Google
This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project
to make the world's books discoverable online.
It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject
to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books
are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover.
Marks, notations and other maiginalia present in the original volume will appear in this file - a reminder of this book's long journey from the
publisher to a library and finally to you.
Usage guidelines
Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing tliis resource, we liave taken steps to
prevent abuse by commercial parties, including placing technical restrictions on automated querying.
We also ask that you:
+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for
personal, non-commercial purposes.
+ Refrain fivm automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the
use of public domain materials for these purposes and may be able to help.
+ Maintain attributionTht GoogXt "watermark" you see on each file is essential for in forming people about this project and helping them find
additional materials through Google Book Search. Please do not remove it.
+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner
anywhere in the world. Copyright infringement liabili^ can be quite severe.
About Google Book Search
Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web
at|http: //books .google .com/I
kf ^-^.^•i^.
IGNEOUS ROCKS AND THEIR ORIGIN
■S
' 1 '
3
1
•
4 '
u
1 1
2
»w4
N
H
>
^
•
X
-4
m ^ '
M
■< ^■
K,
^ »
•
■ • "^
"1
•*
i
'1 c
, -^^
'^■
- ^
•-4
■*■ *.
*
, ^
•~4
•^
1
r —
i**
■i *^
71
N
• •
" •
"•*
^*
:^ ti
v.!
te
'■:
•-1
■
-' -':
•
N "a.
\
.^
^ c.
\1
<fU\{
.. *^
ta
-."I
■ -i
• ■
^•»
■s
D
-::
l'
1
^
.■*
■^
■^
•J
H-^
^
H
• 1
:j
*
^
:ai
^
H
-^
'•:
N
^1
**
t'i
'f
(»
v
^ •
a4
o
■t
"^
^
N
:q
■^
H
■ •
V
>4«
•
IGNEOUS ROCKS
AND
THEIR ORIGIN
BY
REGINALD ALDWORTH DALY
8TURGIB-HOOPEB PBOrESSOB OW QEOLOGT
HABVABD UNIVBBSITT
McGRAW-HILL BOOK COMPANY, Inc.
239 WEST 39TH STREET, NEW YORK
BOUVESIE STREET. LONDON, E. C.
1914
■L^'i. -. :+
M.(iKl»-llll.L lt<><l
Co my Wife
^aspirins bellow Worker
^his 5&00K l5 ^e6icate6
PREFACE
This book gives the substance of a course of lectures prepared for
students in Harvard University and in the Massachusetts Institute of
Technology. Its writing was begim at the Institute and finished at
the University. The preliminary work that led the writer into the
igneous-rock field was a study of Mount Ascutney, Vermont, twenty
years ago. The fimdamental problems in that small area are largely
identical with those encoimtered outdoors during each succeeding
field season. An attempt at their partial solution became more hope-
ful as facts were accumulated, both from rocks themselves and from
the literature of geology. The last decade has been specially prolific
in publications affecting the philosophy of eruptive rocks. Many of
these recent memoirs have described definite proofs of vital principles
which had been no more than suggested as possibilities during the
preceding century. The combination of established principles, new
and old, has led to the following general explanation of igneous activity.
It is offered as a working hypothesis, which may have value in helping
to indicate the truly important problems among the infinite number of
those still unsolved. The history of science shows that it is generally
harder to ask a significant question than to answer an insignificant
question; and that in the bettering of working hypotheses the truth is
approached.
At intervals since 1897 the writer has published on special igneous-
rock themes. In a few instances the matter of such papers, after
revision, has been incorporated in this book. In preparing those
original papers as well as the material here published for the first time,
much assistance has been given by colleagues in the Massachusetts
Institute of Technology — Professors W. C. Bray, T. A. Jaggar, G. N.
Lewis, A. A. Noyes, and C. H. Warren; and in Harvard University —
Professors G. P. Baxter, P. W. Bridgman, H. N. Davis, L. S. Marks,
C. Palache, T. W. Richards, and J. E. Wolff. The writer has profited
from many discussions on the general subject with Professors A. C.
Lane, F. D. Adams, and A. C. Lawson, and with Dr. F. E. Wright.
Special acknowledgments are due to Dr. S. J. Schofield and Dr. J. A.
Allan for unpublished information; to Mr. I. Friedlaender and Dr.
A. Harker for permission to reproduce illustrations from their works;
and to several publishers, especially Justus Perthes, Macmillan and
vu
viii PREFACE
Company, The Macmillan C*ompany, and U. Hoepli, for similar
courtesies. In the actual preparation of the manuscript the writer is
chiefly indebted to hLs wife, who performetl much of the manual labor
and, with rare tact and judgment, guidetl him through many a difficult
passage in thought and exprowion.
CAmaiDoc, MAfMArHrncTTK,
Juiy 2.V 1913.
CONTENTS
Chapteb Page
Introduction xxi
I. Abstract 1
PART I
II. Classification op the Igneous Rocks 9
Mode and Norm Classifications 9
Rosenbusch System 12
Adopted Mode Classification 13
Average Composition of Igneous-rock Species 13
Average Specific Gravities 38
Division According to Mode of Occurrence 39
Igneous-rock Clans 40
III. General Distribution and Relative Quantities op Igneous-rock
Species 42
Need for Quantitative Study 42
Relative Quantities of Igneous-rock Species in the United States. . . 43
Relative Abundance of the Alkaline Rocks, including the Syenite
Clan 46
Relative Abundance of the Subalkaline Clans 49
Rock Species known only in Extremely Small Areas or Volumes at the
Earth's Surface 50
Summary of Conclusions regarding Relative Abundance 51
Maximum Size of Individual Bodies 52
General Remarks 63
IV. Eruptive Types and the Geological Time Scale 55
Eruptive Sequences 66
Recurrence of Types belonging to the Gabbro Clan 56
General Recurrence of Other Clans 57
Time Relations of the Granite, Diorite, and Granodiorite Clans 57
Time Relations of the Alkaline Clans 58
Special Time Relation of Anorthosite. . 58
Dike Rocks in Geological Time 58
Modes of Eruption and Geological Time 59
Summary 60
V. Injected Bodies 61
Classification of Intrusive Bodies 61
Igneous Injections 64
Intrusive Sheet 64
Sill; Differentiated Sill 64
Multiple Sill 65
Composite SiU 66
Interformational Sheet 69
Laccolith. 69
ix
Multipla LMnilitb.
futnpuulB I^rnJitli.
Intorfo-nMilHinal l.>rr(il)th.
IliknJith.
l>ike: l>illerentiat«<l Dike.
Apujiliyau, TutiRua
X'okMllF Xprk.
HyMnsltih.
Choiuilith.
Ktbmdlilh.
Hphrmilith.
VI. St-BJAlEXT IlllIllC!-
LMiniiionii-
B«th.ilith.
SUrk. ti<m>
(ranera) Chuwierutmi.
I^iraiKin in Zunm <i( Moiuitai»4KuUinii.
KUrtiRBlHtn I'Mftlk-I luTartOOM Ah*.
Tin.t' h- :'ilii-n 1" MMUIUib4>lliUtll||.
('rcMt-ouliinc ttrUtuMi lo lh« nv^l'-d Kii«
IVinnp*! 1 1 utiiH* li Had UmI WbIU
DuwnWBnl I'tiLryameat. .
.' M Tiv»<Ia1 FuriiMtiitn
i-.>-iiii>n i>( »)ut>iBrviii Ih-lm
'r..i.-Bn.l lUibnlilh*. .
VII. KxTKoiiF Itoiiirx
Ivu.ti'Mix'i "f ItiM-k IIikIm'k kiiil tit I>r|>iTM>Hin ¥t»
N'lilrMiir Nprk*
Viilranir I'luiti.
Tiimuti.
lU'€n\Ui»
Viilnuiir Ciiiieii.
<'une Clunlerii. Cimr Chitinii.
ItUl«.h.lt<M>
A'lvrnttve ('ralri
Nrntol Oalrni,
CONTENTS XI
Chaptkb Pagb
Nested 148
Sunken 160
Volcanic Sinks 150
Simple 162
Nested 163
Volcanic Rents 153
PART II
Vlll. CosiacAL Aspects 165
lYincipal Sources of Magmatic Heat 155
Planetesimal Hypothesis in Relation to the Heat Problem 156
Density Stratification of the Earth 160
Earth's Sedimentary Shell 162
Earth's Acid (Granitic) Shell 162
Earth's Basaltic Shell 164
Earth's "Peridotitic" Shell 166
Relation of Planetary Shells to Petrogenesis 167
"Average Igneous Rock" 168
Speculation as to Primitive Differentiation of Earth's Silicate Mantle. 170
Physical Condition of the Substratum 171
General Conclusion 173
IX. Abyssal Injection 174
Introduction 174
Contraction of the Earth 175
Shells of Compression and Tension 176
Secular Accumulation of Tensions and of Cooling Cracks 178
Injection of Magma into the Shell of Tension 181
Relief of Tensions through Abyssal Injection 183
Downwarping of the Surface as a Result of Abyssal Injection. . . . 185
Orogenic Effects 188
Renewed Abyssal Injection; Development of Batholiths 188
Volcanic Action Subsequent to Mountain-building 191
Summary 192
X. Magmatic Stoping 194
Development of the Theory 194
Marginal Shattering 198
Relative Densities of Xenolith and Magma 201
Sinking of the Shattered Blocks 203
Roof Foundering 206
Stoping in Sills and Laccoliths 206
Abyssal Assimilation of Stoped Blocks 207
XI. Magmatic Assimilation 209
Introduction 209
Heat Supply and Magmatic Temperatures 210
Observed Temperatures at Volcanic Vents 211
Observed Liquefaction of Country Rocks 212
Fluxing by Concentrated Volatile Matter 212
Influence of the Mixture of Rock Matter on Liquidity 213
Magmatic Stoping in Relation to the Low Temperatures of Consoli-
dation for Batholithic Rocks 214
Av»ilal>l<> Ilrat r<ir .VMiiiiiUlixn in il*1h<ilithir Muw-. :i
MwiiinKl .Vwimilation '2
AttyaMi AMimiUlinn. J
.\iMimilat)iin in Ininuivc ShM'in. . i
XII Mai.mati'- l>irrt«iK\Ti*Ti"!(. J
InltiHlurtiiin. 'J
Molrrulmr I)ifl>»i..n J
Krarlion*! C'nutalluktion J
Liquation. :■
<iravilatire DiHrrMi nation. 2
Diflrrenlialiiin at Cwitral Vrnla. 'J
Chrmiral ('(HitraM ul llulonu- itiirk and ('orrt^Kindtnc Kffiuive
DilTrrrntialiiin in Ijtrrulithii an<l Inlrunive MhfVtn. i
(iravilalivti Diflcmitiatinn in M'K-kii anil Itatbiililha. 2
Kxpuliiiiin of Itrimlual Mamna. J
<iaii and Vs|>iir DifferpntintM. J
(laivilUK Tranitfpr '2
XIII. Mki'Has'ihu or Vuu'ANir Vrvrs or the I'estm^l Ttpe. 2
intrmlurliMn- . 'J
SiiniP Dirn-t ("iinji«|iirnr«i oC Aliyiwal Injrrtiiin. 'J
(ipuotir ClaMifiration tU Viilr&nn- (Ja»e». J
<>|i«nin|| and Ijuralitatinn i>( the Vcnl J
EnUntol KiMiurp* 'i
Diatrrmn. 2
llutonir ('uiKilaa. J
('onliniianrc (if Artivity at Centra) Vrnl« 2
iUtc (if llrat-liiafi throiwh ('<in<lii<-iii)n into lh« Wa)li>. 2
ItatP ft llrat-lom thnnich Hailialiiin at the Crater. 2
Methodn ut llrat Tranofpr 2
Two-phax^ C*onv«cliun. 2
l^vn Fountain*. . 2
CmdinR hj- KiainiE Juvpnilc (iaii. 2
The Volranir KurnarP J
Sumniar)- un the Ileal I't<iI>Ii-iu of an Artive Ontral Vent. . . 2
Revival of Arlivity at ihr Knd tif a Dormant IVrioil 2
Small Swe of CentJ-al Vrnli. .'
Kiploaive Ty|>n; Mattmalir anil I'hreatir J
Maitmatir Differentiation at Central Vent*. . J
l^^^gew in Ktplmuvenenii at the tlrrater Vents, 2
l^va Uutfluw at Central Vrni*. 2
Th« two Type* of I^va Khiw*. 2
Vulraniam OriKinaiinti in Satellitir Injerliua*. J
\ N«cci««ar>- DiviMon of Central Vent*. .i
(jeneral Summary. .>
XIV. tViit-nc Tkkokt or the I(.\coii. H.mh. .1
Summary ii( the I'>lertie Theory . .1
I»ewinnon-I.*j<iing'«*leneralThc«irj-. 9
(^Mtetir CUa«iKntMMi of Ijpteotu Korka. t
1^
CONTENTS xiii
PART III
CHA.]>-rER Page
XV. Gabbro Clan 313
Included Species 313
Primary Basaltic Magma 315
Normal Olivine-free Species 316
Quartz Diabases and Their Allies 316
Norites 318
Hypersthene Basalts and Enstatite Diabases 319
Hornblende Gabbros 319
Iron Basalt 321
Anorthosites 321
Abnormal Features of the Anorthosites 322
Anorthosite a Differentiate of Gabbro 324
Mode of Intrusion 328
Special Conditions for the Formation of Anorthosite 335
Rocks Syngenetic with Anorthosite 336
Conclusions 337
Pillow (Ellipsoidal) Basalts and the "SpiUtic Suite." 338
Transitions to Other Clans 340
XVI. Granite Clan 341
Included Species 341
General Statement 342
Species Derived from Syntectics of Sediments and Basaltic (Gab-
broid) Magma 344
Purcell Sills 344
MarysviUeSiU 346
Minnesota Cases 346
Sudbury Sheet 347
Insizwa Intrusion 349
Other South African Cases 350
Preston Laccolith. 353
Medford Dike 353
Globe District Intrusions 354
Swedish Cases 354
Scottish Intrusions 354
Intrusions of British Guiana 355
Species Derived from the Syntexis of Non-sedimentary, Acid Rocks. 355
Syntexis in Feeders of Fissure Eruptions 356
Conclusions 357
Transitions to Batholiths 358
Origin of Normal Batholithic Granites 359
Granitic Magmas Differentiated from Magmas Belonging to Other
Clans 360
Influence of Resurgent Gases 361
Eruptive Sequences 362
Differentiation from Dior i tic Magmas 363
Differentiation from Granodioritic Magmas 366
Dif iation from Syenitic Magmas 367
Gran Aputes and Pegmatites 368
p 1 of the Rhyolitic Types .370
XVII. Dioun Clan . St*
Indudod Speriet. ... 374
AndMitca. . . . . 37S
AuRiie Andesiic. . 37&
llyp«t«then« Andente. . 380
\Ue* AndcMtm and lIurnbkiHln Atuleaitca. 3U
Diorite*.. ... 3H3
XVIII. GBANODioKm Clan. . . 3KS
Inrluded Hpfim. . 385
Oriicin- ■ 387
XIX. BvrMTa Clan. 383
I Deluded Kitcrim. 303
(feneral SlAteroent ut Onipii. MS
AMoriktion with the <iabbro Clan. . 3B5
Swlinentary Contr»l.. 3M
Statialira tJ field AnorialiuM. 403
Diffrrratiatjon of Hyntartin in llare. 40A
.■^maU Siie of Bodiea ttalontpng to the Hyentt« Clan. 40K
Chemira) Contraat of Plulonin and EfTu«ive« of the Clan. . 400
XX. At-KAUXt Clann. 410
iDcludadSpariea.. . . 410
Ci«neral Statement of OriKiii. 412
Amoriation with CariHinate Hwlu 41&
Kteld .^Moriation vith the Cabbro Clan. 421
(General Chetniral KITwIb of the Alian«i>iioo uf Cart>onal« Hock*. 430
KvidetK«a from the Mineralricy of Alkaline Kurka. . 434
DiH(T«nIiatioa of .Mkalina Horka in Plare.. . 437
Chrmifal <*untra«i of Alkaline Volraoic Specm and the CofTe«|iond-
init Ilulonic Sperii^ , . 443
Kruplive ScquMirea in .Alkahnc IViivinm. 443
Complrtnenlarv Dike* uf the Alkaline Clann. 444
XXI. I>EaiiKmTi Clan a.M> Maomatti <>aE». 44«
InrludeH (t|>er*ni 440
Cfeneral Statmirnt tJ < >ri|tin 446
Kelatiun to the tralibro Clan. 447
I'ltra-feiTomaitneaian and t'llra-rafemic Differentiate* of Sjntertira 4-'iO
Sperim Fdrmeal by (.ianpoiu Transfer. . 4A2
Maitmatic On*. . . 454
XXII. EcLKi-nr Thkumt Arrucu nt tmb Numth AMtaMAN CuBDitXEMA. 4M
ArraNDix
A. (Table XX) Shuwinit Number uf Separate DetermioatMiM UMalin
Computinc the Averace (Quantity of Each Osida in the Rock-
type* Iwiol in Table II. 466
B. (Ta)>le XXIi Showinit Onler of l'>upli<in in Different Hettiiina. . 411V
C. (Table XXIIj iiivinit ImI of Dmtnrta Cliararteriieil by Memberaof
theSyeniteClan; wiihN«tnii>n(hrNaiiirenf Cuuntr>- Horka. . 4M
D. (Tal>le XXIlIj (.iivinit IJM of Diftruu. Chararteruwl by AlkaUM
Itork-typw; with Notes on the Nature uf Country Koda.. . . 511
Imwx. 899
ILLUSTRATIONS
Paob
Plate I (Frontispiece). Mount Baker from the Fraser River at Mission Junc-
tion.
Plate II. Vesuvius in 1911 Opposite page 136
Fig. J 1. Outcrop of lamprophyric sili in Colorado 64
Fig. 2. Dolerite sills cutting basalts, Isle of Eigg 65
Pig. 3. Section of area shown in Figure 2 65
Fig. 4. Multiple siUs of dolerite cutting basalts, Isle of Skye 66
Fig. 5. Sills at the Kettle River, British Columbia 67
Fig. 6. Section of a composite sill in Skye 67
Fig. 7. Composite laccolithic sills in Skye 67
Fig. 8. Laccoliths of the Judith Mountains, Montana 68
Fig. 9. Sections of area shown in Figure 8 69
Fig. 10. Sketch of Warm Spring laccolith, Montana. 70
Fig. 11. Section of Kelly Hill laccolith, Montana 71
Fig. 12, Section of Burnett Creek laccolith, Montana 71
Fig. 13. Section of Warm Spring laccolith, Montana 71
Fig. 14. Laccolith of Bear Lodge Mountains, Wyoming 72
Fig. 15. Section of composite laccolith in Skye 72
Fig. 16. Section of composite laccolith at Black Buttes, Wyoming 72
Fig. 17. Interformational laccolith. Little Rocky Mountains, Montana. . . 73
Fig. 18. Asymmetric interformational laccolith at Barker Mountain, Mon-
tana 73
Fig. 19. Asymmetric interformational laccolith at Black Butte, Montana. . 73
Fig. 20. Section of Nigger Hill laccolith, Wyoming 74
Fig. 21. Dolerite phacolith cutting Ordovician strata, Corndon, Shropshire, 77
Fig. 22. Differentiated dikes in the Trusenthal, Thuringia 77
Fig, 23. Differentiated dike in the Elmenthal, Thuringia 78
Fig. 24. Grreat differentiated dike at Brefven, Sweden 78
Fig. 25. Multiple dike following fault plane in Cowal 79
Fig. 26. Multiple basaltic dike cutting granophyre, St. Kilda Island. ... 79
Fig. 27. Composite dike, Broadford, Skye 80
Fig. 28. Composite dike, Tormore, Arran 80
Fig. 29. Map showing Scottish dikes and the Whin sill 81
Fig. 30. Map of dike system, Spanish Peaks, Colorado 82
Fig. 31. Dike system composed of rhyolite, Corsica 83
Fig. 32. Section of chonolith near Glen Coe, Scotland 85
Fig, 33. Section (hypothetical) of chonolith, Monzoni, Tyrol 85
Fig. 34. Map of chonoliths. Monarch and Tomichi districts, Colorado. ... 86
Fig, 35. Sections along the lines in Figiu-e 34 87
Fig. 36. Diagrammatic section illustrating an ethmolith 88
Fig. 37. Map showing distribution of batholiths in North America 91
Fig. 38. Map showing position of the great Patagonian batholith 92
Fig. 39. Map showing stocks and batholiths in Brittany 93
Ao. 40l Map showing elongation of an Irish batholith 95
y-m^ XV
'4J||
xvi ILLVSTRA TIOSS
Flu. 41. Map ibowicig batbolitlii and ftocks in Cornwall and Oeiroa.
Pio. 42. Map abowing batholtthf and ttorkii in the P>Ten6eii. 97
Fio. 43. Map of Castle Peak stork, Britiiih ( olumbia W
Fio. 44. Map of the 8hap fcranite «t€>rk. KngUnd 101)
Fiu. 45. Map of monsonite stork, Tellurnle quaiirantle. Colorado 101
Flo. 46. Flast-weft sertion throuKh utork in Fifnire 45. 101
Vrn. 47. Plan of bathohth in Bidwell Bar quaiiranftle. California 102
Fig. 4H. Diaicram showinic features of an ideal bath«ilith. 103
Fio. 49. Roof pendants in the Simtlkameen fcranorliorite bathobth. 101
P^o. 50. Map of part of 8noqualmie bathohth, Waflhington. 106
Fig. 51. Transverse sections in a Sierra Nevada bathohth. 106
P^o. 52. Map showing incipient unroofing in a granite stock. Saxony . 106
Fto. 53. Map showing partial unroofing in a granite stork. Saxony. 106
Fig. 54. Map showing partial unroofing of a granite stork. Alaska 107
Fto. 55. Map and sertion of a granite stork, Selkirk Mountains. 107
Pig. 56. Sertion showing downward enlargement of a German granite
batholith. 108
Fig. 57. Demonstrated profile of the Lausiti granite bathohth. 106
Fig. Sk5. I>emonst«*ated profile of a granite batholith. Fichtelgebirge 108
Fig. 59. Plunging contact at south side of Castle Peak stork. 109
Fig. 60. Plunging contact at north nde of same stork as in Fig. 59 110
Fig. 61. Canyon-wall sertion of same st(x>k as m Figure 59 111
Fig. 62. Map showing replacement of sediments by a granite batholith. Brit-
tany. Ill
Fig. 63. Map showing replacement of sediments by the Cauterets granite of
the P>Tenees. 112
Fig. 64. Map and section showing replacement by the compcjsite stork at
Mount Ascutney. 113
Fig. 65. Section of Okanagan com|Mi«ite batholith at 49th Parallel. 115
Fig. 66. Section of the composite batholith of New England. New South
Wales. 1 15
Fig. 67. Map showing distribution of the Derran traps.. 118
Fig. 68. Sections showing Tertiar>' fissure eruptions of basalt. Oregon. 119
Fig. 69. Sertion of fissure eruption in Wdliams Canyon, Aruona. 119
Fig. 70. Map of dike feeders of fissure eruptions near Mount Stuart, Wash-
ington. 121
Fig. 71. Ideal section Ulustrating roof foundering. Yellowstone Park. 122
Fig. 72. Map of Yellowntone Park. 123
Fig. 73. Section of volcanic tuff neck. Faroe Islamls. . 126
Fig. 74. Section of Carboniferous neck, (last (irange. Perthshire. 126
P^G. 75. l*lan and sertion of explosion fissure in Fifeshire. 127
Fig. 76. Sketch of the Cabeson basaltic neck. New Mexico. 127
Fig. 77. l*lans of IVrmian tuff necks, Ayrshire. 128
Fig. 78. Ground plans of composite necks, Stirlmi^hirr 128
Flo. 79. Twin volcanic necks of Carboniferous date. Scotland. 129
Flo. 80. Section of composite necks, StirUngahire. 129
Pto. 81. Volcanic vent and crater, Ice Spring cluster, I'tah. 129
P^o. 82. Sections showing erosion of crater chargeil with cong^aktl lava. . . 130
Flo. 83. Sections showing four stages in the recent histor>* of Mt. Pel^.. . 131
Flo. 84. Sketches of the Mt. Pel^ spine, showing its changea. . 132
Flo. 85. Sketch profiles sbowing ebnngM in Bogoslof Islands in 13 months. . 132
ILLUSTRATIONS xvii
Page
Fig. 86. Map and section of the new plug-dome, Tarumaij Japan 133
Fig. 87. Tumulus in the floor of Kilauean sink, Hawaii 134
Fig. 88. Tumulus in the floor of Kilauea 134
Fig. 89. Driblet cone near the Kamakaaia cones, Hawaii 136
Fig. 90. Pumice cone breached by outflow of obsidian lava current, Island
of Lipari 137
Fig. 91. Cone cluster of the Velay, France 137
Fig. 92. Map showing relation of Tertiary volcanoes of France to crust
fractures 138
Fig. 93. The cone chains of Java 138
Fig. 94. The neo-volcanic cone chains of Japan 139
Fig. 95. Section of the Tritriva crater, Madagascar 140
Fig. 96. Distant view of a small pit crater. Puna district, Hawaii 142
Fig. 97. Map and section of Amsterdam Island volcano 142
Fig, 98, The nested craters of Vesuvius 143
Fig. 99. Nested craters of Etna in early part of 19th century 143
Fig. 100. The Tarawera Rift and Rotomahana caldera, New Zealand. . . . 146
Fig. 101. Map of the Caldeira of the Sete Cidades, San Miguel Island. . . .146
Fig. 102. Nested calderas at the Masaya volcanoes, Nicaragua 147
Fig. 103. The sunken caldera of Santorin 148
Fig. 104, Maps showing modifications of Krakatoa by the explosion of 1883. 149
Fig. 105. View of the Enclos of Reunion 150
Fig. 106. Map of the Kilauea sink, Hawaii 151
Fig. 107. Volcanic sink at top of Tengger volcano, Java 151
Fig. 108. Section of the sink shown in Fig. 107 152
Fig. 109. Nested sinks at Mokuaweoweo, summit of Mauna Loa, Hawaii.. . 152
Fig. 110. Section along the course of the Cleveland dike, Yorkshire 182
Fig. 111. Section of the Cleveland dike, across the Cross Fell escarpment. . 182
Fig. 112. Diagram illustrating abyssal injection 183
Fig. 113. Sections showing the relations of abyssal injection to geosynclinal
downwarping and to orogeny 185
Fig. 114. Map of quartz diorite stock, Marysville, Montana 195
Fig. 115. Section along the line X-F in Fig. 114 196
Fig. 116. Map of the Glen Coe district, showing location of ^' cauldron-sub-
sidence." 197
Fig. 117. Arrested stoping at roof of Lausitz granite batholith 200
Fig. 118. Shatter-zone at contact of Trail batholith, British Columbia. . . . 201
Fig. 119. Section of sapphire-bearing dike, Yogo canyon, Montana 207
Fig. 120. Plan of Mt. Johnson, Quebec 228
Fig. 121. Longitudinal section of the Lugar sill, Scotland 239
Fig. 122. Map of Long Lake quadrangle. New York 240
Fig. 123. Section iUustrating development of femic contact phases in batho-
liths 244
Fig. 124. Syngenetic granite and diorite in Penobscot Bay quadrangle, Maine. 245
Fig. 125. Section of the Grampian Hills stock 245
Fig. 126. Section illustrating differentiation in some dikes 246
Fig. 127. Artificial diatremes in granite cylinders 251
Fig. 128 Diatreme opened on a fissure. Laws Castle, Fifeshire 252
Fig. 129. Ideal section showing formation of volcanic vent 253
Fig. 130. Map showing the long continuance of volcanic action at the Cantal. . 254
Fig. 131. Map of Halemaumau crater, Hawaii, in July, 1909 255
2
XvHi tlLVSTKATIONS
Flo. 133. SATtioa of llakmnnnxi. inuatming two-phaae eonvKticn. .
Fio. 133. IdnllonRitwliiuU •(><-ji Ti'-{ unst<>".il iitjartiofL
Flo- 134. IdcklcroMHnciion U.r.H.^Mi i.<,l1<T.<,K-i> ■bowiiiaFiit. 133. .
Flo. 13S. fleetion of uppar i-iri i.| 1 ill finMi oin.-
FlO. 136. Rwticin uid pUa of hn-..! -I^m :i>-. L v. tbUot, Wwt Mmu.
FlO. 137. fl<-h«n>»tir pUn ami Hvi>->ti <•! (hi-- KjefiLe^wt
Flo. 138, fl«ctinn at iha RiaJiwt. Auwiti^ iolmnA Wooiiili bmMlb.
Fiu. 139. MaitMli'' bcvauM f<r Ml, Km Adiiai, Omuuu? . .
Flo. 140. --^t.. uwn tt Ita ajptawokmak iomm of h* StaiiUwim Baain. .
FlO. 141. PUn «in<{ wrtHwi of tha nkUara Ai BMMUtAkO.
FlO. 142. Part <J 'V-irmnvrtil RMp of •oulhtawlerti Ttaiia:i ■ .
FlO. 143. ScrtiiHI of tha llT(««b(jnr rulfano, orUml . .
Fia. 144. Mapofpartof ^o^^:! •'SoBincpciituiMof "VitIkai>>Embryon«i".
FlO. I4S, Pland , ' .- ^ : ■■ ,- nrckt.
Ro. 11«. Map of
FlO. 147. S«^ion o( nrcki on thr rout o( t'lrcshir*.
FlO. 14ft. Np« Mountain at I'su^tan, Japan. . ...
Pio. 14fU. Map of t'su-San, Japan . .
Flo. 149. Map of the Dululh lamilith, Minnnola . .
Fto. ISO. Map of lh» Clamornan icitihr", Onlnric .
FlO. 151. Map of thr HaJ Hivrr larroUth, Wiiminiim. .
FlO. 152. Sectiutuat ■ Kiv«' laccolith.
Fio. 153. Map of the Murin dixi
FlO. 154 - I 53
FlO. 155. Map of , Norway.
Flo. 154. Map <i( anorthowte ■ ■ ■
FlO, 157. Map o( anorthomtr York .State.
FlO. 158. Srrtinn.i. U.::^..i.,..i; ; .:i;- .. M River.
Fia. 159. Map nf I't|t<<uP I'otut, MiiiMvuta .
FlO. 100. Map of SudlMirj- diatrirt, •hawiOK inlerfiirmaiional theet
FlO. 181. 8eclittB<Jt»n-.-tirrl m*r'p«l in Fie. IflO. . .
Kio. 162. Map.rf ' T • i !■ ! ': ; ■ ! .
Fto. 163. Map and i«*tiims of tho Ka>>bn> laccolith at I'rwton, Connecticut.
FlO. 164. Section of the Hayimne bathulith, Dritiah Columbia.
Fto. 165. Map of intruiiivr ntorkx, Craiy Mountains, Montana
Fto. 166. Map of intrunivo itockn. Castle Mountains, Montana
Fto. 167. Map of '
FlO. 168. Section
FlO. 169. - ■ * .....J... -■.;. :,i«j..
FlO. 170. Plan of oompoaite dike, Cir Mohr, laland of Arran
FlO. 171. Map of Slnimboli Island
FlO. ITi- Section on bne XY of Fi«. 171.
Flo. 173. Map of part of KUensburfi qiiatlranitle. WashinRton
Fm. 174. Section XY In Kig. 17;f
Fio. I7S. Sketch enibracinn Yellowotone Park
FlO, 176. Map ;.:.i:^i. :; _ , : i ^:' :. -^ quadranRle, UrrKun.
Fio. 177. Section *-f M M»<«*|..n. Vicl.ma.
Fto. 17H. Map nf Ihr Monlcrritian Hilla. Quebec.
Fio. 179. Seetiiin at Tinlic, t'tab
Fto. 180. Section of dike at Karsuarauk, Greenland
ILLUSTRATIONS xix
Page
Fig. 181. Map of Mt. Shefford, Quebec 403
Fig. 182. Section through Nigger Hill laccolith, Wyoming 414
Fig. 183. Map of the Monchique intrusion, Portugal 416
Fig. 184. Map of part of Alno Island 419
Fig. 185. Section in the Uvalde quadrangle, Texas 421
Fig. 186. Map of island of Vulcano 422
Fig. 187. Map of Roccamonfina volcano, Italy 423
Fig. 188. Map of part of the Dunedin district. New Zealand 424
Fig. 189. Section of North Otago Head in area of Fig. 188 424
Fig. 190. Map of northern Madagascar 425
Fig. 191. Section of the Cantal volcano, France 425
Fig. 192. Map of the Bancroft district, Ontario 428
Fig. 193. Section across le Livradois and le Comt6, France 432
Fig. 194. Map of phonolites of the Velay 433
Fig. 195. Map of Inchcolm Island, Scottish coast 435
Fig. 196. Sections of Cnoc-na-Sroine laccolith, Scotland 439
Fig. 197. Map of the Ilimausak intrusion, Greenland 440
Fig. 198. Schematic section of the Ilimausak intrusion 441
Fig. 199. Actual section in the Ilimausak intrusion 442
Fig. 200. Section of composite sill in Greenland 447
Fig. 201. Plan of a rock group in Scotland 448
Fig. 202. Section of Sinni valley, Italy 449
Fig. 203. Map of the Palisades sheet. New Jersey 449
Fig. 204. Map of the Kiruna district, Sweden 453
Fig. 205. Sections through iroD-ore deposits. Eagle Mountains, California. . 454
INTRODUCTION
Geology has been charged with failure to measure up to the in-
tellectual standard of the so-called ''exact" sciences. The reproach
is no longer merited. It originated during a century when the power
of the experimental method in science was first clearly appreciated.
As usually the case with great discoveries, this was soon given exagger-
ated importance by many students of the logic of science. Now that the
intoxication of early, magnificent success in the use of experiment
is succeeded by more sober second thought, it has become clearer
that this method of research is only one of several that are quite
essential and are of coordinate value in scientific thought. The in-
cessant revision of experimental methods, and the inevitable shifts
in the values credited to physical and chemical ''constants," show the
inexactness of the principal "exact" sciences. Their mathematics is
precise; their premises are not. It is diflScult to name a single ex-
perimental result which is not troubled with some degree of uncer-
tainty. Nevertheless, using the principle of the limits of error, the
principle of the compensation of errors, the principle of correlation,
and the principle of direct inference, physics, chemistry, and astron-
omy have produced majestic and indispensable results. In each
case, the fundamental unit of mass — molecule, atom, or star — can
only be understood through the use of all these principles. At bottom
each "exact" science is, and must be speculative, and its chief tool
of research, too rarely used with both courage and judgment, is the
regulated imagination.
Though not so tinctured with mathematics, geology is in essentially
the same position. It is "exact" in the sense that a countless number
of its observations are quantitative, with limits of error so small as
to permit absolutely rigorous deduction. The larger part of the earth
is inaccessible, like molecule, atom, or star, but the principle of in-
ference has already afforded geological results which are as final, if
not as fundamental, as those won in the other sciences. Many lead-
ing facts in geology have been necessarily secured through methods
other than the experimental. The existence of peneplains has been
proved in spite of the obvious impossibility of reproducing them in the
laboratory or of reducing the subject of erosion to mathematical
formulas. Some geologists refuse to consider seriously theoretical
discussions regarding the earth's interior, on the ground that theory
xxi
XXII ISTHOUVCTWS
muAt await the quaiititatK'e data from the laboratory. This m^ital
poHition w not juittifie<l by the master in ph^'sicfl, ehemistry, or astron-
omy, whoHc imaiciiiation or speculative faculty is always working in
advance of his "(>xact" determinations.
Wliat K<H)h»K>'t hl^t' ever>' other science, nee<ls to-day is a frank
recognition that imaginative thought is not dangerous to science but
is the life hloo<i of science. Kven the universities do not fully recog-
niie this fact, and are notoriously failing to develop the stimuli which
are necessar>' for the controlled, scientific imagination. Not only is
geolog}*^ now characterised by rigorous thought; by its nature as
a M*ience involving long excursions into space — inaccessible places —
and time— epochs long passed — geology is peculiarly fitted to stimulate
the regulatinl imagination, a process at the core of the highest education.
S<*ience is built on a long succession of mistakes. Their recognition has
meant progn*ss. Progress, indefinitely more rapid, will be possible when
men of science have more generally lost the fear of making mistakes
in using to the uttermost their powers of correlation and deduction.
Si*ience is drowning in facts. It can only be rescued by the growth
of systems of thought. Better than none are ** little systems" which
**have their day and (*<'ase to be." We can hope that geology, like
every other science, will find its superman who shall show us the build-
ing hidden lH*hind the scafT(»Ming of myriad isolate<l facts of nature.
Meantime, it is the duty of every workrr in science to strive for a
complete mental system in his fiehi of n*s<'arch and. however mistaken
he may be, he shoultl have the si)ecial symi>athy of his fellows. The
l)est sym|)athy is expres?M»<l in constructive criticism. The "facts"
of toHlay are thr hy|H>thf»s4*s of yt»stcrday.
IGNEOUS ROCKS AND THEIR
ORIGIN
CHAPTER I
ABSTRACT
A comprehensive knowledge of igneous rocks is important from
many aspects. The sedimentary rocks could not much exceed a
half-mile (0.8 km.) in average thickness if they were spread evenly
over the earth. The stratified terranes themselves have been derived
from igneous terranes. This planet is essentialfy a body of crystallized
and uncrystallized igneous material. The final philosophy of earth
history will therefore be founded on igneous-rock geology. The earth
has the appearance of being a small, cooled star and its physical con-
stitution and history are problems concerning the nearest of the stellar
host. The formation of continental plateau, mountain range, or ocean
basin is a product of forces developed in the planet below its pellicle
of sediments. The salts of soil, river, and ocean waters, as well as
organic matter and the gases of the atmosphere, are largely, if not
wholly, derivatives of rock materials once in a state of fusion. It is
becoming increasingly clear that most of the world's ore deposits are
genetically connected with igneous rocks. Economic and dynamical
geology, meteorology, climatology, and oceanography are thus deeply
affected by increase of knowledge regarding the natural history of
igneous rocks. Historical geology itself is enriched by a systematic
review of the earth's eruptivity. Volcanic effusions and large-scale
intrusions of granitic types of rock matter can often be used to date
events directly registered in sedimentary formations.
