable that large amounts of heat will be required. 

Those who believe in the actuality of the solution of considerable 
amounts of foreign matter in magmas have usually realized this 
fact and have sought a source of the heat in magmatic superheat 
of great amount, that is, in a large excess of temperature of the 
magma above its crystallization range. A study of the probabilities 
of the case and of the usual effects of magmas upon inclusions leaves 
little reason for believing that magmas can ordinarily have any 
considerable superheat. 

Unquestionably, then, the observed effects of magmas upon 
inclusions are usually to be referred to an action other than the direct 
solution of inclusions in superheated magma. An application of 
the conception of the reaction series to the solution of the problem 
affords an explanation of the effects of magmas, even though satu- 
rated. Certain principles governing the effects of liquid upon 
inclusions belonging to reaction series can be developed by studying 
the equilibrium diagrams of systems involving both continuous and 
discontinuous reaction series. In this manner it can be decided 


568 N. L. BOWEN 


definitely that a liquid saturated with a certain member of a reaction 
series is effectively supersaturated with all preceding members of 
that series. It cannot dissolve such members but can only react 
with them to convert them into the members with which it is satu- 
rated. The reaction is not a simple subtraction from the liquid 
of the material necessary for this transformation, but some precipi- 
tation from the liquid itself is involved and the liquid ordinarily 
maintains its position on the same saturation surface. The products 
of crystallization from the liquid and the possible course of frac- 
tional crystallization are thus unaffected. 

On the other hand, a liquid saturated with a certain member of a 
reaction series is unsaturated with all subsequent members of the 
series. Inclusions consisting of these later members can become a 
part of the liquid by a sort of reactive solution, the heat of solution © 
of inclusions being supplied by the precipitation of their heat equiv- 
alent of the member of the series with which the liquid is saturated. 
It should be noted that the material that can by this reactive process 
become a part of the liquid must consist of a later member of the 
reaction series, that is, must be material toward which the liquid 
could pass spontaneously by fractional crystallization. Thenet 
effect upon the liquid is, then, to push it onward upon its normal 
course. 

In Table II the products of crystallization of subalkaline mag- 
mas are arranged as reaction series, as definitely as may be in such 
complex series. The action of magmas upon foreign inclusions of 
igneous origin may be deduced from this arrangement of the crystal- 
line products as series by application of the principles developed 
from the above study of simple systems. Thus we find that a 
granitic magma saturated with biotite cannot dissolve olivine, 
pyroxene, or amphibole, but can only react with them to convert 
them into biotite, the phase with which it is saturated. Or, stated 
more generally, no saturated magma can dissolve inclusions con- 
sisting of minerals belonging to an earlier stage of the reaction series 
(usually more basic). 

Saturated basic magma, on the other andi will react with inclu- 
sions belonging to a later stage of the reaction series (more acidic), 
the reaction being of such a nature that the inclusions become a part 


THE BEHAVIOR OF INCLUSIONS IN IGNEOUS MAGMAS 569 


of the liquid by precipitating their heat equivalent of the phases 
with which the magma is saturated (basic minerals). . The inclu- 
sions, it should be noted, must be of a composition toward which 
the liquid could pass spontaneously by fractional crystallization. 
Thus saturated basaltic magma can dissolve granitic inclusions by 
precipitating basic minerals and the granitic material passing into 
solution then becomes a contribution to the normal granitic differ- 
entiate that may form by fractional crystallization if the conditions 
are appropriate. 

The behavior of inclusions of sedimentary origin is more compli- 
cated since sedimentary material does not belong in the reaction 
series. A consideration of the extent and nature of the variation of 
composition possible in the crystalline phases formed from a magma 
shows that the incorporation of considerable amounts of sedimentary 
material would ordinarily bring about merely an adjustment in the 
composition and relative proportions of existing phases. Asa result 
of the non-appearance of new phases, the general course of fractional 
crystallization is unaffected. In general, the adjustment noted 
takes place through precipitation of the phases with which the 
magma is saturated. As an example it may be stated that the 
addition of highly aluminous sediments to basic magma should bring 
about the formation of anorthite and enstatite molecules at the 
expense of diopside molecules and should therefore cause the precipi- 
tation of crystals rich in anorthite and enstatite. Such action may 
have been important in the formation of many norites. The foreign 
material becomes a part of the general mass as a result of reaction 
and precipitation rather than by simple solution. 

The Cortlandt series of New York, with its inclusions, affords an 
illustration of the behavior of aluminous sediments in basic magma. 
Such sediments may be regarded as consisting in part of material 
corresponding in composition with igneous material late in the reac- 
tion series, together with a certain excess, which is highly aluminous. 
The former may become a part of the liquid by the method of reac- 
tive solution already described. There results the piling-up of the 
highly aluminous excess in the inclusions, with formation of such 
minerals as sillimanite. Moreover, as a consequence of what may 
be somewhat loosely called the instability of sillimanite in contact 


570 N. L. BOWEN 


with liquid rich in anorthite or magnesian silicates, alumina is set 
free as corundum. ‘This condition is transient, however, and even 
these residues from the inclusions may become a part of the general 
mass as a result of the reactive precipitation noted above. The net 
result is the formation of noritic material with an increase in amount 
of the acidic differentiate normally possible. 

The addition of limestone to basaltic magma may perhaps give 
rise to a liquid capable of precipitating melilite in some cases and 
from such a liquid it is possible that some alkaline rocks may form 
by further differentiation. It does not seem possible that limestone 
inclusions can desilicate a granitic magma in such a way as to give 
rise to a liquid capable of precipitating feldspathoids. However, if 
limestone inclusions were present during the differentiation of the 
more basic liquid from which the granitic liquid may have formed, 
the presence of such inclusions might reduce the amount of free silica 
associated with the alkaline feldspar in this liquid to such an extent 
that the normal tendency of orthoclase to break down into leucite 
would manifest itself. Thus rocks bearing leucite, and possibly 
other feldspathoids, might form, but influences prevailing during 
early stages of differentiation, other than the presence of foreign 
matter such as limestone, may likewise lead to the formation of 
leucite at a late stage. 

In conclusion, it may be stated, therefore, that magmas may 
incorporate considerable quantities of foreign inclusions, both by 
the method of reactive solution and by reactive precipitation, and 
such action may have been important in connection with the pro- 
duction of certain individual masses: Thus some norites may have 
been produced as a result of the reactions discussed above, some 
granites may have had their mass augmented by reactive solution 
of granitic inclusions in the magma from which they differentiated, 
some alkaline rocks may have been formed as a result of the presence 
of limestone inclusions in the liquid from which they differentiated. 
All of these actions are, however, an emphasizing of normal pro- 
cesses possible in the absence of foreign matter. It is doubtful 
whether the presence of foreign matter is ever essential to the pro- 
duction of any particular type of differentiate. 


WASHINGTON, D.C. 
April 25, 1922 


_eitz Petrographical Microscopes 
_ NEW MODELS—WITH ENLARGED FIELD OF VIEW 


These microscopes have been developed 
on the basis of latest suggestions from 
petrographists. Their novel and modern 
design will prove a great aid in petro-- 
graphical studies. 


\ 


Every leading petrographist knows the 
superior optical and mechanical qualities 
of Leitz microscopes and we therefore 

are fully safe if we refrain from enumer- 
ating the many features. 


The variety of models we offer afford 
a suitable equipment for every specific 
requirement. 


The complete assortment of our petro- 
graphical accessories will interest those 
whg desire to add to their re 
equipment. 


BD iiccos cise. “S.M.” with synchronic | 
rotation of the nicols 


Some of the prominent features of the new nibdal. 
ne microscopes are: 


. Improved Gece. Clutch Changer which a the chijective to be 
permanently centered. 


Lik Anastigmatic Tube Analyzer sich permits a most crisp definition, 


Special Illuminating Appar eUs, whereby a wider range of application. iS 
afforded. 


tied Microscope Tube which accommodates eyepieces of wide diameter, © 
: ey the field of view is increased 50% over other types of instruments. 


ae Ask for Catalog II-B (F) 


LEITZ- QUALITY 


d Re OF gE 
\ OPTICAL AND MECHANICAL 
WORKMANSHIP 


Qs NOLS 
SURPASSED a EQUALLED 


ITIs 


/ E.LEITZ\ 
NEW vorK / 


60 East /O™St. 


Second Edition: Revised and Enlarged 


Source Book for the Economic 
Geography of North America 


By CHARLES C. COLBY 


Associate Professor of Geography 
in the University of Chicago 


§ The fundamental idea of the book is to make 
available to the busy teacher the material on this 
subject which is scattered widely through litera- 
ture. The course which has been given for a 
number of years at the University of Chicago 


has resulted in the collection and organization of _ 
the material most suitable for the purpose. 


§ The book contains short. magazine articles 
and materials from Canadian, Mexican, and 
United States government publications. The 
selection of the material has. been governed by 
the concept that there is a direct relation be- 
tween the environment and the economic activ- 
ities of the region. The author aims to survey 
the different sections of the country from the 
viewpoint of industrial opportunity. 


With chapter and maps on Alaska 
Cloth, $3.00, postpaid $3.r0 


The University of Chicago Press 


Chicago - - Illinois - 
Binocular 


SPENCER 


ARE EQUIPPED WITH 


ne Interchangeable Tubes—Bin- 
ocular and Monocular. 


II. Nosepiece Adapter mounted 
independently of the tubes, so 
that the objectives remain in 
place and in center irrespec- 
tive of which tube is used. 


SPENCER LENS COMPANY 


MANUFACTURERS 


_ be carried and used anywhere. Has 


374 Broadway 


Mon-Objective 


Microscopes, Microtomes, Delineascopes, Optical Glass, Telescope 
Discs and Lenses, Optical Measuring tae etc. 


BUFFALO, N.Y. 


Fits in case Na four inches high ant 


standard keyboard, with four rows of keys 

and no shifting for figures. Has the au 
matic ribbon reverse. Has every feature y 
associate with typewriting at its best. 


The extraordinary demand for the Remington 
Portable—which is taxing to the utmost our 
factory capacity—is proof of its leadership, 
in quality and popularity. — : 


Price, with case, $60 ie i 


Remington Typewriter Compe 


(Incorporated) - 


Branches and Dealers Everywhere Ne 


SPENCER MON-OBJECTIVE BINOCULAR 
MICROSCOPE NO. 2H 


A \ SEMI QuaRTERL 


aa #IAR ihe iii iia 


a Nays N 
EDITED BY | 4 
THOMAS C. CHAMBE 5 
; With the Active Collaboration of oa 
_ ROLLIN T. CHAMBERLIN, Managing Editor 
sas Paige : Dynamic Geology « n 
PUAR’ WELLER, Lventeneste Balbsatalgay ALBERT JOHANNSEN, Petrology 


ra 


DSON S. BASTIN, Economic Geology : J HARLEN BRETZ, Stratigraphic Geology 


"ASSOCIATE EDITORS 


D EIKIE,- Great Britain - 2 *JOHN C. BRANNER, Leland Stanford Junior University 
RICHARD A. F. PENROSE, Jr., Philadelphia, Pa. 

. WILLIAM H. HOBBS, University of Michigan 
FRANK D. ADAMS, McGill University 

CHARLES K. LEITH, University of Wisconsin 

ORTH DAVID, ees _ WALLACE W. ATWOOD, Clark University 

sLI \ Stanford Junior University WILLIAM H. EMMONS, University of Minnesota 
WALCOTT, Smithsonian Institution ARTHUR L. DAY, Carnegie Institution 


« 


_ OCTOBER- NOVEMBER 1922 


" SYENITE AND _NEPHELITE PORPHYRY OF BEEMERVILLE, 


- - eae oa ened (es AUROUSSEAU AND eee Sh Wicunee 
D AL CORRUGATED ROCKS | oe Z - = Wirttam J. Mater 
ce a - - - = lp, 168i, L. Voct 

ae - Ba fe - - loys 13 


UNIVERSITY. OF CHICAGO RES? 
-) CHICAGO, ILLINOIS, U.S.A. 


Soe ea CAMBRIDGE UNIVERSITY PRESS, Lonpon 
THE Se eee eT RAISES Tone, Osaka, Kyoto, FuxKuoxo, SENDAI 
THE ‘MISSION BOOK COMPANY, SHANGHAI 


NUMBER 7 


GEOLOGY 


571 
587 


Pas = a cf oe: ~ 


Ortober-November hee ida thee eb 8) as ; : av + 
pa! Ee, hs SL 


THOMAS C. CHAMBERLIN — 
With the Active Collaboration of 


ROLLIN De CHAMBERLIN, Managing Editor 


Dynamic Geology Ry * 
STUART WELLER eral , ALBERT JOHANNSEN ie 


Invertebrate Paidertsigey : Petrology 
EDSON S. BASTIN ; f} HARLEN BRETZ_ % 


- Economic Geology j Sore By 


The Journal of Geology is published by the University of Chicago at the University o hicago 


May 1, ee ine Atos T, is Ab cnien 15, November 1, Diener Ths “The oe ik 
per year; the price of Srele copies is 65 cents. Orders for service of less than a half-yea 
at the eed rate. q Postage is prepaid ‘ the publishers on a orders home 


on Hea ania, (total $4. 50); on single copies 6 cents (total oe centé). 
to make all remittances payable to The University, of Chicago Press in. postal or 
or bank drafts. Re 
The following are authorized to aie the prices indicated: nh ae ae ty 
For the British Empire: The Cambridge University Press, Fetter tae London, Bo: (om < 
subscriptions, including postage, £1 5s. each; single copies, including postage, 4S. each. 
For Japan and Korea: The Maruzen-Kabushiki-Kaisha, 11 to 16 Nihonbashi ‘Tori § 
Tokyo, Japan. Yearly subscriptions, including postage, Yen 9.45 cate single copies 1¢ 
postage, Yen 1.50 each. , ee 
For China; The Mission Book Comper: 13 North Szechuen Bart ‘Shanghai, Ye 
scriptions $4.00; single copies 65 cents, or their equivalents in Chinese money. "Postage 
mailed-direct outside of Shanghai, on yearly subscriptions 50 cents, on single copies 6: cents. 
Claims for missing numbers should be made within the month following the regular month 
cation. The publishers expect to supply missing numbers free only when losses have been sus 
transit, and when the reserve stock will permit. : 
Business correspondence should be addressed to The University ap Chicago te Chicoge 
Communications for the editors and manuscripts should be addressed to the ‘adlttors: oh it fE 
or Groxocy, the University of Chicago, Chicago, Ill. 
Contributors are requested to write scientific and proper names with particular care and i in 
to follow the form shown in the pages of the Journal. eg Seo 
Because of the high cost of printing, this Journal has found it necessary to adopt ihe elle 
4 x. If articles exceeding 25 pages in length are accepted for publication in the Journal, the cost of 
position beyond 25 pages shall be paid by the author or by the institution which he represents, at the ¢ 
rates of the University Press. “{ 2. The cost of illustrative material, in excess of $1.00, per page { 
article, beyond the cost of type wicek would have occupied the same space, shall be borne b 
{| 3. The cost of corrections in proof in excess of ro per cent of the cost of first composition, wall 
to the author. 4 4. While the cost of printing remains as now, the Journal cannot supply reprin 


Fintered a as dccondelabs matter, March 20, 1893, at the Post office at Chicago, Til., yaa ae Re 
Acceptance for mailing at special rate of postage eye for in Sechion 1103, Act of Octo 

July 15, tors. 
PRINTED IN THE U.S.A. 


Among articles to appear in 
early numbers of the Journal 
of Geology are the following: 


The Lava Field of the Parand Basin, South America. 
By CHARLES LAWRENCE BAKER. 


The Physical Chemistry of the Crystallization and Mag- 
matic Differentiation of Igneous Rocks. VI. By 
J. HL. Vocr. 


Origin of the Triassic Trough of Connecticut. By W. G. 
FOYE. 


The Problems of the Anorthosites and Other Monomineral 
Igneous Rocks. By F. LoEwINson-LEssinc. 


In Support of Gardner’s Theory of the Origin of Certain 
Concretions. By LEROy PATTON. 


_ The Pleistocene History of the Lower Wisconsin River. 


By Pau, MAcCtrinTOocx. 


Mud Cracks on Steeply Inclined Surfaces. By GERALD 
R. MacCarrTuy. 


The Cycle of Erosion and the Summit Level of the Alps. 
By W. M. Davis. 


Some New Features in the Physiography and Geology of 
Greenland. By LAUGE Kocu. 


Cambridge CAntversity Press 


The Study of Geological Maps. By Grrreuve L. Extes, D.Sc., Fellow of Newn- 
ham College, Cambridge. Crown 4to. With 7 plates, 64 figures, and a map. Price on 
application. Cambridge Geological Series. & 


A Manual of Seismology. By CuarztEes Davison, Sc.D. With 100 illustrations, ; 
Demy 8vo. Price on application. 


Petrology for Students: an Introduction to the Study of Rocks under 
the Microscope. By Atrrep Harker, M.A., F.R.S., F.G.S. Fifth edition, revised. 
Crown 8vo. $3.75. 


A Treatise on Crystallography. By W. J. Lewis, M.A. Demy 8vo. $5.00. 


The Principles of Stratigraphical Geology. By J. E. Marg, Sc.D., F.R.S., F.G.S. 
Crown 8vo. $2.50. 


An Introduction to Geology. By J. E. Marr, Sc.D., F.R.S., F.G.S. Crown 8vo. 
$2.50. 


Published by the Cambridge University Press (England) 
The Macmillan Co., Agents in the United States 
64 and 66 Fifth Avenue, New York City 


Essentials for the Microscopical Determination of 
Rock-forming Minerals and Rocks | 


By ALBERT JOHANNSEN 


Professor of Petrology 
The University of Chicago 


A laboratory manual designed to provide students of petrology and ‘geologists | 
interested in petrology with information on methods of determining microscopically | 
the points of separation between similar minerals and their modes of occurrence. | 

This valuable volume contains, in addition to a folding alphabetical list so | 
placed that reference may be made to it without turning pages, the following | 
important features: “A Quantitative Mineralogical Classification of Igneous ~ 
Rocks”’; ‘““ASummary of Petrographic Methods”; ‘Modes of Occurrence of Various | 
Minerals”; “Rules for Computing Rocks from Their Modes’’; and “Names Pro- | 
posed for Various Families.” | 

Cloth; $2.00, postpaid $2.10 


THE UNIVERSITY OF CHICAGO PRESS | 
CHICAGO . - ILLINOIS | 


PHYSIOGRAPHIC DIAGRAM 


OF THR 


UNITED STATES 


PHYSIOGRAPHIC DIAGRAM OF 
THE UNITED STATES 


Size 62X44 inches, Scale 50 miles to one inch 


By A. K. LOBECK 
Department of Geography, University of Wisconsin 


This diagram is intended to be used asa guide in teaching the physiography and 
geology of the United States. The topographic features are expressed with the idea 
of revealing or suggesting the underground structure, and the general character of 
the underlying rock. 


An introductory drill such as having students draw cross sections through the 
Adirondack uplift; the Cincinnati or Nashville Domes, across the Appalachians in 
different places, through the Black Hills, Central Texas, the Ozarks, Wisconsin and 
many other regions is followed up with the use of other available material, espe- 
cially state geological maps and the United States Geological Survey topographic 
sheets. 


The student frequently fails to note the significance of a locality with relation to 
the physiographic province or geological region as a whole. This is overcome by 
requiring the student to locate upon the diagram the position of the topographic 
quadrangles under consideration. As practically every topographic map published 
by the Survey was consulted in the preparation of the Diagram, most of the signifi- 
cant features are recognizable in the Diagram too. Seventeen geological provinces 
are discussed briefly on the map. 


PRICES 


Student edition (one color) with one-half of U.S. on each side of sheet, size 31 X44 
inches. 4 


Singleicopyseeeee nee $1.25 Fifty copies, each......... $1.00 
Ten copies, each...... I.10 One hundred copies, each.. .g0 
Two color edition with entire map on same side of sheet, size 62 44 inches. 
Singleicopysseceee seen $2.25 Fifty copies, each......... $1.75 
Ten copies, each...... 2.00 One hundred copies, each.. 1.50 


Prices ontwo color edition, mounted on cloth for permanent class room use, size 
62X44 inches. 


With Common Rollers at Top and Bottom................... $4.50 
On Spring Roller and Board with Dust Proof Cover...... Bb ae auton 
imeSteeliopringsloller| Cases eer ieee receieeereaeiicre Deere ssainre 8.75 


A. JI. NYSTROM & COMPANY 


Publishers 2249 Calumet Ave., Chicago 


Second Edition: Revised and Enlarged 


Source Book for the Economic WEATHER CHARTS 


Geography of North America J. PAUL GOODE 


Professor of Geography in the 
By CHARLES C. COLBY University of Chicago 
Associate Professor of Geography 
in the University of Chicago 


Monthly weather chart, 
§ The fundamental idea of the book is to make 8 eench 1 
available to the busy teacher the material on this x IOz Inches, Iz cents 
subject which is scattered widely through litera- a 2 

ture. The course which has been given for a Climatic chart (yearly), 
number of years at the University of Chicago enc 

has resulted in the collection and organization of I5X 10% inches, 3 cents 
the material most suitable for the purpose. 


§ The book contains short magazine articles 

and materials from Canadian, Mexican, and 

United States government publications. The The University of Chicago Press 
selection of the material has been governed by Chicica a i Illinoi 
the concept that there is a direct relation be- 8 ols 
tween the environment and the economic activ- 

ities of the region. The author aims to survey 

the different sections of the country from the 

viewpoint of industrial opportunity. 


With chapter and maps on Alaska SECONDHAND BOOKS 


Cloth, $3.00, postpaid $3.10 | 

Send postal card request for TANDY’S NEW 

—— pst c ee es oa ART, HISTORY, 
I E EOLOGY, BOTANY, ZOOLOGY, 

The University Chicago Press CONCHOLOGY, LEPIDOPTERA, ete. 


Chicago 5 ; Illinois Address M. TANDY, Dallas City, Illinois. 


e 


THE ORIGIN OF THE EARTH 


By THOMAS C. CHAMBERLIN 


Professor Emeritus in the Department of Geology 
The University of Chicago 


One of “The University of Chicago Science Series” 


FROM THE PREFACE 


“In telling the story of this search for the mode by which the earth came 
into being, we have let the incidents that led the inquiry on from one stage to 
another fall in with the steps of the inquiry itself. It is in keeping with the 
purposes of this series of booklets that the motives which set researches going 
should have their place with the quests that arose from them. .. . . The final 
story of the birth of the earth will come only after a time, when the vestiges of 
creation have been more keenly discerned and more faithfully rendered than is 


possible now.” 
Cloth, $1.75 net, postpaid $1.85 


THE MT = See PRESS 
CHICAGO : 2 ILLINOIS 


YOUR MAP NEEDS 


can be selected quickly and with assurance of genuine satisfaction from the 
most comprehensive assortment of effective geography ‘‘teaching tools’? ever 
offered. Note below a summary of that part of 


“‘The Johnston-Nystrom Line’’ 


applying particularly to geography and geology teaching. 


POLITICAL GEOGRAPHY 
Political Maps—Ten Series comprising 136 
different maps. 


Blackboard Outline Maps in two sizes. 
All continents, U.S., World, and many 
states. 


Desk Outline Maps (for pupils) two sizes. 
Wall Outline Maps. 

State Maps. 

Globes—Political, in four sizes. 
Globes—Slated, in four sizes. 
Atlases—Political and Commercial. 


COMMERCIAL GEOGRAPHY 


Finch Series for United States—34 maps 
on to plates, 50X38’. Shows Products, 
Industries, etc. . 


World Maps (Trade and Industry). 
Atlases. 


PHYSICAL GEOGRAPHY 


Physical Maps—Five Series consisting of 
82 different maps. 


Rainfall Maps, U.S., World and Continents. 


TD) 
( B 
ian 


VY 
cL 


as & 


PHYSICAL GEOGRAPHY (Continued) 
Vegetation Maps, Continents and World. 
Pressure and Winds Map. 

Major Natural Regions Map.- 
Thermal Regions Map. 
Structure Map. 
Timber Distribution Maps. 
Desk Outline Relief Maps. 
U. S. Outline-Map with Topography. 
Graph Charts (slated cloth) two sizes. 
Globes—Physical, 15” diameter. 
Globes—Political. 
Globes—Slated. 
Tellurians. 
Atlases. 

PICTURES 
Land Forms, 24 Illustrations. 
Physical Phenomena, 16 pictures. 
Plants and Trees of Commerce, 30 pictures. 
Types of Nations, 11 pictures. 


Geographical Types and Scenes: two 
series, 70 pictures. 


Tell us what subject you are interested in and we will gladly give you information 
about any item in our line that may meet your needs. 


USE THIS CONVENIENT COUPON TODAY 


I am interested in the subject checked below and request literature describing your 


maps, etc., applying to the same. 
OO Political Geography 

1 Commercial Geography 

HO Physical Geography 


O Other Items as follows: 


Mail 


iil A. J. NYSTROM & CO. 


(Go22) 
O Pictures 
O No. A21 Geography Catalog 


2249 Calumet Ave. 
CHICAGO 


Goode’s Base Maps and Graphs 


A series of outline maps for all classes of work in applied sciences and the various fields 
ofresearch. Prepared by J. Paul Goode, Professor of Geography, University of Chicago. 

These maps are adapted for use in every grade from the university to the common 
school: in geography, including commercial or economic geography, in physiography, 
geology, botany, zodlogy, anthropology and ethnology, sociology, economics, politics, 
and history. The maps have been prepared by being first drawn on a large scale, to 
insure accuracy of detail, and then greatly reduced in the engraving. In the quality of 
the drawing they are superior to most maps used in books and magazines. 

All maps are perforated for insertion in a standard notebook cover with page 8X 103 
inches. The larger sizes fold once or twice into this cover. A new classification has 
been arranged, but it should be noted that the maps themselves have not been altered. 
New maps and graphs are shown in blackface type. Write for prices. 


SERIES I SERIES II SERIES IMl SERIES IV Maps 
I IOI ao Rae The World: on Mercator’s projection. 
*ro1Hc ee 301Hc The World (continents): Homalographic projection. 
ney Zo1HcE, 5... The World (continents, Eastern half): Homalo- 
graphic projection. 
201HcW ..... The World (continents, Western half): Homalo- 
graphic projection. 
*ro1rHo te see The World (oceans): Homalographic projection. 
*IoIS Beer see The World (continents): Sinusoidal projection. 
*ro1P wide ee The World in Polar Hemispheres: Lambert’s 
azimuthal, equal-area projection. 
201PN Abels The World (North polar hemisphere): Lambert’s 
projection. 
201PS oth The World (South polar hemisphere): Lambert’s 
projection. 
201PB Baa Same map as 101P showing Isobars for January 
and July. 
2 Io2 202 shes North America: on Lambert’s azimuthal projection. 
3 103 203 ens South America: Sanson’s projection. 
4 104 204 304 Europe: conic projection. 
5 IO5 205 wees Asia: Lambert’s equal-area projection 
6 106 206 = Africa: Sanson’s projection. 
7 107 207 oe Australasia: Mercator’s projection. 
9 109 Say 309 United States of America: conic projection. 
ee 209E aad United States of America (Eastern half): conic 
projection. 
209W Bate United States of America (Western half): conic 
projection. 
Io 110 Peau ene United States of America: state outlines only: 
conic projection. 
III rate seh Canada: conic projection. 
112 reas rae Mexico: conic projection. 
ie Reis 213 wor Gulf and Caribbean Lands: conic projection, 
I4 II4 aren es The British Isles: conic projection. 
a3 euA 215 set's Europe, Northwestern: conic projection. 
16 116 216 ace Europe, Western and Southern: conic projection. 
17 I17 aon atk France: conic projection. 
18 118 eee Se The Spanish Peninsula: conic projection. 
ste) I1IQ cbi0 sod Italy: conic projection. 
20 aes ete oe Central Europe: conic projection. 
21 121 sae Liens The German Empire: conic projection. 
24 124 state asia The Levant: conic projection. 
nies 226 oun China and Japan: Bonne’s equal-area projection. 
132 232 pete United States of America, by counties with names: 
. conic projection. 
ee 133 teks athe Chicago and vicinity: conic projection. 
51 I51 AI fies North Dakota: conic projection. 
GRAPHS 
80 Bio ieee hs Climatic Chart (yearly). 
181 Monthly Weather Chart. 


THE UNIVERSITY OF CHICAGO PRESS, CHICAGO, ILLINOIS 


ee 


Ward’s Natural Science Establishment 


ROCHESTER, NEW YORK 


Relief Maps 


This map, 29" x 46", scale 32 miles to the inch, with a vertical exaggeration of 20 to 1, costs 
only $30.00. The same map, hand colored to show the geology as given in United States 
Geological Survey professional paper No. 71, is $150.00. Write for circular G-233 describing our 
relief maps. 


Fossils 


We are better prepared than ever before to furnish index fossils and exhibition material. A few 
of our recent acquisitions follow: 


Cambrian material collected by Dr. Wolcot in British Columbia. i 
| Rare Ordovician material from New York and Bohemia, such as Aeglina, Proetus, Cryptolithus, 
' __Illaenus, Cystids, etc. is 

Rare Silurian forms from Bohemia such as Ascoceras and Ophidioceras, Cromus, Acidaspis, etc. 

Rare Devonian material from Germany such as Harpes, Cerarges, Goniatites showing sutures, 

Hexacrinus, Cupressocrinus, etc. 

Fine Solenhofen fish, crustaceans and insects, 

Comanchic material from Texas, 

Eocene leaves from Texas and Louisiana, 

Insects in amber. 


Collections of Rocks 


Our rock collections, varying in price from $1.00 to $200.00, are arranged for Genetic, 
Stratigraphic, Petrologic or Petrographic study. Write for circular No. 220, listing these and 
also our Mineral collections. 


WARD’S NATURAL SCIENCE ESTABLISHMENT 
ROCHESTER : : - - : - - NEW YORK 


LEITZ PETROGRAPHICAL 
MICROSCOPE “K.M” 


for routine investigations 


It is provided with the same features 
as our large research model, viz.: 


Wide tube and eyepieces of enlarged 
diameters which renders a field of 
view 50% larger than those of other 
manufacture. Improved illumina- 
ting systems. 


Its liberal size permits the accommoda- 
tion of all existing accessories, including 
“Federow”’ Universal-Rotating-Stage. 


We offer an extensive line of petro- 
graphical microscopes and accessories 
and are able to meet all requirements 
for research or elementary investiga- 
tions. 


The microscope ‘S.M,”’ for syn- 
chronic rotation of both nicols is our 
latest achievement. 


Leitz Microscopes are recognized throughout the world as leading 
and are invariably selected by petrographists who appreciate quality. 


Write for complete Catalog II-B (F) 


‘~.oF 
OPTICAL ano MECHANICAL 
WORKMANSHIP 


LEITZ-QUALITY 
\ Sune ae cue 
SURPASSED or: EQUALLED 
mts SUPREME 


/ELEITZ\ 
\ NEWYORK / 


60 East /O@St. 


VOLUME XXX NUMBER 7 


Walle, 


IOURNAL OF GEOLOGY 


October-November I 922 


THE NEPHELITE SYENITE AND NEPHELITE 
PORPHYRY OF BEEMERVILLE, NEW JERSEY 


M. AUROUSSEAU anp HENRY S. WASHINGTON 
Geophysical Laboratory, Carnegie Institution of Washington 


OUTLINE 
INTRODUCTION 
THe NEPHELITE SYENITE OF THE BEEMERVILLE MAss 
THE NEPHELITE PORPHYRY 
CRYSTALLIZATION VARIANTS OF THE BEEMERVILLE MAGMA 
THE OCCURRENCE OF ZIRCONIA AND RARE EARTHS 
THE STATUS OF SUSSEXITE 
SUMMARY 
INTRODUCTION 

During the earlier geological survey of New Jersey many outcrops 
of igneous rocks, mainly dykes, were mapped and recorded in the 
state reports without particular description. These rocks are 
nearly all located in Sussex County, and, owing to the small size 
of most of the outcrops, their scattered nature, the maturity of the 
topography, and the extent of weathering of the rocks themselves, 
their interesting nature has only been recognized gradually, though 
the first record of the largest mass dates back to 1868. 

Through the work of Emerson, Kemp, and Wolff these rocks 
are now widely known and are regarded in the local sense as a suite 
of genetically connected alkalic intrusions, and more generally as 


571 


572 M. AUROUSSEAU AND HENRY S. WASHINGTON 


showing affinity with the rocks of a dominantly sodic zone or 
comagmatic region which finds expression in the eastern United 
States from New England to Texas.* 

The large mass of nephelite syenite situated to the northwest 
of Beemerville was first described in detail by Emerson, who 
recognized its true character and described it microscopically. 
He believed the mass to be a large dyke.2 Emerson also described, 
under the name of mica-diabase, the series of dykes penetrating 
the ore-bodies at the Buckwheat Field at Franklin Furnace.3 

Kemp examined the mass known as Rutan’s Hill, which forms 
a prominent landmark east of the northern end of the main mass 
of nephelite syenite at Beemerville, and described it as a boss of 
porphyrite, giving two analyses of the rock, and one of the biotite 
isolated from it.4 He also recognized the nature of a number of 
similar ‘‘bosses”’ to the southeast of the main mass, south of 
Beemerville and Plumbstock. On a more detailed investigation 
Kemp mapped and described the main mass and its immediate 
satellites, demonstrating clearly the inhomogeneity of the Beemer- 
ville nephelite syenite and proving the occurrence of a nephelite 
porphyry, probably as a dyke, within it. Here also he showed 
that the satellitic bosses are basic alkalic rocks, allied to the ouachi- 
tites and fourchites. He provided two analyses of different facies 
of the nephelite syenite and one of the nephelite porphyry. These 
investigations established the fact that the Beemerville mass and 
its associates resemble the rocks of the Magnet Cove complex, 
Arkansas. 

Kemp next examined a basic dyke two miles northwest of 
Hamburg, and from microscopical investigation, and the analysis of 
certain pseudomorphous spheroids which it exhibited, he offered 
the tentative suggestion that the rock was a leucite tephrite, and 
concluded that it was related to the Beemerville nephelite syenite.® 


«H. S. Washington, Jour. Franklin Inst., CXC (1920), p. 796. 
2B. K. Emerson, Amer. Jour. Sci., XXIII (1882), p. 302. 

3 [bid., p. 376. 

4J. F. Kemp, Amer. Jour. Sci., XX XVIII (1889), p. 130. 

5J. F. Kemp, Trans. N.Y. Acad. Sci., XI (1892), p. 60. 

6 J. F. Kemp, Amer. Jour. Sci., XLV (1893), Pp. 303- 


SYENITE AND PORPHYRY OF NEW JERSEY 573 


Kemp’s material was studied by Hussak, who had no hesitation in 
terming the rock a leucite tephrite, and considered that the sphe- 
roidal pseudomorphs were pseudoleucite, or were at any rate 
pseudomorphs after leucite.. Kemp then visited some newly 
exposed dykes at Rudeville and obtained material which provided 
definite proof of the existence of leucite, and came to the conclusion 
that the ‘‘mica-diabase”’ dykes of Rudeville, Franklin Furnace, and 
the Hamburg dyke are all related to the Beemerville mass.2, Kemp’s 
papers are illustrated with useful locality maps, but the analyses 
given are all incomplete and inadequate. 

A complete petrological description of the nephelite syenite of 
Beemerville and the large dyke at Franklin Furnace, together with 
good, though incomplete, analyses, is given by Iddings, who recog- 
nized the Franklin Furnace dyke to be a minette.3 

In 1899 Ransome described a small occurrence of nephelite 
syenite, mica syenite, hornblende syenite, and hornblende granite, 
associated with Mesozoic gabbro, at Brookville, in Hunterdon 
County, New Jersey, sixty miles west of south from Beemerville. 
The relations of these rocks to the gabbro were not clearly ascer- 
tained, and they were regarded by Ransome as inclusions, though 
he considered the possibility of an intrusive relation.4 Our colleague, 
Dr. N. L. Bowen, has collected specimens from this locality which 
suggest differentiation of the nephelite syenite and, on a small 

scale, an intrusive relation toward the gabbro. 

In 1902 Wolff described an undoubted leucite tinguaite dyke, 
which cuts the Beemerville nephelite syenite mass near its southern 

end, giving a detailed account of the pseudoleucite, and a good and 
complete analysis of the rock. This is the most satisfactory existing 
analysis of any rock from the alkalic series of Sussex County.5 
Finally, in 1908 Wolff provided a co-ordinated description of the ig- 
neous rocks of Sussex County,° in which he points out that the same 


1E, Hussak, Newes Jahrbuch, IL (1802), p. 153. 

2J. F. Kemp, Amer. Jour. Sci., XLVII (1894), p. 339- 

3J. P. Iddings, U.S. Geol. Surv., Bull. 150 (1898), pp. 209 and 236. 
4F.L. Ransome, Amer. Jour. Sci., VIII (1899), p. 417. 

5 J. E. Wolff, Bull. Mus. Comp. Zool. Harvard, XX XVIII (1902), p. 273. 

6 J. E. Wolff, Geol. Atlas, New Jersey, Franklin Furnace folio (1908), p. 12. 


574 M. AUROUSSEAU AND HENRY S. WASHINGTON 


pyroxene, a zoned aegirite-augite, and large crystals of titanite and 
biotite characterize nearly all the rocks of the series, and that 
the main nephelite syenite mass of Beemerville, with its transgressive 
dykes of nephelite porphyry and leucite tinguaite, has a definite 
relation to the disposition of other rocks of the series. Close to it, 
to the east and south, are the bosslike or necklike ouachitite brec- 
cias; farther to the southeast is a zone of nephelite syenite and 
bostonite dykes, which, like the main mass, are concordant with 
the bedding of the intruded series (the Ordovician Martinsburg 
shale); finally, at some distance to the southeast, are the lampro- 
phyric dykes, which are disposed radially toward the Beemerville 
mass, intruding the Ordovician Kittatinny limestone and the 
pre-Cambrian Franklin limestone. Wolff concludes that the alkalic 
rocks are post-Devonian in age and probably much later. 


THE NEPHELITE SYENITE OF THE BEEMERVILLE MASS 


The main mass of nephelite syenite forms a long, narrow intru- 
sion of elliptical outcrop, lying between the Silurian Shawangunk 
conglomerate and the Ordovician Martinsburg shale, at the foot 
of the Kittatinny Ridge, the southern extremity of the mass being 
two miles to the northwest of Beemerville. It is most easily 
accessible from the town of Sussex (formerly called Deckertown, 
and referred to by that name by Emerson and Kemp). 

The formal relationships of the mass are obscure. Both 
Emerson and Kemp regarded it as a large dyke, but Wolff is inclined 
to regard it as a sill, or an irregular, flat laccolithic mass. Washing- 
ton visited the locality in r901 in company with Professors Kemp 
and Broégger, and is in agreement with Wolff’s opinion. It was 
examined by Aurousseau in the summer of 1921, with special 
regard to this point, but no evidence of a decisive nature is obtainable 
on the ground. As the mass has been studied by a number of com- 
petent geologists at intervals over a long period of time, it is improb- 
able that any fuller information will be forthcoming, the outcrops 
being poor and the contacts obscured by thick soil and driit. 
In particular, no variations of dip are to be observed in the massive 
Shawangunk conglomerate which overlies the mass. To our minds 
the occurrence of the body (which can hardly be younger than 


SYENITE AND PORPHYRY OF NEW JERSEY 575 


early Tertiary and is probably much older) at the junction of the 
Shawangunk conglomerate and the Martinsburg shale, is critical, 
and, taken in conjunction with the fact that long, narrow intrusions 
of nephelite syenite and bostonite lie parallel to the bedding of the 
Martinsburg shale farther to the east, inclines us to the opinion 
that the Beemerville mass is a lenticular sill, or a flat laccolith. 

The nephelite syenite is a somewhat basic foyaite, of the ‘‘foy- 
aite range”’ as recently defined in the classification of the nephelite 
syenite family proposed by Shand." Itis very variable along the mass 
and, although the bulk of the exposure is a fairly constant foyaite 
of the Magnet Cove, Arkansas, or the Umptek type, it grades 
locally into other facies, which are often more basic than the main 
mass. Ditroitic and ijolitic modifications may be collected, and 
especially along the eastern border it becomes foliated, or lujavritic, 
in character. Near the center of the exposure, small, local facies 
with abundant titanite may be found. These variations have been 
admirably described by Emerson and by Kemp. Kemp’s descrip- 
tion may be quoted to illustrate this point: 

The dike varies considerably along its course. The typical elaeolite- 
syenite forms the northern third and the southern extremity, but between these 
points its character changes. Near the southern part of the middle third 
elaeolite-porphyry appears, and forms a most beautiful example of this rock. 
It may come from dikes, as no actual exposures are available. Further south 

_a basic holocrystalline rock comes in which is exposed in place; and, as subse- 
quently shown, contains less silica and more biotite than the typical syenite. 
But on the extreme south where the highway crosses the dike, the rock is much 
like that on the north. It is, however, greatly decomposed, and fresh, firm, 

‘pieces are hard to find.? 


The variation, even of what appears to be the predominant rock, is 
well shown by comparing the analyses by Eakins and Aurousseau 
(Table I). The Martinsburg shale, along the eastern contact, 
has been metamorphosed to a hornstone, the aureole being narrow. 

To the very complete petrographic description of the normal 
rock, given by Iddings, we have little to add. One slight correction 
is necessary. The mineral identified as sodalite belongs to the 
hauynite-noselite series, as is indicated by the analysis here given. 


1S. J. Shand, Trans. Geol. Soc. South Africa, XXIV (1921), p. 117. 
2J. F. Kemp, Trans. N.Y. Acad. Sci., XI (1892), p. 64. 


576 M. AUROUSSEAU AND HENRY S. WASHINGTON 


The most noteworthy chemical characters of the nephelite 
syenite are its low silica percentage, the approximate equality in 
amount of soda and potash, the high content of titanium and 
zirconium, and the comparatively large amount of SO, as compared 
with chlorine. Though it undoubtedly belongs to the highly sodic 


TABLE I 
I II Ii IV Vv 
SIOFE Wein erent ee 47.19 53-50 53-00 52.25 Z 0.18 
TNO Paaie nig gar ere eee 23.01 24.43 Prt. 10(0) 22.24 Or 33.08 
HE Ose ae ne crontasiert: Ber .I9 1.89 42 An 39 
BeOi ye sien: DDR » 22 2.04 1.098 Le II.99 
OY es ON alee cesta i 1.07 0.31 0.32 0.96 Ne 32.66 
CAO a eens cite 2.93 1.24 3.30 I.54 Th 0.71 
DEE O eee Mic ieacutsinarls 7.97 O43 ie0..30 9.78 Nec 0.95 
UO er ree be tensa clots 8.23 9.50 8.42 6.13 Di 4.32 
1a OR eee rs tenet See On5sil iF fee 5 0.98 
1B HO r a do una lads 0.04f 93 0.24f eile Mt I.39 
CORRS irom cre: On38) |. veces ee. On82 iil hte Il 4.10 
WiQa ance" eee BRT Oy [resents O.I1 0.60 Hm 2.08 
ZIO ps ah nae oe eke Cov EC || are eer aac (o)a(0) Wn et ects oc Ap I.O1 
PLO cue nance ONS OI leita st 0.15 0.53 Water ©.59 
SOp ieee one nee OF AY aillsnonieemer None: |iisusie 2 > s.2cif ceca eee | See 
Clee emer ane OLOLY |lomene terrae O02 Wales cocci] cee | ee 
Ie ce eter PERL ReAe joni en Mae An MMOS CG Gh Go ctlloodobcocc 
See ectain iG een eee thal cee eertcel eaetepaneyateae oR; Ennis Man nioicecilacaoooo he 
CrO Reece Nome } os ils eea| le eae) 6 ass eels ea el ee 
(C&W) OF O06.) Woo alee dicted Seta ave ole lllse ak Saves oll oe ee eee | eee 
TAO) tee are Vetere peda. ©.16 ©.10 ©.80) sis 5 os dillon crores | eee 
IBA OES Sela ree ae OOO). oils eae OWE loiehscs. 2 stance ll ete chanel | Ce eee 
Sum ee ee 100.16 99.96 100.48 00316: eae 100.43 


I. Nephelite syenite, Beemerville, N.J. M. Aurousseau, anal. 
II. Nephelite syenite, Beemerville, N:J. L. G. Eakins, anal. U.S. Geol. Surv. 
Bull. 150 (1808), p. 211. 
III. Foyaite, Magnet Cove, Ark. H. S. Washington, sual Jour. Geol., IX (1901), 
p. 611. 
IV. Lujavrite, Rabots Spitze, Kola. V. Hackmann,anal. Fennia, XI (1894), p. 132. 
V. Norm of I. Symbols, (1) II. 7. 1’. 3. Janeirose. 


comagmatic region of the eastern United States, in these respects 
it differs remarkably from the well-known nephelite syenites 
of Massachusetts, Connecticut, New Hampshire, and Maine, and 
also from the nearest similar exposure, that is, from the nephelite 
syenite of Brookville. The last-named rock resembles the nephelite 
syenites of New England and shows the essential differences 
between them and the Beemerville rock (see Table III). The other 
nephelite syenites of the northeastern United States, and, in 


SYENITE AND PORPHYRY OF NEW JERSEY Lv 


general, those of eastern Canada, are characteristically dosodic, con- 
tain less titanium, and where it has been determined, less zirconium 
than that of Beemerville: they are also more silicic, and some of 
the Canadian rocks have a marked tendency to show an excess of 
alumina. The material analyzed by Eakins is termed “the average 
rock” by Iddings. That used for the analysis here presented was 


— Alii 
S, All ieil ly see7 
= ALA L/ 
SSS SrA MMT ee 
eS 


iy 


Fic. 1.—The character of the Beemerville intrusion, as inferred from the distribu- 
tion of rock types: (1) Shawangunk conglomerate. (2) Martinsburg shale and con- 
tact aureole. (3) Normal nephelite syenite, with occasional nepheline-rich, and lujav- 
ritic (marginal) facies. (4) Transition of normal nephelite syenite into darker variety 
with biotite and titanite. (5) Nephelite porphyry, with tinguaitic and sussexitic 
facies. (6) Leucite tinguaite. (7) Ouachitite breccia of Rutan’s Hill. Diagram 
partly idealized. Main boundaries from the Franklin Furnace folio. The probable 
occurrence of a dyke of nephelite porphyry in the northern part of the main mass is 
suggested by an observation of Emerson, Amer. Jour. Sci., XXIII (1882), p. 308. 


collected by Dr. Wolff and is representative of the northern third 
and the southern extremity of the mass. The non-determination 
of TiO, and P.O, by Eakins of course affects the figures for Al,O,. 

As has been stated already, the Beemerville nephelite syenite 
resembles most closely the foyaite of Magnet Cove, Arkansas. 
An analysis of the latter rock is given in Table I for comparison 
together with a lujavrite from Umptek, described by Hackmann. 
The Arkansas rocks are characteristically sodi-potassic, like those 


578 M. AUROUSSEAU AND HENRY S. WASHINGTON 


of the Beemerville, though slightly more silicic. The most note- 
worthy differences are in the titanium and zirconium, and in the 
ferrous-ferric relationships. The Arkansas syenite apparently 
contains less of the acmite and more of the diopside molecule than 
that of Beemerville, and is distinctly poorer in titanium and zir- 
conium. ‘The minor differences among the volatile constituents also 
are not without interest. 


THE NEPHELITE PORPHYRY 


The nephelite porphyry, referred to in Kemp’s description of 
the variations of the main mass, is not exposed in a way which 
permits of a determination of its relationships. It is found in 
the neighborhood of Mr. T. Conroy’s house, near the middle of the 
main mass, and specimens have been collected from the north of 
the house, and from the peach orchard southwest of the house. 
Again quoting from Kemp: 

At the middle of point 4 C on the map the character of the dike changes, as 
is indicated by the float fragments, for no actual exposures occur. Porphyritic 
facies appear, and an excellent elaeolite porphyry was found. .... Another 
porphyritic rock occurs along this portion of the dike, which lacks the large 
phenocrysts of elaeolite. It has, however, others of feldspar, and in the slide 
shows the same tinguaitic base with a much more prismatic development of 
the elaeolite in the groundmass.t 
A small dyke of the same rock occurs two miles northwest of the 
town of Sussex, and is shown on the map of the Franklin Furnace 
folio. The nephelite porphyry of the main mass of Beemerville 
does not reveal its contacts. We believe it to be a dyke of some 
size, intruding the main mass, and both coarse- (definitely porphy- 
ritic) and fine-grained modifications, the latter suggestive of marginal 
relationship, may be collected north of Mr. Conroy’s house. 

The rock is typically porphyritic with nephelite crystals, and 
occasionally with orthoclase. In one specimen small phenocrysts 
of blue fluorite are quite visible to the naked eye. The groundmass, 
which has the typical, dull-green color of the tinguaites, is variable 
in texture in different specimens, and in the finer-grained variety 
consists of a mosaic of interlocking grains of orthoclase and nephelite, 
penetrated by minute aegirites. In the more normal variety the 
aegirites exist in two generations, those of the groundmass forming 


1J. F. Kemp, Trans. N.Y. Acad. Sci., XI (1892), pp. 66-67. 


SYENITE AND PORPHYRY OF NEW JERSEY 579 


minute tufts. Orthoclase also shows its twinned development 
here. ‘The larger aegirites are associated with small but numerous 
crystals of titanite and a little biotite. A few patches of ilmenite 
are noticeable, and under high power a little perovskite and apatite 
may beseen. Cancrinite and hauynite are quite minor constituents, 
and there is very little development of carbonates. .The general 
mineralogical development is quite similar, except as regards 
texture, to that of the nephelite syenite itself. Though orthoclase 
exists In minor amount in some cases, ordinarily it is an important 
constituent. 


TABLE II 
I II Til IV Vv VI VII VIII 

SiO... .| 50.08 | 49.84 | 49.96 | 55.31 | 45.64 | 50.16 | 46.48 | Z 0.32 
ANLOR, « ol] TPsOA | 1)? |) UO) 62 |) Bie oGe |! wolfe) |] m@).975) |) M@)g@e) || (Oke 4I.14 
He Owes Ae 70 4.40.) 4.55 G9 Ml Botley || MoS | | AGL |) LAND) 4.19 
EO). a ool) Bow 2.24 yo) TOD || Bost || BsO2 Dogte) || ING 30.10 
INGOs 5 al) en EO |) £23] OAH | BOL | won 2.49 | Hl 0.12 
(CAO). soo) Ba@O |) Aces |) aos ey || Bhatiby I Some | Zoe) || Alm 0.71 
Na One Onto So7 SoCs | (3.977 |b wey 7.63 8.46 | Ne 0.32 
IK). s cull WoO | Ookpat WoOe || OZ) || O09 |) 6.78 || Os73 |) AC 3.70 
ISL Se cleo 38 38 : 
peer | 2 co]| 224] 226)) 565] oaar| Di || 6-70 
COM Aly eiaree ss OMT] AlMnonn2 ilo Tore Warts les Lil apurenitye 0.36 | Wo Bagi 
TiO, 1.75 1.27 TS 0.07 DiMA Na s39'05 i222 Mt 3.48 
LEONI Ne oie OM TOn MOTO See cyl wll amiuie ate. alte te retest | tee Aer Il 2.89 
P20; OMG AG [loose ore OLGA) ale Nesoaeos ©.83 || O.kg || lehan 0.80 
SOjna- 4. Os EOVil|c ombaoo's GZeXO\ elena cttectal ape sacdes| He o.19 | Ap 1.34 
Che ctenciall eect ©.04 || ©s@H || O2OO Ie oscccsilocosaac 0.08 | Water] 0.44 
1266 6 cdicte eto eee | CeeOeEan ETD a ar actece Paces ae lca cae Layee ea ibm a 
CEOs siallneceraere INiorve mM MINIOne lies isict as |erees Sane. Al nae los geal eres tes all ee 
(CENO)AO} eatcranae OO I CCN |e eee erence (mateo oped Lan 0s SM agenne a ITA EA 

TL Opa ORT Failtst esc 2s OM LF iilercersee OT Oyaliseccane kts ADS Pe Wh care ewcley ea vereie 
BaO OBS 2ilacta cos ON SiO LR aes alee! al eveultas os [eshte sls es ler one gears lhe a tap tet ay | AO fev a 

Stanly lio gious area arene 100.2I | 99.86 |100.76 |100.48 | 99.91 }....... 99.96 

IEaS (OE cas, saci lapse enesens QUOT Py Oma [evs te ss alentyn ew OPO2I iw hatalael | Mbemeiltes 

SHEN, lla lateral eaaemaenne HOO, || OGG |uCO. 70 |e. || OOO) |loncsocaellansooas 


I. Nephelite porphyry, Beemerville, N.J. H.S. Washington, anal. Incomplete. 
II. Nephelite porphyry, Beemerville, N.J. M. /SNOTSSEZ, anal. Incomplete. 
III. Average( excluding Al,O, of I) of I and II. 