This book is intended to summarize and correlate the facts known
about igneous rocks, with special emphasis on their field relations.
Knowledge of petrography and a moderate acquaintance with the
physics and chemistry of rock-melts are assumed, but the treatment
of the subject is essentially geological.
The work is divided into three parts. The first of these (Chapters
II to VII inclusive) broadly considers the facts which need explana-
tion in a philosophy of the igneous rocks. The second part (Chapters
VIII to XIV inclusive) contains a general, eclectic theory on the sub-
1
2 las'Eors RfH'Ks asd their ohicis
jrrt. Till* thini ((*haptcr5 XV to XXII indii!<ivrj outlinoM the n*i«iiltf(
of npplyiiifc the Krnrrnl tli<t»r>' to tin* farts so far <l<*t<*nniniHl.
ChaptiT II Kivrs. for roiivfiiiniri* of n'frrrnre. a claMtfification of
iKnf*ous-nN*k s|M*ru*s. whirh U a sliKlitly Tno<lifi<*4l form of that in
Kos4*nhiisrh's ** Mikroskopisrh<* PhysiographH* Avr MaHsifcon (in^teinc'*
(StiittKart. 11K)7- IlKW). Ki»asoiis an* static! for prfffrring a claHKi-
firation fouiiclfMl on actual niinrral roin|M>sition to the ''norm'* classi-
firation. The calculated average chemical coni|)ositions of 1 16 of the
most im|M)rtant sp^^cies are fciveii. In all. nearly 700 namcnl varieties*
of igneous nM*ks are recogiiized by p^^trographers. KoHenbuM*h
gniui>s thes4* into families. Familic*fi which an* nearly identical or
are s|M*cially related in chemical composition are here f^rouped into
*'clans.*' Thus, hiolite f^ranite is a species; all granites compose a
family: granites, granite porphyrin's, liparitcff, etc., form a "clan."
These* sulKlivisions an* far from lN*ing sharply limited but they have
considerablf* value in facilitating discussicm of ro<*k genesis.
(*hapter III indicates the present lack of sufficient information
regarding the sfmtial distributi<m of igneous tyi)es. An estimate of
their relative quant iti(*s in the United States shows some highly note-
worthy contrasts among clans, families, and species. Mo»t of these
contrasts are (*ssf*ntially pn*s<*rv(*<l in a still niugher estimate of relative
quant itif*s throughout the world. The maximum volumes known to be
assumed by each sp€*cies in separate Ixnlies an* also compared, with
instructive n*sults.
(^hapter IV sketches the n*lation of igni*ous ty|)e9 to g<*ological time.
OI)servt*tl eruptive s<*<|uences, in numlM*r sufficient to illustrate their
variation rather fully, are tabulated (.\p|x*ndix R). The peniistence
of the gabbroid clan in the earth's eniptivity is a prominent fact.
All the important clans are repn*sented in eruptions of Iwth the earliest
and latest grand divisions of geological time — the pre-<^ambriaii and
the Onozoic. Neverthelc*s8. the relative abundance of the clans has
notablv, and in some cases svstematicallv, varied with the march of
time, tjich of the principal mo<les of eruption has characterised both
pre-('ambrian and post-C'ambrian time.
Chapter V <liscuss<*s the classification of the int rani ve iMxlies which
have l)een emplaced l»y simple injection. The classification is baaed
on field n*lations. The range of size in th(*se liodies is illustrated by
approximate values.
Chapter VI gives a classification of the '* subjacent " intrusive bodies
(stocks and batholiths) and indicat4>s their leading structural relations.
The field proofs that they have replaced or incorporated the country
rocks are empha^tized.
Chapter VII contains a discussion and classification of extrusiire
ABSTRACT 3
bodies. In the interest of completeness a theoretical group, "de-
roofing eruptions," are entered in the classification along with the
well-demonstrated groups, fissure eruptions and central eruptions.
The second principal part of the book, the outlining of a general
theory of the igneous rocks, begins with Chapter VIII, which dis-
cusses the cosmogonic relations of the subject. The planetesimal
hypothesis of the earth's origin, like the older nebular hypothesis
associated with the names of Kant and Laplace, logically seems to
imply a former molten stage for the surface layer of the globe. Before its
solidification, that layer became stratified by gravity, an acid (granitic)
shell above and a more basic (basaltic) shell beneath. The sedimentary
rocks form a third shell, interrupted and everywhere relatively thin,
which has been formed during the recognized geological periods.
Petrogenic theory is chiefly concerned with the interactions among
these three shells. Since the late pre-Cambrian at least, igneous
action has been initiated by injections of the molten substratum
basalt into the primitively solidified acid shell.
Chapter IX is devoted to a speculative attempt to state the con-
ditions for such "abyssal injection." The hypothesis favored is
that based on the conception of an earth which is contracting through
the joint action of several causes. The lower and thicker part of the
crust is subject to tensional stresses. Those stresses, together with
the weight of the crust, are held responsible for the rise of the primary
basalt. It has difliculty in rising higher, into the shell of compres-
sion, and thus normally the basaltic mass has the form of a vertical
wedge. These primary injections are called "abyssal magmatic
wedges." Their relations to geosynclinal downwarping and to moun-
tain-building are briefly indicated. During strong orogenic move-
ments the shells of tension and compression are sheared apart over
wide areas, thus permitting of abyssal injection on the largest scale.
The great wedges then injected perform magmatic replacement and
assume the structural relations of batholiths. Vulcanism of all types
is likewise produced by the activity of abyssal wedges.
Chapter X reviews the mechanism of magmatic stoping, a process
partly explaining batholithic replacement. Its significance chiefly ap-
pears in connection with three different considerations. The principle
affords a reason for the general absence of hybrid or mixed rocks, formed
through the solution of invaded rock by batholithic magma at their
mutual contact. Failure to find these shells of diffusion has been a
leading cause for long-continued scepticism of many petrologists as
to the assimilation of the intruded rocks by batholithic magma.
Secondly, stoping accounts for the enlargement of batholithic (abyssal
wedge) chambers. Its most important consequence is the deep-
4 lUSEOVS R(M'KS AM) THEIR ORIGIN
neaifMl Holuiiun of (lowii-st(»|MMl lilcK-ks in hatholithic magma, in«
volving th(* nf*<'<*ssity of lM-li4*f in the secondary origin of much of
tlu* worM'n iniiKHiHtir and ign(*<iii'«-rork matter.
Chapter \I i** orcupicd with a theorrtiral study of magmatic
a.K.*iiniihition in ^«n(*ral. It is of two kinds, marginal and abyssal.
The* diffirult pmlilcm of its (|uaiititative im|Mirtanre is attackeii and
the ronrlu*«ion drawn that no intrusive lK>dy is too large to be explained
as the wtirk of th«* primary liasaltie wedge interacting on the two shelU
overlying the l»a>altir substratum.
Chapter XII is a general outline of the more im|M>rtant phages of
magmatie differentiation. The mixed magma.s due to assimilation
(*'synteetif*s**) anil, under certain conditions, the primar>' liasalt
itself a!e not ^*tal>le s^ilutions l)Ut break up into submagmas. These
"non-c«»nsulute*' fraction'^ are 8(*gregated in two chief ways: u.««ually
by the direct action of gravity; and, on a much smaUer pcale, thmugh
the upward transfrr of magmatic fi^ases which have brought together
silicate or oxidt* materials ** entangled *' with the rising gases. In
general, the unit of difTen^ntiation is a small licpiid mass and true
fractitinal cry^tallization is r«-garde«| xx< a very sulionlinate mode of
magmatic splitting. A general statement is given of demon.strate<l
splitting in volcanic vents, sills, laccnliths, and lijitholith.^. The
most instructive illustration are those derived from sills and laccoliths,
a numlHT of wliich are taliulated, with concisi* statement of the
facts.
Chapter XIII treats of the theory of volcanic acticm at central
vt*nts. Among thi* topics consid«*red an*: the localization ami opening
of the vent ; the jxT^istenci' of il«* erupt ivity for long |M'rio(U: the alter-
nation of active and tiormant stages; the rhythmiccharacter of eruption
during an active stage; tin* origin of the hc-at in the vent: the rate of
he:it-loss during ai-tivity; tlie principle of ** two-phase convection,"
wliich is held to U» the chief cause of the tran^ifer of heat fromthedeptha;
the systematic chauKes during the life of a central vent, with nwpect
U\ expl(»siv«-nc^'« and t4» the jx'troRraphic nature of its lavas: the origin
of block lava and of ropy lava: the cause of lava outflow; the genetic
cla»itication (»f volcanic gases: the iiistinction Ix'twei^n magmatic and
phreatic explo-^iini^: and that b«'tween "principaT* and •'suliordinate"
volcan<M*s of tht* ri-ntral ty(>e.
ChaptiT XIV •*unnuari/e<^ thi- Keiif-ral t henry, which is se<*n to be
eclectic in character >ini*r it includes th«* id«*as of manv workers in
|H*trogeny. Thr nnly nthcr ruiMlrrn attrinpt to form a stable theory
of approxiinatfly **iniil:ir >c(ipe i<« tliat of L«M'winson-Ii4'ssing. whoiie
|iosition is brii'tly di-cu-^seil.
I'hapter XV o|>ens the third and concluding part of the volume.
ABSTRACT 5
The members of the gabbro clan are here listed and some of the more
important are considered in their relation to the eclectic theory.
The composition of the substratmn basalt is approximately calcu-
lated, and in succession, the olivine-free basalts, the quartz diabases
and their allies, the norites and related types, the hornblende gabbros,
iron basalt, the anorthosites, and the pillow basalts and spilites are
described in their genetic relations.
Chapter XVI discusses the granite clan. The origin of the granites
is most clearly indicated by the facts known concerning many sills
and laccoliths which have invaded sedimentary rocks. Assimilation
combined with differentiation has there produced granitic types
which are generally of somewhat abnormal composition. The
abnormality is that expected by theory. The genesis of rocks showing
the usual granitic composition has been considered in previous chapters.
They are generally gravitative differentiates in gigantic abyssal wedges
which are walled principally by the primitive acid earth-sheU and are
cross-cutting bodies, but otherwise are perfect homologues of large,
more fully exposed laccoliths and sills. Many rocks of granitic com-
position (chemically speaking) are clearly differentiated from dioritic,
granodioritic, syenitic, or monzonitic magmas. Gaseous transfer
(pneumatoly tic differentiation) is held responsible for the development
of certain small-scale bodies of aplite, pegmatite, liparite, etc., from
intermediate and even subsilicic magmas.
Chapter XVII, treating of the diorite clan, outlines the facts show-
mg a double mode of origin. Most pyroxene andesites are concluded
to be direct differentiates of the primary basalt. Many diorites
appear to represent syntectic or mixed magmas, such as those normally
expected by the solution of rock from the acid earth-shell with the
jHimary basalt. Mica andesite and hornblende andesite find their
theoretical place as either syntectics or, more commonly, differentiates
of syntectics.
Chapter XVIII contains an abridged statement of the relation of
the eclectic theory to the granodiorites and their allies (tonalites,
many quartz diorites, many dacites, etc.). They have the same origin
as that of most granites, but differ from the latter rocks chemically
because of the large amounts of argillaceous sediments and associated
mediosilicic rocks which, together with the acid earth-shell, have
been assimilated in granodioritic batholiths.
Chapter XIX indicates the great variety of species included with
the syenites proper and their allies. It is recalled that these types
never form the principal rocks in very large bodies, implying that their
parent magmatic wedges were small. A table, found in Appendix C,
illuBtrates the rule that members of the syenite clan are very generally
n KiS'EOl'S HfH'KS AM} THEIR ORIGIS
iTUptivf into nH*(lioNilirir or hulir^ilirir NMlimrnt.s. On account of \\w
f«p€*rially gn*at fluxinfc ixiwcr of ^urh H*fiinH*nts, thnr awimilation
nrar thr top of a narrow aliys>al w«(l)C(* must trnd tf» counteract the
aci(iifirati«in tiuc to Milution of the arid carth-shrll U»noath. The
final diffcrfntiatc of th<* wh(»lc mixturi' should. th<Tc*fore, be k*iv
sihciouH than that normally cxiMTtcd in a gn'at«T wedg<*. which, on
account of its larK«*r supply of heat (lcss^uddcn^hillinKnt contact8.etc.\
diw*o!vcs relatively more of the thick aeiil shell. Svenitc ij* thus in-
t<*rpr<*ted as a (h*^iiicated Kninite. The field and chemical relationj« of
the Hyenites are found to corresfMind to the theory. The influence of
the volatile matter ahsorlNMl with or from the S4*diments is emphasized.
A few sills, showing syenite :i.«i a .«imall-scale differentiate of lia.«(altic
magma invading basic s<*diments. are n'garded as excellent corrol)ora-
tions of the general theor>'.
In (*hapter \X nearly one-third of the recognized igneou.H-rock
.•«fN*cies are c<msidered togethtT, under the name ** alkaline clan.«."
Thes4» include most of the so-called ** alkaline" rocks. Thev are ex*
plaimnl liy the same principles as tho>e usi'd for the granoiliorite and
syenite clans. As a rule, the alkaline rocks are difTtTentiateii from
mixe<l magmas which an* controUed in their comfxisition and in their
splitting by abs<»rlN*d carlninate MMliments. Herause of its infinitely
low c<mtent of silica, liniesttme or dolomite must tend to defiilicate
markedly the total s<ilution in an aby»al basaltic wedge. Herein is
the preferred explanation of the chara(*teri>tic crystallization of mineraU
like leucit<'. nephelite. S4Mlalit<*. etc.. in alkaline rocks. Vet more
signally than witli the syenites, thi* highly alkaline rocks show the
exi>e4*t<Ml influener of gaseous transf«T in segregating submagma4i.
(*arlH»nat<' control in th«* formation of mo>t alkaline rocks is strongly
suggeMed by th«* tablr of their fiehi relati(»n<. given in App«*ndix D;
bv the mineralogy and chemi>try of the rorks; and by the relatively
small vttlume assignalih* t4» every p'corded Inidy of this kind. A
multitude of facts >ub>tantiate the thesis that thecarlNinatesvniectica
have U^en fonned in magmas which were originally of basaltic com*
|H»sition. However, a** eX|H'rted by the tlu'ory. some alkaline rocks
an* manifcMly due tf> >egregation of MMlinientary origin by water-gas
and it i** probabh' that the "juvenile" or primary gaM»s of the sub-
stratum material hav<* >imii:irly functioned in the dey<*lopment of
some alkaiin<*-rork lH>di«'s.
Chapter XXI ci>ntains a short sk(*tch of the |)erid<»tite clan (in-
clu<ling the pyro.xenites and horid)lendit<'s) anti the niagmatic ores.
The ech'ctic theory explain** their very common ficM relations to mem-
lH*rs of the gabbro clan; the rorks an* interpretetl a.s, in part, direct
differentiates t»f ba.Htdtic magma; in (»thcr |mrtp they are diffi
ABSTRACT 7
tiat€s of syntectics. The splitting is often clearly gravitative. In
many cases, however, the segregation has been more or less evidently
due to gaseous transfer from intrusive magmas, especially those af-
fected by the solution of sediments.
The last chapter sketches the result of matching the eclectic theory
with the geology of the North American Cordillera and thus with a
very extensive assemblage of igneous clans. This region offers all the
important problems in petrogenesis.
PART I
CHAPTER II
CLASSIFICATION OF IGNEOUS ROCKS
The greater part of the earth's visible rock-matter is crystalline;
only a minute percentage is composed of glass. The nature and
relative proportions of the constituent crystals or minerals determine
the essential nature of each holocrystalline rock. Actual mineralog-
ical composition is a natural basis for a classification of the rocks and
it must always remain the working basis for field classification. Yet
there are two chief difiiculties standing in the way of a perfect applica-
tion of this principle. No direct method has ever been devised for the
accurate measurement of the proportions of minerals in fine-grained
rocks, nor is it likely that such a method is at all possible. Secondly,
even if such a measuring device were in hand, the results of its use must
be imperfect, since, with few exceptions, each mineral species in rocks
is itself of variable composition. The principal minerals generally
occur in "mixed crystals." Each of these is composed of mixtures
of two or more different molecules, and the proportions in each mixture
form an infinite series within the chemical limits set by the pure mole-
cules. Examples are now familiar in the highly important feldspar,
pyroxene, amphibole, and mica families. Quartz is the only principal
constituent of igneous rocks which always shows the same composition.
These facts have long been recognized by petrographers and the basis
for an ultimate classification is now universally found in the chemical
analysis (total analysis) of the rocks.
Mode and Norm Classifications
The leading petrographers of Europe have not been disheartened
by the general failure to read out the exact nature of an igneous rock,
that is, its chemical composition, from its mineralogical composition.
They have shown abimdantly that there is usually a certain degree
of corresi>ondence between the mineralogical composition of a rock
and its total analysis, so that, in most cases, a general idea of the one
can be obtained if the other is known. The world leader, Rosen-
busch, has prepared an elaborate classification founded on mineralog-
9
1(1 ids KOI S horKS AM} rilKIH OHH;l\
irul r(inip<iMtifin. wliirh Iihs In-^ti hir^fly kept ijiifU*r tlu* contml of
rhemical aiialvsi**. This «*vsi<-iii injiv U- nillfil the "Moilc" rhis.M-
firatidii. ^in^(* it i*« lta>c(i on th«- artiril infMii* in whit'h the rcmstitui-nt
niiiiiTals an* afcer^Katfil in the nwk fahrirs.
On th<' otliiT han<l. certain Anirrican iN^trD^raphors hav<* rut thr
(inniian knot l>y i^norin^ tli*' "n)«i<l«*'* in tlirir primary rl:is*»ifi<*ation.
In>tra<i of t)i«' actual niinrraN. "stan<iarii'* minerals or ninli-rul«-<4
arc ralrulat«*<i from the total anal\>«*s. an<i tlu* ro<*ks an* rlsi.s>ifn'<i
m
arronliuKto the iiatup' of tli«' ''norm*' or whole ^roup of ^tamlard min-
rraUcait'ulatefl fur each rork.* This>v>tem mav lH*ralleil the *' Xonn"
clas^sifiration.
The n-aMer must Im* referreil to other work** for ilisru-«sion jim to thi*
relfttiv«* merits of these two >V'<tem»i.* One or two remarks onlv will
here Ih* ofTer«'<l tu **UKKe*it thi* full rea^m why th(* Mo<le ei:i.ssifieation
will Ih* u>eil in this lHM>k.
In the fir>t plac-e. the Norm system, as announced and prartiM*d
hy it> authi»rs and hv a eonsiderahle numU'r of followers, is laricf'ly
founded not oti proved facts, l>ut on assumptions concerning chemical
"affinities," and the march of molecular f4trmation in natural maipnas.
The >uh>tructure of the system is thu«» in«»ecure; yet very .-liRht chanKon
in this >uhstructure mu>t entail dra>tic chani^es in the pres^'nt classi-
fication its«'lf or ruin it entirelv. In a word, the Norm svstem in
almost sun* to In* found too srn**itive to the discov«Ty of new faeis
concerning mi»lecular (h-velt»pment in >ilic:it«* melts. We may he
reasonahly >ure that pre^-nt opinion re^ardin^ this fundamental
matt<*r i** lH»und to In* m(»re (»r ie**s rhanR«-il in tin* near future. The
danger is ^n^at that the inuenious hut much t4M) (h'licate Norm elaiwi-
ticatiun. together with a unat ma**s of time-consuming labor in
calcuhition** and dc'^criptions. will turn out to have litth* |K*rmaneiit
value as a >V"«tem. however valuahle its invention has U'en in Mtimulat-
infc |N-tro)(raphic thou^lit aion^ <'ht'mical lines.
.Xycain, thi-* >y^tem i> defective, as i-ven its minor sulMli virions
intlivitluallv include ruck tvpfs which are^tron^lv contrsLsted chrmic-
ally anti M-parate other> whidi are almost alik«* chemically, mineralogje-
ally. and genetically. Kxampli > may In- takt-n almost at randum. On
the '•ame paue t»f \Va>hin|kjton'> tahles i p. \i\\ of Prof. PajHT 14, U. S.
> VN<if,ri.'<j.'ii* ('('•i«-i/ti-ij/iii>i 11/ /i/»,#if<.i A\h^.. |iy W. t'msj*, J. p. IfldinicB, L. V.
PirsMHi. ]in«l II S W:i.*!iiiii!ti»n. <'hiiMi!i». VM\'.\ ."^■«' i-iiiHM-ally pA|[f* • "' for
ilrll lilt loll.** iif "liliMlr" :ti)i| "linnii
•A Harkrr. 7''r .\.i/ir,j; //i.,:,.- 7 ..^ l^jum tA fiiH-k<, Niw Vnrk. 1M»U. p. 302, aM..
Citil M:iK.V..l 10. l!io.{. |i 17.;. I W (*l:irkf. Mull till, t' S(;i-«il. Survey, 1911,
p 4at; W ('ruHM. i^it.irt J.mr (ifl Sn- , v,.| tWi. p.iin. p 470: II. S. WMhiaglo^
rruf P»p.T II. r S ('..-..I Siir\.'v. P.NH. p 19; J. V McJinipi, Igneous Rock9, VoL
I, \rw Y«»rk. P.HN. p 4U7
CLASSIFICATION OF IGNEOUS ROCKS 11
Geol. Survey, 1903), which are based on the Norm classification, we
find that the subrang, "toscanose," includes types containing princi-
pal oxides as follows:
Aplite
Trachyte
SiO,
76.03
62.33
Al,0,
13.39
17.30
Na,0
2.98
4.21
K,0
5.18
4.46
On page 320 of the same work, the subrang, "camptonose," includes a
typical camptonite with 43.98 per cent, of silica and a quartz basalt
with 54.56 per cent, of silica.
In the Rosita Hills district of Colorado two syngenetic rhyolites
give these analyses:
Round Mountain
Silver Cliff
SiO,
75.20
75.39
A1,0,
12.96
13.65
Fe,0,
.37
.38
FeO
.27
.18
MnO
.03
.14
MgO
' .12
.15
CaO
.29
.51
NajO
2.02
1.84
K2O
8.38
6.81
H,0
.58
1.13
100.22 100.18
The Round Mountain rhyolite appears in the Norm classification as
a member of the subrang, omeose, of the rang, liparase, in the persalane
order, brittanare. The Silver Clifif rhyolite falls in the subrang,
magdeburgose, of the rang, alaskase, in the persalane order, columbare.^
These few illustrations sufiice to show that this system disregards
vital principles of scientific classification. From the standpoint of
the geologist it is like a zoological system which would place in the
same species the Newfoundland cod, the North Sea herring, and the
Louisiana alligator, while specifically separating the New England cod
from the Labrador cod of slightly different size. As one of the authors
of the Norm classification remarks, ''the norm is primarily a means of
comparison, and has in itself nothing to do with system."^
T^ 3 field geologist has little choice as to his method of rock classi-
fication. He must judge a rock by its actual mineral constitution.
The scale of his operations is usually much too great that he can
hope to secure, from government laboratories or otherwise, the
' H. S. Washington, op. cit., pp. 125 and 143.
«W. Cron, Quart. Jour. Geol. Soc, Vol. 66, 1910, p. 496.
3
12 IGSEOCS ROCKS ASD THEIR ORIGIN
numlior of chemiral aiialvAcs requif^ito to map his igneouts rocks by
the Norm HyHtem. If he did possess unlimited access to the resources
of ehemiral lalM)rutories, his norms calculated for the difTerent phases
of the uveraffi* iHrK«' InhIv of i^nrnius rock would Ik* found to correspond
to many >|MTi«*s in the pn*«f*nt Nonn classification. Actual experience
has already shown that so many sp€»ci(*s apiH*ar in such a IxMiy that
their mapping is impracticable. Further, the prc^sent system contains
no comiK'Hing principle to K^ide the fieM geoloKist in combining the
** sp<*cies *' of this classification into larger units suitable for the practical
geological mapping of the world. In any ca.s<*, the Norm classification
as now develo|MMl is not workable in the practical mapping of the
larger rock iHxIii-s enc*ounten*d in n^ular government surveys. On
the other hand, the Minle classification has borne this test, on the
whole, V€*rv well, mi tlnit many esH*ntial facts i*oncerning the igneous
rocks of the world iire already amply nM<orde«l in existing maps and
memoirs.
This i*«, in reality, the obvious ground for preferring the Mo«le
classification as a ba>is fur the pn*s4*nt discus>ion of igneous rocks.
Through it the facts of nature have Imm^u recorded and only in that
H'cord can th<* material Ih* found f4»r a svnthetic studv.
Rosenbusch System. Tin* Mode classification has Inh^u issued,
so tosfx'ak. in several editions. Zirk<'l, Uos4*nbusch. Michel L6vy,
I^H*wins4Ui-Ii«*ssing. and others have constructed as many systems,
difTering in tietail but all foundeii <m mineralogical com|)osition
supph'menteti by chemical analysis. The latent and most compre-
hensive svsteni. that of KoM*nbusch. has Imn^u the edition most ex-
tensively useil liy g(M»logi<.ts aufl they hav<* used it with fair consistency.
It will Ih* emplttyed in tlie f4»Ilowing chapttTs Inith in the quotation of
facts liiul in the di*icu^>i4in of those fact<. For a description of
Uosenbusch's •ivsiem itself the ri*ader is rt*f erred to his two masterlv
Works, the " Mikro**kopisrhe Physiographic der Massigen Ciesteine,'*
vierte Autiagt*. StuttKart. 11H)7 t)8. and the •*Kh'm«'nte tier Ciesteins>
lehre.** dritte .\utlag«*. Stuttgart. IIMO. In many essential r«v
>IMM'ts the >y«.tem«* di'scrilwd in ZirkePs great lichrbuch der Petn^
graphii*. in Teall's classic work on British Petrography, and in
Idding>*s nion* n*cent InNik on I^iu'ous Itocks are similar to that of
Ho<4'nbu'*cli.
As Hi»<>eiiliusch himself, with >ucc«->>ive editions of his handlKKiky
ilevelo|HM| and cliangeil lii< rIa**>ifieation. so we must ex|M*ct the new
knowN'dge of the future to modify the ^y^^tt-m. To lM*come an ideal
Mode clarification it >hiiuM be made (piantitative. For the plutonic
sfMH^ies ^gt'iietically by far the mo»t im]Hjrtant) Hosiwars method of
CLASSIFICATION OF IGNEOUS ROCKS 13
optical measurement has made an approach to this ideal already
possible; and the writer believes that Rosenbusch's present classifica-
tion of the plutonic rocks could be made quantitative without an
intolerable amount of changes in his definitions.
Adopted Mode Classification. — However, we are here not primarily
concerned with the present merits or future prospects of this system.
The important point is that it is the vehicle which bears most of the
facts now known about the character and distribution of terrestrial
rocks. As a convenient guide Table I, based on F. E. Wright's state-
ment of Rosenbusch's classification, has been prepared. At some points
the table differs from that which would simimarize his hand-book with
perfect fidelity. For example, the granodiorites are here recognized
as forming a distinct family. This has been done partly in order to
emphasize the enormous importance of the granodiorites and their
allies in the two Americas. A thorough-going division into an
alkaline series and a subalkaline (lime-alkali) series is not made, for
two reasons; first, to save space, and secondly, to emphasize the
writer's belief that there is no petrogenic reason for making so formal,
clean-cut separation. (See Chapter XX.) In spite of such subjective
elements in the table, it is offered as a mnemonic aid in appreciating
the actual diversity of igneous rocks.
Average Composition of Igneous-rock Species
On the chemical side the classification is illustrated by Table II.
Its nature is explained in a few paragraphs here quoted from a publica-
tion issued by the writer in 1910.^
Since compiling the averages stated in the 1910 paper, the writer
has been able to improve certain of them. In most cases the number of
analyses averaged for each of these species has been increased, and
often a better selection has been possible through the use of additional
material. In both respects the recent (third) edition of Rosenbusch's
"Elemente der Gesteinslehre" gave exceptional aid in the preparation
of the new averages. The list of types for which the averages have
been improved includes rhomb-porphyry, Iherzolite, pyroxenite,
wehrlite, harzburgite, dunite, all peridotites, picrite, essexite, trachy-
dolerite, iimburgite, ijolite, Hawaiian basalt, and melilite basalt.
Averages for nineteen types, n0t entered in the 1910 table, have been
* R. A. Daly, Proceedings of the American Academy of Arts and Sciences, Vol.
45, 1910, pp. 211-240.
mSKfirs HtH-KS AM) THEIR ItRHllS
lIU II
111 II
i If
J I
i 1
\i i1
ji
^1
S--
XS
ii! =:J
;:^i
«ll IL
nil
I ,
= i:
UU?HliiiUfltliill fill
i;;
CLASSIFICATION OF IGNEOUS ROCKS
£
1
1
1
-
|:|| 3s|l|.
llipiis
iirf iHi
4
1
tl
!«
1
i
i
%
i
|K
\
\
J
■J
1
4
1
1
1
It
j
i
'
1
1
1
|IIIU
Isls
ill
a.
,
1
+
,
1
s
i
3
s
Si
1
1
a
16 IGSBOVH ROCKS ASD TUBIH ORIGIN
inserted in Table II. ThcM* an*: alkaline granite, Hubalkaline granite,
comendite, aulmlkaline syenite, subalkaline hornblende syenite,
subalkaline mica syenite, subalkaline augite syenite, umptekite,
alkaline trachyte, subalkaline tra(*hyte, pantellerite, tonalite, quarts
monsonite, mica p€*ridotite, amphil)ole peridotite, ''pyroxenite of
the alkaline Maries." "pyroxenite of the subalkaline series,'' be-
kinkinite, and melilite-nephelite basalt.
Rosen)>usch and his followers recognize some latitude of variation
in the composition of each rock-t^'pe. The variation is both mineralog-
ical ami chemical, two rock specimens referred to a type showing
differences in the pro[K>rtions of the chemical elements found by analy-
sis of the two ro<*ks. In fact, no two analyses of granite, andesite, or
any other one ty[N* have ever given prf*cisely the same proportions of
the dosen or more oxides which regularly make up an igneous rock.
It is o))vious that the student of map and memoir should, for many
problems, have at hand the actual figures showing the most typical
chemical composition of the rock-types to which his study is directed.
In numerous cases an analysis of a single specimen is not so useful as
that which could Ix* made from a thorough mixture of specimens of the
same roi'k-variety from all places on the glolie where that variety
occurs.
For obvious reasons such ideal analyses have never l)een made. In
their stead the writer )H*lieves that the investigator of petrogenic and
other world problems may well use the averages calculated from the
manv excellent chemical anidvsc^s of r(M*ks made since Rosenbuscb's
system of naming and classification has lH*en in general use. It may,
indt*<Ml. I>e arguni that such averages would more nc^arly represent the
chemistry of Kosenbusch's typ<*s than the corresponding individual
analyses which he has publish«*d in his treatise. These averages would
be ch(*mical ''center-points** in his system of classification as aduaUjf
applied to the terranes of the worhl.
So far as the writer is aware, the preparation of these averages has
not hitherto l>e(*n attemptetl Xo such an extent as to cover the chief
familic*s and species of igm^>us ro<*ks. .\n approximation to the
desired results is offered in the following table.
The work of computing the averagt^s has l)een h»ssened ver>' greatly
by the publication of i >srtnn*s *' Heitriige zur chemischen Petrographie"
(2nd fmrt, Stuttgart. 11NK>). This remarkable lHM)k contains, in con-
venient arrangement, th«* statement of most of the eruptive-rock
analys€\< (over 2f(M) in nunilier) published in the inter\'^al between
1883 and liX>l. The {mtkhI of seventivn years lies within that during
which systematic iH*trography has Ikh^u dominated by Kosenbuscb't
CLASSIFICATION OF IGNEOUS ROCKS 17
names and definitions. In general, the number of analyses for each
rock-species is so large that their average would be but slightly modi-
fied by the inclusion of the analyses made since 1900. In many cases,
therefore, the extended labor recjuired to search out from the literature
the additional analyses has not been considered necessary for the
preparation of useful averages. For other averages it was necessary
to include analyses published since 1900. The sources of such infor-
mation are indicated below. Fortunately for the purpose, nearly the
entire period since 1884 has seen the application of more or less re-
fined methods of analysis; so that errors of observation for the leading
oxides are relatively small.
The method of computation used is essentially like that employed
by Washington and Clarke in their respective calculation of the
"average composition" of all igneous rocks. In general, only the
twelve more important oxides (including MnO) are recognized in the
following tables. Distinctly "inferior" analyses were not considered.
In each case the average was computed according to the actual numbers
of determinations made by the analysts. Table XX, appendix A, shows
these numbers for the respective rock-types, each column being headed
by a key-number which corresponds with the named types of Table II.
For some of the rocks BaO and SrO were computed. Their sum
appears in the averages for CaO, as indicated in the tables. Similarly,
COj and CrjOs were sometimes averaged and entered with H2O and
FeiOs respectively. As expected from the method employed, the
average totals nearly always ran well over 100 per cent. All
averages were reduced to 100 per cent, and entered in Table II.
Each average analysis was then recalculated to 100 per cent, after
HjO (and CO2) had been subtracted. The results are also given in
Table II, in which plutonics and corresponding eflfusives are grouped
together. Magmatic relationships are often less obscured if these
volatile oxides, which may be wholly or in part of exotic nature, are
excluded. Finally, in order to facilitate reference to the tables, an
index to the diflferent rock-types was prepared and may be found
below Table II.
It will be observed that certain rock-types have been omitted from
the tables. The large class of " aschistic " dike-rocks is not represented
because of their chemical similarity to the corresponding plutonic
species. Other named varieties are omitted since their analyses are
too few to give useful averages. In a few cases the mineralogical and
chemical variations within each variety are so great that it has not
seemed advisable to regard their averages as worthy of entry. Many
other subordinate varieties of rock, though given special names, are
18 IGSEOUS ROCKS AND THEIR ORIGIN
rhemically almost idontical with the more important types entered in
the tables and therefore have been exrhided.^
* The iinmediate tiuiirces of the analytical statements used in the oomputatiooB
are as follows:
1. Beitragc zur (.'hemischen Petrographie, zwciter Tell, by A. Osann. Stuttgart,
1905.
2. Chemical AnalysoH of Igneous Rocks published from 1884 to 1900, by H. S.
Washington. Prof. Paper, No. 14, U. S. Geological Survey, 1903.
3. Elemente dcr Gesteinslchrc, 2n(l and 3rd editions, by H. RosenbuBch. Stutt-
gart, 1901 and 1910.
4. Studien iiber die Granite von Schwedcn, by P. J. Holmquist. Bull. Geol.
Institution, University of Upsala, Vol. 7, 1906, p. 76.
5. Some Lava Flows of the Western Slope of the Sierra Nevada, California, by
F. L. Ransome. Ainer. Jour. Science, Vol. 5, 1898, p. 355.
6. A. Lacroix, Mat^riaux pour la Min^ralogie de Madagascar. Nouv. Archives
du Musdum, (4), Vol. 5, Paris, 1903.
7. Geologj' of the Yellowstone National Park, by A. Hague and others. Petrog-
raphy by J. P. Iddings. Monograph No. 32, Part 2, U. S. Geological Survey,
1899.
8. Analyses of Rocks from the laboratory of the United States Geological Survey,
1880 to 1908, by F. W. CUirkc. Bulletin 419 of the Survey, 1910.
9. Geological and Petrographical Studies of the Sudbury Nickel District, by
T. L. Walker, Quart. Jour. (Jeol. Soc., Vol. 53, 1897, p. 40.
10. Petrography and Geology of the Igneous Rocks of the Hig&wood Mountains,
Montana, by L. V. Pirsson. Bull. 237, U. S. Geological Survey, 1905.
11. Geology of the North Ameriran Cordillera at the Forty-ninth Parallel, by
R. A. Daly (Memoir No. 38, Geol. Survey of Canada, 1912. pp. 793-799).
The sources of the analyses used in each average are indicated by the authors'
names at the head of the corresponding columns in Table II. In the future all
the averages can be improved by considering also the thousands of analyses
published since 1900 and not here employed. Such new averages can be advan-
tageously made if the data of Tables II and XX are combined.
CLASSIFICATION OF IGNEOUS ROCKS 19
TABLE II.— SHOWING THE AVEHAGE COMPOSITIONS CALCULATED FUK THE
PRINCIPAL IGNEOUS-ROCK TYPES
GROUP I
m
^X'^il
AliO,
FcO,
CaO
Brf)
■SI
71.06
69. SI
69.73
.48
.54
.34
14.10
13,76
14.98
1.4fi
2.17
1.62
1.63
1.87
1.66
.18
.26
.11
.59
.84
1.08
1.97'
2,20
2.20'
3.24
3.17
3.28
4.50
4.38
3.95
.69
.74
.78
.10
.26
,27
14.78
1.62
1.67
II
1^
72,60 72.90
72.62 72.36
,26 .33
13.77 14,17
1.29 1.55
.90 1.01
2.85
4.56
i.og
Calculated as Water-free
SiOi
71.56
70.33
70.28
70.47
73,72
73.89
73.76
73.16
TO,
.48
.54
.34
.30
.30
.49
.25
,33
A1,0,
14.20
13.86
15.10
14.90
14.10
14.37
13.99
14.33
Fe^.
1.47
2.19
1.63
1,63
1,45
1.67
1,31
1.S7
FeO
1.65
1.89
1.67
1.68
,83
.31
,91
1,02
MnO
.18
.26
.11
.13
.12
.13
.12
.09
MgO
.59
.85
i.og
.98
.40
.41
.39
.63
C*0
1.98'
2.22
2.22"
2,17'
1,34
1.14
1.46
1.39
Na^
3,26
3.19
3,31
3.31
3.59
3,59
3.60
2.88
K^
4.53
4.41
3.98
4.10
4.09
3.99
4.16
4.61
P^.
.10
.26
.27
.24
.06
.01
.07
.09
'Includes .08 per cent. BaO and ,01 per cent. SrO,
■ Includes .06 per cent. BaO and .02 per cent. SrO.
' Includes ,06 per cent. BaO and ,02 per cent. SrO.
L
ao
IGNB0V8 BOCKS AND THBJR ORIGIN
GROUP II
Plutonic
Effusive
9 10
11
12
Subalkaline Alkaline gran-
Comendite
Quarts kerato-
Rranite ite
(RoeenbuBch) (Roeenbusch)
(Rosenbusch)
phyre
(Rosenbusch)
No. of analyses
20
12
6 13
8iO,
69.21
73.30
74.44
75.45
TiO,
.41
.11
.19
.17
A1,0,
14.41
12.33
11.27
13.11
Fefiu
1.98
2.58
2.78
1.14
FcO
1.67
1.28
.94
.66
MnO
.12
.02
.08
.29
MgO
1.15
.26
.35
.34
CaO
2.19
.46
.21
.83
Na/)
3.48
4.55
4.18
5.88
K,0
4.23
4.20
4.95
1.26
H,0
.85
.86
.59
.69
P,0.
.30
.05
.02
.18
Calculated as Water-i
69.82 73 94
'roe
SiO,
74.88 75.98
TiO,
.41 .11
.19 .17
A1,0,
14.53
12 44
11.34 13.20
Fe,Oi
2 00
2.60
2.80 1.15
FcO
1.68 ' 1.29
.95 .66
MnO
.12 , .02
.08 .29
MrO
1.16 : .26
.35 .34
CaO
2.21 1 .46
.21 ; .84
Na,0
3.51 4.59
4.20
5 92
K,0
4.26 4.24
4.98
1.27
P,0.
.30 .05
.02
.18
Each sum - 100.0
0
CLASSIFICATION OF IQNEOUS ROCKS
Plutonic
Effusive
IS
14
IB
IS
M 1
17
18
19
^1
4
S 2
:i
c
No. of
115
3 II
is
•3
^gO
°1
r
i
malyaes
7
2
2
11
50
48
10
SO,
60.79
59.25
51.59
68.65
60.19
60.68
63.91
TiO,
.80
.79
.61
.86
.67
.38
,59
AW,
16.10
15.28
18.77
16.38
16 28
17.74
15.88
FnO.
3.21
2.59
6.11
3.65
2.74
2,64
3.22
FeO
2.92
3.47
3.26
3.09
3.28
2.62
2.23
M»0
.11
.24
.15
.14
.06
.01
MiO
2.20
5.07
4.11
3.06
2.49
1.12
1.14
CtO
3.87
3.68
7.35
4.46
4.30
3-09
2-81
S^O
3.37
3 10
4.35
3.48
3.98
4.43
3.08
Kfl
5.43
4.41
2.99
4.79
4.49
6.74
5.80
Hrf)
.90
2 06
.28
1.13
1.16
1.26
1.28
PA
.30
.30
.36
.31
.28
.24
.05
Calculated aa Water-free
SiO,
61.34
60.49
51.72
59,32
60.90
61.46
84.74
TK),
.81
.80
.61
.87
.68
.38
.60
Al/),
16.25
15.60
18.82
16.66
16.47
17. B7
16.09
Fe^.
3.24
2.66
6.13
3.60
2.77
2.67
3.26
FeO
2.96
3.56
3.27
3.13
3-32
2.66
2.26
MnO
.11
.24
.15
.14
.06
.01
MgO
2.22
5,17
4.12
3.10
2.52
1.13
1.15
CO
3.90
3,76
7.37
4.50
4.35
3.13
2.85
Na^
3.40
3.17
4.36
3.52
4-03
4.49
3-12
K^
5.48
4.50
3.00
4.85
4.54
5.81
5-87
P^.
.30
.31
.36
.31
.28
.24
.05
la.SEOIS RIKKS AM} THEIR ORiC.IS
Plutonic
i
II
11
i
1';^';
£
!
lis
a _
' 1
1
No- of
•>.
<=-€
2
wXyM*
7
5
8
5
23
19
7
12
SiO,
M 3*1
til M
61 96
fiO 01
61 99
62 63
61 51
68 «3
T.().
4.".
15
W
A4
56
62
45
.35
AW.
16 HI
It) 07
17 07
It) t>5
17 93
17 06
17 37
10.30
IV^).
1 m
J (M
2 35
2 41
2 22
3 01
1 92
a.<o
Fr4>
■2 71
1 4B
3 37
3 H5
2 29
I W
3 35
2 61
MnO
IS
01
.M
IS
09
13
01
-21
Mr«J
7J
M
I 3tt
97
.96
63
1 2«
37
r»(i
1 .S5
1 47
3 41
2.62
2 55
1 51
1 W
1 07
x.fl
4.7rt
« 45
4.65
6 5.3
5 54
6 -26
5 23
6 14
K,ll
a «2
5 75
3 Ml
5 47
4 98
5 37
5 29
4 17
ll,l>
,70
.47
93
50
76
71
2 45
63
p/).
<»
OH
17
14
09
OR
.02
SiO,
A4 HI
tii 15
tl2 55
fiO 32
62 4ti
63 09
63 06
69 00
TiO,
45
15
1 00
tu
56
.62
.46
.is
Mih
Ifi M
Ift Irt
17 23
16 74
IS 07
17 IH
17 81
10 36
y-jo.
t.<»
2 tWl
2 37
2 42
2 24
3 03
1 97
6 63
M>
■2 7:i
1 SO
3 40
3 H7
2 31
2 00
3 43
2 62
MnO
u
1)1
00
IS
OH
13
.01
21
MKI
73
55
1 3»
97
97
63
I 29
37
(■•o
1 sa
1 48
3 44
2 ta
2 57
1 52
1 II
1 08
X.,C)
5 so
0 4H
4 119
6 5«l
5.58
6 30
5 36
ft. 17
Krfl
5 00
5 7H
3 S4
5 50
5 02
5 41
5 43
4 19
p,o.
.00
(W
17
14
09
08
02
F:arh><>im-IOUl|(>
CLASSIFICATION OF IGNEOUS ROCKS
23
GROUP V
Plutonic
Effusive
28
29
No. of
analyses
SiO, ~
TiO,
A1:0,
Fe,0,
FeO
MnO
MgO
CaO
Na,0
K,0
H,0
P.O.
Laurvikite
(Osann)
Rhomb-porphyry
(Washington)
Plutonic
30
Monzonite
(Osann and
(Washington)
SiO,
TiO,
A1,0,
Fe,0,
FeO
MnO
MgO
CaO
Na^
K,0
PiO.
57.45
1.06
4.10
5.89
3.87
.70
.54
57.85
1.07
4.13
5.93
3.90
.54
7 ,, 12
56^36 I' 55.25
.48 .60
20.10 16.53
2.86 ' 3.03
2.01 4.37
.01 i .15
1.15 i 4.20
2.73 i 7.19
7.65 3.48
4.97 I 4.11
1.20 .66
.48 ' .43
Calculated as Water-free
^Includes .16 per cent.
'Includes .14 per cent.
57.06
55.62
.48 1
.60
20.35
16.64
2.90
3.05
2.04 1
4.40
.01
.15
1.16 !
4.23
2.76 1
7.24
7.74
3.50
5.01
4.14
.49
Each sum = 10(
.43
).00
BaO and .07 per cent. SrO.
CO2.
Effusive
31
Latite (Ran-
some and
Daly)
10
57.65
1.00
16.68
2.29
4.07
.10
3.22
5.741
3.59
4.39
.91*
.36
58.18
1.01
16.84
2.31
4.11
.10
3.25
5.791
3.62
4.43
.36
34
lONBOUS ROCKS AND TUBIB OBIOIN
M
Plutonic
SS $4
— ■ —
EffunTc
U
ST
U
4 w
v^a V *
Leuciie
Leoctt4>-
No. of
analy-
royaite
(Onann
and
UoM*n-
bujirh)
Trtitr
((Mann)
I^urda-
lite
UHmnn)
Nephe-
litc «ye-
nite
(Osann)
Phonolite
(Otann,
Clarke,
and
Lacroix)
phonolite
(Otann
and
Waahint-
ton)
phyre
(WarfiiDC.
1 too and
ROMH
bnarh)
H^M
10
3
3
43
25
4
8
8iO,
50 11
4.5 61
.54 36
54 63
57 45
54 80
40 83
TK)t
4.5
21 33
27 76
1 30
10 00
.86
10 80
.41
20 60
.71
AW).
21 28
10 00
hWh
1 87
3 67
2 70
3 37
2 35
3.04
3.17
FH)
1 47
.50
2 .5H
2 20
1 03
1 40
3.69
MnO
M
1.5
.18
35
.13
.01
.17
McO
«
10
1 72
87
30
.66
1.70
(V)
i.ri
1 73
2 06
2 51
1 50
2 31
5.60
Nay()
H 4H
16 25
8 28
8 26
8 84
5 62
7 10
K,4)
i\ 4tV
3 72
4 08
5 46
5 23
8 30
6 15
IM)
1 5()
42
22
1 35
2 04
2 31
1 03
IV >.
01
64
25
12
78 _
( 'alriilati'd an Wa
tcr-frcc
HiOt
M W
45 W
.54 4S
.5.5 38
.58 65
56.10
50 82 ~"
TK)t
4A
21 (»
27 HH
1 :io
20 (M
87
20 16
42
21 ai
.72
Al,<>.
21.78
10.38
Fr,C),
1 W '
3 6S
2 S4)
3 42
2 40
3.11
3.28
KH)
1 40
.50
2 50
2 23
1 a5
1 53
3 66
MnO
O.*!
15
IH
35
13
01
.17
MkO
.VI
10
1 72
88
31
68
1.83
i'aO
1 7."!
I 74
2 07
2 .54
1 .53
2 36
5 80
\ii,0
H t\l
16 32
8 :<o
H :i8
0 02
5 75
7 33
K,4>
t\ .V\
3 74
4 00
5 .54
5 34
8 .50
6 27
IV »•
01
(M
25
12
70
^ • *
FjM'h
Kuni - KM
).00
CLASSIFICATION OF IGNEOUS ROCKS
25
GROUP VII
Plutonic
Effusive
39
40
41
42
No. of
Tonalite
(Osann, Rosen-
busch)
Quartz monzon-
ite (Clarke)
Granodiorite
(Osann, Clarke)
Dacite (Osann,
Rosenbusch)
analyses
5
20
12
30
SiO,
66.57
66.64
65.10
66.i91
TiO,
.54
.50
.54
.33
A1,0,
14.57
15.57
15.82
16.62
Fe,0,
2.36
1.91
1.64
2.44
FeO
4.12
1.94
2.66
1.33
MnO
.01
.06
.05
.04
MgO
1.72
1.41
2.17
1.22
CaO
3.27
3.50
4.66
3.27
Na,0
3.22
3.41
3.82
4.13
KtO
2.22
3.72
2.29
2.50
H,0
.93
1.15
1.09
1.13
PtOs
.47
.19
.16
.08
Calculated as Water-free
SiO,
67.20
67.41
65.82
67 67
TiO,
.54
.51
.55
.33
\Wz
14.71
15.76
15.99
16.81
Fe,03
2.38
1.93
1.66
2.47
FeO
4.16
1.96
2.69
1.35
MnO
.01
.06
.05
.04
MgO
1.74
1.43
2.19
1.23
CaO
3.30
3.54
4.71
3.31
Xa,0
3.25
3.45
3.86
4.18
K,0
2.24
3.76
2.32
2.53
PjOi
.47
.19
.16
.08
Each sum = 100.00
IGNEOUS ROCKS AND THBJR ORIGIN
CHOIP VIII
a "Si
il-lf
il!l.
«5' i\ Itj
1 =
KiO,
M 17
58.38
.W-77
59 59
57-50
59 48
61.12
62-25
TiO,
,W
-80
.84
-77
.79
.48
.42
l.OS
A1,0.
16 52
16.28
16.67
17.31
17.33
17.38
17.65
16.10
Fe^,
2 «:(
2 08
3 16
3,33
3 78
2.96
2.89
3-62
FcO
4 11
4 11
4.40
3-13
3 62
3,67
2.40
2.20
MnO
,0S
,13
13
18
-22
.15
.16
.21
MkO
:i 75
3. 88
4.17
2.75
2.86
3-28
2.44
2.03
CaO
6 24
6.38
8.74
5 80
5.83
6.61
5.80
4.05
Xa,0
2 OR
3.34
3.30
3.58
3 53
3 41
3.83
3.55
K^)
1 93
2 00
2 12
2 04
2 36
1.64
1.73
2.44
H,0
1 30
1 37
1 36
1 26
1.88
74
1,43
l.SO
P,().
26
.26
Iciilutod
.26
OS WatP
-30
-free
.20
,15
.40
SiO,
60.31
59 19
57.56
60,3.5
58-65
59.92
62.01
"m!20
TiO,
.6,5
.81
.85
.78
.80
.48
,43
1-67
Ai,(),
lfi,7.i
10 SI
16 90
17 54
17.67
17.51
17,91
16-35
Fo^).
2 67
3 02
3 20
3.37
3,8.5
2,98
2.93
3-67
KeO
4 17
4 17
4 46
3.17
3.69
3.70
2,44
2.23
Mn(>
OK
13
13
.18
.22
.15
,15
.21
Mk<)
A m
3 93
4 23
2.78
2 90
3.31
2,48
2.06
Cat)
R 3:(
6 47
6.83
5.87
5 92
6.66
5,88
4.11
Nh,()
3 m
3 39
3 44
3 63
3 60
3 44
3,88
361
K,()
1 w>
2 12
2 15
2 07
2 40
1 65
1,74
2-48
i',o.
.26
26
25
26
,30
20
.15
.41
Kach HU
1 = 100-00
CLASSIFICATION OF IGNEOUS ROCKS
Plutonic
"EffuBive '
tlj
53
63
S4
6S
06
67
66
1
i\B
ih
1
j
3 g J
1
1
^
1
s
i
basalt, includ
la, 17 olivine
laphyres, and
Basalt, as named
(including also
tachylitf, etc.)
{
1
1
1
1
1
^
<
^IsS
3
o
(3
NVof
tsilyiKB
7
41
^^98
161
20
17
50.10
U
50.60
9
SO,
50.16
48.34
49.06
48.78
50.12
49.50
TiO,
1.64
.97
1.36
1.39
1.41
1.25
.68
1.42
AIK).
18.51
17.88
15-70
15.85
15-68
14.43
17.40
14.37
T,ia,
1.88
3.16
5.38
5-37
4.55
5.06
4,57
6.55
Frt
9.29
5.95
6.37
6-34
6-73
6,31
C 29
5.84
MnO
.14
-13
.31
.29
.23
.25
.46
.17
MiO
5-97
7.51
6.17
6.03
5.85
7.32
4.89
7.75
CO
7-90
10.99
8.95
8.91
8,80
S-53
8,09
9,96
K«0
2.72
3.55
3.11
3.18
2.95
2.75
3 23
2-60
K,0
-80
.89
1.52
1,63
1.38
.73
1.76
.84
H,0
.76
1.45
1.62
1.76
1.93
2.00
1.83
.66
P,0.
-23
.28
,45
,47
.37
,27
,20
.44
SiO,
50.54
48-95
4
TiO,
1.65
-98
.U,0.
18-65
18-15
1
Ke,0.
1.90
3.21
Kd)
9-36
6.04
MnO
.14
-13
MbO
6-02
7,62
CaO
7-96
11.15
XM>
2-74
2.59
K,0
,81
.90
P,(-).
.23
.28
Calculated
49,87^
1.38
15.96
5.47
,47
Water-free
49.65 I 51.11 I
1.41 , 1.44 j
16.13 I 15.99 i
5.47 I 4.64 I
6.45 I
.30 I .23 I
M4
5.96
8.97 I
3,01 '
1 51.12
51,54
49.83
i 1.27
.69
1-43
14,73
17.73
14.47
5.16
4,66
6.59
6.44
6.41
5.88
.25
.47
.17
. 7.47
4.99
7.80
9,73
8.24
10.02
1 2,81
3.29
2.52
1 ,74
1,78
.85
i \ .28
.20
,44
28
IGNEOUS ROCKS AND THEIR ORIGIN
GROUP X
69
No. of
analysra
SiOs
TiO,
AWt
FcOi
FcO
MnO
Mk()
CaO
Na,0
K,0
H,0
p,o.
SiO,"
TiO,
Al,0,
Fc,(),
FcO
MnO
MrO
C'aO
Xa,0
K,0
PfO»
bC
c
•s e
50 31
.85
iK.:io
2.85
5 89
.12
G.73
10 81
2 86
1 00
.28
50 51
50.78
2 05
.15
18 32
28.51
73
1.07
10.38
1.13
.20
.05
6.33
1.26
7 93
12.55
3.19
3.70
1.02
.76
.34
.05
CLASSIFICATION OF IGNEOUS SOCKS
GROUP XI
Plutonic
Effu-
U
66
«6
67
68
89
70
71
11!
d
-2
2
c"
O
a
■Nf^
^2
.U
No. of
1?
III
1
s
1
.S J
a
6
II
II
•ulym
3
7
5
5
5
31
14
810,
35.62
40.91
45.07
43.14
42.71
38.68
41-09
41-30
TiO,
3.2fi
.65
.64
,12
I- 16
.81
A1,0.
7-78
5.00
5.75
4.72
3.11
,94
4,80
9,43
F«*.
5-72
4,64
3,43
1.23
4.97
5.47
3.96
5,30
FeO
9.42
7,97
9,53
7,46
4 67
8.44
7.12
8,86
MnO
.16
.07
,26
07
.06
.17
.10
,29
MgO
22-70
30.82
22. S8
37.89
32.31
42.51
32,25
19,94
CO
4.56
4.41
7,48
3-01
6,51
1.06
4,42
8.01
N.,0
.43
.68
1,14
1.02
.03
,49
1.20
Kfl
3.32
.36
.67
.29
.04
,96
.39
H,0
6.15
4.56
3 10
2 43'
3.32
2.63
3.63
4.27
PO.
.88
,03
.16
.01
.03
.12
.20
Calculated
as Wftter-free
8iO,
37.96
42.87
46 51
44.22
44.18
39.72
42.60
43.14
TiO,
3.47
.68
,66
.12
1.20
.85
Al^,
8.30
5,25
5,93
4.84
5.28
.96
4,97
9.85
Fe,0,
6.09
4-87
3.54
1,31
5.14
5.61
4,10
6.64
FeO
10.04
8,34
9.84
7.64
4,72
8.67
7,39
9.26
MnO
.17
,07
.27
.07
.06
.17
.10
.30
MgO
24.19
32,30
23-61
38.83
33.45
43.68
33,43
20,83
CaO
4.86
4.63
7.72
3-09
5.69
1.09
4.59
8.37
Na,0
,45
,59
1.17
1.05
.03
.51
1.25
K,0
3.53
,37
.59
.30
.04
.99
.40
P,0.
.94
,03
.16
.01
.03
.12
.21
IGNEOUS ROCKS AND THBIR ORIOIN
1
1
ill
■srD
111
HI
4 =
1 1
11
■
«
1^
1
h -
* .
:
No. ot
=
;:
<
anklyK«
4
10
n
20
34
14
6
SiO,
53.05
51.29
44.30
48 64
49 20
41.25
42,25
TiO,
14
,58
2.31
1 86
1.68
1.59
2 52
Al,0,
I M
3 52
8 93
17.96
16.65
12.03
IS, 26
Fe,0.
1.90
1 82
7 94
4 31
4,76
5.65
8 43
F«0
5.35
0 00
7.75
5.58
5.36
7.29
5.4G
MnU
.17
.13
.23
.19
.55
.54
MkO
22 57
21 06
10 20
4 00
4.43
11.22
S.49
C»0
13.37
13 8S
15.27
8. SO
7,74
11,88
9.75
N.<1
.20
.30
.74
4 :iO
4 54
3.40
445
Kfl
.07
.1A
1.05
2.28
3 19
1,30
1.92
ii,o
.M
1 20
1 IH
1 34
1 30
3 20
2.43
IV>.
.07
.IW
01
-65
.60
,65
1.04
CuWInlol 01
WatpT-fre
SiO,
54 11
51 01
44 S3
. 49 31
49 85
42 62
43.30
TiO,
14
..W
2 34
1.S8
1 70
1,65
2.58
Al,0,
1 fi7
3 .'.7
!> IKt
18 20
16 88
12 43
16.67
Fr,(),
1 92
1 S4
S 04
4 37
4 82
5 84
8 64
FcO
5 40
ti 07
7 S4
5 66
5 43
7 53
6,59
MnO
i:
2-J 7rt
■J I .12
23
Id 41
.19
4 05
,56
4 48
,56
11 60
Mirll
5 63
r«o
13 49
14 (»
1.-. 4li
9 01
7 S4
12 28
9 99
Na,0
2rt
31)
.75
4 M
4 60
3 48
4,56
K,(l
07
Ifi
1 IV>
2 :il
3 23
1 34
1,97
r,(i.
"7
IW
01
r-i
.61
,67
1,07
CLASSIFICATION OF IGNEOUS ROCKS
CROUP XIII
Plutonic
"
80
ri
'■i
e
t
So. ot
«.il?«»
6
6
48.66
SiO,
45.61
TiO,
1.96
.97
AI,0,
14.35
12.36
F.,0.
6.17
3.08
FeO
4.03
5.86
MdO
.19
.13
MeO
6.05
S.Ofl
C«0
0.49
10.46'
N.rf3
5.12
2.71
M
3.69
5.15
Hrf)
2.60
1.46
Pfl.
.74
1.07
2.57
3,39
2.00
,14
8.20
10.12
3.81
2.37
2.42
aa
84
88
£.1
•f-
z
•1
55
4
20
16
46.91
49.90
44.20
1.81
.16
1,64
16.25
16.94
15,64
7.70
3.02
4,35
4,06
7,15
6.14
1.43
,23
.19
2.95
4.22
8.89
0.36
10,04
9.74
4.25
2,24
4,03
2.63
3 57
1.83
2.51
1,74
2.67
1,14
.79
.88
Calculated as Water-tree
4.76
.01
5-43
11,64
2.93
4.55
1.12
.50
SiO,
46.83
49.38
50,15
45.51
48.12
50.79
45.41
46,86
TiO.
1.98
.98
1-02
1.60
1.86
.16
1.68
1,31
A1,0,
14.73
12.66
16.90
16.20
15,65
17.24
16.07
16,78
Fe.0,
6.34
3.12
3,72
4.78
7.89
3,07
4 47
5-90
FeO
4.14
5,95
6,82
5.99
4.16
7.28
6,31
4.81
MdO
.19
.13
.31
.14
1.47
.23
.20
.01
MgO
6.22
8.21
4,06
8.41
3.02
4,30
9.13
5,49
CaO
9.75
10,62'
10,08
10,37
9.60
10.22
10.01
11,77
N8.0
5-27
2,75
2.62
3,90
4.36
2.28
4.14
2.96
K,0
3,79
5,23
3,46
2.43
2.70
3.63-
1.88
4,60
Prf).
.76
1,08
.86
.67
1.17
.80
.70
.51
' Includes .40 per cent. BaO and .09 per cent. SrO.
• Includes .41 per cent. BaO and .09 per cent. SrO.
32
IGNEOUS ROCKS AND THEIR ORIGIN
• Inclihlrsi 'X^ \*cx ctMii W%0 9X\\\ jC |mt crni. Sri>.
• 1nohi<io!t 4S \\tx mil Hat* ami l> ycx cent, Sri>.
■ Inrhuiogi M \M^t i^^tit U.ii> .^mi H7 |*«t ot'ni. SH>.
• In^'liKioit .M> j^r oriii aii*i 1\> imt *^nt Sil>.
87
Ff»rgujiitc
(Pinwon)
I
51 70
ORrtrP XIV
Plutonic
88
i Miflsourite
1
(Pinwon, Daly)
1 2
44.27
1
Effusive
89
Leucite basalt
(Osann, Rosen-
buach)
7
46.47
90
No. of
Leueitite
(Osann, Rosen-
buseb)
AnnlyiMw ,
7
HiO,
47.72
TiO,
.23
1 37
1.33
.52
Al,().
\\ m
10.73
15.97
18.19
Fci()«
5 07
3.63
5.97
4.74
FH)
3 58
5.87
4.27
3.90
MnO
.01
.06
.01
.06
Mic<)
4 55
13 a5
5.87
3.45
CnO
7 40»
1 1 46>
10.54
7.27
Na.()
2 03
1 07
1.69
4.51
K,()
7 «)
4 43
4.83
7.66
H,() '
2 25
3 23
2.32
1.51
P.(>.
IS
83
.73
.47
C'alrulAtiHl AM
WaI
tor-frce
SiOt
52. S9
45 75
47 58
48.45
TiO,
.24
1 41
1 36
.53
Al,Oi
14.83
11 00
16 35
18.47
Kr,t>,
5 IS
3 75
6 11
4.81
Fr<>
3 . m
6 07
4 37
3.96
MnO
01
06
01
.06
M||l>
4 lU
13 40
6 01
3 50
i^*0
7 o7»
11 S5«
10 79
7 38
Nii,0
3 (H)
1 10
1 73
4 58
K|0
7 79
4 :>7
4 94
7.78
r.o,
IS
M'%
75
.48
V^ch Mim
-10
0 00
CLASSIFICATION OP IGNEOUS ROCKS
33
Plutonic
91
Ijolite
(Rosen-
busch)
92
Bekinki-
nite
(Rosen-
busch)
GKOUP XV
93
Effusive
94
Ncpheli-
nite
(Rosen-
busch)
Nephelite
basalt
(Osann)
SiO,
TiO,
A1,0,
Fe,0,
FeO
MnO
MgO
CaO
NaiO
K,0
P,0.
96
96
Melilite-
Melilite
nephelite
basalt
basalt
(Rosen-
(Rosen-
buschy
busch)
Osann)
5
5
37.56
35.72
2.66
4.78
10.08
9.56
6.82
5.41
5.94
6.55
.06
15.32
15.46
13.82
14.20
3.11
3.35
1.53
1.67
2.52
2.67
.58
.63
Calculated as Water-free
43.18
1.56
19.12
3.89
4.88
.19
3.19
10.56
9.72
2.28
1.43
42.47
2.75
14.77
5.22
7.22
9.42
12.42
3.66
1.20
.87
42.19
1.38
17.25
7.79
6.81
.17
3.81
10.37
6.61
2.55
1.07
40.77
1.53
13.88
6.86
6.57
.21
10.73
12.65
3.94
1.90
.96
Each sum = 100.00
38.54
2.73
10.35
7.00
6.09
.06
15.70
14.17
3.19
1.57
.60
36.70
4.91
9.82
5.55
6.73
15.89
14.59
3.44
1.72
.65
» Includes 0.29 per cent. COj.
lONBOVS ROCKS AND THEIR ORIQW
EfTiuire
No. of ^
111
III
!m
S -
•3-'
li
ll
1
II
P
■ea 1
3
10
4
10
20
4
S
5
3
8iO,
76 47
62 21
50.34
74.04
48.90
52.04
53.56
50 11
47.45
TiO,
,07
.60
.34
.18
1.71
.76
,82
.06
.81
Alrf).
13.03
16.45
14.75
13.19
13 94
17.65
17.88
13.04
11.43
F.rf),
},.«{
2. S3
4,18
1 35
5 59
4.66
4,51
4.58
3.22
FeO
2 89
2.75
1,01
8-63
2.76
3-05
3.04
6.78
MdO
.01
.02
.11
,04
,53
.13
.07
.11
.12
M,rf>
oe
3.32
4 23
.32
6 39
3 33
3,62
9.27
14.60
CO t
,45
4.96
10.43
1.19
9.05
5.11
6.45
7.63
8.18
N.rf)
3 53
3. US'
5 27
3.88
3 22
4,10
3.41
1 94
2.32
K,0
4 SI
2.21
5 -l
3 75
1,03
5-03
3.76
4.15
3,99
ll,o
r,i
.so-
1,20
1 02'
.73
3 74
2 32
3.58
3.60
P.O.
.01
ls
1.19
03
.28
.70
.55
.69
.60
Calr.iliilM M W
ater-fire
SiO,
7t> S7
62 71
50 •)■'•
74 SO
49 27
54 06
54 84
51 97
48.67
TiO,
07
Ml
3.-.
IS
1.72
,79
.84
1.00
.83
Al,t>,
13 10
hi 5-i
14 'J3
13 33
14 04
18 34
18.31
13.52
11.73
FeiO,
' 1 03 ^
2,5-1
4 23
I 37
5 63
4 84
4.62
4.74
3,30
FeO
2 92
2 7S
! 02
S 69
2 85
3.12
4.08
6.B3
MoO
01
W
11
04
.54
.14
07
.13
.13
M«0
06
3 35
4 2-;
.32
6 43
3 46
3 70
9 62
14.97
c«o
45
5 00
10 .V.
1 20
9 12
5 31
6.60
7 91
8.39
.\.,o
3 M
3 91'
•i 33
3 91
3 24
4 26
3.49
2 01
2.38
KMl
4 s;t
2 23
5 27
3 :»
1 01
5 22
3.S5
4 31
3.06
r,o.
01
l:t
I -.M
.03
2-;
,73
.56
.72
.62
K.ich
*uin - 10«> 00
' lo-hdfti 07 por .vnt. Uil>
■ InduJrs li;> )MT ivnt. (.'1 ■
CLASSIFICATION OF IGNEOUS ROCKS
35
GROUP XVII
106
Granite-
aplite
(Osann,
Washington)
15
1 75.00
Dike-rocks
107 108
109
110
Xo. of
analy-
Bostonite
(Rosenbusch,
Washington)
5
61.32
Grorudite
(Osann,
Washington)
5
1 70.91
Solvsbergite
(Osann,
Washington)
Tinguaite
(Osann,
Washington)
ses
sior
8
62.16
15
55.02
TiO,
.30
.89
.48
.31
.36
AljO,
13.14
18.43
11.50
17.58
20.42
Fe,0,
.58
3.84
4.58
3.05
3.06
FeO
.40
1.60
1.88
1.80
1.82
MnO
.07
.01
.39
.18
.22
MgO
.30
.46
.11
.48
.59
CaO
1.13
1.45
.39
1.11
1.67
\a,0
3.54
5.75
5.43
7.30
8.63
K/)
4.80
4.94
4.08
4.95
5.38
H,0
.71
1.31
.25
1.04
2.77
PtO»
.03
.04
.06
Calculated as Water-fre
e
SiO,
75.54
62.14 71.09
62.82
56.59
TiO,
.30
.90
.48
.31
.37
A1,0,
13.23
18.67
11.53
17.77
21.00
Fe,0,
.58
3.89 4.59
3.08
3.15
FeO
.40
1.62
1.89
1.82
1.87
MnO I
.07
.01
.39
.18
.23
MgO
.30
.47 .11
.49
.61
CaO ;
1.14
1.47 1 .39
1.12
1.72
Xa,0
3.57
5.82 5.44
7.37
8.87
K,0
4.84
5.01 4.09
6.00
5.53
P/)»
.03
1
.04
.06
Each sum = 100. 00
36
IGNEOUS ROCKS AND THEIR ORIGIN
GROUP XVIII
112
Kersantite
(Osann,
Rosen-
busch)
20
50.79 '
Dike-roclu
118
Vogesite
(Osann)
4
52.62
114
Campton-
ite
(Osann)
116
Ill
Minette
(Osann,
Clarke)
10
49.45
116
No. of
analy-
Monchi-
quite
(Osann)
16
AlnAite
(OMim,
WMh-
ingtoo)
scs
15
40.70
6
"SiO,"
45.17
32.31
TiO,
1.23
1.02
.54
3.86
1.90
1.41
Al,0,
14.41
15.26
14.86
16.02
14.78
0.50
Fe,0,
3.39
3.29
3.60
5.43
5.10
5.42
FeO
5.01
5.54
4.18
7.84
5.05
6.84
MnO
.13
.07
.84
.16
.35
.01
MkO
8.26
6.33
8.55
5 43
6.26
17.43
CaO
6.73
5.73
5.86
9.36
11.06
13.58
Na,0
2 54
3.12
3.21
3.23
3.69
1.42
K,0
4.69
2.79
2.83
1.76
2.73
2.70
H,0
3 04»
5.7P
2 70
5 59>
3.40
7.60*
p,o,
1 12
.35
.21
.62
.51
2.38
50.99
Calcul
53.87
ated as Wal
54.08
ter-free
43.10"
8iO,
46.76
34.03
TiO,
1 27
1 08
.56
4.09
1.96
1.52
A1,0,
14 86
16 18
15 28
16.97
15 30
10.27
Fe,0,
3 50
3.48
3 70
5.76
5 28
5.86
FeO
5.17
5.88
4 29
8.30
5.23
• 6.85
MnO
.13
, .07
.86
.16
.36
.01
MrO
8 53
6.71
1
8 79
5 76
6 48
' 18.84
CaO
6 95
6.09
6 02
9 92
11 45
! 14.68
Na,0
2 (*>2
3 31
3 30
3 42
3 82
1.53
K,0
4 81
2 96
2 90
1 86
2.83
2.02
P/).
1.14
.37
.22
66
53
2.50
Ei
kch sum » 10
0.00
* Inchi<le8 .01 imt cent. CO,.
* Includes 2.97 per rent. CO,.
« InchiileH 2.61 |)er rent. CO,
* Includes 4.35 per cent. CO,
CLASSIFICATION OF IGNEOUS ROCKS
37
INDEX TO TABLE II
Figures refer to column.
Absarokite, 104
Akerite, 22
AUskite, 97
AIndite, 116
Amphibole andesite, 49
peridotite, 65
Andesite (all), 46
Anorthosite, 63
Augite andesite, 47
syenite (subalkaline), 15
Augitite, 78
Banakite, 102
Basalt (all), 53
as named by authors, 54
of Hawaiian Islands, loi
Basanite (all), 82
Bekinkinite, 92
Bostonite, 107
Cdunptonite, 114
Comendite, zz
Dacite, 42
Diabase, 55
Diorite, including quartz diorite, 44
excluding quartz diorite, 45
of Electric Peak, 98
Dolerite, 58
Dunite, 69
Eleolite syenite, 35
Essexite, 75
Fergus! te, 87
Foyaite, 32
Gabbro (all), 52
excluding olivine gabbro, 59
Granite (alkaline), zo
(subalkaline), 9 *
of all periods, 4
younger than the Pre-Cambrian, 3
Granites (Pre-Cambrian, including 16
analyses of Swedish types), z
(Pre-Cambrian of Sweden), 2
Granite-aplite, zo6
Granodiorite, 4Z
Gronidite, zo8
Harzburgite, 67
Hornblende andesite, 49
syenite (subalkaline), Z3
Hypersthene andesite, 48
Ijolite, 9Z
Keratophyre, 26
Kcrsantite, ZZ2
Latite, 3Z
Laurdalite, 34
Laurvikite, 28
Leucite absarokite, Z05
basalt, 89
basanite, 86
phonolite, 37
tephrite, 84
Leucitite, 90
Leucitophyre, 38
Lherzolite, 68
Limburgite, 77
Liparite (all), 5
as named by authors, 6
Malignite, 99
Melaphsnre, 57
Melilite basalt, 96
-nephelite basalt, 95
Mica andesite, 50
peridotite, 64
syenite (subalkaline), Z4
Minette, zzz
Missoiuite, 88
Monchiquite, zz5
Monzonite, 30
Nephelite basalt, 94
basanite, 85
syenite, 35
tephrite, 83
Nephelinite, 93
Nordmarkite, 20
Norite (all), 5z
excluding olivine norite, 6z
Olivine diabase, 56
gabbro, 60
Doritei 6a
L
38
JGSEOUS ROCKS AND THEIR ORIGIN
Pantellerite, 27
Pcridotito (all), 70
Phonolite, 36
Picritc, 71
Pulaflkito, 21
Pyroxenite (rttihalkalino Hcrics), 73
(alkaline norios), 74
Quart I diorite, 43
kcratophyrCy 12
monionite, 40
porphyry, 8
Rliomh-porphyry, 29
Rhyolitc, an namrd by authors, 7
of Yellowstone Park, 100
Saxonite, 67
Shonkinite, 80
Shoehonite, 103
SdlvflberRite, X09
Syenite (all), 17
(alkaline), 24
(all types, subalkalinc), 16
•
Tephnte (all), 81
Theralite, 79
Tinguaite, no
Tonalite, 39
Trachydolerite, 76
Trachyte, 18
(alkaline), 25
v'subalkaline), 19
Umptekite, 23
Urtito, 33
Vogosite, 113
Wehsterite, 72
Wohrlite, 66
AvERAOE Specific Gravities
A fundamental datum concerning an igneous rock is its average
specific gravity. I'nfortunately most authors have neglected to ap-
|K»nd to each chemical analysis a statement of the density of the
analyzed specimen. This is notably the case with the many hundreds
of magnificent analyses published by the oflScers of the United States
Geological Survey. It would Ix^ a valuable contribution to petrology
if the specific gravities of all original analyzed specimens (whole hand
specimens) in the colk'ctions of that survey and of similar institutions
were determinetl and tlie result.s published. Becke has quite recently
imlicated the im|>ortanct^ of such determinations for petrology.*
On account of this gimeral failure of record during the last three
decades, it is not iK>s8ible to give satisfactory' average specific gravities
for most of the holoorvstalline ignet>us-rock t>7>es. It goes without say-
ing that rooks containing abundant glass, which from specimen to
specimen always varies greatly in relative amount, can rarely give use-
ful averages. However, certain neeiLs in the following discussion are
tolerably siitisfied by a compilation of the average specific gravities
of the plutonicsixHMt^ which have Ihmmi measureil. The more impor-
tiint aveniges ap|H\ir in Table III. in which the averages more recently
caloulateil by RtH'ke. from siHHMally seUn'teil material, aiv also entered.
In each case tlie nuniUT of s|HHMmens avcnigiMl is shown in brackets.
> F Beckc, SiUuii|t*lH*r k Ak^a Wls*., M.Mh Natun* . Kl . Vol, 120. 1911, p.