IV. Tinguaite, Monte Mulatto, Predazzo, Tyrol. M. Dittrich, anal. J. Romberg, 
Sitzb. Preuss. Akad. Wiss., I (age), p. 748. 
V. Nephelite porphyry, Wudjavrtschorr, Upmtek, Kola. V. Hackmann, anal. 
V. Hackmann, Fennia, XI (1894), p. 151. 
VI. Tinguaite, Hooper’s Inlet, Dunedin, New Zealand. P. Marshall, anal. P. 
Marshall, Quar. Jour. Geol. Soc., XLII (1906), p. 396. 
VII. Leucitite, Etinde Volcano, Kamerun. M. Dittrich, anal. Sztzb. Preuss. 
Akad. Wiss. (1901), p. 299. 
VIII. Norm of III. Symbols, II. (6) 7. 1.3. Janeirose. 


580 M. AUROUSSEAU AND HENRY S. WASHINGTON 


Chemically the rock is a foyaite and differs in no essential 
manner from the nephelite syenite. Indeed, the similarity of the 
analyses of the nephelite syenite and of the nephelite porphyry is re- 
markably striking. The latter contains less alumina and titanium 
than the former, but is a very close parallel to it otherwise. What has 
been said, therefore, of the affinities of the nephelite syenite applies 
likewise to the nephelite porphyry. The specimen analyzed was 
selected with care, and comes from the neighborhood of the peach 
orchard mentioned. It is, as nearly as possible, the average 
porphyry. 

The rock was analyzed independently by each of us. As the 
summation from Washington’s figures was low, and inspection 
showed that the figures for alumina were probably at fault, the 
average of the two analyses was taken, excluding the alumina from 
I, and correcting that of II only by the deduction from it of one-half 
of the difference between the two determinations of TiO,. Column 
III represents the accepted values. On comparing the rock with’ 
others of a similar nature, it is seen that it resembles the nephelite 
porphyries of other localities. In particular may be mentioned the 
nephelite porphyry of Julianehaab* (Fox Bay type), which is very 
different mineralogically, however, and the nephelite syenite 
porphyry of the Val dei Coccoletti, in the Tyrol. The last-named 
rock is practically the chemical equivalent of the tinguaite of 
Monte Mulatto, Predazzo, and indeed, so great is the chemical 
resemblance of the Beemerville nephelite porphyry to certain 
tinguaitic and leucitic rocks, that we quote the tinguaite of Monte 
Mulatto (IV of Table II) in preference to the nephelite porphyry. 
Column V is the nephelite porphyry of Umptek, a more sodic 
rock, but otherwise similar; while VI and VII are respectively the 
tinguaite of Hooper’s Inlet, Dunedin, New Zealand, and a leucitite 
from Kamerun. The similarity in chemical composition between 
the Beemerville magma and the magma which has produced the richly 
leucitic leucite phonolite of Poggio Muratella, Lake Bracciano, has 
already been pointed out elsewhere. We desire to stress the simi- 

tN. V. Ussing, Geol. Julianehaab, Meddel. om Grinld., XXXVIII (1911), p. 275. 


2J. Romberg, Sitzb., Preuss. Akad. Wiss., I (1911), p. 748. 
3H.S. Washington, The Roman Comagmatic Region (1906), p. 47. 


SYENITE AND PORPHYRY OF NEW JERSEY 581 


larity of the nephelite porphyry of Beemerville to other nephelite 
porphyries and similar rocks, because the Beemerville rock, on the 
basis of a poor analysis, was made the type of the species sussexite. 
This matter will be discussed below. 


CRYSTALLIZATION VARIANTS OF THE BEEMERVILLE MAGMA 


The nephelite porphyry and the leucite tinguaite described by 
Wolff occur within the main mass of nephelite syenite. We 
believe the first to be a dyke of some magnitude, while the second 
is a dyke only fifteen inches wide. ‘The leucite tinguaite, also, 
differs from the nephelite syenite in no other way than in the 
reversal of the ferrous-ferric relation and a change in the rdle of 
sulphur (see II of Table III below). The main mass crystallized 
completely as orthoclase, nephelite, and aegirite. The smaller 
mass (dyke?) of nephelite porphyry is only a textural variant of 
the same magma, while the smallest dyke, of the same chemical 
composition, is a mineralogical variant, having produced a certain 
amount of leucite and little or no primary orthoclase. The symbols 
of the three rocks indicate clearly that there has been no chemical 
differentiation. They are as follows: 

Nephelite syenite, Beemerville, (I) I1.7.1’.3. | 

Nephelite porphyry, Beemerville, II.(6)7.1.3. + Janeirose 

Leucite tinguaite, Beemerville, I1.7(8).1.3. { 

_ They all fall in the same subrang. It may be mentioned here that 
the subrang Beemerose was established from Eakin’s analysis, 

which does not seem to be so representative of the mass as the 
- new analysis here presented. The crystallization variants appear 
to be due to differences in the rates of cooling of the three rocks, 
an assumption based upon the respective volumes of the masses 
concerned. That anephelite syenite magma is capable of producing 
leucitic rock is a matter of great interest, and the presence of 
nearly 12 per cent of leucite in the norm of the Beemerville nephelite 
syenite may be significant in this respect. The great similarity 
of the Beemerville magma to certain tinguaites, as already men- 
tioned, is a matter of like nature, that is, the expression of magmas 
of similar composition in feldspathic and feldspathoid form. In 
view of some results obtained by Morey and Bowen, on the thermal 


582 M. AUROUSSEAU AND HENRY S. WASHINGTON 


relations of leucite and orthoclase, there is no difficulty in accept- 
ing the leucite tinguaite of Beemerville as the rapidly cooled 
equivalent of the nephelite syenite. This interpretation also tends 
to confirm Kemp’s diagnosis of leucite in the true differentiates 
of the magma, the dykes at Rudeville and Hamburg. 


THE OCCURRENCE OF ZIRCONIA AND RARE EARTHS 


In the Beemerville rocks the amount of zirconia is rather high. 
This illustrates the fact that the region east of the Appalachians, 
from Essex County Massachusetts, through New Jersey as far as 
North Carolina, and possibly beyond, is a region rich in zirconia. 
Numerous localities for zircon have been discovered in New Jersey 
(personal communication from Dr. J. E. Wolff), and its distribution, 
and that of the rare earths in places, is so well known in Virginia 
and North Carolina as to need no great comment here. The 
zirconia is not necessarily confined to sodic rocks, and indeed most 
frequently occurs in zircon pegmatites, like the well-known peg- 
matite at Tuxedo, near Hendersonville, North Carolina. The rare 
earths occur mostly in allanite, which has a fairly wide distribution 
in Maryland and Virginia. Our determination of the rare earths 
in the nephelite syenite and nephelite porphyry is the first record 
of them in northern New Jersey. A number of unpublished 
analyses of aegirites, by Washington, indicate that the rare earths 
of the Beemerville rocks occur in the aegirite. The bulk of the rare 
earth precipitates in our analyses was too small to admit of any 
separation being made, but the chemical behavior during the 
determinations suggests that yttrium preponderates over cerium, 
and that thorium is present. 

As the literature of the alkalic rocks of northern New Tee 
is scattered, and in part somewhat old, we append the superior 
analyses of other rocks of the district, with which we have not 
dealt directly here. We have included an inferior analysis of the 
ouachitite of Rutan’s Hill, by Kemp, as there is no other chemical 
information extant concerning the basic lamprophyres. The 
summation of this analysis is low, in spite of the fact that the iron 
is all expressed as Fe,O,, and the water and CO, are merely repre- 


t G, W. Morey and N. L. Bowen, Amer. Jour, Sci., TV (1922), p. I. 


SYENITE AND PORPHYRY OF NEW JERSEY 583 


sented as “loss on ignition.” As regards the other analyses of the 
table, the MnO of II is probably too high, and the non-determination 
of P.O; and TiO, in III of course render the figures for alumina 


too 


high. All other analyses of the alkalic igneous rocks from this 


region we have discarded as unfit for use. 


° TABLE III 
I II Ir IV 
‘Si1O)35 6 baat Ae Si aN 54.68 50.00 40.71 40.47 
EXIWOs 5 cele Ea Gea Ie ene ee 21.63 20.03 19.46 11.86 
12210). 3 SAS A AR ae 2.22 0.98 7.46\ 
IDO), ies ote er aera 2.00 3.98 6.83f BU es 
IMO) sidiae 6 6c oar eae it Dy °.69 One BELO 
CTO eee ie toilsliyoecnts 2.86 3.41 11.83 16.80 
INFAO) 6, ol co Rene ao ana co ea 7.03 8.28 1.80 1.90 
ne 5, Sic Si aR Ae ee eat 4.58 ce 3.26 4.21 
A OTP agit BBP OREN NG con ERCRRE 1.88 I.50 
1EL(OE> 5 5 Gee ae eee 0.27 o.10f T.53 ge 
(COs: cuunacobessdgeuore bc None 0.22 Ronse lst cal | PEP Tay atte 
UNOS; oS See eee eee ©.79 O1OOS hall ecraeek te ee nalll eerie erate 
AO Roe es br Bowes 0.28 Ro Ida arty ie, mr eA UY Ath Lautan nrean anal 
SO, ooo od aCCbod boo oODKoOoOOE OsOQVF  Wecocvcccgccclloacasaccadaoooudsaoana08 
Cle 3.675 Go alone ie eens None ERI) AM eencients aeues aoe aren eee 
EI saat cpasusisn curio’ 0.22 FO1G Ie ie, ese acre orn A Ae tate 
IBC o 6 odo leh Sateen a eRe Re eree ete OlsG Ai 3 Meena ee ane vel Eye eyen Pete 
NIM OMe ns eee ‘Wie, 0.50 CASwN il otras sidan oe 
a O ean eee iN Sollee. 0.05 DNIKG} CV Su URL [Shay a cher Greve ell entarn aia seme oe 
SUT ame neice pein: 99.81 99.87 100.01 99.38 
WES PR eric naa: OOO: leecctrere is 0 ole ee errr wyotin | Beene para naae 
Sumamee seers ectocn 99.72 99.87 100.01 99.38 
I. Nephelite syenite, Brookville, Hunterdon Co., N.J. G. Steiger, anal. Amer. 


Kri 


Jour. Sci., VIII (1899), p. 423- 


. Leucite tinguaite, Beemerville, N.J. J. E. Wolff, anal. Bull. Mus. Comp. Zodl. 


Harvard, XXXVIII (1902), p. 276. 


. Minette, Franklin Furnace, N.J. L.G. Eakins, anal. U.S. Geol. Surv. Bull. 150 


(1808), p. 238. 


. Ouachitite, Rutan’s Hill, Beemerville, N.J. J. F. Kemp, anal. Amer. Jour. 


Sci., XXXVIII (1889), p. 133. 


THE STATUS OF SUSSEXITE 


Brégger, in working out the grorudite-tinguaite suite of the 
stianiagebiet, extrapolated from his analyses of the series, 


and calculated an end member for the suite, the composition of 
which is shown in column II of Table IV. No rock corresponding to 
this hypothetical composition was found in the Kristianiagebiet, 


584 M. AUROUSSEAU AND HENRY S. WASHINGTON 


but Brogger considered that Kemp’s analysis of the Beemerville 
nephelite porphyry, together with the description, indicated that a 
rock of the hypothetical, calculated composition, and corresponding 
to an end member of this series, actually existed. He therefore 
defined the species now known as sussexite,’ making the Beemerville 
rock the type, in the following terms: 


Gesteine wie diejenigen von Beemerville waren aber friiher nur wenige be- 
kannt, jedenfalls nur wenige analysirt; sie bilden einen ganz distincten chem- 
ischen Typus, wie sie auch geognostisch durch ihre haiufige Verkniipfung mit 
Nephelinsyeniten charakterisirt sind; dementsprechend sind sie reich an Alka- 
lien, arm an Kalk und Magnesia und massig reich an Eisenoxyden, aber mit 
sehr hohem Al,O; Gehalt. Es ware entschieden irreleitend, diese Gesteine als 
Nephelinite zu bezeichnen, nur deshalb weil sie aus Nephelin (und Aegirin) 
bestehen. 


A sussexite, according to Broégger’s definition, is a nephelite 
porphyry either poor or lacking in orthoclase, and therefore a 
persodic or dosodic rock, resembling the urtites and ijolites in com- 
position. 


TABLE IV 
I II Til 

SiO ps hatnig eae 45.18 45.0 47.43 
FAIL OF ase whet vegans ota ale 23.31 25.0 23.60 
Fe,03\| 4.59 
FeO { oD O00 COG Oo Co 6.11 6.5 { I.20 
IIE Oech cho eee Aitene 6 otee I.45 To§ i, 26) 
CaO Pe ponee sa ceenta eet os 62 2.0 0.67 
IN aOR eto ee Ir.16 12.0 15.08 
Ke OFS O oarer ena: 95 TO 2.00 
1 Ec Sh aes eR So I.14 TiO. | tt See 
DEO Pere sre See srs ae Bae cote | Mek MCDA e sizes castrate leet eee Rees ©.10 

SHON es cust oie o 98.92 100.0 99.09 


I. Nephelinitic facies of nephelite porphyry, Beemerville, N.J. J. F. Kemp, anal.(?) 
J. F. Kemp, Trans. N.Y. Acad. Sci., XI (1892), p. 67. Brégger’s type for the 
species Sussexite. 

II. Hypothetical Sussexite, calculated by extrapolation from the Grorudite-Tinguaite 
series. W. C. Brégger, Eruptivgest. des Kristianiagebietes, I (1894), p. 172. 

III. Sussexite, Penikkavaara, Kuusamo, Finland. M. Dittrich, anal. V.Hackmann, 
Bull. Comm. Géol. Finlande, XI (1900), p. 22. 


Kemp’s analysis (see I of Table IV) indicates that the sample 
he analyzed corresponded fairly closely to Brégger’s definition of 


«W. C. Broégger, Eruptivgest. des Kristianiageb., I (1894), p. 173- 


SYENITE AND PORPHYRY OF NEW JERSEY 585 


the type. It is certainly dosodic. We have shown here that the 
nephelite porphyry of Beemerville is a sodipotassic rock, in no 
important respect different from other nephelite porphyries, and 
we can only conclude that the sample chosen for analysis by Kemp 
was not representative. Kemp’s own descriptions of the nephelite 
porphyry show that the rock he described was not abnormally 
poor in orthoclase. Consequently the Beemerville rock cannot 
maintain its position as the type sussexite. Only one rock cor- 
responding to Brogger’s original definition has been analyzed so 
far, it being the sussexite of Kuusamo, Finland, described by 
Hackmann (see III of Table IV). As it establishes the existence 
of the species (the existence of which may be said to have been 
predicted by Brogger), the name sussexite should remain in use, 
in the sense of Brogger’s definition. Sussexite is essentially a 
nephelite porphyry devoid of feldspar, or, in other words, a porphy- 
ritic urtite. 

Rocks of the nephelite syenite family tend to lack homogeneity 
within the mass, and too much care cannot be exercised in the 
selection of material for analysis, which will correspond well with 
the material upon which the petrographic descriptions are based. 
Another instance showing lack of correspondence between the 
chemicai analysis and the mineralogical description is the mariu- 
_ polite described by Morozewicz.t The analysis of this rock does 
not permit of the existence of the amount of nephelite it is said to 
contain (according to the description). 

The Beemerville nephelite porphyry has been widely accepted 
as the type of sussexite. Iddings calculated the ratio 


Na,O+K.0 
SiO, 
for Broégger’s grorudite-tinguaite series and found that the Beemerville 
rock, from Kemp’s figures, lay upon a prolongation of the approxi- 
mately straight line representing the series in the diagram.* The 
Beemerville rock, however, is not a differentiate at all, but, as we have 
shown, a textural variant of the main mass of the nephelite syenite. 


tJ. Morozewicz, TMPM, XXI (1902), p. 230. 
2J. P. Iddings, Jour. Geol., II (1895), p. 357. 


586 M, AUROUSSEAU AND HENRY S. WASHINGTON 


SUMMARY 


The scattered contributions to the geology and petrology of 
the alkalic igneous rocks of northern New Jersey are reviewed in 
chronological order, and a general account of these rocks is given. 

The large mass of nephelite syenite northwest of Beemerville 
is described and is interpreted as a lenticular sill or a flat laccolith 
of foyaite, intruded by a mass of nephelite porphyry (probably a 
dyke) and by a small dyke of leucite tinguaite. 

New analyses of the nephelite syenite (foyaite) and of the 
nephelite porphyry are presented, and the affinities of these rocks 
and of the leucite tinguaite are discussed. It is concluded that 
these three rocks are textural and mineral variants, without chemical 
differentiation, of the same magma. 

It is shown that the nephelite porphyry is not a sussexite, as 
formerly supposed, and the status of sussexite as a rock variety 
is considered, with the conclusion that the name should be retained 
in its original sense, but that the nephelite porphyry of Beemerville 
can no longer be regarded as the type of the variety. 

The presence of zirconium and the rare earths in the Beemerville 
' rocks has been demonstrated, and the wide distribution of these 
elements in the region east of the Appalachians is briefly discussed. 


Wasuincton, D.C. 
April, 1922 


INTRAFORMATIONAL CORRUGATED ROCKS 


WILLIAM J. MILLER 
Smith College, Northampton, Massachusetts 


CONTENTS 

INTRODUCTION 
DIFFERENTIAL MOVEMENT ACCOMPANYING THRUST FAULTING 
DIFFERENTIAL MOVEMENT ACCOMPANYING NORMAL FAULTING 
DIFFERENTIAL MOVEMENT ACCOMPANYING REGIONAL FOLDING 
DIFFERENTIAL SQUEEZING ACCOMPANYING REGIONAL FOLDING 
DIFFERENTIAL MOVEMENT UNDER THE ACTION OF GRAVITY 
SUBAQUEOUS GLIDING OR SLUMPING 
ACTION OF ICE 
DIFFERENTIAL WEIGHTING 
CRYSTALLIZATION AND HyDRATION 
PRESSURE OF INTRUDING MAGMAS 
ACTION OF MacmatTic INJECTION 
MacmatTic FLOWAGE 

_ INTRODUCTION 


It is the purpose of this paper to enter into a general discussion 
of intraformational contorted rocks, with emphasis upon the modes 
_ of origin of the different types. During the last eighty years various 
examples have been described, but it is surprising how few are the 
cases which have been discussed in sufficient detail to furnish the 
- data necessary for their classification on the basis of origin. Evi- 
dently these interesting and often puzzling structures have not 
received the attention which they deserve. In the present attempt 
to make a tentative genetic classification of intraformational corru- 
gations, examples, mainly from American localities, are given to 
illustrate the various types. Because of the lack of certain critical 
data, it is difficult or impossible to be very sure of the proper 
classification of some of the cases. 


DIFFERENTIAL MOVEMENT ACCOMPANYING THRUST FAULTING 


Excellent examples of corrugated strata between practically 
undisturbed strata occur in the walls of the postglacial gorge at 


587 


588 WILLIAM J. MILLER 


Trenton Falls, New York, where the contorted beds lie at two dis- 
tinct horizons within the Trenton limestone formation, there about 
300 feet thick. The lower contorted zone, 4 to 6 feet thick, is visible 
only near the crest of the lower part of High Fall and in the upper 
end of the gorge near Prospect Village where the strata are highly 
inclined against a thrust fault surface. ‘The upper contorted zone, 
5 to 15 feet thick, is well exhibited along the path opposite High 
Fall (Fig. 1, upper part) from which place it is clearly traceable in 
the walls of the gorge for nearly 2 miles northward to Prospect 
(Fig. 1, lower part). 


ZF 
DSS SER 


eae 
(3 Gee) Gee GS 
ZB Sa SE 
fos A EE) 


Fic. 1.—Upper figure: sketch of part of the upper contorted zone at Trenton 
Falls, New York. Lower figure: north-south structure section showing the positions 
of the contorted zones within the Trenton formation and their relation to the thrust 
fault at Trenton Falls, New York. 


In both the undisturbed and the corrugated portions of the 
formation the impure limestone layers average only a few inches 
in thickness, and they are separated by thin partings of shale. 
Within the contorted zones the strata are in some cases scarcely 
disturbed; in some cases they are only gently folded or tilted; 
but most commonly by far they are highly folded, contorted, and 
even fractured. 


INTRAFORMATIONAL CORRUGATED ROCKS 589 


The explanation offered by the writer is that the contorted 
zones were produced by differential movements within the mass of 
the Trenton limestone. The displacement (140 feet) of the thrust 
fault at Prospect Village was sufficient to cause the beds of the 
middle Trenton to be shoved over the upper Trenton. Figure 1 
(lower part) shows the relation of the contorted zones to this fault. 
It is easy to see how, when the force of compression was brought to 
bear in this region, the higher Trenton beds on the upthrow side 
must have moved more easily and consequently faster than the 
lower Trenton beds. For instance, the portion A in Figure 1 
(lower part) being separated from C by an intermediate mass B 
of possibly slightly less rigidity moved over C and caused the 
portion B to become ruffled or folded and fractured because this 
portion took up most of the differential movement. The portion B 
needed to be only slightly less rigid than the adjacent strata and 
this slightly reduced competency is due to somewhat thinner 
limestone layers separated by relatively thicker shale partings. 
A similar explanation applies to the lower contorted zone. Accord- 
ing to this explanation the corrugated zones indicate horizons along 
which the differential movements took place, and no great amounts 
of differential movement were necessary to produce the contortions. 

Grabaut' states that 

These disturbances at Trenton Falls have been variously explained, the 
general conclusions of geologists being either: (1) that they were truly tectonic 
—lateral pressure having resulted in the folding of certain strata while others 
took up the thrust without deformation, or (2) that they were due to squeezing 
out of certain layers under the weight of overlying rock masses. Both explana- 


tions are unsupported by the detailed characteristics of the folds and their 
relationship to the enclosing layers. 


The writer agrees that these two explanations must be ruled out, 
but he does not agree with Grabau who accepts Hahn’s? subaqueous 
slumping or gliding hypothesis, described later in this paper, without 
even mentioning the tectonic differential slipping hypothesis (above 
outlined) which was first applied by the writer? to the Trenton 


tA. W. Grabau, Principles of Stratigraphy (1913), p. 784. 
2. Hahn, Neues Jahrb. Beil., Vol. XXXVI (1913), pp. 1-40. 
3 W. J. Miller, Jour. Geol., Vol. XVI (1908), pp. 428-33. 


590 WILLIAM J. MILLER 


Falls contorted zones in 1908, and further elaborated in 1915.' 
Arguments against Hahn’s hypothesis are there entered into some- 
what in detail. In the opinion of the writer, differential movement 
(not always as an accompaniment of thrust faulting) is involved in 
many, if not most, cases, of intraformational corrugations. 
Intraformational contorted beds of essentially the same origin 
as those at Trenton Falls, only on a larger scale, occur within the 
straight-bedded limestones of Turtle Mountain, Frank, Alberta, 


} 


TURTLE MF 


. 
i 


< 
SoM, fissures 


SN 
SS 
c> 

{> 


SSS 
S 
~ 


i 


old bed of Old Man River 


Lake 


Veet MN _ 
YM 


Paleozoic limestones Cretaceous shales end sandstones 


RA/ 


Fic. 2.—Section showing the position of the contorted limestones and the relation 
of the contorted beds to the great thrust fault at Frank, Alberta, Canada. (After 
R. W. Brock.) 


Canada. Accompanying Figure 2 after Brock? shows the position 
of two corrugated zones of more shaly material intercalated between 
straight beds of limestone and parallel to a great thrust fault on its 
upthrow side. The writer would explain these corrugations as 
due to differential slippings within the limestone with resultant 
crumpling of the more shaly layers during the process of thrust 
faulting of the Paleozoic limestone over the Cretaceous strata. 
Many years ago Logan,3 in his description of a section 1,210 
feet in thickness of Devonian strata on the Forillon peninsula of 
W. J. Miller, N.Y. State Mus. Bull. 177 (1915), pp. 135-43. 


2R. W. Brock, Dept. Interior Canada, Ann. Rept., Part 8. 
3 W. Logan, Geol. Can. (1863), p. 391. 


INTRAFORMATIONAL CORRUGATED ROCKS 591 


Gaspé in the Gulf of St. Lawrence, called attention to a remarkable 
example of intraformational corrugations. Logan says: 


It would appear as if the layers, after their deposit, had been contorted by 
lateral pressure, the underlying stratum remaining undisturbed, and had then 
been worn smooth before the deposition of the next bed. Where the inverted 
arches of the flexures occur, some of the lower layers are occasionally wanting 
as if the corrugated bed had been worn on the under as well as the upper side. 
The corrugations are precisely in the direction of the dip, and the peculiarity is 
not confined to a small part of the deposit. 


He states that the same structure occurs at localities a mile apart. 


Fic. 3.—Contorted strata within Devonian limestone at Cape Gaspé, Quebec. 
(After J. M. Clarke.) 


_ John M. Clarke, who has observed the Gaspé occurrence, and 
who has kindly permitted the use of the accompanying picture 


(Fig. 3), says: 

Crinkled strata lying between strata which show no evidence of dislocation 
are not of infrequent observation but, in most of the recorded instances, the 
crinkled layer is of softer stuff (that is, a highly aluminous mud rock) than the 
rigid beds above and below. The brilliant exhibition of this phenomenon on - 
the cliffs of Cape Gaspé, first sketched by Sir William Logan, is not of this 
character. Here the middle deformed beds are of thin limestone leaves like 
those which bound them. They are crumpled into sharp, much involved and 
overlapping curves in which the limestone plates are broken sharply across. 
It seems very doubtful if any other explanation can be brought forward for 


592 WILLIAM J. MILLER 


this exceptional occurrence than that generally adopted for those of the first 
named category; a sliding of soft sea bottom deposits on a sloping surface under 
gravity, helped forward perhaps, if on a large scale, by earthquake shock or some 
other jolt-like impulse. ... . It follows from the conception of these structures 
that the deformation was contemporaneous, and preceded the deposition of the 
overlying beds.? 


For the reasons below listed the writer is strongly inclined to the 
conclusion that the corrugated zone at Gaspé developed as a result 
of differential movement within a great block of strata while it was 
being thrust faulted. (1) If we accept the hypothesis of subaqueous 
sliding, it is necessary to attribute the nearly straight upper surface 
truncating the contorted zone to erosion, but the very character 
and uniformity of both the contorted and inclosing strata seem to 
render this extremely improbable; (2) the worn character of both 
the upper and lower surfaces of the contorted zone, with certain 
beds locally missing as noted by Logan, are best explained on the 
basis of differential movement within the mass of limestone; (3) that 
the conditions were very favorable for differential movement is 
borne out by the fact that the contorted zone lies not far from the 
bottom of a large block of Devonian strata which has been exten- 
sively thrust faulted over Cambro-Ordovician strata; (4) the strike 
of the contorted zone is approximately parallel to the strike of the 
thrust fault; (5) the corrugations conform to the direction of dip of 
the inclosing beds; (6) according to Logan’s section, the corruga- 
tions developed just where the conditions were most favorable for 
differential movement, that is, in the weakest (most shaly) part 
of a mass of the strata 200 feet thick with appreciably more arena- 
ceous and resistant strata above and below this mass; and (7) the 
rather puzzling occurrence of the two nearly straight, thin beds 
within the contorted zone (see Fig. 3) may be more reasonably 
explained on the basis of differential movement than on the basis of 
subaqueous gliding because, as revealed by careful examination 
of the figure, these layers, as well as the layers capping the con- 
torted zone, show clear evidence of having been deformed and even 
cut across locally by the corrugations, thus indicating that the con- 
tortions took place after deposition of those layers. 


t Personal communication from Dr. Clarke. 


INTRAFORMATIONAL CORRUGATED ROCKS 593 


DIFFERENTIAL MOVEMENT ACCOMPANYING NORMAL FAULTING 


Up-drag of strata commonly takes place as a result of friction 
during movement on the downthrow sides of normal faults. Under 
proper conditions drag folds of the nature of intraformational corru- 
gations develop as a result of differential movement of the bending 
strata. A good example has been observed by the writer in the bed 
of the Connecticut River just below the dam at Holyoke. A normal 
fault is clearly traceable in the bed rock across the river a few rods 
below the dam. From the fault south for 150 yards the strata 
(sandstone and shale) show dips of 25 to 40 degrees to the south due 
to up-drag, and they strike parallel to the fault. About 150 yards 
south of the fault there is a notably disturbed zone of thin-bedded 
dark shale with strike parallel to that of the fault. It shows clearly 
along the strike for 200 yards. The disturbed zone, varying in 
thickness between 4 and 1o feet, is overlain by fairly well-bedded 
red sandstone, and underlain by thin-bedded dark shale much like 
that of the disturbed zone. Next below there is sandstone. The 
folded zone does not terminate abruptly at either summit or base, 
but the top is much the more regular, coming close to the over- 
lying sandstone. Even the sandstone, for a foot or so above the 
shale contact, is locally somewhat bent. The whole body of the 
rock in the bed of the river south of the fault plainly shows the effects 
of differential movements which took place during the process of 
normal faulting. One belt of weak, thin-bedded shale overlain 
by relatively rigid sandstone became moderately corrugated, the 
corrugations being of the nature of drag folds produced by differ- 
ential movements. That the corrugations must have developed 
after the deposition of the overlying sandstone layers is proved by 
the fact that the lower part of the sandstone is in many cases moder- 
ately bent just like the immediately underlying shale. It seems 
impossible to escape the conclusion that this corrugated zone is of 
tectonic origin, that is, the result of differential movement accom- 
panying normal faulting. 


DIFFERENTIAL MOVEMENT ACCOMPANYING REGIONAL FOLDING 


The principle of differential movement of strata is well illus- 
trated in many regions of notably folded rocks. Differential move- 


504 WILLIAM J. MILLER 


ments take place between the relatively more competent beds with 
not uncommon development of corrugations of the nature of drag 
folds between them. According to Leith: “‘The stronger beds tend 
to assume the ‘parallel’ type of folds in which the principal adjust- 
ment is between the beds rather than within them. This readjust- 
ment or slipping is concentrated in the intervening weaker layers. 
The folds of the weaker layers are really ‘drag folds’ due to differ- 
ential movement between the controlling harder layers.”* 

Excellent examples of intraformational corrugated strata 
believed to have resulted from differential movement accompany- 
ing regional folding were observed by the writer some years ago 
at Baldhead Cliff near Ogunquit, Maine. The perfectly stratified 
thin-bedded rocks are there interbedded quartzite and phyllite. 
That the region has been subjected to severe lateral compression 
is evidenced by the fact that the strata stand in nearly vertical 
position. Figure 4 is a ground plan sketch showing the detailed 
structure of one of these corrugated zones about g feet thick, 
although the quartzite layers are really less conspicuous than indi- 
cated in the diagram. The corrugated zone, consisting very largely 
of phyllite, is not sharply delimited from the adjacent straight layers 
of predominant quartzite on either side. In most cases slight 
faulting has taken place along the axes of the sharp folds, but, as a 
rule, individual sharp folds or faults rarely extend all the way across 
the contorted zone. The folds are uniformly overturned toward the 
east. Fracture cleavage is well exhibited in the phyllite layers a 
few inches thick which lie within the quartzite on either side of the 
contorted zone. The cleavage cracks are uniformly inclined toward 
the east, as are the folds. ‘‘When a slate or shale is folded between 
two competent layers, such as quartzite, the cleavage produced in 
the slate affords clear evidence of slipping or shearing between the 
quartzite beds.’” 

All the evidence points to the tectonic genesis of the above- 
described intraformational corrugated strata. The corrugations 
must have developed after all the strata were deposited because 
the folded zone grades into the non-folded strata on either side. 


«C.K. Leith, Structural Geology (1913), Pp. 114. 
2 Ibid., p. 119. 


INTRAFORMATIONAL CORRUGATED ROCKS 595 


Under conditions of differential movement the wide belt of predomi- 
nant phyllite yielded by development of corrugations, while the 
thin bands of phyllite between much quartzite on either side yielded 
by development of fracture cleavage. 

In the Marquette synclinorium of Michigan intraformational 
corrugations are shown on large scales in the slate formations which 
lie between quartzite formations. 

During the summer of 1921 the writer was impressed by the 
fine exhibitions of intraformational contorted strata of Proterozoic 


Fic. 4.—Ground-plan sketch of intraformational corrugated phyllite and quartzite 
near Ogunquit, Maine. The black lenses are quartz. Contorted zone about 9 feet 
wide. 


age on both small and large scales in Glacier National Park. In 
the wall of the great Swiftcurrent cirque, only a few rods from the 
trail, the strata are notably folded and even crumpled through a 
thickness of 30 to 4o feet, and for a distance of a few hundred feet, 
while on all sides the strata are undisturbed except for the moderate 
regional tilting. The folding rocks are not sharply separated from 
the others. On the grand scale the strata are very irregularly con- 
torted through a thickness of hundreds of feet in the face of the 
mountain between Gunsight Lake and Gunsight Pass with non- 
contorted strata above, and also at the same general horizon on 
either side. These contorted beds have quite certainly resulted 
from local differential movements within the great body of Protero- 
zoic strata either during the development of the synclinal structure 


596 WILLIAM J. MILLER 


of the park, or during the tremendous process of thrust faulting of 
the district, or both. 
In a discussion of the highly folded gold-bearing series of Nova 
Scotia, Faribault" has figured and described some interesting 
cases of intraformational corrugated quartz veins in slate lying 
between beds of quartzite. He says: ‘‘Interstratified (quartz) 
veins often exhibit a remarkable folded or corrugated structure 
within the beds of slate that contain them. The corrugations, or 
crenulations, usually occur at or near the apex of the anticline, and 
run parallel with one another and in a direction approximately par- 
allel with the axis of the fold.’”’ He believes (1) that the veins were 
formed during the folding of the region; (2) that, due to differential 
motion within the relatively weak or plastic slate containing veins 
which were formed early in the folding process, the veins and inclos- 
ing slate were corrugated; and (3) that such motion resulting in 
corrugations took place mainly at the apexes of the folds. 


DIFFERENTIAL SQUEEZING ACCOMPANYING REGIONAL FOLDING 


Lateral pressure may result in the folding of certain weaker 
strata while adjacent more resistant strata take up the thrust either 
without so much folding or by being fractured instead of folded. 
Intercalated beds of limestones are especially likely to yield in this 
manner. Interesting effects of differential squeezing in the folded 
Algonkian strata of the Marquette district of Michigan have been 
described by Van Hise, Bayley, and Smyth? who say: 

Along the contacts of the (Kona) dolomite beds and the quartz(ite) layers 
accommodation was necessary, and in places a bed of limestone may be seen 
bent into a series of anticlines and synclines, the overlying quartzite not being 
similarly bent, but being compressed and brecciated, thus making a pseudo- 
conglomerate. .... When the series was folded the more plastic limestone 
yielded to the pressure, in both a major and a minor way, by folding, while the 
brittle quartzite was fractured through and through, the movement of the 
fragments over one another, and of the beds as a whole, being sufficient to 
truncate the minor waves of the marble. 


«EF. R. Faribault, Can. Geol. Surv., Guide Book No. 1, Part 1 (1913), pp. 174-88. 
2 Van Hise, Bayley, and Smyth, U.S. Geol. Surv., Mon. 28 (1897), pp. 242-43. 


INTRAFORMATIONAL CORRUGATED ROCKS 597 


Even in this case some shearing action or differential slipping was a 
factor in the process. 
Prouty* has described the crumpling of thin beds of marble 
between thick beds which latter yielded by fracturing and faulting. 
The principle of differential squeezing appears to be not uncom- 
mon in various regions of folded strata. 


DIFFERENTIAL MOVEMENT UNDER THE ACTION OF GRAVITY 


Fine examples, believed by the writer to belong in this category 
of intraformational contorted strata, are to be found in the post- 
glacial clays of various regions. The following observations made 
by the writer upon the clays in and near Northampton, Massachu- 
setts, give a fair idea of the nature of the foldings and their origin. 
In the bank of the Connecticut River 2 miles east of Northampton, 
12 to 14 feet of nearly horizontal, thin-bedded, perfectly stratified 
clays are overlain by 10 to 15 feet of stratified sands. The clay 
contains two contorted zones—a lower one 1 to 2 feet thick, and an 
upper one 4 to 8 inches thick—separated by 8 or 9g feet of the 
ordinary non-contorted clay. ‘These corrugated zones are clearly 
traceable for several hundred feet. Immediately above each con- 
torted zone, the non-contorted layers are in many places somewhat 
wavy or slightly folded. Different portions of the same contorted 
zone show different degrees of folding, the clay beds in some cases 
‘being only moderately folded, while in others they are intensely 
twisted, pulled apart, and even overturned. Some of the straight 
beds contain notable amounts of very fine sand, but the contorted 
‘beds consist of distinctly less sandy clay. The corrugations almost 
invariably have strikes parallel not only to each other but also to 
the notable dip (several degrees) of the clay beds in general. The 
under surface of each contorted zone is usually very straight, while 
the upper surface is commonly somewhat irregular (Fig. 5). 

In the South Street clay pit of Northampton the writer has ob- 
served a very fine highly contorted zone of clay sharply inter- 
calated between beds of clay whose stratification surfaces are almost 
straight. The straight beds consist of alternating, very fine-grained, 


1W. F. Prouty, Geol. Surv. Ala., Bull. 18 (1916), p. 170. 


598 WILLIAM J. MILLER 


sandy clays and pure clays, while sandy clays are distinctly less 
conspicuous among the contorted beds. Figure 6 represents a 
detailed sketch of part of this corrugated zone, a feature of excep- 
tional interest being the only slightly disturbed layer of fine-grained, 
sandy clay lying in the midst of the contortions. This contorted 
zone about 8 inches in thickness may be traced for a number of rods 
in the walls of the clay pit, but its full extent is unknown. Above 
it about a foot there is another corrugated zone lying between prac- 


Fic. 5—Highly contorted zone (8 inches thick) of clay between practically 
undisturbed beds of clay and sandy clay in the bank of the Connecticut River 2 miles 
east of Northampton, Massachusetts. 


tically undisturbed beds. Within the clay pit the beds show a very 
appreciable dip of at least several degrees to the southeast. 

It seems impossible to explain intraformational contorted clays 
like those just described except as a result of differential movements 
after the clays overlying the contortions were deposited. An 
explanation commonly given for such phenomena, but rarely if 
ever supported by anything like reasonable proof, is that the con- 
tortions were caused either by ice thrust, or the bumping or crowding 
of icebergs on surface layers which were afterward covered by more 
clays. Some facts opposed to such a hypothesis are: (1) the 
remarkably uniform thinness of the corrugated zones of such wide 
extent which could hardly have resulted from ice action upon 
surface layers, the development of such zones under considerable 


INTRAFORMATIONAL CORRUGATED ROCKS 599 


weight of overlying materials being far more plausible; (2) there 
are no notable irregularities or depressions at the tops of the corru- 
gated zones such as must have developed in the case of crumpling 
of surface layers with these depressions first filled by the succeeding 
deposits; (3) the moderate disturbance of the beds immediately 
overlying the corrugated zones, while those just underneath are 
distinctly straighter, strongly point to differential movements after 
the overlying beds were laid down; and (4) the only slightly dis- 
turbed thin bed of fine sandy clay in the midst of the contorted 
South Street clays, as well as the relatively straight beds of similar 
material just above and below the contortions, are best accounted 


Fic. 6.—Contorted zone in South Street clay pit of Northampton, Massachusetts 


for on the basis of differential movements, the thin, clay-rich, 
very plastic beds having yielded by crumpling, while the much less 
_ plastic sand-rich beds did not crumple. It is out of the question 
to look upon the upper surfaces of the corrugated zones as erosion 
surfaces because these perfectly stratified, thin-bedded, postglacial 
clays everywhere plainly show that they were deposited without a 
break in very quiet water. 

The hypothesis of differential movement does not necessarily 
preclude the possibility of subaqueous slumping for it is plausible 
to think of differential movements within masses of the clays which 
may have shifted more or less down the gently sloping delta fronts 
in the postglacial lake of the Connecticut Valley. The action of 
gravity alone, or of gravity aided by an occasional earthquake, 
may have caused the movements. It is more likely, however, that 


600 WILLIAM J. MILLER 


the movements mostly took place much later, that is, since the 
river has cut deeply into the distinctly dipping clay deposits so 
that, under the action of gravity or gravity aided by earthquake 
shocks, overlying beds have moved differentially over lower-level 
beds in the general direction of the dip. In some local cases 
differential movements within clays may possibly have been caused 
by the crowding action of ice against upper portions of clay deposits 
as explained later under another caption. 

A few examples of apparently similar intraformational clays 
from other regions will be cited, with various explanations which 
have been offered to account for them. In regard to corrugations 
in clays near Boonville, New York, Vanuxem’ eighty years ago said: 
“These interesting forms of disturbance were no doubt the result 
of unequal, local, and lateral pressure.”’ A very similar excellent 
example of intraformational sand and clay in the Devil’s Lake 
region of Wisconsin is figured and described by Salisbury and 
Atwood,? who say: “‘The grounding of an iceberg on the surface 
before the overlying layers were deposited, or the action of lake 
ice, may have been responsible for the singular phenomenon.”’ 

M. E. Wilson has described and figured} interesting cases of 
intraformational contorted and broken clays in Timiskaming 
County, Quebec, and in regard to their origin he says: ‘‘ Whatever 
the cause of these peculiar deformational structures, it is evident 
that they were contemporaneous with deposition, for the stratifica- 
tion is uniform in both the overlying and underlying beds.” 

In Albany County, New York, according to Nason, 


A layer of blue clay about a foot in thickness and one hundred feet long is 
crumpled and gnarled, appearing as though its laminae had been disturbed by 
some dragging or shoving weight, while above and below the layers are exactly 
parallel and wholly undisturbed. ... . Bearing in mind the fact that the clay 
banks are underlain by sand, the water circulating through these sands gradually 
undermines the clay bank and tilts it to such an angle that one part of a bed 
would slide over the other, only leaving visible marks along the particular 
stratum disturbed, and in the form of crumplings. Many of the clays lie at an 
angle to the horizon, and only a slight tilt would suffice to give rise to a slip.4 


tL. Vanuxem, Geol. N.Y., Part 3 (1842), p. 215. 

2 Salisbury and Atwood, Jour. Geol., Vol. V (1897), p. 143. 

3M. E. Wilson, Can. Geol. Surv., Mem. 103 (1918), p. 142 and Pls. r5—16. 
4F.L, Nason, N.Y. State Mus. Rept. 47 (1893), p. 465. 


INTRAFORMATIONAL CORRUGATED ROCKS 601 


It is to be noted that Nason’s explanation clearly involves the 
principle of differential movement which is advocated by the 
present writer as the cause of most intraformational contorted clay 
beds. 

SUBAQUEOUS GLIDING OR SLUMPING 


The hypothesis of subaqueous gliding has been elaborated by 
Hahn‘ who has considered most cases of intraformational contorted 
strata to belong in that category. According to Hahn’s hypothesis 
a lenslike mass breaking loose from any cause (e.g., earthquake 
shock) would glide down a subaqueous slope and, because of the 
striking of some obstacle on the bottom and increased friction and 
water pressure, the gliding mass would come to rest only after it had 
become considerably deformed or contorted. Sediments would then 
be deposited in normal order on top of the crumpled layer. The 
most intense folding would be toward the front of the transposed 
mass, and of course the strike of the folds would be at right angles 
to the direction of the moving mass. Conditions for such gliding 
are regarded as favorable at many places on the marginal sea bottom. 

In the paper above cited, Hahn especially refers to the intra- 
formational contorted zones at Trenton Falls, New York, as typical 
examples of submarine slumping among ancient strata. In a 
paper already published the writer has given reasons for believing 
_ that Hahn’s hypothesis cannot possibly account for the Trenton 
Falls occurrences.2 Hahn has regarded most cases of intercalated 
corrugated strata as results of subaqueous gliding, while the present 
- writer regards most of them by far as results of differential move- 
ments within the masses of strata. 

T. C. Brown: has described what appears to be a clear case of 
intraformational folding in Paleozoic strata near Bellefonte, Penn- 
sylvania. In regard to this occurrence Brown says in part: “At 
periodic intervals these beds of calcareous mud and intermingled 
pebbles slumped or slid along the bottom under the influence of 
gravity. At the time of the slump or slide the matrix around the 
pebbles consisted of incoherent lime mud or paste. As it moved 


« F, Hahn, Neues Jahrb. Beil., Vol. XXXVI (1915), pp. 1-41. 
2W. J. Miller, V.Y. State Mus. Bull. 177 (1915), Pp. 140-43. 
3T. C. Brown, Jour. Geol., Vol. XXI (1913), pp. 241-43. 


602 WILLIAM J. MILLER 


it developed unsymmetrical waves or ripples in its mass.” After 
the mass came to rest more lime mud was deposited upon its surface. 
Since then the whole has been solidified. The contorted zone shows 
notable variations in thickness from a few inches to several feet 
(Fig. 7). Certain criteria which seem to rather definitely place 
this occurrence in the category of subaqueous gliding, and which 
particularly distinguish it from the above-described examples 
believed to have resulted from differential movements after deposi- 
tion of the overlying layers, are the following: (1) the notable 
variations in thickness of the contorted zone locally, even within a 
few feet; (2) the very irregular upper surface of the folded zone, 


Fic. 7.—Folded limestone and limestone-conglomerate, several feet thick, between 
non-folded beds near Bellefonte, Pennsylvania. (After T. C. Brown.) 


and the rather regular under surface; (3) the bulging of the immedi- 
ately overlying strata over the little anticlinal folds; and (4) the 
distinct evidence of the filling of the depressions on the upper surface 
of the corrugated zone before the general layers of overlying ma- 
terials were laid down. 

D. W. Johnson has kindly allowed me to reproduce a picture 
(Fig. 8) of a moderately corrugated zone within the cross-bedded Tri- 
assic red beds near Kanab, Utah. Regarding the occurrence he says: 


The crumpling and faulting must have taken place during the process of 
deposition, for the erosion plane beveling the deposit a few inches above the 
corrugations, and upon which the next layer of cross-bedded sand was deposited, 
shows no disturbance. I have therefore attributed the corrugations and 
miniature faults to slumping or settling of the deposit as it was built forward, 
delta-like, under the influence of current action.t 


t Personal communication from Professor Johnson. 


INTRAFORMATIONAL CORRUGATED ROCKS 603 


In view of the fact of the very moderate degree of corrugation, 
a slight amount of subaqueous slumping or forward settling, per- 
haps under earthquake impetus, would have produced the struc- 
tures. There may have been no actual gliding over the lower 
erosion surface at all. Erosion intervals, under the conditions 
of shifting currents during the deposition of the cross-bedded 
sandstones, would be expected. The corrugations of the disturbed 
zone might possibly have resulted from slight differential slipping 


Fic. 8.—Intraformational cross-bedded, corrugated, Triassic sandstone near 
Kanab, Utah. (After D. W. Johnson.) 


along the erosion surfaces but, in spite of certain outward resem- 
blances to the intercalated contorted zones above described as due 
to differential movement after deposition of the immediatley over- 
lying beds, the writer believes that the structures in the Kanab 
occurrence are essentially different, and that they are correctly 
interpreted by Johnson. 

Norton’ has discussed subaqueous gliding as a cause of a certain 
type of breccias, but examples certainly coming under this category 
are apparently not common. 


tW. H. Norton, Jour. Geol., Vol. XXV (1917), pp. 182-85. 


604 WILLIAM J. MILLER 


ACTION OF ICE 


Intraformational contorted clays have, by various writers, been 
attributed to the action of ice but, as a rule, there has been little or 
no attempt to really analyze the structures involved. Salisbury 
and Atwood,’ in their discussion of an extinct glacial lake near 
Devil’s Lake, Wisconsin, figure and very briefly describe intraforma- 
tional contorted clays. They say: “The grounding of an iceberg 
on the surface before the overlying layers were deposited, or the 
action of lake ice, may have been responsible for the singular phe- 
nomenon.” In accordance with criteria set forth in the foregoing 
discussion of the Connecticut Valley clays, the writer believes that 
this corrugated zone must have resulted from differential movement 
after deposition of the overlying beds. The remarkable uniformity 
of thickness of the contorted zone; its relatively regular (nearly 
straight) upper surface; and the gently bent immediately overlying 
beds all strongly indicate that the corrugations developed under 
weight of the overlying beds. If the corrugations were caused by 
thrusting action of ice upon surface layers would not the con- 
torted zone show notable variations in thickness and irregularity 
of its upper surface, and would not the overlying beds fail to show 
appreciable evidence of having been deformed? It is, however, 
conceivable that the corrugated zone may have resulted from 
differential movement brought about by the crowding action of 
ice against the upper portion of the whole body of clay, thus setting 
up a differential motion within its mass. Either this, or differential 
movement brought about under the action of gravity (in case the 
clays are at least moderately tilted), appears to have produced the 
corrugations. 

J. Geikie,? in his description of the early postglacial deposits of 
the basin of the Forth in Scotland, says: 

Here and there also the beds (sands and clays) are much crumpled and 
confused, great sheets of clay being rolled over and over, and involving the 
associated sands for considerable distances. ... . These are exceedingly 


irregular, and are just of such a character as we should expect would result 
from the grounding of ice rafts. 


t Salisbury and Atwood, Jour. Geol., Vol. V (1897), p. 143. 
2J. Geikie, The Great Ice Age (1894), pp. 271-72. 


INTRAFORMATIONAL CORRUGATED ROCKS 605 


Such contorted zones were quite certainly produced before the 
overlying clays were deposited, as proved by the notable variation 
in thickness of the contorted zone, its very irregular upper surface, 
and the manner in which the perfectly undisturbed overlying clays 
were laid down on the irregular surface. 

W. A. Johnston" has described and figured an interesting case 
of notably crumpled sand forming a zone of variable thickness under 
till near Fort Frances, Ontario. He says that the crumpling of the 
sand was due to the over-riding action of glacial ice. 


DIFFERENTIAL WEIGHTING 


Kindle,” in his discussion of deformation of unconsolidated beds 
on the Avon River, Nova Scotia, describes ‘‘a section of finely 
laminated horizontal silts, which, for a thickness of one foot or 
more near the middle, have been distorted into a highly convoluted 
zone.’ He advances the hypothesis of differential weighting to 
account for the phenomenon and says: 

If a heavy load of sand were deposited over a portion of an area in which 
very soft beds were interpolated between more coherent strata, the more mobile 
would be likely to squeeze outward away from the sand pressure toward an 
unsupported edge, if one were developed by stream or wave cutting. This 


might occur without disturbing firmer beds above and below through the more 
yielding character of the soft beds. 


According to Kindle, current scour would remove the heavier and 
coarser beds, after which horizontal layers would be deposited over 
the disturbed beds. He describes experiments in which clay beds in 

glass tanks were notably deformed by differential weighting with 
shot. 

Some reasons for thinking that the foregoing explanation is not 
applicable to the Avon River occurrence, and quite certainly not to 
intraformational contorted clays in general as typified by the Con- 
necticut Valley occurrences above described, are as follows: (2) 
Quite generally, in nature, there is no evidence of anything like 
notable current scour, but rather there has been continual deposition 
of the clays; (2) in the experiments the surfaces of the deformed 


tW. A. Johnston, Can. Geol. Surv., Mem. 82 (1915), p. 43 and PI. 8. 
2 E. M. Kindle, Geol. Soc. Amer. Bull., Vol. XXVIII (1917), pp. 323-32. 