765
CLASSIFICATION OF IGNEOUS ROCKS
39
TABLE III
Granite,
Granodiorite,
Tonalite,
Syenite,
Monzonite,
Nephelite syenite,
Diorite,
Gabbro,
Olivine gabbro,
Anorthosite,
Peridotite and pyroxenite
Essexite,
Theralite,
Malignite,
Average Specific Gravity
Daly Becke
2.660(58) 2.682(43)
2.740(5)
2.723(7)
2.773(11) 2.775(27)
2.805(2)
2.617(10) 2.655(9)
2.861(17) 2.855(13)
2.933(19) 2.975(27)
2.948(4)
2.715(6)
3.176(21) 3.307(13)
2.844(5) 2.915(3)
2.917(3) 2.940(4)
2.884(4)
Division According to Mode op Occurrence ,
Rosenbusch has done special service to petrology and to the theory
of petrogenesis in retaining the time-honored division of igreous
rocks into the plutonic, volcanic, and dike classes. The separation
of the plutonic and volcanic classes is obviously necessary in laying
the basis of fact for a general petrogenic discussion. Table II is so
arranged as to show a general contrast between those two classes.
On the average a plutonic species is less salic, that is, lower in silica
and alkalies and higher in iron oxides, lime, and magnesia, than the
corresponding extrusive species. The same contrast has been stated
by Rosenbusch.^ It is sometimes illustrated in individual localities
where a plutonic rock and the contemporaneous volcanic equivalent
have both been discovered and chemically analyzed; but in most
cases the characteristic variability of rock-bodies makes it difficult
to discern any systematic contrast without a prohibitive number of
analyses. In this regard world averages must have special value.
An explanation of this fundamental contrast is ofifered in Chapter
XII (page 229). For the present it will suffice to point out that chem-
ical analysis, like the necessities of geological mapping, shows the
justice of this old grouping of igneous-rock species.
In spite of recent adverse criticisms, Rosenbusch's conception of
the remaining dike class is still the best method of grouping a consider-
able number of species. Brogger's division of the rock bodies of this
class into aschistic and diaschistic is sometimes hard to apply, but its
underlying idea is clearly helpful. Aschistic dikes are directly apo-
^ H. Roflenbusch, Mikroskopische Physiographie der massigen Gesteine, Vierte
.Vuilage, Stuttgart, 1908, p. 717.
40 IGNEOUS ROCKS AND THEIR ORIGIN
physal from larger intrusive masses and have essentially the same
chemical composition as these. A diaschistic dike is an offshoot
from a larger intrusive mass but represents the effect of more or less
thorough splitting or differentiation of that mass, whereby the dike
comes to differ chemically from the parent mass.
IC.NEOUS-ROCK ClANS
In the index of Rosenhusch's hand-book, more than 660 varieties
of igneous rocks are named, and this number excludes many types
which, though dignified with special names, are simply weathered or
otherwise altered equivalents of species in the first-mentioned group.
If there were no obvious genetic or chemical relationships among
these hundreds of species, the problem of origins would be, in very
truth, a difficult one. However, field and laboratory observations
without number have already simplified the problem by showing that
the many types can be grouped in chemical series of relatively small
number. For temporary, convenient use in this book, these series
may be called "claas.'' Each clan is composed of families, distin-
guished less by chemical composition than by mode of field occurrence,
by mineralogical composition, or by rock structure.' The genetic
problem is most concerned with the explanation of chemical diversities
and it is greatly simplified by recognition of the fact that these clans
represent so many chemical groups, each of which includes syngenetic
families. For example, the problem of origins is essentially the same
for magmas of the syenite family, the 8>'enite porphyry family, and
the trachyte family. The chemical similarity of gabbro, gabbro
porphyrite, diabase, diabase porphyrite, basalt, dolerite, melaphyre,
and other species, suggests, though of course it does not prove, their
common magmatic origin. On the other hand, the diorite family, like
some other families, includes species which clearly have different lines
of descent; but, for convenience of treatment, such a family, together
with its effusive and dike equivalents, is here regarded as forming a
clan. Thus, while a clan generally represents a group of rocks formed
by similar magmati(* processes, it may also include some mineralogical
and chemical allies formed by quite different processes.
The more important groups are the granite clan, the granodiorite
clan, the diorite clan, the gabbro clan, the syenite clan, the nephdite-
syenite clan, and the peridotite-pyroxenite clan. The minor groups
are the essexite clan, the theralite clan, the ijolite-bekinkinite clan,
and the missourite-fergusite clan. The last four groups mentioned,
' The term "structure" is here employed with its long-established meamnc •■
used by Zirkel, Rotenbusch, and the majority of petrographers.
CLASSIFICATION OF IGNEOUS ROCKS 41
together with the nephelite-syenite clan, will be discussed together
under the caption "the alkaline clans," in Chapter XX; the other
six clans will be treated in as many separate chapters (XV to XIX,
and XXI). In those chapters the reader will find lists of thp principal
rock varieties constituting each of the clans.
CHAPTER III
GENERAL DISTRIBUTION AND RELATIVE QUANTTTIBS OF
IGNEOUS-ROCK SPECIES
Need for Quantitative Study
It is probable that all or nearly all of the elans are represented in
each of the seven continents. Most of them are represented in the
islands of the great ocean basins, but the granite family seems, accord-
ing to present knowledge, to be wanting in the islands of the deep seitf
except in the regions where there is independent evidence of compara-
tively recent fragmentation and subsidence of former continental
ATQfis, The recent attempts to establish a rule localizing the alkaline
clans in the ''Atlantic'' portion of the globe, and the subalkaline clan:^
in and around the ** Pacific" region, have not met with general favor
among petrologists. Nor has better success attended the attempt to
connect the distribution of the alkaline rocks with the type of crystal
deformation — foundering or large-scale normal faulting — which iq>eci-
allv characterizes the Atlantic basin. The recurrence of most of the
clans on ever}' continental plateau and along every ocean border is
becoming increasingly evident as the methods of modern petrography
are applied to tlie study of the world's igneous terranes.
Partly owing to this fact, but still more to the circumstance that
the rarer rock species are the more ''interesting'' to most petrog-
raphers, the tendency has long reigned in petrographical literature
to emphasize the diversity of igneous rocks. Like every other science,
petrography has had to be analytic before it could be healthfully
8>'nthetic. But there is no little danger of a false perspective if, in
the search for specific distinctions, a considerable effort is not made to
estimate the actual value of those distinctions. Above all, petrog-
raphy needs to be ever more closely linked with areal and structural
geology, in order that the problem of rock origin may be phrased in
terms of the actual proportioas of the different species.
Obviously, the data for such a quantitative study of the visible
igneous matter in the earth fall far short of being complete enough
for the ultimate needs of petrogeny. Yet the documents already
published suflSce for a few important conclusions from a synthetic
study. Certain of these will doubtless stand fast when the whole
earth is as well known as the most thoroughly explored parts of Europe
or America are now known.
42
DISTRIBUTION AND QUANTITIES OF IGNEOUS-ROCK SPECIES 43
A quantitative review of the myriad facts recorded in petrographic
works might, along special lines, be undertaken by such a powerful
body as a representative committee of the International Geological
Congress. Until that body or a similar group of properly equipped
specialists undertakes the great task, it would seem futile for any one
to try to state the genetic problem of the igneous rocks in all the
fullness and clarity now possible. The writer has found at hand only
a few quantitative tests which can be applied to the modern geological
maps i^-ithout a prohibitive amount of labor. Among the significant
questions are the following: 1. What are the relative areas of the
earth's surface covered by the different intrusive-rock species?
2. How do the respective total areas of the alkaline and subalkaline
bodies compare?
3. What is the largest known rock-body representing each of the
rook-clans?
Relative Quantities of Igneous-rock Species in the United
States
For the petrographer as for the geographer, North America is in
many respects "the typical continent." A first approximation to an
estimate of the relative areas covered by igneous species throughout
the world would be feasible if complete measurements were now at
hand for this one continent. Although these can only be made by the
work of many future generations, the scanty record of the present day
points, with a high degree of probability, to certain principal facts of
distribution.
The igneous rocks of North America are almost entirely confined
to three provinces, the Pacific Cordilleran system, the Appalachian
mountain system (Newfoundland to Alabama), and the pre-Cambrian
shield of Canada. In the United States the geology of the Cordilleran
and Appalachian provinces has been sampled in the preparation of the
folios issued by the Federal geological survey. The number and
distribution of these areas are such as to show that the sampling so far
accomplished roughly indicates average quantitative relations among
the exposed igneous rocks of these two provinces. On the whole, too,
the standard of detail and accuracy set for the folios is fairly uniform;
in all cases the field conclusions have been checked with more or less
thoroughness by microscopic and chemical examinations of the rocks.
The total area of the United States proper is 3,030,000 square miles.
Of this, about 159,000 square miles are covered in the folios published
up to January 1, 1912. The Cordilleran area is about 800,000 square
, miles, and in this belt the fifty-nine published folios, in which igneous
I formations are mapped, cover a total area of 60,990 square miles. The
44
IGNEOUS ROCKS AND THEIR ORIGIN
Appalachian area is somewhat less than 200,000 square miles; within
it the sixteen folios showing igneous formations cover a total area of
13,221 square miles. The writer has measured the total areas covered
by each of the igneous-rock species named and mapped in these groups
of folios. The results are shown in Table IV.
TABLE IV.— PLUTONIC ROCKS
Pacific
Cordillera,
square miles
Appalachian
System,
square miles
T6UI,
scpiaie
mOes
Pre-Cambrian granite
Paleosoic and later granite
Total granite
Granodiorite
Quarts monsonite
Quarts diorite
Diorite
Gabbrodiorite
Gabbro
Anorthoeite
Syenite
Monsonite
Nephelite syenite
Shonldnite
Fergusite
MisBOurite
Theralite
Peridotite
Pjrroxenite
Totals
2089.0
402.0
(2491.0)
2040.0
11.0
45.3
103.5
98.5
226.4
52.0
24.4
17.5
3.5
8.7
<1.0
.1
0.3
73.3
2.2
5204.7
1151.0
194.0
(1346.0)
10.0
47.5
.3
0
.0
(8886.0)
9010.0
11.0
45.3
118.5
5
9
0
,4
6
.8
7
378
52
34.
17.
8.
8.
<1.0
.1
6.8
78.8
2.2
1402.8 I 6607.6
Granite porphyry
Quarti porphyry and rhyolitc.
Dacite porphyrite
Quarts-hornblende |x>rphyritc
Quarts monionitc porphyry . .
Diorite porphyrite
Hornblende porphyrite
Quarts diabase
Diabase.. . .
Syenite porphyry
Monsonite porphyry
Nephelite syenite pon>hyr>'. . .
Pbonolite
Peeudo-leucite porphyry
Total8
•HYPABYSSAL HOCKS" (INTRUSIVE)
I Pacific
Cordillera,
square miles
17.9
26 5
7.8
2.0
4 6
20.1
10
3 0
150.0
3ft 4
0 4
< 1
2 7
.5 _
284.0 I 125.1
Appalachian
Total,
System,
square
square miles
miles
2.0
19.9
1.0
27.6
7.8
2.0
4.6
1.6
21.7
1.0
8.0
118 0
268.0
2.5
40.9
B ■ •■• ■■■■•••
9.4
* ••■>■«■
< .1
2.7
'
.6
409.1
DISTRIBUTION AND QUANTITIES OF IGNEOUS-ROCK SPECIES 45
EXTRUSIVE ROCKS
Rhyolite
Dacite
Mica andesite
Hornblende andesite
Pyroxene andesite (chiefly)
Augite porphyrite
Basalt
Trachyte
Latite
Phonolite
Trachydolerite
Teschenite
Nephelite basalt (Texas)
Nephelite-melilite basalt (Texas)
Limburgite
Quartz basalt
Totals
Pacific
Cordillera,
square miles
~ 2145T7
82.1
3.0
21.6
3966.0
255 . 0
3079.0
6.5
4.6
5.5
.3
1
2
2
2
2
8
5
8.0
9584 . 0
Appalachian
System,
square miles
1.0
130.0
131.0
I
Total,
square
miles
Total igneous-rock area mapped . 1 5,072 . 7
1,658.9
2146.7
82.1
3.0
21.6
3966.0
255.0
3209.0
6.5
4.6
5.6
.3
.2
1.2
2.8
2.5
8.0
9716.0
16,731.6
The measurements for each sheet were not made with extreme
nicety, for that would have involved much labor useless for present
purposes. However, the order of magnitude is believed to be correctly
stated. The areas of the rock types showing small extension were
measured with special care. Composite terranes, respectively includ-
ing several igneous species but mapped under one color, were neglected.
In some cases terranes of greatly altered igneous rocks were similarly
excluded from consideration.
In estimating the relative volumes corresponding to the areas
listed, it should be noted that many of the bodies of "plutonic rocks"
are really sheets, laccoliths, or irregular intrusions, and hence do not
extend to great depths. This is true, for example, of all the theralite
mapped, of some diorites, and of many shonkinitic, monzonitic, and
syenitic bodies. It is not certain that any of the gabbro masses ex-
tends, with undiminished length and breadth, to a depth of more
than a few thousand feet. On the other hand, most of the granite,
granodiorite, and quartz diorite bodies have depths to be estimated
in miles. Hence, several of the plutonic types, which show small
total areas, are yet more clearly subordinate in respect of total volume.
Again, observation shows that for the efiFusive types small total
area is generally accompanied by small average thickness.
i
46 IGNEOUS ROCKS AND THEIR ORIGIN
Hence, whether we compare the plutomcs inter se or the effuave
rocks inter se^ the volumes of the less extensive eruptives in the folio
quadrangles are likely to have even lower ratios to the volumes of
the other eruptives than those between their respective surface areas.
Rklativk -Abundance of the Alkaline Rocks, Includinq the
Syenite Clan
The ** alkaline provinn^s'' of the Conlilleran belt include some
which are among the most extensive in the world and are also very
rich in typc^. In this part of the world the shonkinites, missourites,
fergusites, theralites, and latites were first named and described.
Perhaps in no other region are the monzonites represented by more
numerous distinct bodies than in the United States portion of the
Cordillera. There is no reason for believing that the relative abund-
ance of the alkaline rocks in the Cordillera as a whole is greater than
that illustrated in the Conlilleran quadrangles of the Geological
Sur\'ey folios so far published.
Hence, the following conclusions directly derivable from Table
IV are highly significant. The combined area of all the syenites,
nephelite syenites, monzonites, shonkinites, mi.ssourites, fergusites,
and theralit(*s of thesi* Cordilleran cpiadrangles is only about 61 square
miles out of a total of the 520.") scjuan* miles covered by all the plutonic
types mapped in these (juadrangles. The combined area of the
syenite por]>hyri<»s. monzonite porphyries, nephelite-s>'enite por-
phyries, pseudo-leucite porphyries, and intrusive phonolitc is only
ol sf]uare miles out of a total of 2SI .^piare miles of hypabyssal rocks.
The combined area of the extrusive trachytes, latites, trachydolcrites,
phonolites. tcschenites. nephelite bjisalts, ami limburgites is only
about 23 sipiare miles out of a total of 0584 srpiare miles of cxtrusives.
The combincnl area of all the mapped alkaline rocks (including syenites
and trachytes) in the ( 'ordill<Tan cpiadrangles is only about 135 square
miles out of a grand total of about 15,000 srpiare miles of igneous
rocks.
The totals for the alkaline rocks will doubtless be enlarged as more
detailed petrographi<* work is done* in the Cordillera, but it seems
certain that all totals for this group of rocks must, in this vast belt,
always remain extremely small as <'ompared with the totals for the
subalkaline types.
The sixteen Appalachian folios show only 0.3 sc|uare mile of
nephelite sy(»nit(» and 2.5 xpiare miles of syenite* porphyr>' to represent
the entire alkaline group of rocks (including syenites and trachytes),
although 1G59 s^iuare miles of igneous rocks are mappeil in these folios.
DISTRIBUTION AND QUANTITIES OF IGNEOUS-ROCK SPECIES 47
In all the folios the combined area of all mapped alkaline rocks
is about 140 square miles out of 16,700 square miles of igneous forma-
tions mapped.
Considering the relatively small thickness represented in most of
the alkaline bodies mapped in these folios (sheets, laccoliths, lava
flows, beds of pyroclastics) , it is tolerably certain that the total maxi-
mum volume of alkaline rock exposed in these quadrangles is far less
than one-half of 1 per cent, of the total maximum volume of sub-
alkaline rock.
The same truth is to be inferred from a general review of North
American petrographic geology. Including the large syenite, nephelite
syenite, and malignite masses mapped in New York State (150 square
miles), New Hampshire (21 square miles), Ontario (100 square miles),
and British Columbia (200 ± square miles), the total area of alkaline
rocks (including syenites) actually mapped in North America, outside
the United States Geological Survey folio quadrangles, is slightly over
500 square miles. In all, then, about 700 square miles of alkaline
rocks have been mapped on this continent.
In the Cordilleran belt there are about 170,000 square miles of
post-Cambrian plutonic rocks indicated on the new map of North
America published by the United States Geological Survey, 1911.
These rocks are chiefly granodiorites and granites. .In the same belt
over 400,000 square miles of post-Cambrian volcanic rocks are shown
on the map; most of the extrusive bodies are basalts and the rest are
chiefly basaltic andesites.
In the Appalachian mountain system the same map shows about
30,000 square miles of post-Cambrian extrusives, which are known
to be almost entirely subalkaline and chiefly granitic.
The post-Cambrian intrusives and extrusives of this continent are
known to cover total areas whose orders of magnitude are, respectively,
200,000 square miles and 400,000 square miles. It is safe to say that
the pre-Cambrian granites (with the orthogneisses) cover at least
1,000,000 square miles.
We may conclude that all the known alkaline rocks of North
America (including the syenites and monzonites) have a total area less
than .05 per cent, of the total known area of the igneous rocks. In
view of the striking field appearance of most of the alkaline types and
in view of their superior ** interest*' for petrographic geologists, one
may readily believe that these rocks are somewhat over-emphasized
in the existing maps and memoirs. The sampling of this continent
80 far accomplished indicates, in fact, that its visible alkaline bodies
have a total volume probably less than one-tenth of 1 per cent, of the
total volume of its visible subalkaline bodies.
48 IGNEOUS ROCKS A\D THEIR ORIGIN
A partial canvass of the geological maps of Europe has convinced
the writer that the eorresponding ratios f<»r that continent are of the
same order of magnitude. Without exception the large-scale maps
of European countries are much l(»ss siitisfactor>* in a study of this kind
than are the fr»iios of the United States Geological Sur\'ey. The
writer found that the time and labor required to assemble the maps
an<l memoirs and to weigh them properly, so as to allow for variable
petrographic standards. (»tc.. were practically prohibitive. It is much
to be desired that quantitative studies of the igneous terranes be under-
taken by each geological commission in Europe.
The number of alkaline bo<lies in Europe is large, but, i»nth a few
except icms, they are all very small. Though phonolite was first
named in (iermany. th(» eye is much strained to find the tiny spots of
color for phonolite on Lepsius's wonderful map of the empire. The
Tertiar>' nephelite basalts, n.elilite basalts, and limbuncitm are
exceptionally well developed in Ctermany. but, taken together, the>'
are probably far inferior in volume to the ordinary- plagioclase basalts;
and, in any cas<', cannot be compared in inferretl volume with the pre-
Tertiary eruptive rocks of the empire. Though trachyte was first
named in Fnmce, neither this well-known species nor phonolite could
be profitably shown with separate color on the new, beautiful wall-map
of France published by the government sur\'ey. Inspection of that
map or of the large-scale she(»ts of th(» French sur\'ey will readily
in<licate to the reader that the total volume of the P^rench alkaline
bodies must be much less than 1 per cent, of the combined subalkaline
bo<lies in that count rv. A similar statement mav be made for Great
Britain and Ireland, for Switzerland, for Finland and for Spain. In
Scandinavia. Italy, and perhaps Portugal, the alkaline rocks are rela-
tively mon» voluminous. l»ut it is questionable that, in any one of
thcM' count rii's, they have a total volume which is 5 per cent, of the
total for the subalkaline bo<|ies.
For Europe as a whole the writer suspects that this ratio is less
than one to one hundred.
Though so little is known of Asia. .Vfrica. .Vustralia, ami Antarctica,
it st'cms highly probable that for each of thes<» continents the ratio is
iigain less than one to (»ne hundred: in fact, there is nothing to indicate
that it is greater than one to out* thou*<and.
Whatever criticism may be a<lvan<*ed agaMist this *' extrapolation**
fr(»m so 'ncomplete a body of known facts, a definite conclusion on this
matter is aln'ady clear. The ratio of V'.»lumes is certainly extremely
small for North AnitTii-a and then* i^ no reason to doubt that this
continent is a fair >aniple of all the laiuN i»f the gh»be. When it is
furtiier renu^mbered that the number of known liK*alitics for the alka-
^u^^^D
DISTRIBUTION AND QUANTITIES OF IGNEOUS-ROCK SPECIES 49
line rocks has been greatly and specially increased during more than
thirty years of enthusiastic search, we are prepared for the view that the
alkaline clans, taken together, are, as it were, only incidental products
of a planet whose eruptions have been overwhelmingly of the subalka-
line quality. The significance of this basal fact will be noted in
Chapters XIX and XX.
Relative Abundance of the Subalkaline Clans
Returning to Table IV, the reader will observe significant quantita-
tive relations between the diorites, gabbros (including gabbrodiorites
and anorthosites), and peridotites on the one hand, and granite and
granodiorite on the other.
Though the first three plutonics named are found in a large number
of distinct bodies, their total areas in the folio quadrangles are re-
spectively less than 4, 10, and 2 per cent, of the total area of granite;
and respectively less than 6, 19, and 4 per cent, of the total area of
granodiorite. Like the many alkaline species as a whole, each of
these basic plutonic types is quantitatively very subordinate to the
add types. A review of the maps and memoirs concerning other
parts of North America, as well as the other continents need not be
absolutely complete to convince one that diorite, gabbro, and perido-
tite, either singly or collectively, have, when compared to granite and
granodiorite, subordinate place in the world's terranes; the ratio of their
total areas, over the whole earth, will have the same order of magnitude
as that deduced from the United States folios. In principle this state-
ment is obvious to any informed petrographer, but the figures will tend
to keep these fundamental facts always in view as the genetic problem
is discussed.
Equally evident and equally important is the fact that the extrusive
members of the basic subalkaline clans excel the extrusive members of
the granite and granodiorite clans, both in total areas and in average
thickne^ises. It may be observed that the United States folios cover
most of the Yellowstone rhyolite plateau, almost certainly the greatest
body of this type of lava in the world. A fairer estimate of the true
relative areas would be gained by deducting the 1750 square miles of
rhyolite recorded in the Yellowstone Park folio. If, however, this
rhydite be retained in the total, and if there be added to it the enor-
mous areas of basalts and andesites in the part of the Cordillera not
OQfvered by the folios, the relative insignificance of the volimie of acid
lavas, as compared with that of the basalts or with that of the basic
andesites, is made even clearer. It is questionable that the total
volume of the visible rhyolites of North America is as much as 1 per
I cent of the total volume of the visible basalts or that of the visible basic
60 IGNEOUS ROCKS AND THEIR ORIGIN
andeaites. The writer believes, though on less secured evidence, that
the same order of magnitude characterizes the ratios for the rest of the
world.
These general conclu»ions may be conveniently illustrated in
tabular form:
20
Ratioof total area of world's granite to total area of world's diorite. > ~p
Ratio of total volume of world's rhyolite to total volume of worid's ^ _L
andesite 1^
Ratio of total area of world's granite to total area of world's jM)
gabbro 1
Ratio of total volume of world's rhyolite to total volume of world's ^ J_
basalt 60
Though the exact figures of this table are, of course, now im-
possible of verification, they serve to indicate one of the most vital facts
in petrolog>'. The basic subalkaline clans predominate in theextnisive
phase of igneous action; the acid subalkaline clans predominate in
its intrusive phase.' This indubitable fact needs explanation and» in
the writer's opinion, that explanation must be at the very heart of a
general theor}' of petrogenesis.
KocK Species known only in TIxtremely Small Areas or Volumes
AT THE Earth's Surface
Out of the ten families of plutonic rocks discussed in Rosenbusch's
hand-book (4th edition), four families are but feebly represented
among the world's terranes as known at the present time. The total
known area of the bodies corresponding to each family is very probably
less than 60 square miles. A c1os<t statement of their respective knon^Ti
quantities is:
Superior limit of
Family known total area
Essexites 50 square miles.
Shonkinitcs (largely sheets and
lacoolitliM) 20 square miles.
Theralitos (largely sheets and
laecolithfl) 20 square miles.
... .. . r .. _ I Missourites 1 square mile.
Missountcs and Fergusites. ^ v^^^,-.^ i «^ -i
I rergusites .1 square mile.
»i !•* J n \,'^v;^:4^ ^ Ijolites. . . .5 square miles.
Ijobtes and Bekinkinites ^ Bekinkinites 1 square miW.
'Though plutonic and effusive representatives of the granite clan are ex-
tremely abundant in Fennoscandia, Sederholm states that "it is astonishina not
to find any abyssic rooks certainly connected with the rocks of the diabase which
have erupted during every time of quiet sedimentation in Fenno-Scandia, either
forming effusive beds, or being intercalated between the strata of the sediments."
(J. J. Sederhohn, Bull. Comm. g^l. Finlande, No. 22., 1907, p. 108.)
Shonkinites and Theralites . *
DISTRIBUTION AND QUANTITIES OF IGNEOUS-ROCK SPECIES 51
A close quantitative view of the dike-rocks is, of course, impossible,
but, on the average, no type in this class of rocks is likely to have
a total volume as much as 5 per cent, of the volume of its parent plu-
tonic chamber. For most of the species the ratio is probably less
than 1 : 100. These figures, as well as the actual field relations, show
that the genetic problem of the dike-rocks is largely bound up with the
problem of the plutonic masses.
Difficult as it is to obtain an idea of the volume of plutonic bodies
from the surface outcrop, it must be at least as impracticable to
estimate the true volumes of the world's extrusive masses, each of
which is of ever-varying thickness. Of the fourteen families listed in
Rosenbusch's hand-book, eleven families represent rock masses
now known only in small total volumes respectively, as compared with
the family of basalts (melaphyre and diabase). The family of
andesites and porphyrites may have a total volume of the same order
of magnitude as that of basalt. The family of the "quartz trachytes
and quartz porphyries*' (including rhyolites) comes next to those
two families in respect of total known volume, but is clearly quite
subordinate to them, as already indicated. No one of the other eleven
families is known to have a total volume as much as 1 per cent, of the
basaltic masses already demonstrated on the globe. The families of
relatively small total volumes may be listed as follows:
Trachytes and quartz-free porphyries. Leucite rocks.
Phonolitic rocks. Nephelite rocks.
Dacites and quartz porphyrites. Melilite rocks.
Picrites and picrite porphyrites. Limburgites and augitites.
Trachydolerites. Lamprophyric effusive rocks.
Tephrites and basanites.
Summary of Conclusions regarding Relative Abundance
All of the above-mentioned estimates are phrased in terms of the
total volumes of the rock masses actually mapped and studied by
modem petrographic methods. They are all, of course, subject to
the criticism that future work may tend to increase the relative volumes
of the subordinate families. This is likely to be the case with the
extnisives, though the writer believes that the orders of magnitude
represented in the foregoing estimates will always apply. An atten-
tive study of petrographic literature, as it issues from the press year by
year, shows that new discoveries do not tend to alter the position of
the rock families noted above as dominant in the world. This is
true, notwithstanding the tendency of petrographers to publish
material concerning the rarer types and to neglect the others. As
already noted, this psychological factor has thus long tended to place
52 . IGNEOUS ROCKS AND THEIR ORIGIN
misleading emphasis on the qualitative side of petrology. The more
objective product of the great government surveys is a useful C5orrec-
live to this tendency. P'or this reason, the writer has laid special stress
on the work of the United States Geological Survey, the Geological
Survey of Canada, and the surveys of Western Europe, where the
microscope and chemical analysis have been systematically implied.
The assumptions are here made: that each survey has fairly sampled
its territory, giving a picture of the average quantitative relations
of the rock in each country; and that North America and Europe are,
in their turn, fair samples of the whole area of the continental plateaus.
These assumptions are already so well founded on recorded facta that
certain generalizations seem to be unshakable, even though they are
built on limited exact knowledge concerning the earth.
1. The visible alkaline rocks of the world (including the syenite
and monzonite clans) probably have a total volume less than 1 per
cent, of that of the visible subalkaline rocks (granite, granodiorite,
diorite, gabbro, and peridotite clans).
2. Among the visible intrusive rocks, the granites and granodio-
rites together have more than twenty times the total area of all the
other intrusives combined.
3. Among the vi.^^ible extru.sive rocks, basalt has probably at least
five times the total volume of all the other extrusives combined;
basalt and pyroxene andesite together have at least fifty times the
volume of all the other extrusives combined.
4. The granite clan is highly predominant among the intrusives
and is one of the more subordinate clans represented in the extrusives.
5. The diorite clan is decidedly subordinate among the intrusives
but rates next to the gabbro clan among the extrusives.
6. The gabbro clan is likewise subordinate among the intrusives,
but distinctly predominates among the extrusives.
7. Quantitatively considered, the igneous rocks of the globe
chiefly belong to two types, granite and basalt. This truth was
long ago recognized by Durocher, von Cotta, andBunsen, and later by
Michel I/»vy, Lo<'winson-Lessing, and others. But petrologists have
still to become unanimous in fully appreciating that the one of these
dominant types is intrusive, and the other extrusive. To declarethe
meaning of this fact is to go a long way toward outlining petrogenesia
as a whole.
MaXIMIM SiZK of iNniVIDlAL B0DIE.S
To visualize the grnrtic problrm ideally it would be further nec-
es8ar>' to know the average siz<' of the rock bodies corresponding
respectively to the intrusive and extru.^ive members of each of the
DISTRIBUTION AND QUANTITIES OF IGNEOUS-ROCK SPECIES 53
great clans. For obvious reasons this is now impossible even for the
thoroughly mapped regions of the earth. However, it is well to
supplement the foregoing statements of relative quantities by an esti-
mate of the actual areas covered at the earth's surface by the represen-
tative largest known masses of the important intrusive types.
TABLE V
Rock type
Location of rock bodies (intrusives)
Area in
square miles
Granite (pre-Cambrian)
Granite(post-Cambrian)
(laccolith).
Granodiorite
''Quartz monzonite^'
Quartz diorite (tonalitc)
Diorite (stock)
Gabbro (laccolith)
Anorthoeite (laccolith?). . . .
Peridotite
Pyroxenite
Subalkaline syenite
Pulaskite
Nordmarkite
Nephelite syenite (lacco-
lith?).
Laurvikite
Malignite
Monzonite
Essexite
Shonkinite (laccolith)
Miaaourite (stock)
Fergusite (stock)
Theralite (laccolith)
Ijollte
Bekinkinite
Post-Bottnian "central" granite of
Finland.
Bush veldt, Transvaal
Sierra Nevada
Bitterroot Range, Idaho
Southern Alaska
Little Belt Quadrangle, Montana. . . .
Duluth and northeastward, Minnesota
Saguenay district, Quebec'
Mt. Stuart Quadrangle, Washington . .
Mt. Diablo, California
Ceard, Brazil
Coryell batholith. West Kootenay
district, British Columbia.
Nordmarken, Norway
Kola peninsula, Lappland
Laurvik, Norway
Pooh Bah lake, Ontario
TeUuride Quadrangle, Colorado.
Shefford Mountain, Quebec
Square Butte, Montana
Shonkin stock, Montana
Amoux stock, Montana
Northern Crazy Mts., Montana.
Kuusamo parish, Finland
Bekinkina Mts., Madagascar. . .
9,100
10,000=*=
20,000=*=
3,000=*=
5,000=*=
25
2,400
5,800
40 +
2
500=*=
100
800*
700*
(?)
250=*=
15
8
1.5+
4
< .7
<1
4
<1.0(?)
1.0(?)
This table also emphasizes the relative insignificance of the alka-
line rocks; the clear dominance of the more silicious types among plu-
tonic bodies; and the subordinate place of basaltic (gabbroid) magma
in the world's intrusions.
General Remarks
Two of the laws of distribution are truisms to geologist and
petrographer. Magma of basaltic composition is clearly the most
^An area east of the Moisie River, Quebec, is estimated to cover 10,000
iquare miles, if mapped correctly.
64 IGNEOUS ROCKS AND THEIR ORIGIN
widespread, occurring in plains, in mountains, and in all of the ocean
basins. Granites, granodiorites, and most diorites are continental
rocks and, more speeifically, mountain rocks.
Petrographical provinces exist, but one such province may be
more or less closely duplicated in another continent or ocean basin. The
rock types found in the Monteregian Hills, Quebec, correspond in
many essential details ynth those of the Adirondacks, of Mount As-
cutney, Vermont, or of the Christiania Region, Norway. The qiecial
combination of anorthositic and syenitic species in the Adirondacks
is remarkably like that of western Norway. The rock associations
of the Bohemian alkaline province are more or less fully repeated at
the Great Rift of Africa, in central France, in Tahiti, in the Canary
Islands, in the Azores, in Hawaii, etc. Many more examples will be
noted in the following chapters. They all go to show that the petro-
genie mechanism has worked in practically the same manner beneath
each of the continents and, with exceptions to be further discussed,
beneath each ocean basin. Sameness in the midst of variety is a prin-
cipal result of the earth's igneous activity.
This fundamental fact is obscured by belief in the existence of
"Atlantic'' and "Pacific" branches into which igneous rocks are to
be divided. The inappropriateness of these terms is manifest from
many considerations already dealt with by different writers; little need
be added on the present occasion. (See pages 338 and 412.)
CHAPTER IV
ERUPTIVE TYPES AND THE GEOLOGICAL TIME SCALE
Eruptive Sequences
Having obtained at least a limited view of the space relations of
igneous rocks we now turn to the subject of their time relations.
To bring illustration of the essential facts within practical limits the
writer has constructed Table XXI (Appendix B).
The table is intended to be representative of the many thousand
sequences actually determined for local areas of the world's igneous
rocks. The first part exemplifies the pre-Cambrian succession of
those rocks. The second part states the case for typical districts where
the order of eruption is known for the pre-Cambrian as weil as for the
later periods. The third part relates to eruptive sequences occurring
only in post-Cambrian time. Some of the sequences refer to igneous
development in whole countries, like the British Isles or Germany;
others refer to very small areas, like the Eolian Islands or the single
island of Vulcano. Either kind of statement, supported by the de-
tailed facts of the original monographs, has its special value in petro-
f^enic theory.
The order of eruption is indicated partly by the geological dates,
and in all cases the statement begins with the oldest eruptive. Within
a given geological period the order is shown by numerals. In some
instances a half-dozen or more types are referred to the same geological
period, >^athout any note as to the relative ages of those types. Gen-
erally this means that information on such a group is lacking; much
more rarely it means that the eruptions are known to be essentially
contemporaneous.
In an extensive region affected by igneous action, the eruptions
jjenerally, if not always, form groups which are separated in time by
long intervals. The intervals may be one or more of the standard
geological periods, like the Triassic or the Eocene; or they may be
^separated by a long epoch of erosion. For example, the twenty-
five rock species listed in the Upper Huronian and Keweenawan of
the Lake Superior district were erupted after the development of the
peneplain of the "Eparchean Interval." During so long a time as
that represented in peneplanation, the magmas which had been in-
jected into the earth's crust must have crystallized to a depth of several
miles below the roofs of their chambers (dike, stock, or batholitb).
55
56 IGNEOUS ROCKS AND THEIR ORIGIN
Ah shown in Chapter IX, all basic cruptives at least are due to injec-
tion of the crust from beneath. This was undoubtedly the case with
the dominant magma erupted after the close of the "Eparchean
Interval." Similar crustal injection occurred in the Lake Superior
district at least ivricc before this peneplanation. The visible igneous
rocks of the region are, therefore, to be referred to at least three
jietrogenic periods or cycles. Many of the series in Table XXI also
show long time intervals during which igneous action was dormant.
The positions of these time-gaps are indicated by solid lines in the
table. The gaps have there been recorded conservatively, so that
the total numl>er of petrogenic cycles indicated is a minimum.
Recurrence of Types Belonging to the Gabbro Clan.
The chemical type most often represented in past eruptions is the
basaltic. In North America the oldest recognized igneous formation,
the Keewatin, is largely a greenstone, originally of basaltic composition.
In Europe the oldest recognized members of the "Archean" contiun
''metabasites" of similar original nature. There is no younger igneous
rock on the globe than the basalt which at this moment is splashing and
solidifying in the crater of Kilauea. This already well known fact, the
persistence of the basaltic type in past eruptivity, Ls illustrated in Table
XXI. Gabbro, cliabase. basalt, or their chemical equivalent appears in
66 of the 62 series listed in the table; and in 96 of the 116 dilFerent
petrogenic cycles there indicated. This type of magma is recorded in
the rock group initiating th(» sef|uence for each of 68 cycles.
The immediate areas covered by the named authors dealing with
the Belknap Mountains and Ked Hill of New Hampshire, with the
Goldfield district of Nevada, with the Ouray-Silverton quadrangles
of Colorado, and with the* trachytic areas of the Carpathian Moun-
tains, have no stated outcrops of rock belonging to the gabbro clan.
In each of these tracts, however, such rocks apiM'ar at liistances of a
few miles from the limits of the tract. It is, in fiu*t. highly probable
that for any area as large as the state of Massachusetts (8,000-|- square
miles), the eruptive swiuence includes rocks referable to the gabbro
clan.
Further, the geological record shows that, from an early pre-
Cambrian epoch to the pres<»nt. basaltic lavas have been extruded
not only the most often but also always in the greatest average
volume, during any one of the longer divisions of the time record.
This statement is obvious to any inforineil worker with the world's
geological maps. Sufhcirnt cvidencr for its truth is the long list of
dated fissure eruptions to be found in a later paragraph of this chapter.
ERUPTIVE TYPES AND THE GEOLOGICAL TIME SCALE 57
The masses formed by such eruptions are overwhelmingly basaltic and
are almost always greater in volume than the other volcanic masses
of the same geological period.
On the other hand, the intrusive members of the gabbro clan, in
the pre-Cambrian and in each of the following geological periods,
have always been subordinate to the granites. Thus, the dominant
intrusive type has always been granite, while the dominant extrusive
type has always been basalt. We have already found the same
important relation registered in the present distribution of igneous
rocks, regardless of age.
General Recurrence of Other Clans
Supplementing the information contained in Table XXI with that
embodied in the thousands of modern geological maps and memoirs,
a general conclusion regarding rock types other than basalt and its
chemical equivalents, becomes apparent.
Excepting the missourite-fergusite and theralite clans, all the
clans are represented by eruptives in both pre-Cambrian time and
in Cenozoic time. The two clans named are not shown to be rep-
resented among the pre-Cambrian eruptives, but their discovery in
that oldest terrane may be made at any time, and in any case
these clans are of almost negligible quantitative importance in the
record of igneous rocks now determined for all the geolog;ical periods.
Thus, in the first of the greater divisions of geological time, as in the
last division, the eruptives were of the same quality. Similarly, most
of the clans, including all the dominant ones, are also represented in
the Paleozoic eruptive bodies and in the Mesozoic bodies. An
ultimate theory of the igneous rocks must, therefore, recognize a mech-
anism which has produced eruptions of each of the principal chemical
types at intervals from at least later pre-Cambrian time to the present
time.
Time Relations of the Granite, Diorite, and Granodiorite Clans.
—Yet, for some of the clans, nature has varied the accent in the record
of earth history. Regarding area alone, more than nine-tenths of the
world's granite (including orthogneisses, excluding the granodiorites)
is pre-Cambrian in date. Allowing for the pre-Cambrian granites
underlying the later sediments, this fraction is likely to be nearer
nmety-nine one-hundredths.
Notwithstanding the enormous erosion which has affected the
pre-Cambrian terranes, they are also specially rich in the effusive
Kpresentatives of the granite clan. The visible roots of the pre-
Cambrian (extrusive) leptites and quartz porphyries in Fennoscandia
58 IGNEOUS ROCKS AND THEIR ORIGIN
and of the similar types in the Canadian shield have total area and
inferred volume so great as to imply special intensity of rhyolitic erup-
tion in the pre-Cambrian.
In spite of the incompleteness of petrographic surveys it seems
fairly probable that the visible masses belonging to the graoodiorite
clan (including many quartz monzonites and quartz diorites) are chiefly
of post-Cambrian dates.
Throughout post-Keewatin or post-Bottnian time, the intrusive
members of the diorite clan seem always to have been very subordinate
in volume. On the other hand, the effusive members of this clan, rep-
res<»ntod by many porphyrite and greenstone masses of pre-Cenozoie
age, as well as by the less altered andesites, have very generally rated
next to the basalts in average volume, if not in frequency of eruption.
Time Relations of the Alkaline Clans. — ^Jensen has recently argued
that the alkaline rocks of the world are almost entirely of Tertiarj*
dates.* In vi(»w of the multiplying discoveries of pre-Tertiar>'
nephelite syeniti^s, malignitcs, munzonite, etc., this statement must
be seriously qualified. It is only natural that the pro-Tertiary
alkaline mnss<'s, because always of relatively small volumes, should
have bet»n d(?stroyo<l or gn^atly diminished by Tertiary or earlier
erosion, or else c<>mpl«*t<»ly buried uucUt Tertiar>' or older sediments.
Nevertheless, it mav still be ii fact tfiat the Cenozoic era was a time for
the spe<'ial development of alkaline eruptives, lx>th in breadth of
geographical distribution and in total volume.
Special Time Relation of Anorthosite. — Anorthosite represents
the one plutonic tyix* whose cruptioas seem to be more or less defi-
nitely confiniMl to one part of geological time. By far the greatest
amount of this rock constitutes mass(»s t)f pre-Cambrian date. So far,
the writ(*r has <lisc()vere<l in the Iiteratun» references to only one region
when* anorth(»sit<» of post-Cambrian date has lK?en found on any
notable s<'ale. That is tin? rcKion of the Norwegian anorthosites
describe<l by KoMenip, who refers them to a late Silurian or post-
Silurian i»|H)ch. SiiK'c this tyjM' is so conspicuous in the field, its re-
currence in the pre-( *anii>rian an<l it> very small volume in the younger
terranes is not to be explaiiieil as due to the ac(*i<lents of geological
discovery. The problem of orijiins must, then, take account of this
relation of the anorthosit<-s to geological age. (See Chapter XV.)
Dike-rocks in Geological Time. What has been .^^tated regarding
time relations of the r\iii\< in geiuTal will, of course, apply to the
aschistic dik<'-rocks. The diax'histic types ne<Ml a special induction.
This caiuu»t vi't In* carried nut with even tli:it tlegree c»f completeness
|H>ssil)h' for the priiu-ipal claii**. Tlie exact dating of <likes in the geo-
>H. I. Jenflcn, Prot*. Linn. Sui-., New South Wales, Vol. 33, 1908, p. 401.
ERUPTIVE TYPES AND THE GEOLOGICAL TIME SCALE 59
logical record is generally more diflScult than that of the larger intru-
sive masses. Moreover, it is only within the last twenty-five years
that much attention has been given to the "complementary" dikes.
So far, the diaschistic dikes known to be of pre-Cambrian age are
almost wholly aplites and pegmatites. A malchite from the Marquette
district of Michigan and a tinguaite (heronite) from Ontario are other
pre-Cambrian dikes of this class noted by Rosenbusch in his hand-
book. A few other types, including the camptonite, alnoite, monchi-
quite, and tinguaite, of Alno, Sweden, may date from the late pre-
Cambrian, but absolute certainty on that point is not yet assured.
Thus, in spite of the enormous development of igneous rocks in the
oldest terrane, diaschistic dikes other than the commonest kinds,
granitic aplites and pegmatites, seem to be very rare in it. The
rock types which are known only, or almost wholly, in post-Cambrian
representatives are : paisanite, bostonite, gauteite, solvsbergite, minette,
kersantite, vogesite, odinite, cuselite, spessartite, fourchite, and
ouachitite. The many expert petrographers who have ranged widely
over Fennoscandia, the Canadian shield, the Adirondacks, the
numerous pre-Cambrian areas of the North American Cordillera,
the peninsula of Hindustan, and other smaller areas of the ancient
terrane, have certainly not been oblivious to the attraction which
diaschistic dikes have long had for workers in the younger rocks.
On the contrary, it appears safe to hold that these particular dike-
magmas have been chiefly developed during post-Cambrian time.
Modes of Eruption and Geological Time
As shown in Table VI, page 98, intrusion of the batholithic type has
been recorded in the pre-Cambrian (at least three distinct epochs), at
the close of the Ordovician(?), and in the Devonian, the Carboniferous,
the Jurassic, the Early Tertiary, the Miocene, and (probably) the
Pliocene. Each of the major geological periods has doubtless witnessed
the injection of dikes and of sills or laccoliths, as well as volcanic erup-
tions of the central type. Fissure eruptions of basaltic magma were
incidents in the history of many periods, as illustrated in Table X,
page 191.
It is generally agreed that the pre-Cambrian was an era of specially
extensive and prolonged batholithic intrusion. Of late years evidence
has accumulated that it was also a time of intense vulcanism. Gigantic
fiasore eruptions and pyroclastic developments then formed many
basaltic masses which are probably without parallel among later for-
mations. In Chapter XV the probability is stated that the enormous
pd>bro and anorthosite bodies of the pre-Cambrian are of laccolithic
60 IGNEOUS ROCKS AND THEIR ORIGIN
origin. If that be true, it follows that conditions were extraordinarily
favorable for the transfer of these magmas into the stratified terranes.
Moreover, the lit par lit injection of acid magma is an amasinf^y per-
sistent feature of the British Columbia pre-Cambrian and the writer
has come to suspect that some of the so-called batholiths of other
Archean areas are really thick sills. In brief, igneous action in all its
principal phases peculiarly characterized the pre-Cambrian era not only
as regards persistence in time and wide extension in space, but also with
respect to the scale of the magmatic movements.
SuiOfARY
Incomplete as induction must be at the present day, certain general
conclusions appear to be justified.
1. In a qualitative way the intermittent development of igneous-
rock bodies through the geological record has followed uniformitarian
lines. In their chemical diversity as in their modes of eniption, the
visible pre-Cambrian, Paleozoic, and Mesozoic bodies are, in general,
similar to the corresponding Tertiary types.
2. Exceptions to these rules are to be found: (a) in the excessive
development of true granites in the early pre-Cambrian; (b) in the
restriction of all important masses of anorthosite to pre-Devonian,
and usually pre-Cambrian, daten of eruption; (c) in the special de-
velopment of alkali no rocks and of the granodiorite clan in post-
Cambrian, especially post-Paleozoic, time; (d) in the special diversifi-
cation of diaschistic dikes in post-Cambrian time; (e) in the special
strength of the earth's eruptivity — batholithic, volcanic, and lacoolithic
— during the pre-Cambrian. The explanation of these certain or
highly probable facts are so many items in the petrogenic problem.
CHAPTER V
INJECTED BODIES
Classification of Intrusive Bodies
To complete the summary of the facts which it is the duty of any
petrogenic theory to explain, it is necessary to review the types of
form assumed by igneous rocks. These types almost always show
some degree of transition into one another. An intrusive sheet or
laccolith is generally continuous with its feeding dike or dikes. Many
sheets, laccoliths, or dikes communicate with their parent batholiths.
Even the distinction between intrusive and extrusive bodies is oc-
casionally difficult to make in the field, where, for instance, a dike
passes at the top into a lava flow, or where the material of a batholith
is poured out on the surface. Nevertheless, the existence of such
transitional phases cannot aflfect the supreme value of a classification of
igneous masses on the basis of their forms and field relations.
On account of their greater volume and higher importance in
petrogenesis, the intrusive bodies will be here treated before the
extrusive bodies. In 1905 the writer published a paper on the classifi-
cation of the intrusive bodies. A scheme was there outlined which
seemed to combine the best elements of geological tradition on this
subject. The reader is referred to the paper itself for a statement of
the grounds on which the classification and corresponding definitions
are based.* The needs of this book will be met by a brief statement
of the definitions, accompanied by fuller illustration of the world's
intrusive masses. Some paragraphs of the 1905 paper will be quoted.
The general adoption of a consistent, well-defined scheme of types —
a scheme as complete as possible, but elastic enough to admit of new
types — would tend to make field descriptions more scientific than
many of them are at present. Such general adoption would mean a
lain in precision, in the ease with which a description of igneous intru-
sions would be understood, and in an economy of words. The filling-
out of the scheme of classification to an extent quite beyond that
now prevailing in standard text-books of geology would further have
the effect of sharpening the eyes of the field observer. He may perhaps
not be content to describe a given granite intrusion as simply a " mass,"
or an "area," or an ''outcrop," if it be possible by the study of its
» R. A. Daly, Jour. Geology, Vol. 13, 1905, pp. 485-608.
61
62 IGNEOUS ROCKS AND THEIR ORIGIN
^contacts to indicate the real form and relations of the granite body.
The use of the term ''mass" in that sense is often excellent because
of the apparent impossibility of determining the true shape of the
granite l>ody; but such justifiable employment of the term implies
that that particular body cannot as yet l>e thoroughly classified. A
rather negative name has in such a case a distinct value. Of
yet greater value is the positive reference of intrusive bodies to
definite categories. A good observer always feels the pressure of
the category. If his classification be systematic, his observing power
is quickened, his report enriched; if his classification be that in gen-
eral use, his descriptions will be of the greater service to the science.
For a given body the method of intrusion is the most important
criterion that could be used in classification. If it might be determined
in every detail just how the igneous mass reached its present position,
the form of the body and its relation to structural planes in the
country rock would therewith be kno\ni. A genetic, and therefore
natural, classification would thus \xi founded on the method of in-
trusion. In the present state of geological science it is, however,
impossible to apply this fundamental principle throughout the
established list of intrusive bodies.
The greater number of recognized tyjn^s are those of lM>dies of
magma which is exotic except for a small, variable i>ortion of it due to
contact fusion. In each of these cas«»s the magma has come into its
chaml)er through chann(»Is which 1jjiv(» fed the growing lx)dy from larger,
deeper-lying, generally invisible res<Tvoirs. The chamlx»r is due to a
parting of the country rock into which the magma is injected. .\n
injected body is thus one which is entirely inclosed within the inva<led
formations, except along the relatively narrow openings to the chamber
where the latter has b(^»n in communication with the feeding rcser\'oir.
On the other hand, stocks, boss<»s, and batholiths never show a
true floor. They appear to communicate <lirectly with their respective
magma res<»rvoirs. Kach of thcs(» !)odics shows field relations sug-
gesting that it is a part of its maf^ma reservoir and that communication
\iith the magmatic interior of the earth is not established by narrow
openings, but by a huge, downwardly enlarginf^ o])4>ning through the
country rock. In relation to the invade<l formations a stoi'k, boss, or
batholith is intrusive, but is suhjacent rath«T than injected.
How a batholithic res«»rvoir is enlarged along its (x>ntacts
is a matter permitting as yrt of no absolute certainty. In .«'parat-
ing intrusive bodies into two primary divisions, one including all
injected Iwdies. the other including subjarent bodies, a classificatioti
will do good service in emphasizing the need of further investigation
into the mechanics of intrusion. Xo one has yet proved that MU
INJECTED BODIES 63
«
granite mass over 200 square kilometers in area and characterized by
vertical or outwardly sloping contact surfaces, is due to injection.
Whatever may be the probabilities, all geologists are not agreed that
such a mass has been intruded by any kind of assimilation of the
invaded formations. Some light has been shed on the origin of
batholiths and stocks, but they are certainly not understood as are
dikes and sills.
So far as the method of intrusion is concerned, therefore, stocks,
bosses, and batholiths belong to a primary division of intrusive
bodies which may be defined as not demonstrably due to injection.
The principle is negative, it leaves the method of intrusion unstated,
but it brings into clear relief a principal contrast existing between the
greatest of intrusions on the one hand, and dikes, sheets, laccoliths,
etc., on the other.
The subdivision of the two classes is given in the following table,
which is a somewhat altered reprint of that in the 1905 paper.
A. Injected Masses.
I. Concordant injections (injected along bedding planes).
1. Intrusive sheets, homogeneous and differentiated.
a. Sills.
(1) Simple.
(2) Multiple.
(3) Composite.
b. Interformational sheets (at unconformities).
2. Laccoliths, homogeneous and differentiated.
fl) Simple: symmetric and asymmetric.
(2) Multiple.
(3) Composite.
(4) Interformational (at unconformities).
3. Phacoliths.
II. Discordant Injections^ (injected across bedding planes).
1. Dikes, homogeneous and differentiated.
(1) Simple.
(2) Multiple.
(3) Composite.
2. Eruptive veins: contemporaneous veins.
3. Apophyses or tongues.
4. Necks, homogeneous and differentiated.
5. Bysmaliths.
6. Chonoliths.
'Called "transgressive intrusions," by A. Harker, The Natural History of
Rocks, New York, 1909, p. 61.
64
IGNEOUS ROCKS AND THEIR ORIGIN
7. Ethmoliths.
8. Sphenoliths.
B. Subjacent Masses
I. Stocks and bosses, homogeneous and differentiated.
(1) Simple.
(2) Multiple.
(3) Composite.
II. Batholiths, homogeneous and differentiated.
(1) Simple.
(2) Multiple.
(3) Composite.
Igneous Injections
An intrusive sheet is a tabular injected body lying parallel to the
bedding plane of the country rock.
In accordance with A. Goikie's definition, the standard in general
usage, a sill is a sheet of igneous material which has been injected
into a sedimentary series and has solidified there, so as to appear more
or less regularly intercalated between the strata (Figs. 1, 2, and 3).'
KL
J Mil
L^ K,
Fio. 1. — Outcrop of lamprophyric sill in flat Laramie sediments (KL), Colorado.
(After Spanish Peaks Folio, U. S. G. S., No. 71, 1901.)
Ideally, it would be well to distinguish the class of sills which have made
room for themselves by lifting flat, overlying strata, from another
class which have l)een forced into vortical strata. The former class
represents a mechanism like that of the laccoliths; the latter represents
a mechanism like that of dikes as hereafter defined. But experience
shows that the ideal subdivision cannot always be applied in nature and
some authorities are content to use the term **siir* without directly
' Reprint, with originals of Figs. 2 and 3, kindly supplied by Mr. Barker.
INJECTED BODIES
implying the nature of the intrusive mechanism. Most of the greater
recorded sills are of gabbroid or diabasic (basaltic) composition,
Via. 2.— Dolerite Bills cutting basaltfl, lele of Eigg. (After A. Harker, Quart.
Jour. Geol. Soc., Vol. 62, 1906, p. 44.)
thoi^h some of these have minor non-basaltic phases and illustrate
the class of differenliaUd sills. (See Fig. 158 and pages 229 £f and 344.)
Fio. 3.— Section of area shown in Fig. 2, from 8, 10° W. to N. 10" E.
A multiple sill is a compound intrusion of sill form and relations,
aadis the result of successive injections of one kind of magma along a
66 IGNEOUS ROCKS AND THEIR ORIGIN
bedding plane. Marker has doKeribcd remarkable examples in the
^- iaiand of Skye; Fig. 4 is reproduced from his
\ \ 2 drawing.
i\ \ A composite Bill is a compound intrusioD of
II >^ Hill form and relations, and is the result of sue-
V- ccBsive injections of more than one kind of
'^ magma along a bedding plane (Pig. 5). Again
"s Skye furnishes good examples' (Figs. 6 and 7).
J Another is figured in a recent paper on the coaitt
^ geology ot Greenland' (Fig, 200, page 447).
s » Sills vary in thickness from sheets of micro-
-B ^ scopic dimensions to those more than 1000 feet
^ . in thickncst!. In all cases it is necessary that a
" S sill shall hold its major thickness, at least ap-
-a £ proximat«'ly, for long distances along its roof or
J Jt* floor; but it is obvious that there is no sharp line
K<^ between nills and laccoliths.
3 .3 ^i"^ '"Ay '>^ extremely abundant in a rangle
u J outcrop. The writer has seen more than one
^ a hundred sills in n cliff section, 2500 feet high, in
^ £ the pre-< 'ambrian sediments (Shuswap series)
^JS o^ ihf Columbia mountain range in British
c t Columbia. He has counted twenty-five sills in
I I a 30-foot cliff in the same tcrrane. The sills of
^ ^ the Purcell Range in the same Canadian province
J 5 arc ns notable for their number as for great indi-
^ 5 viilual tbi<'knesses.
S ^ The famou.s U'hin sill has a maximum known
= i thickne.is of but 150 feet (average thickness
1 _. SO 100 feet), but its mapped outcrop is more
5-5 than 80 miles in length (Fig. 29). The greatest
I ?i of the Trin.-wic sills in New Jersey reaches 1000
± " feet in thickness and ha.s about 100 miles of
■= residual length (Fig. 203). The Cape Province,
3- South Africa, is rich in enormously extended
-S intrusions of thi.s class. The lowest of the do-
is Icrit^" sills in Calvinia is continuous over at least
I~. .3000 square miles, and one near Hopetown
1. covers more than 5000 square miles. Another
_; ciolontc sheet Iwtwfcn I.angcljergen and Tanqua
■Z ViiUf y. thoiigli reaching u. maximum known tbick-
' A. Harker, Tertiary Uneoiw Hocks of Skye, 1904. pp. 204 and 267.
■ A. Helm, MeUdclcber om Gruolund, Vol. 47, 1911, p. 203.
INJECTED BODIES 67
ness of only 300 feet, has an outcrop more than 100 miles long. The
Rooi Hoogte sheet fronts the Great Karroo for nearly 50 miles and
has an outcrop more than 70 miles in length.'
FiQ. 5. — Simple and composite (Jf) sills cutting Tertiary strata at the Kettle
Riv(T, British Columbia. (R. A. Daly, Memoir 38, Geol. Surv. Canada, 1912,
Fi)t. 25.) I, rhomb-porphyry; S, pulaskite porphyry, cut by f; 3, Oligocene sand-
stone and shale.
Fio. 6. — Section of a composite sill in Skye. (After A. Harker, Tertiary Igne-
ous Rocka of Skye, 1904, p. 204.) The stratilied Lias was out by the sill of basalt;
this was cut by the later sill of granophyre. The basic aill may have been double.
Scale, 1:8,500.
Fio. 7. — Composite laccolithic sills of Skye. (After A. Harker, Tertiary
Itneous Rocks, Skye, 1904 p. 257.) 1, basalt lavas; t, dolerito sills; S, olivine
dolrrit*; 4, mugearite. Scale, 1 : 12,700.
Such statistics have considerable importance inasmuch as they
make it easier to believe that the colossal injections of gabbro in
' A. W. Rogers and A. L. du Toit, Geology of Cape Colony, Ixindon, 2nd
Edition, 1909, pp. 261-264; cf. same authors in Ann. Rep. Geol. Comm. Cape of
Good Hope, 1903, pp. 37-39.
68 IGNEOUS ROCKS AND THEIR ORIGIN
Minnesota (Duluth), in tlic Bustivcldt, Soutli Africa, etc., are lacco-
liths, orgreatly thirkoned 6h(>et!!, vrhth true floors. (See TaUe XIV,
page 230.)
Sills vary (treatly in clicniicat romposition, from peridotlte to highly
Fu:. S.--l^>.v.>liths of III!' .lo.litli M.miitain^. Montana. (After W. H. Weed
Hn.lL, Vrirwui. iMh Ann It.-,.. IS C. S . I't 3. IS9S. PlatpTS.) SUidUaek.
ai'iilir porphyni^ of th<' kr.-nUihs. i'-trii,-l r-ireif. ftranii« porphyry; S, Byenite;
r. tinitiMiti': til*. Ilfiiton :iiwl l):ik.>i^i ^li^iW :iTi.l rvind-otonm: K. KooUnie •hftla
nnil iinnilHtoniv: J, Jiir:i."ii- ^hjili':! :inil Imic.-ioin-s; I'B, Carbonitefous limeatone;
l>S, IVvimiiin iini! Siliiri;iii luiii--.tiiii.'-. (", l':ii[ilirKin sb.iW sml limtvtoDea.
$ili(-i«u^ apliti-s, hut tli.- crojitiT, iium' [i.TM^tvnt sills are in almont
all .■a.<(>s iiial..vi,- or K^iM-r-i.l l.;L-;iHi.-
CUvn-iy ri'httctl to sill* i* a taliular intr\i>iiiii which has been in-
jiM.-toil aloD^t H plaiu' ol um-onfonnity, in such (a.-'hion that th« nuM
INJECTED BODIES
69
lies parallel to the bedding-planes of one of the invaded formations.
Such bodies have been named (intrusive) interformational sheets. The
greatest recorded example with well-determined roof and floor is
that in the Sudbury District of Ontario.* (See Figs. 160, 161, page
347.) Like most of the great sills of the worid, the magma of this
Sudbury sheet has been differentiated and the process has here taken
place on a scale seldom equalled in demonstrably injected bodies.
Gilbert originally described the typical laccolith in the following
words:
"The station of the laccolite being decided, the first step in its formation
is the intrusion, along a parting of strata, of a thin sheet of lava, which spreads
until it has an area adequate, on the principle of the hydrostatic press, to the
deformation of the covering strata. The spreading sheet always extends
Gurnett
Triongulation
Peak
Alpine
Creek
Crystal Peak
Ridge
Alpine Creek
GP Judith Peak
0
Mis.
5 Km.
Fio. 9. — Sections of area shown in Fig. 8. Solid black, acidic porphyries of
the laccoliths; GP, granite porphyry; M, Mesozoic shales, limestones, etc.; C,
maanTe Carboniferous limestone; jD, Devonian limestone; CA, Cambrian shales
and thin-bedded limestones. '
itself in the direction of least resistance, and, if the resistances are equal on
all sides, takes a circular form. So soon as the lava can uparch the strata,
it does so, and the sheet becomes a laccolite. With the continued addition
of lava, the laccolite grows in height and width, until finally the supply of
material or the propelling force so far diminishes that the lava clogs by con-
gelation in its conduit and the inflow stops
"As a rule, laccolites are compact in form. The base, which in eleven
localities was seen in section, was found fiat, except where it copied the curva-
ture of some inferior arch. Wherever the ground plan could be observed,
it was found to be a short oval, the ratio of the two diameters not exceeding
that of three to two. Where the profile could be observed, it was usually
' A. E. Barlow, Ann. Report, Geol. Survey of Canada, Vol. 14, Part H, 1904;
A. P. Coleouui, Report of Bureau of Mines, Ontario, Vol. 14, Part 3, 1905.
70 IGNEOUS ROCKS AND THBIR ORIGIN
found to be a siin)Jc curve, convex upward, but in a few cases, and espenally
in that of the M&n-ine laccolitc, the upper surface undulates. The height
is never more than one-third of the width, but is frequently much less, and
the average ratio of all the measurements I am able to combine is one to
seven.
"The ground plan approximates a circle, and the type form is probably
a solid of revolution — such as the half of an oblate spheroid
"The laccolite ia a greatly thickened sheet (sill) and the sheet |sill| is a
broad, thin, attenuated laccolite
"The laccolite in its formation is constantly solving a problem of least
force' and its form is a result A laccolitc grows 'by lifting ita
cover.' "'
Thoee who have made actual researches among laccoliths, and
have preserved Gilbert's (l(>finitton, arc agreed on the following char-
acteristics: (a) What<!ver the origin of the force involved, a laccolith
^
i
^ - — ^ ^^==;>-- .
^^^P
l^^^^m^
S
■^r~^-j^»?:a^j^';;''"'^^£^
■ ^^^*^5^^f|^^^
Pio, 10. — Slcrpofcrsphic sketch of tlio Warm SprinR Iwdotith, Montana.
(W, 11. Wceil and L. V. rirsaon, l»th knn. R(^p., U. S. G. »., Pt. 3, ISQS, p. 519.]
P, porphyry of laccolith, dotted; K, Crrtareous Mtrata; J, Juraaiiic strata; it, road.
is always injected. (1>) A laccolith is always in sill-relation to the
invadetl. stratified formation; that is, the injection has, in the m^n.
followei) a bedding-plane; but. like sills, laccoliths often locally brvak
across the bedding, (c) A laccolith has the shape of a plano-convex
or doubly convex lens flattened in the i>lane of bedding of the invaded
formation. The leaf may be symmetric or asymmetric in profile.
circular, oval, or irr<>gular in gniuml jdan. (d) There are all traasi-
tions between sills and luccolithit.
Vcrj' large laccolit li.-t shnw a feature which is not characteristic of
thoee in the American West. The scdiment.s underlying the Bush-
veldt laccolith dip inward on all sid<-s (Fig. ]f>2, page 351). Both
I Ci. K. Gilbert, itcpnrt on the Gralogv- of I he Henry Mountains, 1877, pp. 30,
U. 91, 9S.
INJECTED BODIES 71
roof and floor of the iaccolithic sheet at Sudbury have centripetal dips.
Another example is found in the multiple laccolith of the Cuillin
Hills, Skye.' The sandstones and porphyry flows in the roof of the
Fio. 11. — Section of Kelly Hill laccolith, Montana. (Same ret. aa tor Fig. 10,
p. 499.) Solid blach, porphyry ot laccolith; upper broken littet, Meaozoic strata;
larger reetangki, Carboniteroua; smaUer rectanglet, Siluro-Devonian; loujer brofcen
lint*, Cambrian.
Ilimausak "bathollth," Greenland, similarly dip inward* (Figs.
197-9). The floors of the Duluth and Bad River laccoliths in the
LakeSuperiorregion are concave (Figs. 149, 151, and 152). These cases
Fto. 12. — Section ot Burnett Creek laccolith, Montana, showing erosion of roof
more advanced than that at Kelly Hill. (Same ret. as for Pig. 10, p. 490.) Same
Intend as for Fig. 11.
suggest that the large-scale transfer of magma into a laccolithic
L-hamber regularly causes subsidence of the solid rock beneath the
chamber. An intrusion so deformed is best r^arded as a true laccolith
Warm Spring Creek
Fio. 13.— Section of Warm Spring laccobth, Montana, showing erosion of root
and of laccolith more advanced than that at Burnett Creek. (Same rrf. aa for
Fig. 10, p. 519.) Solid Uack, porphyry of laccolith; Kd, Dakota aandatone; Kt,
Kootanie sandstone; J, JurasMc; Kde, Dakota and Colorado groups.
though in form it differs systematically from the Henry MountMna
type.
' .K. Barker, Tertiary Igneous Bocks of Skye, 1904, pp. 86 and 423.
' N. V. Uasing, Meddelelser om Grflnland, Vol. 38, 1911, p. 322.
73 IGlfBOUS ROCKS AND THBIR ORIGIN
lllustrfttions of aimpU laceUilht are given in Figs. 8, 9, 10, U, 12,
13, and 14. The Tenmile District of Colorado represent* UceoUtht
Fio. 14. — Laccolith of Bear Lodge Mta., Wyoming {inUrformfttionAl). (From
Suadance Folio, U. 8. G. S., No. 127, 1905.) S, sy«nit« porphyryi M, Lower
MMotoic; C, CarboniferoiM; Ca, Cambmoi j4|7, "AlgonkianT" granite. 8c*le,
1 : 105,000.
in special abundance and, as well, exaniplea of the tranntioD to alls.'
A mulftple laceolilk may be conceived, the name being f(Mmied on
Fio. l.j.— Section of composite kccolilh in Hkye, (After A. Harker, Twtiarr
Iltneoiu Rocka of Skyc, 1904, p. 209.) B, baaalt; C, ^anophyre. The grano-
phyre pot) euU the intruHivc basalt whirh is ii>ieU intrusive into baaaltic lava*.
The maKimum thicknens of the lacrolith in LW ft.
the analog)' of "multiple dike" and "multiple sill." It would differ
from a compound laccolith ualy in the fact that the deformation of the
Fio. 18. — Section of coniponitc larnilith at Blark Butlee, WyomtDK. (SMoe
ref. aa for Fin. H) ^. Tcrliftry phonolitc: .HP, Tfrtiarj- montonite pocphyry;
r, Triawic formation; /., CarlMinitprinw Minncktihta iimrotono; M, Carbooifflroua
Minneliua sandstone; I'l', C'arl>unifcriiii:i }'iihnMap» jiinnttone; C, Cambri&Q itnta;
S, pre-Cambrian schintn.
strata, while again amilor in character to that i>n>dueed during the in-
trusion of a simr' HlQplith, lias lieen due to distinctly successive
I Teiu"it- DWh^ ^V V- S. Ofl0> ^"^gf. No. 48, 1898.
INJECTED BODIES
iiqections of the same kind of magma. Harker describes the gabbro
laccolith of the Cuillin Hills, Island of Skye, as of this origin.*
Composite laccolitha differ from multiple laccoliths in the respect
FlO. 17.^-Section of interiormational laccolith, Little Rocky Mta., Montana.
(W. H. Weed aod L. V. Pirsson, Jour. Geol., Vol. 4, 1896, p. 412.) Solid black,
laccolith; roof, Paleoioic sediments; ^oor, pre-Cambrian schiata. Length of aec-
tioD about 10 milea (16 km.].
Fio. 18.' — Section of asymraetric interformational laccolith at Barker Mt.,
HoDtana. (W. H. Weed, 20th Ann. Rep. U. S. G. S., Ft. 3, 1900, p. 356.) Solid
iladi, porphyry of laccoUth; L, Carboniferous limestone; C, Cambrian shale;
P, pre-Cambrian. Scale, 1:97,500.
Fra. 19. — Section of asymmetric interformational Liccolith at Black Butte,
Montana. (Same ref. as for Fig. 10, p. 556.)
that the micces^ve intrusioas are here composed of two or more difiFer-
ent magmas. This type is illustrated in Figs. 15 and 16.
' A. Haiter, Tertiary Igneous Rocks of Skye, 1004, p. 8S.
IGNEOUS ROCKS AND THSIR ORIGIN
k.-^ In the 1905 paper s class of "compound lao-
colithH" was recognized and an example in the
Judith Mountains of Montana has been figured
and so named l>y Weed and Pirason.' In the
Judith Mountain type, as in the larger examples
of laccoliths in the Henry Mountains, the whole
intruxivc body is divided by strong beds of the
invaded formation. This gives the appearance
of a number of distinct intrusons, one of them
dominating, the others submdiary, in siie, but all
of them composed of the same kind of material.
If the magma had all been intruded at practi-
cally the name time, we have the "compound
laccolith" of Weed and Rrsson. Actual (Hwctice
has »hown that it is difficult for students to re-
member the distinctions between "compound,"
"multiple," and " compofflte" laccoliths. Rence,
it is here Kuggestod that the rare bocUes of the
Judith Mountain type be described as "divided"
simple lacTolitha.
Like int<'rformatiunal aiiecb*, a few bodies of
liU'culitliic form and origin have been ii^ected
along I'-lano!) of unconformity. For want of a
better name these may Ite called "infef/ormafuma/
laccolidi/'." Examples have l>een found by Weed
i? - 5 and I*ir.**on.* J. I). Ir\ing,* and by Dartoa*
J A.l (Figs. 17. 1ft. Ill, 201.
— ■,- ^ Vcrj- M-ld<mi are tlio fec<ling channds of
= f ~ liircoliths ex]Mi!ic-il. Students of these bodies
B'l I agree in ixn^Iulaling narrow channels, actually
-.7 %"% dikes or of dike-like pruixirtioas.
.- g ? Laccoliths vary greatly in dimensJons, the
2 "t S thicknesses rniigiiig from a fraction of an inch
•vend miles. Cilhcrt Idls estimated that 10
cubic miles is tin- approximate volume of the
Mt. Iltller.i intrusive, the Iarge:it of the classic
Ilenn.- Mountain laccoliths in Utah. Its depth is
. alH)iit 7(HI0 feet, and its diameters arc four miles
S S
ii!
:n
I W. 1). Wrttl and L. V- I'Jnwon, 18th Ann. R^.
V. S. (iwl. Sunrv. I'art :!. IS'Js, p. .'iSO.
• W. II. W»-«J and L. V. I'irwmn. Jimr. (:.-..1..e.v. Vol. 4, ls9«. p. 402.
' J. ». IniiiR. Annnls New V«rL .Vcml. S.-i<'n.-™, Vid- 12. ISM, p. 208.
■ N.II. Darton, Sundance Folio. Xo. 127, 1'.S.liooI. Survey, atructuren
INJECTED BODIES 75
aiKl three and three-quarter miles. ^ One of the largest bodies for
which a probable laccolithic origin has been indicated is the Duluth
gabbro of Minnesota. Van Hise and Leith are among those who favor
such a mode of intrusion for this body.^ At a minimum estimate it
has an outcrop covering 2400 square miles, with a length of 125
miles, and a maximum width of about 25 miles. (See Fig. 149, page
325.) If, as assumed by Van Hise and Leith, the av^age dip of the
intrusive body is ten degrees, its maximum thickness where exposed
is alwut 22,000 feet. These estimates do not include the area of the
"rod rock/' which is very closely associated with the gabbro, and may
possibly be regarded as an acidic, upper phase of the laccolith. Still
more imposing is the Bushveldt laccolith with its 250 miles of length
and 80 miles of width (Fig. 162, page 351).
These bodies also vary greatly in chemical composition. Cross
has recorded the most thorough canvass yet made of the laccoliths
of Colorado, Utah, and Arizona.' His list of the petrographic types
includes: augit^ porphyrite, hornblende porphyrite, porphyritic augite
(liorite, quartz porphyrite, and quartz porphyry. Some of these
species would now, probably, be called monzonite porphyries. In the
Highwood Mountains of Montana the rock types forming laccoliths
include sodalite syenite, shonkinite, basic syenite, and leucite-basalt
porphyry.* Granite porphyry, rhyolite porphyry, syenite porphyry,
and diorite porphyrite compose the many laccoliths of the Judith
Mountains, Montana.^
The types represented in the laccoliths of the Black Hills, South
Dakota, include grorudite, phonolite, rhyolite porphyry, dacite,
andesite porphyry, syenite porphyry, and diorite porphyrite.* Hills
describes laccoliths of doleritic rock, found in Huerfano Park,
Colorado.^ Gabbro laccoliths are reported from the island of Skye,
and they seem to have been developed on a large scale in the area
covered by the Roseburg folio of the United States Geological
Sur\'ey, as well as in Minnesota and other states of the union. At
least ten of the laccoliths underlying the thirty-two domes of the
Kunn of Cuteh are composed of quartz-bearing diabase.' An ultra-
> G. K. Gil5ert, Report on the Geology of the Henry Mountains, Washington,
1877, p. 30.
*C. R. Van Hise and C. K. Leith, Monograph 52, U. S. Geol. Survey, 1911,
p. 372.
* W. Cross, 14th Ann. Rep. U. S. Geol. Survey, Part 2, 18H p. 166.
«L. V. Pirsson, Bull. 237, U. S. Geol. Survey, 1905, p. 57 ff.
» W. H. Weed and L. V. Pirsson, 18th Ann. Rep., U. S. Geol. Survey, Part
3. 1898, p. 557 ff.
• T. A. Jaggar, Jr., 2l8t Ann. Report, U. S. Geol. Survey, Part 3, 1901, p. 182.
' R. C. Hills, Proc. Colorado Scientific Society, Vol. 3, Part 2, 1889, p. 226.
' J. F. Blake,|Quart. Jour. Geol. Soc., Vol. 54, 1898, p. 12.
L
76 IGNEOUS ROCKS AND THEIR ORIGIN
femic, wehrlitic, intrusive body at Kilauea, Hawaii, has arched the
overlying ash beds after the manner of a laccolith and has been so
classed.' True granite is the most conspicuous phase of the differen-
tiated laccolith in the Bushveldt, Transvaal. Theralite forms lacco-
liths in the Crazy Mountains of Montana.' Ijolite, urtite, and
nephelite syenite compose the supposed laccolith of the Kola peninsula.
Without further multiplying examples, it is clear that all, or nearly
all, the igneous clans are represented among the laccolithic injections.
For some years geologists were inclined to emphasize the view that
a chemical composition of intermediate (mediosilicic) character was
generally an essential feature of laccolithic rock. So many instances of
highly acid types as well as of strongly basic types have now been
recorded that it is no longer safe to say how general the rule may be.
At any rate, the relation of the peculiar mechanism of laccolithic
injection to the chemical composition of the intruding magma is now
too obscure to be of great use in petrogenesis.