606 WILLIAM J. MILLER 


zones are very irregular, while in nature they are usually very regu- 
lar, or even straight, for long distances; (3) such regular surfaces 
could hardly have resulted from vigorous current scour because the 
contorted zones are usually remarkably uniform in thickness for 
long distances; (4) there is almost invariably no evidence for the 
present or former existence of materials of such arrangement and 
character directly over the contorted zones as to give rise to very 
appreciable differential weighting; and (5) the contorted beds are 
seldom very much softer or more mobile than the inclosing strata. 


CRYSTALLIZATION AND HYDRATION 


In certain types of rocks, like gypsum and salt, there is strong 
evidence for the development of intraformational deformative 
effects by crystallization (or hydration) after their deposition. In 
the Zechstein salt of Germany, 
where the enclosing rocks are undisturbed, the layers of brightly colored bittern 
salts and of gypsum often show a remarkable flexuous, sinuous, or disrupted 
Character warmer In the Salina deposit of central New York, some of the 
alternating salt and gypsum layers occasionally show a pronounced flexing 
and overfolding, while others are wholly undisturbed. 

In his discussion of the salt beds of western central New York, 
Luther? has reproduced an interesting picture of a small sharply 
overturned fold of rock salt between practically undisturbed beds. 

The above-described examples occur in regions of non-folded and 
non-faulted strata, and it seems quite certain that “the main 
force was the endogenetic one due to the crystallizing force of the 
salts and to metasomatic process” (Grabau after Arrhenius). 

Very fine examples of corrugations and crenulations occur within 
the gypsum deposits at Hillsborough, New Brunswick. According 
to Ami3 “‘the gypsiferous deposits present evenly banded structure, 
between which there occur neatly folded layers in the form of 
ribbon-like corrugations” (Fig. 22). Kramm* states that ‘‘the 
gypsum rests upon a bottom of anhydrite and reaches a maximum 
thickness of perhaps 125 feet,’’ and that gypsum was derived by 

tA. W. Grabau, Principles of Stratigraphy (1913), p- 757+ 

2D. D. Luther, V.Y. State Mus. Rept. 50, Part 2 (1896), Pl. 4. 


3H. M. Ami, Geol. Soc. Amer. Bull., Vol. XXV (1914), Pp. 37- 
4H. E. Kramm, Can. Geol. Surv., Guide Book No. 1, Part 2 (1913), p. 364. 


INTRAFORMATIONAL CORRUGATED ROCKS 607 


hydration of anhydrite, as can be observed in many places. Since 
the gypsum beds occur in a very shallow syncline, the highly 
crenulated intercalated layers cannot have been caused by regional 
folding. The cause was, no doubt, the force exerted within the 
whole mass as a result of the great expansion during the trans- 
formation of anhydrite to gypsum. Differential stresses and 
strains set up locally in the whole mass caused localization of the 
corrugations (Fig. 9). 


Fic. 9.—A specimen of gypsum from Hillsborough, New Brunswick, showing 
highly folded layers between less folded layers. 


PRESSURE OF INTRUDING MAGMAS 


Among the pre-Cambrian rocks of the Adirondack Mountains, 
New York, the writer has observed excellent examples of intra- 
formational corrugations which are believed to have been direct 
effects of the pressure of magmatic intrusions. He has presented 
evidence’ to support the view that the very ancient Grenville 
stratified series has never been subjected to severe regional folding 
throughout most or all of the Adirondacks. During the intrusion 
of the tremendous volumes of syenite-granite magma, the Grenville 
series was, however, badly cut to pieces so that many masses, 
both small and great, were tilted about in the rising magma and in 
many places subjected to notable differential pressure. The lime- 
stones, especially where they are near contacts with the syenite- 


=W. J. Miller, Jour. Geol., Vol. XXIV (1916), pp. 595-06. 


608 WILLIAM J. MILLER 


granite, much more commonly exhibit local corrugations or crump- 
lings than the other more rigid strata. Where a mass of the Gren- 
ville strata contains zones of well-bedded limestone intercalated 
between more rigid strata, and the whole has been subjected to 
differential pressure by the invading magma, the limestone layers 
have not uncommonly become contorted by differential movement 
within the mass while the adjacent beds above and below have been 
deformed little or not at all. In Figure 10, which well illustrates 
such a phenomenon in northern New York, the contorted beds of 


Fic. 1o.—Crumpled beds of impure, thin-bedded, crystalline limestone between 
. beds of only slightly disturbed garnetiferous gneiss north of Hermon, St. Lawrence 
County, New York. 


very plastic impure limestone lie between heavy beds of rigid 
garnetiferous gneiss. This contorted limestone and its associated 
gneiss form part of a long narrow body of Grenville strata which 
was included in, and subjected to differential pressure by, a large 
body of granite magma, causing the more plastic limestone beds to 
crumple. 

Spurr, in his discussion of the Silver Peak region of Nevada, 
publishes a fine picture of intraformational contorted strata which 
occur on Mineral Ridge. In this region large volumes of granite 
magma invaded and cut to pieces Paleozoic strata made up of 


tJ. E. Spurr, U.S. Geol. Surv. Prof. Paper 55 (1906), p. 108 and Pl. 2r. 


INTRAFORMATIONAL CORRUGATED ROCKS 609 


a comparatively thick series of thin-bedded, shaly, and calcareous sediments 
intercalated with beds of pure limestone, now metamorphosed into marble. 
The thin-bedded sediments are sometimes carbonaceous, sometimes streaked 
with sandy material..... The limestone-slates are contorted on a minor 
scale. Most of this crumpling was probably due to the intrusion. 


Beyond the last sentence quoted, Spurr says nothing regarding the 
cause of the intraformational crumpling. The available data 
rather clearly indicate that these corrugations, like those above 
described as occurring in northern New York, have been caused by 
the magmatic intrusion where a block of strata was more or less 
completely surrounded by the magma and subjected to differential 
movement, causing the more yielding layers to crumple. 

Between 5 and 6 miles north of Northampton, Massachusetts, 
the writer has observed excellent examples of local contortions 
within the Leyden argillite (Paleozoic) formation near its contact 
with a basic phase of the Williamsburg granite. For about 2 miles 
parallel to the contact, irregularly distributed contortions are highly 
developed in the argillite for ro to 20 rods out from the contact. 
Beyond that they rapidly diminish to disappearance. From the 
field evidence it seems clear that the corrugations resulted from 
differential movements within the argillite, caused by the shoulder- 
ing pressure of the rising magma. 


ACTION OF MAGMATIC INJECTION 


In certain regions magmatic injection schists and gneisses con- 
tain local portions which are highly contorted. Many observations 
of such phenomena have been made by the writer in his study of 
the Adirondack Precambrian rocks. The following brief description 
of a certain district may best serve to illustrate the main principles 
involved. Extending 2 miles northeastward from just north of the 
village of Russell, St. Lawrence County, New York, there is a wide 
belt of mixed rocks containing fine exposures of amphibolite and 
garnetiferous gneisses intimately cut and injected, mostly parallel 
to the foliation, by moderately coarse-grained granite and pegma- 
titic granite, the whole mass being conspicuously banded. Just 
north of the village most of the mixed rock is notably contorted, 
while one-fourth to one-half of a mile farther east most of the rock 
is relatively straight-banded, but contains local highly contorted 


610 WILLIAM J. MILLER 


zones. Still farther east the mixed rocks are nearly all straight- 
banded. It is believed that, during the forcing in of the magma, 
the whole mass of rock was notably plastic, and that local differential 
movements within the mass, caused by unequal pressures of the 
rising magma, resulted in the local corrugations. 


MAGMATIC FLOWAGE 


Finally, in our discussion of intraformational contorted rocks, 
differential magmatic flowage should be mentioned as a cause. 
Within certain areas of plutonic igneous rocks which exhibit primary 
foliation, there are not uncommonly local zones or bands in which 
the gneissoid structure, accentuated by dark minerals, appears to 
be irregular, wavy, or even contorted. The writer’s experience in 
the Adirondacks shows’ such local, contorted, primary flow-struc- 
tures to be very common there, especially in the great syenite- 
granite series, and it is believed that they are essentially the result 
of varying magmatic currents under differential pressure, principally 
during a late stage of magma consolidation. 

Lawson? has noted similar structures within certain granites of 
the Rainy Lake region of Ontario. He says: ‘‘The lines of streak- 
ing are very often not straight, but are wavy or contorted, sometimes 
intricately so, and are evidently due to slow movements in the 
magma prior to final consolidation.” 


tW. J. Miller, Jour. Geol., Vol. XXIV (1916), pp. 611-12. 
2A. C. Lawson, Can. Geol. Surv., Mem. 40 (1913), p- 93- 


THE PHYSICAL CHEMISTRY OF THE CRYSTALLIZATION 
AND MAGMATIC DIFFERENTIATION 
OF IGNEOUS ROCKS 


J. H. L. VOGT 
Trondhjem, Norway 


(Continued from page 649 of Vol. XXIX) 


V 
THE INFLUENCE OF PRESSURE 
A 


The dependence of the melting point on uniform! pressure is 
designated, as is well known, by the formula established by Clausius- 
Chapeyron: 

ASTD T (Viiquid— Veolia) 
SD Diag e men Glan 


AM 2. : ? ; Bait ‘ 
AP designating the alteration of the melting point in C° per increase 


of atmospheric pressure, T the melting point in absolute tempera- 
ture (starting from — 273°), q the latent melting heat in gr. cal. pr. 
gr., v the specific volume (in cm3 per gr.) at the melting point— 
Viiq. — Vso. accordingly meaning the difference of specific volume 
between the liquid and the solid phase at the melting point—and 
H=425. 

In by far the most substances Vjig. —Vso1. 1S positive at common 
and at moderate pressures. Accordingly the minerals, when 
melting, must, as a general rule, expand in volume or become of 
lower specific gravity. The case is entirely the reverse of that of 
ice and of bismuth. : 

t Within a liquid, such as a magma, also in the case that more or less mineral has 
already crystallized, the pressure always becomes uniform or hydrostatic, and in this 
case the formula quoted is applicable. Quite different is the case of non-uniform com- 


pression (and stress) in the solid phase. I beg to refer to a series of publications of 
recent years (see the review of John Johnston in this Journal, XXIII [1915], p. 732). 


611 


612 Jn 2. VOGE 


G. Tammann, as is well known, has pointed out that for many 
salts, Viig. — Vso. decreases with increasing pressure so that, when the 
pressure is extraordinarily high, for instance 5000—-10,000 atmos- 
pheres, it approaches zero; at still higher pressure the difference 
may be negative. However, this is not the case of substances in 
general. Thus P. W. Bridgman™ has experimentally ascertained 
that, in a great many substances (metals and some salts, etc.), 
on pressures rising to 12,000 atmospheres (corresponding, for a 
magma, to a depth of about 40 kilometers), Vig. — Vsol. Femains posi- 
tive—indeed with gradually (somewhat) decreasing magnitude 
for increasing pressure. Some substances, even at the enormous 
pressures just mentioned, show but an inconsiderable decrease of 
the difference of volume, and J. Johnston and L. H. Adams? showed, 
for a number of metals, under a pressure of up to 2000 atmospheres 
(corresponding, for a magma, to a depth of about 7.5 kilometers), 
a rectilinear course of the difference of volume. 

Under a pressure of one atmosphere all the silicate minerals 
hitherto examined show, at room-temperature, a lower specific 
gravity for the glass, that is the extreme viscous fluid phase, than 
for the crystalline, that is the solid phase. 

According to what we have stated above, we must take it for 
granted that for rock-forming minerals Vjiq. — Vso. is in all cases 
positive, and that the difference of volume existing for a pressure 
of one atmosphere will remain nearly constant, at any rate down to 
depths of a few kilometers, and that even to depths of 5 or Io 
kilometers it will show practically no or only a little decrease. 

As far as the common rock-forming silicate minerals are con- 
cerned, the latent melting heat is very high throughout. In fact 
for minerals melting at about 1200° to 1400° or 1500°, it amounts 
to about go or 100 gr. cal. pr. gr.—tfor some a little more and for 
some a little less. 

= Proc. Amer. Acad., XLVII (1911-12) and Physic. Rev., III (1914). 

2 Zeits. f. anorg. Chemie, 72 (1911), and Amer. Jour. of Sci., XXXI (r1o11). 


3As for anorthite (melting point 1550°), diopside (1391°), akermanite (about 
1310°) and fayalite (about 1075°),I have in earlier publications (Silikatschmelzlés. 
II [1904], with correcting calculation in my publication on slags in Doelters, Handb. d. 
Mineralchemie, I [1912], p. 942) determined the latent melting heat at respectively 
about 105,94, 90, 80 cal., with an error limit of +15 or 20 cal. For anorthite Bowen 


MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 613 


Thus the divisor of the formula becomes very great, and con- 
sequently the melting point of the silicate minerals rises but very 
little on increasing pressure.’ As an illustration I cite a case 
treated in my publication just quoted: 

For a mineral with melting point = 1200°, melting heat = 100 cal., 
density in the solid phase=3.000 and in the fluid phase=2.887, 
Viig. — Vsol., at the melting point accordingly amounting to 0.013 or 
3.9 per cent we calculate the melting point at higher pressures: 


Depth below the 


Pressure Melting Point 


Surface 
i QUOC capoeooseeoescenea o kilom. 1200° 
Dr ORALIMOSPHCLESHM ere: oe ele lon © 1201-35 
27Oo atmOspheres...........-.-- to kilom. RAIS 4 
10,000 atmospheres............. 37 kilom. T250° 


We have assumed, in this instance, in accordance with an early 
‘examination carried out by C. Barus,? a percentage value of 3.9 
per cent, for the difference (at the pressure of one atmosphere) 
between the specific volume in the fluid and in the solid phase of 
the melting point, and we assume also the same value at higher pres- 
sure. 

A series of determinations of glass and of crystalline substance 
at room-temperature (15° or 20°) show, for rock-forming minerals 
and for rocks, the following percentage differences of specific 
volume 33 

In some cases only about 3 per cent (2.9-3.5 per cent), in most 
cases between 5 and 8 per cent, exceptionally up to 10.6, 11.4, and 
in a single case 13.6 per cent. 


(in his study on plagioclase, Joc. cit.) has calculated 104.2 cal. (I found 105 cal. 20 per 
cent.) For diopside W. P. White (Amer. Jour. Sci., XXVIII [1909], p. 486, footnote), 
according to a preliminary determination, states 106+ 15 cal. (I found 9415 percent.) 

tT beg to refer to my statement in Tscherm. Mitt., XX VII (1908). 

2 Phil. Mag. London, XXXV (1893), and U.S. Geol. Surv. Bull. 103. 

3 Most of these statements are grouped from Doelters, Handb. d. Mineralchemie, I 
(1912), p. 672. 

4 The last named value concerns CaMgSiO.O¢ with density of the mineral diopside 
=3.275, and of the glass=2.830 (Allen, White, etc., Amer. Jour. of Sci., XXVII 
[1909], Vlig.—Vsol. accordingly = 0.3533 —0.3053 =0.050. 


614 TMD VOCE 


These figures, however, cannot be transferred so as to be applied at 
the melting point without being corrected, as the crystalline and the 
glassy phase will,as a general rule, differ a little as to volume dilation. 

The only precision investigation known to me of the specific 
volume of a silicate at the melting point, has been carried out by 
A. L. Day, R. B. Sosman and J. C. Hostetter, who determined for 
a diabase: 

At 20° in the crystalline phase dens. = 2.975 (specific volume = 
0.3362) and in the glassy phase dens.=2.763 (specific volume= 
0.3620), the percentage difference at 20° accordingly, amounting 
to 7.1 per cent, the glass being taken as the starting point. 

This percentage difference increased at the melting point to 
9.1 as a minimum and Io.9 as a maximum, and next to the last 
figure, accordingly to about 1o.5 per cent corresponding to a differ- 
ence=ca. 0.041 of the specific volume. 

From what we have stated above, we take it for granted that, 
at the melting point of the common rock-forming minerals (melting 
at about 1200-1500), Viig. — Vso. Will, as a general rule, vary between 
the limits of about 0.015 and 0.045. Assuming a melting heat of 
too cal. and a melting point of respectively 1200° and 1500°, we 
shall have, for a pressure of 1000 atmospheres, corresponding to a 
depth of about 3.7 kilometers, a rise of temperature of respectively 
about 5-6° and about 15-18°. ‘That is to say, in igneous flows, at 
depths of up to o.5 or 1 kilometer, the melting point of the different 
minerals only rises between about 1° and about 4 or 5°, while 
in deep-seated rocks, in which the crystallization takes place at 
depths of 5, 10, or 15 kilometers, there may be involved a rise 
varying respectively between about 6° and 30°, between about 12° 
and’40°, and between about 15 or 20° and 60”. 

Thus, even in deep-seated rocks crystallizing at very great 
depths, the rise of the melting point is rather inconsiderable. 
And it should be particularly emphasized that the difference of 
rise of the melting points of the various minerals crystallizing in 
one and the same magma, will only amount to some few degrees; 
even in deep-seated rocks the difference is rather small, in fact 
rarely amounting to more than 10, 20, or perhaps in some cases 40°. 


t Amer. Jour. of Sci., XXXVII (1914). 


MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 615 


As to the dislocation of the eutectic by uniform pressure, H. W. 
Bakhuis Rooreboom' has stated that in this case also the formula 
of Clausius-Chapeyron is applicable. The rise of temperature of the 
binary (and the ternary) eutectic of the rock-forming minerals 
by pressure may consequently be measured by the same small 
measures that have just been indicated for the minerals themselves. 

Moreover, it is to be noted that the tangent to the melting curve 
near the origo (respectively 100 per cent a and 100 per cent 0) is 
determined by the factors: T 
(absolute temperature), q (melt- 
ing heat), and the electrolytic 
dissociation which three factors 
are but little dislocated by pres- 
sure, and finally the molecular 
weight which is constant if no 
polymerisation takes place. In 
other words, the melting curve 
in a binary system (Fig. 50) near 
the origo will, at high pressure, 
take a course parallel to that of 
the melting curve at low pressure. 


a 6 


Fic. 50.—Illustrating the unessential 


The further course of the melting 
curve is also essentially deter- 
mined by the constants just 
mentioned, which are generally 


dislocation of the composition of the 
binary eutectic by uniform pressure. 

ie oon =melting point and E,,, E,,= 
binary eutectic by respectively low and 


: E high pressure. 
very little dislocated by pressure. 


This relates to the general experience that pressure has but little 
influence on the solubility of the mutual solutions, provided that 
no vapor phase is present. 

The contents of a and 0 at the intersecting point of the curves, 
in other words the composition of the binary eutectic, (Ep, at low 
pressure, and E,, at high pressure) will accordingly be but very 
little dislocated by pressure. And this accords with the fact that 
the temperature of E,,, as just stated, is only very little higher than 
that of E,,. This reasoning can also be extended so as to be applied 
to the ternary and still more complex eutectics. 


t Heterogene Gleichgewichte, II (1904), p. 715. 


616 JH. E. VOGL 


The error made by transferring the determination of the compo- 
sition of the eutectic on a pressure of one atmosphere, to be applied also 
to the pressure prevailing during the crystallization of the eruptive 
rocks, will accordingly be rather unimportant provided that the minerals 
involved are formed independently of pressure. 

I have demonstrated this fact by some petrographic instances 
in my publication in Tscherm. Min. Petrogr. Mitt., XXV (1906), 
and XXVII (1908). I will here only point out that the analyses 
of the final eutectic quartz: feldspar-product from igneous flows, 
formed at a comparatively low 
pressure, accord within the 
limits of error exactly with the 
corresponding final product from 
granitic deep-seated rocks and 
also with the graphic granite in 
the Archean granite-pegmatite 
dikes formed at very great depths 
(see the analysis No. 1-40). And 
I deem it right to employ the 
experimentally determined eu- 
tectic Qu:An, at a pressure of 

i - one atmosphere, for parallellism 
st Se fusing ie amesntsl with the eutectic Qu: Abin deep 
tween the liquidus and the solidus curve of seated rocks or granite-pegmatite 
the An: Ab-system by low and high pres- dikes (Fig. Be 
ae _ Thesame general reasoning as 
to the inconsiderable influence of 
pressure on the dislocation of the 
melting point and on the eutectic may be transferred, in all essen- 
tials also to mix-crystal systems, or at any rate, to the continuous 
mix-crystal systems of type I. 

For example, let us consider Ab:An. Even if the melting point 
of An should, at high-pressure, rise a little more or a little less than 
of Ab, it must be presumed a priori that the course of the liquidus 
and the solidus curves, and, what is petrographically of the greatest 
importance, the horizontal distance between the liquidus and the solidus 
curves, will remain almost the same (Fig. 51) at a high pressure 


1400 


1500 


1004m 80An 6047 40An 204n adn 
0.Ab 2046 40Ab ——60Ab 8046 10046 


ee Tpn = Smelting point by respec- 


tively low and high pressure. 


MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 617 


(pn) as at a low one (p;). This is also verified by petrographic 
experience. Thus in effusives and dike-rocks with a proportion 
Ab:An, fixed from the analysis of the whole rock, the mix-crystal 
first separated had nearly exactly the same composition as in a 
melt of Ab+-An crystallizing at a pressure of one atmosphere.’ 


B 


The formula of Clausius-Chapeyron quoted (p. 611) is to be 
applied also to the dislocation, at uniform (hydrostatic) pressure, 
of the znversion point between two reversible solid phases of one 
and the same substance. 

The volume difference v,— vz, in this case, involves the difference 
between the solid phases stable at higher and lower temperature, 
and the melting heat must be replaced by the inversion heat. 

Also here v;— vz is in most cases positive, and only exceptionally 
negative.” 

The inversion heat between two solid phases of one and the same 
substance is, so far as we now know, always positive and generally 
very small, sometimes low even almost to zero, and in most cases it 
amounts only to a small fraction of the melting heat.’ 

For metals, sulphides, silicates, etc., the divisor of the tempera- 
ture: pressure formula thus becomes, as a general rule, very small, 


tI refer, on this subject, to a comparison between my account in Tscherm. 
Mitt., (t905), of the mix-crystal system Ab:An in igneous rocksand the diagram of 
Bowen (cf. Fig. 2, dating from 1912-13) for the Ab:An melt at the pressure of one 
atmosphere. The analyses of the plagioclase first separated in the rocks concern partly 
-plagioclase somewhat zonally constructed, and partly plagioclase that has grown a little 
poorer in An and richer in Ab on account of partial equilibrium of the solid and the 
liquid phase. The plagioclase mix-crystal first separated in andesites, dacites, etc., 
will, accordingly, have been somewhat richer in An and poorer in Ab than is indicated in 
the analysis tables on pages 503 and 512 in my publication in Tscherm. Mitt., XXV (1905). 

2 As an instance of negative v:—vz is mentioned: The density of the three modi- 
fications of Ca,SiO, amounts to: of a (at a high temperature) =3.27, of 8=3.28, and 
of y (at a low temperature) =2.97. The density of common stannum is =7.3, that of 
the gray tin formed at a low temperature (undercooled much below +20°) is only =5.8. 
Both in the passage from 6 to y-Ca2SiO,, and from the common white stannum to the 
gray one, so great an increase of volume takes place that the substance is disintegrated 
by itself. 

3 However, to this rule also there are some exceptions. The most striking instance 
is formed by lithiumsulphate having an inversion heat even five times greater than the 
melting heat (Hiittner und Tammann, Zeit. f. anorg., Ch. 43 [1903]). 


618 J. ok. VOGH 


and consequently the inversion temperature, even if v,—v, only 
amounts to a medium positive value, will have a not quite incon- 
siderable rise on increasing pressure. And if v,—v. is considerable 
(and positive) the rise of inversion temperature owing to pressure 
will be exceedingly great.* 

This has been experimentally examined for monoclinic and 
orthorhombic sulphur (monocline S, on pressure of one atmosphere 
having its melting point at 119.25°, density =1.98, melting heat, = 
12.5 cal. r inversion temperature between monoclinic and ortho- 
rhombic S, on pressure of one atmosphere = 95.6°, inversion heat = 
2.52 cal., density of orthorhombic S = 2.07 and melting point = 112.8). 

On pressure the inversion temperature between monoclinic and 
orthorhombic sulphur rises more than the melting curve of mono- 
clinic sulphur. The two curves intersect at a pressure of 1320 
kilogrammes pr. cm? (=1275 atmospheres) and at a temperature 
of 151°, and at still higher pressure only orthorhombic sulphur may 

be formed.” 
Accordingly it will not be surprising that, in the case of a rock- 
forming mineral which at low pressure has an a-form (at high 
temperature) as well as a 6-form (at lower temperature), it should 
always be the 6-form,3 or the form stable at lower temperature, that 
crystallizes in eruptive rocks formed at high pressure. 

Let us consider particularly some of the most important modi- 
fications of S20,: 

The inversion from a-cristobalite into a-tridymite at 1470+ 10°.4 

The inversion from a-tridymite into a-quartz at 870+ 10°).4 

t Therefore, the doctrine, advanced by me in 1908, of the inconsiderable rise of the 
melting point of silicate minerals cannot also, as a general rule, be transferred to the 
inversion point. I refer, on this subject, to the instructive remarks, made by V. M. 


Goldschmidt, Die Kontaktmetamorphose im Kristianiagebiet (1911), p. 112, and by 
C. N. Fenner, Jour. of the Wash. Acad. of Sct. (1912), 2. 

2See G. Tammann Krystallisieren und Schmelzen (1903), and several publications 
by Roozeboom reviewed in Doelter’s Phys.-chem. Mineralogie (1905), p. 31. 

3 According to the terminology used by Boeke and by many other mineralogists, 
I employ a to designate the modification stable at a higher temperature, 8 and y to 
designate the modifications stable at lower temperatures. To avoid misconception, 
it should be noted that some investigators, among them Wright and Larsen, have 
employed quite the reverse terminology (a designating lower, and 8 higher temperature). 


4C. N. Fenner, Amer. Jour. of Sci., XXXVI (1913). 


MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 619 


The inversion from a-quartz into 6-quartz at 575+ 2°. 

According to the statement made by A. L. Day, R. B. Sosman, 
and J. C. Hostetter’ the specific volume of 8-quartz is at 20° 0.3775 
(=density 2.649), but rises by heating, and at 561°, or just below 
the inversion point, it has risen to 0.3922. At the inversion point 
there is a sudden rise, which, however, is not precisely stated, and 
then we pass on to a-quartz at 585° with a specific volume of 0.3972. 

According to the statements of F. E. Wright and Esper L. 
Larsen the inversion heat, when $-quartz is transformed into 
a-quartz, amounts to 4.31 cal. If we set the volume difference at 
the point of inversion at 0.003 and the inversion heat at 4.3 cal., 
there is a rise of 0.015° per atmosphere, which makes at a pressure 
of 1000 atmospheres, corresponding to a depth of about 3.7 kilo- 
meters, a rise of about 15°. This affords a measure for the order 
or quantity which has to be taken into account. Accordingly the 
difference between a-quartz and 6-quartz, if depths of more than 
5 kilometers are not involved, can, at any rate with only a little 
modification, be employed as a geological thermometer. 

As to the relation of inversion point and pressure between 
a-quartz and tridymite the case is quite different. At room tem- 
perature tridymite has a specific volume of 0.4329 (density = 2.31, 
medium of 2.282 and 2.326) and 6-quartz a specific volume of 0.3775 
(density =2.649); v:—v2 accordingly amounts to 0.0554, which is a 
very high value. By heating, 8-quartz, as has just been mentioned, 
expands its volume considerably and then changes into a-quartz, 
which, according to the statements of Day, etc., has at 850° a specific 
volume of 0.3957. According to the law of the expansion of bodies 
by heating, tridymite also must expand its volume at higher temper- 
ature, but how much is not stated. If we estimate its specific 
volume at 875° at 0.445, the volume difference v:—v. at 875° will 
amount to 0.05 (or perhaps a little more). 

The inversion heat is not stated, but must be supposed to be 
tolerably low. A rough estimate, rating the inversion heat at 10 
and 5 cal., respectively, gives a rise of inversion temperature per 


tF, E. Wright and E. S. Larsen, ‘‘Quartz as a Geologic Thermometer,” Amer. 
Jour. of Sci., XXVILI (1909). 
2 Amer. Jour. of Sci., XX XVII (1914). 


620 Tip) Eh, IGS OEIE 


atmospheric pressure of respectively 0.115 and o0.23°._ A correspond- 
ing rough calculation has been made by Fenner,’ who presumed 
Vi—V2=0.057 (as at room temperature) and inversion heat=15 cal., 
which makes a rise of temperature per atmosphere of 0.105°. 
Starting from these figures (15, 10, 5 cal.) we should have, on a 
pressure of 270 atmospheres (or at the depth of 1 kilometer), a rise 
of the inversion point of respectively 28, 38 and 76°, or from 875° 
to respectively 908, 913 and 951°—and it may be that even the last 
number is too low an estimate. On the pressure prevailing at the 
depth of a few kilometers, the inversion point quite certainly will 
most likely prove to be considerably above 1200°. 

The binary eutectics Qu:Or and Qu:Ab, at a pressure of one 
atmosphere, I estimate at about 1075°—1100°, repectively 975°—1000°. 
The ternary eutectic Qu:Or:Ab must lie somewhat lower, prob- 
ably at about 950° or 925°. 

If we choose for instance a common granite, quartz porphyry, 
or rhyolite, the final crystallization must here take place as a 
complex eutectic Qu:Or:Ab+An, with a little ferric oxide and 
Mg, Fe-silicate, at a pressure of one atmosphere at a temperature 
of presumably about goo°-950°. And even if there occurs some 
surplus quartz, so that the crystallization of this mineral begins 
at a comparatively early stage, still the first crystallization of SiO, 
will scarcely take place at a higher temperature than 950° or 1000°. 
These values of temperature reckoned on the pressure of one atmos- 
phere must certainly be increased somewhat, but only a little, for 
crystallization at great depth and under high pressure, but, as we 
shall mention later, must be decreased by the content of water, 
etc., especially in the granitic deep-seated magma. 

ie on the other hand, the temperature of the inversion point 
between a-quartz and tridymite rises so considerably, SiO, will 
never crystallize in deep-seated rocks as tridymite, but always as 
quartz, and this quartz in granite, quartz-porphyry, and graphic 
granite is, as pointed out by Wright and Larsen, always an a-quartz. 

Tridymite as a primary” formation in igneous rocks is, as is 
well known, limited to certain rhyolites, trachytes, andesites, etc., 

Jour. of the Wash. Acad. of Sci., 2 (1912), p. 479 and Doren Jour. of Sci., XXXVI 
(t913), P- 347- 

2 The hydrothermal formation of tridymite is here left out of consideration. 


MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 621 


consequently to effusive rocks and to some dike rocks crystallized 
at a small depth. The crystallizing temperature will here have 
been above, probably indeed only a little above, the inversion point 
between a-quartz and tridymite in force at the pressure in question, 
and accordingly at depths of some hundred m., viz., at about goo° 
or perhaps a little higher. 

By stating the depths at which SiO,, in the effusives and dike 
rocks here mentioned, occurs as quartz or as tridymite, our knowl- 
edge of the influence of pressure on the rise of the inversion point 
between a-quartz and tridymite will be increased. Fenner (Amer. 
Jour. of Sci., XXXVI, p. 348) describes a volcanic rock in which 
quartz phenocrysts were first formed and were afterward trans- 
formed into tridymite. Fenner suggests the explanation that the 
quartz phenocrysts were formed at higher pressure a little below 
the inversion point in force at the pressure in question, and that the 
magma with its phenocrysts was afterwards brought nearer the 
surface, viz., to a diminished pressure, lying a little above the 
inversion point at the given temperature. The phenocrysts of 
a-quartz first formed at the greater depth must consequently now 
be transformed into tridymite. Fenner, however, expresses some 
doubt as to this explanation, which, after all, in my opinion, is the 
natural one. 

According to what has been stated by Wright and Larsen, the 
quartz in the graphic-granite from granite-pegmatites was originally 
formed as a-quartz, and accordingly at a relatively high tempera- 
_ture, viz., between the inversion point between 6- to a-quartz and 
a-quartz to tridymite. But, besides, there sometimes occurs, in the 
granite-pegmatites, quartz that “‘in all probability has never been 
heated above the inversion temperature between a- and 6-quartz.” 
“These large masses of quartz were in certain cases definitely stated 
by the field relations to be the last portions of the pegmatite to ' 
crystallize out.’’ This 6-quartz which is formed below the inversion 
point between a- and 6-quartz, and the formation of which, at the 
enormous pressure, must have taken place at a temperature below 
600 or 625°, must be supposed to have been separated from a par- 
ticular solution of SiO,+-H,O (see following chapter on the influ- 
ence of the light volatile compounds). 


622 Te VENV OGE 


According to the investigations made at the Geophysical Labora- 
tory in Washington, the inversion point between pure a=CaSiO, 
(pseudowollastonite) and B=CaSiO, (wollastonite) has been stated 
at about 1200°. Some amount of MgSiO, entering isomorphously 
into the silicate makes the inversion point rise; thus about 6 per 
cent MgsiO, brings about a rise up to 1300° and 17 per cent 
CaMgSi,O¢ (=ca. 8 per cent MgSiO,) even up to 1345+ 10° C.t 
The inversion point of the mineral wollastonite occurring in the 
contact zones, containing as as general rule a little MgSiO,, must, 
accordingly, be estimated, at the pressure of one atmosphere, at 
about 1250-1300°, and at a high pressure even a little higher 
temperature must be presumed, though, on account of the small 
difference of density between the two forms, the rise will probably 
be quite inconsiderable.? 

As will be brought out in a later paper, the eruption of the 
magmas, as a general rule, took place at a temperature which was 
for the deep-seated rocks almost exactly identical with the tempera- 
ture of the beginning crystallization. For the porphyry rocks we 
may in many cases assume a temperature of the eruption even 
somewhat below that of the beginning crystallization. 

If we leave out of consideration the peridotites, anorthosites, 
and analogous anchi-monomineral deep-seated rocks, the eruption 
temperature, accordingly, must only exceptionally have been so 
high as about 1300°. For many anchi-eutectic rocks, I estimate it 
at about 1250°, and for granitic rocks, even much lower still, as 
1o00° or somewhat less. Thus, it is easily explained that in the 
contact zones there always occurs wollastonite and never poets 
wollastonite. 


MgS10;.—According to the investigations made by O. Andersen 
and N. L. Bowen MgSiO, has, at usual pressure, no true melting 
point, as, at 1557°, MgSiO, is divided into solid Mg,SiO, (forsterite) 
and liquid. ‘The weight proportion of the two phases is 5.5 per cent 


«T. B. Ferguson and H. E. Merwin, Amer. Jour. of Sci., XLVIII (1919), p. 165, 
and the earlier investigations here cited. 


2 White and Larsen, Amer. Jour. of Sci., XXVII (1909), p. 421; cf. V. M. Gold- 
schmidt, ““Die Kontaktmetamorphose im Kristianiagebiet” (1911), p. 110. 


3 Amer. Jour. of Sci., XXXVII (1914), and Zeits. f. anorg., Ch. 87 (1914). 


MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 623 


forsterite to 94.5 per cent fluid. With continued heating the 
forsterite dissolves in the course of time, so that all is transformed 
into fluid at 1577°. 

On cooling a melt of MgSiO, there first crystallizes (at 1577°- 
1557°) a little forsterite, which on continued (and slow) cooling is 
resorbed under new formation of solid MgSiO, (at high tempera- 
ture clinoenstatite). 

The inversion point, at a pressure of one atmosphere, between 
clinoenstatite and enstatite amounts to about 1100°. The point 
cannot be settled exactly. The above reference to the treatise of 
Andersen and Bowen is for pure MgsSiO, (at one atmospheric 
pressure). 

For (Mg, Fe) SiO,, (Mg, ¥*) SiO, and FeSiO, the following must 
be taken into consideration: 

Mg,SiO, has melting point 1890°. 

Fe,SiO, has melting point about rroo°. 

For the intermediate mixtures there are intermediate melting 
point intervals (Fig. 22). 

Fe,SiO, and Mg,SiO, with relation about in the middle between 
Mg and Fe, has a considerably lower melting point interval than 
I550°, and consequently cannot be separated in the solid face at 
the first mentioned temperature from a (Mg, Fe) SiO, melt. This 
makes it probable that the splitting by one atmosphere’s pressure of 

solid pure MgSiO, into olivine-mineral and fluid may be transferred 
to (Mg, ¥*) SiO, with only a little Fe, but not to metasilicate with 
predominant Fe or with a middle relation between Mg and Fe. 

_ Clinoenstatite (respectively clinobronzite), as well known, 
sometimes occurs in meteorites, but has never been determined 
certainly in terrestrial igneous rocks. If the metasilicate in the 
rocks originally had existed in the clino-modification (clino-enstatite, 
-bronzite, -hypersthene) we might, according to the experiences from 
the meteorites, suppose that the original mineral, in any case, 
occasionally would still have been in existence. But while this, 
as far as we know up to date, never is the case, the explanation 
seems to be that the crystallization of the Mg, Fe-metasilicate 
in the common igneous rocks—of composition between 0.08 FeSiO,, 
0.92 MgSiO, and about 0.4 FeSiO,, 0.6 MgSiO;—always took place 


624 SE VOGT 


for the existing pressure at a temperature lower than the inversion 
point between the clino- and the ortho-modification of the meta- 
silicate. This view is strengthened by the fact that in the rocks 
where the metasilicate has crystallized at an early stage it shows 
the crystallographic contour of the orthopyroxene. 

It is probable that the inversion temperature between the clino- 
and the ortho-modification rises not quite inconsiderably with the 
pressure. The difference in density between clinoenstatite and 
enstatite is indeed very little, but if the inversion heat is minimal, 
as only a small fraction of 1 cal., the inversion temperature will 
nevertheless increase not inconsiderably with the pressure. In 
the common anchi-eutectic norites, etc., the crystallization interval 
for the most part lies at about 1275—1200°, which must be lower 
than the inversion boundary at high pressure between the two Mg, 
Fe-metasilicate stages. In bronzite rocks and bronzite-carrying 
olivine rocks a not inconsiderably higher crystallization interval may 
be presumed, but even in these rocks we cannot observe any struc- 
tural indication that here originally occurred a clino-modification. 

By melting MgSiO, at 1557° to a liquid in connection with solid 
olivine—the latter only in a small quantity—and at 1577° only toa 
liquid, v; — v2 will, in accordance with what happens in all previously 
examined silicates, doubtless be positive. That is to say, the 
“incongruent”’ melting point of MgSiO,, in the same way as of 
other silicates, must be supposed to rise somewhat with the pres- 
sure. The splitting of MgSiO, to an olivine (very poor in FeO) in 
the igneous rocks might thus have taken place at a temperature 
somewhat above 1557°. But the crystallization of the hypersthene- 
bearing gabbros, syenites, granites, etc., occurred, as just mentioned, 
at a considerably lower temperature. The conclusion of this, 
according to my view, is that the proved crystallization of forsterite 
in a melt of MgSiO, at a pressure of one atmosphere cannot be 
transferred to take place at the crystallization of the bronzite- or 
hypersthene-bearing deep-seated rocks. I cannot agree with more 
of the opinions maintained by Andersen and Bowen (loc. cit.) 
and Bowen‘ since these investigators have not taken into considera- 
tion the influence of pressure. 


t The crystallization of haplobasaltic and haplodioritic magmas (1915), and The 
later stages of the evolution of the igneous rocks; (1915). 


MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 625 


As an argument for the primary crystallization of olivine with 
following resorbtion of olivine under new-formation of orthopy- 
roxene, several investigators have referred to the well-known kely- 
phitic or coronation zones around olivine, with hypersthene in the 
inner zone facing the olivine. As stated above (Vol. XXIX, pp. 645- 
49) we are dealing here, however, with a reaction in the solid face be- 
tween olivine and plagioclase at a maximum temperature of about 
1250 —1200° and generally somewhat lower—thus at least 300° lower 
than the temperature (1557°) at which, under a pressure of one 
atmosphere, olivine may crystallize from a melt of MgSiO;. The 
kelyphitic hypersthene zone next to the olivine may thus have 
nothing to do with the primary segregation of olivine from a melt of 
MgsiO, and the later transformation from olivine to orthopyroxene. 

As maintained by Andersen (Joc. cit. [1915], p. 453) the olivine, 
as a consequence of the splitting of MgSiO,, in the orthopyroxene- 
bearing igneous rocks, should in every case have crystallized earlier 
than the orthopyroxene, and the crystallization of the olivine should 
have been finished even before the beginning of the solidification 
of the orthopyroxene. But petrographic investigation of rocks 
rich in orthopyroxene but poor in olivine, shows that this is not so. 
We refer to the facts mentioned above (Fig. 25). 

The conclusion from this is that bronzite and hypersthene, under 
high pressure, in the common deep-seated magmas (norites, etc.), 
crystallized directly from the magma in the same way as the other 
common silicate minerals. How it may be with enstatite in an 
almost pure MgSiO, magma, remains, however, an open question. 
In the igneous rocks, indeed, we never find orthopyroxene with less 
than ca. 7 per cent FeSiO, (or more than 93 per cent MgSiO,). 


Cc 


1. Many minerals, such as olivine, monoclinic pyroxene, the 
feldspars, spinel, magnetite, etc., may be formed in melts at a 
pressure of one atmosphere, as well as in effusives and in deep- 
seated rocks—consequently at low, as well as at middle high, and 
very high pressure. 

2. Some minerals, as melilite for instance, which crystallize 
out of melts at a pressure of one atmosphere, occur, moreover, in 
effusive rocks (melilite basalts, etc.) and in certain dike rocks (as 


626 IHL. VOGH. 


alndite), but on the other hand are not—as yet (1920)—known from 
deep-seated rocks. 

If we compare a number of analyses of troctolite (“forellen- 
stein’? and feldspar-peridotites) with for instance the analyses 
of slags given in my work “Silikatschmelzlés,” I, p. 16 (containing 
ca. 30-42 per cent SiO,, 8-28 per cent AlO,, 20-40 per cent CaO and 
up to 13 per cent MgO with some FeO, MnO, etc.), in which a 
melilite mineral (melilite-gehlenite) is crystallized, it is apparent 
that by re-melting at one atmosphere some of the deep-seated rocks 
mentioned, which consist only of olivine and anorthite-bytownite, 
they must recrystallize with more or less melilite besides several 
other minerals of which one is spinel (Vol. XXIX, p. 524). 

Gehlenite occurs in some contact zones; it must consequently 
under certain conditions be formed at high pressure. And melilite 
occurs, as just mentioned, as a rarity in some dike rocks, which 
might have been solidified at a tolerably great depth, thus also at a 
tolerably great pressure. 

It is worth noticing, however, that melilite, so far as known, is 
lacking in the common deep-seated rocks, and especially that, 
instead of a melilite mineral, we find in the troctolites the combina- 
tion olivine and anorthite-bytownite. 

A similar case is that of leuczte which can crystallize in melts at 
a pressure of one atmosphere, and which occurs in many effusive 
and some dike rocks, but has only rarely been found in deep-seated 
rocks. When microcline and biotite are melted together in certain 
proportions, leucite results. And in many deep-seated rocks which 
by re-melting at one atmosphere would give leucite, we find instead 
of leucite other minerals, as microcline, biotite, etc. 

3. On the other hand there are some minerals which in part 
preferentially and in part exclusively belong to deep-seated or other 
rocks formed at a very high pressure. As an instance we may 
choose garnet. 

By re-melting garnet there results, as is known, not this mineral, 
but a mixture of other minerals, according to the composition of 
the garnet—melilite, anorthite, olivine, spinel, etc. On the other 
hand, we may get garnet by using certain chlorides, for instance 
AICl,, as the solution medium. Garnet—or at least some varieties 


MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 627 


of garnet—may thus also be formed under high temperature at the 
pressure of one atmosphere. 

Garnet occurs as a primary formation in some effusives and 
often in various deep-seated rocks. The mineral further is formed 
as known inter alia by contact-metamorphism and by intensive 
dynamo-metamorphism. 

Amphibole, as is known, does not crystallize ova anhydrous 
silicate melts at one atmosphere but it is generally supposed to be 
conditioned by high pressure. ‘This depends rather certainly on 
the fact that the magmas only at high pressure may carry the 
sufficient quantity of water (or hydroxyl) which seems to be a 
constitutional component of amphibole. 

Biotite has, as is known, been produced synthetically by melting 
with fluorides, and the magnesia-mica phlogopite has also been 
established as crystallized out of common anhydrous slags contain- 
ing a little fluorine.t Phlogopite may thus be formed by crystal- 
lization of silicate melts at a pressure of one atmosphere. But 
common mica occurs in the igneous rocks preferentially in deep- 
seated rocks; it occurs also, however, in effusives. On the other 
hand, muscovite never occurs as a primary mineral in effusives, but 
occurs in certain granites and especially in granite-pegmatite dikes. 

As we shall show later, in granites of exactly the same composi- 
tion—excluding the original content of H,O, etc.—hypersthene, 
biotite or muscovite may crystallize. Which of these three minerals 
is formed, may depend on the content of H,O, etc., in the magma 
(when least, hypersthene is formed; when most, muscovite). 


t See my cited treatises from 1884-85 and 1888-90. In slags from the Kafveltorp 
copper works (containing 41-46 per cent SiO2, 8-11 Al,O;, 7-15 FeO including ca. 0.5 
Fe,O;, 13-20 CaO, 10-18 MgO, a little ZnO, Cu,S, etc., also 3-5 K,O+Na.0 and a little 
F) there has crystallized 10-15 per cent mica, in great leaves up to 5-6 mm. in diameter, 
with the following characteristics: pseudohexagonale thin leaves, with just as good 
cleavage as natural mica; elastic flexible; pressure figure as in natural mica; prism 
angle ca. 121°; optical biaxial and negative; acute bisectrix almost perpendicular on 
oot, differing only by 1-1.5°; 2 V only a very few degrees; colorless; very vivid inter- 
ference colors; very resistent to acids; chemical composition: 42.20 SiO., 11.30 Al.O;, 
5.92 FeO (with a little Fe.O;), 22.903 MgO, 2.29 CaO, 1.40 ZnO, 0.33 CuzS (mechanical), 
total 86.37 per cent, rest ca. 13 per cent K,0, Na,Oand F. The mineral is thus a MgO- 
mica rather poor in FeO and Fe.0,, viz., a phlogopite, whose content of H,O (or HO) 
is replaced by another component, F. 


628 Jae VE ViOGIE 


The influence of pressure with regard to the minerals amphi- 
bole, biotite, and muscovite must then be indirect, since the forma- 
tion of these minerals depends essentially on the magmatic content 
of H.O, etc., which further, to a certain extent, depends on the 
pressure. 

By great differences in pressure, H.O content, etc., a magma 
otherwise of the same or almost the same chemical composition 
may give rise to the crystallization of totally different minerals or 
combination of minerals. P. Niggli, in his work (1920, p. 207) 
cited below, mentions as a typical instance of this the two rocks, 
(1) a durbachite (rich in biotite and hornblende and further ortho- 
clase, plagioclase, a little quartz and titanite bearing boundary 
facies of granite), and (2) a glassy leucite basalt. 


1. Durbachite 2. Leucitebasalt 

SIO Rs. Sia he oe Bae ection ome see ales 51.05 51.43 
IS Ose ve Se ae cesn ce a nee ae ere 1.76 1.12 
AU O PSone < poe mene cue ee ier, arene 14.49 14.88 
INH OF ee Seaeiceaonecrte redacted Scale 4.10 6.30 
1. O eee psneea aeniy arenes caretaer ints aes 4.37 Beit 
INGOTS tia.b alee thie Rivanert eects ee 8.16 6.67 
CAO ah E ie es cotioegeec an nae eae 5.11 5 Ou 
IN a Oar vata ahe uae Sealenensieua ea ere selicg te 1.85 1.83 
2 OAD ie Mee etna Nestea dA cep te PAM sk os Q.22 
Po Oe ise etsy ena tity masa hacen ark ere et ©.70 0.51 
NTO ay ei Nev slay meine oon atnoue seers I.05 0.74 
Mo talent bce caress eric 99.94 100.85 


1. Durbachite from Durbach, Schwarzwald. A. Sauer, Miti. Badische Geol. 
Landesanst., II, 1892 (1890). 2. Leucitebasalt from Gaussberg, Kaiser Wilhelm IL 
Land. R. Reinisch, Deutsche Siidpol-Exposition II (1), 1906. 


Undoubtedly, on account of the high content of water, biotite 
and hornblende, and further orthoclase, have been formed in the 
durbachite, while on the other hand in leucite basalt which solidified 
at lower pressure leucite was formed instead of orthoclase and K,O- 
rich biotite. 

RS dissolved in silicate melts crystallizes at a pressure of one 
atmosphere as monosulphid (oldhamite, alabandine, sphalerite, 
pyrrhotine ortroilite, etc.), but may at high pressure by pneumatolytic 
processes (in the crystallization of magmas as described by Brogger* 

t Zeits. f. Kryst. Min., XVI (1890), I, p. 161. 


MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 629 


for the nepheline syenite-pegmatite dikes, further in contact zones, 
etc.), give rise to the formation of helvine (and danalite). Ata 
moderately high pressure also some double-compounds of silicate 
and NaCl or Na,SO, (sodalite, nosean and hauyn) crystallize. 
But these minerals—and according to my experience this is also 
true of scapolite—may not be reproduced by melting the respective 
silicate with admixture of NaCl or Na,SO, at common low pressure. 

In this connection it may be further mentioned that the mag- 
matic crystallization of minerals as FeS, and CaCO, imply a high 
pressure. 

This réview proves that the formation of many minerals, or 
combinations of minerals, may take place at low pressure as well as 
at moderately high or very high pressure, and further, that the 
determinations undertaken at one atmosphere pressure in regard to the 
eutectics and mix-crystal systems, so far as these minerals are concerned, 
may also be made valid with only slight corrections at high pressures, 
even at the high pressure prevailing during the crystallization of 
the deap-seated magmas. 

But this may not be applied to all minerals, or combinations of 
minerals. The formation of some minerals, as melilite and leucite, 
is favored by low pressure; the formation of other minerals, as 
garnet, is favored by high pressure. For some minerals, as 
especially muscovite, further H,O-bearing biotite and hornblende, 
a more or less high pressure may be an indispensable condition for 
crystallization from the magma; indeed indirectly depending on 
the fact that the presence of sufficient H.O in the magma implies a 
high pressure. For pyrite and calcspar, which at one atmosphere 
are dissociated by moderate heating, high pressure is necessary for 
their crystallization from the magma. 

It is to be noted that garnet has a much higher density (Mg-AL- 
garnet 3.7-3.8, Ca-Al,-garnet 3.9-4.0, Ca-Fe,-garnet 3.8-4.1, etc.) 
than the combination of those minerals which result from re-melting 
it at one atmosphere. On the other hand, Jeucite has a much lower 
density (2.45-2.50) than the combination of those minerals, espe- 
cially orthoclase (2.55) and the various K,O-rich micas (2.76-3.0) 
which generally replace leucite in the deep-seated magmas. And 
melilite (2.90-2.93) has somewhat Jower density than the combina- 


630 Toil, ib. WOE 


tion of a medium mixture principally of olivine (3.22) and anorthite 
(2.765) which in most cases replace it in the deep-seated rocks. 

In all of these three cases we thus observe the fact that a com- 
bination of minerals, respectively a single mineral (leucite, melilite) 
in the effusives are replaced in the deep-seated rocks by a single 
mineral (garnet), respectively a combination of minerals, charac- 
terized by increased density. 

This recalls the well known “volume law” valid for dynamo- 
metamorphism, viz., that various minerals under high pressure 
form new combinations with the minimum total volume—a law 
whose physicochemical cause, moreover, is not exactly elucidated. 