Among the special students of laccoliths the hypothesis prevails
that great viscosity is an essential prere({uisiteinthis mechanism. If
it be true, this view has an important consequence. Some of the
Montana laccoliths (Square Butte, Shonkin Sag) clearly show evidences
of magmatic viscosity which after injection was still low enough to per-
mit of truly spectacular differentiation. The prevalent hypothesis, if
applicable to these cases, implies that thorough-going differentiation
can take place in highly viscous magma. It would seem, in fact, that
the thick Square Butte intrusive is a fairly typical laccolith, and that
this hypothesis should apply there, if anywhere. The Shonkin Sag
l)ody is more like a sill, suggesting a relatively low initial viscosity
for its magma. The whole problem is worthy of investigation becau.<e
of the light it may throw on the physical conditions for magmatic
differentiation, as well as on those for laccolithic injection.
Phacoliths. — Harker has introduced the name '*phacolite" (here
*'phacoIith,'' meaning literally *' lens-rock*') for a third class of con-
cordant injections. His description may be quoted.
"In the ideal case of a system of undulatory folds there is increased
pressure and compression in the middle limbs of the folds, but in the crcDt«
and trouglis a relief of pressure and a certain tendency to opening of the bed-
ilinR-surfaces. A concurrent influx of molten ma^ma will therefore find iw
way along the crests and troughs of the wavcMike folds. Intrusive bodies
corresponding more or less chiscly with this i<ieal ca>e are common in folde<l
districts. Since >c»nie distinrtivr name sterns t<» l>e ncHMled, we mav call
them phacolitefi. The n:unc hircnlito has often been cxtcmled to include
* R. .\. Duly, I*nw Aiiut. Arjul Arts ami Scifiires, Vol. 47, 1911, p. 115.
' Personal comnmnirntion from J. K. Wolff.
INJECTED BODIES
77
such bodies, but this is to confuse together two things radically different.
The intrusions now considered are not, like true laccolites, the cause of the
attendant folding, but rather a consequence of it. The situation, habit,
magnitude, and form of the phacolite are all determined by the circumstances
of the folding itself. In cross-section it has not the plano-convex shape of
Fig. 21. — Dolerite phacolith (D) cutting Ordovician strata, Comdon, Shrop-
shire. (After A. Harker, The Natural History of Igneous Rocks, 1909, p. 78.)
Bf Stapeley ash and andesite; A, Mytton flags and Hope shales.
the laccolite, but presents typically a meniscus, or sometimes a doubly con-
vex form. Except where the folding has the character of a dome, a phacolite
does not show the nearly circular ground-plan of a laccolite, but has a long
diameter in the direction of the axes of folding. As regards the mechanical
conditions of its injection, the phacolite resembles rather the small subsidiary
Fig. 22. — Differentiated dikes in the Trusenthal, Thuringia. (After H.
B&ddng, Jahrb. k. preuss. geol. Landesanst., 1887, Taf. V.) 5, granite porphyry;
*, syenite porphyry; /, melaphyre; (7, granite. Width of dikes somewhat exag-
seraied. Scale, 1:5,000.
intrusions which sometimes accompany a laccolite, and are consequences of
the sharp flexure caused by the primary intrusion.
"The ideal type of phacolite is subject to many modifications, in accord-
tnee with the varying mechanical conditions of intrusion. Some bodies of
tUs nature, in the Alps and elsewhere, attain large dimensions. According
78 IGNEOUS ROCKS AND THEIR ORIGIN
to Battier, the Alctach mass is 18 or 19 miles long and 2 milH t»o*d, with t
visible thickness of 2600 to 3200 feet, while the St. Gotthard man has a
length of 45 mile^t and a breadth of 2 or 3 miles."'
Fig. 21 is copicil from Harkcr's illustration of this intniUTe type.
According to what scetnH to be the commonest usage, a nmpfe dike
(a) is an injected body, (b) has nearly or quite parallel walls, (c) ia
Fio. 23. — Section of differentiated dike in the EUnenthal, Thuriagia. (8*iiM leg-
end and ref. as for Fig. 22.)
narrow in proportion to its outcropping edge, (d) cuts across tbe
bedding when the invaded formation is BtratJ6ed, and (e) has any
angle of dip.
^^'hpn stratification and cleavage or schistosity are not coincident
such an intrusive Ixxly i^ generally called a <like, even though it fol-
lows the planes of cleavage ur »chistosity. This u.iage is adopted in
the present classification.
^ 1
L-] a
^■"^^X. 0
•«.-
P^^^
^^'^^ii^ii^^^bfc,
Fio, 24. — I'Inii of IIip jtrciit
K. Wingi*. CitH.!, Farm. FOrh..
tnternicdi.ilp rork: S, olivini" din
lifTiTontiatwl diki- M Brctven, Sweden. (Aft«
ol. IS, ISDO, p. 187 ) (, BTanite' porphyiy; 1,
Fume simple dikes exhibit s<'gri>gation of magmatic elements
after injection and may be describwi as differmtiaied dikes (Figa. 72,
23. and 24).
MuUi'iilt dikes are intrusi<ia-< of dike form, due to successive in-
jections of one kind of niagnm into the s^inie fissure (Figs. 25 and 26).
■ A. Baikcr, The Natural llialorj- of Iftneous Hooka, New York, 1901^ pp. 77-7S.
INJECTED BODIES
79
The instances described in geological literature seem to be much more
numerous than those of either multiple sills or multiple laccoliths.^
Composite dikes are intrusions of dike form, due to successive in-
jections of chemically different magmas into the same fissure. This
nomenclature brings out the analogy with ''multiple" and " composite"
sills and laccoliths — types already well named and established. Dikes
of this class are illustrated in Figs. 27 and 28.
Marker has published a useful set of diagrams illustrating certain
irregularities in the forms of dikes.^
FAULT
FAULT
Fig. 25. — Multiple dike following fault plane in Cowal. (After W. Gunn, C. T.
Clough, and J. B. HiU, Geology of Cowal, 1897, p. 144.)
No lower limit can be safely assigned to the possible width of a
dike. For dikes of small length the limit is certainly less than 1 milli-
meter. Emerson describes a glassy (diabase) dike in Pelham, Massa-
chusetts, only .9 millimeter wide, with apophyses respectively .5
millimeter and .02 millimeter wide. A second tachylitic dike, also
cutting gneiss in the same region, is 2 millimeters wide, with apophyses
about .1 millimeter wide.' No sharp line can be drawn between
true dikes and mineral veins deposited by water or by other volatile
'■'•itiii'i a%iiriiiiiiiit<'Miiii
Fio. 26. — Multiple (triple) basaltic dike cutting granophjrre, St. Kilda Island.
(After A. Geikie, Ancient Volcanoes of Great Britain, 1897, Vol. 2, p. 417.) 1, B,
and 5, separate intrusions.
fluids, which are capable of searching out the minutest crevices in
rock. Yet multitudes of dikes of " dry," specially basic magma have
outcrop lengths of hundreds of feet, with widths of much less than
1 foot. Such dimensions imply both low magmatic viscosity and great
* For good examples see: T. Thorrodsen, Petermann's Geog. Mitt., Erg. Heft,
Vol. 32, 1905, p. 249; A. Geikie, Text-book of Geology, 4th ed., Vol. 2, 1903, p.
746.
•Tlie Natural History of Igneous Rocks, New York, 1909, p. 74.
* B. K. Emerson, Monograph 29, U. S. Geol. Survey, 1898, p. 416.
80
WNB0V8 ROCKS AND THBIR ORIOIN
rapidity of injpctiun. In many casen the injection aecenarily Beuns to
have progressed with almost explosive violeace. Barrel) has ably
discuflsetl this matter, which is evidently of importance to the theory
of intrusive mechaiusm in general.'
- - -1-;
Fio. 27. — Conipositc dike, Ilroailfonl, Skyc. (After A. Geikie, nuna n£. -.
for Fig. 26, p. 162.) 6', Rrsnophyre; if, biuialt, cut by 6'; S, TorridoaiAB ■
It is instructive to note the great lengths of outcrop determined
by some of the largest known dikes. Two diabase dikes in northern
Marj'Iand are 8hown on the State map to have respective lengths oS
38 mile.-! and 4.5 miles.' HogerH and du Toit describe a dolerite dike
I!
— 8
s
FlU. 2S.~runi|H>Kil(> iliki'
Ueol. Soc.. Vul. 19, l(»a;i. |.
fclditel, cut by /■; A, awRitc a
, An-nn. (Aftrr J. W. Judd. Quait. Jour.
. pitrhKlonc (ducitc); D, dadte (quarts
t by D; S. tiandBtoiie. Scale, 1:300.
in Mataticlr. (npc Provnncf, which is 15 miles long and up to 1 mile in
width: another, runninft through Beukos Funtein, is 13 miles long and
• J. narrell. Frof. I'sprr, No. 57. t'. S. Ceol, Surrey, 1907, pp. U7-UB.
< Maryland Cral. Sun-e)-. Vol. tt, 1906.
INJECTED BODIES
81
about 100 feet wide.^ A third dike, over 45 miles long, runs be-
tween Mt. Fletcher and Mt. Frere in the Drakensberg region.^
A. Geikie states that the Cleveland dike of northern England is
at least 110 miles long and may be as much as 190 miles. Scottish
dikes, respectively 25, 30, 36, 47, 50, 58, and 60 miles in length, have
been recorded.' (See Fig. 29.) The well known Brefven dike of
Sweden is over 20 miles in length and reaches a width of more than a
mile* (Fig. 24). Thoroddsen states that some of the basaltic dikes of
Iceland may reach lengths of from 30 to 65 miles. Many of them
are more than 10 miles long.*
Fio. 29. — Map showing position of long dikes (D) and of the Whin sill. (After
Government geol. map of Scotland, pub. by Stanford.)
One of the most remarkable assemblages of dikes yet mapped is
that in the Spanish Peaks district of Colorado (Fig. 30; see also Figs.
70 and 165).
These examples show that the lengths of dike outcrops are of the
same order of magnitude as the lengths of the greatest batholiths yet
mapped. On the other hand, the width of dikes, namely, tabular,
cross-cutting injections with the characteristic chilled contacts, is
seldom as much as a mile. It is questionable that the width of any
mapped dike is as much as 3 miles. The average width of mapped
' A. W. Rogers and A. L. du Toit, Geology of Cape Colony, 2nd ed., London,
1909, pp. 231 and 260.
* A. L. du Toit, 15th Ann. Rep. Geol. Comm. Cape of Good Hope, 1910, p. 99.
' A. Geikie, Ancient Volcanoes of Great Britain, London, Vol. 2, 1897, p. 143.
* P. J. Holm uist, Bull. Geol. Inst. Upsala, Vol. 7, 1906, p. 107.
* T. Thofodi nn's Mitt., Erg. Heft No. 152, 1905, pp. 250-251.
83
IGNEOUS ROCKS AND THBIR ORIGIN
dikes is probably well under 100 feet, if, indeed, it is not len than 40
feet.
A dike system 13 a local group of roughly parallel dikea iqjected in
the same intrusive epoch. As a rule these follow master jtunta in the
country rock. Such sj-stcms can sometimes be seen, after suffident
eronon, to have been the feeders of major fissure eniptJons. (See
Figs. 31 and 70, page 121.)
It is not neccssar>- to repeat the statement included in Chapter II,
Fio. 30. — Map of ilike xyKtem, ftpopbynal from itocks, Colonula (SfMoUi
Peaks Folio. No. 71, U. S. G. S., 1901.) OP, (tranitc porphyrj-; AP, aiigit«-pw>-
ite porphyry; D, sugite diorite.
as to the chemical range maiufestcd in the dike rocks. All the p«at
clans arc represented in the a.schistic cla-sti, which are often simply
physical satellites of bathoIithH or stocks. The diaschistic dikes are
clearly modified offshoots uf the magma in larger magma chambers
and may be deflcritie<l as chemical satellites of these.
Intrufivt vt-inn were lung ago <Iefined by Jukes in the fdlowng
words: "When the injected mass ha.s arisen along an o 3eni
and solidified there as a walMikc intrusion, it is called a A
0 3enadJM|kg|
INJECTED BODIES 83
its path has been leas regularly defined, and penetrates the surrounding
rocks in a wavy thread-like fashion, this irregular protrusion is called
a vein."^
As suggested to the writer by Mr. R. W. Brock, Director of the
Geological Survey of Canada, there would be distinct advantage if
the term "vein" were restricted as much as possible to the tabular
bodies formed by deposition from solutions with a high proportion of
volatile matter. Such definition seems advisable even in spite of the
difficulty of distii^juishing veins from some kinds of dikes.
According to A. Geikie's definition, a contemporaneous vein "forms
part of the igneous rock in which it occurs, but beloi^s to a later
Fio. 31. — Dike system composed of thyolite {iolid black) in Corsica. (After
J. Deprat, Bull. serv. carte ggol. France, t. 17, 1907, PI. I.)
period of consolidation than the portion into which it has been
injected."'
Apophyses or tongxies are dikes or veins which, either directly or
by inference from field relations, can be traced to larger intrusive
bodies as the source of magmatic supply of dike or vein.
Volcanic Neck. — The solid lava io a volcanic vent must be con-
adered as intrusive into the wall rock, where this is non-volcanic, or is
I composed of older iava or pyroclastic material like tuff or agglomerate.
I Ducuamon and illustration of necks will, however, be postponed to
I Chapter VII, which deals with the forms of volcanic bodies.
a Jlj^^^s, Manual of Geology, edited by A. Geikie, 1872, p. 263.
W "^^^^^ "> >ook of Geology, 4th ed., London, 1003, p. 738.
84 IGNEOUS ROCKS AND THEIR ORIGIN
Bysnudiths. — Iddings has descril>ed a 'Miysmalith*' as an injci*to<l
IxKly filling a ''more or less circular cone or cylinder of strata, having
the form of a plug, whi(*h might l)e driven out at the surface of the
earthy or might terminate in a dome of strata resembling the dome over a
laccolith." The downward termination of the original type bysmalith
(Mt. Holmes) is found in a hypothetical Archean floor on which the
porphyry of the bysmalith rests. This body is sectioned in Plate 5
of Monograph 32, Part 2, of the U. S. Geol. Survey (1899). No other
bysmaliths appear to have been described under that name.
Chonoliths* — There remains for distinction a class of injecte<l
igneous bodies which arc not included in any of the above-mentioned
categories. In the dislocation of rock formations such as is brought
about during mountain-building, actual or potential cavities are formed
within the earth's crust. These are occasionally filled with igneous
magma squeezed into the individual cavity from below, from the side,
or, it may l)e, from alx>ve. Dikes, sills, and bodies of laccolithic
form (though not strictly of the laccolithic mode of intrusion, as
designated by Gilbert) may thus originate. Yet very often the shape
of the intruded mass is so irregular, and its relations to the invade<l
formations so complicated, that the body cannot be classified in any of
the divisions so far named. Again, irregular injected bodies of a
similarly indefinite variety of form are due to the active crowding-
aside and mashing of the country rock which is forced asunder by the
magma under pressure. Or, thirdly, such bodies may be due to a
combination of the two primary causes — orogenic stress opening
cavities, and hydrostatic or other pressure emanating from the magma
itself and widening the caviti(»s.
The numl)er and total volume of these irregular intrusions possi-
bly rival the numl)er and volume of all the true laccoliths of the
world. In the average mountain range the geologist is more likely
to encounter injected bodies of the former kind than he is to discover
true laccoliths.
No generally accepted name has yet lK»en proposed for such irregular
intrusions. '* Laccolith*' cannot l>e used, since that term denotes a
definite form, and also implies a sfx^cial mode of intrusion different
from that here conceived. The writer has not l)een able to find a
simple English word for the purpose, and suggests a name formed
from the Greek on the analogy of ** laccolith," ** bysmalith," and
** batholith," It is **chonoIith," deriv(»(l from x*^*'^! a mold used in
the casting of metal, and Xtl^, a stone. The magma of a '*chonolith"
fills its chamber after the manner of a metal casting filling the mold.
Like a casting, the *'chonolith'' may have any shape.
The writer is not entirely satisfied with this invention. El^
INJECTED BODIES
85
mologically it errs in being too broad, since laccoliths, sills, and dikes
are bodies molded gainst their wall-rocks. This objection is, how-
ever, more or Jess formal and is not so important as the objection that
the chonoliths as defined include masses formed under two highly
FiQ. 32. — Section of chonolith (P), near Glen Coe, Scotland. (After drawing
by Clough, Maufe, and Bailey, Quart. Jour. Geol. Soc, Vol. 65, 1909, p. ft59.)
P, intruaive porphyrite; L, lavae with conglomerate at base; S, schiete; felute
dikesfaown; ABCD, hypothetical top ot porphyrite intrumon. Scale, 1:19,000.
contrasted conditions. In the one case the magma is active, as in a
laccdith; in the other it is largely or wholly passive during intrusion.
However, the general impossibility of distinguishing the two cases
in nature renders a "Sackname" useful to the field geologist. It is,
Fio. 33. — Section (hypothetical) oS chonoLth at Monioni, Tyrol. (After Mrs.
Ogihrie Gordon, Trans. Edin'. Geol. Soc., Vol. 8, 1903, pp. 141, 175.) Solid black,
Uoumii intrusion; QP, quarts porphyry; S, stratified rocks. Scale nearly
1:«,000.
of course, not intended that the use of this term shall discourage the
further invention of good descriptive names for injected bodies, hke
" 'laccolith," "sill," etc. So far as such new classes become
d ai 1 named, the range of the chonolithic class, as covering
86 IGNEOUS ROCKS AND THEIR ORIGIN
ftll the irregularly shaped injected botiien, will be restricted. If tiie day
ever comes when the essential mechanism of each injecUon becomes
understood, the chonolithic bodies will merit more significant name*
in systematic clas-sification. Meanwhile, in Hpite of its shortconuDgs,
the blanket name, "chonolith," can serve a useful purpose. As
above noted, fur cxamplu, it can l>e employed in many cases where
bodies have been descrilwd bh "laccohths," though these masses have
Fici. 34. — Pliin or rliiiiiolillii) of <|iiartz Ulite potphyry (QL) and kikImIa
{A)e uttinR qimrfi iniinKmitR iiii'i i|iiiirtz nionzonitc porphyry (Q.V), post-
Ckrhnnifrroiia icraiiito {(ITi. Va,\nniM pcilimcntH (/'), and pre-Cambrian gnuiitc
(C); MotiarrU and Tomi'-hi .iistri<-t:i. C'ohirjido. (After R. D. Crawford, BuIL
4, Color&do StiilP Ural. Surv., ViVA, f'kte 2.) Tlin chonolilhi &re profalMy ol
Tertiftr; age.
neither the forms nordemun.-'lr.tlily Die mode of intrusion of true lacct^
litha. Such overloading of (iilhcrt's term must tend to injure it
for scientific puriHJses. In a nepiitive way the sumewhat negatively
ilefine<l word " chonDlilh " has dirftinct value: in this book it will have the
positive value of makiii);! pos.-iilile a brief formuf reference to some of the
world's injected masses.
A "chonolith" may be thus defined: an igneous body (a) injected
INJECTED BODIES
87
into dislocated rock of any kind, stratified or not; (b) of shape and re-
lations irregular in the sense that they are not those of a true dike,
vein, sheet, laccolith, bysmalith, or neck; and (c) composed of magma
passively squeezed into a subterranean orogenic chamber or actively
forcing apart the country rocks.
The chamber of a chonolith may be enlarged to a subordinate
degree by contact fusion on the walls, or by magmatic "stoping."
The writer^s preliminary paper (1905) on this subject contains
reference to manj*^ bodies which seem to belong to the chonolithic
class. The cases there cited were discussed simply from the maps,
sections, and reports of government geologists, working in Montana,
Washington State, Colorado, and South Dakota. Other instances
♦ + ♦-•• + -^^^\' • -^ • ' , ,
^S^ • • •
♦ ♦ + -•• + + "*'^v • * '
♦ •••-••♦■»• + -»■■»• +S. • * '
-•• -•• -I- -1- •♦- +v
+ -I- -•• ■!• + -••
-••♦ + + -•• +
v^ 0
QM \ .*
QM
Co
i Mis. ^.
J>
Km.
Fig. 35. — Sections along the lines A-B and C-D in Fig. 34. Underground con-
tacts partly determined by mining.
are illustrated in Figs. 32 and 33. Still others have been described as
occurring in British Columbia, Pennsylvania,^ Colorado^ (Figs. 34
and 35), and New South Wales.*
Etfamolith. — Salomon has interpreted the tonalite of the Adamello
group as an injected, partially cross-cutting body. The described
structural relations and mode of intrusion are those of the chonoliths,
yet the form of the whole body as deduced from its outcrop is, in Salo-
mon's opinion, definite enough to warrant a distinct name. He has
accordingly called this body an "ethmolith" (literally funnel-rock).
» F. Bascom, Bull. 360, U. S. Geol. Survey, 1909, p. 663.
» R. D. Crawford, Bull. 4, Colorado Geol. Survey, 1913, p. 107.
* E. C. Andrews, Records Geol. Survey of New South Wales, Vol. 8, Pt. 3,
1907, (reprint) p. 13.
88 IGN80VS ttOCKS AND THBIR ORIGIN
It is dflfiaed as a plutonic mass which narrows downwardly and is so
situated that the younger beds of the (sedimentary) oountiy-roek are
bent down into contact with the igneous body.' Fig. 36 will convey
this author's meaning better than a lengthened tact description.
No other masses seem to have been described under this name.
Fia. 36 — Diai^&inmatic section illuttrAtiiiR an etbmoUtb (E), with il
aunccstiona as to the nature of it* fcolinK channel (Mte Adamello, aftfr W.
Salomon, SiUunpbcr. k. prcum, Akad. Wim., Vol. 14, 1903, p. 310.) StratificktMMt
of invaded aedimenta shoira by lines.
Spbenolith. — Thi.s term wa.t invented by Burclchardt to disUn-
gui»h the special form and relation.*) of the dacitic intruaon at Las
Parroquias, Mexico.* This l>ody is clearly of the injected olasa. It is
partly concordant, like a thick sill, and partly discordant. The coun-
try rocks have been displaced even to overturning and some of the
muvemcnt is to be cre<lited to pressure from the magma itself.
> \V. Salomon, SitiungHber. k. preuu. Akad. Win., phya-matb. CtsMa, Vol.
14, 19(>3. p. 310.
> C. Burckbardt, Guide. Cong. Gdol. Internat. Mexico, 1900, Part M, p. 33.
CHAPTER VI
SUBJACENT BODIES
Definitions. — Incomparable as they are in individual volume, the
subjacent masses are the least understood of all the intrusive bodies.
If they were truly understood there would be no ''problem*' of the
igneous rocks. The failure of accurate knowledge is, of course,
natural. Direct observation on a batholith and direct inference as to
magmatic processes within its chamber are alike dependent on un-
roofing. In many cases erosion has bitten thousands of feet into
batholithic rock but seldom, if ever, tens of thousands of feet. In
most cases each plutonic mass is fairly homogeneous throughout its
known depth, which may be more than 6500 feet.^ There is, thus,
no indication that, if in any instance erosion could have penetrated
to twice the depth actually attained, the batholithic mass would change
its lithological character in marked degree. In brief, the geologist
has access only to the upper part of batholith or stock. For several
reasons which will gradually appear in the sequel, that fact is of pri-
mar\' importance. Properly appreciated, it will greatly aid in reach-
ing a sound explanation of the physical and chemical changes which
have occurred in the most voluminous masses. This book is, per-
force, chiefly occupied with intrusives of the batholithic class. The
present chapter attempts to state and illustrate merely the leading
structural and related features as set forth by field geologists.
** Batholtthy^^ the designation for the larger subjacent bodies, was
introduced by the elder Suess at a time when his own conception of
their origin was undergoing noteworthy changes. His final definition
of the term is here given in free translation: "A batholith is a stock-
shaped or shield-shaped mass intruded as the result of the fusion of
older formations (orig. Durchschmelzungsmasse). On the removal of
it^ rock-cover, and on continued denudation, the mass either holds
its diameter or grows broader to unknown depths (orig. bis in die
iv^ige Teufe).^^^ This definition has a strongly subjective element,
a^ it is based on a theory of intrusion which is still in active discussion.
^ Cf. R. A. Daly, Bull. Geol. See. America, Vol. 17, 1906, p. 372.
' E. Suess, Sitzung^berichtc der K. K. Akad. der Wissensohaften, Vol. 104,
1S95, p. 52.
89
90 IGNEOUS ROCKS AND THEIR ORIGIN
Many authors, without implying adhesion to that or any othet theory,
have since used the name to denote the great bodies othonMriae referred
to as "central granites," ** intrusive mountiun-cores/' " Fusi^granit/'
etc. For present purposes it seems best to adopt the same negative
position and emphasize in the definition only those features which are
nearly or quite independent of petrogenic theory. On account of the
supreme signifirance of these facts concerning subjacent bodies, they
will be illustrated in some detail.
At the outset, it should be observed that all the essential character-
istics of batholiths, except size, are also represented in those bodies
which have long been called *^ stocks^' or *' bosses.'' The terms are
often used synonymously. According to its general meaning, "boss"
should, apparently, refer only to such stocks as are of nearly circular
ground-plan at the surface of exposure. A boss is, thus, a variety of
the chiss of stocks. Stocks themselves are simply small batholiths.
The limit between these two classes cannot be other than arbitrar>\
In the 1905 paper the writer proposed that the upper limit of mse for
stocks be placed at 200 square kilometers. This figure was chosen so
as to include bodies approaching those dimensions and named as
stocks by certain authors. The obscr\'ation of actual usage among
writers during the last seven years has suggested that the limit set in
the preliminary i)aper is too high. It is more in accord with general
usage to confine the term ** batholith " to those subjacent masses which,
at the outcrop, cover more than 100 s(]uare kilometers or about 40
square miles; a stock is of smaller outcrop area.
Since batholiths, stocks, and bosses are similar in field relations,
the description of those relatioas will be chiefly phrased in terms of the
batholiths alone, thus avoiding useless repetition. A batholith has
the following characteristics:
1. Ix>cati()n in orogenic belts.
2. Cicnerally, elongation parallel to the tectonic axis of the moun-
tain range.
3. A date of intru.sion which follows, more or less closely, an ante-
cedent period of mountain-building.
4. Cross-cutting relations.
5. An irregularly domical roof,
t). Steeply inclined walls.
7. H«»lative sm<)othn«»ss of walls.
8. Downward enlargemtMit; no floor vi>ii)lt».
9. The appearand* of having rephKM'd t ho invaded formation during
its intrusion.
10. Composition usually granitir.
SUBJACENT BODIES
91
General Characteristics
Location in Zones of Mountain-building. — Without known ex-
ception, the post-Cambrian subjacent bodies are all located in orogenic
belts. This is not due merely to the greater depth of erosion and con-
sequently more perfect exhibition of deep-seated formations in the
Fig. 37. — Map showing distribution of batholiths (solid black) in the North
American Cordilleran (C), Appalachian (A), and Antillcan (An) mountain systems.
LargeHBcale overthrusting is demonstrated for zones marked with hachures and the
s}TnboI X.
mountainous regions. The denudation of some high areas character-
ized by the plateau type of structure has been very profound, but in
no one of them has a post-Cambrian batholith cutting the flat-lying
i^edimentaries yet been discovered.
In a given mountain chain, the abundance and observed sizes of
8
92
IGNEOUS ROCKS AND THBIR ORIGIN
batholiths are in direct proportion to the intensity of the orogenic
crumpling. These rules are illustrated in the North America Cordil-
lera (Fig. 37) . The largest and most numerous subjacent bodies occur
in the western half of this belt, from Southern California to Bering
Sea, where, on the average, the deformation of the invaded formations
is much more advanced than in the eastern half of the huge belt.
Similarly, the compressed folds of New England are penetrated by
numerous post-Cambrian batholiths and stocks, while no subjacent
Fio. 38. — Map showing position of the in'cat Patagonian batholith ($aUd
black), (After P. D. Quensel, Bull. Gcol. Inst. Univ. Upeala, Vol. 11, 1011.)
Mapping approximate.
body is known to cut the (K]ually ancient Paleozoic strata in the open
folds of Pennsylvania.
The same rules apply also to the pre-Cambrian batholiths. The
North American Laurentian granites and orthogneisses always cut
sediments or other supra-orustal rocks whose degree of defommtion
constitutes a now famous major difficulty in structural and historical
geoIog>'. The younger Huronian sediments, less deformed, are also
less affected by batholithic intrusioas. In Fennoscandia the crumpled
pre-Kalevian sediments and volcanics are cut by the largest bi^o-
SUBJACENT BODIES
94 rONEOrS ROCKS AND THEIR ORIGIN
liths of that roKion; while the Kalevian, Jatulian, and Jotnian sedi-
ments, Huecessively younj^er and less deformed, are successively less
inteiTUi)te(l by subjacent ma**ses.
For the same mountain belt one may often observe that batho-
lithic intrusion has varied in amount according to the intensity of
orojijenic crumpling; at different f^eological epochs. For example, no
intrusion of this kind has yet been connected with the moderate post-
Pennsylvanian deformation of the coastal half of the North American
Cordillera; while such intrusion, on an unrivalled scale, followed the
late Jurassic orogenic revolution in this same belt.
On the other hand, zones of intense crustal deformation by no
means always include visible batholiths. Granitic rocks are relatively
subordinate in the outcrops of the European Alps, the Carpathians,
the Caucasus, the Himalayas, and the New Zealand Alps, as in the
Allegheny zone of close folding and overthrusting. In part, the ex-
planation of this fact may be found in the local failure of sufficient un-
roofing by erosion, but it is cjuite possible that, in some of these cases,
large-scale intrusion has never affected the actual mountain ranges.
Perhaps the absence or subordinate development of batholiths in the
Al|)«, the Carpathians, the (*anadian Kockies. etc., may be connected
with the fact that each of these ranges exhibits .strong overthrusting
as an essential structural feature. (See Chapter IX, page 190.)
Since mountain-systems an», in general, due to the crumpling of
geosynclinal sediments, it follows that batholithic intrusions are
usually situated in belts where such sedimentary prisms have been
develope<i. The th«M»retical significance of this fact will be suggested
in Chapter IX. It should l)e noted, however, that the rule does not
seem to api)ly to most of the older pre-Cambrian batholiths, doubt-
less the most ext<'nsive of all in the earth. The roof rocks of the
oldest batholiths (»f Finland, like those of the Laurentian batholiths
in Canada, now include few sediments, and it is very unlikely that,
in either case, the failure of s^^ctions showing truly geosynclinal thick-
nesses is to be explaine<l liy batholithic replacement or by deep ero-
sion or by both processes together.
In other words, the con<litions lea<ling to the intru^on of these
huge masses into the upper part of the earth's crust have changed
notably between early pn»-Cambrian time and post-Cambrian time;
and one of the new. imlispensable conditions seems to have been the
anteccilent dev«»lopnicnt (>f thi<'k g<*osynclinal prisms.
Elongation Parallel to Tectonic Axes. -Where erosion has stripped
off much of the cover, the longer axis of a visible batholithic mass is,
ver}' generally, nearly or quite parallel to the tect<mic axis of the
mountain-built zone in which the nK<iss is situated. This typical
SUBJACENT BODIES
95
relation is, of course, likely to be more or less concealed where the
removal of the cover has only begun; in such cases the exposure of the
batholith is due to the accidents of denudation, and the shape of the
intrusive has no necessary relation to the ground-plan of its outcrop.
Examples of parallelism between batholithic axes and tectonic
FiQ. 40. — Map showing elongation of an Irish batholith (G) paT&llel to orogenic
uea dittgr&mmaticaJly shown in the invaded Ordovician (0), and Cambrian (Co)
itnta; Cb, Carboniferous, and D, Devonian.
axes in the country rock are to be found in great abundance. It will
suffice to illustrate the rule by reference to standard cases in the North
American Cordillera (Fig. 37) ; the South American Cordillera (Fig.
381; the Hercynian Mountains of Brittany {Fig. 39), Ireland (Fig. 40),
tad England (Fig. 41); the Pyrenees (Fig. 42); the Ural Mountains;
06 IGNBOUS SOCKS AND THBIR ORIGIN
the Himalayas; the Atlaa Mountaina; the mountaina of New South
W&lfts and of New Zealand.
Those examples arr all taken from areas of poBtrCambrian intrunoa.
The pre-C'atnbrian Imtholiths have very often developed "periphenU
schistosity" in their respective country rock terranes. Thereby it
may become difficult to determine the relation of a batholithie szia
to the average strike of the invaded formations. Exceptionally, these
pre-Cambrian intrusive masses are arranged parallel or en axe, so aa to
BUggeat that, in these cases, they were intruded under coaditioBS much
Ftn. 41. — Map showinK Btifcnnimt of Kranite batholitha And stocks (G) in Cora-
wall anil Devon, to oroRcnir axu which are shown diagrammaticallj'. T, TriMne;
r, Corlxiniferoua; D, Devonian, Ordovician, and Cambrian (f); 5, Ordovkisji
(Lower Silurian), B, Brrjientine, otr. Map Kt-nrrahsed.
like those of post-Cambrian batholiths. Such relations are illiutrated
in Ontario and in Sweden, whore the invaded sediments are locally of
verj- Kreat thickness.
Time Relation to Mountain Building. — Without known exception,
each batholithie invasion ha'' followed more or less closely a period (rf
stroni{ cru.-4tal deformation afTectinf; the older formaUona of the
»ame rettion. Tliis rule, which seems really to have attuned the
diFEnity of a law, is generally rccoRnized by geologists, but do one
ha-i hitherto published » statement showing how full is the evidence.
Space cannot be taken for details. The following table (VI) wilt
SUBJACENT BODIES
98
IGNEOUS ROCKS AND THEIR ORIGIN
TABLE VI
Onigrnir periwl
Epi-Keewatin.
Epi-Lower HuronUn
Epl>Upper Huronian
Epi-GreDville
Pre-CambrUn
Epi-"AlconkUn"
Epi-'Teraandan"
Epi-"AlcookUn" .
Epi-Shutwap terranr . .
Epi-Viabnu
"Loww pre-Cambrian " (a)
•'Lowrr pre-Cambrian" (b).
E|^-JatuUaD
Epi-Bottntan .
Epi-Kalevian
Epi-Jatulian. .
Pre-Cambrian
Prv-Cambrian.
Pmt-Malinrabury
Early pre-Cambrian
Epi-Ordovirian (?)
Caledonian
Caledonian
Epi-Devonian
Poat-Niacara . .
Hrrrynian.
Ilrrrynian
Herrynian . .
Carbomferiiun
Carboniferijun
Epi-Carbouifrrnu.« . . . .
Epi-Kanirh iMenotdir^)
Epi-TnaMic
Late Juraswir
Late Juraiwic
Late Jura*Mr .
Late Juraniiic
Late JuraaMc
Epi-Cretaceou^
Epi-Cretaceou!«
Epi-OnomaniAn
Tertiary .
Mid-Mii>r«'nf
Mi«xM«n«»
Correspondinc batholithic
period
lUgioo
Laurentian
Middle Huronian
Keweenawan .
Laurentian
Lat«r pre-Cambrian . .
Later pre-Cambrian . . .
i Lat«r pre-Cambrian . .
■ loiter pr»-Canibrian . .
loiter pre-Cambrian . .
I loiter pre-Carobrian .
Uter"Arrhean"
Serarcheao
Poat-JatuUan interval.
Poet-Bottnian interval.
Poat>Kalevian interval
Jotnian
loiter pre-Cambrian .
loiter pre-Cambrian . .
Later poet-Malmeebury.
Later pre-Cambrian . .
j Taconic (?)
, Devonian (?)
Devonian (?) ....
I Ijite Devonian or Early Car-
bonif«*rou>i.
I..ate Silurian or Devonian .
Early C'arboniferoun . .
Carboniferous
I Carboniforoun (or Permian?)
Permo-<"'arlM>niferou«
Permo-Carbonifer»>u!i
Permian. . .
PfMt-Kanirri (Tertiary?)
I^ate Juraaiiir. .
ClfMie of Jurawir
Close f>f JurAMir
Late JuraiKUr or Early Cre-
tarcf»U!i.
Clo*» of Jura«(<ur
CliMW of JuriuMic
Early TiTtiary . . .
Early Tertiary
Early Ti-rtiary or Lati* C'rfta-
ct«ou*.
I*at*»r Tertiank'
l.atf* Mior<»n**
IMi«ir»'ne
16 daatiieta of h»km 8«ptiior
RegioB: CanadUn ■hfeld im
fencrml.
7 diatrjeu of hakm Sapwior
Recion.
2 diatrieto of h»km 8«p«ior
Region.
AdiroDdacka, Ontario.
Laramie Mta.. Wjromlac.
Black HiUa, 8. Dakota.
Central Tesaa.
Colorado Front Range.
Britiah Columbia.
Grand Canyon, Aria,
flweden.
Sweden.
Sweden.
Finland,
Finland.
Finland (Rapakiri. batbo-
Uthie?).
Brituny.
Mt. Lofty Rangea, Bevtli
Australia.
Cape Prorinee.
Shan-Tung and oihor di^
tricU, China.
New York and New Eaglaiid.
Chriatiania region.
Weetem Norway.
Canadian AppalaeUana.
Fox lalanda and Perry bmia.
Maine.
Brituny.
Germany.
Devon. Cornwall.
New South Walea.
Queensland.
Pyreoeea.
New Zealand.
Caucaaua.
Hierra Nevada.
Cascade Range.
Coast Range of Britiah Col-
umbia and Alaaka.
West-Kootenay Diatrlet.Brit-
iah Columbia.
Idaho- Montana.
Patagonian Cordillera.
New Mexico.
Pyrenees.
TyrcJ.
Washington State.
YrllowNtooe Park (batbo-
lithir?).
SUBJACENT BODIES 99
partly summarize the history of the world's batholithic intrusions in
relation to orc^enic movements,'
Cross-cutting Relation to the Invaded Formations. — A leading
characteristic of stock or batholith is that its contact-surfaces usually
truncate the planes of stratification or of schistosity in the invaded
formations.
This rule is specially patent in the published maps of subjacent
bodies. Examples are reproduced in Figs. 43, 44, 45, 46, 55, 62, 64,
and 164.'