Since the above section, in the main, was written, in the winter 
1919-20, a treatise on “The Mineral Facies of Rocks,” by P. 
Eskola,? was published, wherein he tries to establish some (five) 
facies for the mineral combinations depending on the pressure not 
only for the metamorphic, but also for the igneous rocks. Eskola 
classes with the last, the facies of the igneous rocks formed under 
the highest pressure, among others those especially investigated by 
him, namely garnet-rich (with up to 75 mol. per cent pyrop- 
component Mg,Al, SiO,0,.) eclogites, which “occupy a volume about 
15 per cent smaller than that of the corresponding gabbros.”’ 
Eskola thus maintains the same views as those I have pointed out 
above. I call attention to this treatise, in which are given some new 
instances of the interdependence of the formation of certain mineral 
combination in the igneous rocks upon the pressure. Further on 
Eskola treats of the interdependence of the metamorphic new- 
formations upon the pressure, which matter I do not touch in this 
treatise. 

[To be continued] 


™ Norsk geol. Tidsskrift, Vol. VI (1920). 


EDITORIAL 


There has come to the editors of this Journal, Volume 1, Number 
1, of the Japanese Journal of Geology and Geography, issued under 
the auspices of the National Research Council, Department of 
Education, Japan. The journal is published in Tokio and will be 
issued as a quarterly. The Journal of Geology extends to this new 
confrére in the service of the earth sciences its hearty congratulations 
and good wishes. For a number of years Japan has been sending 
many of her young geologists to the United States for their special 
training, and the associations thus established have been mutually 
pleasant and profitable. The appearance of this journal, published 
in English, will extend and strengthen the comradeship in science 
between Japan and the United States and with all other English- 
speaking countries. The new journal consists in part of original 
contributions and in part of abstracts of articles appearing in other 
Japanese publications. Many of these cover articles originally 
published in Japanese that would not otherwise be utilizable by 
English-speaking readers. Many of the abstracts are prepared by 
the authors, and others come from such authoritative reviewers as 
Takeo Kato, well known to American geologists. While most of 
the abstracts are in English, one of them is in German. 

135 Tio 1835 


631 


PETROLOGICAL ABSTRACTS AND ReEvIEWws 
ALBERT JOHANNSEN 


Mitter, Witiram J. ‘Geology of the Blue Mountain, New York, 
Quadrangle,” Bull. 192, N. Y. State Mus., 1916 (1917). -Pp. 68, 
map I, pls. 11, fig. 1. 

The Blue Mountain Quadrangle in the Adirondack region lies in northern 
Hamilton County, N. Y. From oldest to youngest, the formations are: the 
Grenville series, limestones, and quartzites, followed by two small intrusions 
of anorthosite. The most widespread rocks of the region are syenite and 
granite with basic phases which are intrusive into both the Grenville series and 
the anorthosite. Following this came gabbro, still later pegmatite and a few 
dikes of diabase. Glacial and postglacial deposits complete the series. The 
rocks of the Grenville series are thought to be sedimentary. Twelve quartz- 
syenites are described, three of which are said to be “practically monzonites.” 
The quartz ranges from 12 to 20 per cent, consequently the reviewer would 
prefer not to call them quartz-syenites, which name he would limit to syenites 
with less than 5 per cent quartz, but granites. In the reviewer’s system ten of 
these rocks are classed in 226’, granites (or in limited sense monzo-granites), 
the remaining two are 226” or quartz-monzonites. The basic phases of the 
syenite are 228, 328, 227’’, 2212, 3211’, tonalites, quartz-monzonite, diorite, 
and monzo-diorite. Of fifteen granitic-syenites and granites, thirteen are 226’ 
(granites), one 126’ (a leuco-granite), and one near Daly’s Moyie sill rock, at 
the intersection of Families 1, 2, 5, and 6’, in Class 2, Order 2. Of six “‘typical 
gabbros,” one (No. 17) is a garnet-bearing melagabbro, and one (No. 52) isa 
garnet-bearing norite. The other four are said to contain oligoclase-labrador- 
ite and andesine-labradorite which, without further descriptions of the feld- 
spars, prevents their classification. 


Mitier, Witiiam J. “Adirondack Anorthosite,” Bull. Geol. Soc. 
Amer., XXIX (1918), 399-462, figs. 3. 


Miller ee exception to Bowen’s statement (Jour. Geol., XXV [1917], 
242) that “Anorthosites are made up almost exclusively of the single mineral 
plagioclase.” Bowen’s statement is and is not true, depending upon whether 
the rock anorthosite or the anorthosite formation is meant. Hunt’s original 
definition applied to a whole series of rocks which are ‘‘composed chiefly of a 
lime-soda feldspar, varying in composition from andesine to anorthite, and 


632 


PETROLOGICAL ABSTRACTS AND REVIEWS 633 


associated with pyroxene or hypersthene. This rock we shall distinguish by 
the name of anorthosite. .... In some cases the . . . . dark mineral is en- 
tirely wanting.” (Geol. Canada, 1863, 22.) According to modern petrographic 
usage the term anorthosite has come to mean a basic-feldspar rock, which is 
practically free from dark minerals, say with less than 5 per cent, yet the 
Canadian anorthosites are spoken of in the sense of Hunt. Bowen, therefore, 
if he speaks of the whole formation, is incorrect when he says that ‘“ Anortho- 
sites are made up almost exclusively of the single mineral plagioclase,” for as 
Miller shows, the dark mineral averages 10 per cent. He says, ‘The main 
bulk . . . . contains 5 to 10 per cent of minerals other than plagioclase ... . 
in many places there are 10 to 20 per cent, or even more of dark mineral. It is 
also true that some portions of the mass contain less than 5 per cent of dark 
mimerals. .. . . Conservatively estimated, I believe the average... . 
anorthosite carries fully 1o per cent of minerals other than plagioclase.” 
Attacking Bowen’s theory that the anorthosite may not have been at one time 
in a molten condition, Miller says: “‘The Adirondack anorthosite would have 
formed a melt of notably more complicated composition than the artificial melt 
with ro per cent diopside and [it would have been formed] under deep-seated 
geologic conditions. Is it safe to say, therefore, that such a melt may not have 
been a true magma with a high percentage of liquid? . . . . Another impor- 
tant consideration is the almost certain presence of very appreciative amounts 
of dissolved vapors, particularly water vapor, in the magma. .... Also the 
presence of about 2 per cent iron oxide in the typical anorthosite should not be 
overlooked... . . All things considered, therefore, I not only think it very 
reasonable to apply the mutual solution theory to the anorthosite, but also to 
regard the anorthosite to have existed in magmatic condition at a moderate 
temperature.’ He says further: “I consider the main steps in the develop- 
ment of the anorthosite to have been as follows: First, intrusion of a lacco- 
lithic body of gabbroid magma . . . . second, relatively rapid cooling of the 
marginal portion to give rise to the chilled gabbroid border phase; and, third, 
‘settling of many of the slowly crystallizing femic minerals in the still molten 
interior portion of the laccolith, leaving a great body of magma to gradually 
crystallize into anorthosite. Thus, at the bottom, and probably nowhere 
visible in the field, lies a mass of pyroxenite or peridotite ... . ” (p. 457). 
Miller says in reference to a syenite (p. 438), “Labradorite and andesine are 
always present and oligoclase usually.” The reviewer must again make the 
statement which he has made a number of times before, that he doubts whether 
in rocks which are composed of crystals of a single generation, two different 
plagioclases can occur together except as zonal growth. 
It is impossible in this abstract to give all of Miller’s conclusions. Briefly 
they are: 
The Adirondack anorthosite is a great laccolithic intrusive body—older 
than the accompanying granite-syenite series. The average rock contains 
fully ro per cent of dark minerals, and is differentiated practically 7m situ from 


634 PETROLOGICAL ABSTRACTS AND REVIEWS 


an intruded gabbroid magma, developing a chilled gabbroid border facies and 
an upper zone of anorthosite from a magma which was to a very considerable 
degree at least, actually molten. The anorthosite-gabbro and gabbro associ- 
ated with the anorthosite represent local differentiates. Syenite and granite 
are not differentiates from the anorthosite, although transition rocks (Keene 
gneiss) were produced locally by magmatic assimilation of still hot, but not 
molten, anorthosite by the syenite or granite magma. 


Mitrer, Witi1AM J. “Geology of the Schroon Lake Quadrangle,” 


Bull. 213, 214, N. Y. State Mus., 1918 (1919). Pp. 102, map 1, 
pls. 14, figs. 9. 

The Schroon Lake quadrangle represents an area of about 215 square miles 
in the central eastern portion of the Adirondack mountain region. The oldest 
rocks are those of the Grenville series which were intruded by a relatively stiff 
gabbroid magma which differentiated and formed the anorthosite. The intru- 
sion of the anorthosite lifted the Grenville strata but also engulfed fragments of 
it. Following this came the intrusion of the syenite-granite series which par- 
tially domed, partially broke and tilled, the Grenville. The metamorphism of 
the Grenville took place probably before or during the period of igneous 
activity. During the succeeding period of uplift there was great erosion and 
some igneous activity indicated by the intrusion of certain diabase dikes. In 
late Cambrian time a gradual submergence took place with deposition of sand- 
stones and dolomites. A long period of erosion from Ordovician to late in the 
Mesozoic reduced the land to a peneplain rising to moderate heights above the 
general level, and this was followed by uplift and active erosion, which con- 
tinues to the present time. During the Ice Age the entire area was covered by 
the ice sheet. Toward its close there was a subsidence of several hundred feet 
below the present level, and finally a differential uplift with greater elevation 
at the north. 


Mitrer, WititaM J. ‘‘Banded Structures of the Adirondack 
Syenite-Granite Series,” Science, XLVIII (1918), 560-63. 


Both assimilation and differentiation contributed to the banding. 


Mitter, Witiiam J. ‘‘Silexite: A New Rock Name,” Science, 
XLIX (1919), 140. 

The name silexite is proposed for any body of pure or nearly pure silica of 

igneous or aqueo-igneous origin which occurs as a dike, segregation mass, or 


inclusicn within or without its parent rock. (See also ‘“‘Pegmatite, Silexite, 
and Aplite of Northern New York,” Jour. Geol., XXVII [1919], 28-54.) 


PETROLOGICAL ABSTRACTS AND REVIEWS 635 


Moore, E. S. “‘Pele’s Tears’ and Their Bearing on the Origin 
of Australites,” Bull. Geol. Soc. Amer., XXVII (1916), 51-55. 


Australites are thought to be of volcanic origin since bodies of distinctly 
volcanic origin show that similar forms can originate in the atmosphere from 
rotating liquid bodies. 


Moore, Raymonp C. “The Relation of the Buried Granite in 
Kansas to Oil Production, Bull. Amer. Assoc. Petroleum Geol., 
IV (1920), 255-6r. 

This paper is petrographically of interest from the short description of a 
buried ridge of granite extending as an elongated mass from north of the 
Nebraska state line into Kay County, Oklahoma. The ridge reaches its high- 
est elevation near the north boundary line of Kansas in Nemaha County, where 
it is less than 500 feet below the surface. It descends gradually to the south 
and forms a saddle in northwestern Wabaunsee County, then rises in Morris 
and Chase counties in underground peaks of different elevations. 


O’Harra, CLeopHAS C. “A Bibliography of the Geology and 
Mining Interests of the Black Hills Region,’ South Dakota 
School of Mines Bull., XI, 1917. Pp. 216-17, map 1. 


A very important and useful bibliography of the Black Hills, containing not 
only titles and references, but abstracts of each of the 1,187 items listed. 


Osann, A. ‘Der chemische Faktor in einer natiirlichen Klassifika- 
tion der Eruptivgesteine, I.” Absandl. d. Heidelberger Akad. 
d. Wiss., Math.-naturw. Kl., Abh. 8, 1919. Pp. 126, pls. 5. 


This is another important contribution to the chemical classification of 
_ rocks, and one which will be welcomed by everyone who uses Osann’s system. 
In former publications the determination of the s, A, C, F, a, c, and f values, 
and their plotting in triangular diagrams ended the attempt at classification. 
Here regular pigeonholes are established, so that it is a very simple matter to 
locate analyses of similar rocks. The first 23 pages of this work are devoted 
principally to general discussions and various modifications of the previous 
methods of calculation. It is to be regretted that nowhere is there given a 
definite set of revised rules for the computation of the formulae. The various 
modifications here proposed can be adopted by those already familar with the 
system, but a beginner will find it necessary to follow the discussion through 
many pages of print to obtain for himself a workable set of rules. It is true 
that the system has gradually developed, and various changes have been 
introduced, but it would seem that it is now in such form that well-defined 
rules could be given, and it is to be hoped that in the near future Professor Osann 


636 PETROLOGICAL ABSTRACTS AND REVIEWS 


will publish such a set. From page 23 to the end various type rocks are 
calculated and the relations between their mineral composition and their posi- 
tion in the classification are discussed. The general divisions of the new 


‘groups are as follows: 


c=0—-0.5 c=I-2 C=3-4 c=5—6 etc., etc. c=I13—14 


a=30 —28 


a=27 —25 
a=24 —22 
etc:, etc. 

CMO rt iC) 
a= 6 = 5 
a= 4.553 
= 2 = i 


a= 0.5-0 


The values of a have intervals of 3 between 30 and 7, from 6 to 1 the interval 
is 2: the values of c, except the first, have intervals of 2. 
The limiting values of the older rock types are: 


eee —————————————e 
EE —— SSS EEE 
LimItInc VALUES 


Rock TYPE 
5 A k Ee he 

Granite and Quartz-diorite.......... 82.5-69 | 26-27 1.8 -1.2 63 
Syenite and Diorite..............-. 74. -57 | 24-23 iy =).(5) 65-51 
Essexite and Gabbro.............-- Jo) Sy || = ©) O19) Sin 7/ 53-46 
Mafic Essexite and Gabbro, 

Hornblendite and Pyroxenite..... 46 —42 2-0 Oo7) =o)2(5 44-42 
Dunite, Peridotite, and 

basic Elornblendite. ......------- Ai S(O) 2-0 O10.-Ons 40-35 
Nephelite-syenite.............--+-- V7 oe || Acaig Ons SOs 7 50-54 
Theralite and Shonkinite........... 53 -49] 12-5 0.65-0.5 46-43 
Alkalic feldspar-free Rocks......... 53 = || At 0.86-0.39 | 50-39 
Anorthositete: nines sii es ae peeetaner Oy =e |) wu oh |} 2.6) =o.@) 57-44 


The first three groups included the acid, neutral, and basic igneous rocks. 
According to Rosenbusch, the acid have 65 per cent, the neutral between 64 
and 52 per cent, and the basic below 52 per cent of SiO,. The fourth and fifth 
groups, with their extrusives are mafic, basic, and ultra basic differentiates of 
the third group. The sixth group is an alkalic side-group of the second, the 
seventh of the third. The eighth group includes rocks of very different mineral- 
ogical character and can be regarded only as a basic side-group of the sixth 
and seventh. The last group is connected with the second and third but 


belongs to the alkali series. 


PETROLOGICAL ABSTRACTS AND REVIEWS 637 


Osann, A. “Der chemische Faktor in einer natiirlichen Klassifi- 
kation der Eruptivgesteine, II,” Abhandl. d. Heidelberger 
Akad. d. Wiss., Math.-naturw. Kl., Abh. 9, 1920. Pp. 50. 


In the first part, reviewed above, the plutonic rocks are discussed; 
in this the extrusives are considered. One hundred and fifty-one types, 
based on 973 analyses, are given, but owing to present conditions of pub- 
lication, only the type rocks are shown, and the number of analyses 
falling in each group is indicated. The tabulation follows that given 
for the plutonites. 


Patton, Horace B. ‘Geology and Ore Deposits of the Platoro- 
Summitville Mining District, Colorado,” Bull. 13. Colo. Geol. 
SUurv., 1917. Pp. 122, maps 3, pls. 4o. 

The region here described lies between Creede, Alamosa, Chama, and Pagosa 
Springs. The rocks which occur are very similar to those described in the 
folios of the San Juan region. Short descriptions are given of rhyolite, latite 
(two chemical analyses), andesite (one analysis), basalt, monzonite (one 
analysis), quartz-monzonite-porphyry and diorite. 


PETROGRAPHIC COMMITTEE. ‘‘Report on British Petrographic 
Nomenclature,” Mineralog. Mag., XIX (1921), 137-47. 

A committee, consisting of Watts, Elsden, Flett, Teall, Thomas, Tyrrell, 
Evans, Hatch, Holmes, Prior, Rastall, and Smith, from the Geological Society 
of London and the Mineralogical Society, report on ninety rock-names and 
_ petrographic terms which have been used in more than one sense by British 
authors, and make recommendations as to those which are to be rejected and 
definitions of those to be retained. Some of the recommendations are in 
direct opposition to the recommendations by the Committee on Petrographic 
' Terms of the U.S.G.S. in 1897 and 1898. For example the British committee 
recommend the use of the terms porphyry and porphyrite in the sense used in 
Germany, namely porphyry for rocks with dominant alkali-feldspar and 
porphyrite with dominant plagioclase. The U.S.G.S. recommended that 
“Porphyry and its derivatives are to be used as purely textural terms, without 
limitation to mineralogical groups. Porphyry will thus apply to all rocks, 
whatever their composition, containing phenocrysts in a distinct groundmass, 
and without regard to the size of the grains of the groundmass. Porphyrite 
is discarded as superfluous. ... . ” Further in regard to the use of the 
hyphen, the U.S.G.S. recommends that only similar terms be hyphenated, thus 
two mineral terms or two rock terms, as biotite-muscovite granite, granite- 
syenite, etc., but mineral and rock, or rock and texture, as unlike terms, are 
not united. The British committee say: “‘When a mineral-name, or names, 


638 PETROLOGICAL ABSTRACTS AND REVIEWS 


and a rock-name are compounded to form a name of ‘specific’ signification, 
these should be joined by a hyphen; e.g. biotite-granite.”” With the usage of 
porphyry in the sense of the U.S.G.S., porphyritic rocks are defined as granite 
porphyry, diorite porphyry, etc., but the British committee says, ‘The name 
granite-porphyry is ambiguous, and should not be used.” Why ambiguous 
is not clear, unless it is thought possible that it may be confused with porphy- 
ritic granite. But porphyritic granite and granite prophyry are quite different 
things. It is recommended that adamellite be discarded, or only used for acid 
members of the monzonite series, and monzonite is to be restricted to rocks of 
the type occurring in the Monzoni district. If monzonite is used in the original 
sense of de Lapparent for the Monzoni rocks, it is a collective name and 
embraces monzonites Brégger, gabbros, and pyroxenites. Personally the 
reviewer would like to see the intermediate rocks, monzonite and adamellite, 
dropped (see Jour. Geol., XX VII [1919], 38, and XXVIII [1920], 229), and the 
term granodiorite returned to its original sense (Jour. Geol., XX VII [1919] 168). 
The latter recommendation is also made by the British committee. The usage 
of the term panidiomorphic by the British committee (following Rosenbusch) 
as given in the definition of aplite, is regarded by the reviewer as incorrect. A 
panidiomorphic rock would be one in which all of the constituents have their 
own crystal boundaries. Such a rock is almost inconceivable: and the texture 
of aplite is as far removed from this as it can possibly be, for it has a saccharoi- 
dal texture, that is, one in which all of the constituents are xenomorphic 
(allotriomorphic); it is, consequently, panxenomorphic or panallotriomorphic. 
The British committee recommend the terms allotriomorphic and idiomorphic 
in preference to xenomorphic and automorphic. The latter, however, have 
priority by one year. The term essexite “is retained for rocks practically 
identical with, or which show but slight divergence from, the original type of 
Salem Neck, Essex Co., Massachusetts,” which is a large order since the 
original locality shows such a divergence of type. It is recommended that 
diabase be dropped and dolerite used. One is as bad as the other. Diorite 
and gabbro are given the meanings accepted in the United States (namely 
separated on the basis of the feldspar), and Harker’s usage is not followed. 
Leuco- as a prefix for leucite, as used by Lacroix, is put in the discard. The 
less exact terms basic, intermediate, and acid are used in preference to sub- 
silicic, mediosilicic, and persilicic of Clarke. 


POWERS, SIDNEY. A Lava Tube at Kilauea. Private Publication. 
Pps 7, pls. 5: 


Describes a lava tube, extending from the Kaluaiki pit crater in a north- 
easterly direction for 1,494 feet with a drop of 73 feet. The maximum height is 
20 feet, the maximum width 22 feet. In the roof of the tube there are more 
than a dozen small, conical cupolas “blow-piped”’ by gas escaping from the 
lava. These cupolas vary from 1 to 8 feet in height and with similar basal 
diameters. Thirty-three cross-sections, drawn to scale, are given. 


PETROLOGICAL ABSTRACTS AND REVIEWS 639 


POWERS, SIDNEY. ‘‘ Volcanic Domes in the Pacific,” Amer. Jour. 
Sct., XLII (1916), 261-74, figs. 5. 
Gives data on various volcanic domes, showing size, composition, and 
important features. Viscosity is the principal factor which determines 
whether the magma shall appear as a flow or dome. In some cases a flow 


comes first, then a dome; in others the dome comes first and the lava later 
breaks through the crust. 


Powers, SIDNEY. ‘Granite in Kansas,’ Amer. Jour. Sct., XLIV 
(1917), 146-50. 
An earlier report on the granite ridge mentioned above (Moore, Ramond C., 
Bull. Amer. Asso. Petro. Geol., TV, 1920). 


QUENSEL, Percy. “De kristallina Sevebergarternas geologiska 
och petrografiska stallning inom Kebnekaiseomradet,”’ Geol. 
Foren. Forhandl., XLI (1919), 19-52, figs. 15, pls. 2, profile r. 


Describes the crystalline schists from Kebnekaise, in Lapland. An 
analysis of a feldspar-rich ‘‘ gneiss-mica-schist”’ is given. 


QUENSEL, PERcy. ‘Zur Kenntnis der Mylonitbildung, erlautert 
an Material aus dem Kebnekaisegebiet,” Bull. Geol. Inst. 
Upsala, XV (1916), 91-116, pls. 4. 

The mylonites from Kebnakaise are regarded as primarily derived from 

_ igneous rocks, though some sediments may occur among them. 


QUENSEL, Percy. ‘Uber ein Vorkommen von Rhombenpor- 
phyren in dem prikambrischen Grundgebirge des Kebnekaise- 
gebietes,” Bull. Geol. Inst. Upsala, XVI (1918), 1-14, pl. r. 


Describes a rhombic porphyry in which the feldspar is a microcline anti- 
perthite (Or,Ab,An2). It is microperthitic and zonal, with a central portion 
of plagioclase surrounded by pink potash feldspar or white albite. The change 
from center outward is not gradual, but abrupt, and seems to indicate a sudden 
change in the chemical character of the magma. 


QUIRKE, TERENCE T. ‘Espanola District, Ontario,” Mem. 102, 
Canadian Geol. Surv., 1917. Pp. 92, map 1, pls. 6, figs. 8. 

The Espanola district is 43 miles west of Sudbury, and comprises an area 
of 116 square miles. The rock formations are principally Huronian, with older 
metamorphosed sedimentary schists and slates, and intrusive greenstones and 
granite. The Huronian rocks are cut by diabase dikes and sills, more or less 


640 PETROLOGICAL ABSTRACTS AND REVIEWS 


contemporaneous with the faulting of the region. Pleistocene formations lie 
directly upon the Pre-Cambrian, and consist of fluvioglacial deposits, till, 
clays, and lake sands. 


QUIRKE, TERENCE T., and FINKELSTEIN, LEo. ‘‘ Measurements 
of the Retinadeenite of Meteorites,” Amer. Jour. Sci., XLIV 
(1917), 237-42. 

Gives the radium content of twenty-two meteorites not previously deter- 
mined. It is found that the average stony meteorite is considerably less radio- 
active than the average igneous rock, probably less than one-fourth that of 


an average granite, and that the metallic meteorites are almost free from 
radioactivity. 


REINHEIMER, SIEGFRIED. Der Diorit vom Buch bei Lindenfels im 
Odenwald mit einem Anhang iiber einige mikroskopische 
Methoden. Inaug. Dissert., Heidelberg, 1920. Pp. 63, Figs. 8, 
Photomela: 


Two “‘diorites” and a gabbro are described. No modal percentages 
are given, but in general it may be said that the main rock from the Buch 
consists of more amphibole than plagioclase with accessory biotite and a 
little quartz. The plagioclase is zonal with cores Ab,,Angs and outer 
zones Abjo—46 ANgo-s,- There is a pale to colorless amphibole and one 
that is green, the latter usually surrounding the former. The author 
says a “‘gabbro tendency” is shown, and that diallage is proxied by an 
amphibole of similar chemical composition. The second rock, from 
Kreuzer’s quarry, near Winterkasten, is similar, although generally the 
amphibole and plagioclase are in equal amounts. The feldspar is 
approximately Ab,,Ang,, biotite is rare, and quartz is wanting. The 
third rock from near Laudenau differs from the other two. It is de- 
scribed as a gabbro of the type of the Scandinavian hyperites. The dark 
minerals consist of green amphibole poikilitically intergrown with plagio- 
clase, and diallage and olivine. The plagioclase is Ab,Ang;. Since 
the plagioclase in all of the rocks is labradorite to bytownite, the reviewer 
would call all the rocks gabbro (amphibole gabbro) in spite of the state- 
ment of the author that the magma must have been poor in lime. (No 
chemical analyses are given.) 

Associated with these rocks are certain diaschistic rocks, among them 
diorite-pegmatite, schlieren of “‘needle-diorite,” and beerbachite. 

In the Appendix the determination oi refractive indices in cleavage 
fragmenis of amphibole by the immersion method, and the determination 
of 2V with the Federow stage are discussed. 


PETROLOGICAL ABSTRACTS AND REVIEWS 641 


RicHarps, H. C. “The Building Stones of Queensland,” Proc. 
Roy. Soc. Queensland, XXX (1918),-97-157, pls. 3, figs. 10. 
The physical, chemical, and mineralogical characters of various available 
building stones, with their good and bad qualities, and a list of structures with 
the stones used, are given. 


Ricuarps, H. C. ‘‘The Volcanic Rocks of Springsure, Central 
Queensland,” Proc. Roc. Soc. Queensland, XXX (1918), 179- 
98, pl. 1, figs. 6. 

The oldest rocks of this area are Paleozoic sediments, consisting of sand- 
stones, gravels, and shales, and some limestone. The volcanic rocks, with a 
total thickness of 1,000 feet, are divided into three groups: a Lower Series 
which consists of basaltic agglomerate and basaltic flows, followed by trachyte 
tuffs and flows, then a return to basaltic flows. The Middle Volcanic Series 
consists of trachytic tuffs and flows which are “really phonolite in the strict 
petrological sense.” In the weathered material, precious opal has been 
obtained. The Upper Volcanic Series consists of basaltic flows to a thickness 
of 600 feet. Three new chemical analyses are given. The upper and lower 
basalts are similar and closely comparable with the composition of the average 
basalt of the world. Richards says: ‘The basaltic rocks. . . . may represent 
outpourings of a comparatively undifferentiated primary basaltic magma..... 
Gravitative differentiation may have gone on to some extent as the lower series 
is olivine free, while the upper series is rich in olivine (fayalite). ... The 
intruded terrane almost certainly contains limestone, and the solution of this 
material to a small extent would be regarded by Daly as sufficient to result in 
the production of the phonolitic material from the calcic basic magma... . . 
-The writer, however, in dealing with the origin of the volcanic rocks of south- 
eastern Queensland, regarded them as being differentiates of a single original 
magma.” 


-RicHARDSON, W. ALFRED. ‘“‘The Marginal Features of a Basic 
Dyke at Peldar Tor, Charnwood Forest,” Geol. Mag., LVIII 
(1921), 170-77. 

A much altered greenstone (dolerite) dike which intrudes dacite shows a 
chilled marginal phase and a crystalline center. At the contact the margins 
are laminated, resembling columnar basalt, but actually consist of alternating 
sheets of country rock and dike. Three possible explanations for the lamina- 
tion are given. 


RicHARDSON, W. ALFRED. “A Method of Constructing Rock- 
Analysis Diagrams on a Statistical Basis,” Mineralog. Mag., 
XIX (1921), 130-36. 


Diagrams for plotting the chemical analysis of a rock are given. 


642 PETROLOGICAL ABSTRACTS AND REVIEWS 


RICHARDSON, W. ALFRED. ‘‘A New Model Rotating-Stage Petro- 
logical Microscope.” Mineralog. Mag., XTX (1920), 96-08. 
Describes a new Swift petrographic microscope. 


RICHARZ, STEPHAN. ‘‘Die Basalte der Oberpfalz,” Zeitschr. d. 
deutsch. geol. Gesell., LX-XII (1920), 1-100. PI. 2, Figs. 8. 


The basalts of the Oberpfalz are very similar megascopically, but 
under the miscroscope three types are recognized; pure nephelite- 
basalts, pure feldspar-basalts, and nephelite-bearing feldspar-basalts. 
No melilite-basalts were found although they occur at Steinberg near 
Hohenberg in Oberfranken. Some of the basalts are rich in both endo- 
genic and exogenic inclusions, others rarely contain any. Among the 
inclusions are olivinefels, pyroxenites, fritted sandstones, and basalt- 
jaspis. In addition to the usual minerals, labradorite, nephelite, augite, 
olivine, magnetite, some glass, there occurs in some of these rocks, 
biotite. In the inclusions aegirite, katophorite, sanidine, oligoclase- 
andesine, quartz, and the recrystallization minerals, natrolite, phillipsite, 
calcite, aragonite, opal, and a new mineral called magnalite, occur. 
Field observations show that the basalt occurs in the form of dikes much 
more commonly than previously thought. The width varies from 50 
to 200 meters. 


RinnE, F. Gesteinskunde. Leipzig, 1921. 6th and 7th (double) 
ed., 8vo. Pp. 365, Figs. 510. 

It is almost impossible in a short review to do justice to a general 
textbook which does not aim to present any startling new theories. 
One can do little more than give a summary of the table of contents. 
The first 120 pages of this book, which is not only a petrography but a 
petrology, are devoted to general geological modes of occurrence of igneous 
rocks, sediments, and crystalline schists, jointing, parting, and other 
structures, petrographic methods, and the mineral constituents of the 
rocks. ‘The final 235 pages are devoted to descriptive petrography of 
igneous, metamorphic, and sedimentary rocks. As an introduction to 
the igneous rocks there are thirty-eight pages devoted to their chemical 
composition and modes of expressing these graphically, differentiation, 
gases in magmas, sequence of crystallization with many diagrams, and 
textures and structures. Following the general descriptions of the 
igneous rocks are a few pages on meteorites. Introducing the sedi- 
mentaries are sections on origin, weathering, transportation, deposition, 


PETROLOGICAL ABSTRACTS AND REVIEWS 643 


and diagenesis. Among the descriptions of the sediments, that on 
salt deposits is especially detailed, occupying sixteen pages. The crys- 
talline schists are preceded by chapters on origin and textures. The 
book is profusely illustrated with photographs, undoubtedly excellent 
in the originals, but not well reproduced. The book may well serve as 
a text for students who have a sufficient command of German. 


ROSENBUSCH-WULFING. Mukroskopische Phystographie der Min- 
eralien und Gesteine. Bd. I. Die petrographisch wichtigen 
Mineralien. Pt. 1. Untersuchungsmethoden. 5th ed., Stutt- 
gart, 1921. Pp. 252, Figs. 192, and a colored plate. 

This well-known work again has been revised, enlarged, and prac- 
tically rewritten by Wiilfing, so that it bears very little resemblance to 
the third edition of Rosenbusch. The present instalment of the book, 
which is the first half of the first part, all so far published, deals with 
methods of preparation, and with general theories of optics. Much 
new material has been added, and some of the old has been omitted to 
keep the size of the book within reasonable bounds, but how much has 
been omitted it is impossible to determine, in many cases, on account 
of the rearrangement and the appearance of only the first part as yet. 

Following the Introduction, which is somewhat condensed, there is 
a section on preparation methods. ‘The history of microscopical research, 
the descriptien of stereographic projection, and the graphical methods 
and formulas, which followed in the fourth edition, are entirely omitted. 
Thirty-five pages on the preparation of thin sections come next, instead 
of being inserted between the chapters on optical principles and optical 
instruments, a decided improvement in arrangement. Here are given 
a number of new devices for cutting, grinding, and polishing. The 
cutting of oriented sections is described in considerable detail, occupying 
with the instructions for cutting plane surfaces and polishing, some fifteen 
pages. 

Optical methods are introduced by a general discussion of theories 
of light. In the preceding edition, following the discussion of the 
indicatrices, the Fresnel ellipsoid, and uniaxial and biaxial ray and 
wave surfaces, came a section on lenses, .microscopes, and various acces- 
sories. In the present edition this is omitted, probably to come later, 
and the discussion of optical phenomena continues unbroken. The 
same arrangement holds throughout the book. All of the theoretical 
material is brought together and the determinative methods are omitted, 
undoubtedly to be collected in the second half. This makes the different 


644 PETROLOGICAL ABSTRACTS AND REVIEWS 


parts of the book much more unified and gives a better appearance, but 
will it be so easy for the student ? 

After a discussion of interference phenomena there is a new section on 
the dispersion of birefringence and the departure of the interference 
colors of crystals from the pure colors of Newton’s scale. Section 46, on 
the movement of light, has been entirely rewritten, and the figures have 
been redrawn. The material in this portion of the work has been so 
radically rearranged that it is difficult to compare it with the old edition, 
especially since some of the material may have been transferred to the 
future second part. 

In the preceding edition, following the theoretical discussion of the 
intensity of transmitted light, came a chapter on the practical methods 
of determining extinction angles. In the new edition this is omitted, 
apparently to be given later, and the theoretical part is followed by an 
explanation of the phenomena in convergent light, and there are given 
the formulas for isochromatic curves, isogyres, and so on. The steps 
in obtaining Neumann’s formula for calculating the values of birefrin- 
gence in any section have been increased from a half to four pages, a 
very considerable help to the student who cannot refer to the original 
article. The section on interference figures also has been much extended, 
the explanation of the cross and rings obtained in uniaxial crystals alone 
having been increased from four and one-half to nine and one-half pages. 
Becke’s skiodromes are given in illustration of both uniaxial and biaxial 
figures. Under dispersion the old cuts have been discarded and new and 
much better drawings, as well as photographs showing dispersion, have 
been inserted. The subject of pleochroism has been extended from 
twelve to sixteen pages, and a new section of two pages on luminescence 
has been added. 

Polarizing prisms, which formerly came before the discussion of 
microscopes, and before the discussion of interference, pleochroism, etc., 
now comes near the end of the first half and takes up twenty-one pages. 
Finally, there is the concluding section on Monochromatic Light, 
increased from six to fourteen pages. Here are now given liquid color 
filters, more material on monochromatic flames and the monochromator, 
and a new description of the mercury lamp. 

The book is now, as it has always been in the past, the one big indis- 
pensable work which begins where others end. Author and publisher 
are to be congratulated on its appearance. Press work, paper, and illus- 
trations are excellent and, in spite of the difficulty of obtaining good 
paper, are fully up to the standard of previous editions. 


PETROLOGICAL ABSTRACTS AND REVIEWS 645 


SCHEURING, GEoRG. “Die mineralogische Zusammensetzung der 
deutsch-siidwest-afrikanischen Diamantsande,” Bettrdge 2z. 
Geol. Erforsch. d. Deutschen Schutzgebiete, Hf. I, 1914. Pp. 4o, 
map e We. 1: 

Describes the minerals associated with the diamond in German Southwest 

Africa. Of petrographic interest from the description of a separating appar- 


atus which might possibly be applied to the preliminary separation of rock 
constituents. 


SCHLOSSMACHER, K. “Zur Erklaérung der Becke’schen Linie,”’ 
Centralbl. f. Min. Geol., etc. (1914), 75-70, figs. 2. 

A theoretical discussion of the Becke line. The writer shows that with 
vertical contacts the phenomenon is to be explained by Snell’s law; with 
inclined contacts between minerals the effects are more complex, and may be 
accounted for in part by the explanation given by Grabham. 


REVIEWS 


The Problem of the St. Peter Sandstone. By CHARLES LAURENCE 
Dake. Bulletin of School of Mines and Metallurgy, Uni- 
versity of Missouri, Vol. 6, No. 1. Rolla, 1921. Pp. 228, 
pls. 30. 

Professor Dake finds the St. Peter sandstone in Minnesota and 
Wisconsin equivalent to the upper part of the Chazy, and in Oklahoma 
and Arkansas, to all of it. It is unconformable with the Potsdam sand- 
stone in Wisconsin, and with strata above the Potsdam farther south, 
up to the Arbuckle limestone in Oklahoma. 

A study of the characteristics of the sandstone, such as composition, 
texture, and structural features was undertaken with a view to determin- 
ing the origin of the sand. The author appears to have started with 
hospitable attitude toward the hypothesis that the sand is of eolian origin, 
but in the end he was led to the conclusion that several of the criteria 
usually held to indicate an eolian origin for sand (1) are “‘of less positive 
significance than is generally believed’; that (2) they are “significant 
only of conditions of transportation, and not of deposition”; that (3) 
they are “sometimes inherited from an older formation’; and that 
(4) they are “not present in the St. Peter in any appreciably greater 
perfection than in other sandstones of the same region known to be 
marine.”’ He also holds that structural features imposed on a formation 
at the time it is laid down are ‘‘the only positive criteria as to condi- 
tions of deposition. These criteria point rather definitely to the marine 
origin of the [St. Peter] formation.”’ 

Of special significance in this connection is the basal conglomerate 
present in many places, for in it there is “no sign of wind polish or of 
faceted forms, and nothing comparable to desert varnish” (p. 187). 

This conclusion as to the origin of the St. Peter sandstone is not only 
interesting in itself, but seems to suggest that the “‘continental deposi- 
tion” idea, long neglected, has of late been overworked. In sundry 
recent publications it has almost seemed that if a marine origin for a 
formation is not proved, a non-marine origin is assumed. This volume 
is a wholesome check to this tendency. 


646 


REVIEWS 647 


Some of the author’s detailed conclusions are as follows: 


1. The composition and texture of a sandstone may furnish criteria regard- 
ing its derivation and transportation, but not regarding its method of depo- 
sition. 

2. The history of the sand grains of a sandstone is usually so complex 
including transportation successively by winds, streams, and waves, that 
textural criteria afford no proof whatever of the nature of transportation, even 
to the Jast deposit in which the sand is found. ... . 

3. The structural and stratigraphic relationships in the field, including 
such features as the character of bedding, cross-bedding, unconformities, lateral 
gradation and similar associated phenomena, constitute the only valid criteria 
for determining the conditions under which a deposit was last laid down, 
and these may sometimes give a clue to the method of transportation fo that 
particular resting place. 

4. The purity of the St. Peter sandstone, while very remarkable, as com- 
pared with that of average sandstones, is . . . . not sufficiently different from 
that of associated marine sandstones to demand any essentially different 
explanation of origin;.... 

5. Size of train, i pure quartz sands, in general, is limited by the size of 
quartz grains in average igneous rocks, and is not a satisfactory criterion of 
wind-blown sands. 

6. The size and uniformity of grain in the St. Peter is so near that of the 
Roubidoux marine sand, that no discriminations as to origin can be made on 
such a basis. 

7. The degree of rounding and frosting of grains, which has been used as 
one of the chief arguments for eolian origin of the St. Peter, may often be 
‘masked by secondary quartz enlargement, but making due allowance for such 
modification, the St. Peter cannot be distinguished, on this basis, from the 
marine Roubidoux, or from older Cambrian sandstones. ... . 

8. The St. Peter shows bedding better developed than cross-bedding, and 
‘does not show typically developed dune-structure, even in the protected basal 
layers in the valleys of the old erosion surface... . . 

rz. Limestone layers occur at many horizons, particularly at the south, 
but are known as far north as north central Iowa and northern Illinois, and 
indicate marine deposition. 

12. Oscillation ripple-marks in sand layers, marine fossils in limestone beds 

. .-occur in Arkansas and Missouri, next above the unconformity [at the 
base of the series], showing submergence before the advance of the sand into 
the region. 

13. Marine fossils are found in the St. Peter as far north as Minneapolis, 
not only in the uppermost transition layers, but also at three horizons, more than 
60 feet below the top. These would not appear to have resulted from working 
over of dune deposits. 


648 REVIEWS 


14-19. The St. Peter appears to have been derived from a relatively 
low land mass to the northward. This land is believed to have sloped 
southward, in which direction its rivers flowed, to have been affected 
by a moderately humid climate, but not to have been clothed with 
vegetation, because land plants had not yet developed. The land 
included pre-Cambrian crystalline rocks and a broad fringe of Potsdam 
sandstone. 

20. The derivation of the St. Peter, largely from this Potsdam belt in which 
the sands were already well assorted and rounded, together with the added 
sorting and rounding by wind work in the supply area, and by waves in the 
sea, explains in a wholly satisfactory manner the high degree of purity and 
rounding of its grains. 

21. These sands were delivered to the sea both by rivers and to a minor 
degree directly by winds, and were distributed chiefly by waves and currents. 

22. The shores of this sea were fluctuating, but during middle and late 
St. Peter time, were for the most part north of the Iowa-Minnesota line. 

23. North of that line there is quite probably a small amount of St. Peter 
that is truly unmodified terrestrial deposit. .. . 

24. South of the Iowa-Minnesota line, conditions of both transportation 
and deposition were almost wholly marine, and in this area there did not exist 
during any part of St. Peter time, a great interior desert of drifting sand. 

A discussion of the geographic conditions under which this and other 


early Proterozoic formations were made, closes the volume. 
REDS 


Deposits of Manganese Ore in Arizona. By E. L. JoNnEs, JR., and 
F. L. Ransome. Bulletin 710-D, United States Geological 
Survey, Government Printing Office, Washington, D.C., 1920. 
Pp. 92, pls. 6, figs. 8. 

The production of manganese ore as such in Arizona dates from 
1915. The producing district lies in the more southern part of the state, — 
The greater part of the ore worked bears at least 35 per cent manganese, 
and not more than 4 per cent iron. ‘The ore is shipped east to Illinois, 
Alabama, Tennessee, and Pennsylvania, and lately also to California. 
Perhaps the chief difficulty encountered in production lies in the inacces- 
sibility of the mines to railroads, which necessitates ‘‘packing” the 
manganese out of the mining district, a tedious and expensive process. 

Various scattered manganese have been studied by Mr. Jones in the 
preparation of this paper. Dr. Ransome describes those at Bisbee and 
Tombstone. In the latter district, the sequence extends from the 
pre-Cambrian Pinal schist through Cambrian, Devonian, Mississippian, 


REVIEWS 649 


Pennsylvanian, Triassic (or Jurassic), and Comanchean rocks. These 
are mostly limestones, though the early Mesozoic is marked by porphy- 
ritic intrusions and the lower part of the Cambrian series is quartzitic. 
The manganese ore, largely psilomelane, occurs in irregular bodies in 
close association with fissures in the Carboniferous limestones; the 
deposits follow the fissures, or extend laterally from them along certain 
beds of limestone; they seldom descend to depths greater than fifty 
feet and are worked by open cuts or shallow inclines. With the hard 
psilomelane are lesser amounts of barite, quartz, a green copper-arsenic 
compound (new species), and soft black pyrolusite; chalcolite is 
occasional. 

In the Tombstone district the manganese grades into ores rich in the 
precious metals; it occurs in irregular pipelike masses or chimneys 
distributed along fault zones. 

Almost certainly these manganese deposits are related to the copper 
ores, as they are generally closely associated. The manganese ores are 
unquestionably supergene, being generally found only in the oxidized 
zone. Psilomelane deposition seems to have been conditioned chiefly 
by fissuring. In the Tombstone district, manganese-silver ores are as 
common as manganese-copper ores in the Bisbee region, and possibly 
the manganese zone here represents a leached silver zone. The deposits, 
on account of their irregularity, can only be worked under unusual 
conditions. Possibly not more than 60,000 tons of 4o per cent ore are 
available in these two districts combined. 

Elsewhere in the state, in Coconino, Graham, Greenlee, Maricopa, 

Mohave, Pinal, Santa Cruz, Yavapai, and Yuma counties, there are 
smaller manganese deposits. Here veins, brecciated zones, bedded 
deposits, and irregular deposits with travertine, all furnish greater or 
lesser amounts of manganese ores. The ores are in pre-Cambrian 
granites and gneisses, Tertiary rhyolites, andesites, and dacites, and 
Quaternary basalts, as well as in limestones and quartzites of Paleozoic 
age, sandstones of Tertiary age and course clastics of the Quaternary. 
The manganiferous silver veins occupy an important place among the 
vein deposits; they are well shown in the Hartshell shear zone ores and 
in the Globe district; in such cases the manganese oxides are psilomelane, 
pyrolusite, braunite, manganite, and wad. There may be more or less 
iron oxide associated, as in the Globe district, where the ores are intimate 
mixtures of manganese- and iron-oxides. Where braunite is the chief 
ore mineral, it is commonly associated with cerusite, vanadinite, and 
wulfenite. 


650 REVIEWS 


Veins barren of silver are widely scattered throughout the southern 
part of the state and include most of the deposits examined. They are 
most common in Tertiary lavas. The ore-shoots vary greatly in size. 
The ore consists of oxides derived from the weathering of vein material, 
hence its depth depends largely on the permeability of the rock to 
circulating water. The ore minerals are psilomelane, pyrolusite, and 
manganite; these are accompanied by barite, calcite, and iron oxide. 

Bedded deposits vary as to character and associated rocks. They 
may be contained in tuffs, or they may be the result of replacement of 
sandstone. They are generally of Tertiary age. Such deposits do not 
extend to great depths and are worked through shallow pits and shafts. 
The manganese minerals are psilomelane, pyrolusite, manganite, and sub- 
ordinately braunite. Quartz, feldspar, iron ores, and calcite are the chief 
gangue minerals—partly secondary, partly the unreplaced minerals of 
the rock. Much of this ore, developed in sandstones and only partially 
replacing the country rock, is siliceous. 

Manganese ore associated with travertine is known from one locality 
only; here the travertine and the clayey manganese-bearing beds are 
capped by basalt. The manganese mineral is principally botryoidal 
and vesicular psilomelane. 

A detailed description of the geography, geology, and manganese 
deposits of each of the districts is given; for these the reader is referred 


directly to the carefully prepared paper itself. 
C. H. B., Jr. 


World Ailas of Commercial Geology; Part I, Distribution of Mineral 
Produciton. United States Geological Survey, 1921. Pp. 72, 
pls. 72. 

The purpose of this atlas, prepared by more than a score of geologists, 
is “to set forth graphically and to describe concisely the basic facts 
concerning both the present and the future sources of the useful 
minerals.” Part I deals chiefly with present sources; later parts will 
exhibit, so far as practicable, the mineral reserves of the world. The 
maps of Part I, which deal with the most important thirty mineral 
commodities, are arranged in groups of eight, each group containing (1) 
a map of the world showing production and, for major commodities, 
consumption by countries in 1913, the last year of normal production; 
(2) a map of each of the continents, indicating production by countries, 
districts, or fields, in percentages of the world’s production in 1913; and 
(3) a map of the United States, exhibiting production in 1918 by states, 
districts, or fields, in percentages of the total output of the country. 


REVIEWS 651 


The accompanying text presents briefly and effectively fundamental facts 
concerning the various minerals, under headings such as uses, geologic 
occurrence, geographic distribution, technology, centers of consumption, 
and the like. 

The atlas is an outgrowth of investigations of mineral problems begun 
during the war, investigations of a type too long delayed in this country, 
for the facts concerning the distribution, quality and quantity, availa- 
bility, and commercial and political control of the world’s mineral 
resources are destined to affect increasingly our trade and industries 


and our relations with other peoples. 
Haran H. Barrows 


An Introduction to Paleontology. By A. Morey Davies. Lon- 
don: Thos. Murby & Co.; New York: D. Van Nostrand & Co. 


Mr. Davies has planned his book with the intention of making it, 
above all else, a practical, usable textbook for courses in the elements of 
paleontology. To this end, he begins with those animals which are most 
common as fossils, and which can be studied most easily by the beginner— 
the Brachiopoda. He first describes certain common species, in order 
to give the student a clear idea of the general characters of the group, 
and then presents a brief, but tolerably adequate, account of the entire 
class. 

From the Brachiopoda the author carries his text along the ascending 
scale, through the vertebrates. He then returns, begins with the 
Echinodermata, and follows the descending order, finishing with the 
Protozoa. The system of treatment violates tradition, and certainly 
has the disadvantage of leaving a beginning student in something of a 
muddle as regards classification. But it has the advantage of beginning 
with the easy, and proceeding to the difficult, and parallels the system 
of treatment used in several of the more recent and progressive high- 
school texts in zodlogy. 

On the other hand, Mr. Davies has adopted a few innovations that 
are neither advantageous nor, so far as can be seen, justifiable. There 
is no clear ground for separating the Molluscoidea into two groups, and 
putting the Bryozoa with the corals; neither is it plain why the Pythono- 
morpha have been omitted entirely, and the Aves reduced to an order 
among the Reptilia. These are points that an instructor may correct, 


but it is not clear why he should be forced to do so. 
Cire 


652 REVIEWS 


The Mineral Resources of the Philippine Islands for the Years 1917 
and z918. Contributors: Ermer D. MERRILL, VICTORIANO 
ELICANO, LEOPOLDO A. Faustino, W. H. OvERBECK, and T. 
Dar Juan. Issued by the Division of Mines, Bureau of 
Science, Bureau of Printing, Government of the Philippine 
Islands, Manila, 1920. Pp. 75, pls. 8, tables 12, and analyses. 


This issue of Mineral Resources of the Philippine Islands deals with 
the years 1917-18, following the plan expressed in the first issue of 1908, 
which was continued by annual publications until 1916. Of great 
interest in the publication under review is the increase in the importance 
of coal, the slump in gold mining, the appearance of asbestos as a mineral 
product of the Islands, and the discovery of more deposits of iron. 
Lesser resources treated are sulphur, manganese, and asphalt. The gold 
production in 1917 was about 1,990,000 grams; in 1918 it was about 
50,000 units less. Silver production increased from 81,000 odd fine 
grams in 1917 to 128,o00-odd in 1918. A still greater advance is recorded 
in the coal production—from 5,000 tons (round numbers) in 1917, to 
15,000 tonsin 1918. Since the year 1907 there has been a steady increase 
in mineral yield except in r1o10. 

In spite of the increasing gold production of the Islands, it must be 
admitted that gold mining is still in the formative stage. The largest 
production comes from Masbate and the Mountain Provinces, and the 
largest single mine is the Benguet Consolidated Mining Company, in 
the Mountain Province district. 

Three sources of iron ore have been exploited in the islands, and 
magnetite sand, briquetted, is also being experimented with. Lead and 
zinc deposits in Marinduque Island are described. They are fissure 
veins, 4 to 10 feet wide, running as high as 60 per cent lead or 45 per cent 
zinc; the minerals are galena and sphalerite, the gangue is quarts, and 
the country rock andesite. Manganese has been produced to some slight 
extent, and copper is mined in Mancayan. 

Most Philippine coal is really lignite, but good bituminous coal is 
known from Polillo Island and in the Zamboango district. At the close 
of 1918 twenty-three mines were producing coal. In the better mines 
a modified room and pillar method is used, the room being also an entry. 
Considerable waste, especially on the part of small operators, is the rule. 
Mine gases are rare, the walls are good, and ventilation problems in 
these shallow mines are simple; transportation, however, is a real diffi- 
culty in coal mining. An interesting feature of coal production is the 
development of producer gas plants on the Islands. 


REVIEWS | 653 


Asbestos manufacture is a new industry in the Philippines. At 
present only one district—that of Ilocos Norte—is productive. Both 
amphibole and chrysotile asbestos are known. An asbestos plant is 
now operating in Manila. 

Oil exploration is so far merely preliminary and confined to the Lake 
Lanao-Cotabato district (Mindanao) and to the Tayabas field. The 
Mindanao oils have a specific gravity of .93 to .gt as analyzed; the base 
is paraffin. 

Sulphur is found in solfataras and in the impure state, mixed with 
volcanic ash, in several localities. The production of cement has 
virtually ceased because of the failure of the largest cement plant. 
Fire-clay, lime, sand and gravel, stone, salt, and mineral and artesian 
waters are the other resources treated. A separate chapter is devoted 
to glass-making; this demonstrates the accessibility of all the material 
necessary to the process—lime, silica, and alkali. 

The report urges the revision of the federal mining laws and the 
establishment of a school of mines. The adoption of a leasehold system 
is advocated; present mining law in the Philippines requires that 200 
pesos worth of development be performed annually on located, unpat- 
ented claims, which “does not always accomplish the purpose sought 
either by the Government or the claim holder.” 

The report contains a directory of mine-owners, lessees, and operators, 
and a transcript of the mining laws of the Philippine Islands. Several 
good photographs accompany the report, but unfortunately others are 

carelessly mounted—possibly the error of the publisher—and still others 
_ show little or nothing of the very features they are presumed to illustrate. 
Some of these shortcomings may no doubt be laid at the door of the 
smallness of the funds available, which has compelled a reduced staff. 
- to disperse its energies over a large field. Similarly, no doubt, the 
general lack of detailed geological descriptions may be accounted for. 


All in all, the report is of distinct value. 
COED Bee 


Deposits of Manganese Ore in Cosia Rica and Panama. By JULIAN 
‘D. Sears. Bulletin 710-C, United States Geological Survey, 
Government Printing Office, Washington, D.C., 1919. Pp. 
31, pls. 1, figs. 28. 
This bulletin is actually two separate papers—one dealing with the 
manganese of Costa Rica, the other with that of Panama. 
The known manganese deposits of Costa Rica are all in the Province 
of Guanacaste, on the Pacific Coast. They are widespread, but generally 


654 REVIEWS 


either of low grade or of small extent. Only two really important depos- 
its are worked. 

The important manganese deposits of Costa Rica are in the Nicoya 
Peninsula, which is very hilly, with a ‘‘backbone”’ running in a generally 
northwesterly direction. The east coast (Nicoya Gulf) is low, with 
swamps and estuaries, while the Pacific coast is high and rugged. ‘The 
rocks are chiefly sandstone, shale, conglomerate, and limestone. Most 
of the sediments have undergone considerable dynamo-metamorphism, 
the greater part being now iron-pigmented quartzite, but other less 
highly colored quartzite occurs higher in the sequence. Silicified lime- 
stone, shale, and breccia are also reported. Igneous rocks include basic 
fine-grained types, largely flows (?). Some intrusions are also thought 
to characterize the region though no plutonic rocks are actually de- 
scribed. Structurally the area seems to indicate an igneous basement, 
with superjacent sediments that have been intricately folded and faulted 
since deposition. 

The ore-bodies are manganese oxides, partly soft (pyrolusite ?), 
partly hard and crystalline. Iron oxide is generally low, but silica occurs 
mechanically admixed. The oxides are found in pockets or troughs 
between the red metamorphosed rocks and lighter colored sediments, or 
may be in direct contact with igneous rock. Generally the deposits are 
too small to merit another term than “pocket,” but they may be as 
large as 500 by 100 feet, averaging 5 feet in thickness. The exact size 
of these ore-bodies is not determinable, and estimates of a reserve based 
on a 40-45 per cent ore are therefore not dependable. 

The ores are related to fault zones but not all the faults of the region 
are ore-bearing. ‘The ore is attributed by the writer to hydrothermal 
action, the hot, ascending solutions passing along the faults and spreading 
on planes of contact between formations and depositing the manganese 
as a carbonate or silicate which was later oxidized. ‘The great silicifica- 
tion of the wall rock and close relation between the ore and fissures are 
supposed to lend credence to this view. : On the other hand the manga- 
nese may be the product of downward concentration, deposited because 
of the impermeability of the highly metamorphosed rocks. 

A detailed discussion of the mines and prospects emphasizes their 
economic insignificance. ‘Two only are of importance as producers at 
present, those at Playa Real and at Curiol, and one prospect (Pavones) 
may prove to be productive in the future. 

In Panama two manganese deposits are known on the west side of 
the Boqueron River, about 20 miles east of Colon. The country near 


REVIEWS 655 


the mines is hilly, and in places reaches altitudes of 1,000 feet. The rock 
immediately adjacent to the deposits is fine-grained sediment containing 
much siliceous cement, or hard, gray, siliceous limestone. Associated 
with the ore-bodies are shales and breccias, and a dark igneous rock, 
probably basalt. 

At Mine No. 1, the manganese occurs as mixed oxides, largely in 
bowlders, or segregated in lenses and sheets in varicolored clays. The 
manganese ore may be in stringers or beds in the clay. It appears to be 
the result of concentration and segregation. Taking all the possible 
sources of manganese together, about 10,000 tons are available at this 
locality. 

At Mine No. 2 the ore is also in bowlders, which lie on the surface, 
or in clay banks; here too are sheets of manganese ore, ranging in thick- 
ness to 15 feet. Manganese is also segregated in a zone in bedded 
breccia, formed apparently through concentration in the residual clays 
that weathered from the breccia. 

In general, therefore, the manganese ores appear to be residual, not 
unlike those of the Piedmont district of southeastern United States. 

(C5 IEE, 1a} o6 Ire 


The Iron and Associated Industries of Lorraine, the Sarre District, 
Luxemburg, and Belgium. By ALFRED H. BROOKS AND 
Morris F. LaCroix. United States Geological Survey, 
Bulletin 703, 1920. Government Printing Office, Washington, 
D.C. Pp. 131, pls. 2, figs. 12, and numerous tables, including 
statistics on Belgian iron and coal production. 

This report was prepared at Paris for the use of the American 
Commission to Negotiate Peace. It illustrates again the value of geology 
in fields normally considered foreign to the science. It calls to mind 
too the desirability of making peace-terms on the basis of such carefully 
organized facts with a view to stabilizing world-industry, rather than 
on the principle that to the victor belong the spoils. 

The purpose of the original report was to lay before the commission 
certain facts relating to the pre-war use of Lorraine iron ore and thereby to 
forecast the probable future of the metallurgical industry in Lorraine as 
modified by the new national control which were under discussion when the 


report was submitted. .... The original report was in effect an argument 
for the adoption of certain policies with reference to the iron and coal industries 
GmcentraleHurope: a. -)- For these reasons the reader will find that certain 


parts of the report are presented as arguments rather than as expositions. 


656 REVIEWS 


The importance of the iron ores of Lorraine Annexee and French 
Lorraine may be shown by the fact that they furnished in 1913 34 per 
cent of the total iron consumed in Europe. 

Lorraine Annexee, that part which Germany controlled subsequent 
to 1872, produced in 1913 75 per cent of the entire German output of 
iron ore. The reserves aggregated about 1,830,000,000 tons of ore 
averaging about 30 per cent iron. More than two-thirds of the coke 
used in Lorraine Annexee came from the Westphalian and Aix-la- 
Chapelle districts; the remainder, only some 1,500,000 tons, was from 
the nearby Sarre field. The European iron reserves in other fields were 
being rapidly depleted, and it would thus have been greatly to the 
interest of Germany to obtain control also of the French Lorraine field. 
German capital already owned 10 to 15 per cent of the entire iron district 
by purchase before the outbreak of the war, and if French Lorraine had 
been annexed, Germany would have controlled 50 per cent of Europe’s 
iron resources. As it is, however, ‘the Treaty of Versailles has left 
Germany with only 7 per cent of Europe’s iron reserves, while France 
owns 48 per cent. Moreover, the deposits of iron ore in the German 
Republic are widely scattered, and some of them are not favorably 
located for economic development. Therefore any large production of 
iron and steel in Germany must be based on imported ores.” Only her 
Westphalian coking coals prevent the immediate annihilation of 
Germany’s metallurgical industry; this coal, it is shown in the report, 
is necessary to assure the economic utilization of France’s Lorraine ores. 

In general the iron deposits of Lorraine occur in a belt extending 
northward from Metz along the pre-war frontier between France and 
Germany in an area averaging 60 kilometers long and 20 kilometers 
wide. The more southerly Nancy iron district lies entirely within the 
pre-war French territory and forms an outlier of the main field. The 
dip of the beds is gently westward, though modified by slight folds and 
faults. | 

The ores are mined at a low cost; this, taken with their great extent, 
their proximity to coal fields and markets, and their composition, 
which adapts them to the basic process, gives them their great value. 
The phosphorus content is 1.5 to 2 per cent, fairly constant, and yields 
valuable slag fertilizer. ‘The iron mines in the occupied parts of Lorraine 
were but little damaged by the Germans, but the furnaces, which might 
later be expected to compete with German ones, were injured or 
destroyed. 

About 74 per cent of all coking coal that is sufficiently near for 
economic use in Lorraine lies in the Westphalian fields of Germany. 


REVIEWS 657 


The Sarre fields can only contribute some 22 per cent of the needed coal, 
so that with the restoration of Lorraine Annexee, France holds much 
smaller coal reserves than either Germany or Great Britain. 

In the Westphalian field of Germany the coal is close to tidewater 
and is connected by rail and by waterways with the iron and steel 
centers of Lorraine and Belgium. ‘The district produces about half of 
the total German output of pig iron, and in 1913 some 45 millions of 
tons of Lorraine ore were smelted there. About 180 coal mines are 
operated. The Sarre coals lie near the Lorraine field—some thirty 
kilometers east—but are far inferior in coking qualities to those of 
Westphalia. There are from 27 to 32 workable seams, aggregating 
about 4o meters in thickness. The total reserve, estimated to a depth 
of 2000 meters, is about 16 million tons. The coal cokes but poorly, 
yielding on the average 50 per cent coke; in blast furnace practice, 
therefore, it is customary to make the charge of equal quantities of Sarre 
and Westphalian fuel. The Sarre coal is really best used for steam, gas, 
and domestic purposes. 

There are no large coal reserves in Belgium, excepting possibly in 
the Campine Basin. The Campine coals, after development has pro- 
ceeded a little further, may, with the Sarre, supply enough coal for all 
the Lorraine ores; but in an open market, they could not compete suc- 
cessfully with the higher grade Westphalian coal. 

Luxemburg bears reserves of iron that should last about thirty-five 
years. It supports large furnaces; there were in 1913, forty-six blast 
_ furnaces and six steel plants. Of the six large corporations that control 
most of the stock, four are German, one Belgian, and one mixed capital. 

The entire compilation is to be commended, first for its purpose 
_ —the application of economic facts to international problems—and 
second for its accuracy and completeness. Two good maps illustrate 
the geographic relations of the coal and iron districts in question; many 
graphs, tables, and diagrams make the salient points doubly clear. 

(Os Jaleo Bay) IRs 


The Earlier Mesozoic Floras of New Zealand. By E. A. NEWELL 
ArBER, M.A., Sc.D., F.G.S., F.L.S. Wellington: New Zealand 
Geological Survey, Palaeontological Bulletin No. 6, 1917. 
ps7 2) pls: 14). 

This memoir is concerned with an account of the earlier Mesozoic 
floras of New Zealand, with which very little work has hitherto been 
attempted. A majority of the species described are new. One result 


658 REVIEWS 


of the work has been to show that, despite assertions to the contrary, 
no trace of any Paleozoic flora has been found in these islands. Rumors 
of the presence of Glossopteris-bearing rocks have no foundation in the 
material studied. Even in Permo-Carboniferous times, when the 
southern continent of Gondwanaland included a large part of the 
Southern Hemisphere, New Zealand did not, on the basis of the known 
evidence, form any part of that continent. Whether beds of Permo- 
Carboniferous age do or do not occur, is not definitely known. 

So far as may be concluded from present evidence, the Mesozoic 
land connections between Antarctica and the temperate regions of the 
Southern Hemisphere appear to have been chiefly in the direction of 
New Zealand and Australia. Although somewhat similar Wealden 
floras are known in South America, the evidence is too meager to warrant 
conclusions concerning its connection with other southern lands. 

The portion of the paper devoted to systematic paleobotany 
includes the description of forty-eight species, all of which are figured. 


The report includes an extensive bibliography. 
; A.C, Mer: 


ERRATA 


Journal of Geology, Vol. XXX, p. 269, footnote, line 2, “Oct., 1922” 
should read ‘“‘Sept., 1922.” 

P. 269, footnote, line 8, ‘‘pp. 36” should read “‘Vol. 36.” 

P. 286, footnote 1, should read ‘‘Cf. footnote 3 on page 270.” 


RLD-ROMIC. diiie 


pam TO ALL LANGUAGES 


BBMOTLEDUE Penh 


Volume 1915-1 


TEXTES PHONIQUES 


La Maison 


Pierre ; a batie 


Fi 

_ Robert Morris Pierce _ 

aaa Soca fot Ameria ef ihe Inernteesl Phonic Avecerog 

“Seay cmt Tp a Tae Powe Sa, Eee 
(Cote en 


Professor of Romance Languages in Harvard Universtty 


Laxcuaces Pusisiina Company 
9 Warr @?ru Sruurr, New-Your 


Lancoaces PusisHinc Company 
ans sraurt, Naw Your 


: KAS é 
ARTS, by RosBert Morris Pierce: four, each measuring (paper-size) 24 by 15 inches, and 
led respectively: Chari of the Universal Alphazom, Chart of the Organs of Speech, Chart of 
nch Alphagam, Chart of the Spanish Alphagam: folded and bound in paper 6% by 43 inches, 
nts each, postpaid; mounted on cloth and rollers, for use as wall-charts, $3.88 each, postpaid. 


‘TION ARIES, with pronunciations in World Romic, edited or selected by ROBERT MorRIs 
CE: (unilingual) English, French, or German; (bilingual) French-English and English-French: 
6 each, posipaid. 


ONUNCI ATION-TABLES, special alphagams, showing, in World Romic, the elementary 
ds of 89 languages, edited by RoBeRT Morris PIERCE, 30 cents postpaid, each language: 
an, Albanian, Old Armenian, Bearnese (Arrens), habe: (Garlin), Bohemian, Bragancian, 
hui, Burgundian, Cantonese, Catalan, Charentese, Cumbrian, Dauphinese, Babien (Bournois), 
ian (Pontarlier), Dutch, Emilian (Badi), Emilian (Bologua), Emilian (Novellara), Tuscan 
jan (Carrara), Tuscan Emilian (Gombitelli), Tuscan Emilian (Sassalbo), English, American 
ish, British English, Scotch English, Middle English (Orm), Middle English (southern dialect, 
century), Middle English (East Midland, time of Chaucer), Old English, Finnish, French, West 
an, Galician, German, Modern Greek, Aeginian Modern Greek, Cycladic igdees Greek, Tachic 
bdern Greek (Ajasso), Lesbic Modern Greek (Mitilini), Lesbic Modern Greek (Honea 
acedonian Modern Greek, Thracian Modern Greek, Old Greek, Hungarian, Old Icelandic, 
an, Japanese, Kanara, ote we Kui, Kurdish, Tahale Languedoc, Latin, Laz, Lien 
bard, Lorrainese, Marchesian, Mingrelian, Molisian, Norman, .Norwegian, Pahari, Picard, 
uguese, Provencal, Old Provencal, Rhetian, Rumanian, Sanskrit, Scotch, Spanish, Andalusian 
anish, Bosnian Judeo-Spanish, Bulgarian Judeo-Spanish, Mexican Spanish, Tridentan, Umbrian 
ubbio), Umbrian (Rieti), Venetian (Grado), Venetian (Murano), Venetian ree) Walloon 
(il ag “ORG Stayelot); Welsh, Yorkshire. 


REMINGTON PORTABLE | 


{ The most complete portable 
typewriter. Has complete 
standard keyboard and other 
“big machine”’ features. 


9 Small, light, simple and 
convenient—fits in case only 
four inches high. _ 


4 These surpassing features 
have made it the leading 
portable typewriter—in sales . 
and popularity. Subscription, $r -50 per weir See copy, 

_ three months’ trial, 25 cents — 
Price, complete with case, $60 2S So aaa 


Remington Typewriter Company 
374 Broadway New York 


Branches and Dealers Everywhere 


“ Microscopen 
66 IlI-M 99 


This simplified and inexpensive stand is in’an deal manner adapted for Stu 
Laboratory purposes. It embraces the high quality of optical .an 
workmanship to the same extent as -the more elaborately pay 
Petrographical Microscopes. 


Object Stage: Rotating, 10 cm in diameter. The einen of the! 'S 

with scale and vernier for correct reading of degrees and fract: 
Micrometer Screw: ‘Ball type’ fine adjustment for microscope tube 
Coarse Adjustment: By rack and pinion. : gt 
Tube Analyzer: With correction lens, to compensate for the differences I S Ss. 


“Amici Bertrand” Lens: In perm ye mounting, attached to ee (to | 
connection with eyepiece I 


Objective Clutch Changer: New patented form, objertives remain inc i 
permanently centered. i 


Three Objective Adapters: With individual ates mount. 
Two Stage Clamps. . 


Illuminating Apparatus: With ‘‘Nicol’’ prism and three-lens condenser, 
attached to polarizer, No. 1504. 


Write for Catalog IT -B (F) 


ee NC. 
NEW YORK / 
60 East 10™St. 


> 


>, a, 5 
OPTICAL ano MECHANICAL 
WORKMANSHIP 


ee eS NUMBER 8° 


ee OF | GEOLOGY 


A SEMI-QUARTERLY 


EDITED By \@ APR 
THOMAS C. CHAMBERLIN. I. 


With the Active Collaboration of 


“ROLLIN T. CHAMBERLIN, Managing Editor 
_Dynamic Geology 


j ELLER, Invertebrate Paleontology - ALBERT JOHANNSEN, Petrology 
BASTIN, Economic Geology J HARLEN BRETZ, Stratigraphic Geology 


ASSOCIATE EDITORS 


Great Britain 2 : *JOHN: C. BRANNER, Leland Stanford Junior University 
: _ RICHARD A. F. PENROSE, Jr., Philadelphia, Pa. 

WILLIAM H. HOBBS, University of Michigan _ 

FRANK D. ADAMS, McGill University 

ensie ices ; CHARLES K. LEITH, University of Wisconsin 

[H DAVID, Australia . é WALLACE W. ATWOOD, Clark University 

St WILLIAM H. EMMONS, University of Minnesota 

' ARTHUR L. DAY, Carnegie Institution 


OF IGNEOUS ROCKS SS (2 oe ee EE Voc (be 
OCENE HISTORY OF THE LOWER WISCONSIN RIVER Pavr MacCumtocx 673 
F THE TRIASSIC TROUGH OF CONNECTICUT - - [Witsur G. FovE 690 


ae GARDNER’ S THEORY OF THE ORIGIN OF CERTAIN CONCRETIONS 
LERoy PATTON 700 


-. GeratD R. MacCartuy 702 


Q 
‘on 
mM 
© 
4 
S 
ica 
iss] 
Lar} 
Ist 
KA 
= 
a « 
i@) 
iva 
i 
Z 
fe 
o 
mn 
is 
wW 
Les] 
> 
@) 
eI 
i) 


ROLOGICAL ABSTRACTS AND REVIEWS - - - - - - - =. = 703 


pas i ge ene te toe aie 


“THE UNIVERSITY: OF ‘CHICAGO PRESS 
CHICAGO, ILLINOIS, U.S.A. — 


: THE CAMBRIDGE UNIVERSITY PRESS, Lonpon 
w THE MARUZEN-KABUSHIKI-KAISHA, Toxyo, Osaka, Kyoto, FuKuoKko, eae 
THE MISSION BOOK COMPANY, SHANGHAI ; 


betta vm ber, tye 
aS i > Paar ah Been “Ai. 
Ps ae 4 A, 


November-December 1 1922 Ee pean Me 


EDITED By 
THOMAS C. CHAMBERLIN | 


With the cae Collaboration of 


ROLLIN T. CHAMBERLIN, Managing Editor ste 


Dynamic Geology ie 
STUART WELLER gy _ ALBERT JOHANNSEN * 


Invertebrate Paleontology — j ; 
EDSON S. BASTIN, ee scr cape HARLEN BRE 
Economic Geology 


The Journal of Geology is published by the University of Chicago at the Univer: 
5750 Ellis Avenue, Chicago, Tllinois, semi-quarterly, on or about the following dates: 
May 1, June 15, August 1, September 15, November 1, December rs. 
per year; the price of ate copies is 65 cents. Orders for service of less than a h 
at the single-copy rate. 1 ix is prepaid o the gene on. ae orders. 


Tauris, q Postage is cHaeeedd extra as fellows: for Ganaen 2 5 cents on ‘win ays Pp 
on single copies 3 cents (total 68 cents); for all other countries in’ the Postal Uni 
subscriptions (total $4.50), on single copies 6 cents ie 71 cents). 


The following are authorized to quote the’ prices indicated: e 
For the British Empire: The ‘Cambridge University Press, F etter Lane, onde 1, E 
subscriptions, including postage, £1 5s. each; single copies, including postage, ‘4S each. 
For Japan and Korea: The Maruzen-Kabushiki- Kaisha, tr to 16. Nihor i 7 
Tokyo, Japan. Yearly subscriptions, including aoe Yen o. 45 each; SS copi 
postage, Yen iT 1.50 each. 


cones: eoiiding postage, $4.50; single copies, mudine postage 71 pane or ‘their quiva 
Chinese money. Rea 
Claim’s for missing numbers should be made within the month following the rege ar m 
cation.. The publishers expect to supply missing numbers free only when losses shave 
transit, and when the reserve stock will permit. — sy ae 
Business correspondence should be addressed to The University of Chicago 1 Press, 
Communications for the editors and manuscripts should be addressed to the Editors 
OF GroLocy, the University of Chicago, Chicago, Til. ie 


to cue author. { 4. While the cost of printing remains as: now, a TSA? canno ane 
out charge, but reprints will be furnished Woe at cost, if ordered in advance ~ 


Entered as second-class matter, March 20, 863," at the Post aifite at Chieago, Ill, under hel by 
Acceptance for mailing at special rate of postage provided for in Section 1105, Act of (0) 


July 15, ro18. 
PRINTED IN THE U.S.A. 


x 


Among articles to appear 
in early numbers of the 
Journal of Geology are 
the following: 


Bere Lava Field of the Parana Basin, 
South America. By CHARLES 
LAWRENCE BAKER. 


The Problems of the Anorthosites and 
Other Monomineral Igneous Rocks. 
By F. LOEWINSON-LESSING. 


The Cycle of Erosion and the Summit 
Level of the Alps. By W. M. 
DAVIS. 


Some New Features in the Physiogra- 
phy and Geology of Greenland. 
By LAUGE KOCH. 


Ward’s Natural Science Establishment 


ROCHESTER, NEW YORK 


Relief Maps 


This map, 29" x 46", scale 32 miles to the inch, with a vertical exaggeration of 20 to 1, costs 
only $30.00. The same map, hand colored to show the geology as given in United States 
Geological Survey professional paper No. 71, is $150.00. Write for circular G-233 describing our 
relief maps. 


Fossils 


We are better prepared than ever before to furnish index fossils and exhibition material. A few 
of our recent acquisitions follow: 


Cambrian material collected by Dr. Wolcot in British Columbia. 

Rare Ordovician material from New York and Bohemia, such as Aeglina, Proetus, Cryptolithus, 
Illaenus, Cystids, etc. : 

Rare Silurian forms from Bohemia such as Ascoceras and Ophidioceras, Cromus, Acidaspis, etc. 

Rare Devonian material from Germany such as Harpes, Cerarges, Goniatites showing sutures, 
Hexacrinus, Cupressocrinus, etc. 

Fine Solenhofen fish, crustaceans and insects, 

Comanchic material from Texas, 

Eocene leaves from Texas and Louisiana, 

Insects in amber. 


Collections of Rocks 


Our rock collections, varying in price from $1.00 to $200.00, are arranged for Genetic, 
Stratigraphic, Petrologic or Petrographic study. Write for circular No. 220, listing these and 
also our Mineral collections. 


WARD’S NATURAL SCIENCE ESTABLISHMENT 
ROCHESTER : . - : = - : NEW YORK 


Goode’s Base Maps 
and Graphs 


Prepared by 


J. PAUL GOODE 


Professor of Geography in the 
University of Chicago 


A series of outline maps for all classes of work: in every grade from 
the common school to the university, in all branches of study, and 
for private use in applied sciences and the various fields of research. 


The desk maps come in three sizes, from 8X 1035 to 2115 inches, 
all perforated for insertion in a standard notebook cover with page 
8X 103 inches. The larger sizes fold into this cover. Several wall 
outline maps are issued, 48 X 36 inches. 


GOODE'S SERIES OF BASE MAPS AND GRAPHS. THE WORLO ON THE HOMALOGRAPHIC PROJECTION (INTERRU! 


Miniature of Map No. 101 He. The homalographic projecticn, interrupted to present the continents 
The entire earth’s surface is shown without distortion of area 


The maps are made from the best drawings, printed with a good quality 
of ink on bond paper, and the presswork is of the highest order. Recent 
issues have been made by the photo-lithographic process, thus preserving 
an absolutely smooth surface of paper on which to work. These maps 
are considered by the many who use them as the best quality of base 


maps on the market. 
Complete list with prices will be sent upon request 


The University of Chicago Press 
5750 Ellis Avenue Chicago, Illinois 


YOUR MAP NEEDS 


can be selected quickly and with assurance of genuine satisfaction from the 
most comprehensive assortment of effective geography ‘“‘teaching tools” ever 


offered. Note below a summary of that part of 


**The Johnston-Nystrom Line’’ 


applying particularly to geography and geology teaching. 


POLITICAL GEOGRAPHY 


Political Maps—Ten Series comprising 136 
different maps. 


Blackboard Outline Maps in two sizes. 
All continents, U.S., World, and many 
states. 


Desk Outline Maps (for pupils) two sizes. 
Wall Outline Maps. 

State Maps. 

Globes—Political, in four sizes. 
Globes—Slated, in four sizes. 
Atlases—Political and Commercial. 


COMMERCIAL GEOGRAPHY 


Finch Series for United States—34 maps 
on to plates, 50’’"X38’’. Shows Products, 
Industries, etc. 


World Maps (Trade and Industry). 
Atlases. 


PHYSICAL GEOGRAPHY 


Physical Maps—Five Series consisting of 
82 different maps. 


Rainfall Maps, U.S., World and Continents. 


1 We 


\ CAB 


PHYSICAL GEOGRAPHY (Continued) 
Vegetation Maps, Continents and World. 
Pressure and Winds Map. 

Major Natural Regions Map. 
Thermal Regions Map. 
Structure Map. 
Timber Distribution Maps. 
Desk Outline Relief Maps. 
U. S. Outline Map with Topography. 
Graph Charts (Geted OLE) two sizes. 
Globes—Physical, 15” diameter. 
Globes—Political. 
Globes—Slated. - 
Tellurians. 
Atlases. 

PICTURES 
Land Forms, 24 Illustrations. 
Physical Phenomena, 16 pictures. 
Plants and Trees of Commerce, 30 pictures. 
Types of Nations, rr pictures. 


Geographical Types and Scenes: 
series, 70 pictures. 


two 


Tell us what subject you are interested in and we will gladly BiVe you intemal 
about any item in our line that may meet your needs. 


USE THIS CONVENIENT COUPON TODAY 


I am interested in the subject checked below and request literature describing your 


maps, etc., applying to the same. 
O Political Geography 

11 Commercial Geography 

[1] Physical Geography 


O Other Items as follows: 


(Go22) 
O Pictures 
O) No. A2z1 Geography Catalog 


Mail A’ J. NYSTROM & CO. 


2249 Calumet Ave, 
CHICAGO 


The University of Chicago 


Science Series 


EDITORIAL COMMITTEE 


ELIAKIM HASTINGS MOORE, Chairman 
JOHN MERLE COULTER PRESTON KYES 


The following volumes are bound uniformly in maroon cloth, 
small 12mo: 


The Evolution of Sex in Plants. By JoHN MERLE COULTER. x-+140 
pages. $1.25, postpaid $1.35. 

Individuality in Organisms. By CHARLES MANNING CHILD. x-+212 
pages. $1.50, postpaid $1.65. 

The Origin of the Earth, By THomas C. CHAMBERLIN. xii+-272 pages. 
$1.75, postpaid $1.85. 

Finite Collineation Groups. By Hans F. BLICHFELDT. xii+-194 pages. 
$1.50, postpaid $1.60. 

A Chemical Sign of Life. By Sutro TasHiro. x-+142 pages. $1.25, 
postpaid $1.35. 

The Biology of Twins. By Horatio Hackett NEWMAN. x-+186 
pages. $1.50, postpaid $1.60. 

The Physiology of Twinning. By Horatio Hackett NEWMAN. x+232 
pages. $1.75, postpaid $1.85. 

Food Poisoning. By EDWIN OAKES JORDAN. viii+116 pages. $1.25, 
postpaid $1.35. 

The Electron: Its Isolation and Measurement, and the Determination of Some 
of Its Properties. By ROBERT ANDREWS MILLIKAN. xii+-268 pages. 
$1.75, postpaid $1.85. 

The Living Cycads. By CHARLES JOSEPH CHAMBERLAIN. xiv+172 
pages. $1.50, postpaid $1.60. 

Problems of Fertilization. By FRANK R. LILLIE. xii+278 pages. $1.75, 
postpaid $1.85. 

The Origin and Development of the Nervous System from a Physiological 
Viewpoint. By CHarLes M. Caitp. xvii+296 pages. $1.75, 
postpaid $1.85. 

IN PRESS 


The Antiquity of Disease. By Roy L. Moopir. 
The Maize Plant. By PAUL WEATHERWAX. 


VOLUMES FOR LATER PUBLICATION 
Mechanics of Delayed Germination in Seeds. By W. CROCKER. 
Black-Body Radiation. By CHARLES E. MENDENHALL. 
The Rigidity of the Earth and of Materials. By A. A. MiIcHELSON. 
Linear Integral Equations in General Analysis. By Et1akim Hastincs Moore. 


ULE: SERS Y OE CHIERG® PRESS 
CHICAGO : : ILLINOIS 


Ge 
7Pin 
Pe 


_ Ct 


PHYSIOGRAPHIC DIAGRAM OF 
THE UNITED STATES 


Size 6244 inches, Scale so miles to one inch 


By A. K. LOBECK 


Department of Geography, University of Wisconsin 


This diagram is intended to be used asa guide in teaching the physiography and 
geology of the United States. The topographic features are expressed with the idea 
of revealing or suggesting the underground structure, and the general character of 
the underlying rock. 


An introductory drill such as having students draw cross sections through the’ 
Adirondack uplift, the Cincinnati or Nashville Domes, across the Appalachians in | 
different places, through the Black Hills, Central Texas, the Ozarks, Wisconsin and | 
many other regions is followed up with the use of other available material, espe-’ 
caly state geological maps and the United States Geological Survey topographic 
sheets. 


The student frequently fails to note the significance of a locality with relation to 
the physiographic province or geological region as a whole. This is overcome by” 
requiring the student to locate upon the diagram the position of the topographic. 
quadrangles under consideration. As practically every topographic map published’ 
by the Survey was consulted in the preparation of the Diagram, most of the signifi- | 
cant features are recognizable in the Diagram too. Seventeen geological provinces’ 
are discussed briefly on the map. 4 


PRICES 


Jf 

Student edition (one color) with one-half of U.S. on each side of sheet, size 31 X44, 
inches. lp 
Singleicopyaneeeeeneeee $1.25 Fifty copies, each......... $1.00 

Ten copies, each...... One hundred copies, each.. .90 

Two color edition with entire map on same side of sheet, size 6244 inches. 
SHMBEA KVM cocachocoe $2.25 Fifty copies, each......... $1.75 

Ten copies, each...... 2.00 One hundred copies, each.. 1.50 


Prices on two color edition, mounted on cloth for permanent class room use, SiZ¢” 
62X44 inches. 


With Common Rollers at Top and Bottom................... $4.50 
On Spring Roller and Board with Dust Proof Cover...... Ra iici!'s 
In Steel Spring Roller ‘Case... 5.0 c.c0.. soe eee eee 8.75 


Publishers 


Cambridge Gntversity Press 


The Study of Geological Maps. By Gerreune L. Ettes, D.Sc., Fellow of Newn- 
ham College, Cambridge. Crown 4to. With 7 plates, 64 figures, and a map. Price on 
application. Cambridge Geological Series. 


A Manual of Seismology. By Cuartzs Davison, Sc.D. With 100 illustrations. 
Demy 8vo. Price on application. 


Petrology for Students: an Introduction to the Study of Rocks under 


the Microscope. By Atrrep Harker, M.A., F.R.S., F.G.S. Fifth edition, revised. 
Crown 8vo. $3.75. 


A Treatise on Crystallography. By W. J. Lewis, M.A. Demy 8vo. $5.00. 


The Principles of Stratigraphical Geology. By J. E. Marr, Sc.D., F.R.S., F.G.S. 
Crown 8vo. $2.50. 


An prec onction te to Geology. By J. E. Marr, Sc.D., F.R.S., F.G.S. Crown 8vo. 
2.50 


Published by the Cambridge University Press (England) 
The Macmillan Co., Agents in the United States 
64 and 66 Fifth Avenue, New York City 


Essentials for the Microscopical Determination of 
| Rock-forming Minerals and Rocks 


By ALBERT JOHANNSEN 


Professor of Petrology 
The University of Chicago 


A laboratory manual designed to provide students of petrology and geologists 
interested in petrology with information on methods of determining microscopically 
the points of separation between similar minerals and their modes of occurrence. 

This valuable volume contains, in addition to a folding alphabetical list so 
placed that reference may be made to it without turning pages, the following 
important features: ‘“A Quantitative Mineralogical Classification of Igneous 
Rocks”’; ““ASummary of Petrographic Methods’’; “Modes of Occurrence of Various 
Minerals”; “Rules for Computing Rocks from Their Modes”; and “Names Pro- 
posed for Various Families.”’ 

Cloth; $2.00, postpaid $2.10 


THE UNIVERSITY - CHICAGO PRESS 
CHICAGO - - ILLINOIS 


The BOOK of LAKE GENEVA 


By DR. PAUL B. JENKINS 


J.W. MC CARTNE 


Yerkes Observatory of the University of Chicago 


If you have a home at Lake Geneva, 
Wisconsin, or have ever visited there 
you will want this handsome volume. 
It is a swiftly-moving story of the 
history of the lake, a nature-guide for 
the tourists who visit it, and a splen- 
didly illustrated gift book, describing 
the homes and institutions on its 
shores. Your dealer has it on sale 
or you may purchase direct from us. 


$4.00 net, postpaid $4.20 


Published for the Chicago Historical Society by 


THE UNIVERSITY OF CHICAGO PRESS 


VOLUME XXX NUMBER 8 


THE 


HOURNAL OF GEOLOGY 


November- December 1922 


THE PHYSICAL CHEMISTRY OF THE CRYSTALLIZATION 
AND MAGMATIC DIFFERENTIATION 
OF IGNEOUS ROCKS 


J. H. L. VOGT 
Trondhjem, Norway 


VI 


THE INFLUENCE OF THE LIGHT VOLATILE COMPOUNDS 
(H,0, CO2, ETC.) ’ 

The physico-chemistry of the light volatile compounds in the 
‘magmas has lately been explained in two treatises published at the 
same time and independently of each other, namely: 

P. Niggli. Die leichtfliichtigen Bestandteile im Magma. Preisschrift 
‘von der Fiirstlich Jablonowkischen Gesellschaft zu Leipzig, 1920. 

- Th. Vogt. Lecture in the Society of Science, Kristiania Meeting, 
April 16, 1920. 

Referring to these two treatises and to H. E. Boeke’s ‘‘ Grund- 
lagen der physikalisch-chemischen Petrographie”’ (1915, especially 
pp. 256-63), we will begin with a general orientation. 

We will discuss a binary system at constant pressure, in which 
we think of a very high pressure, between A and B, with melting 
point, respectively critical temperature, 5,, K, and S,;, K, where S, 
lies notably higher than K, (illustrated. by Fig. 52). The “critical 
curve’’ cuts the curve of the gas phase. In this system, when 


650 


660 Ws ele Ib, WADE Ih 


applied to the combination H,O:silicate, the water at a low tem- 
perature will only keep dissolved a trifle of silicate. ‘The eutectic 
point (E in Fig. 52) of the water solution will consequently practi- 
cally coincide with S, which is illustrated by the schematic Fig. 53, 
copied from Th. Vogt’s treatise where, under the signature B, we 
keep together the different sili- 
cate components, for instance, in 
a granite. 

We may here distinguish 
between three temperature 
stages: 

1. From S, (the melting point 
of the silicate) to Q. 

We here get a crystalliza- 
tion of the silicate by continu- 
ously decreasing temperature, 
the quantity of H,O increasing 
in the mother liquid (the rest- 
magma). ‘There is at last pro- 
duced a magma relatively 
strongly enriched in H,O, of the 

Fait wh composition Q. 
ee A nae At the same time some of the 
ie site quiets perien ght volatile compounds escape, 
temperature as ordinate. So we get a gas phase which can 
effect pneumatolytic formations. 
Concerning this we refer to the explanation on the fundamentals 
given by Th. Vogt and Niggli in Beyschlag-Krusch-J. H. L. Vogt, 
Erzlagerstitten, II, second edition (1921), pp. 555-61. 

The gas pressure (the pressure of the escaping gas phase) at S, 
(thus for pure B) is very little and increases with the quantity of 
the light volatile compounds dissolved in the magmatic solution (up 
to Q). The gases escape when the gas pressure exceeds the 
external pressure. At a relatively low external pressure, as for 
the crystallization of a magma at a little depth, the light volatile 
compounds will consequently escape at higher temperatures, thus 
also at a relatively earlier stage than at higher pressure, by the 


MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 661 


solidification of the magma in a depth, for instance, of 5 or 10 
kilometers. And in the last case the escaping gas must dissolve 


in itself more of the sili- 
cate compounds. Th. 
Vogt applies this to very 
instructive geological 
examples. 

2. At the tempera- 
ture interval Q—P there 
exists a gas (H,O in su- 
percritical condition 
with some dissolved sili- 
cate in gas-formed con- 
dition) from which more 
or less silicate may cry- 
stallize directly from 
the gas. 

3. From P (the criti- 
cal point of H,O con- 
taining some B in 
solution, which point 
lies somewhat higher 
than the critical point, 
~K,=374°, of pure H,O) 
to E (Fig. 52) or S, (in 
Fig. 53, where E and S, 
practically fall together) 
we get a liquid very rich 
in H,O, which may effect 
divers hydrotermal for- 
mations. Further on 
we get a gas phase con- 
sisting of quite predomi- 
nant H,O and only a 
very little B. 


Fic. 53.—System A:B, where A=H,0 and B= 


silicate (silicates of the granite). Qu=quartz point 
(inversion points between a and 6 quartz, at one at- 
mosphere at 575° and at pressure somewhat higher). 

The pressure may be indicated by axes perpendic- 
ular to the plane of the paper. At relatively low 
pressure in the deep-seated magmas and conse- 
quently at high temperature, contact gas escapes, 
giving as a result the contact metamorphic ore de- 
posits of the Kristiania type. At higher pressure 
and consequently at a somewhat lower tempera- 
ture, contact gases escape with a higher solution of 
silicates, etc., giving, as a result, contact metamor- 
phism of the Orijarvi type (‘‘Orjv. gas”). After 
Th. Vogt. 

By far-advanced crystallization of the granite 
magma escape “‘cassiterite gases” (Sn F,, etc.), 
giving as a result the minerals of the cassiterite 
deposits. 


We shall in the following pages only treat the crystallization 
in a magma, containing light volatile compounds, and not employ 


662 fH. £. VOGT, 


ourselves with the pneumatolytic, pyrohydatogen, and hydrotermal 
processes effected by the escaping gas. 

During the crystallization of the magma the light volatile 
compounds will conduct themselves in different ways: 

1. Some will, if the gas pressure is sufficiently high, escape from 
the magma. 

2. Some will be inclosed in the crystallizing minerals as gas- or 
liquid-inclusions (for instance, the well-known pores in quartz, at 
low temperature chiefly consisting of liquid CO,). 

3. Some may enter in solid solution into the crystallizing miner- 
als. The quantity calculated by weight—or molecular weight— 
of these ‘‘occluded”’ gases is, however, very little. 

4. Some may, under certain circumstances, enter into the crys- 
tallizing minerals, for instance H,O in muscovite, biotite or horn- 
blende, CO, in calcspar, etc. 

5. The rest of the gas will remain in dissolved condition in the 
magma and be concentrated in the mother-liquid, by continuously 
decreasing temperature until it reaches Q. 

These light volatile compounds dissolved in the magma exist 
in the reciprocal solution in the same manner as the other solution- 
components. If we think of a magma containing mix-crystal 
components, as Ab and An besides dissolved water, the water will 
not influence the mix-crystal system. And if we have a ternary 
system of two components independent of each other, a and 6 
and also a little water (in lesser amount than that equivalent to Q 
on Figs. 52-53), the sequence of the crystallization will be between 
a and b, dependent on the quantitative relation between a and 0 
in relation to a eutectic boundary curve, beginning at the binary 
eutectic E,_}. The further extension of the curve will be stipu- 
lated by the third component, H,O, present in small amount. With 
a slight quantity of H.O the relation between a and 0 on the bound- 
ary curve will be almost exactly the same as in the binary eutectic 
E,_»- That is to say, the sequence of the crystallization between 


By heating to below the melting point of the rocks these occluded gases may 
escape and thereby break the rock into pieces. In this manner many rocks (granite, 
syenite, gabbro, etc.) may be completely desintegrated by fire. 


MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 663 


a and 6 will be displaced only very inconsiderably by the presence 
of some dissolved water.* 

We have here supposed that H,O enters as an independent com- 
ponent (H.O). But it may also be thought that H,O, CO., HCl, 
etc., under certain conditions, partly also may form special combina- 
tions (for instance H,SiO,, H,SiO,, H Al SiO,, etc.), which may be 
broken up during the run of the crystallization. Hereby there is 
a possibility for complications which will be very difficult to explore. 

In a binary system (under high pressure) of a light volatile 
component, as, for instance, H,O, and a silicate, as Ab, An, Or, 
Ca Mg Si,0,, etc., even a little H.O will effect a relatively consider- 
able depression of the melting point of the silicate, and this depres- 
sion is nearly proportional to the quantity of H,O, etc. We refer 
to the explanation given by Niggli (1920), pages 34-35, and illus- 
trated by his Figure 2, [I and III. And in general, the light volatile 
components (H,O, CO., HCl, etc.) will effect a tolerably consider- 
able depression of the crystallization interval. 

From theoretical reasons it must be presupposed that the light 
volatile components may effect a very considerable increase of the 
thin-fluidity. ‘This is, as far as I know, not verified by laboratory 
experiments, but that it is the case, may inter alia be proved by the 
fact that the magma of granite-pegmatite dikes, in spite of the low 
_ crystallization interval (about 800°—700°) must have been extremely 
thin-fluid. 

With regard to the quantity of the light volatile components in 
_ the various eruptive magmas, we may present the following general 
consideration. 

By crystallization-differentiation in a parent-magma with a 
certain percentage of light volatile components (as, for instance, 
H,O), the first separated crystals will, in most cases because of their 
increased density, sink into a deeper-lying and _ higher-heated 
magma-layer where they are dissolved (or resorbed). As we shall 

1 Tn this connection we will cite a portion of the conclusion in the treatise of P. 
Niggli, “Die Gase im Magma” (Centralbl. f. Min., etc. [1912], pp. 337-38). “Es 
ergiebt sich aus den (physikalisch-chemischen) Versuchen dass in vielen Fallen (hoher 
Druck) die Gasmineralisatoren wie eine andere Komponente behandelt werden 


kénnen, dass also Vogt’s Ansicht, dass einfache Schmelzfliisse schon viele petro- 
graphische Probleme erleuchtern, richtig ist.” 


664 JH EL VOCE 


see in a later paper, there results from this process, as a rule, after 
repeated crystallization-differentiations, the anchi-monomineral 
magmas (for example, anorthosite, dunite), where the original quan- 
tity of the light volatile compounds in the parent-magma must 
be considerably diluted. 

By the segregation of the anchi-monomineral magmas, the light 
volatile components, so far as they do not escape, will remain in 
the rest-magma. Extensive crystallization-differentiation results 
in anchi-eutectic magmas (most gabbros, norites, syenites, granites) 
and as the final product of the differentiation running in the anchi- 
eutectic direction are brought out the granitic magmas. 

Consequently, we must a priori presume that the light volatile 
components on an average will be in the smallest quantity in the 
anchi-monomineral magmas which must have been “dry” or 
“almost dry melts.’’ A somewhat higher percentage may generally 
occur in the anchi-eutectic magmas and indeed especially in the 
granitic magmas. And in the last ones result, as the final solution 
at a very great depth (after Th. Vogt) where the light volatile com- 
ponents cannot, or can only in part, escape the granite pegmatitic 
magmas where we may expect a relatively extensive concentration 
of the volatile components. 

As support to this theoretical deduction we shall first point 
out that miarolitic druses, according to my own field observations, 
generally are completely lacking in anorthosites, dunites and 
petrographically related rocks. On the other hand, they are 
very common in many granites, quartz-syenites, etc., and in the 
miarolitic druses we often find, as well known, a supply of 
pneumatolytic minerals proving that these druses must have been 
genetically connected with the volatile components. 

Further, according to my own field observations, tourmaline 
and other pneumatolytic minerals are completely or almost com- 
pletely wanting in the Norwegian massives of anorthosite and also 
in the numerous but certainly, as a rule, very small massives of 
peridotite. By far the most magmatic-epigenetic (pneumatolytic, 
pyrohydatogen, and hydrotermal) ore deposits are connected with 
acid or intermediate igneous rocks (granite, quartz monzonites, 
etc.). These ore deposits are, in relation to the extension of the 


MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 665 


igneous rocks, relatively much rarer with basic igneous rocks, and 
they are, so far as I know, entirely lacking in the often very large 
massives of anorthosite. As more fully treated in Beyschlag- 
Krusch-Vogt (Die Erzlagerstatten, Vol. II, 2d ed. [1921], pp. 564-65), 
this may be explained by the relatively high content of volatile com- 
pounds in the graniticmagmas. And as far as the granite-pegmatite 
dikes are concerned, these are, as is known, frequently character- 
ized by a very considerable supply of pneumatolytic (or magmatic- 
pneumatolytic) minerals. 

We further get very instructive information in investigating the 
relations between orthorhombic and monoclinic pyroxene on the one 
hand, and hornblende, biotite and muscovite on the other, in the 
various rocks. 


As is known, muscovite is noted by Tschermak with the formula 
KH.AI,(SiO,); and biotite (meroxene) with K,HAI,(SiO),-nMg,SiO,, where 
n is 3 or at times somewhat lower. Some of K usually is replaced by Na, and 
especially in biotite some of Al by Fe and some Mg by Fe. Further some F 
commonly enters into the mica, replacing O (or HO?). The relation between 
K and H moreover, is subject to certain variations. These standard formulas 
prove that the muscovite contains considerably more H,O' than the biotite. 
Primary muscovite from granite-pegmatite dikes (and granite) contains, accord- 
ing to the analyses at hand, mostly 5-8 per cent H,O and 11-9 per cent K,O 
+Na,O, and the biotite from granite and other acid igneous rock, mostly 
1-2 4 2.5 per cent HO and 9-7 per cent K,0+Na,0. 

Also the hornblende commonly carries some H,O (or HO), viz., in the 
igneous rocks at most 2 per cent, usually considerably less, and tremolite 
up to 2.5 per cent. 


As to the conditions for the formation of biotite in the igneous 
rocks, we refer especially to the account given by N. L. Bowen? 
and to the treatise of P. Niggli,3 “Die gasformigen Mineralisationen 
im Magma.” 

In the previously (pp. 430-35) described orbicular quartz-norite 
from Romsaas, consisting of ca. 63 per cent hypersthene, 8 per cent 
biotite, 24 per cent plagioclase (on an average Ab,,Or,An;5), 4 per 


«Here and in the following I do not enter upon the question on whether it is 
hydroxyl, HO, or H,O, which appears. 

2 “The Later Stages of the Evolution of the Igneous Rocks,” Jour. of Geol. (191 5). 

3 Geol. Rundschau, II (1912). 


666 JH ES VOGT 


cent quartz, and a little rutile and apatite, the individualization 
began with the crystallization of hypersthene in large quantities. 
Then followed, during a short stage, a simultaneous crystallization 
of hypersthene and biotite, while the Mg, Fe-silicate present in the 
magma during the last stage entered totally into the biotite. 

The anhydrous meta-silicate hypersthene thus during the later 
stage of the crystallization—even after the percentage of SiO, in 
the residual magma was quite considerably increased, viz., to about 
61 per cent—was replaced by the hydrous ortho-silicate biotite. This 
must be due to the fact that the original percentage of H.O of the 
magma was so small, that at first hypersthene could be formed. 
But when a considerable quantity of this mineral was segregated 
H,O became so strongly concentrated that biotite also could be 
formed. And during the last stage, when the remaining mass was 
reduced to ca. 33, of the entire rock, the quantity of H.O was thus 
very considerably increased, so that the formation of hypersthene 
ceased, and biotite was formed instead. 

As we shall treat more particularly in a later paper, there occur, 
in the quartz-norite massive at Romsaas, a number of pegmatitic 
“Schlieren” and dikes of nearly the same mineral composition as the 
intervening mass between the orbs of the orbicular norites, but 
with the plagioclase (AbssAn,.) somewhat richer in Ab and with 
somewhat more quartz. These pegmatitic “Schlieren,” etc., 
must represent the end-magma, resulting from a very late stage 
where the quantity of H,O was still more concentrated. This is in 
accordance with the pegmatitic structure and moreover with the 
fact that the Mg, Fe-silicate here only enters into biotite, while 
hypersthene is entirely lacking. 

In several norites from Norwegian localities examined by me, 
biotite is entirely lacking in almost all thin sections from one single 
field (Ertelien on Ringerike) while a little biotite occurs in most 
fields, mostly 2, 3, 4, or 5 per cent, and only as a rare exception, as in 
the quartz-norite from Romsaas, as much as 8 per cent. Where 
both hypersthene and biotite occur, the last one always, as just 
described from Romsaas, belongs to a somewhat later stage than 
the hypersthene. It is especially to be emphasized that neither 
the absolute quantity of biotite nor the quantitative proportion 


MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 667 


between biotite and hypersthene stands in any relation to the com- 
position of the rocks determined by the quantitative chemical 
analysis. The formation of biotite thus is not dependent on 
the content of K,O in the rocks nor on the proportion of K,O to 
MgO (or MgO+FeO). In some norites with 0.5-0.7 per cent K,0 
the entire quantity of K,O enters into the feldspar (as KAISi,O, 
in the plagioclase). In other norites with the same percentage 
of K,O, as much as ca. $ of the content of K,O of the rock may 
enter into the biotite and only + into KAISi,O; of the plagio- 
clase. When we consider this in the light of all the other obser- 
vations here treated, the conclusion clearly is justified that the 
cause of the greatly varying quantity of biotite in these rocks 
must be due to the variations in the quantity of H,O in the magma. 
But even in those norites that carry as much as ca. ¢ of biotite in 
proportion to the sum of hypersthene and biotite, the H,O per- 
centage in the magma must have been rather small. 

As earlier mentioned, many Norwegian norites and gabbros 
do not contain any primary hornblende at all, and where this 
mineral occurs it is somewhat younger than the pyroxene (p. 521). 
We may here apply the same considerations as those regarding the 
relation between the hypersthene and biotite in the quartz norite 
from Romsaas. 

The anorthosites constantly carry, as is well known, a small 
- admixture of hypersthene, or sometimes of augite (diallage) and 
olivine, while primary hornblende seems to be entirely lacking. 

As we shall show in a later paper, the peridotite series in the 
_ first stage of concentration—carrying about 35-50 per cent olivine 
and with a chemical composition 41-49 per cent SiO,, 5-10 Al,O;, 
6-10 CaO, 0.25~-2 alkalies, 1o-15 FeO, and 15-25 MgO, thus about 
0.75 MgO:0.25 FeO—in almost every case is characterized by some 
primary hornblende, at times also by some primary biotite. In the 
progressive concentration of the olivine—and simultaneously with 
diminishing percentage of Al,O;,, CaO and alkalies and increasing 
Mg,SiO, in proportion to Fe,SiO,—the hornblende is, on an aver- 
age, diminishing in quantity and in peridotite rocks with at least 
85-00 per cent olivine hornblende as a primary formation is entirely 
or almost entirely lacking. 