FiQ. 43. — Map Bhowing reUtian of the Castle Peak stock (British Columbia-
IVathinKton) to the invaded Cretaceous aandatones and argillites (C). Strike and
flip Uate solid; fault lines broken; figures show values of dip. Scale, 1 : 125,000.
Studies on batholiths and stocks which have been deeply canyoned
by streams, and on others which have been explored underground by
deep mining, show that subjacent bodies cfoss-cut the country rocks
>l«i in vertical sections. (See Figs.-46, 115, 168, and 169.)
'Id the table the prefix "epi" means that the corresponding orogenic period
ioMiy succeeds the Bedimentary period, the name of which follows the prefix in
tbe Gnt column. The "intervals" noted for the Sweniieh and Finnish cases are
>liue which elapsed after the respective orogenic movements were completed
ud before the next recognized sedimentary series was deposited.
'Sec B. K. Emeraon, Mon. 29, U. S. Geol. Survey, 1898, Plates 25, 28, 3!,
32, ud 34; J. C. Branner et tU., Santa Cms Folio, U. S. Geol. Survey, No. 163,
IKNI, map and plate of structure sections.
100
IGNEOUS ROCKS AND THBIR ORIOIS
In exceptional csuicti the country rockn of even poHt-C'ambrian
batholiths have Itoon motamorplioxod in kucIi a way that these rocks
show a new schiittosity or clfavaRC running parallel to the batholithic
contact (Fig. 47). The cross-cutting quality of the intrusions is then
not so quickly obviouR, hut seldom, if ever, does it fail to appear on a
close examination of the metamorphosed terrane.
Fio. 44. — Mnp of thf Shap iQ-iiiiitc- nltirk, Knitland, HhowinK it* croNbcuttinit
cb&r&ctcr. (.\f1i'r \. Iliirkcr and J. K. Mnrr, (jiiiirt. Jour. Ceol. Sac., Vol. 47,
1891, Plalc 10.) Faults Hhnwn in hriivy limf.
Principal Features of Roof and Walls.— The contact-surface or
roof of a liatliolith or stock characteristically hoA the shape of a dome.
(See Fig. 48.) The dome is, however, usually divemified l>y salients
and re-entrants. Very <iftcn great salients of the country rocks, with
the approximate shap«-s of inverted ])yraniids or of downwardly di-
rected wedges, hang from (he gent-ral roof (Figs. 49, 50, and 51).
Kueh have been called ''m»f-pcndaiits."' Vm\\ may be recogniiedai
a projection of the roof rather than as inclusions of country roek,
partly by its great size and, still more clearly, if the average ■rtril
> R. A. Dsly, Bull. Geol. Soc. America, Vol. 17. 1S06. p. ]
SUBJACENT BODIES
structures 1 es parallel to the average strike of the country rock sur-
roundu^ the batholith If such a mass were an inclusion in the once-
molten magma t would regularly be sh fted or rotated out of the
Fig. 45. — Map ot monionite stock in the Telluride quadrangle, Colorado,
iboiriiig cro«-cutting character. (After Telluride Folio, No. 57, U. 8. G. S., 1899.)
If, monsonite; D, diorite porphyry; R, Potoai rhyolite; /, lotermediate series
(volcanic breccias); S, San Juan tufTs; E, Eocene conglomerate; K, Cretaceoua
ihale and sandstone; J, Jurassic shale, sandstone, and limestone. M, D, R, I, and
SsK Tertiary; the monzonjte outs all the bedded rocks.
regional strike. Further, as noted in Chapter X, it would be prac-
tically impossible for a very lat^e inclusion to remain suspended in a
molten batholith, even if it were as viscous as hard pitch.
,„^_£3<a^g— —
^:v;n^-TT-r
i,^fe
^^*i^^,\'.'.
.'.'."X"-"*''''
r:r" ■'"::::}!%%•
Fig. 46, showing cross-
Pia. 40. — East-west section through stock mapped i
cutting in vertical plane.
Again, batholithic roofs often deviate systematically from the
pore domical shape because of the presence of re-entrante composed
of tbe batholithic material. These projections of the igneous ma&s
102 IGNEOUS ROCKS AND THEIR ORJOIN
have been callett "cupolaa," from the analogous relatioD of an artificial
cupola to the building of which it ib a part.' Many stocks are cupolas
on batholiths. P'xampIeH aro noted in FigH. 50 and 51.
rocka basic rocks
I'lo. 47. — Plan of hiithulith in l)i<ln-t>)l nur quadrangle, Cftlifomut {di
haehurtK\. ahowitiK nn (>\iiiii|ilc iif piriiihi^al rleuV3|[>^ or schiatCMBtjr (brotvn Urm
in country Tockt). (Attcr Bidwrll [l:ir Folio, No. 43, U. S. G. S., 18118.) Snk.
1 : 4flO.(»00.
Pos.-*iliK- iHjme cu|Kilas have l)con open to the sky, so that a subja-
cent hmly may l)c lorally rooflvtw. This has bocn a rare case, except,
■ It. A. Daly, Troc. Amer. Aod. Arts and Sciennii, Vol. 47, 1911, p. (».
SUBJACENT BODIES 103
of course, where volcanic vents of the central type have been fed directly
from a bathoUth.
Only quite seldom is it possible to follow a plutonic contact
vertically for more than a few hundred feet as it pitches into the
invisible depths. In general, however, the roof is marked off from the
"walls" of the igneous body by a relatively sharp increase in the
average dip of the contact-surface at the edge of the roof. Below
that edge the average angle of dip is always more than forty-five degrees,
and in most cases seems to average well over sixty degrees. In brief,
the walls of a typical subjacent body are highly inclined.
Though the main walls of a batholi th are often broken by apophyses,
they must he described as, on the whole, remarkably smooth. On the
Fio. 48. — Diagram showing features of an ideal batholith. P, roof-peudanta;
CA, aureole of contact metamorphism in folded sediments; S, satellitic stock.
accurate maps now published, the contact lines of deeply eroded
stock or batholith are almost always flowing lines. This is another
oF the fundamental facts which has to be explained by astable petrogeoic
theory. Illustrations are given in Figs. 39, 43, and 44.
Downward Enlargement. — In general, the average dips of the
main (molar) contact surface of stock or batholith are quaquaversal.
The few recorded exceptions to that rule refer to outcrops which are
limited both vertically and horizontally. It is quite possible that these
contact surfaces would also show quaquaversal dips if they were
better exposed, especially in depth. In any case, the rule is clear that
the greater intrusive bodies enlarge downwardly. The enlargement
in the horizontal section is rapid at first, that is, to the "edge" of the
(wf; it is usually then much slower to the deepest levels exposed by
natural erosion or inferred from mining or drilling. In all the greater
lONEOVS ROCKS AND THSIR ORtOttT
SUBJACENT BODIES 105
subjacent bodies the enlargement doubtless contiaues for one or more
miles in depth, but it is hardly probable that the increase in the hori-
Fio. SO. — Map showing roof-pendant in, and cupola Btock aatellitic from, the
SDoqualmie batholith, Washington State. (After Saoqualmie Folio, No. 139,
U.S. G. S., 1906.) GD, granodiorite; D, diorite; MA, andeeite; MR, rbyolite;
XS, date, etc.; EB Teanaway basalt (Eocene). All formations except EB are of
Miocene age. Scale 1 210 000
Kio. 51. — TrftDBverse sections in the Sierra Nevada, California, showing
tupoU stocks and roof-pendants in a granodiorite batholith (G). (After Truclcee
Folio, No. 3B, U. S. G. S., 1897.) V, Neocene volcanios, chiefly andMitea; J,
Jun-Tiios (T) slate and schist.
wntal section continues indefinitely. Figs. 52, 53, 54, 55, 56, 57, 58,
40, 43, and 49 are maps and sections (in part hypothetical) of a series
IGNEOUS ROCKS AND THEIR ORIGIN
Kin. M. — Mii|i slmwinR idripifiil iiiiroolinK of ii uranitc Htock in 8«xoDy
(AftiT Cli-yrr-KlircnfririlerwWf Siktlon ol the Gcol. Kpoiiatkftrt«.) /nrertfd
tiirrlK, (craiiitp: A, iiictumdrphir mirpolc in iiiiiwovitc itrhist; P, phyllite; M, mui-
covilp whist ;''i'.W. Riicissip miraHrhiHi. TwcibuHHC«of thefO'anitebody urezpoMd
liy criMiion: thi'Kr<'»t wiillh nr A in tn Imt pxptnincd by the rapid cnlarftement of theM
boaacs in depth. (Cumparc K. Didmer, Zeil. fUr prak. Gcol., Ud. 8, 1900. p. 297.)
'Wi
■ ^"^
TBEUtN
_.
^■^ ^^^S^
^
.^
1
t'i'i. -'i.'i. — .\hip slum jriK p:ir(i:d unriHirmE uf s l:ir(£p ftrunite stork in SAXony.
(.Vfter i;.-..!. S|wii;ilk;.r(.> .!<■* K<.n, S;i.Ii>,i,h— Tr.iicn nn.i npJKhborinK SektkNieli.}
a. KTHiiii.-; ri: ( atiihrinn :>n.l ulU-r artcillil.' and phylhtp; a. inner meUmocpUc
aurm.l..; I.. ..iiI.t nii>i:iiri..rplii.- ;iiirr..!r-; .1, iniii-r mirc^lc of Kiirhberx-ScfanMbcTl
KTanit)- liathnliili; H, uuUt aiircok' til tb>- Hump. Tlic Krcat width of a uid A mtf-
KMta that much of the riMif uf the atuek Htill remains. Scale, 1:275,000.
SUBJACENT BODIES
Fio. 54 — Sect on show ng pai'tial unrooRiig of & granite stock (5) in the Seward
Peninsula, Alaska {After A Knopf Bull 358, U. S. G. S., 1908, p. 27.) 1,
Bchiats; 2, contorted limestone Hor zontalscale, 1:84,000; vertical scale, 1:60,000.
'1 'm
A ^ jtt^'yyii iiii""'i'
UJ
ffl
^^B
3
a
%M
p-y^^^m
l§
^l
Wf
"^^r^
Flo. 55. — Map and section showing croBS-cutting chsfacter and downward
nilirjraient trf a granite stock {inverted carets) in the Selkirk Mts., near the 49th
PinUel. (R. A. Daly, Memoir 38, Geol. Surv. Canada, 1912, p. 299.) The broad
■oreole of contact metamorphism and shattering in the nearly vertical etrata ia
•kowD with dots.
108 IGNEOUS ROCKS AND THBIR ORIGIN
of stocks and batholiths showing a<lvance in erouonal unroofing, wHh
the implication of downward enlargpment. Figa. 59, 60 and 61 illiu-
SCHNCCBERG
Fio. AC. — Section Hhowinft how mining h&a proved downwud enlargMnent for
& German Rranitp batbolith (ijvtrttd eareli). (After Geol. Speiw]k«rt« dM Kta-
Sactuteni, Sckt. Srhnpcberft, Bnd R, LepsiuB, Geologie von DeuUehltind (Tngnl
mann),) t, outer metamorjihic aureole, in phyllite; t, inner metcmacpliie
aureole, in phyllite; S, mining shaft«i F, Fldtie.
tratc well exposed cuntaet» Mhowini; quaquavorsal dips to considerable
depths.
Fio. .17. — Dfrnonstratpd pmfilp of th*" T.auxiti K>^nit« hatholith (p), (
which is intrusive into metamorphosed fp'iiywa<:kea (A'J. (Same nt, am (cr F!(.
56, Sekt. Kloster St. MarienBtem.) Horiiontal scale, 1:86,000; vertioiJ snle
1:172,000.
The attempt to "rxtrapolatr" from the linown facts, so as to
construct the probahle type for Htock or hatbolith in great depth, is,
Fig. S«. -D<-iiion.»tr:itr.l pn>filr of .1 in-iinitc l)!ilhrilith (ff) in the Firhtfflittbirie.
(After R. RUdumunn, Neuu) Julirb. (ur Min'-r., I). It. 5, 1887, p. 674, ud R.
Lepaiufl, Geologie von DeutKhlan<l (EnRelmann).) C, Cambrun dfttM uid meU-
morphosed equivalent (rfoff); P, Phrllitr and metamorphosed vquiraloit (dsti)_
of course, to enter the realm of hypothosi.-!. N'cvertheless, this proUem
ia at the ver>- heart of petroftenic philosophy. It must be attacked if
SUBJACENT BODIES 109
petrology is to make essential progress. The writer believes it can be
attacked indirectly and that the positive results can justify this
venture into the field of speculation. The conception that batholiths
are bottomless (in the sense that its visible country-rocka do not
extend beneath it) seems to accord with the vast multitude of facts
embodied in igneous petrography and geology. No rival conception
has ever succeeded in so thoroughly explaining these myriad facts.
The preferred hypothesis has, in the writer's opinion, a sanction which
can probably become of the same order as the wave theory of light or
the atomic theory of matter. Yet the conception that the typical
Pig 59 — Plunging contact (AS) at south mde of granodionte stock (left)
n ling Cretaceous sedimentfi (nght) at Castle Peak Bnt ah Columbia (R A.
Dtlj Memoir 38 Geol Suit Canada 1912 p 498 ) The vertical distance be-
tveen A and B levels is 800 ft. "H^ced from a photograph.
batholith ia bottomless is hypothetical and, as such, its consideration
>s deferred to later chapters in this book.
Replacement of bivaded Formation. — As a guide to further
thought along this line, another fact may here be appropriately men-
lioned. Batholiths and stocks appear to have replaced their respec-
lirp country rocks in the act of intrusion. This statement has the
^mock of the subjective, but its subjectivity is that of common sense
ttacting on the facts of nature. Just as surely as the roof-rock of a
^Accolith appears to have been displaced in the formation of the lacco-
no
IGXEOUS ROCKS AND THEIR ORIGIN
litliic chamU'r, m> the roof-rock of a batholith appears to have been
replaced in the dove lop mont of tho upper part of the batholithic
t'liamtxT. The larcnlittuc iiiiiciim has Hfled or thrust away the roof-
rock to innkc room fur itself. In no ca.so has the magma of a t>'pi('al
Iiathohth rt'achod its level reRistcreii in the general mass of viniljle
plutonic rock I>y nien-ly liftiuft thi> roof-rock or by thrusting it aiiiilp.
The emplacement of each l>atholitli ha-t lieen a relatively quiet procp8!>,
uiiac<'om}>aiu(>il by any sueli <-olussal movements in the other parts
Fi<i. TiO.— I'liinRinK contiirt ABfl> M north »f Mmc stock u Uwt ibowa in
Fir. Kl. fJriiiif»iiiiriti' mi rlu- rialii ; (ntin-i^Hi.i M<-<liiiiiniii on theMt. Tbevcr-
tirul <)iMiui'-(< h<-iw.-.-n thi' A :iii<l H levels is l,T{K)f(. Traced frum a pbotop^ih.
of the earth's crust ns would Ik- n-riuired to form the huge ehambrr
liy the "lnci'i)lithi<'" niuile of intrusion.
Plutonic Keology has nlrcit'ly su]>plicd hundreds of examples.
Some which have undergone dclitiled stmly may 1m> recalled: the com-
l>ositr stock at .\siutney Mountain, Vermont (Fig. 64); the wonder-
fully c\i)r>scil stui-k at Casile Peak, Iloicomeen Kai^^ in British Co-
lumbia I I'lKs. 1:1. -V.l, lid, and (ill ; a stock in the Telluride quadran^,
( 'ohira<lo ( l-'iKs. I.-|. Ill) : t he slock at Marysvillc, Montana (Figs. 114 and
Hi)); many ^^tocks and small I>athotiths in Hrittany and in the Pyie*
SUBJACENT BODIES
nees (Figs. 39, 42, 62, and 63) ; and the satellites of the Bayonne batho-
lith of British Columliia (Fig. 55). A typical record is here quoted
//.> t.u t/'.i.U'.t t.Sti.t ■*.*.
Fio, 61.— Cany on- wall section at north of eamc stock as that shown in Fig.
59; from a field sketch. (?, Granodiorite; C, Crctaceoua argillite. Elevations
in feet; dips in degrees.
From Cushing's report on the geology of the Thousand Islands region of
New York State. Concerning the Picton granite, he writes: "Our
-^ Mil.
Fig. 62. — Map showing replacement of sediments by a granite batholith
ICj in Brittany. (After C. Bairois, Annates soc. giktl. du Nord, Vol. 21, 1893, p.
UO.) Se, Silurian and Devonian slates; S, Silurian sandstones; P, St. L6 phyl-
lit«. Note the preservation of strike in the sandstonea.
strikes and dips, read on the rocks in the field, gave absolutely concor-
dant results as we passed from one inclusion [roof-pecdant] to another,
112
IGNEOUS ROCKS AND THEIR ORIGIN
results also concordant with the readings obtained on the aame rocks
I>cyond the roach of the intruKion. We were able to map the original
I>elt8 of (ironvillc quart zile and tichist, and the iotrusionfl of Laureatiao
granito gneiss, as accurately a» though the Picton granite was not pres-
ent, so little had thoy been disturbed by the intruaon."'
The crosw-cutting quahty of the batholiths, where viewed in plan
or in Hcctiun; the nature and position of roof-pendants and of mag-
matic rupolaH; the Hlight influence of batholiths on the tectonic axes
of their respective countr>' rocks; the general fulure of circumferen-
tial faults or axcH of warping almut batholiths, and the fact that none
Kio. 03. — Map ahon-ing rcplu^cmmt of smliiacntB by the C*ul«reU irautc
{G) of the Pyrenees. (After A. BmwoD, Bull. acrv. carte gfol. Fruuie, Vol. 14,
1003, PL V, Fir. 1.) C, C&rt>onireruuHquartiitc,ctc.; S, CarbonifennMilate, etc.;
L, Devonian limcutonei D, Dcvoniiin Kraywocke, etc.
of these IkkUcs hat yet been aliuwn to. have a Imttom — all theae facts
are oppos«>d to the application of the " laccolithic" mechanism in the
interprelntion of subjacent bodies. On the other hand, the direct
lielil observutions favor the interpretation that the upper part of each
subjacent Iwdy lias l>ecn emplttrcd liy a kind of rorroaion on a huge
.•*cftle. IIow tlii.-* replacement has taken place is a matter for theoret-
ical discus-sion and therefore nne outside the special scope of the pres-
cnl chapter, which is primarily encased in pointing out the eaaential
observations wliieb have l>een actually made concerning subjacent
I«.dies.
■ H. P. Cj^ihini;. Bull. 143, N'ew Vark Suto Miueum, 1910, p. 43.
SUBJACENT BODIES
113
Chemical Composition of Subjacent Bodies. — The chemical varia-
tion in a stock or batholith is often considerable but the average com-
position is approximately determinable so far as the outcrop is con-
cerned. In the following section of this chapter the statements regard-
ing chemical composition refer to such averages. Some discussion
of the variations observed in individual bodies will be found in later
chapters.
The larger subjacent masses are usually composed of granite,
though granodiorite, quartz diorite, and syenite batholiths are known.
. , _J^<^
^^^^
(fev:%?;dfev:--:-:-->^^^^
0 3 ... 0
J
HE?
!v.*v!v/!'
'II i 1 1 \\\l ul
Fio. 61. — Map and section showing replacement of phyllitee and gDeisBea by
tbe composite etoclt at Mt. Ascutney, Vermont. (R. A. Daly, Bull. 209, U. S. G. S.,
1W3.) G, biotite granite stock; N, nordmarkite atock; NF, nordmarkite poi-
ptiyry and paisanite; P, pulaskite; D, diorite stock, with gabbroid and eeseiitic
^utsee; Pk, phyllit«fl with thin limestone interbeds; On, gaeissee, with thick Ijmo-
itooe pods interbedded. Symbol for strike and dip.
It is an open question whether any other plutonic type composes a
trae batholith, that is, a subjacent body more than 100 square kilo-
meters m total area of exposure. The great gabbro mass mapped in
the Roaebui^ folio of the United States Geological Survey covers more
than 300 square kilometers (150 square miles), but its relations to the
surrounding Tertiary sediments suggest that this body may be a huge
laccohth. (See atructiue sections by J. S. DiUer in the (oUo mentioned.)
The Duluth gabbro covers about 6000 square kilometers (2400 square
114 IGNEOUS ROCKS AND THEIR ORIGIN
miles) but it is regarded as a laccolith by Van Hise and Leith. ' Ram-
say and Hackman make a similar interpretation of the huge mass of
nephelite syenite, urtite, etc., in the Kola peninsula of Lappland.*
Possibly the vast Inxlies of anorthosite in eastern Canada, in New
York State, and elsewhere, are also laccolithic, but little can yet be
definitely affirmed as to the intrusive mechanism of these extraordi-
nary Ix>dies, whos<» true character is, in this and other respects, one
of the deepest mysteries of petrology. (See Chapter XV, page 321.)
The large intrusions of alkaline rock near Julianehaab, Greenland,
are descril)ed by Ussing as batholiths; yet certain of their structural
relations and the character of their magmatic differentiation seem to
indicate a chonolithic or laccolithic origin for each of these masses.'
These magmas have performed a limited amount of magmatic sloping
and it was here that Ussing Ijecame an independent author of the
stoping theory. Yet the enlargement of the magmatic chambers by
that method was probably moderate. One must l)elieve that very
long-continued stoping of such silicious rocks as the sandstone and
ancient granite intruded by the dominant foyaitic magma would
generate acid phases in volumes much greater than those actually
found.
A review of oth(»r examples of the abnormally constituted, greater
ma.Hsifs, with respect to such structural relations as can actually be
observed, shows that the* mere areal extent of a "plutonic" mass can-
not be taken as a sure evidence of its **l)ottomless," batholithic char-
acter. The sill at HoiM»town, South Africa, is known to cover more
than 5()00 square miles. ^ If it were a few thousand feet thick, instead
of a few hundred f(>et, that body could doubtless have crystallised
(like the actual Bushveldt laccolith) with the texture and coarse grain
of the normal plutonic rock. If, in addition, erosion had not exposed
the floor of the sheet, it might have deceptively resembled a true
batholith.
In average composition individual stocks have more variety than
that characterizing the larg(T subjacent Ixxlies. Besides granites,
granodiorites. quartz diorites, and syenites, the list of types recorded
as forming stocks includes gabbro, norite, quartz g^bbro, diorite,
nephelite syenite, shonkinite, or their porphyritic equivalents.
1 C. R. Van IIimo and C. K. I^uth, Monograi>h 52, U. S. Geol. Sunr^, 1911,
p. 372.
' W. RaniHay ami V. Hackman, FVnnia, Vol. 11, Part 2, 1894, p. 96.
*Seo N. V. rK8inK. MccMrlolnor om Gronland, Vol. 38, 1911, pp. 38, 49,50,68,
251-5, 290, and IMate i\.
* A. W. Rogora and A. L. du Toit, Geology of Cape Colony, 2d edition, LondoB,
1909, p. 261.
SUBJACENT BODIES
115
Classification of Stocks and Batholiths
Simple stocks are composed of material intruded in one period
of irruption. Types are illustrated in Figs. 39, 42-46, 50-55, 59, 60,
and 61.
On the analogy of multiple dikes, a multiple stock may be con-
Fio. 65. — Section of Okanagan composite batholith at 49th Parallel. (R. A.
Daly, Memoir 38, Geol. Surv. Canada, 1912, p. 426.) S, Paleozoic schist, lime-
stone, quartzite, etc.; C, Cretaceous sandstone and volcanic breccia; /a, Chopaka
peridotite; /6, basic complex (gabbro, hornblendite, etc.); ^, Ashnola gabbro;
5a, Remmel granodiorite (gneissic); 56, Osoyoos granodiorite (gneissio); 4,
Kniger nephelite syenite and malignite; 5, Similkameen granodiorite; &a. Cathe-
dral granite, older phase; 6b f Park granite; 7, Cathedral granite, younger phase.
The igneous rocks are numbered in the order of intrusion. Vertical exaggeration
of about 5 to 1.
ceived, namely, one which is composed of uniform material demon-
strably intruded in two or more distinct stages of one eruptive period.
Apparently no example is on record.
A composite stock is composed of materials demonstrably intruded
Fig. 66. — Section of the composite batholith of New England, New South
Waka. (After E. C. Andrews, Records, Geol. Surv. N. S. W., Vol. 8, Pt. 2, 1905,
Fig. 7.) P, Permo-Carboniferous sediments; A, Dark porphyries; B, "Blue
granite"; C, Sphene-granite porphyry; D, "Acid granite." The rocks are
DAined in the order of mtrusion. Scale approximate.
io two or more distinct stages of one eruptive period, the magmas of
the different intrusions having different compositions. A number of
examples are known. (See Figs. 64 and 181.)
In corresponding fashion, simple and composite batholiths may be
distinguished. No multiple batholith has been described and there is
1 16 IGNEOUS ROCKS AND THBIR ORIGIN
m
reason to doubt that one shall ever be discovered. During the ex-
tremely long time represented in the crystallization of part of a large
magma ehaml>er, the forces leading to the differentiation of the residual
liquid would tend to produce magma differing chemically from that
already solidified. In general, therefore, the result of renewed in-
trusion during a single petrogenic cycle should be a compositei not a
multiple, batholith.
Simple batholiths are illustrated in Figs. 40-2, 47, 57, and 58.
The greatest exposed post-Cambrian composite batholith is doubt-
less that forming the greater part of the Coast Range of British
Columbia and Alaska (Fig. 37). Another example, perhaps an
outlier of the vast body just mentioned, has been sectioned and
descril>ed by the writer under the name, " Okanagan composite batho-
lith*'^ (Fig. 65). Nearly all of the larger plutonic masses in the
North American Cordillera seem to be of the composite class. To
this class the writer is disposed to refer the intrusive complex of the
Christ iania Ilegion, made classic by the writings of Brdgger.^ Figs.
00 and 116 illustrate composite batholiths in Australia and ScoUmnd.
> R. A. Daly, Bull. Gool. 8oc. Amer., Vol. 17, 1006, p. 320.
' W. C. Brogger, Die Eniptivgcsteine dee Kristianiasebietet, Vol. 1, 18M,
and Vol. 2, 1895.
CHAPTER VII
EXTRUSIVE BODIES
Classification
As far as possible the classification of the extrusive bodies should
be rigorously tied to field observations. But experience shows that
little progress can be made without some immediate interpretation
of the simple observed facts. It is obvious that a complete classifica-
tion of the known extrusive bodies cannot be made without a prelimi-
nary, very thorough exercise of the faculty of interpretation. That
means the use of principles which, by their very nature, are not to be
deduced from field facts alone. The classification to be outlined in
the present chapter, being founded primarily on field observations,
can be neither complete nor wholly genetic. It will serve, however,
to exhibit the leading facts which, with a multitude of others, are
capable of fuller and more systematic description after the theory of
igneous action has been reviewed.
Three principal groups of extrusive masses have been recognized,
the distinction in each case being founded on the mode of extrusion.
Two of the groups, respectively formed by "fissure" eruption and by
"central" eruption, are regularly named and discussed in modern text-
books on geology. The third group of volcanic masses is not generally
appreciated at its apparent value; it includes those bodies of lava
formed by what is here called "de-roofing" eruption. As indicated
below, these three types of eruption are transitional into each other,
but the recognition of the pure types is none the less helpful in forming
a basis for volcanic theory.
I. Fissure Eruptions
The simplest mode of extrusion is illustrated in the greatest
basaltic fields. As a rule, the lavas have there issued quietly, without
explosion of violence sufficient to form dominant layers of tuflf or other
p>Toclastic material. However, occasional ash-beds locally interrupt-
ing the lava flows represent temporary eruption of the familiar "cen-
tral" type. The famous eruption at the Icelandic Skaptar J5kull
fissure during 1783, the most imposing historic example, typified this
subordinate part of pyroclastic deposits in basaltic plateaus.
117
118
lONBOUS ROCKS AND THEIR ORIOIft
Table X, page 191, liHtn some of the more important ftssure erup-
tions. Those of Tertiary date are generally little deformed and merit
the common name, plateau l>aiialt». The fields celebrated for their
extent are those in India (Fig. 67) and the northwestern United
States. Their only rival among the post-Jurassic fields is probably
that of the North Atlantic. Even the remnants of the plateau basalt"
Fin. n7. — Map Hhowing dial ritiiit ion of the Decran trapd (doile^,
still visible aft<-r the late-Tertiary sulwidencc of this ocean bam
considerable, a." .-ihown in Thoroddsen's table:
tjitiouud Armw
(aq. km.)
Scotland and Ireland
. 10,000
Faroe lalands
1,326
Iceland
104,786
KasI coaat of Orpcnlantl. .
20,000+
It is possible that the North Atlantic Tertiary basalts have covered
an area totalling 500,000 sttuarc kilometers. Those, together with the
basalts of the northwestern United States and the Decran, reprearat
EXTRUSIVE BODIES
119
a grand total of nearly 1,500,000 square kilometers or one-third of the
area estimated by von Tillo to be covered by all the "young" volcanic
masses of the continents and islands.
Observed thicknesses for the plateau basalts reach high figures.
The maximum proved for the Teanaway basalt (Eocene) in Washing-
ton State is 6000 feet (1830 meters). The overlying Yakima basalt
rtrr
•9IIOO
ifiy
>5000
•4400
-3800
-SCOO
•t«oo
Fig. 68. — Sections showing great thickness of Tertiary fissure eruptions of basalt in
Oregon. (After G. A. Waring, Water Supply Paper 220, U. S. G. S.. 1908, Pl.lO.)
(Miocene) is more, perhaps much more, than 2000 feet (600 meters)
in thickness at the Yakima River canyon. The Oregon basalt is also
very thick (Fig. 68). A. Geikie estimates the maximum thickness of
the Iceland basalts at 3000 meters. That of the plateau basalt in
northwestern Greenland is more than 1200 meters. The Deccan traps
are locally more than 6000 (1830 meters) thick.
Fig. 69. — Section of great fissure eruption in Williams Canyon, Arizona. (After
W. T. Lee, Bull. 352, U. S. G. S., 1908, p. 54.) C, crystalline rocks; S, sand and
gravel; B, massive flow of basalt, 800 ft. thick; 7, vent of the flow, 400 ft. in
diameter.
The thickness of individual flows in such terranes seldom surpasses
100 meters. In the United States the average thickness is probably
less than 15 meters. The observed averages for the Deccan flows in
the Bhor and Thai Ghdts are respectively 19 meters and 26 meters,
but these values are said to be too high, on account of the difficulty
of distinguishing the flows in many cases. ^ Thoroddsen states that
* R. D. Oldham, A Manual of the Geology of India, 2nd ed., Calcutta, 1893,
p. 261.
120 IGNEOUS ROCKS AND THEIR ORIGIN
the Icelandic flows are generally from 5 to 10 meters thick. The
South African flows seem to have nearly the same average dimension.
Where, however, the eruption takes place in a valley the lava may
attain much greater depth. Lee has mapped a remarkable case at
Williams canyon, Arizona, where the basalt is locally 800 feet (240
meters) thick and covers an area 14 kilometers broad by 22 kilometere
in length. This flow has a welU'xposcd feeder with the unusual width
of 400 feet (120 meters) (Fig. 69).
Thoroddnen gives some illustrations of the lengths, areas, volumes,
and surface slopes of sample Icelandic flows:
^ Length, Area, Volume, at
Flow I I A Slope
km. sq. km. cu. met. ^
LakTfiBSure (1783 eruption) 90 5G5 12,320,000,000'........
Veidivatnahraun (prehistoric) 150 1,080 43,160,000,000 5'
Frambnini (prehistoric) 110 465 23,250,000,000 M^V
Eldjgd (about 930 A.D.) 0,325,000,000_ ._^. . . .
He also estimates the volume of the greatest lipari tic flow in Iceland
(Hrafntinnihraun) to be about 500,000,000 cubic meters and thus
insignificant when compared with the basalt floods.
The foregoing table and that on page 290 illustrate the note-
worthy contrast in volume between flows in a basaltic plateau and
those emanating at "central" vents.
The feeding dikes of the plateaus are numerous but narrow; the
width probably averages loss than 10 meters and seldom reaches 50
meters (Fig. 70). The narrowness of the feeding fissures is a direct
evidence of the relative rapidity of these extrusions. If the tnmgmm
remained stagnant many days or weeks in such narrow vents it would
necessarily solidify and seal the fissures. These could then be re-
openod for continued eruption only by explosion or by a remelting
of the congealed rook. Noithor of these events has regularly occurred
in the fields of plateau ba^^alts. It seems to follow that, in general,
fissure eruption is a sudden aot.
The relation of basaltic fl(M)ds to crustal stresses is briefly discussed
in Chapter IX, p:iges 18G and 191.
Wliilo the vast majority of fissure eruptions are basaltic, other
I)etrographic types aro roprosonted. The rhyolite of Corsica appears
to have welled up through numerous fis.sures rather than pipes (Fig. 31.
page 83). (iregory states that trachyte, andesite, and basalt are all
represented in the fissure oru))tions of the Great Ilift in Eastern Africa.'
Many small phonulitic flows, both in Kurope and in America, have em-
anated from fissures. In fact, it is quite possible that aU of the dans
> J. W. Gregory, The Great Rift Valley, Londoo, 1896, p. 335.
EXTRUSIVE BODIES
121
are represented ia extrusioiis of this kind. Yet basalt doubtless
composes at least 95 per cent, of the total Icnown volume of fissure
eruptives. The sgnificance of this important fact will be noted in
Chapter VIII.
Via. 70. — Map of dike feeders of Tertiary fisaure eruptiona near Mt. Stuart,
ffwhiDgton. (After Mt. Stuart Folio, No. 106, U.S. G.S., 1904.) PC, pre-Tertiary
complex (peridotite and greenstoDe) ; G, Mt. Stuart granodiorite; SS, Eocene
Snuk sandstone; B, Eocene basaltic fissure eruptions; their feeders shown by
bn; RS, Eocene Roelyn sandstone and shale; R, Pliocene (?) rhyolite.
II. Extrusion bt De-roofing
We have seen that the covers of most batholiths must have been
td moderate initial thicknesses, to be measured in hundreds or thousands
of feet but not in tens of thousands of feet. In one part at least the
roof rock of the great Boulder batholith of Montana was about 1000
feet thick.' The original cover of the Snoqualmie batholith of
Waslungton is stated to have had locally no greater thickness than
4000 feet.'
Notwithstanding the great areas covered by these magmatic
boiSes, and also in spite of the fact that roof rock is generally denser
than acid magma, geolc^sts are agreed that batholithic roofs have
generally remained intact during the respective magmatic periods.
In fact, the geological text-books make no mention of the posability
' R. W. Stone, Bull. 470, U. 8. Geol. Survey, 1911, p. 78; cf. J. Bwrell, Prof.
Piper No. 67, U. S. Geol. Survey, 1907, p. 166.
■ G. O. Smith, Snoqualmie folio, U. S. Geol., Survey, 1906, p. 12.
122 IGNEOUS ROCKS AND THEIR ORIGIN
of partial foundering of the roof of a subjacent body, and it is not likely
that it has taken place in the case of the great majority of visible
stocks and batholiths. Nevertheless, there are evidences that some
subjacent masses have been opened to the sky over areas much more
extensive than those represented in apophysal dikes.
Scottish geologists have described a case, at Glen Coe, where part
of a batholith's roof has sunk along peripheral faults (a "cauldron-
subsidence")- As the sinking progressed the magma was squeesed
up, following the faults on nearly every side of the sunken area (Fig.
116, p. 197). If the faulting had progressed still further it seems in-
evitable that foundering would have occurred.' The present writer
has outlined the facts which suggest that the rhyolite plateau of the
Yellowstone National Park represents a broad cicatrix in the roof of a
Arv€7 €>/
PHASE OF BATHOLITH
Fin. 71. — Idoal Roction illustrating tho hypotheflis that the rfayolHe sad the
thermal phenomena of the VeUowstcme Park are directly related to the fouaderinK
of part of the roof of a lati'-Tertiary hatholith.
Pliocene butholith; that over hundreds of sc|uare miles the batholithic
roof was ver>' thin when the rhyolite plateau was formed; and that this
thin roof was locally swallowed up (Hg. 71). The field habit and field
relations of the rhyolite, and particularly the long persiatence of
geyser activity on the plateau, are evidences supporting this hjrpothe-
si.-2 (Fig. 72).
In the basin of the Fox Kiver. Wisconsin, the granite of a pre-
(*ambriun batholith is gradually transitional, through a nouMTe
rhyolite porphyry (keratophr>'e), into a rhyolite in which all the
evidence of former flowage and rapid surface cooling b apparent. The
areas of granite are contiguous and the areas of keratophyre lie m
a zone bordering the granite on the east and south, while the rhyolite
> C\ T.CMoiifch, H. B. Maiife, and K. B. Bailey, Quart. Jour. Gaol. 8oe., Vol
(>5, 1909, p. t)70.
* R. A. Daly, Proc. Amer. Acad. ArU) and Sciences, Vd. 47, 1911, p. 68.
EXTRUSIVE BODIES
areas lie in an outer zone beyond the keratopbyre and fartbest removed
from the granite.'
Fia. 72. — Map of the Yellowstone Park, showing the unexampled profusioD of
bot springB (.S) and geyser basina. (Prom Yellowstone National Park Folio, No. 30,
I'. S. G. S., 1896.) P, pre-Tertiary formationB; N, pre-rhyoLte, Neocene volcanic
breccias; R, rhyolite {PUocene or Miocene); SB, Snake River basalt (Pliocene);
Nodk ipolt, sinter deposits. I, Mammoth Hot Springs. 2, Terrace Mountain.
'^, NottIb Geyser Basin. 4, Lower Geyser Basin. S, Upper (jeyser Baein. 6,
Shoshone Geyser Basin. 7, Heart Lake Geyser Basin. 8, Hot Spring Basin.
At several localities in Sweden the acid porphyries (included in the
"leptJtes") seem to pass gradually into graiutes (also ortht^ndsses),
> W. H. Hobbe and C. K. Leith, Bull. Univ., Wisconsin, No. 1S8, 1907, p. 266.