668 JE. VOGE 


The formation of the minerals in both the anorthosites and the 
olivine rocks—with at least, respectively, ca. go per cent plagioclase 
(labradorite-bytownite) and 85-90 per cent olivine (poor in iron)— 
thus indicates a crystallization of a magma very poor in H,O. 

As to the relation between hypersthene, biotite, and biotite+-mus- 
covtte in the alcali granites, we shall as a beginning cite a very instruc- 
tive statement by H. Rosenbusch (Mikroskop. Phys. d. Massigen 
Gesteine, II, 1 [1907], p. 71): “Die Analyse eines Hypersthen- 
granites und eines gewohnlichen normalen Alkaligranite sind nicht 
sicher zu unterschieden.”’ 

In granites with composition 


73-77  ~per cent SiO, 

TI—-15 2 ANON 

1 BS > FeO reo 
©) =0)59) Sy ERO) 

On25 an ae ea@) 

5-8 <  KCO--NaO; 


(with varying proportion of K,O and Na,O) we find, in some cases, 
though rather rarely, hypersthene—in by far the most cases biotite 
—and at times biotite-+-muscovite. 

In the hypersthene-granite from Birkrem and environs in the 
Ekersund Soggendal-field, the hypersthene (opt. neg.), according 
to my determination, shows axial angle ca. 70°, the composition is 
thus ca. 0.64 MgSiO,:0.36 FeSiO, (equivalent to about 23 per cent 
MgO and 18 per cent FeO). The quantity of hypersthene in this 
rock (with ca. 73-75 per cent SiO.) according to microscopical 
examination is quite small, about 1 per cent, corresponding to ca. 
0.2-0.25 per cent MgO-+o.2 per cent FeO. An analysis published 
by C. F. Kolderup' shows 73.47 per cent SiO,, 0.12 TiO., 15.42 
Al,O;, 1.02 Fe,O, (including FeO), 0.20 MgO, 1.35 CaO, 5.57 Na,O 
and 3.64 K.O, thus stoechiometric 0.70 Na,O:0.30 K.0. The three 
analyses of charnockite (hypersthene-granite) from Madras? (India) 
with 75.3-77.5 per cent SiO,, on the other hand show in part a middle 

t Das Labradorfelsgebiet bei Ekersund und Soggendal, Bergens Museums Aarbog 
(1896), p. 96. 

2 Cited from H. S. Washington, Chemical Analyses of Igneous Rocks, 1884-1913 
(1917, pp. 88 and 956). 


MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 669 


proportion of Na,O to Ka,O, and in part predominantly K,O, viz., 
almost 0.7 K,0:0.3 Na,O. We shall group the contents of MgO 
and K,O in the just-mentioned analyses of hypersthene-granite 
with 73.5-77.5 per cent SiO,: 


per cent MgO....0.20 0.43 0.69 0.60 
i KAO) 6 ao 6BsOA 4.14 BoM Omia 


It appears from this that the formation of hypersthene in these 
alkali-granites is not dependent on any especially high percentage 
of MgO (or MgO+Fe(O) nor on an especially low percentage of K,O. 

According to my examinations of some Norwegian “white gran- 
ites,” relatively rich in acid plagioclase (according to the nomen- 
clature of V. M. Goldschmidt! ‘“‘Trondhjemite”’) from the north 
of Norway carrying both muscovite and biotite, the muscovite occurs 
exactly in the same way as the biotite. Especially it is to be empha- 
sized that the muscovite-individuals frequently are congested in 
small aggregates—they thus show “together swimming structure”’ 
(synneusis structure) indicating formation at a very early stage— 
and they are in more places deposited on the small apatite-crystals 
which serve as ‘‘Fixkérper.’’ Some individuals show idiomorphous 
contours parallel to oo1, as well as perpendicular to oor, against 
the quartz and the feldspar. Most frequently occur the usual 
lobed outlines, however, as in most of the individuals of biotite 
‘In granite. 

In some of these muscovite-bearing granites, as, for instance, ina 
rock from Narvik—with twice as much muscovite, in leaves up to 
3 mm. in size, as biotite—we observe crystallographically parallel 
growths, at times in alternating strata of the two mica-minerals, 
as described by Rosenbusch (op. cit., p. 57). 

In other samples, however, as in a rock from Fustervand near 
Mosjoen, with nearly twice as much biotite as muscovite—the last 
one in leaves only ca. 1 mm. in size—we observe individuals of 
muscovite with idiomorphic contours inclosed in the biotite, 
indicating that the muscovite at least for a great part was formed at 
an earlier stage than the biotite. 


« “Geologisch-petrographische Studien im Hochgebirge des Siidlichen Norwegens,” 
Ges. d. Wiss., Kristiania (1916). 


670 IMHO ES VOGE 


Rosenbusch (of. cit., p. 51) maintains the view that the forma- 
tion of muscovite in the granite must be explained as a pneumato- 
lytic process' “worauf auch seine weite Verbreitung in den Pegma- 
titen deutet.”” This may perhaps to a certain extent be adequate 
for the muscovite, which belongs to miarolitic druses, but it cannot 
be applied to the common muscovite evenly distributed in the 
granite. The structure proves that the muscovite has crystallized 
from the magma at a very early stage, in part at the same time and 
in part somewhat earlier than the biotite. 

Whether hypersthene, biotite, or biotite and muscovite shall 
crystallize in acid granites with ca. 75 per cent SiO, does not depend 
upon the presence in the magma of those compounds that are shown 
by the quantitative chemical analysis, nor upon certain variations 
with regard to pressure, time of cooling, etc. When to this are 
added the facts that muscovite as a primary formation in igneous 
rocks is limited to granite-pegmatite dikes and to some granites, 
and that muscovite is characterized by relatively much H,O, 
biotite by a smaller percentage of H.O, and hypersthene by no 
HO at all, we must conclude that the decisive factor is the varying 
content in the magma of H,O (eventually also other volatile com- 
pounds). 7 

In the two-mica-granites examined by me, there is up to about 
twice as much muscovite as biotite. If we turn to the granite- 
pegmatite dikes, however, we find at times, though very rarely, as in 
the case of some districts of Smaalenene in Norway, muscovite 
only, without any biotite at all, and the quantity of muscovite 
may here rise to as much as about to per cent. 

The great majority of alkali-granites are characterized by bio- 
tite. Two-mica-granites, in Norway as elsewhere in the world, 
are rare, and hypersthene-granites, so far as yet known, are still 
more rare. 

The content of H,O in the granite-magma thus in most cases 
must have been lying within the interval that gives biotite. As a 


* While H.O dissolved in the magma appears in the same manner as the other 
components, I do not find it natural or right to extend the meaning of the term pneu- 
matolysis to include the formations of minerals, as biotite or muscovite where the 
magmatic dissolved H,O was co-operant. 


MAGMATIC DIFFERENTIATION OF IGNEOUS ROCKS 671 


rare exception only, the quantity of H.O may have been so small,’ 
that hypersthene has been formed (or, principally in somewhat 
more basic granites, a monoclinic pyroxene, besides or instead of 
hypersthene), and relatively seldom only the content of H.O, 
etc., was somewhat higher, so that muscovite was formed together 
with biotite. In granite-pegmatite dikes of analogous chemical 
composition the content of H,O, etc., throughout must have been 
considerably higher, as we here never find hypersthene or diopside, 
but mica and not seldom muscovite together with biotite, occasion- 
ally even muscovite alone without biotite. 

And if we further draw a parallel between the granites (with ca. 
75 per cent SiO,) on the one hand and norites, gabbros, anorthosites, 
etc. (with ca. 50 per cent SiO,), on the other, we will find that mica 
plays a predominant part in the first-mentioned acid rocks, while in 
the last-mentioned basic rocks biotite is much more subordinate 
than the pyroxenes, and muscovite is entirely lacking. 

By three different methods of investigation, viz., in the study of 
the distribution of miarolitic druses, in the study of the magmatic- 
epigenetic formations of ore deposits, etc., connected with igneous 
rocks, and in the study of the relation between mica (biotite, even- 
tually also muscovite), and in part also hornblende, and pyroxene, 
we thus get a confirmation of the theoretically drawn conclusion 
that the granites on an average contain the highest percentage of 
' HO, etc. 

Then we come to the question of how much H,0O, etc., there may 
have been present at the start of the crystallization in the magma 
-and especially in the granite-magma. 

In the muscovite granite-pegmatite dikes there may occur about 
Io per cent muscovite containing 5-8 per cent H.O. In proportion 
to the entire magma there thus entered into the early crystallizing 
mica ca. o.5-0.8 per cent H,O. But still there must have been 
present so much H,O that the later separated minerals might also 
obtain the pegmatitic structure. This quantity of H.O (with CO., 
etc.) escaped, except those small quantities that entered into the 
microscopic pores or was occluded in the separating minerals. This, 
including the content of H,O in the mica, will only make 1 per 
cent by weight of the entire magma. The gas that escaped in the 


672 JL) VOCE 


course of the crystallization, at least in part, will have formed 
miarolitic druses. These play a rather subordinate part in regard 
to the quantity in the granite-pegmatite dikes, however. 

Even in the granite-magma the quantity of HO will hardly have 
amounted to much more than a few per cent. And the crystalliza- 
tion of the igneous rocks must, as pointed out by many earlier 
investigators, among them also myself, be conditioned by the 
decreasing temperature of the magma and not by the diminution 
of volume occasioned by the escaping H,O, etc. 

According to Clarke’s well-known calculation the earth’s crust 
consists to a depth of 10 miles (=ca. 16 km.) at a medium density 
of the rocks of 2.7, of 


93.39 per cent solid crust, with 2.02 per cent H.O 
6.58 a ‘““ ocean, with ca. 97 per cent H,O 


0.03 - ‘* atmosphere 


If we were to presume that the entire quantity of water of the 
ocean was supplied from the igneous magmas, and assume a medium 
thickness from the solid crust of three times 10 miles=ca. 48 (or 
50) km., we would get a quantity of H.O: 


4X6.58X0.97+97.81X2.02=4.06 per cent H,0. 


This value, ca. 4 per cent, must indicate the maximum original 
content of H,O in the initial parent-magmas. But of this again a 
considerable part must have escaped before the parent-magmas 
could have been cooled so far that crystallization-differentiation 
began. The various partial magmas, resulting from magmatic 
differentiation, which, crystallized to form the igneous rocks, must 
thus, on an average, have contained less, very likely quite consider- 
ably less, than 4 per cent H,0. 

In concluding, I wish to note that I have learned much fran Py 
Niggli’s great work: “Die leichtfliichtigen Bestandteile im Magma” 
(1920), but I cannot endorse his statement (pp. 123-38) that the 
volatile compounds have been the important factor in magmatic 
differentiation. This will be more closely treated in a later paper. 


[To be continued] 


THE PLEISTOCENE HISTORY OF THE LOWER 
WISCONSIN RIVER? 


PAUL MacCLINTOCK 
University of Chicago 


INTRODUCTION 


The Wisconsin River, rising among the glacial lakes in the 
northern part of the state of Wisconsin, flows almost due south 
nearly to Portage, where it turns and flowing westward for a dis- 
tance of about 80 miles joins the Mississippi just south of Prairie 
du Chien (Fig. 1). It is this lower, east-west part of the river 
valley which is discussed in this paper. The terminal moraine of 
the Wisconsin glacial epoch crosses the valley just east of Prairie 
du Sac and marks not only the eastern boundary of the region here 
considered but also that of the driftless area. Since glacial drift is 
found in Iowa opposite the lower end of the valley, it may be said 
that the Wisconsin River traverses from east to west the entire 
driftless area. It is thus seen that drift remnants which are found 
in this part of the valley are of important significance in the history 
of ancient ice invasions in bordering regions. 

These remnants of glacial drift in the valley fall naturally into 
two divisions: First, there are terraces of Wisconsin age: (a) 
remnants of the valley train sloping from the terminal moraine, 
_ where it crosses the valley near Prairie du Sac, to the Mississippi 
River, and (5) a lower terrace standing only 15 feet above the pres- 
ent river flood-plain; and second, standing well above the preced- 
ing terrace, are rock benches covered with much older drift. These 
upper benches have a gentle slope toward the east (Fig. 2). 


| PART I. OLDER DRIFT 

Stated in the simplest terms, there are six areas of the older 
drift: (1) near Lone Rock, (2) near Muscoda, (3) from Muscoda 
to Boscobel, (4) from Boydtown to the Kickapoo River, (5) at 


t Condensed from Ph.D. thesis submitted to the Department of Geology, the 
University of Chicago, 1920. 


673 


I ‘OI 


0z61 
PonyPoyy jog 
NIV1d ssoyuo 
ddd 1V1s0d ‘sn 5220121 12p1Q | 
WOus 
VI sony jonny FE 
N3345 ONIYdS dvW asva Ea 
Sai 20urY YUEN AY YS} B 
> = 
STE) SINTON] SV Eee ens 011 upvory yy 0077 


(.-L¥od390gINg 


NN 


Pung pues 


713809s509g 
vaoosnw 


8O0IS739x3 


YSLN59 anvjHo 


PLEISTOCENE HISTORY OF LOWER WISCONSIN RIVER 675 


Wauzeka, (6) at Bridgeport. The first four of these are similar 
in topography, constitution, and amount of weathering, while the 
last two differ from the others in that they contain not only much 
striated material (at Bridgeport) but also a large percentage of 
calcareous material. 
I. SUBDIVISION 

It appears from the evidence that the older drift is not all of the 
same origin or age. 

a) A lithologic study of some 300 characteristic rock specimens 
collected from the different exposures, and examined in the labora- 
tory, shows 37 common to Wauzeka and Bridgeport, 17 common to 


Prairie _du_Chien 
Bridgeport 
Wauzeka 

Blue River 
Muscoda 
Avoca 

Lone Rock 
Mazomanie 
Prairie du Sac 


Ss ssssyem Older Terraces 
—-— High Wisconsin Terrace 


— — Low Wisconsin Terrace 


Fic. 2.—Profile of the Wisconsin River Valley from Prairie du Sac to Prairie 
du Chien showing the bedrock, the drift partly filling the valley, and the levels of 
the three terraces. 


Orion (Port Andrew) and the nearest Illinoian drift at Verona, 
9 miles southwest of Madison, 27 common to Orion and Wauzeka, 
and 25 common to Bridgeport and Iowa (near McGregor). These 
facts show that there is close similarity between the drifts of Illi- 
noian age and that at Orion on the one hand, and between the 
drift at Wauzeka, Bridgeport, and Iowa on the other. 

b) The drift at Wauzeka and Bridgeport contains much lime- 
stone and dolomite, while farther up the valley there is neither. 
Since the Illinoian and pre-Illinoian drifts east of the region contain 
calcareous material, it is likely that these terrace deposits in the 
mid-course of the valley originally contained the carbonates which 
have been subsequently leached. 


676 PAUL MacCLINTOCK 


c) There is well-developed cross-bedding at Blue River dipping 
westward, while at Wauzeka, less well-developed but still recog- 
nizable cross-bedding in sandy layers dips eastward (Fig. 3). 

d) While there are numerous bowlders in the drift of the mid- 
course of the valley, there are more to be seen at Wauzeka and 
Bridgeport. 


ETRE 


Fic, 3.—Westward dipping gravel on the high terrace two miles northeast of 
Blue River. 


e) On the Bridgeport terrace the stones are not only more angu- 
lar than elsewhere in the valley, but numerous subangular and 
striated ones are found. In fact these glaciated stones are as 
numerous on this terrace as in the till either in Iowa or at the 
eastern end of the region. 

The suggestion from this evidence is that the drift in the mid- 
course of the valley is fluvio-glacial and from the east, while that 
at Bridgeport is glacial, and that at Wauzeka is fluvio-glacial and 
both the latter are from the west. 


PLEISTOCENE HISTORY OF LOWER WISCONSIN RIVER 677 


2. MID-COURSE DRIFT 
a) Origin.—The cross-bedding at Blue River makes it evident 
that the drift in the mid-course was brought in from the east 
(Fig. 3). This being the case, the upper surface of these terrace 
deposits should slope conspicuously toward the west, for, while 
large bowlders so common in most of the exposures may have been 
‘carried in bergs, the mass of the material is fluvial and must have 
been transported by a glacial river—a river having powerful current 
and fairly steep gradient. A steep gradient may well be postulated 
for such a glacial river, for there appears, if the bedrock floor of 
the valley be considered, to have been down-warping at the eastern 
end of the region. The bedrock in the valley bottom near Prairie 
du Sac has an altitude of something less than 500 feet,! while at 
Prairie du Chien, near the mouth, its altitude is 490 feet, making 
a gradient of only two inches to the mile (Fig. 2). The preglacial 
river with so low a gradient as this could not have eroded so deep 
and narrow a valley as the rock bottom of this part of the Wis- 
consin Valley appears to be. The unavoidable inference is that the 
eastern end of the region must have been higher in preglacial and 
possibly early glacial times, and must have subsided before the last 
glacial advance. Data from neighboring regions also suggest that 
such warping has taken place.” 
b) Age.—Old drift has been described on the eastern margin 
of the driftless area by Leverett, Weidman, and Alden,’ and called 
by them pre-Illinoian in age. The absence of calcareous material in 
_ the old drift in the mid-course of the Wisconsin Valley, even where 
seen 10 and 12 feet below the surface, suggests that it is as old as 
Kansan and probably older, i.e., the outwash from the first ice 
advance. 
3. WESTERN DRIFT 
a) Origin.—The drift on the Bridgeport terrace must be either 
glacial or fluvio-glacial in origin.. The large number of striated and 


™W.C. Alden, U.S. Geol. Surv. Prof. Paper 106 (1918), Plate II. 


2,W. C. Alden, op. cit.; F. Leverett, Journal of Geology, Vol. III (1895), p. 740; 
E. W. Shaw, Bull. Geol. Soc. Amer., Vol. XXVI (1914), p. 67. 


3F. Leverett, U.S. Geol. Surv. Monograph 38 (1899), pp. 109-10; S. Weidman, 
Wis. Geol. Surv., Bull. 16 (1097), p. 433; W. C. Alden, op. cit., p. 168. 


678 PAUL MacCLINTOCK 


subangular stones found nowhere else in the outwash material of 
the valley, together with the patchy character of the drift, suggests 
deposition directly by the ice. If this view is correct the glacier 
must have extended from Iowa across the Mississippi into the - 
lower end of the Wisconsin Valley. 

On the other hand, at no place where this older drift occurs was 
a glacial pavement seen. ‘The drift lies in most places on deeply 
eroded and weathered dolomite, while at other places several inches 
of blue-black clay, weathered from the bedrock, lies at the base of 
the brown drift. It is not strange that, in exposures so limited, 
no pavement was seen; none has been found in Iowa in this vicinity, 
where the ice is known to have stood to the very edge of the Miss- 
issippi Valley. It seems probable that if a tongue of ice projected 
into the Wisconsin Valley for a distance of 4 miles—a condition 
called for by this hypothesis—it would have been at least as wide as 
the mouth of the valley (13 miles) so that its shoulders would 
have rested against the valley walls near the mouth, and have left 
there glacial material. Some material of this kind is seen for a 
distance of 13 miles north of the lower end of the Wisconsin Valley. 
It is however small in size, meager in quantity, and found not strictly 
on the shoulders but on the lower slopes at heights of never more 
than 100 feet above the flood-plain. Glacial material on slopes so 
steep as the shoulders present would not have remained there but 
would have soon been washed to the flat below. 

The crucial points are: (1) the Bridgeport drift is much higher 
in altitude than any of the older drift farther up the valley, 
(2) it is composed of striated, subangular, and grooved material, 
and (3) it is both stratified and unstratified—the latter material 
indistinguishable from till. The conclusion then is that this drift 
is glacial in origin and was deposited by a tongue of ice. It seems 
clear that the drift at Wauzeka is the outwash material from the 
same ice invasion, for it is closely akin to the Bridgeport drift in 
many ways, has the lense and pocked structure of outwash material, 
and has a suggestion of eastward dipping cross-bedding (Fig. 4). 
None of this calcareous drift is found farther up the valley because 
the decline of this old valley train would have brought its top 
below the level of the rock benches where drift is now found. 


PLEISTOCENE HISTORY OF LOWER WISCONSIN RIVER 679 


This hypothesis involves a damming of the Mississippi River 
by the ice tongue at the mouth of the Wisconsin. Under this 
condition the Mississippi must then have flowed between this ice 
tongue and the northern wall of the Wisconsin Valley and thence 
eastward to the Rock River or more probably the Lake Michigan 
Basin. Such an eastward flowing river, if the bedrock divide 
near Portage had about the present altitude of 600 feet, would have 


Fic. 4.—View of the outwash material on the high terrace at Wauzeka showing 
lenses of sand in the gravel. 


had a gradient of about 2.5 feet per mile, or a foot per mile greater 
than that of the present Wisconsin River. This suggests that the 
down-warping of the eastern part of the area, mentioned above, had 
already taken place. It seems, in fact, quite probable that this 
eastern part of the region was lower than it is at present, for post- 
Champlainic uplift and warping in the Great Lakes area probably 
raised the divide from some lower elevation to its present altitude. 
Since such down-warping as first mentioned has been found in 


™W. C. Alden, op. cit., Plate II. 


680 : PAUL MacCLINTOCK 


neighboring regions, it seems well to note it here, and by so doing 
possibly to fix the date of the warping—after the first ice invasion 
at the east of the area and before the second invasion on the west. 

Additional evidence of such a displacement of the Mississippi 
might be expected in old channels. In the eastern part of the 
valley, Wisconsin glaciation has destroyed any possible trace, 
while at the west the river was either displaced for so short a time, 
or subsequent erosion has been so great that there is no evidence 
of a channel occupied during the displacement. No channel is found 
in Jo Daviess County, Illinois, where a similar tongue of ice pushed 
across the Mississippi Valley from Iowa and left drift near Hanover.* 

b) Age.—Two drift sheets, the Kansan and the pre-Kansan, 
are thought to be present in Iowa near the mouth of the Wiscon- 
sin River.2. The western drift in the Wisconsin Valley must be 
correlated in age with one of these. The former drift sheet is less 
weathered than the latter, which is represented, according to present 
determinations, only by scattered and very much weathered errat- 
ics. Judging from the thickness (20 to 50 feet), and from the large 
content of limestone and dolomite, it seems most probable that 
the Bridgeport and Wauzeka drift is of Kansan age. 


PART II. WISCONSIN DRIFT 


1. The terminal moraine of the last glacial invasion extends 
southward from the Baraboo Range, crossing the Wisconsin River 
1+ miles northeast of Prairie du Sac. On the south side of the 
river it maintains a southerly direction to Black Earth Creek which 
it crosses 1+ miles east of Cross Plains. North of the river this 
moraine is a belt showing morainic topography, while south of the 
river it is in most places a narrow distinct ridge strewn with 
bowlders. Where it crosses the river the section shows 60 feet of 
cross-bedded sand with small lenses of fine gravel, overlain, with 
a sharp contact, by 30 feet of till. The sharp contact shows no 
weathering. This sand may be outwash from an early Wisconsin 
moraine farther east or may be the outwash deposited in front of 
the advancing late Wisconsin ice. 


tE. W. Shaw and A. C. Trowbridge, Jll. State Geol. Surv. Bull. 26 (1916), p. 87. 
2 A. C. Trowbridge, Bull. Geol. Soc. America, Vol. XXVI (1914), p. 76. 


PLEISTOCENE HISTORY OF LOWER WISCONSIN RIVER 681 


2. Sloping gently westward from the moraine north of the river 
a sandy outwash plain extends to an irregular boundary against 
the sandstone hills of the country rock. The western edge is irreg- 
ular, for the fluvio-glacial material is found up the valleys of Honey 
and Otter creeks. Of special interest are the erratics found in 
the south branch of Honey Creek as far west as Blackhawk and 
Plain. They lie on an upper terrace, corresponding in elevation 
to that of the outwash plain across the mouth of the creek at its 
eastern end. This position, 17 miles beyond the terminal moraine, 
implies that the bowlders were carried to their positions while 
frozen in blocks of ice floating on a lake. 

Such a lake may have been formed in one of two ways: 

a) The edge of the ice may have extended beyond the terminal 
moraine and dammed the mouth of Honey Creek. No evidence 
was found to substantiate this possibility. 

6) As the outwash plain was being built, the glacial waters 
issuing from the ice-front between Prairie du Sac and the South 
Range swept their load southward across the mouth of Honey 
Creek. Outwash material may thus have dammed the mouth of 
Honey Creek, forming a lake upon which icebergs may have floated 
the bowlders. While this suggestion involves the difficulty of 
getting the bergs swept across the outwash plain and into the lake, 
it still seems the more probable of the two. 

3. The valley train, now represented by terrace remnants, once 
filled the bottom of the valley from the terminal moraine to the 
Mississippi River. It was go feet above the present flood-plain 
‘near the terminal moraine, 30 feet in mid-course, and 4o feet at 
the western end of the valley. As this outwash deposit was grow- 
ing, the glacial waters constantly deposited material across the 
mouths of the tributary valleys, causing them in turn to aggrade 
their channels. Terrace remnants of these slack-water deposits are 
to be seen in most of the tributary valleys, serving to project the 
level of the valley train even where it has been removed from the 
main valley by subsequent erosion. 

The most easterly remnant of the original valley train lies near 
Mazomanie at the mouth of Black Earth Creek. It is an irregular 
area a mile wide by 6 miles long, separated from the south bluffs 


682 PAUL MacCLINTOCK 


by Black Earth and Halfway Prairie creeks. The surface of this 
area is gently rolling and marked here and there by patches of low 
sand dunes. This terrace level extends eastward into Black Earth 
and Halfway Prairie valleys while the two intervening shorter 
valleys have this high fill only at their very mouths. This relation- 
ship is of importance in connection with the problem, later to be 
considered, of the age of the terrace. 

From Mazomanie for a distance of 30 miles to the west, the 
upper part of the valley train has been entirely removed from the 
main valley. The terrace level is, however, present in most of the 
tributary valleys; notably Blue Mounds, Wyoming, Otter, Pine, 
Eagle, and Kickapoo creeks. But there are several tributary 
valleys (see Fig. 1) lacking this terrace level, a fact whose signifi- 
cance is later to be considered. 

The remnants of this level, the high Wisconsin terrace, are 
again found in the main valley near Muscoda and Blue River where 
the bench is protected by a subjacent ledge of sandstone against 
which the river is at several places flowing. At Boscobel a large 
terrace remnant lies against the south wall of the valley. In these 
latter terrace patches the material is smaller in size and contains 
fewer limestone pebbles than farther up the valley. 

4. Twelve to 15 feet above the Wisconsin River flood-plain and 
extending short distances up the valleys of many of its tributaries 
there is an extensive sandy terrace—the low Wisconsin terrace. 
From the terminal moraine at Prairie du Sac to Wauzeka, a distance 
of 65 to 70 miles, it is nearly continuous in the main valley on one 
side of the river or the other, while from Wauzeka to the Mississippi 
it occurs only in small detached areas. Remarkable uniformity in 
height above the river is one of its most notable characteristics, for 
the variation is not more than a foot or two throughout the whole 
distance. A second notable feature is that the material, where 
seen in shallow cuts, is uniform in size and constitution through 
the whole length of the valley, being mostly sand with small pebbles 
scattered rather uniformly through the mass. ‘The surface of this 
terrace is in general very flat, but in detail it is seen to have irregu- 
larities produced by the wind, such as sand dunes and “blow- 
holes.” Considerable dune areas are found in the neighborhood 
of Lone Rock and Spring Green. In fact, the whole terrace is so 


PLEISTOCENE HISTORY OF LOWER WISCONSIN RIVER 683 


sandy and so poor as farm land that it is called locally “Prairie” 
or ‘‘ Barrens.” 

There are three possibilities to be considered in discussing the 
origin of the low terrace: It is either the valley train of the late 
Wisconsin ice advance or was cut from the early Wisconsin valley 
train by waters from a glacial lake, or is a combination of the two. 


; lance tee ee renee 
yen tane He () 
“Terminal ZEA 


° 
IE ue J S 3 

ENO ae 
32 Wah Wisconsm Iereace ~~ oe ; 
KES : EE] how Wiscensin Terrace Re y ; : 

B Late Wisconsin lerminal Moraine = 2 \C7o Te S oe BS 
Jz = z 
tC} 


Fic. 5—Diagram showing the relation of the terraces in the four valleys east of 


Mazomanie. Black Earth and Halfway Prairie creeks contain the high terrace while 
. the two shorter valleys do not. 


- @) Alden is of the opinion that the low terrace is the result of 
deposition by glacial waters from the late Wisconsin invasion, while 
the upper terrace resulted from the early Wisconsin advance.t' The 
evidence is as follows: Of the four small valleys east of Mazomanie, 
the two longer ones contain the upper terrace while the shorter 
ones do not.2 This, Alden interprets as meaning that the ice of 
the early Wisconsin invasion did not reach the heads of the shorter 
valleys, discharging its waters only through the longer ones and 
building in them the high terrace (Fig. 5). 

™W. C. Alden, of. cit., pp. 191-93 and 244-45. 


2 The northern one does contain several small patches (see Fig. 5). 


684 PAUL MacCLINTOCK 


Following the retreat of the early Wisconsin ice came a period 
of erosion during which part of the fill was cut away. The late 
Wisconsin ice advanced farther and stood across the heads of all 
four valleys, building valley trains at the level of the low terrace. 
At the same time the low terrace was being built in the main valley. 

Evidence adverse to this suggestion must be summarized under 
several heads: 

z) If deposited by glacial waters, the terrace should decline 
westward more rapidly than the present river, unless the western 
end had been raised by postglacial tilting. But, judging from the 
tilting of the glacial beaches of the Great Lakes, this latter possibility 
is unlikely. 

it) If deposited by glacial waters the material should be notice- 
ably coarser at the eastern end grading to fine at the west. This 
is not the case. 

uit) There are three conspicuous examples of small valleys 
containing only the low terrace, west of the farthest ice advance 
(just west of Black Earth, south of Arena, and west of Avoca), 
while there are but two such cases among the valleys heading in 
the terminal moraine, as previously cited. It would seem that 
the mere fact that two of the four valleys east of Mazomanie do 
not have the high terrace is not conclusive one way or the other. 
In fact, it would be just as plausible to suppose that there was but 
one Wisconsin advance in this region and that, since the longer 
valleys drained the ice both earlier and later than did the shorter 
ones, they received more outwash, and so were more aggraded. 
The thin terminal moraine crossing the valleys means a short stay 
of the ice-edge at this place, or poverty of débris in the glacier. 

iv) No evidence was seen of weathering of the stratified drift 
underlying the till of the terminal moraine, as would be expected 
somewhere in the region if it were early Wisconsin and the till were 
late Wisconsin, since this interglacial interval is considered by many 
to be fairly long. The stratified drift may as well be outwash depos- 
ited by waters which flowed out in advance of the oncoming ice. 

v) The evidence from the larger amount of leaching of the out- 
wash plain, suggested by Alden and Weidman," as showing that 

tW. C. Alden, op. cit., p. 192. 


PLEISTOCENE HISTORY OF LOWER WISCONSIN RIVER 685 


the high terrace is older than the terminal moraine and outwash of 
the low terrace, was not verified. For cuts on and directly west of 
the moraine show about the same amount of leaching as do the 
exposures farther west on the high fill. 

vt) The high terrace marks a time of great filling, while the 
low one is much less important in this respect. Evidence from 
other regions has led to the generalization that the ice of the late 
Wisconsin substage was the most energetic of all the ice advances, 
building higher and more rugged moraines; eroding more deeply 
and more conspicuously; dumping more sediment into the drainage 
lines leading away from the ice-front, and so building larger valley 
trains. This line of evidence would point rather to the late Wisconsin 
than the early Wisconsin substage as the builder of the high terrace. 

vii) Since erratics in Honey Creek Valley rest only on the high 
Wisconsin terrace, it seems clear that a glacial lake stood in this 
valley during at least part of the time when the slack-water fill of 
which these terraces are remnants was being deposited. It would be 
inferred that the lake was dammed during the maximum extent of 
the ice, rather than when the ice-edge stood farther east, as it did 
in early Wisconsin time if the early Wisconsin ice affected this 
immediate region. ‘This piece of evidence suggests that the high 
Wisconsin fill in Honey Creek Valley was deposited when the ice- 
front stood at least as far west as Prairie du Sac. 

The weight of this evidence is seriously against the possibility 
that the low terrace was deposited as a valley train of the late 
Wisconsin invasion. 

b) When the ice-front of the Green Bay lobe had withdrawn east 
of the Portage divide, the ice-dammed lake, Jean Nicolet, was 
formed with its outlet down the Wisconsin Valley. These outlet 
waters were clear, and probably cut the upper part of the valley 
train down to the level of the low terrace. Evidence that the low 
terrace was cut by waters from Lake Jean Nicolet follows. 

4) The uniform height, 12 to 15 feet, of the terrace above the 
flood-plain all the way from the terminal moraine to the mouth of 
the valley, suggests strongly an erosional rather than a depositional 
origin. 

tW. Upham, Amer. Geologist, Vol. XXXII (1903), p- 330. 


686 PAUL MacCLINTOCK 


ii) The material, in at least the upper few feet of the terrace, 
throughout the length of the valley is uniform in size, shape, and 
structure. The river having a uniform gradient would handle 
sediment of uniform size through its whole length. This would 
result in the coarser material in the eastern part of the high terrace, 
when cut by these outlet waters, being buried below several feet 
of finer re-worked material covering the low terrace. 

iit) Its similarity to the Brule—-St. Croix outlet of Lake Duluth 
is noticeable. This latter is also a broad sandy plain with dunes 
and blowholes upon its surface." 

wv) The low gradient, 1.75 feet per mile, for so large a volume of 
water, would favor a wide rather than a deep cut. 

Against this mode of origin the following points may be 
registered: 

z) It would be expected that the upper few feet of the terrace 
would be re-worked by the running water and the material therefor 
assorted. But this is, as a rule, not the case, for the pebbles are 
scattered indiscriminately through the sand. 

it) Weidman states, in relation to the Brule-St. Croix outlet 
that ‘‘. . . . Aside from cutting down a few drift dams that lay 
across the outlet, there was not much erosion.’ It is possible, 
then, that there was not enough cutting by the waters of Lake 
Jean Nicolet to cut the upper terrace to the level of the lower one. 
However, the rapidity of cutting, depending upon the volume and 
the velocity of the river as well as the kind of material cut, may 
not have been the same in the two cases. So the slight amount of 
cutting of the Brule-St. Croix outlet would not carry a necessary 
implication against great cutting in the Wisconsin Valley. | 

iii) The relation of the terraces in the four tributary valleys 
just east of Mazomanie, previously discussed, is significant but not 
conclusive. 

From the weight of this evidence, the low terrace appears to be 
a degradational level cut by waters from the glacial lake. 

c) A third suggestion presents itself which combines the first 
and second in such a way as to obviate many of the difficulties 

* Moses Strong, Geology of Wisconsin, Vol. III (1880), p. 387. 

2 Samuel Weidman, personal communication. 


PLEISTOCENE HISTORY OF LOWER WISCONSIN RIVER 687 


inhering in each. With the advance of the early Wisconsin stage, 
the outwash valley train was deposited. During the subsequent 
period of ice withdrawal, the pounded waters of the Fox River 
Valley flowed across the Portage divide and down the Wisconsin 
Valley, cutting away a large part of the valley train. This period 
must have been rather long, or erosion excessively rapid by a large 
and powerful river, for more erosion took place then than has taken 
place since the last withdrawal of the ice. This would not appear 
to be improbable, for, during this partial withdrawal, the ice may 
have dammed the lake for a much longer period of time than it did 
in the final deglaciation. Then when the late Wisconsin ice 
advanced to the region of Prairie du Sac, the outwash partially 
filled the channel cut below the early fill. Later, as the ice with- 
drew, the lake was again dammed east of the Portage divide 
and the waters flowed westward down the Wisconsin Valley, cutting 
the lower fill to the level of the low terrace. This would involve 
less cutting at any one stage than the first suggestion, and at the 
same time would allow the terrace to stand, as it does, at a con- 
stant elevation above the present river level, for the waters from 
the lake probably would cut to the same gradient as do those of 
the present Wisconsin River. 

While the hypothesis of two Wisconsin advances will explain 
the presence of the high terrace in two of the valleys east of Mazom- 
anie and its absence from the other two, it will not account for 
the absence of this high terrace in the tributaries farther down the 
Wisconsin River. And since it is evident that the former case can 
be explained on the basis of one advance into this region, the idea 
of two ice invasions in Wisconsin time may be discarded as needless. 

SUMMARY 

The terraces of Wisconsin age may be best explained on the 
hypothesis that they are connected with one glacial advance—that 
of the late Wisconsin ice-sheet—and that the lower terrace was cut 
from the higher by waters issuing from Lake Jean Nicolet. 


PART III]. THE PLEISTOCENE HISTORY 


The first glacial invasion in Pleistocene time advanced on the 
eastern side of the region to a position somewhat east of the Wiscon- 


688 PAUL MacCLINTOCK 


sin terminal moraine. The eastern end of the region at this time 
stood relatively higher than it does now, so that the glacial waters 
flowing down the Wisconsin Valley had a steep gradient and trans- 
ported coarse débris. The glacial drainage from at least a hundred 
miles of ice-front to the north must have flowed southward to the 
vicinity of Portage, and then westward down the Wisconsin River. 
In its course, along the ice margin, the river must have cut against 
the edge of the ice, at least in places, and must have broken off 
blocks of débris-laden ice, floating them into the Wisconsin Valley. 
Here many of them must have grounded and, upon melting, have 
deposited their loads. The adequate source of bergs, the abundant 
supply of glacial material, and the swift and powerful glacial river 
seem sufficient to account for the older drift deposited on the terraces 
of the mid-course of the valley. After the ice had stood long enough 
to build a valley train to a height of at least 75 feet above the present 
flood-plain in the mid-course of the valley, it withdrew and erosion 
cut away the valley train till all that remained were the terrace 
remnants on rock benches along the sides of the valley. It is not 
known how deeply this erosion progressed, but probably the valley 
was largely re-excavated. ; 

At some time after this first valley train was built and before 
the next glacial advance, the eastern end of the region was depressed 
relative to the western end. A depression of 150 to 250 feet would 
not have been unlikely and would account for the phenomena 
observed. 

The Kansan ice advanced across Iowa, crossed the Mississippi 
River in the neighborhood of Prairie du Chien, and projected a 
tongue of ice into the lower end of the Wisconsin Valley. The 
Mississippi was dammed, diverted into the Wisconsin Valley, and 
flowed eastward, carrying with it not only great quantities of coarse 
and fine outwash material, but abundant icebergs broken from the 
ice-front farther north as it encroached upon the Mississippi Valley. 
An eastward sloping valley train of coarse material was built. 
When the ice withdrew, erosion cut away the moraine and valley 
train, save where remnants are left on rock benches at Bridgeport 
and Wauzeka. The depth of this erosion is not known accurately, 


PLEISTOCENE HISTORY OF LOWER WISCONSIN RIVER 689 


but the valley was probably again re-excavated to about its maxi- 
mum depth. 

There are no terraces in the valley which correspond in age to 
the Illinoian drift found at the eastern end of the region, so the 
assumption is that the outwash from this glacial advance did not 
fill the valley high enough to be above the present surface of the 
river. The evidence of five well records' shows that at one time the 
valley floor stood, in the main and also the tributary valleys, 30 
to 50 feet below its present Jevel long enough to accumulate a bed 
of peat. It is probable that the outwash from the Illinoian invasion 
filled the valley only to this level, 30 to 50 feet below the present 
surface. Then ensued a period during which the vegetation accu- 
mulated on the swampy surface of this outwash. 

During the Iowan epoch loess was blown on to the western part 
of the area, burying the drift with a blanket of eolian material. 

In Wisconsin time not only the moraine at Prairie du Sac, but 
the valley train in the Wisconsin Valley, was deposited. As the 
ice withdrew east of the divide near Portage, ponding produced a 
lake which drained westward down the drift-filled Wisconsin 
Valley. This was for a time the main drainage for at least a hundred 
miles of ice-front lying toward the north, consequently a large quan- 
tity of clear water flowed down the valley. The upper part of the 
fill was largely cut away, leaving remnants which now constitute 
the upper terrace in the Wisconsin Valley. The down-cutting 
river reached grade at the level of the top of the lower terrace. 

After the ice had withdrawn and the glacial lake was drained, 
| the postglacial Wisconsin River cut away large parts of the Loe 
terrace to form its present flood-plain. 

1 Well records which show peat 30 to 50 feet below the surface; fair grounds at 
Richland Center; schoolhouse 13 miles northeast Richland Center; Bear Creek ~ mile 


north of junction with Little Bear Creek; Little Bear Creek 2 mile north of junction 
with Bear Creek { mile southwest of Leland. 


ORIGIN OF THE TRIASSIC TROUGH OF 
CONNECTICUT 


WILBUR G. FOYE 
Wesleyan University, Middletown, Connecticut 


The paper by Professor W. M. Davis on “The Triassic Forma- 
tion of Connecticut’’* has long been regarded as a masterpiece in 
geologic literature. So well was the work done that little has been 
added to the knowledge of the Newark formation in Connecticut 
since its publication. Professor Davis did not, however, reach a 
definite conclusion concerning the origin of the trough within which 
the Newark sediments collected. ‘Two hypotheses have been most 
widely held to explain the origin of the depression. It is the pur- 
pose of this paper to state briefly the field facts which bear upon 
these hypotheses and to suggest a method of research which may 
aid in the solution of the problem. 

The two hypotheses referred to are: (1) the depression was 
formed by a gradual bending downward of a canoe-shaped trough 
without faulting movements (Fig. 1), and (2) the depression was 
developed by faulting movements on each side of the depression 
sedimentation (Fig. 2). Both of these hypotheses were suggested 
by Professor Davis in his report.?, The fundamental hypothesis 
may be modified by certain limiting conditions. Professor Davis 
was inclined to believe that the formation of the trough was not 
accompanied by faulting. He also held that the original area 
covered by the Newark deposit was not much greater than that over 
which the series outcrops today. Professor Grabau, while agreeing 
with Professor Davis in part, believes that a vast geosynclinal 
wedge extended from the eastern folds of the Old Appalachian 
Mountains, and that the present areas “‘are mere erosion remnants 


« Fighteenth Annual Report, U.S. Geol. Surv. (1897), Part II, pp. 1-192. 
2 Ibid., pp. 37-38. 3 [bid., p. 19l. 
690 


ORIGIN OF THE TRIASSIC TROUGH OF CONNECTICUT 691 


of once much more extensive deposits . . . . preserved by being 
faulted beneath the level of erosion.’ 

Again, Professor Davis suggests that the trough may have been 
developed by faulting movements on each side of the depression 
(Fig. 2), whereas Professor Barrell? has limited the faulting to the 
eastern border (Fig. 3). 

It is believed that there is general agreement with Professor Davis’ 
statement that the “ancient mountains of Western Upland must 


 _— =< Vy 
a 


LA 


Fic. 1.—Diagram representing a depression formed by a gradual bending down- 
ward of a canoe-shaped trough, without faulting. 


Fic. 2.—Diagram representing a a developed by faulting which was 
continuous during the period of sedimentation. 


have been worn down to a peneplain, or at least reduced to hills 
of moderate elevation and gentle slope, at the time the accumu- 
lation of the sandstones began.’’3 It is further agreed that “‘the 
basement on which the Triassic strata rest”’ was worn “‘so low that 
no great additicnal amount of waste could be worn from it” had 
there not been depression of a central area accompanied by “‘corre- 
lated elevation of the adjoining areas on the west and east.’’4 
Professor Davis goes on to say that “‘two suppositions may be 
made as to the character of these correlated elevations. The 


tA, W. Grabau, Text Book of Geology, Vol. II, pp. 612-13. 


2 Joseph Barrell, ‘‘Central Connecticut in the Geologic Past,” Bulletin No. 23, 
Conn. State Geol. and Nat. Hist. Survey. 


3 Eighteenth Annual Report, U.S. Geol. Surv., Vol. I, p. 25. 
4 [bid., pp. 37-38. 


692 WILBUR G. FOYE 


trough may have been bent down between two arched areas on 
either side, as in Figure 1, or the trough may have been faulted down 
between two uplifted blocks alongside of it, as in Figure 2. While 
it did not seem advisable to make a final choice between the alterna- 
tives, the conditions illustrated by Figure 1 were favored, chiefly 
because the centripetal dips there shown would give, after general 
eastward tilting with more or less faulting, moderate dips for the 
lower strata in the east and stronger dips for the same strata in the 


+ 
+. 
“Apt aN 


¥o 
++ 


Al 
= Me | 
ahs 
ist 
* > Ih &/) 
os = 
ONS 


WT 
v= 


. 
s 
10 


Fic. 4.—(After Barrell) 


pa 


SS=] Triassic Sediments and lavas. >A Paleozoic Sediments. 

Paleozoic intrusive granite-gneisses. DH Pre-Paleozoic complex gneisses. 
——— ed 

0. Scale in miles, horizontal and vertical. 10. A-A Depth reached by later cycles of erosion. 


west. Professor Davis states that the field evidence showed an 
average dip to the east for the basal beds of 20° to 30° along the 
western border, and seldom more than 20° to the east for the 
analogous beds where exposed along the eastern.’ His section, so 
widely copied in textbooks (Fig. 5), is therefore based on Fig. 1. 
Professor Barrell, in his well-known study of ‘‘ Central Connecti- 
cut in the Geologic Past,” gives his conception of the origin of the 
depression. His idea is illustrated by Figures 3 and 4. A marginal 
fault of gradual development along the east side of the Connecticut 


t Fighteenth Annual Report, U.S. Geol. Surv., Vol. II, p. 39. 


ORIGIN OF THE TRIASSIC TROUGH OF CONNECTICUT 693 


“Mandngy p0I1G0j0aH “SQ fo jaoday jonuupy yyuaaysrg ut omn8y staeq Jaye ‘Kaedng C40jstzT oanjoNT Puy 409180024 ayojG ynayoouuoyD ‘QO ‘ON 
UYayng WOIY (‘WOl}edg1}7"1}s Jo souvid yuasordal Jou soop sour jayyered YIM SuIpeys 9YL) “SuUNMeT pue Surypy oy} s10Joq UoTyeUAIO; IT} 
jo 9pnztj72 pasoddns ay} smoys UOrT}D9S IOMOT OYJ, *po10}SeI ov UOIsOIa Aq PoAOUIOI Useq sARY YIM sjivd oy, ‘poyney pue poz} 
SB UOlJeUIOJ 9Y} JO 9pnziz}e jussord oy} smoys uoT}oes Joddn oyy, ‘oIssetay, Ao[eA INIIoUUO0D IY} sso1oe suOT}ZIBgG—'S ‘OI 


12120'sy20y Sosy 


EUTEy Pease ‘sy20yy vary porn yous 2/s°04 P24OjSA4 DJOUO 21SSO/4L Wt We auyyposkay 4 P2syses-sy20y auyjeyskig 
CE 


=i QUOT 
SS — 


WAS 


634 
72,644, 

“ 
ALO 


~ 


4, 
CALL 
43th, 


694 WILBUR G. FOYE 


depression tilted the accumulating sediments toward the east and 
quickened the streams. Later smaller faults broke the trap flows 
and initiated the present topography (Fig. 4).! Professor Barrell 
supported his conception by the one statement that ‘“‘the dominant 
segregation of conglomerates near the eastern margin is even more 
marked in the beds above the lava flows than in those below, and 
this greater average coarseness of the upper sediments indicates 
the intermittent regrowth of the mountains whose perennial waste 
kept supplying material for the basin.’ 

For reasons which will now be briefly stated the writer believes 
that Barrell’s diagram represents most accurately the structure of 
the Connecticut depression during the Triassic. 

1. A warping movement that distorts a peneplain surface with- 
out faulting must proceed very slowly. It is difficult to imagine 
that such a movement would revive the streams flowing into the 
Connecticut basin and cause them to transport bowlders of such a 
size as may be found not only in the edge but also toward the center 
of the trough. Cobbles 6 or 8 inches in diameter are found near 
the center of the valley, north of Meriden.’ 

2. Arkoses are common along the western border of the basin 
but are almost lacking along the eastern border, whereas coarse 
conglomerates are common along the eastern border but are seldom 
found along the western border. ‘The inference is that the streams 
from the west were carrying the exfoliation products of a desert 
topography, but those from the east were carrying bowlders snatched 
from the wall of a growing fault scarp. 

3. A consideration of the geometry of the geosynclinal hypothesis 
of Davis and the fault-monoclinal hypothesis of Barrell leads to 
conclusions which are more favorable to the latter. In Figures 6 
and 7, let W represent the width of the Connecticut Valley. In 
northern Connecticut this width is approximately 21 miles; at 
Middletown it is 17 miles, or, if the Pomperaug Valley area is 
included within the larger basin, the width becomes 33 miles. Let 
D represent the depth of sedimentation within the basin. Compe- 

t Bulletin No. 23, Conn. State Geol. and Nat. Hist. Survey, p. 28. 

aTbid., p 29. 

3 Eighteenth Annual Report, U.S. Geol. Surv., Part II, p. 33« 


ORIGIN OF THE TRIASSIC TROUGH OF CONNECTICUT 695 


tent opinion places this thickness between 12,000 and 13,000 feet 
or between 2% and 2} miles. Let a represent the angle of dip 
developed in the basal beds by the gradual depression of the trough. 
At the sides of the geosnycline the actual angle would be greater 
than that indicated whereas, at the center, the beds would be flat, 
since the basin would be concave (Fig. 5). In either Case I (Fig. 
6) or Case II (Fig. 7) it should be noted that the beds laid down at 


——_——_—___—_———"_- Tw LYV/\V_ ——— 


a 
D 
Fic. 6.— 

Width Depth a 
17 24 143 
17 2s 164 
21 24 12 
21 2 134 
33 2% 7% 
33 23 83 


Fic. 7.— 
Width Depth a 
17 2% 12 
17 23 8% 
21 2t 64 
21 23 7 


the close of the period of sedimentation were approxiamtely hori- 
zontal, a fact overlooked by some writers. The diagrams are drawn 
to scale for a width of 17 miles and depths of 2{ and 25 miles, 
respectively. 

Whether the geosynclinal or the fault-monoclinal hypothesis be 
accepted, the present dips of the sediments of the basin were pro- 
duced at the time of the post-Triassic faulting movements which 
tilted the rocks to the east. It would be advantageous to compare 


696 WILBUR G. FOYE 


the dips of the beds at the same horizon on the two sides of the 
trough, but only basal beds are exposed on the western side of the 
valley and upper beds on the eastern side. It is, however, possible 
to compare the known dips of the upper beds to the east with the 
dips of the basal beds to the west. Allowing for an erosion of 500 
feet, the beds at the eastern side of the valley cannot lie far below 
the top of the series. Their original attitude at the close of the 
Triassic sedimentation was approximately horizontal. In the case 
of the geosynclinal hypothesis there may have been a slight dip 
to the west; in the case of the fault-monoclinal hypothesis, possibly 
a slight dip to the east. At present they have an average dip of 
15° to 20° to the east. It is difficult to believe that the post- 
Triassic tilting was very dissimilar on the two sides of the valley. 
If it is assumed that the present dip of the upper beds was developed 
entirely at the time of the post-Triassic faulting, and that the tilt 
was, therefore, approximately 20° to the east, then, by adding 20° 
to the angle a, the present angle of dip of the basal beds at the 
western side of the valley should be obtained. Davis (see above) . 
states that the average dip of the basal beds on the western side of 
the valley is 20° to 30° In Case I (Fig. 6) by the foregoing method, 
the dips should be from 30° to 40°. In Case II (Fig. 7), angles from 
20° to 30° are to be expected. It is true that, if the Pomperaug 
Valley area is included within the main basin, the results are incon- 
clusive, but few authorities believe the original basin was much 
larger than it is today. 

Assuming that the present dip of the upper beds of the Newark 
series at the eastern side of the valley represents the approximate 
angle of tilt to the east developed at the time of the post-Tniassic 
faulting, then referring to Davis’ diagram (Fig. 5), and conceiving 
that the present width of the valley was its approximate width in 
the Triassic period, in Case I the present dips should vary only 
slightly from the east toward the center of the valley, averaging 
perhaps a little lower about midway between the two points, but 
they should rise to a maximum at the western border. No normal 
dips near the center should be greater than the dips at the eastern 
border. In Case II there should be a progressive increase in dip 


ORIGIN OF THE TRIASSIC TROUGH OF CONNECTICUT 697 


from the eastern to the western side of the valley: The data 
concerning the strikes and dips within the Connecticut trough 
have never been assembled. Professor Davis’ general statement 
quoted above would lead one to suppose, however, that there is 
an increase in the dips from the eastern toward the western side 
of the depression. 

4. Pebbles in the upper conglomerates at the eastern border of 
the basin are known to be similar to rocks exposed at the very edge 
of the eastern upland. This condition is especially true near Lake 
Quonnipaug in Durham, Connecticut. East of the lake a chlorite 
schist, which is not common within the metamorphic rocks of the 
upland, outcrops for a mile or two. West of the lake the coarse 
‘“fan-glomerates” are filled with pebbles of this rock. The evi- 
dence indicates that the eastern limit of the Newark formation is 
at its ancient boundary, as Barrell’s hypothesis would postulate, 
and that the basin sediments never extended over the eastern up- 
land. If the eastern and marginal fault developed after the period 
of sedimentation, the chlorite schist at Lake Quonnipaug could 
not have been exposed to erosion at the time the ‘‘fan-glomerates”’ 
were being deposited. 

5. The abundant development of those rocks so aptly named 
‘“‘fan-glomerates” by the western geologists along the eastern 
side of the basin is, in itself, strong evidence of the early initiation 
of faulting movements along this boundary. Such conglomerates 
are common throughout the exposed thickness of the Totoket block, 
but are not known on the western side of the valley.” 

_ 6. Finally, there is good evidence of the localization of vulcanism 
along the eastern fault line long before the end of the period of 
deposition within the basin. The writer has recently discovered a 
volcanic neck, in the southern part of Durham, north of Totoket 
Mountain, which lies within a stone’s throw of the eastern fault 


margin. 


t Excessive dips to the east are known near the eastern boundary fault. They 
are in the opposite direction from the drag dips which one would expect in this 
vicinity and have not been explained. 


2 Cf. C. R. Longwell, Amer. Jour. Sci., IV (1922), 234-35. 


698 WILBUR G. FOYE 


For the reasons stated, the writer conceives of the Triassic 
basins of eastern Canada and the United States as a series of troughs 
of the basin range type which were developed during the collapse 
of the ancient land of Appalachia after the Appalachian mountain- 
building episode. 

The Vale of Eden at the western base of the Pennine escarpment 
in northern England offers an interesting parallel to the inferred 
structure of the Connecticut Valley., Kendall has described the 
geology of the vale as follows: 


The succession in the Vale of Eden is of particular interest from the evi- 
dence that it furnishes of the physical conditions of the period and their changes. 
The valley is bounded on the east by the Pennine escarpment which owes its 
existence to a tremendous series of faults truncating the Permian and later 
rocks. The succession from west to east is: Carboniferous Limestone and 
Millstone Grit, covered unconformably by massive calcareous conglomerates, 
“Lower Brockram,” usually dolomitized; bright red Penrith Sandstone about 
300 m. (1,000 ft.); “Upper Brockram” interbedded in the upper part of the 
Penrith sandstone; Hilton Plant Beds with Noeggerathia, 45m. (150 ft.); 
Magnesian Limestone o-6 m. (o~20 ft.); Marls with gypsum having, locally, a 
basal conglomerate, 90 m. (300 ft.); St. Bees Sandstone (Trias) 600 m. (200 ft.) 


The materials of the Lower and Upper Brockrams respectively furnish 
evidence of contemporaneous movement of the adjacent fault zone. The 
Lower Brockram consists exclusively of fragments of Carboniferous Limestone 
and the writer (Kendall) infers that it represents gravel-fans washed by torren- 
tial rains from the uplifted fault country, when the displacements had exposed 
only that division of the Carboniferous series. The Upper Brockrams were 
laid down after the deposition of 300 m. (1,000 ft.) of Penrith Sandstone, which 
should have covered up an equivalent portion of the faulted area, yet these 
Brockrams consist in large measure of the Basement Conglomerate of the 
Carboniferous series, with occasional pebbles of the underlying Ordovician 
rocks. This is interpreted to mean that between the formation of the two 
Brockrams a great further movement of the faults took place bringing the 
base of the Carboniferous up within the action of surface erosion. 


The eastern upland of Connecticut consists of such a tangle of 
metamorphic rocks that the rock succession is difficult to interpret. 


t “The British Isles,’ Handbuch der Regionalen Geologie, Band III, Abteilung 1, 
p. 188. The writer is indebted to Professor Fearnsides, of Sheffield University, for 
calling his attention to the parallelism here described. ‘‘Brockram” is a local term 
used in the Vale of Eden for the rock known to the western geologists as a “‘fan- 
glomerate.”’ 


ORIGIN OF THE TRIASSIC TROUGH OF CONNECTICUT 699 


However, the description of the geology of the Vale of Eden sug- 
gests a possible problem in sedimentation. A detailed study of the 
rocks of the eastern upland near the fault zone, combined with a 
_ microscopic study of the Anterior, Posterior, and Upper sandstones 
of the Connecticut Valley deposits might yield further evidence 
of the progressive growth of the eastern fault during the period of 
sedimentation. . 


IN SUPPORT OF GARDNER’S THEORY OF THE 
ORIGIN OF CERTAIN CONCRETIONS: 


LEROY PATTON 
New Concord, Ohio 


In an article in the Journal of Geology, Gardner? has maintained 
that certain concretions are formed in supersaturated or overloaded 
water carrying fine clay particles. He believes that the particles 
are pressed together and are gathered in lumps just as the finely 
disseminated particles of butter are gathered together in churning 
and that these particles grow by accretion and gain their spherical 
form by being rolled along the bottom. He bases his opinion on 
observations of aggregations of mud balls in the bed of a stream 
after a flood in the Rio Chaco region of the San Juan Basin, New 
Mexico. 

In the summer of 1921 the writer observed similar phenomena 
in the bed of the North Fork of the Red River, Beckham County, 
Oklahoma. As a result of a series of severe rains, this river, which 
is usually an insignificant stream flowing in sand-choked channels, 
had been flowing bank full. After the flood had subsided the writer 
observed on one of the sandy flats in the river bed a remarkable 
collection of clay aggregations similar to those described by Gardner. 
They consisted both of clay balls and cylinders, the former being 
much more numerous. The balls varied from less than an inch 
to about six inches in diameter. The cylinders were from four to 
six inches in diameter and a foot or more in length. Both balls 
and cylinders were composed of fine clay, with a small amount of 
sand and gravel in them or imbedded in the outer portion. The 
cylinders were apparently the result of two balls becoming stuck 
together and being rolled along the bottom, as several cases were 
observed showing the steps in this process. The considerable 

* : Published by permission of the Director of the Oklahoma Geological Survey. 


2 J. H. Gardner, “‘ Physical Origin of Certain Concretions,” Jour. Geol., Vol. XVI 
(1908), pp. 442-58. 


700 


LEROY PATTON 701 


number of these aggregations at the place described seemed to be 
due to the fact that the flat was on the inside of a rather sharp bend 
where the current would be slackened. 


Fic. 1.—Concretionary-like aggregations of fine clay deposited on a sandy flat 
in the bed of the North Fork of the Red River, Oklahoma, during a flood. 


From the very perfect resemblance of these clay aggregations 
to ordinary concretions and the rather large number of them found 
after this one flood, the writer is inclined to agree with Gardner 
that this method of formation of concretions may be more common 
than is ordinarily supposed. 

Figure 1 gives some idea of their size and distribution. 


MUD CRACKS ON STEEPLY INCLINED SURFACES 


GERALD R. MacCARTHY 
University of North Carolina, Chapel Hill, North Carolina 


It is generally accepted that well-developed mud cracks or 
desiccation fissures are formed only on level surfaces that have been 
covered by shallow stands of water. Hence the presence of this 
phenomenon would be supposed to indicate level, low-lying mud 
flats; estuarine flood-plain, or playa in origin. 

In the spring of 1922 it was my fortune to observe at Williams- 
town, Massachusetts, well-defined mud cracks in what I believe 
to be an atypical position. A small stream had undercut a bluff 
composed of finely laminated glacial-lake clays. Down the face 
of this bluff several mud streams had flowed, solidifying before 
reaching the brook. The surface of these mud streams was seamed 
with sun cracks which reached depths exceeding 8 inches, and whose 
intersections produced irregular polygons varying from 6 to 18 
inches across. Clinometer readings carefully taken on those por- 
tions of the surface which exhibited the best-defined polygons 
ranged from 11° to 38°, with an average of about 22°. Some of 
the best polygons appeared on the steeper slopes. 

Had any one of these mud flows been covered by later deposits 
and induration taken place, the presence of these sun-cracked poly- 
gons along the bedding-plane would have unquestionably been taken 
as evidence that it marked the contact between two horizontally 
deposited beds of clay. While the above-noted phenomena may 
not be at all unusual, it would seem that sufficient attention has 
not been called to occurrences of this type which might lead to 
serious stratigraphic errors after consolidation of the inclosing 
sediments. 


702 


PETROLOGICAL ABSTRACTS AND REVIEWS 
ALBERT JOHANNSEN 


SCHLOSSMACHER, K. “Die Sericitgneise des _ rechtsrheinischen 
Taunus,” Jahrb. d. Preuss. Geol. Landesanst., f. 1917, XX XVIII 


(1919), Th. 1, 374-433, pl. 1, figs. 2. 

The sericite-gneisses of the Taunus are dynamo-metamorphosed quartz- 
keratophyres and felso-keratophyres. Ten chemical analyses are given, of 
which four are new. A general description and a discussion of the chemical 
relationships are followed by descriptions of the various localities. 


SCHLOSSMACHER, K. “Ein Verfahren zur Herrichtung von schie- 
frigen und lockeren Gesteinen zum Diinnschleifen,” Ceniraldl. 
fio As GED, Boxy WOMO)S JED, WSO, ls, in 

The usual method of boiling porous and schistose rock fragments in Canada 
balsam is not efficient. Here is described an apparatus by means of which the 
pore spaces in the rock may be filled with balsam. A tube with a stoppered 
side opening is partially filled with balsam, placed in a water bath, and the air 
exhausted. Any time thereafter chips to be sectioned are placed in the side 
tube, the air is exhausted, the tube is tilted so that the chip falls into the 
balsam where it is left until the bubbles cease. 


SCHLOSSMACHER, K. “‘Keratophyre und ihre dynamometamor- 
phen Aquivalente aus der Umgegend von Bad Homburg im 
Taunus,” Jahrb. d. Preuss. Geol. Landesanst. f. 1919, XL (1920), 
Th. 1, 460-505. 

The metamorphic rocks of the Taunus are keratophyres and soda-kerato- 


phyres; tuffs were nowhere found. Detailed petrographic descriptions and 
two new chemical analyses are given. 


SCHLOSSMACHER, K. “ Einige nichtmetamorphe paléiovulkanische 
Eruptivgesteine aus dem Vordertaunus,” Zezischr. d. Deutsch. 
Geol. Gesell., LX XII (1920), 25-27. 

Unmetamorphosed paleovolcanic albite-trachytes and trachy-andesites, 
that is, keratophyres and keratophyre-porphyrites, are described from eight 


793 


704 PETROLOGICAL ABSTRACTS AND REVIEWS 


localities. The phenocrysts are usually albite, in two occurrences orthoclase 
with intergrown albite; the groundmass is trachytic albite with small amounts 
of magnetite, chlorite, and a little sericite. 


SCHNEIDERHOHN, H. ‘Die Methoden zur mikroskopischen Unter- 
suchung kristallisierter K6rper,’’ Handbuch der mikrosko pischen 
Technik, Stuttgart, X (1914), 45-94, figs. 68. 


Gives a brief but very good summary of petrographic-microscopic methods, 
well illustrated by figures. Attention is called to the fact that these methods 
are applicable not only to the determination of rocks, but that they may be 
used in the determination of natural and artificial salts, synthetic minerals, 
cement, etc. 


SCHNEIDERHOHN, Hans. “Uber Methoden, um rasch und einfach 
aus Photographien Strichzeichnungen.herzustellen,” Sencken- 
bergiana, I (1919), 190-93, figs. 2. 


While this is not a petrographic article, the method here given for rapidly 
reproducing photographs may be of interest to petrographers, especially in 
these days of poor print-paper. A developing-paper print of the thin section 
or other petrographic subject is “inked-in” with waterproof ink. It is then 
immersed in subdued light for a few minutes in acid hypo, and then, without 
washing, placed in a rather concentrated solution of about equal parts copper 
sulphate and potassium bromide. In a few minutes the silver image will 
become altered to a yellowish image of silver bromide. When of a yellow color, 
it is washed for a few moments and re-immersed in the hypo until the yellow 
color disappears and leaves a white background. Without first immersing in 
hypo the operation is somewhat slower. Instead of copper sulphate, potassium 
ferricyanide may be used. A second method, less desirable on account of the 
poisonous nature of the material, but more rapid, is to dissolve the silver in a 
dilute solution of potassium cyanide. The print should be thoroughly washed 
afterward. 


SCHNEIDERHOHN, Hans. ‘Die mikroskopische Untersuchung un- 
durchsichtiger Mineralien und Erze im auffallenden Licht und 
ihre Bedeutung fiir Mineralogie und Lagerstéttenkunde,’’ 
Neues Jahrb., B. B. XLIII (1920), 400-438. 

Here is a most excellent summary of work done on the determination of 


opaque minerals by means of incident light under the microscope. A long 
bibliography is given. 


PETROLOGICAL ABSTRACTS AND REVIEWS 705 


SCHNEIDERHOHN, Hans. “‘Beitrige zur Kenntnis der Erzlager- 
statten und der geologischen Verhiltnisse des Otaviberglandes, 
Deutsch-Siidwestafrika,” Abhandl. d. Senckenbergischen Naturf. 
Gesell., XX XVII (1921), 221-321. Figs. 16, figs. 40 in photo- 
gravure, and colored map 1. 


The greater part of this report is economic and geologic. Only a few 
igneous rocks are mentioned, namely, aplite, olivine and mica kersantites, and 
microgranite. The aplite occurs in a dike-like mass widening into lens-shaped 
masses in several places. It is cut by the younger kersantite. The micro- 
granite forms a laccolite intruded between strata of the dolomite, and in all 
probability was derived from the same source as the aplite which may represent 
the channels through which the larger mass was intruded. 


SCHURMANN, H. M. E. ‘“‘Beitrage zur Petrographie der éstlichen 
arabischen Wiiste Agyptens,” Centralbl. f. Min., Geol., etc., 1921, 
449-58, 481-90. 

Very brief descriptions are given of the various igneous rocks of Gebel 
Mogul, between Gebel Mogul and Um Dalfa, and between Gebel Gharib, 
Gebel Dara, and Gebel Mogul. Neither chemical analyses nor modal per- 
centages are given, though the various minerals are named. The rocks de- 
scribed are various granitites, pegmatitic granite, hornblende-syenite, and 
tonalite as plutonic rocks occurring in stocks; pneumatolytic granite, granite- 
pegmatite, quartz-diorite, and quartz-augite-diorite, plutonic rocks in dikes; 
pegmatite, graphic-granite, quartz, aplite, riebeckite-aplite, quartz-bostonite, 
malchite, minette, augite-kersantite, amphibole-vogesite, granite-porphyry, 
riebeckite-granite-porphyry, and quartz-diorite-porphyrite in dikes; and the 
following extrusive rocks, also in the form of dikes: granophrye, quartz- 
porphyry, riebeckite-quartz-porphyry, felsite-porphyry, various porphyrites, 
andesite, diabase-porphyrite, and diabase. 


SCHUSTER, Ernst. ‘Calcitfiihrende Auswiirflinge aus dem Laacher 
Seegebiet,”’ Neues Jahrb., B. B. XLIII (1919), 295-318, pls. 2. 


The calcite-rich ejected blocks of the Laacher Sea region which occur in 
the leucite-phonolite-tuff are alkali syenites which must have formed the 
country rock in the deeps. The calcite is regarded as a magmatic mineral, as 
is also cancrinite and the rare calcium apatite. The rocks may be called 
calcite-pegmatites and calcite-syenites. In other fragments melilite was 
developed as well as calcite. 


706 PETROLOGICAL ABSTRACTS AND REVIEWS 


SEDERHOLM, J. J. “On Synantetic Minerals and Related Phe- 
nomena. (Reaction Rims, Corona Minerals, Kelyphite, Myr- 
mekite, etc.),”” Bull. Comm. Geol. Finlande, No. 48, 1916. Pp. 
148, pls. 8, figs. 14. 

Synantetic minerals are those which are characteristic at the contact between 
two definite minerals in igneous rocks, kelyphite rims being one form. Myrme- 
kite is applied to intergrowths of plagioclase and vermicular quartz. In this 
paper the various forms and the different minerals occurring are discussed in 
great detail, and the literature is fully summarized. 


SHAND, S. J. ‘‘The Pseudotachylyte of Parijs,” Quart. Jour. 
Geol. Soc., LX XII (1917); 198-221, pls. 4, figs. 13. 

In the granite from the neighborhood of Parijs, Orange Free State, there 
occur abundant veins and networks of a dense black rock, to which, from its 
resemblance to tachylite, the name pseudotachylite is given. Numerous 
sketch maps and two photographs show the nature of the occurrence in the field, 
and eight photogravures show the appearance as thin sections. The rock is 
very opaque, due to innumerable inclusions of very fine black specks of magne- 
tite. In some of the widest veins there is less magnetite but many polygonal 
spherulites of dark-brown color in a felt of feldspar-microlites. Several analy- 
ses are given. The writer concludes that the pseudo-tachylyte originated from 
the granite itself through melting, which was caused, not by shearing, but by 
shock or gas-fluxing. 


SHAND, S. J. “‘The Principle of Saturation in Petrography,”’ 
Geol. Mag., I (t914), 485-93; II (2915), 339-40. 

Mr. Shand replies to certain critics of his system of classifying rocks on the 
basis of saturated or unsaturated minerals. (The former minerals are those 
which are stable in the presence of free silica under magmatic conditions, the 
latter those that are unstable.) 


SHanp, S. J. “A System of Petrography,” Geol. Mag., IV (1917), 
463-609. 


Gives further ideas as to desirable features in a classification of rocks. 
Shand proposes the following factors: (1) degree of saturation, giving five 
divisions; (2) the double ratio of Or-Ab-An, giving about eight families within 
each division; (3) the color ratio, giving from two to ten, but preferably four 
groups in each family; (4) crystallinity, giving two sub-groups within each 
group; (5) ratios of specific minerals or groups of minerals, giving the types to 
which “specific” names will be attached; (6) trivial characters of mineralogy 
and texture, giving varieties. 


PETROLOGICAL ABSTRACTS AND REVIEWS 707 


SHAND, S. J. ‘‘The Norite of the Sierra Leone,” Geol. Mag., V 
(1918), 21-23. 
Describes two norites from Sierra Leone. One an olivine-rich norite, belongs 


to 2312 (new form) of the reviewer’s classification; the other melanocratic, 
and without olivine, belongs to 3312. 


SHANNON, Eart V. ‘‘Petrography of Some Lamprophyric Dike 
Rocks of the Coeur d’Alene Mining District, Idaho,” Proc. 
U.S. Nat. Museum, LVII (1920), 475-95, pls. 3. 

The various dike-rocks from the Coeur d’Alene district, collected by Ran- 
some, Calkins, and Umpleby, are here classified and described. Among the 
rocks are various minettes, spessartites, and vogesites, and one odinite. From 
the widespread occurrence of these dikes the conclusion is reached that the 
district is underlain by a granitic batholith which is so far down that none of 
the complementary aplite reached the surface. The dikes and ore veins belong 
to substantially the same period. 


SKOETSCH, CaRL. ‘‘Die Einschliisse in den Basalten zwischen 
Godesberg und Remagen,” Ceniralbl. f. Muin., etc., 1921, 
353-63. 

In this paper are described all the different minerals which have been found 


in inclusions in the basalts of this region, as well as their mode of origin, and 
the alterations produced in them by the basaltic magma. 


Situ, W. CampBeLt. “ Riebeckite-Rhyolite from Northern Kordo- 
fan, Sudan,” Mineralog. Mag., XIX (1920), 48-50. 

Describes a riebeckite-rhyolite from which certain ancient stone imple- 
ments found at Beraeis are made. Two specimens of tinguaitic dikes from 
Kadoro, described by Linck, represent the only previously mentioned soda- 
rich rocks in Kordofan. 


SPANGENBERG, K. ‘Die Einbettungsmethode,” Fortschr. d. Min. 
Krist. u. Petr., VII (1920), 397-458. 

Under “Immersion Methods” are included all those methods for determining 
refractive indices based upon certain appearances at the contact between a 
known and an unknown medium. Three groups are discussed: (1) Disap- 
pearance of the border, (2) Tépler’s method of inclined illumination (often 
spoken of as Schroeder van der Kolk’s method); (3) Becke’s method of raising 
or lowering the tube of the microscope. A general summary is given of all 
methods, the reasons for the phenomena are discussed, the relative accuracy 
shown, and the cause of variation under different conditions pointed out. 


708 PETROLOGICAL ABSTRACTS AND REVIEWS 


SPANGENBERG, K. ‘“Einige Anwendungen und Erweiterungen der 
Einbettungsmethode,” Centralbl. f. Min., etc., 1920, 352-62, 
406-14. 


Gives various applications of the immersion method. 


STEIDTMANN, EpwarD. “Origin of Dolomite as Disclosed by 


Stains and Other Methods,” Bull. Geol. Soc. Amer., XXVIII 
(1917), 431-50, pls. 7. 

Most dolomites were deposited in the sea. A minority were formed by the 
replacement of limestones by underground waters. Pure dolomites and lime- 
stones are far more abundant than mixed beds of limestone and dolomite. 
The occurrence of calcitic casts in dolomite, or of hollow casts bounded by 
perfect molds, indicate that the casts were deposited in dolomite. Dolomite 
rhombs, imbedded in a hornlike impervious mass of fine-grained marine calcite, 
were evidently formed in the ooze contemporaneously with the calcite. 


Tarr, W. A. ‘‘Odlites in Shale and Their Origin,” Bull. Geol. Soc. 
Amer., XXIX (1918), 587-600, pls. 2, figs. 2. 


Describes certain odlites found in shale in the Wind River Mountains, near 
Lander, Wyoming. They are believed to be due to direct precipitation of 
colloidal silica by the electrolytic and saline character of the shallow waters 
into which they were introduced by streams from the adjacent land. 


Tarr, W. A. “Origin of the Chert in the Burlington Limestone,” 
Amer. Jour. Sci., XLIV (1917), 409-52, figs. 13. 

Believes the widespread chert which occurs in the Burlington formation 
of Mississippian age has been formed from colloidal silica derived from inflowing 
streams and deposited by electrolytic action. The ellipsoidal form of the chert 
is attributed to the flattening of the colloidal mass under its own weight and 
later by the weight of overlying sediments. 


TittEy, C. E. “The Petrology of the Granitic Mass of Cape 
Willoughby, Kangaroo Island, Part I, Trans. Roy. Soc. South 
Australia, XLIII (1919), 316-41, pls. 2, sketch maps 2. 


The granitic rocks of Cape Willoughby, Victor Harbor, and Port Elliot are 
thought to be chonolites connected below with a single batholith. They were 
intruded at the close of the orogenic movements in the region. The dominant 
rock is granite with minor intrusions of aplite and pegmatite. Interesting 
rocks are the albitites, quartz-albitites, and muscovite-albitites, which are 
regarded as the final differentiates from the residual magma. The first rock 


PETROLOGICAL ABSTRACTS AND REVIEWS 709 


consists essentially of albitite, with accessory apatite, zircon, and rutile, and 
with small amounts of muscovite and quartz. The albitite has the character of 
the ‘‘chequer”’ albite of Flett. The quartz-albitite is similar to the preceding 
but contains a blue opalescent quartz. The muscovite-albitite contains essen- 
tial muscovite, some of which is regarded as primary, though some is secondary. 
The amounts of quartz and muscovite are not stated. The first and third rocks 
belong to 1112 (new form) of the reviewer’s system, the second is 118 if the 
amount of quartz is over 5 per cent, as it presumably is since these rocks are 
contrasted with quartz-bearing albitites. 


TittEy, C. E. ‘‘The Occurrence and Origin of Certain Quartz- 
Tourmaline Nodules in the Granite of Cape Willoughby,” 
Trans. Roy. Soc. South Australia, XLIII (1919), 156-65, pls. 2. 


Certain nodules, consisting essentially of quartz and tourmaline, occurring 
in an aplite intrusive in granite, are considered as having developed by the 
replacement of albite and microcline by tourmaline. Says the writer: ‘“ Micro- 
scopic and other evidence tends to show that they are strictly penumatolytic 
products. In the slides is to be seen the very act of replacement of feldspar by 
tourmaline.” 


TsuBOI, SEITARO. ‘‘On the Determination of the Limiting Values 
of the Medium Refractive Index of a Finely Crushed Biaxial 
Crystal by the Immersion Method,” Jour. Geol. Soc. Tokyo, 
XXV (1918), 38-41, fig. 1. 


Maximum and minimum values of refractive indices are readily determin- 
able but the intermediate value must be obtained from a carefully oriented 
section or be computed. In the latter case the angle between the r axis and 
one of the optic binormals must be known. In the present paper is given a 
method of determining limiting values for 8, based on the fact that it must 
always lie between the two values observed in a crystal section of any orienta- 
tion. By making observations on many grains, the difference between upper 
and lower limiting values may be made very small. Using basic plagioclase, 
the author made determinations to 0.003. 


TsuBOI, SEerTaRO. ‘Notes on Miharaite,”’ Jour. Geol. Soc. Tokyo, 
XXV (1918), 47-58, pls. 2. 


The term miharaite is given to a lava from the volcano Mihara on the island 
of Oshima, Idzu. It is a basalt characterized by abundant phenocrysts of 
bytownite with a few of hypersthene and clino-hypersthene, and a very small 
amount of augite. The groundmass contains labradorite-bytownite microlites, 


710 PETROLOGICAL ABSTRACTS AND REVIEWS 


augite, magnetite, rare apatite, and negligible glass in the gray varieties, and 
plagioclase and augite in brown glass in the slaggy kinds. Five chemical 
analyses are given, all giving high SiO, (51.94, 51.13, 51.32, 51.40, and 51.45). 
The rock is given a new name on account of its occult quartz and normative 
bytownite. Mineral percentages are not given. 


TsuBOI, SEITARO. “A Diagram for Determining Plagioclases.”’ 
Published in the Japanese language in Jour. Geol. Soc. Tokyo, 
XXVII (1920). 

Since cleavage pieces of plagioclase are used in determining their refractive 
indices, a table giving the values in (oro) and (oor) is of much greater value 
than the usual one giving a, 6, and y. These values, computed by Tsuboi 
and plotted as a curve, are reproduced in the reviewer’s Essentials in the Deter- 
mination of Rock-forming Minerals and Rocks. 


TsuBOI, SEITARO. ‘On a Leucite Rock, Vulsinitic Vicoite, from 
Utsuryoto Island in the Sea of Japan,” Jour. Geol. Soc. Tokyo, 
XXVII (1920), g1-104. 

Describes a porphyritic rock with abundant phenocrysts of sanidine and 
labradorite, the former slightly more abundant than the latter, less hornblende, 
augite, and titanaugite, and microphenocrysts of biotite, olivine, and apatite. 
The groundmass consists of laths of orthoclase and plagioclase and round leu- 
cites, with prisms and grains of aegirite-augite, some magnetite, and a trifle 
glass. An analysis is given which, recast into the norm, gives 4.63 per cent 
nephelite, 39.59 orthoclase, but no leucite. The analysis is readjusted to give 
leucite with the approximate proportions orthoclase 21.5, albite 42, anorthite 
9, leucite 14, diopside 4.5, magnetite 2.8, ilmenite 1.2, apatite 1.1, olivine 2.7, 
and zircon o.11. Compared with the description, however, this does not 
represent the actual mode (in which the plagioclase is stated to be labradorite), 
consequently it cannot be classified in the reviewer’s system. 


TsuBolI, SEITARO. ‘Volcano Oshima, Idzu,’’ Jour. Col. Sci., Tokyo, 
XLITI (1920), art.8. Pp. 148, 24 photomicrographs on 4 plates, 
map I, plate profile 1, figs. 42. 

Part of this report was published in the preliminary papers described in the 
second and third preceding articles. Here is given a geological and historical 
sketch of the volcano Oshima as well as descriptions of the rock types. These 
are basaltic bandaites, miharaites (which resemble the preceding but have no 
olivine), and basalt. Various analyses are given, and some beautiful photo- 
micrographs. 


PETROLOGICAL ABSTRACTS AND REVIEWS Tas 


A modification of Becke’s method for determining 2V, here given, greatly 
simplifies the process. The isogyre is first placed parallel to the horizontal 
cross-hair of the microscope, and the position of the melatope is determined by 
two angles, one measured in azimuth from the vertical cross-hair to the melatope 
by rotating the stage, the other measured from the center by means of a grad- 
uated eyepiece and any method similar to that employing a Schwarzmann’s 
axial angle scale. The position of the melatope (A) is marked on a stereo- 
graphic net and a great circle is drawn through it and the ends of the horizontal 
line. So far the method agrees with that of Becke. Change the conoscope into 
an orthoscope and rotate the stage until the section is at extinction and read 
the angle through which the stage was rotated. This locates one of the vibra- 
tion directions which is now drawn at the proper angle in the projection. Lo- 
cate, on the great circle previously drawn, the other melatope by means of an 
angle equal to that between the first melatope and the line representing the 
vibration direction. Measure 2V in the projection. In the older methods it 
was necessary to observe a second point on the isogyre, which was difficult. 
In the Tsuboi method only the position of the melatope is needed. Further, 
there is no need of using the refractive index in the new method. In Becke’s 
original, five great circles were necessary, and in Wright’s modification, four. 
In Tsuboi’s, only one great circle is drawn, and one straight line. 


TYRRELL, G. W. “A Contribution to the Petrography of Ben- 
guella, Based on a Rock Collection Made by Professor J. W. 
Gregory,” Trans. Roy. Soc. Edinburgh, LI (1916), 537-59, pl. 1. 


Benguella is one of three provinces of the Portuguese West African colony 
of Angola. Chemical analyses and complete descriptions are given of the rocks, 
which are granite, charnockite, dellenite, nephelite-sodalite-syenite, akerite, 
shonkinite, solvsbergite, ouachitite, and various basic intrusives. The two 
“‘charnockites”’ of Table I are 227’ (new form of reviewer’s system) or monzo- 
tonalites (granodiorites in limited sense), while the type charnockite from India 

‘is 226/ , or typical granite. The hornblende-hyperites belong to 3312, as they 
should. The granite of Table II is 216’, typical granite; the granodiorite is 
227’, granodiorite in sense usually used but better monzo-tonalite. The two 
dellenites computed in Table IV are 227’ for the Angola rock, and 227” for the 
one from Sweden. That is, the former is the extrusive equivalent of a grano- 
diorite, while the latter is the extrusive equivalent of a quartz-monzonite. The 
shonkinite of Table V is 2113, which is not according to definition of shonkinite 
as originally give by Pirsson, for in that the dark constituent must form 
more than half the rock, consequently it must be in Class 3, as actually is 
the type Montana specimen given in the same table. The Angola rock falls 
into the group with pulaskites although the feldspathoid in the latter rock was 
given by Williams as nephelite or its decomposition product analcite. 


712 PETROLOGICAL ABSTRACTS AND REVIEWS 


TyRRELL, G. W. ‘Further Notes on the Petrography of South 
Georgia,’ Geol. Mag., III (1916), 435-41. 

Describes various rocks from South Georgia. The sediments are slates 
and phyllites, arkoses and grits. The igneous rocks are epidorite, dolerite, 
basalt, alaskite, quartz-felsite, lavas and tuffs of doubtful affinities, epidosite, 
and augitite. So far as petrographic evidence goes, the question whether South 
Georgia belongs to Suess’ ‘Southern Antilles,” or whether it is a remnant of an 
old sunken continental land remains unsettled. 


TyRRELL, G. W. “The Petrography of Arran,” Geol. Mag., III 
(1916), 193-96. 


Pitchstone xenoliths in a basalt dike throw some light upon the ques- 
tion of the temperature of lavas. The phenocrysts of quartz have suf- 
fered hardly at all, the andesine has had a softening on the margins 
and fissuring in the interiors, while the orthoclase shows fusion around the 
margins and along cleavages, producing a yellow or grayish glass, differing from 
that of the groundmass. The temperature of the intruding lava is therefore 
thought to have been between 1170° C. and 1375° C. 


TyrRELL, G. W. ‘Some Tertiary Dykes of the Clyde Area,” 
Geol. Mag., IV (1917), 305-15, 350-56, figs. 3. 

Describes a dike-rock consisting of phenocrysts of anorthite in a ground- 
mass of labradorite, enstatite and augite, and much glass. The glass indicates 
orthoclase, silica, and albite. Chemically this rock approaches andesite, from 
which it differs in its more basic phenocrysts. It is here called Cumbraite. It 
differs from Thomas and Bailey’s Inninmorite in containing enstatite as well 
as augite. While the name cumbraite is proposed by Tyrrell, he says: 
‘“‘Whether these terms should obtain a circulation outside the discussion 
of the British Tertiary petrographic province is a question beyond the scope 
of this paper. My own opinion is that they should not.” 


TYRRELL, G. W. ‘The Trachytic and Allied Rocks of the Clyde 
Carboniferous Lava-Plateaus,” Proc. Roy. Soc. Edinburgh, 
XXXVI (1917), 288-99. 

Among the lavas of the Scottish Carboniferous, true andesites and rhyolites 
are absent, trachyte and allied rocks are present in subordinate quantities, 
while basalts predominate. In this paper are brief descriptions of albite- 
bostonites, albite-trachytes, albite-keratophyres, bostonites, keratophyres, 
quartz-keratophyres, felsite, and phonolite. Ten analyses, one of bostonite 
previously unpublished, are given. 


PETROLOGICAL ABSTRACTS AND REVIEWS 713 


TyRRELL, G. W. “The Igneous Geology of the Cumbrae Islands, 
Firth of Clyde,” Trans. Geol. Soc. Glasgow, XVI (1916-17), 
244-74, figs. 5. 

Most of the igneous rocks of the Cumbraes are of Lower Carboniferous age. 


They are predominatly basaltic and originally covered from 2,000 to 3,000 
square miles. 


TyrRELL, G. W. ‘The Picrite-Teschenite Sill of Lugar,” Quart. 
Jour. Geol. Soc., LX XII (1917), 84-131, pls. 2. 


The Lugar sill in the west of Scotland is found to be made up of a complex 
of rocks belonging to the analcite series. It forms a mass 140 feet thick and 
was intruded into cold rocks, as shown by chilled contacts at top and bottom, 
giving a fine-grained teschenite for a thickness of 10 feet. Beyond the mar- 
gins, both top and bottom, the rock passes into coarse teschenite. In the 
interior the sill is divided into at least three bands by some process of differen- 
tiation or by successive intrusions, giving first a band of ultra basic rock— 
picrite and peridotite of coarse texture—occupying the major part of the whole 
mass. The picrite forms the upper part of the ultrabasic stratum, the perido- 
tite the lower. Above the picrite is a band about 10 to 15 feet thick of fine- 
grained, basic, nephelite rock of the theralite family. Overlying the picrite, 
in places, is a peculiar rock to which the name Jugarile has previously been 
given. It appears to be intrusive in the picrite, for veins of similar material 
traverse the latter rock in various places. Three of the teschenites are 23109’ 
of the reviewer’s system, and two are 3320, the difference being the absence 
of orthoclase and the predominance of the dark constituents in the latter. Two 
chemical analyses are given as well as five more calculated from Rosiwal 
measurements. Two lugarites are 2320. (A rock described as lugarite in 
Geol. Mag., 1915, 363, is here called a lugarite-like rock. It is 2218’.) One 
chemical analysis and two calculated analyses from Rosiwal measurements are 
given. As for the cause of the differentiation, the author thinks that the hy- 
pothesis of sinking of heavy crystals is well attested in the Lugar magma as a 
whole, but that the differentiation took place prior to its emplacement. The 
material was injected in successive intrusions, the teschenite first in cold rocks 
and formed the fine teschenite borders. While still cooling, but probably 
already solid, the picrite was intruded along its center plane. Here differentia- 
tion took place mainly by the sinking of olivine-crystals. Later, while probably 
still partly liquid, it was intruded by a small mass of lugarite. 


VELDE, Luise. ‘Die silikatischen Einschliisse im Basalte des 
Biihls bei Kassel,” Abhandl. d. Senckenberg. Naturf. Gesell., 
SOOO (azo), MiB G, ToS, ah 


The silicic inclusions in the Biihl basalt in many cases preserve the charac- 
teristics of the original rocks from which they were derived, namely, sandstones 


4 


714 PETROLOGICAL ABSTRACTS AND REVIEWS 


and slates. The great majority of the rocks are strongly metamorphosed and 
in them have been developed sillimanite, corundum, spinel, magnesium-diop- 
side, scapolite, cordierite, and plagioclase. In this work the various inclusions 
are petrographically described and several chemical analyses are given. The 
most abundant inclusions are quartz-sillimanite. 


C. H. BEHRE, JR. 


Voct, J. H. L. “Die Sulfid-Silikat-Schmelzlésungen,” Norsk. 
Geologisk Tidsskrift, IV (1917). Pp. 97, figs. 13, and several 
tables and analyses. 


VoctT, J. H. L. Die Sulfid-Silikat-Schmelzlosungen: Die Sulfid- 
schmelzen und die Sulfid-Silikatschmelzen. Christiania, 1919. 
Pp. 131, figs. 45, and numerous tables and analyses. 


The first of these two papers is essentially a résumé, written in 1917, of 
extensive work on sulphide-silicate solutions, giving the results obtained to the 
date of its publication. The second paper presents in detail the data of 
the earlier one, and embraces additional facts gleaned through two more years 
of work on the same subject; it is more detailed than the earlier publication 
and will be reviewed here first. The reviewer believes that with such com- 
prehensive work as this, adequate abstracts are impossible. He strongly 
advises a careful perusal by metalographers, economic geologists, geochemists, 
and physical chemists. He wishes to commend the completeness of these 
studies. 

Previous experiments have shown that certain sulphides, such as Sb,S;, 
Bi,S;, and Ag.S, have a lower melting point than even those silicates with the 
lowest melting points. Other sulphides, such as those of lead, copper (Cu,S), 
and iron (FeS) and pyrrhotite, have melting points about like those of the 
least refractory silicates and slightly higher than some of the eutectic mixtures 
of silicates with low melting points. Other sulphides finally, such as those of 
zinc, manganese, barium, and calcium, have melting points markedly higher 
than those of the more common natural silicates. Under-cooled sulphide 
mixtures or solid solutions of sulphides (sulphide glasses) are unknown. 

From a study of the latent heat of fusion it appears that the sulphides PbS, 
Ag,S, Cu.S, FeS are not highly polymerized. _It is found, further, that Fe,O, 
is only slightly soluble in melts of Cu.S; this is corroborated by the crystalliza- 
tion sequence as observed in magmas, for magnetite (and ilmenite) crystallize 
very early indeed from a pytrhotite- or pyrite-bearing magma. Silicates are 
soluble in FeS or Cu,S melts to only a very minor degree. 

After a study of the relations between the various sulphides and their 
eutectics, the writer demonstrates that a eutectic is also possible in solutions of 
calcium sulphide (or manganese sulphide) in various silicates, such as melilite 


\ 


PETROLOGICAL ABSTRACTS AND REVIEWS TES 


and olivine. For example, in sulphide-rich melts with melilite, much of the 
mass of sulphide crystallizes out before the corresponding spinel, and the 
remainder crystallizes synchronously with the spinel. 

In a solution of calcium magnesium silicate, calcium sulphide in the amount 
of 2.3 per cent lowers the melting point about 50°. By calculating the molecular 
depression it is found that polymerization of the sulphides of calcium and 
manganese in silicate melts is essentially nil; in fact, in the silicate solutions 
an extensive electrolytic dissociation more probably takes place. 

Somewhat similar results as to the presence of a eutectic in solutions of 
zinc, aluminum, and iron oxides are discussed. In a melt bearing zinc sulphide, 
zinc spinel, and melilite, the order of crystallization follows the order of nam- 
ing, as above; if, however, olivine be present instead of melilite, and only a 
very little sulphide, the order of crystallization is spinel, olivine, and sulphide. 

In other slags the presence of copper and its relation to chondri-like 
structures, and to iron sulphide-bearing silicate solutions were studied. An 
interesting feature is the concentric arrangement of iron sulphide inclusions 
in hexagonal plates of biotite. From these observations it is found that Cu,S 
is wholly insoluble or at best only slightly soluble in silicate melts rich in iron 
sulphide—attributable possibly to the presence of a common ion. Various 
conclusions drawn in this part of the investigation are interesting not only to 
the geochemist, but to the metallurgist as well. 

Some space is also devoted to the crystallization of apatite and ore—the 
so-called ‘“‘telechemical”’ minerals—those only distantly related to the silicate 
minerals. 

The foregoing tacts may be gleaned from the publication of 1919. The 
paper of 1y17 briefs most of these observations, and adds considerable material 
on the part of sulphides in eruptive magmas—an application of physical 
chemistry to systematic petrogenesis. 

Pyrite appears to crystallize very early from magmas, and has a higher 
melting point than magnetite. Pyrite generally precedes pyrrhotite in the 
normal sequence of crystallization; it also precedes chalcopyrite, which gener- 
_ally follows magnetite when all three sulphides crystallize from a melt. As 
previously observed, magnetite is only very slightly soluble in pyrite-rich 
melts; hence iron oxides are rare or absent in such melts. 

A study of the norite-type magmas leads to the conclusion that the pre- 
dominant mineral generally crystallizes first; thus, in plagioclase-rich norite, 
plagioclase is automorphic; in olivine-plagioclase, plagioclase-rich rock, plagio- 
clase; in olivine-rich olivine-plagioclase rocks, olivine is automorphic, and so 
on. The physical chemistry of the several systems normally present in a gab- 
bro-norite magma is summarized, and the writer believes in a eutectic for 
rocks of the gabbro-norite group, this eutectic, however, being expressed by a 
line, rather than by a point. 

The rather constant appearance of pyroxenites and peridotites in close 
association with nickel-pyrrhotitic deposits, as at Sudbury, is thought to be 


716 PETROLOGICAL ABSTRACTS AND REVIEWS 


indicative of a eutectic mixture. Various conclusions regarding the order of 
crystallization in such magmas are drawn from petrographic data. The 
crystallization temperature for the normal norites is calculated (estimated) 
to lie between 1200° and 1300° at an atmosphere’s pressuie. From norite 
magmas acid dikes are distinct differentiation products, hence very generally 
associated with nickel-bearing norite. Petrographically the writer seeks to 
establish in such basic rocks (of the picrite-norite series) the increase of the 
nickel content with increase in hypersthene or bronzite; all the nickel deposits 
associated with peridotite have an exceptionally high nickel content. 

The sulphides in the norite masses are supposed to be precipitated from the 
melts with decreasing temperature and then to settle to the bottom; the 
small amount of sulphides remaining later crystallized with the silicates and was 
not whoily segregated. 

This work has demonstrated: (1) that the origin of nickel-pyrrhotite 
deposits may be explained by the laws affecting a system liquid-liquid; (2) 
that the leucocratic acid pyritiferous dikes correspond to the end-product of 
crystallization in a noritic magma bearing a slight amount of free quartz; (3) 
that the numerous occurrences of a chalcopyrite-rich sulphide mixture in 
locally distributed veins, especially such as are limited to the border phases 
of the rock, depend upon the fact that the chalcopyrite was concentrated in 
the end magma through progressive solidification of the sulphide constituents. 


REVIEWS 


Description géométrique des Alpes Frangaises, Annexe du Tome 
second. By P. HELBRONNER. Six panoramas. dessinés et 
peints par l’auteur, 23 planches dans un carton in-folio. Paris: 
Gauthier-Villars, 1921. 

Gauthiers-Villars of Paris have, in this portfolio, put on the market 
colored panoramas of the French Alps which for both accuracy and 
beauty excel anything which has hitherto been produced. These pano- 
ramas have been drawn and colored as aquarelles by Paul Helbronner 
upon the basis of a trigonometric survey which this accomplished map- 
maker, Alpinist, and artist has made. In the course of this work, he 
found it necessary to ascend with his instruments all the high and diffi- 
cult peaks of the region. In all, the panoramas consist of twenty-three 
plates in color joined in six groups. The largest of the panoramas is 
a complete tour of the horizon from the summit of Mont Blanc and 
consists of thirteen plates making a picture 19 ft. 6 in. in length. Two 
others—Mont Blanc from the Mont-Maudit and Mont Blanc from the 
Col du Géant—consist each of three plates of the same size as the 
others. Under each of the panoramas is an outline-drawing from which 
each peak may be identified. ‘The coloring of these remarkable panora- 
mas is superb while faithful to nature, as will be testified to by anyone 
who has climbed in these regions. The publication of the portfolio 
marks an epoch in the history of reproduction in color. The panoramas 


are all well suited for framing. 
W. H. Hosss 


‘Coal. By Exzwoop S. Moore. New York: John Wiley & Sons, 
March, 1922. 

This is primarily a volume for the mining student and the engineer 
who is in charge of coal-mining operations, but it should also be of inter- 
est to every geologist. It brings together the best of a varied literature 
and places it at the disposal of all in usable form. 

Neither time nor space is wasted in the introductory chapter. From 
a brief history of the subject the author starts out energetically to discuss 
the megascopic and microscopic properties of coal and then passes 
naturally into a discussion of its chemical composition. The best and 
latest methods of sampling are given in sufficient detail to enable a man 


717 


718 REVIEWS 


to follow them satisfactorily. A similar statement might be made 
regarding the methods of chemical analyses and, in later chapters, on 
mining methods. 

It is unfortunate that with all our classifications of coal none seem to 
meet with general approval. Probably the best is that adopted by the 
Twelfth International Congress of Geologists, and which was originally 
developed by D. B. Dowling of the Canadian Geological Survey. The 
author favors a classification based upon more than two factors as best 
meeting the requirements, but admits the inadequacy of those now in use. 

The origin of coal, through the transformation of vegetal matter 
accumulated in swamps, is discussed in-considerable detail. The amaz- 
ing rapidity of the metamorphic processes in any such accumulation is 
illustrated by the pebbles of coal occurring within the Coal Measures and 
the alteration of the upper end of a mine prop which had been subjected 
to high pressure for thirty years, during which time it also felt the effects 
of the heat from a fire in an adjacent part of the mine. The upper part 
of the prop and the cap wedge had a jet black color, a bright glossy 
luster, and conchoidal fracture. Evidently the intensity of the other 
factors may compensate for the lack of a large portion of the time ele- 
ment usually regarded as essential in the formation of the higher grades 
of coal. 

A chapter is devoted to the vegetation of the coal periods and deals 
chiefly with the extinct forms. The author then takes up the structural 
conditions existing in coal seams, the location and determination of 
thickness of beds, and the value of coallands. The latest mining methods 
and the preparation of the coal and coal products for the market are 
given due consideration. Finally a summary of the geology of the coal 
fields and the coal resources of the world completes the volume. 

The book is adequately illustrated by well-chosen cuts, maps, and 
photographs. ‘The plan of the text is admirable and the author handles 
his subject with good clear English, which is free from useless repetition. 
This has made it possible for him to get a remarkable amount of informa- 
tion into a single volume. Considering its size and scope, it is certainly 


one of the best texts on coal that has been published. 
Cu. Rags: 


Structure in Paleozoic Bituminous Coals. By REINHARDT THIESSEN, 
United States Bureau of Mines, Bulletin 117, Washington, 1920. 
Pp. xilit+296. Pls. CLX. 
An extensive historical review of previous studies of woody structure 
in coal is followed by a detailed description of the technique of the 


REVIEWS 719 


author’s investigation. A study of the origin of peat follows, which 
throws some light on the way in which coal was formed in Paleozoic 
times. The author’s study of the structure of coal embodies the results 
of his examinations of a number of different Paleozoic coals. Particular 
attention is given to microscopic studies, many of which were made with 
a magnification of 1,000 diameters, approaching the limit of visibility. 
The text is accompanied by 160 plates, many of which contain several 
illustrations,and a bibliography of publications on the composition of coal. 
An enormous amount of valuable information on the composition of 
coal has been accumulated in this bulletin. Many side lights on plant 
life during the Paleozoic are brought out by the study of the spores and 
other morphological elements of the coal. Biologic factors like the origin 
of rootlets and the existence of fungi in Paleozoic times, are revealed. 
There are a number of theories concerning the origin of coal, but we 
are yet unable to form a conclusive conception of this interesting geo- 
logical process. Thiessen’s paper supplies a great amount of information 


which brings us a step nearer to a satisfactory conception. 
ACC NE 


Contact-metamorphic Tungsten Deposits in the United States. By 
FRANK L. Hess and Esper S. LARSEN. United States Geo- 
logical Survey, Bulletin 725-D, 1921. 

Of the 5,000 tons of tungsten concentrates (reckoned as 60 per cent — 
WO,) produced in the United States in 1918, about 1,400 tons was in 
the form of scheelite (CaWO,) from contact-metamorphic deposits. 
Most of these deposits are along the western side of the Great Basin in 
California and Nevada but there are scattered deposits near Great Salt 
Lake and in Arizona, New Mexico, and Oregon. Their development has 
been recent, the first of the type being discovered in 1908. During the 
European war most of them were active producers but by 1920 all had 
lapsed into idleness because of the severity of competition with imported 
concentrates and richer American ores, combined with the great depres- 
sion in the steel industry. 

The contact-metamorphic tungsten deposits are nearly all at or near 
the contact between quartzose igneous rocks, principally granodio- 
rites, and limestones. In a number of districts the deposits are clustered 
about several small granite outcrops close together, which suggest the 
presence of a larger granite body beneath. 

The silicate minerals are those usual in contact-metamorphic deposits, 
except that minerals carrying boron appear to be absent and magnetite 
and hematite are notably rare. 


720 REVIEWS 


The eccentric distribution of the deposits along some igneous contacts 
is correlated with the presence or absence of fractures which could 
serve as channels for the metamorphosing solutions, but other eccentrici- 
ties such as the presence of chunks of unaltered limestone in the intrusive 
are not readily explained. Where the intrusive rock is in contact with a 
large body of sediments the contact-metamorphic rocks may follow the 
contact rather regularly, or may replace certain beds in the sediments, 
or may follow fissures crossing them. Unlike many contact-metamorphic 
deposits which are notably compact, these commonly contain vugs 
some of which attain the dimensions of caves in which a man can stand 
upright. These are not the result of solvent action subsequent to 
metamorphism but were formed during metamorphism, and are charac- 
terized by crystals of quartz some of them a foot long projecting inward 
from their walls. 

Contact metamorphism has usually affected the igneous rock as 
well as the limestones, but the metamorphic zone is much narrower in 
the intrusive than in the limestone and is much more siliceous being made 
up mostly of quartz with subordinate dark silicates of the same varieties 
that characterize the adjacent altered limestone. A distinct zoning is 
usually perceptible in the alteration of the limestone. The zone nearest 
the intrusive is characterized by dark-colored iron-bearing silicates— 
iron-bearing garnet, epidote, pyroxene, hornblende, etc., with calcite and 
quartz and minor amounts of other minerals. This zone is the chief 
host of the scheelite. 

Beyond this zone comes a zone characterized by light-colored sili- 
cates poor in iron, tremolite and wallastonite being the commonest of these 
silicates. Colorless diopside, scapolite, colorless garnet, and other silicates 
are present in less abundance. Calcite is very abundant. Scheelite 
is absent. The zone of light-colored silicates grades outward into the 
marble that forms the outermost part of the contact-metamorphic aureole. 

Evidences are present that in many of the deposits the minerals were 
not deposited contemporaneously, but that one mineral followed another 
in a regular sequence. The details of the successive replacements were 
not worked out microscopically, but in general it appears that garnet 
was one of the first to be deposited and that sulphides were among 
the last. 

The authors propose the name “tactite”’ for the rocks and ores de- 
veloped by contact metamorphism nearest the intrusive (Latin, tactus, 
“touch’’?). This term seems well chosen and may prove a convenience. 
It should be noted that the authors designate as ‘“‘tactite”’ only the zone 


REVIEWS 721 


of dark silicates next the intrusive, and do not apply the term to the 
zone of light-colored silicates or to the marble. 

No evidence of enrichment through the action of meteoric waters 
was noted in any of the deposits. Leaching is shallow. 

The average tenor of the deposits worked near Bishop, California, 
is about 0.5 per cent of WO,; ore mined in the Mill City, Nevada, 


district averaged about 2 per cent of WO . 
E. S. BAsTIN 


Geology of the Non-metallic Mineral Deposits Other Than Silicates. 
Vol. I. Principles of Salt Deposition. By A. W. GRABAU. 
First edition. New York: McGraw-Hill Book Co., 1920. 
Pp. xvi+435, figs. 125. 


This volume is a treatise on applied stratigraphy. The author 
designates it as ‘a hand-book of salt-geology,” using the term salt to 
include phosphates, nitrates, borates, and similar deposits, as well as 
common salt. 

The author discusses first the sea as a source of saline deposits, basing 
a number of his conclusions on the careful work by Van’t Hoff on the 
famous Stassfurt deposits. This is followed by a well-illustrated chap- 
ter concerning the conditions by which sea salts are deposited in nature, 
emphasis being placed on their organic deposition. 

In the discussion of lagoonal deposits, it is concluded that many of 
the older salt deposits, generally believed to have been formed after 
the manner postulated by the bar theory, cannot be accounted for in 
this way, inasmuch as they are non-fossiliferous. Lagoonal deposits, 
such as those forming in the Karatugas Basin, contain fossils. 

A chapter is devoted to the classification of terrestrial salts, and the 
larger portion of the book is devoted to a discussion of their origin, method 
of concentration, and distribution. Throughout the book numerous salt 
lakes and salt basins of various types are described, and their deposits 
discussed. 

Attention is given to the mooted questions of the origin of nitrates, 
phosphates, and dolomites, and the several theories which have been 
advanced to explain them are stated. 

In the chapter on “Deformation of the Salt Bodies,” the author 
favors the view of Rogers that the salt domes of Louisiana and Texas are 
of exogenetic origin, but doubts whether all the characteristics of salt 
domes can be accounted for without the action of endogenetic forces. 


722 REVIEWS 


The last chapter is devoted to an interesting discussion of the ‘‘ Condi- 
tions of Salt Deposition in Former Geological Periods.”” In this chapter, 
and indeed throughout the book, attention is directed to the stratigraphic 
and paleontologic characteristics of the various types of deposits, the 
object being to prepare the reader for the stratigraphic chapters which 
are to follow in Volume II. 

The book is a valuable contribution to the science of salt deposits, 
and should prove of great service to students of economic geology. 


Wy Be 


Upper Cretaceous Floras of the Eastern Gulf Region in Tennessee, 
Mississippi, Alabama, and Georgia. By E.W.BERRy. United 
States Geological Survey, Professional Paper 112, Washington, 
IQIQ. 

The bulk of the flora is from the Tuscaloosa formation, and this paper 
is devoted principally to its elucidation and is to be regarded as a pre- 
liminary report, because Professor Berry’s work was of a reconnaissance 
nature. Later, when the abundant workable clays of the Tuscaloosa 
formation are developed for economic purposes, many new localities of 
fossil plants probably will be discovered and additional representatives 
of its flora. One hundred and fifty-one species, representing eighty-seven 
genera from the Tuscaloosa formation, are described. The three other sub- 
divisions of the Upper Cretaceous, the Eutaw formation, the Selma chalk, 
and the Ripley formation, contributed but meagerly to Berry’s collection. 

In comparing the Tuscaloosa flora with the European Upper Cre- 
taceous floras, Berry finds that eighteen identical species are found in 
the European Cenomanian, two in the Turonian, and four in the Se- 
nonian. ‘Therefore, the evidence of the floras is overwhelmingly in favor 
of the pre-Senonian and pre-Montana age of the Tuscaloosa and allied 
floras. Thirty Tuscaloosa species are found in the Atane beds of Green- 
land, and ten in the Patoot beds of the same island. The book is amply 


illustrated with plates. 
A.. Gag INe 


The Mineralization of the Copper Shales. By F. BEYSCHLAG. 
Zeitschrift fiir praktische Geologie. January, 1921. Pp. 1-16. 
In this article Dr. Beyschlag brings forward very cogent arguments 
in support of the view that the famous copper deposits of Mansfield, 
Germany, are not of sedimentary origin, but are younger than the rocks 


REVIEWS 723 


which inclose them and where deposited by solutions from magmatic 
sources and later enriched through the agency of descending solutions. 

The copper ores occur in three positions: First, in the highly bitu- 
minous and pyritic beds of the copper shale (Kupferschiefer); second, in 
the immediately underlying Zechstein conglomerate and the Weisslieg- 
ende—the latter forming the upper, bleached portion of the Rotliegende; 
third, in “‘rucken” or veins cutting any or all of these Permian sediments. 

The dominant metallic minerals in all three situations are chal- 
copyrite, bornite, and chalcocite. Minor components are pyrite, ga- 
lena, niccolite, safflorite (CoAs.), and some others. In the copper shale 
the ore is mainly the so-called ‘‘Speise,” a fine dust of sandlike ore par- 
ticles in the rock. Strings and thin plates of sulphides also occur usu- 
ally parallel to the bedding but elsewhere cutting it at small angles or 
uniting to form networks of sulphides. Bean and kidney-shaped masses 
of sulphides also occur. The microscope shows clearly that these sul- 
phides have developed by replacement of the shales. Nowhere are the 
sulphide masses rounded as might be expected if they were original 
components of a mechanical sediment. In the Zechstein conglomerate 
and Weissliegende the ore minerals characteristically form ‘“‘sand ore” 
1-3 cm. thick immediately beneath the copper shale. In the sand-ore 
spaces between the sand grains are occupied by chalcopyrite and more 
rarely bornite and chalcocite. Heavy impregnations grade into mere 
sprinklings of sulphides. In more clayey beds segregations and plate- 
like masses of sulphides occur. The ‘“‘rucken” or veins occupy faults of 
a few centimeters to 100 meters displacement cutting the sediments. 
Many oi them are richly mineralized while others are lean or barren. 
The maximum mineralization of the veins is in the vicinity of the copper 
shale and frequently the ore is confined to that part of the vein between 
the displaced portions of the copper shale. Two generations of ore 
minerals are recognizable in the veins, the older generation with nickel 
and cobalt arsenides, calcite, molybdenite, and rarely pitchblende, and 
the younger generation with copper sulphides and barite. 

If it is assumed that the ores in the copper shale and Zechstein were 
deposited on the sea bottom at the time these sediments were laid down, 
it is necessary to assume that copper-bearing solutions were contributed 
to the sea by streams from the land or by submarine springs. The land 
areas of that period contained no rocks which by weathering or other- 
wise can be supposed to have yielded copper-rich solutions. If sub- 
marine springs had been the source of copper-bearing solutions the ores 


724 REVIEWS 


in the sediments should presumably have been less uniform over wide 
areas than is actually the case. 

The author of the paper turns therefore to igneous sources for an 
explanation of the ores. He points out that the Permian intrusive rocks 
of Europe are very generally copper-bearing and attributes the primary 
deposition of copper in the Mansfield deposits to solutions coming from 
the same magmatic source that at an earlier period yielded the eruptive 
rocks that form part of the Rotliegende. Cupriferous mineralizing 
solutions, according to Beyschlag, ascended along the vein fractures 
and spread Jaterally through and beneath the copper shale depositing 
chalcopyrite in the sediments by the filling of pores and by replacement, 
the bitumen and pyrite in these beds being instrumental in the ore pre- 
cipitation. They also deposited chalcopyrite, cobalt, and nickel arse- 
nides, molybdenite and rarely pitchblende in the fault fractures. 

The writer points out that mineralization in the copper shale and 
Zechstein is greatest where veins cutting these formations are numerous 
and close together. 

Although Beyschlag does not emphasize particularly the genetic 
significance of primary mineral composition, to the reviewer the presence 
in the veins of cobalt and nickel arsenides, of molybdenite and of pitch- 
blende is highly suggestive of a magmatic origin for the mineralizing 
solutions. 

Of the primary ores developed by the processes just described only 
remnants now remain. Most of the primary ore minerals have been 
replaced by the rich copper minerals, bornite and chalcocite, through the 
agency, Professor Beyschlag believes, of descending meteoric waters— 
the familiar process of downward sulphide enrichment. As evidence of 
such enrichment, it is shown that much if not most of the bornite and 
chalcocite is a replacement of chalcopyrite and drawings are presented 
showing such replacements. Bornite is characteristically the first 
mineral to replace chalcopyrite; chalcocite then replaces bornite. Fur- 
ther, there is a marked falling off in the abundance of the rich copper 
minerals in depth. 

Of fundamental importance is the conclusion that the ores in the 
sediments and in the veins were formed contemporaneously, a conclusion 
based mainly on the restriction of ore in the veins to the vicinity of the 
ore-bearing beds and conversely the greater development of ore in the 
beds where they are cut by numerous and close-spaced veins. 

E. S. Bastin 


INDEX TO VOLUME XXX 


PAGE 


Adapting a Short-Bellows, Roll-Film Kodak for Detail Work in the 


Field. By Chester K. Wentworth .. 158 
Age of the Domes and Anticlines in the Lost Solaientercis Diceniet 
Wyoming hey Bye. Ee hath = 303 


Allen, R: C., and Barrett, L. P: Gontnnccan to ie Pre (Coleen 
ecoloxy: of Northern Michigan and Wisconsin. Review by. A. C. 


IWICH gees. co 1A: ‘ 85 
American Species of Onhophmiemiia aid Tepidocyeliias The. By 

ijasephyAc Cushimanss sReview by~Aa ©: Mich ..95) i enan en 200 
Ammonite Evolution, Aspects of Ontogeny in the Study of. By A. E. 

Trueman . A : : : ; : 5 WIC) 
Androscoggin River, Roemer Courses ai che. By Irving B. Crosby . 232 


Annotated Index of Minerals of Economic Value, to Accompany a 
Bibliography of Indian Geology and Physical Geography. An. 
Compiled by T. H. D. la Touche, M.A., F.G.S. Review by A. C. 


MCE 5 MMP ely Hak Arche pata ober ii Why ect Ry rca ya i 88 
Apus in the Permian of Oklahoma, On the Occurrence of an. By 

Rudolf Ruedemann . , Bie 
Arber, E. A. Newell. The Beier Mesoreie Bloras, of New Zealanel 

Review by A.C. McF. . ‘ 5 OST 
Aspects of Ontogeny in the Study of Aanimionite Bolen Pe Acris 

Truemann . : 140 
Aurousseau, M., and Waehin pron Henry Ss. The NEDRERFE Gyenite 

and Nephelite Porphyry of Beemerville, New Jersey. aa Ate 5 GAL 


Baker, Frank Collins. Pleistocene Mollusca from Northwestern and 

Central Illinois . : 43 
Barrows, Harlan H. Remiens of World tas of Commerc Geclonve 

Part I, Distribution of Mineral Production, by United States Geo- 

logical Survey . ‘ 650 
Bassler, R. S. Systematic Report on ‘ihe Cane, naa Ondeviigian ai 

Maryland. Review by A.C. McF. . ee ; odes, 0 
Bastin, E.S. Editorial . ; : OR 
Reviews . 83, 263, 268 idea aeOu) Alsi, AWA, Kolo), Gti, Wi), 7/23 
Paka, F. A. Growth-Stages of the Blastoid, Orophocrinus Stelliformis 73 
Beal, Carl H., and Lewis, J.O. Some Principles Governing the Produc- 

fHiom of Oil Wells, IRewigwia7 WeOsG, =. 5 Gk A 


725 


726 INDEX TO VOLUME XXX 


PAGE 


Beger, David B. Geology of Webster County and a Portion of Mingo 
District, Randolph County, South of Valley Fork of Elk River. 


Revewsby AiG. Mehr seen ime. 2 uf a See 
Behavior of Inclusions in Igneous Magmas, ae. By N. L. Bowen) Saaeeee 
Behre, Charles H., Jr. Petrological Abstracts and Reviews . . | AO ois 
Benson, W. N. An Outline of the Geology of New Zealand. .. I 
Berry, Edward W. Marine Upper Cretaceous and a New me 

from the altaplaniciae of Bolivia : 227 


Upper Cretaceous Floras of the asian Gulf Regan in 
Tennessee, Mississippi, Alabama, and Georgia. ReviewbyA.C.N. 722 
Beyschlag, F. The Mineralization of the Copper Shales. Review by 


BeSe Beek 3 .. eee 
Blastoid, Orophocrinus Silane. Growin Sages of ‘he: By Ae 
Bather. . , : ; 73 
Bowen, N. L. The Behavior Re Thelusienes in Tmieous Mauia /«., Mae 
The Reaction Principle in Petrogenesis . 177 
Bridge, Josiah, and Charles, B. E. A Devonian Outliers near ane Crest 
of the Ozark Upliit] ss. 450 


Brooks, Alfred H., and LaCroix, Mons F. Ay te Ton Asta Assactared 
Industries of Bomane. the Sarre District, Luxemberg, and Belgium. 


Review, by €yhy Bsr ce. o, |\ hee eg 
Brown, John S. Fault Rea eures of Saon Peasy Calbinate Me ir 23 
Bruce E. L. The Early Pre-Cambrian Formations of Northern Ontario 

and Northern Manitoba. y Eh ace 
Bryan, Kirk. The Hot Water Supply ise ie Hot Sarno Aveamen® . 9s 
Bulletin No. 36, Illinois State Geological Survey. ReviewbyA.C.McF. 330 
Burwash, E. M. The Pre-Cambrian of Western Patricia : aia 
Cambrian and Ordovician of Maryland, The. By R. S. Bassler. 

Newlewalbyaser © nc haves nee 9 he SAT 
Cameron, A. E. Post-Glacial Lakes in fhe Mackenzie River Basin, 

Northwest Territories, Canada. 3 337 
Canu, Ferdinand, and Bassler, Ray S. North Menenenn Early Tenia 

Bryozoa. Review by A.C. McF.. . 261 


Case, E. C., and Robinson, W.I. The Geology Gi Dimiéstoae Mount 
and Ghenuan Hill in Houghton County, Michigan. Review by 


ACS NiGE sana: 85 
Chadwick, G. H. The Baleorcie Rocks of “the Canton Quaduamers 
Rewmewbyievk ss Were Rae ir af Gs 
Chamberlin, T. C. Memorial Bichronal-“R olin ID). Salicoueyd d . 480 
Review of Life of James Hall, Geologist and Paleontologist, 
1811-1898. By J. M.Clarke. . 175 


Clarke, J. M. Life of James Hall, (eeoleeien itl Paleontalones 1Btr 
ToOS., Aneview by, dua ©. Cen 28d [it 85 : : : 2 Seas 


INDEX TO VOLUME XXX 


Clastic Sediments, a Scale of Grade and Class Terms for. By Chester 
K. Wentworth . : 4 : 4 4 

Coal. By Elwood S. Moore. Review by C. R.S. 7, 

Coal in 1918. Part A: Production. By C. E. Lesher. Review by 
Has. B.. L to 

Coal, Oil, Limestone, ane resin Ore Man NeW Ecition ee By West 
Virginia Geological Survey. Author’s Abstract. 

Colby, Charles C. Source Book for the Economic Graapanciy of North 
America. Review by R. D.S. ; 

Collins, W. H. An Outline of the Physiogaphic isuoss of North- 
eastern Ontario . : 

Contact-Metamorphic auinasten enositey in fhe Waited States! By 
Frank L. Hess and Esper S. Larsen. Review by E. S. B. 

Contact Phenomena between Gneiss and Limestone in Western 
Massachusetts, On. By Pentti Eskola 

Contributions to the Pre-Cambrian Geology of Noeenernt MaenieAn anal 
‘Wisconsin. By R. C. Allen and L. P. Barrett. The Geology of 
Limestone Mountain and Sherman Hill in Houghton County, 
Michigan. By E. C. Case and W. 1. Robinson. Review by A. C. 
MceF. 

Cooke, H. C. Geology i the Matachewan Tiserict: Northern Outan! 
Review byz@. HB. Jr... 

Copper Deposits of ayy and nant evoar The. By ieredericl 
Leslie Ransome. Review by E. S. B. and C. H. B. 

Copper Shales, The Mineralization of the. By F. Raraikie, Revian 
by E. S. B. : 

Cox Ga. Dake, C.L., and Miuilenbure. G. is veld Methods in 
Betrolesm Geology. iReew by W. O. G. , 

Cressey, George B. Notes on the Sand Dunes af Notniestenn maine 

Crinoid Genus Scyphocrinus and Its Bulbous Root Camarocrinus, On 
the. By Frank Springer. Review by A. C. McF. 

‘Criticism of the ‘‘Faunal Relationships of the Meganos Crain? by 
Bruce L. Clark, A. By Roy E. Dickerson . 

Crosby, Irving B. Former Courses of the Androscoggin Ger 
Crystallization and Magmatic Differentiation of Igneous Rocks, The 
Physical Chemistry of the. V. By J. H.L. Vogt . : 
Crystallization and Magmatic Differentiation of Igneous Rocks, T ne 

Physical Chemistry of the. VI. By J. H. L. Vogt. 
Crystobalite in California, A New Occurrence of. By Austin F. Rogers 
Cushman, Joseph A. The American Species of Orthophragmina and 
Lepidocyclina. Review by A. C. McF. a eta te 


Dake, Charles Laurence. The Problem of the St. Peter Sandstone. 
evaewa Wyse ese thly fer) to), el! oan 


727 


PAGE 


SL 
717 


264 
418 

82 
199 
719 


265 


85 
424 
325 
723 


332 
248 


259 


295 
232 


611 


659 


2I1 


260 


646 


728 INDEX TO VOLUME XXX 


Davies,A. Morley. An Introduction to Paleontology. Review by C.L.F. 

Deposits of Manganese Ore in Arizona. By Kok. Jones) jr, ands ele 
Ransome. Review by C. H. B., Jr. si CEES Soe 

Deposits of Manganese Ore in Costa Rica and Panama. By Julian D. 
Sears. Review by C. H. B., Jr. : 

Deposits of Manganese Ore in the Bateeile eee Apianere 
Preliminary Report on the. By Hugh D. Miser. Review by 
Cy Hess air é ive | a 

Deposits of Ton Ore near Stanford Montana: By L. G. Westgate. 
Review by E. S. B. 

Description geométrique des Alpes Franeaisest Rarer da Tonle Seon 
By P. Helbronner. Review by W. H. Hobbs 

Detailed Report on Nicholas County [West Virginia.] By Dana B. 
Reger. Author’s Abstract 

Devonian Outlier near the Crest of the Ozark Uplift, A. By Josiah 
Bridge and B. E. Charles 

Dickerson, Roy E. A Criticism of ans conan Rel aeneeeee of the 
Meganos Group” by Bruce L. Clark $y) eee 

Dinosaur Tracks in Hamilton County, Texas. By W. E. Wrather 

Domes and Anticlines in the Lost Soldier-Ferris District, Wyoming, 
DheyAge ot the, “By A, E. Kath): 


Earlier Mesozoic Floras of New Zealand, The. By E. A. Newell Arber. 
Review by A. C. McF. ‘ 

Early Pre-Cambrian Formations of N: mere Omens fad ‘Northen! 
Manitoba, The. By E. L. Bruce . 

Editorial. By E.S. B. 

Editorial, Memorial. Rollin D. Salisbury. “By T. C. oo : 

Economic Aspects of Geology, The. By C.K. Leith. Review by Edson 
S. Bastin 

Economic Geography of North Navaaten, icounce Bool foe, the By 
Charles C. Colby. Review by R. D. S. 


Eleventh Biennial Report of the State Geologist on the Wines Td 


tries and Geology of Vermont. By George H. Perkins, e¢ al. 
Review by A. C. McF. : 

Eskola, Pentti. On Contact Phenomena between Gueies and Limestone 
in Western Massachusetts . . -«. 


Fairchild, Herman L. Pleistocene Submergence of the Hudson, 
Champlain and St. Lawrence Valleys. Review by J. F. W. 

Fath, A. E. The Age of the Domes and Anticlines in the Lost Soldier- 
Ferris District, Wyoming aa : ae 

Fault Features of Salton Basin, Galion By John S. Brown . 

Fault Troughs of the Antilles, The Great. By Stephen Taber 


PAGE 


651 
648 


653 


501 
335 
717 
419 
450 


295 
354 


393 


657 
459 
631 
480 


511 


82 


86 


265 


504 


303 
217 
89 


INDEX TO VOLUME XXX 


“Faunal Relationships of the Meganos Group” by Bruce L. Clark, 
A Criticism of the. By Roy E. Dickerson . 

Fay, Albert S. A Glossary of the Mining and Mineral iieeustan 
Review by E. S. B. rae 

Ferguson, Henry G. The Morollen iDjsintios iNew Mexicans Review 
byaGw El. Be Jr 

Field Methods in Petroleum Geclogy: By G. ig Con, C. L. Dake. ad 
G. A. Muilenburg. Review by W. O. G. Sha ek (ae 

Former Courses of the Androscoggin River. By Irving B. Crosby 

Foye, Wilbur G. Origin of the Triassic Trough of Connecticut 

Fraser River, British Columbia, Canada, The Character of the Stratifica- 
tion of the Sediments in the Recent Delta of. By W. A. Johnston 


Gardner’s Theory of the Origin of Certain Concretions, In Support of. 
By Leroy Patton 

Genesis of the Zinc Ores of Bamards Diasec, “St. ‘Lawrence County: 
New York. By C. H. Smyth, Jr. Review by J. F. W. i 

Geological Survey of Western Australia. Maps and Sections to Accom- 
pany Report on Contributions to the Study of the Geology and Ore 
Deposits of Kalgoorlie, East Coolgardie Goldfield. Review by 
va (Ce MIE ifiepe pat & yeah, | eee LI Re 

Geological Survey of Western Aereteait. Maps and Sections to Accom- 
pany Report on the Geology and Mineral Resources of the Yilgarn 
Goldfield. Review by A. C. McF. NY ate ata ea 

Geological Survey of Western Australia. Maps and Sections to Accom- 
pany Report on the Geology and Ore Deposits of Meekatharra, 
Murchison Goldfield. Review by A. C. McF. 

Geology and Mineral Deposits of a Part of Amherst Towshin. Ouches 
By M. E. Wilson. Review by J. F. W. 

Geology of Hardin County and the Adjoining Part of Bowe Carma, The. 
By Stuart Weller with the collaboration of Charles Butts, L. W. 
Currier, and R. D. Salisbury. Review by A. C. McF. i 

Geology of New Zealand, An Outline of the. By W.N. Benson . 

Geology of Northeastern Rajputana and Adjacent Districts, The. By 
A. M. Heron. Review by A. C. McF. : ihe gare a 

Geology of the Matachewan District, Northern Onerie IByy Jaly (Ce 
Cooke. Review by C. H. B., Jr. . 

Geology of the Non-Metallic ivEneral IDEDORite giher cher SHlfteaties: 
Vol. I. Principles of Salt Deposition. By A.W. Grabau. Review 
lon? Vie 1, 18. : 5 

Geology of the Tuapeka Deeriet Cael Otago IDeyietons The. By 
P. Marshall. Review by A. C. McF. : 

Geology of the Yellow Pine Cinnabar Teenie Tea! By E. 6 Renee 
and E. C. Livingston. Review by E. S. B. ai ige 


729 


PAGE 
295 
263 
420 
332 
232 


690 


TES 


700 


418 


86 


84 


84 


88 


263 


730 INDEX TO VOLUME XXX 


Geology of Webster County and a Portion of Mingo District, Randolph 
County, South of Valley Fork of Elk River. By David B. Beger. 
Review by A. C. McF. : oy ue oo) soa eae 

Glacial Loess Accumulation, The Time se in its Relation to the 
Climatic Implications of the Great Loess Deposits: Did They 
Chiefly Accumulate during Glacial Retreat? By ais mas 
Visher . : : 

Glenk, Robert. Lenina ipo. Rarer by Cc H. B. ip. 

Glossary of the Mining and Mineral Industry, A. By Albert Ee Fay. 
Review by E. S. B. 

Goldman, Marcus I. lnfhalopie culieutface Corelaionk in the “Bend 
Series’? of North Central Texas. Review by C. H. B., Jr. 

Grabau, A. W. Geology of the Non-Metallic Mineral Deposits other 
than Silicates. Vol. I. Principles of Salt Deposition. Review 
byaWWeee be : : 

Grade and Class Terms for Glastie Sediments. A Scale Be By Chester 
K. Wentworth . Cabs Plat Mane 

Great Fault Troughs of the aeralles The. By Stephen Taber 

Gypsum Deposits of the United States. By R. W. Stone. Review by 

Gypsum in 1919. By R. W. Stone. Review by E. S. B. 


Harder, Edmund Cecil. Iron Depositing Bacteria and Their Geological 
Relations. Review by C. H. B., Jr. : 

Hayes, A.O. The Malagash Salt Dyseaety Cambered Gounies Nova 
Scotia. Review by J. F. W. ; ; 

Helbronner, P. Description penmerrate des Alpes Franieaiceet Anners 
du Tome second. Review by W. H. Hobbs 0 eee 

Helium-Bearing Natural Gas. By G. Sherburne Rogers. Review by 
E. S. Bastin 

Heron Ae Mew bs See) he G. S: ‘The Gealoey, of Nort Weneronn eapanticnn 
and Adjacent TORSEiCtS Review by A. C. McF. : 

Hess, Frank L. Tungstenin1918. Review by E.S. B. and C. H. Boe 

Hess, Frank L., and Larsen, Esper S$. Contact-Metamorphic Tungsten 
Deposits in the United States. Review by E. S. B. 

Hobbs, W.H. Review of Description geométrique des Anes Franeaest 
Annexe du Tome second, by P. Helbronner j 

Hot Water Supply of the Hot Springs, Arkansas, The. By Karle Bae 


Igneous Rocks, On the Representation of, in Triangular Diagrams. By 
Albert Johannsen 

Igneous Rocks, The Berea Chemisty, of tite Crystallivation und 
Magmatic Differentiation of. V. By J. H. L. Vogt 

Igneous Rocks, The Physical Chemistry of the Crystallization and 
Magmatic Differentiation of. VI. By J. H. L. Vogt 


PAGE 


329 


472 
262 


263 


333 


167 


659 


INDEX TO VOLUME XXX 731 


Inclusions in Igneous Magmas, The Behavior of. By N.L. Bowen . 513 
Informational Corrugated Rocks. By William J. Miller. . . . 587 
Introduction to Paleontology, An. By A. Morley Davies. Review by 

Cr AS ies 651 


Tron and Associated Tadueties of Negeenne ihe Sarre ister: Tse 
burg, and Belgium, The. By Alfred H. Brooks and Morris F. 


itaCroix, Review byCob..Bs Jr. iia. 4 Ae Lehn ERIN OT 
Iron Depositing Bacteria and Their Geological Rela norst By Edmund 

SeciltardermnNevicways Colle De-air ae a mills let ii tale n nse EGG 
Jack, R. Lockhart, B.E., F.G.S. The Phosphate Deposits of South 

Australia. Review Be CeMch = Ono 176 
Johannsen, Albert. On the Representation of Tenens Rocks in IB 

angular Diagrams .. ar hi tere eee ay «eu ste ONE 7 
Johannsen, Albert, and Charles H. Beer! ie pe irolowieal Abstracts 

and Reviews . . 170, 252, 319, 401, 482, 632, 703 
Johnson, W. E. Map of the North Baste Review. 420 
Johnston, W. A. The Character of the Stratification of the Sediments 

in the Recent Delta of Fraser River, British Columbia, Canada . 115 
Jones, E. L., Jr., and Ransome, F. L. Deposits of Manganese Ore in 

Arizona. Review by C. H. B., Jr. Be ke cer ESRC eh CaN te oR OS 


Kalgoorlie, East Coolgardie Goldfield, Maps and Sections to Accompany 
Report on Contributions to the Study of the Geology and Ore 
Deposits of. By Geological Survey of Western Australia. Review 


Loar Nn Oa LCE: ett isn v ch cet uci ant ine her ih gies hie Mike een ga ples 86 
Lane, Alfred C. Segregation Granites. . 162 
Larson, E. S., and Livingston, E. C. colony of the svellon pine 

Cinnebar Dyeaeiee Idaho. Review by E. S. B. ‘ 263 


la Touche, T. H. D., M.A., F.G.S. (Compiler). 1a emginted Index 
of Minerals of Economic Value, to Accompany a Bibliography of 
Indian Geology and Physical Geography. Review by A. C. McF. 88 
Leith, C.K. The Economic Aspects of rina Review by Edson S. 


Bastin. ; hee pelvis? 
Lesher, C. E. Coal in rane. Part A: peodcton! Review by 

BS Be. ; 264 
Life of James Hall, Geologist and Paleontologist, tone Taos. By l. M. 

Clarkes” Reviews byeiaiey Ce: 175 
Limestone and Phosphate Resources of New Zealand The. Bare L 

Limestone. By P. G. Morgan. ReviewbyE.S.B. .  . 83 


Limestone Mountain and Sherman Hill in Houghton County, Neha! 
The Geology of. By E. C. Case and W. I. Robinson. Review 
Payer Are GRICE rset ayia her ees hs DIT iatea at, aes tne at Oe Yl aae a Ai 85 


732 INDEX TO VOLUME XXX 


Lithologic Subsurface Correlation in the ‘‘Bend Series” of North 
Central Texas. By MarcusI. Goldman. Review by C. H. B., Jr. 

Longwell, Chester R. The Muddy Mountain Overthrust in South- 
eastern Nevada 

Lost Soldier-Ferris District, ‘Waornine: The Nee af the Domes atl 
Anticlines in the. By A. E. Fath. 

Louisiana Lignite. By Robert Glenk. Review by C. ig B. ae 


MacCarthy, Gerald R. Mud Cracks on Steeply Inclined Surfaces 

MacClintock, Paul. The Pleistocene History of the Lower Wisconsin 
River . Ap) tee iat i ee 4 : : : 

McLennan, J. C., and ecmeeriee Report on Some Sources of Helium 
in the eiiln Empire. Review by E. S. B. 

Magnetite Deposits of Grenville District, Argenteuil Counce, Guenee 
By M. E. Wilson. Review by J. FW. )) |. SH 

Malagash Salt Deposit, Cumberland County, Nova seater The. By 
A. O. Hayes. Review by J. F. W. ; > oe 

Manganese Ore in Arizona, Deposits of. By E. L. Jones, Jr., and 
F. L. Ransome. Review by C. H. B., Jr. 

Manganese Ore in the Batesville Diss, Arlamens, Prelmceee 
Report on the Deposits of. By Hugh D. Miser. Review by 
@ He Ber ; 

Map of the NerH Bache: By W. 1D; Johnsons Review: 

Maps and Sections to Accompany Report on Contributions to the Sandy: 
of the Geology and Ore Deposits of Kalgoorlie, East Coolgardie 
Goldfield. By Geological Survey of Western Australia. Review 
by A. C. McF. 

Maps and Sections ‘oS Accompany Reporee on fie Gecleey and Manoel 
Resources of the Yilgarn Goldfield. By Geological Survey of 
Western Australia. Review by A. C. McF. 

Maps and Sections to Accompany Report on the Geolony. and Ore 
Deposits of Meekatharra, Murchison Goldfield. By Geological 
Survey of Western Australia. Review by A. C. McF. 


Marine Upper Cretaceous and a New Echinocorys from the Altaplame 


of Bolivia. By Edward W. Berry. 

Marshall Re MiPAG DESC: slianG: Sam ulune Geoloen of fhe al aapele 
District, Central Otago Division. Review by A. C. McF. : 

Martin, James C. The Pre-Cambrian Rocks of the Canton Quadrangle. 
Review by J. F. W. . 

Meekatharra, Murchison Goldfield, Maps cra Sections to Accompane 
Report on the Geology and Ore Deposits of. By Geological 
Survey of Western Australia. Review by A. C. McF. 

Mehl, Maurice G. A New Phytosaur from the Trias of Arizona . 

Memorial Editorial—Rollin D. Salisbury. By T. C. C. 


PAGE 
333 
63 


393 
262 


702 
673 
4II 
503 

82 
648 


501 
420 


86 
84 


84 
227 
88 
416 
84 


144 
480 


INDEX TO VOLUME XXX 


Miller, William J. Intraformational Corrugated Rocks . aie: 

Mineral Resources of Michigan for 1914 and Prior Years. Prepared 
under the direction of R. C. Allen. With a treatise on Michigan 
copper deposits by R. E. Hore. Review by D. J. F. ; 

Mineral Resources of Michigan for 1917 and Prior Years. Prepared 
under the direction of R. C. Allen. Review by D. J. F. . 

Mineral Resources of the Philippine Islands for the Years 1917 and 
1918, The. Contributors: Elmer D. Merrill, Victoriano Elicano, 
Leopoldo A. Faustino, W. H. Overbeck, and T. Dar Juan. Review 
loyy (Cola B es re 

Miser, Hugh D. Brelieamery, Reade on ‘the menace of Mianpanese 
Ore in the Batesville District, Arkansas. Review by C. H. B., Jr. 

Mogollon District, New Mexico, The. By Henry G. Ferguson. Review 
bya Cr EB jr 

Moore, Elwood S._ Coal. Rewiew by C. 2. Sis 5 

Moore, Raymond C., and Plummer, Frederick B. Pennsylvanian 
Stratigraphy of Nore Central Texas 

Morgan, P. G. The Limestone and Phosphate Resources of N ew 
Zealand. Part I. Limestone. Review by E. S. B.. ‘ 

Mud Cracks on Steeply Inclined Surfaces. By Gerald R. Mac @arthy, 

Muddy Mountain Overthrust in Southeastern Nevada, The. By 
Chester R. Longwell 


Nephelite Syenite and Nephelite Porphyry of Beemerville, New Jersey, 
The. By M. Aurousseau and Henry S. Washington 
New Edition of Coal, Oil, Gas, Limestone, and Iron Ore Map. Bay 
West Virginia Geological Survey. Author’s Abstract A 
New Zealand, An Outline of the Geology of. By W.N. Benson . 
Non-Metallic Mineral Deposits other than Silicates, Geology of the. 
Vol. I. Principles of Salt Deposition. By A.W. Grabau. Review 
by W. T. B. Bk Sa ee er a a 
North American Early Ber tary avon By Ferdinand Canu and 
Ray S. Bassler. Review by A. C. McF. 


O’Harra, Cleophus C. The White River Badlands. Review by 
ReaD S: : sie) Sie 

Ontogeny in the Study of Ammonite EMoluion: Agneeis a By Avs 
Trueman : ‘ 

Origin of the Triassic T Pomel of Connecticut: By Wilbur G. ipoye 

Orophocrinus Stelliformis, Growth-Stages of the Blastoid. By F. A. 
Bather : 

Outline of the TBherctiomaaaltie aketares af NGitheestenn Ontario! ree By 
W. H. Collins Bd Stic See Wer cess 


733 
PAGE 
587 


332 


332 


652 
501 


420 
717 


702 


261 


33t 


140 
690 


73 


199 


734 INDEX TO VOLUME XXX 


Paleontology, An Introduction to. By A. Morley Davies. Review by 


PAGE 


CaLeE. : SOM ha hidecer es iidy ; (OS ee 
Paleozoic Bituminous Coals:  Serceire in. By Reinhardt Thiessen. 

Review by A. C.N.. og AS ea 
Paleozoic Rocks of the Canton Gunde “The. By G. H. Chadwick. 

Review by J. F.W.. . 415 
Patton, Leroy. In Support of Garduers mbeary, of the Orieits of 

Certain Concretions . «8 Tr 
Pennsylvanian Stratigraphy of N erie Géatral eras By Raymond C. 

Moore and Frederick B. Plummer 18 
Perkins, George H., ef al. Eleventh Biednial Resare of the State 

Geologist on the Mineral Industries and Geology of Vermont. 

Review by A. C. McF. a Mata 7th RE OES 86 
Petrogenesis, The Reaction Principle in. By N. L. Bowen. Uhr) 
Petrological Abstracts and Reviews . . 170,252,319, 401, 482, 632, 703 
Physical Chemistry of the Crystallization and Magmatic Differentiation 

of Igneous Rocks, The. V. By J. H. L. Vogt . 611 
Physical Chemistry of the Crystallization and Magmatic Wiferenndeen 

of Igneous Rocks, The. VI. By J. H. L. Vogt 659 
Physiographic History of Northeastern Ontario, An Outline af the. 

By W. H. Collins 199 
Phosphate Depostis of South eeeralics The. Ry R. Mectnen Tea, 

B.E., F.G.S. Review by A. C. McF. 176 
Phytosaur from the Trias of Arizona, A Non By Viauniee G. Mehl st eee 
Pleistocene History of the Lower Wisconsin River, The. By Paul 

MacClintock ‘ 673 
Pleistocene Marine Submergence ae ihe isniclesia, Chama and St. 

Lawrence Valleys. By Herman L. Fairchild. Review by 

J. F. W. £504 
Pleistocene Mollusca kom Nothivetaen and Conca ior By 

Frank Collins Baker. : ae 43 
Possible Silurian Tillite in Shurieacter Prieta Columbia. By Francis 

P. Shepard : Pe pra) 
Post-Glacial Lakes in the Minelenrie ues Bastin, INoriaeee eminent 

Canada. By A. E. Cameron. ge Re ea eae eae 7) 
Potash Recovery at Cement Plants. By Alfred W. G. Wilson. Review 

by E. S. B. : : 412 
Pre-Cambrian of Western Baers The. By E. M. Barvace 393 
Pre-Cambrian Rocks of the Canton Quadrangle, The. By James C. 

Martin. Review by J. F. W. , 416 
Principles Governing the Production of Oil Wells, Some By Carl H. 

Beal and J. O. Lewis. Review by W. O. G. 264 
Problem of the St. Peter Sandstone, The. By Charles Tinbenes Dake, 

Review by R. D.S.. 646 


INDEX TO VOLUME XXX 735 
Problems in Stratigraphy along the Rocky Mountain Trench. By hes 
Francis Parker Shepard . Des Mean wn Re 0 
Proceedings of the Coal Mining aeticute of AmenGs: Review by 
Jdjs Sy 1B ‘ AIL 
Purcell Range and the Roce Mionarains 6 Cnnedla. The Saree 
Relation of the. By Francis Parker Shepard 130 
Ransome, Frederick Leslie. The Copper Deposits of Ray and Miami, 
Arizona. Review by E. S. B. and C. H. B. Bias 
Reaction Principle in Petrogenesis, The. By N. L. Bowen 177 
Reger, David B. Detailed Report on Nicholas County ee ae 
Author’s Abstract ose A EO 
Report on Mining Operations in the Brovanee of @ucbee: 1919. Review 
by E. S. B. La epee SSS 
Report on Some Sources of ietaiena § in ete British aires ‘ES é (Ge 
McLennan and Associates. Review by E. S. B. A4II 
Reviews. : An UE Dirh. Bos. Alii. (sone, a. 717 
Rocky Mountain rence probleme in Stratigraphy along the. By 
Francis Parker Shepard . 361 
Rogers, Austin F. A New Occurrence @ Gry cronaliren in (Canina 211 
Rogers, G. Sherburne. Helium-Bearing Natural Gas. Review by 
E. S. Bastin : 509 
Rollin D. Salisbury— Memorial Batarell By T. C. CG: 480 
Ruedemann, Rudolf. On the Occurrence of an Apus in ile Permian of 
Oklahoma . 311 
St. Peter Sandstone, The Problem of the. By Charles Laurence Dake. 
Review by R. D.S. 646 
Salisbury, Rollin D Bere morial Bdiveriall ee T. C. ©. 480 
———. Review of Source Book for the Economic Geography of North 
America. By Charles C. Colby 82 
Review of the Problem of the St. Peter cancion By (harles 
Laurence Dake. : ; ; : . 646 
Review of the White ewes Badass By Cleophus C. 
O’Harra ; ; ; LEME 
Sand Dunes of Mowdmesion ineiane Nate ¢ on the. By George B. 
Cressey 248 
Scale of Grade and Glass een foe Clastic seamen A. By Chess 
K. Wentworth . ; 377 
Sears, Julian D. Deposits of iViameanese Heh in Cis Ree ane Banana 
Review by C. H. B., Jr. . 653 
Sediments in the Recent Delta of Taner Raver jones Canis, 
Canada, The Character of the Stratification of the. By W. A. 
Johnston II5 


736 INDEX TO VOLUME XXX 


Segregation Granites. By Alfred C. Lane . 

Shepard, Francis P. Possible Silurian Tillite in Southeastern Brith 
Columbia . 

Problems in Sraieriehy lone ae Roly Mountain Tenge 

The Structural Relation of the Purcell Range and the Rocky 
Mountains of Canada : 

Short-Bellows, Roll-Film Kodak for Deru Wore in the Field, “Adaprime 
a. By Chester K. Wentworth : 

Smyth, Jr., C. H. Genesis of the Zinc Ores i Edwards Diceied St. 
Lawrence County, New York. Review by J. F. W.. oe 

Source Book for the Economic Geography of North America. By 
Charles C. Colby. Review by R. D. S. ; 

Springer, Frank. On the Crinoid Genus Scyphocrinus and Its Bulborte 
Root Camarocrinus. Review by A. C. McF. Pe : 

Stone, R. W. Gypsum Deposits of the United States. Review by 
1g 12s Se 3 

———. Gypsum in rgIo. Review by E. S. 2, 

Stratification of the Sediments in the Recent Delta of Freee Ragen 
British Columbia, Canada, The Character of the. By W. A. 
Johnston : 

Stratigraphy along the ROC Mountain Tien, "problema in. ene 
Francis Parker Shepard . 

Structural Relation of the Purcell Range sine tlie Rocky: Mountains ci 
Canada, The. By Francis Parker Shepard 

Structure in Paleozoic Bituminous Coals. By Revel “hisses 
Review by A. C.N.. 

Systematic Report on the Canaan antl Ordovieul of Marylameh 7B 
R. S. Bassler. Review by A. C. McF. 


Taber, Stephen. The Great Fault Troughs of the Antilles : 
Thiessen, Reinhardt. Structure in Paleozoic Bituminous Coals. 
Review by A. C. N. 


Timiskaming County, Quebec. By M.E. Wilson. Review by J. F. W. 


Triassic Trough of Connecticut, Origin of the. By Wilbur G. Foye 

Trueman, A. E. Aspects of Ontogeny in the Study of Ammonite 
Evolution . : 

Tungsten Deposits in the Waited States, Coming: Metamtona nicl By 
Frank L. Hess and Esper S. Larsen. Review by E. S. B. 

Tungsten in 1918. By Frank L. Hess. Review by E. S. B. and 
CAH Bair: 


United States Geological Survey. World Atlas of Commercial Geology; 
Part I, Distribution of Mineral Production. Review by Harlan 
H. Barrows. 


PAGE 
162 


77 
361 


130 
158 
418 

82 
259 


327 


413 


140 
719 


423 


650 


INDEX TO VOLUME XXX 


Upper Cretaceous Floras of the Eastern Gulf Region in Tennessee, 
Mississippi, Alabama, and Georgia. By E. W. Berry. Review 
by A. C. N. a te Ne es Wines Ge eel 


Visher, Stephen Sargent. The Time of Glacial Loess Accumulation 
in Its Relation to the Climatic Implications of the Great Loess 
Deposits: Did they Chiefly Accumulate during Glacial Retreat? . 

Vogt, J. H. L. The Physical Chemistry of the Crystallization and 
Magmatic Differentiation of Igneous Rocks. V 
VI 


Weller, Stuart, with the collaboration of Charles Butts, L. W. Currier, 
and R. D. Salisbury. The Geology of Hardin County and the 
Adjoining Part of Pope County. Review by A. C. McF. ; 

Wentworth, Chester K. Adapting a Short-Bellows, Roll-Film Kodak 
for Detail Work in the Field . 

———. A Scale of Grade and Class Terms for Cast cecimente! 

Westgate, L. G. Deposits of Iron Ore near Stanford, Montana. 
Review by E. S. B. 

West Virginia Geological SHinrest ‘New Beition ai Coal Oil, limestone: 
and Iron Ore Map. Author’s Abstract 

White River Badlands, The. By Cleophus C. O’ Ears Review be 
IRs IDs Sb : RIG a as 

Wilson, Alfred W. G. Potash Recowasy at Gement elemte Review by 
ESS. Bs : 

Wilson, M. E. Cases rat Manecall Deshet: aie a Burnt af Aenberst 
Township, Quebec. Review by J. F. W. : 

—— . Magnetite Deposits of Grenville District, pierenteuil Conner 
Quebec. Review by J. F. W. 

———, Timiskaming County, Quebec. Review be J. F. W. 

World Atlas of Commercial Geology; Part I, Distribution of Mineral 
Production. By Unites States Geological Survey. Review by 
Harlan H. Barrows . 

Wrather, W. E. Dinosaur T ee in Hamilton county Texas 


Yilgarn Goldfield, Maps and Sections to Accompany Report on the 
Geology and Mineral Resources of the. By Geological Survey of 
~ Western Australia. Review by A. C. McF. 


Zinc Ores of Edwards District, St. Lawrence County, New York, 
Genesis of the. Review by J. F. W. 


UE 


PAGE 


722 


472 


611 
659 


257 


158 
377 


335 
418 
331 
412 
419 
503 


413 


650 
354 


84 


418 


ORLD -ROMIC SYSTEM 
_ MASTERKEY TO ALL LANGUAGES 


| a IMERS, pated or selected it RoBeRT Morris PIERCE, 16 languages, $1.94 postpaid, each 
inguage: Arabic, Chinese, Danish, Dutch, English, Scotch English, French, German, Italian, 
Beier epee Polish, Portnuuese: Russian, Spanish, Swedish. 


Re ty Coe 


Vou, nosed: Thuy combene opine | La Maison 
DLE, we 2p Oamne. jpouiowhy } | 
purceanals | 


Professor of Romance Languages in Harvard University 


PIERRE Be | \epae 


Volume 1917-2 Price 36c (1s 6d) | 


BIBUO HIVE ple be 
Volume 19151 ix Ge 


TEXTES PHONIQUES 


Spanish ‘Alphagam 
Pierre: a batie 


Opes meus Pierce 
Phowene Asoc 


Lani GUAGES Posuisnina Company 
US Warr (710 Srauer, New-Your 


CHARTS, by Poeear Morris Prerce: four, each measuring (paper-size) 24 by 15 inches, and 

entitled respectively: Chart of the Universal Aiaean Chart of the Organs of Speech, Chart of 
the French Alphagam, Chart of the Spanish Alphagam: folded and bound in paper 62 by 4% inches, 
. 37 cents each, postpaid; mounted on cloth a rollers, for use as wall-charts, $3.88 each, postpaid. 


¢ 

DICTION ARIES, with pronunciations in World Romic, edited or selected by RoBERT Morris 
PIERCE: (unilingual) English, French, or German; (bilingual) French-English and English-French: 
$4. 86 each, postpaid. 


: PRONUNCIATION. TABLES, special alphagams, showing, in World Romic, the elementary 
: _ sounds of 89 languages, edited by ROBERT Morris PIERcE, 30 cents postpaid, each language: 
: Afghan, Albanian, Old Armenian, Bearnese (Arrens), Renee (Garlin), Bohemian, Bragancian, 
Brahui, Burgundian, Cantonese, Catalan, Charentese, Cumbrian, Dauphinese, Dubian (Bournois), 
_Dubian (Pontarlier), Dutch, Bilan (Badi), Favelien @aeee Emilian (Novellara), Tuscan 
_Emilian (Carrara), Tuscan Emilian (Gombitelli), Tuscan Emilian (Sassalbo), English, American 
English, British English, Scotch English, Middle English (Orm), Middle English (southern dialect, 
_r2thcentury), Middle English (East Midland, time of Chaucer), Old English, Finnish, French, West 

Frisian, Galician, German, Modern Greek, Aeginian Modern Greek, Cycladic Modern Greek, eae 

Modern Greek (asco Lesbic Modern Greek (Mitilini), Lesbic Modern Greek (Gignaal 
Macedonian. Modern Greek, Thracian Modern Greek, Old Greek, Hungarian, Old Icelandic, 
Italian, Japanese, Kanara, Kanauri, Kui, Kurdish, Lahuli, Languedoc, Latin, Laz, Ligurian, 
Lombard, Lorrainese, Marchesian, Mingrelian, Molisian, Norman, Norwegian, Pahari, Picard, 
Portuguese, Provencal, Old Provencal, Rhetian, Rumanian, Sanskrit, Scotch, Spanish, Andalusian 
Spanish, Bosnian Judeo-Spanish, Bulgarian Judeo-Spanish, Mexican Spanish, Tridentan, Umbrian 
(Gubbio), Umbrian (Rieti), Venetian (Grado), Venetian (Murano), Venetian (Triest), Walloon 
_ (Liége), Walloon (Stavelot), Welsh, Yorkshire. 


LANGUAGES PUBLISHING COMPANY 


toe Publishers of PIERCE’S LIBRARY: A Miscrtiany oF LITERATURE AND WorKs OF REFERENCE IN ALL LANGUAGES 


8 West 40th Street, New York City 


REMINGTON PORTABLE 


g The most comp lete fone A Joy and Benefit to Students: - 
typewriter. Has complete and Lovers of Nature 
standard keyboard and other 2; aig 


“hig machine’’ features. 


§ Small, light, simple and ene Forty-seven years of effective service re f 
convenient—fits in case only a | ae Science, Education and Huan 


four inches high. Be: 

c- ° | 12 | 

4 These surpassing features T Guide to Nature 
. ° Monthly Illustrated Magazine | 
have made it the leading aa 
; me EDWARD F. BIGELOW, Editor || 
portable typewriter—in sales ah 
e,: : ij Vie. ~ ihe 

and popularity. : { E Subscription, $1.50 per year; Sediaie. 15 cents; | 

2 ee three months’ trial, 2 5 cents en 
Price, complete with case, $60 | 
pee 


: : Please write for particulars : 
Remington Typewriter Company. o 


374 Broadway ; New York Ae ARCADIA 3 
Branches and Dealers Everywhere SOUND BEACH "CONNECTICUT 


Second Edition: Revised and Enlarged 


Source Book for the Economic | 
Geography of North America > 


By CHARLES C. COLBY 


Associate Professor of Geography 
in the University of Chicago 
‘ 
The fundamental idea of this book is to make available to the busy aeken thes 
material on this subject which is scattered widely through literature. The course hich te 
has been given for a number of years at the University of Chicago has resulted in the 3 
collection and organization of the material most suitable for the purpose. x , 


The book contains short magazine articles and materials from Canadian, Mexican, e : 
and United States government publications. The selection of. the material has been 

* governed by the concept that there is a direct relation between the environment and _ 
the economic activities of the region. ‘The author aims to survey the different sections _ 
of the country from the viewpoint of industrial opportunity. 5 


Payee “ia aie 
aL PO SE 


‘ : 
’ B mh ao = 
wee f 


' With chapter and maps on Alaska Cloth, $3.00, postpaid $ 3. to. 


: CHICAGO - 


a hed 


é 


wii