124 IGNEOUS ROCKS AND THEIR ORIGIN
which crop out n^-ith batholithic proportions. The masave porphyries
have the fine grain and other characteristics of more or less metamor-
phosed extrusive rhyolites. They are often directly associated with
tuffs and other pyroclastic beds of the same chemical compodtion.*
The phenomena suggest very strongly that some of the pre-Tcrt-
iary batholiths actually reached the earth's surface through orifices
vastly larger than ordinary dike fissures. There the magmas were
chilled in much the same way as ordinary flows of rhyolite have been
chilled. The glassy, scoriaceous, and pyroclastic rocks, however
formed, have been greatly deformed and eroded, so that it is difllicult
to make a final test of this explanation for these older bodies. The
case of the Yellowstone rhyolite would doubtless yield more conclusive
results in a complete investigation of the problem. Whatever be the
fate of the roof-foundering hypothesis, it points to the need of a re-
vision of the conception that all large bodies of granitoid constitution
are ** plutonic" in origin. It must be remembered that the thin, scori-
aceous, glassy, or lithoidal cap, formed by atmospheric chilling in
an area of foundering, would be immediately subject to erosion and
to ultimate complete removal. It would be very difficult to distinguish
such a denuded batholith, of granular texture because cooled slowly
beneath the cap, from one which has had a roof of sediment or schist.
Roof foundering may, in fact, have occurred much more often than is
indicated in the districts so far discussed.
In recognizing this class of truly cicatricial batholiths we have
departed from the general plan of cla.ssifying igneous bodies on the basis
of fairly uniform agreement among geologists as to types of field re-
lations. The departure has been made in the interest of greater com-
pleteness for the classification.
III. Central Eruptions
Ordinary' volcanoes are characterized by piles of extrusive ma-
terial which has been accumulated pericentrically. The main gas
vent of a volcano is also the main vent for the discharge of rock matter,
whether fluent or solid. P'or this reason the type of extrusion generally
most familiar has come to be called *' central " eruption.* (Sec Frontis-
piece.)
The kinds of rock bodies associated with central eruption are
^ A. Ci. HoKborn, Bull. ChhA, Inst. Upsaln, Vol. 10, 1910 (reprint), pp. 36, 46,
«*)(>. In volume 5 of the 8.<inie hullrtin.s (HN)1, p. 19) O. Nordenskjdld notes that
tho extensive hiilleflintas of northea.'ttern SniAland nre transitional into granites
and sugKCi^ts that they are surface phases of those fn'anitcs.
* A valuable account of central vents is to be found in G. Mercalli'i ''I Vul-
cani della Terra," MiL«in, 1907.
EXTRUSIVE BODIES 125
already so familiar to the reader that only a brief description of them
is here necessary. They may be listed as follows:
Rock Bodies
1. Necks.
a. Tufif necks.
b. Lava necks.
c. Composite necks.
2. Plugs, domes (endogenous growths), aiguilles, cumulo-volcanoes,
mamelons.
3. Flows; superfluent, interfluent, effluent streams.
Special phases and features: Block lava, ropy lava, pillow
(ellipsoidal) lava, tunnels, lava cascades, lava scarps, tumuli,
homitos.
4. Cones.
a. Tuff cones, cinder cones, ash cones.
b. Lava cones.
(1) Lava domes (exogenous growths).
(2) Lava rings.
(3) Driblet cones.
c. Composite (normal) cones, breached cones.
d. Cone clusters, cone chains.
The negative topographic forms associated with central eruptions
may also be reviewed in summary form, since many of them are con-
nected with the essential mechanism of igneous eruption. The more
important forms are as follows:
Negative Reliefs (depression forms)
1. Craters.
a. Lava pits.
b. Maars (volcanic embryos).
c. Blow-holie^
d. Adventive craters (parasitic, lateral).
e. Nested craters.
2. Calderas (evisceration by explosion).
a. Simple calderas.
(1) With lava discharge.
(2) Without lava discharge.
b. Nested calderas.
c. Sunken calderas.
3. Volcanic sinks.
a. Simple sinks.
b. Nested sinks.
4. Volcanic rents.
lONBOUS ROCKS AND THEIR ORJOIN
Rock Bodibb
Volcanic Necks.— In many hundreds of cases, secular erouon has
removed volcanic cones wholly or in part and has often denuded
the non-volcanic formations underlying the cones. The feedini;
vents have thus l>ecomc exposed at levels far below the Boors of the
Fiu. 73.— Soctionuf Tolrnnic tufT nock piercing and OTnrUin by pIfttcMi buklta,
Faroe IsUrwlii. (After A. Ccikii-, Ancient Volcanoes of Great Britain, Vol. 2, 1SV7.
p. 205.) Scale, 1:1,300.
craters. The typical vent of th<- <-entral type is roughly cylindrical
or ncck-shapcd. Even at cunsidcrabic depth many of the vent fillings
or nocks are known to bo composed wholly of pyroclastic material,
with which some of tho country rock may be mixed. The«e are gen-
Fi«, 71.— I
of CarboniferoUB neck at East GranKC, Ptirtbahire. (AftCf A.
,Vol.l,18fl7, p. 42a.) ThcftnUtnT-
Htiinp, and Umeetone. Th« neek (A') '»
ndntonp, Hhale, and coal, without igDtoua
The neck ia from 1,500 to 2,000 ft. in
Gcikie, Anoient Volcanoes of Great Brili
erseil are CarboniriTOUs uhalefl, coal, iro
composed ot days, full of fragmcntii of *t
rock; waa it due to phrcatir expliKtion?
diameter.
erally called tuff ncckx [.Fins. 7:i-5). Other vents, lava necfc« (Figs.
7& and 130), arc filled entirely with massive lava; while atiU others,
composite necks, are matic up of pyroclastic rock cut by maaeiTe lava
(Figs. 77-80).
EXTRUSIVE BODIES
127
In certain young, though extinct volcanoes, the lava has risen high
in the craters, so as to form lava lakes with areas much greater than the
average cross-sections of the respective necks. The whole mass may
then solidify or part of the lava may sink away (Pigs. 81 and 82). In
Fig. 75. — Plan and section of explosion fissure cast of the Rock and S
Pifnthire. (After A. Geikie, Geology of Eastern Fife, 1902, p. 211.) The breccia is
romposed of fraginents of shale and sandstone. Fissure about 125 ft. long. Ar-
rows show the dip of the strata traversed.
!^uch cases the word "neck" is specially well chosen for the filled pipe
joining the "body," the subterranean magma chamber, to the "head,"
the relatively large prism of lava in the crater. If the erosion surface
Fia. 76. — Sketch of the Cabezon basaltic neck, New Mexico. (After D. W.
Johnson, Bull. Oeol. Soc. Amer., Vol. 18, 1007, p. 310.) The neck is about 1,400
ft. in dituneter; its siimmit is 2,160 ft. above the valley.
nowhere cuts under that prism, a mistaken idea is possible as to the
order of dimensions for the neck beneath. The vent at Kilauea can
hardly have a cross-section of more than 400,000 square feet, and there
are reasona for considering it as less than 50,000 square feet. At cer-
128 lONSOVS BOCKS AND THBIB ORiailt
tun times, by a rise of the lava, the area of the lava lake is as much as
1,500,000 square feet, perhaps fifty or more times the cross^ection
Fio. 77 — PlanH of Permian tuff necks, Ayrshire. (After A. Geikie, Aneiait Vol-
inoes of Great Britain, Vol. 2, 1897, p. 64.) Solid black in « and fi n
ve lava. Scale, nearly 1 : 12,000.
•
♦
DUMGOYN
w^
m
W
IP
w
^r^-----''".-
-'!ii||f!.,,/
■fir- ■.•:.--■..'
OuMraYM
0
.000 f,
®
0
3J0 „
Fir,. 7S. — Grounil plana of compoiiitc necks, Stirlingshire. (After A. G«ik>e,
Ancient Volcanoes of Greut Britain, Vol. 1, 1897, p. 395.) Shadtd artoM, ammw*
lava; doUrd areat, tuff iiml uRfclo'icratp.
area of the vent. Similarly, the Raring, superficial portion of a vent
may be easily mistaken for a true neck. Often an enormous pit has
EXTRUSIVE BODIES
been developed over a vent by one or more major explosions and then
filled with tuff and breccia. Unless denudation has removed the flar-
Fio. 79. — Twin volcanic necka of Carboniferous date, Scotland. (After A.
Geikie, Ancient Volcanoes of Great Britain, Vol. 1, 1897, p. 396.) Shaded areat,
massive lava; dotted areas, pyroclaatic material; D, dike. Scale, 1 : 14,500.
ing part of the pyroclastic deposit, the outcrop of the latter may exceed
[Qtmy times the average cross-section of the vent. It seems likely that
some of the recorded tuff "necks" of Scotland and elsewhere are seen
SiraihhIaneHms
Fio. 80. — Section of composite necks, Stirlingshire. (After A. G«ikie, Ancient
Volcanoes of Great Britain, Vol. 1, 1897, p. 400.) /, 2, S, PaleoEoic sedimenta; A,
andesitic lava; S, agglomerate; 6, diabase. Scale, 1:16,000.
in the outcrop only at sections passing through the upper flaring por-
UoQ of the pyroclastic filling.
In any case, central vents are always small, even minute, when com-
Fio. 81. — Volcanic vent and crater, loe Spring cluster, Utah. (After G. K.
Gilbert, Rep. Surveys West of lOOtb Meridian, 1875, p. 139.) Note eocentrio
position and oblique attitude of tbe vent, and the shelf of congealed lava. The
crater is about 600 ft. wide at its outer Up.
pared with the sections of the larger intrusive bodies. The accompany-
ing table shows the range of diameters in many typical necka.
130
IGNEOUS ROCKS AND THEIR ORIGIN
TYPICAL SIZES OF VOLCANIC NECKS
Region
Number of
necks
Range of diameten
Feet
Metcn
Cape Province, South Africa:
Namqualand
WodehouBc district .
Barkly East district
Elliot district
Herschel district
Matatiele district
Maclear
Eastern Fife, Scotland
Ajrrahire, Scotland
Swabia
New Mexico
Leucite Hills, Wyoming
16
15
20
17
22
19
15
80
60
132
"several
hundreds*'
6
135-2250
12-5280
40-680
4-1600
120-5280
30-5280
60-4000
1400, largest
described
120-500
36-1600
9-1600
1&-1200
Up to 420
36-150
Fig. 82. — Sections Hhowing orof<ion of crater charged with congealed lara.
(After H. Laspeyrc«, Das Sicbongobirgo am Khein, 1901, pp. llH-9.) D, Devonian
formation; C, trachyte tuff; T, bjiHaltic tuff; B, massive basalt. The upper
tion illustrates a maar-liko crater; the middle section, a lava lake; the lower
tion, a lava mesa loft after prolongecl erosion (Kuppe).
Volcanic Plugs. — Lacroix, Ileilprin, Hovey, Jaggar, and others
have made the "dome,*' "spine.** or ''aiguille*' of Mont PcKe so
famous, and have illustrated it so bountifully that a detailed account
of this marvellous body is not here necessary (Figs. 83 and 84). Its
upthrusting in 1902 has stimulated search for volcanic masses formed
by similar mechanism. Various authors have suggested parallels
among the extinct volcanoes.
EXTRUSIVE BODIES
131
Jaggar has observed the upthrusting of a "dome" of the Pel^an
type at Bogoslof (Fig. 85). The new dome at Tarumai, Japan, is
illustrated in Fig. 86.
In exceptional eases the highly viscous lava of relatively cool
vents has exuded in quantity sufficient to form distinct domes at the
surface, covering and notably overiapping the limits of the vents
(cumulo-volcanoes, mamelons). These domes have grown endogen-
1350 m.
Etaftp ji5<?<^
izrom.
2
4
Fio. 83. — Sections showing four stages in the recent history of Mt. Pel6e
(After A. Lacroix, La Montague Pel6e et ses eruptions, 1904, p. 121.) /, The sum-
mit before the great eruption of 1902. ^, The spine on July 31, 1902. 5, The
spine on Oct. 4, 1902. 4, the spine on Mar. 9, 1903.
mly^ as bodies of unbroken, massive lava. A classic example is
that of the trachy tic Grand Puy of Sarcoui in the Auvergne.
Lava Flows. — The usual massive components of a central volcano
are, of course, flows. Dana has described these as superfluerU, effluerd,
or interfluent J according to the mode of discharge, which is respectively
at the summit of the volcano, at a lateral fissure, or by way of sub-
surface cavities within the cone.^
Certain details of individual lava flows have received special names
' J. D. Dana, Characteristics of Volcanoes, New York| 1891| p. 2.
132
IGNEOUS ROCKS AND THEIR ORIGIN
Fio. 84. — Sketches of the Mt. Pelde spine, showing its changes. (After A.
Lacroix, La Montafoie Pel<^ et sea eruptions, 1904, pp. 124, 126, 127.) i. Not.
22, 1902. e, Nov. 25, 1902. 5. Apr. 3, 1903. 4, June 13, 1903. The respeetiTs
heights of the summit were: 1,566, 1,548, 1,593, and 1,582 meters. The elerv
tion of the crater rim (right) is 1,264 meters.
.—- /
1
Ajdl
Mc
Fio. 85. — Sketch profiles (partly diagrammatic) showing changes in the Bogo^
lof Islands in 13 months. (After T. A. Jaggar, Bull. Amer. Geog. 8oe., VoL 40,
1908.) /, September, 1906. 2, Aug. 7, 1907. 5, Oct. 16, 1907. C, CtMt
Rock (Bogoslof); G, Grewingk; 3/, Metcalf cone; Afc, McCuIlodi eoiie, a
temporary plug-dome of andesitic composition. After the rising of MeCttUocb
dome a major explosion destroyed it and part of the older Metcalf
EXTRUSIVE BODIES 133
and represent features needing explanation by a complete theory
of igneous action. The more striking items may be mentioned.
Block lava, forming the aa fields of Hawaii, lea ckeires of France,
and the Malpais of Mexico, needs no formal description in this place.
Similarly, ropy lam {pahoekoe, FladenUwa, PUUtenUwa), pillow Uwa,
Fio. S6. — Map and section of the new plug-dome (.dolUd) ftt Tanimai, J^wn.
(After H. Simotomai, Zeit. Gee. Erdkunde, Berlin, 1912, p. 433.) Top of dome
■bout 1,000 meters above sea; contour interval, 10 metere. Sc^e, 1:20,000.
and such topographic features as constructional lava scarps, lava eas-
aideg, and lava tunnels are too fatmliar to need comment.
Many congealed flows of the pahoeboe type exhibitswellingsor low
domical hills from 10 to 20 or more meters in length and a few meters
in height (Fig. 87). These may be called tumuii {SchoUeadome of
IGSEOUS ROCKS ASD THEIR ORIGIN
Fig. S7.— a tumulus in ihr flmir of \\\e Kibuntn wnk. Hawaii. (Prom a
pbotoffmph hv I Krir-lbpn'trr. piilv in H Mi»n',\lh'ii I Vulrani Attivi della Twra,
1907. p SI
Fl-i -iv— A tiimiihii in th^ fl.^vr of KiUura.
authur. 1*^ Duriiie the iiiMtommti. lupiid lava is
tcrutic of ttua voKanic (urm.
From a photopajdi by the
unl from th« cneka eharac-
EXTRUSIVE BODIES 135
Friedlaender). Hundreds are to be seen in the lava fields of Hawaii.
They are to be explained by the local hydrostatic pressure of still
fluid lava beneath the already chilled crust of a somewhat inclined
flow. The ropy crust is characteristically fractured by the pressure
and sometimes the liquid lava escapes through the fractures (Fig. 88).
The intumescence is analogous to the deformation of a laccolith's
roof.
A hornito is a gas-emitting vent on, and originating in, a lava flow.
The writer has not been able to find in volcanic literature a clean-cut
definition of this term, but, by usage, it seems to include vents which
have built up driblet cones as well as those developing minute cones
of ash or tuff on the back of the parent flow. Pacheco has recently
figured examples in the Canary Islands.*
Volcanic cones themselves hardly need detailed consideration in
a scheme for their classification as bodies of igneous rock. As with
necks, the basis for division here chosen is the relative importance of
pyroclastic and flow in each cone. Dikes, sheets, and sometimes
laccoliths, cut the extrusive rocks of most of the greater cones but
such material is usually too insignificant to affect the essential form
and contents of a cone.
Tuff cones, otherwise called cinder cones or ash cones, are, when
ideally developed, always small.
Lava cones, those composed wholly of lava flows, are relatively
rare. An appreciable proportion of pyroclastic material is generally
interbedded with the dominant massive rocks of this class of extrusive
bodies.
A pure type is the driblet cone, found in Hawaii, in Reunion, and
in other basaltic fields (Fig. 89).
Lava rings are the greater driblet cones formed by the symmetrical
upbuilding of the walls of a lava lake by the congealing of intermittent
thin overflows of the lake. In 1893 such a self-built rim was to be
seen around the Kilauean lake.^ A good example was found by the
present writer on Mount Hualalai (Hawaii) southeast of its summit.
Lava domes are the greater masses of lava, which, in the form of many
individual flows, have issued from central vents in the proper abund-
ance and proper directions to build a dome-shaped pile of lava. Appar-
ently in all cases some subordinate intercalations of tuff or breccia
are to be found in the actual domes. The world type is Mauna Loa.
Excellent examples have been described by Thoroddsen and others
» E. H. Pacheco, Mem. R. Soc. Espafiola Historia Natural., Vol. 6, 1910, p.
2.51.
*For admirable illustration see W. T. Brigham's "The Volcanoes of Kilauea
and Mauna Loa of the Island of Hawaii,'' Honolulu, 1909, Plate 50.
136 lONBOVS ROCKS AND THSIR OBtOltT
in Iceland. Such bodies are of exogenous growth and are thus con-
trasted genetically as well aa in size with the plug-domes previoodjr
mentioned.
Huge as a lava dome may be, its constituent flows are ali^y^ so
far as known, of quite moderate in<^vidual thicknesses. Ons of tha
most favorable sections for obscn'ing this fact is the gre»t scarp
(accortUng to LindKren, a fault-scarp) limiting the island of Mol<Aai
on the north. There the writer has counted more than one hundred
flows in a part of the cliff where it is about 2000 feet in height. The
average thickness of the flows is, therefore, not far from 20 feet. The
average thickness of the flows in the 2000-foot clifTs of theE
Fid. 89. — Driblet cone near the Kamakaaia rones, Hawaii. (nt>in s pbolopvpk
by II. E. Wilson, July U, 1911.)
rent in Maui is nearly similar. Nciwhero on Hawaii are the exposures
so favorable as in Maui and Molokm, but in the ctiffB studied by the
writer the flows of Hawaii arc probably lesa than 25 feet in mean
thickness. The section in the great canyons of Kauu show the aver-
age thickness of the flows to be again of that order.
It appears, then, that the Hows of the greater lava domes have aver-
age thickn(>s.scs much Ukc those ol>servcd in the greater fissure emp-
tions (lava plateaus) of the world; in lioth cases the lava is generally a
typical bfl-xnlt. (Hoe pages 119-120.)
Composite cones, ''normal" cones, or cones of the " mixed" type
arc, of course, the most abundant, including the celebrated living
f
Platb II. -The CnNS and CsATKt
I. FWBDLABNDBR, PcnRMAKK
. OF SoMUA (aiter a. Caatkilionb a
3oTMA, 1912). ScALE-l: 21,800.
EXTRUSIVE BODIES
— Fumjce-cone breached by the outHow of an obeidian htTa-cuirent,
d of Lipari. (After J. W. Judd, Volcanoee, Now York, 1881, p. 124.)
shiuD and cone clusters of the Velay, France. (After M. Boule,
ill. serv. carte glol. France, No. 28, 1892, p. 223.)
138 IQNBOUS ROCKS AND THSIR OBIOIS
volcaaoes of the Mediterranean, of Mexico, of Java, etc., and tbe
majority of bodies centrally erupted through geolopcat time (Flgl.
171, 186, 191, and Plate II).
Fio. 92. — Map rfiowinR rolation of Tertiary volcanoea (doMei) of e
Franre to rruat fractuTM. (After Le Service Gfologique d« U Furnace.)
1:2,000,000.
Fiij. »:{.— The rone rhainn of Java (Jolltd). (After R. O. M. VcrbMfc u'
K. Fonnema, Dewription Rik>l. de Java et Madoura, 1806.) Uoavj bhck doW
represent prinripal vents. Scale, 1: 10,000,000.
" Many tones formed in tlic firiit instance of acoiitt, tuff, and pumke n4
ftivc rise to streams of lava, before the vent which they Burround mki into
a alate of quiescence. In these cases, the liquid lava in the veat pra forth
EXTRUSIVE BODIES
139
ch quantities of steam that masses of froth or scoris are formed, which are
KUd and accumulate around the orifice. When the force of the explosive
)lion is exhausted, the lava rises bodily in the crater, which is more or less
mpletely filled. But, eventually, the weaker side of the crater-wall yields
BDcath the pressure of the liquid mass, and this part of the crater and cone
Smithb Id.
HotiWife RocM
"a. 94.~The neo-volcanic cone chains of Japan (dotted). (After S. Yoshiwora,
Geol. Mag., Vol. 9, 1902, p. 298.)
'i avept away before the advancing lava-stream. Examples of such 'breached
*im' abound in Auvei^e and many other volcanic districts. A beautiful
'UDipIe of a cone formed of pumice, which has been breached by the outflow
'U lava-stream of obsidian, occurs in theLipari Islands, at the Rocche Rosae.
t ia this locality which supplies the whole world with pumice"' (Fig, 90).
' J. W. Judd, Volcanoes (International Scientifie Series), New York, 1881, p. 123.
140 IGNEOUS ROCKS AND THEIR ORIGIN
Finally, colossal masses of igneous rock are represented in glubtbis
and CHAINS of cones (Figs. 91, 92, 93, 94). This grouping of units
obviously represents one of the chief genetic problems of the igneous
rocks.
Depression Forms
For the proper understanding and illustrating of the proceflses by
which the extrusive bodies are formed, it is necessary to have in mind
the chief topographic forms associated with these bodies. Althou|^
a g^eat literature concerning the negative reliefs of volcanic origin has
been developed, there is little agreement as to fundamental defini-
tions. In standard works the term ''crater" has been applied to
genetically quite distinct things, and the name "caldera" has become^
through conflicts of definition, much impaired as an aid to rigorooi
discussion. Failing the authority of a general systematic usage, the
writer has selected certain elements in existing definitions and hai
Fio. 95. — Section of the Tritriva crater, Madagascar. (Aft«r A. Bonrdariat
and H. J. Johnston-Lavis, Bull. soc. beige de gdol. etc., t. 22, 1906, p. 107.) Scab
approicimate.
combined them to form the basis of a classification which will be nted
in this l>ook.
A crater is a pit forming the normal surface expression of aeentral
vent. Though a few craters have vertical walls, the great majority
have flaring walls (Plate II). The flaring is usually continiMMiB b«t
in some craters the flare is interrupted by subordinate, vertical dib
overlooking the center of the depression. Under conditions of nonnil
activity, the flat floor of the crater (liquid, pasty, or solid lava) is
nearly equal in area to the cross-section of the neck beneath (Fig. 9S).
The flare of a crater is produced in three ways: by exidoiioii, bj
slumping, or by melting of the wall rocks. All three methods are
often combined in action. In normal activity the gases producing
explosion are chiefly containeil in the vent (neck) itsdf and the figure
typically produced by their explosion is an invertechcone. Under the
EXTRUSIVE BODIES 141
conditions, the floor of the depression can differ little in area from that
of the cross-section of the vent beneath. Subsequent slumping of
the more or less shattered walls will tend to increase the flare while
diminishing, for a time at least, the exposed area of the lava column.
Renewed explosion may clear out such debris and restore the typical
relation of floor area to the size of the vent. The terraced craters are
best displayed in the non-explosive, basaltic volcanoes like the Hawaiian
Kilauea, Mokuaweoweo (Mauna Loa), Hualalai, etc. In these cases
the flare is produced by the undermining and slow action of the lava
lakes, which are intermittently formed in each crater by flooding
through the vent (Figs. 81 and 131). During the life timeof each lake,
the solid rock of its shores is softened and weakened at the contact of
the liquid lava, causing rock falls by a kind of lateral stoping. Yet,
even in this type the crater is normally floored with liquid or solid
lava, whose area is quite moderate and not much larger than that of
the cross-section of the feeding vent.
When first formed, a crater may be fissure-like but, as a product of
the same mechanism that keeps it open (explosion, slumping, or melt-
ing), the ground plan soon becomes subcircular.
Craters, as above defined, have apertures always of small absolute
size. The visible, because eroded, rock-filled vents of central erup-
tions are seldom, if ever, more than 1000 feet in diameter; generally
the diameters are much smaller. Observations on hundreds of the
intAct explosion vents show that no explosion of the directly magmatic
gases has cleared the volcanic throats to depths greater than a few thou-
sand feet; rarely is the depth of the visible explosion pit more than
1000 feet. The conical, pure-explosion figure must, therefore, always
have moderate dimensions, including the area of the aperture. The
aperture is regularly enlarged by slumping but, since the angle of rest
for tuffaceous material is nearly thirty degrees, this enlargement is
quite limited.
Again, the flare produced by undermining and melting by the lava
lake m a crater must be restricted by the size of the lake. On account
of the high rate of heat radiation from hot lava, only a small lake is
possible at any crater which communicates with the earth's interior
through ordinary necks. Limited as the flare so caused must be, it
cannot be much enlarged by slumping subsequent to the freezing of the
lake or to its diminution of surface through withdrawal of the lava into
the vent.
There are, thus, good grounds for including in the conception of a
true crater the idea of relatively small size.
Lava pits are those craters which are visibly floored with massive
lava, either liquid or solid. The famous pits of Hawaii are illustrated
IGNEOUS ROCKS AND THEIR ORIGIN
Fia. 06. — DiMant view of a small pit crater in th« Puna district, Ha
(From a photograph by the author, 1909.) The welMike crater ia about 2t
in iltamctcr.
Fio. 97. — Map and HFrtioniif .\ni.stcnlani iHlaix) volcano, showinf adT«ntive<
and cratcra. (After C. V^lain, Miwion de lllo Saint Paul, 1880, PL 2^
EXTRUSIVE BODIES
Fis. 98. — The nested eriitera of Vcaiiviua, from a. sketch by Sir W. Hamilton.
(Campi Phlegraji, 1799.)
Pio. 99. — Nested craters of Etna in the early part of the 19th ceatury. (After
A'. 8. TOD Waltershausen, Der Mina, Vol. 2, 1880, p. 304.) Three stages of the
•ummit crater: 1, years 1804-5; 2, years 1805-9; 3, years 181t>-16.
144 laXEOLS HOCKS A\D THEIR ORIGIN
in Fig!?. 90 and 142. These craters are specially instructive as they
have clearly \K'vn formed in an older lava plateau, not by explosion but
\>y meltinj^ perforation.
Maars are relatively flat-floored explosion craters at vents which are
(»ither coneless or else provided \iith inconspicuous cones. In this
class come most of the famous Vulkan-Embr>'onen, described by
Branco in Swabia/ (Fig. 144).
Blow-holes are the minute craters formed on the surfaces of thick
lava flows. They are often visible on driblet cones.
Advent iv€ (parasitic or lateral) craters are those opened on the flanks
of great cones (Fig. 97).
Nested craters. — Sometimes central vents show the phenomenon
of crater in crater. The smaller crater in each case has evidently been
produced by a temix)rar>' or permanent restriction of the amount of
heat transferred to the vent from the earth's interior, whereby the
cross-section of the feeding pipe has l>een diminished. (For illustra-
tions see Figs. 98 and 99.)^
Calderas. — Few terms in igneous geolog>' have occasioned more
diversity of definition than the name **ealdera.'' All writers on the
caldera agree as to its amphitheatral or circus shape and as to the
necessity of considerable size. Beyond those points the various sug-
gestions concerning the proi)er meaning of the word sharply diverge.
(1) Some authors have used the term as signifying merely a
gigantic <*xpIosion crater.
(2) Others add to that genetic feature the necessity of a large
lateral ofK'ning in x\w wall of the depression, as in the famous Caldera
of La Palina of the Canary Islands. This lateral opening should, by
this (iefiiiitinn, pass into a **barranco" or deep ravine running outward
from the volcanic center.
u3) A third group of writers have followed Dutton in defining a
caldera as a down-faulted, steep-walled depression formed as a result
of volcanic action.
i4) Still others have considered a caldera as chiefly formed by
erosion, which has preatly enlarged a normal crater, thus recognising
the co-ojH'ratii^n of two genetic conditions.
(5) Finally, (lagel has recently made an exhaustive study of La
^ W. Branco, Schwabon's 125 Vijlkan-Krnbr>*oncn und deren tufferfliQte
Au>hnirh>rnhren -<laiJ >n"6sste Clobiot chemalijccr Maare auf der Erde, TObingm,
IslU. For illustrations <if the Eifd niaars s<h» R.Lopsiua, Geokxgie von Detitacb-
lan.l. Stuttgart, br Ttil. 1SS7-92. p. ;W3.
' Also H. n. M. \ crhe^k an<l K. Frnnema, Description g^logique de Java ft
Ma«ioura. Amsterdam, Athu* HjjlaRo IS, Fijf. 71; H. .\bich, ElrUutemde Abbildon-
iivn i:foln^jisch»T Krs<h<*iniinpen beohachtet am Vesuv und ^tna, Berlin^ 1837,
Plates 1 and II ^Vcijuvius in July, 1834, and Etna in June, 1834.)
EXTRUSIVE BODIES 145
Palma caldera and has defined the word without any necessary
reference to volcanic action. He thinks it advisable to regard these
circus-shaped depressions as essentially due to head-water erosion
of streams. He even placed in the class of calderas certain amphi-
theaters which have been eroded out of horizontal sandstones in New
South Wales.
Much of the diflSculty in reaching a common understanding in
this matter is due to the somewhat peculiar conditions at the " Caldera"
of La Palma. So many different origins have been assigned to it that
an unprejudiced person may feel no compulsion in settling on the es-
sential elements, either of form or of genesis, in this one case. Lyell,
himself, who made this depression famous during his discussion of
von Buch's elevation theory of volcanoes, could not decide as to its
mode of origin, but he called the great pit of Mt. Somma (the atrio) a
caldera, while recognizing that it was formed by explosion. In 1860
Hartung described the *'Caldeira das Sete Cidades,'' the "Caldeira
de Santa Barbara," and three other named "caldeiras" of the Azores,
as due to repeated volcanic explosions.^ These and other writers
helped to establish the tradition that a caldera is to be regarded as a
gigantic explosion form, either with or without a lateral opening or
barranco.
It is also true that the Portuguese use the word "caldeira," not only
for depression of explosion, but also with the meaning of "boiling
springs," as in "Caldeiras das Furnas" of the Azores (Hartung).
Thus, as a common noun, the Latin peoples use the word in totally
different senses; always a "caldron" or "kettle," but now almost
literally, and again in a remotely figurative sense. Accordingly,
there is no antecedent, formal objection to introducing the word into
technical geology with a figurative meaning. The one indispensable
condition is that it shall have a definite meaning. Now that there is
pressing need for a term for the greater explosion depressions in vol-
canic regions, geology will certainly follow the line of least resistance
in adhering to the definition implied in the use of Lyell, Hartung, and
other of the older writers. The "caldera" of Button may be called
a "volcanic sink," as noted below. The erosion "Kessel"of Gagel,
if named specifically from a locality type, should be referred to a type
much less subject to difference of interpretation than the Caldera of
La Palma.
Regarding calderas as explosion forms, it remains to distinguish
them from ordinary explosion craters. As the word "caldera" has
been used, it has almost always referred to very large depressions. If
these are correctly interpreted as due to major explosions, they cannot
^ G. Hartung, Die Azoren, Leipzig, 1860, pp. 311-312.
IGNEOUS ROCKS ASD THEIR ORIGIN
Flit. llN).~Th<- TiU-uwi-rn rift and Itoluniahuntt ralilnra, N<Mtr Zotkad. (After
A. P. W. ThoDitw, whose iimj) b reproduced by J. M. Bell, Geoc. Jour., 1906.)
Y\r.. im. 'M:ir> »r lli'' Ci.iaiini •>( thr ^•■u- TMimW. Sun Mipip] Islatid, V
<.\rtiT (1. Iliirlunic. Uk- Aiiircn, 1M)(). iitliu,) Srvtrnl yuung mtcra uc nn
tbe calden.
EXTRUSIVE BODIES
147
safely be described as "craters" in the sense above given to that word.
If the actu^ly exposed " necks " of the world indicate the maximum size
of central conduits, the vents beneath calderas must have cross-sections
much smaller in area than the floors of the corresponding great depres-
sioDS. The writer is, in fact, incHned to make this the criterion for
distinguishing explosion craters from calderas. In each of the latter
the area of the floor is many times greater than the cross-section of
the magmatic column exposed to the air by the explosion.
Illustrations of simple calderas, in the sense here used, are given in
Figs. 100 and 101.
Fia. 102. — Nested calderas at the Maeaya volcanoes, Nicaragua. (After
K, TOD Seebach, Abhand. k. Gee. Wiss. Gottingcn, Phys. Kl., Vol. 38, 1892, p. 58
uid Taf. 9.) The outer escarpment is the eroded lip of the older caldera. Within
it ia seen part of the youngf>r caldera ring, within which are nested the active
Nindiri-Maaaya conee.
If the genesis of these depressions has been correctly stated, there
can be little doubt as to the necessity of distinguishing them from true
explosion craters, for it is scarcely credible that such relatively vast
hollows represent the figures due merely to the explosion of gases
within columns of liquid magma. An extreme case is represented in
the sumnut caldera of Bandai-San, in Japan (1888), which is said to
have beenformed by aprodigiousexplosion, without theexposureofany
148
IGNEOUS ROCKS AND THEIR ORIGIN
liquid magma whatever. ' The proposed dietioction between " craters"
and "caldera»" thus uuggesU a genetic distinctioQ between the two
clasKos of forms. It in hardly necessary to add that the asaignnient of
a small explosion depression to the class of craters or to the clan of
calderas, as hero defined, may occasionally be very difficult, if not
impossibie ; but such a trouble cannot outweigh the advantage of hav-
ng systematic designations for the types which can be de6mt«ly
Fio. 103.— The sunken ealdcra of Santorin. (AttCT F. Fouqu^, Skntcna et
KPS eruptions, 1S79; his niup hiTP coitii^l from a copy in Neumajr's Erdgeachiehle,
pull, by the BiblioBraphisrh<-s In-stilul.) Scale, 1 : 150,000.
assigned. This difficulty of tran.-«itional forms falls to the common lot
of nearly all students of natural olijecta.
.VfNfff/ cahlerna. — The conditions leading to the formation of a
c.ihiera may recur on a smaller scale at the same volcanic center,
so that a second depression of this class is developed within the per-
imeter of the first. The process may be conceived as repeated Mveral
> S. Scluy& uid J. Kikuchi, Jour. Coil. Science, Tokio, 188B, p. 106.
EXTRUSIVE BODIES
ISO
IGNEOUS ROCKS AS'D THEIR ORIGIN
times. In such co^es the form U conveniently (lescnbed under the
name, concentric caidcraij, or more generally, nested caldertu. Several
examples of such compojtite forms have \u^-n described. {See Fig. 102.)
Sunken calderan. — In a coasJdcrahle number of cases subadence
is reported to have followed cahiera explosion. The resulting depres-
sions may be called sunken c'lWcran (FlRa. 103 and 104).
Volcanic Siiik8.^A.s already sujtgestetl, the "cal<iera problem"
can be partly aolveil by a general ap-eement to recognize by actual
names the sharp frenetic distinction between the " caldcras" of Hartung
in the Azores and the" calderas" of Duttonin Hawaii. A considerable
I vol<-anir Kink, in which ia (hr ts-
.M»i>m lie nic Saint Paul, ISSO, W.
A.
T\a. 105.— View of Hip Kmlon of UAiiii.m,
tinct cone, Ic pitun B..ry. I.VtIcr ('. WUiiii, M
10.) A Bmull udvcntivi- i-nilir in the fiiri'itrcjui
number of .American authors, includitiK members of the Uiuted States
deologicul Survey and writers of tfxt-books, have adopted Dutton's
definition. Of the Ilawniian ■caldcni.s" he wrote: "Con.-udered with
reference to their origin thf I'vidt'iicc is conclusive that they were
formed by a dropping of a bluck of tlic mountain cru.'«t which once cov-
ered a reservoir of lava, this reservoir beinK tapped and drained by
eruptions occurrinR at niuili lower levels."' Diller hius followed Dut-
tonin plaeinEtliefaniiiusdejircssion due to en(:ulfment,ut Crater Lake.
Oregon, among the ealderas,-
' C. E. Dmton, 4th .\rr. lti>ii-. l'. S. Clwil. Survey, 1884, p. 105.
» J. S. Diller, Prof. Paper No. 3, U. S. Geol. Survey, 1902, p. 46.
EXTRUSIVE BODIES
^^^^^fiffi^J^N f OiC/ N(.
HOUSE
^^
"^
K U£A IKl
!■ / r-~s~
Fio 106 —Map of the Kilauea sink Hawaii in 1886 (After the Government
map by t S Dodge ) Scale 1 57 600
Fio, 107.— Volcanic sink at top of Tengger volcano, Java. (After R. D. M.
Verbeek, Description g^l. de Java and Madoura, Atlas, sheets CS and C9.)
152
IGNEOUS ROCKS AND THEIR ORIGIN
It is surely better , in the interests of productive science, to retain
the old, established meaning for the word "caldera." The volcanic
basins of engulfment, or down-faulting (each with floor area many
2780
FiQ. 108. — ^^tion of the sink shown in Fig. 107. (Same ref., Fig. 8 in atlfts).
S, Sunken area underlain by ash, tufT, and lava; T, tuff beds of original cone.
Faults shown. Heights in meters. Scale, 1:05,000.
times greater than the cross-section of the associated vent) have been
called '* volcanic sinks" and actual practice, both in writing and teach-
ing, shows that this simple term fills the need and avoids the logical
0
? Mil.
0
y^^^^'i^
summit]
r m
13.675 a|
w
Fio. 100. — Xested sinks at Mokuaweoweo, summit of Mauna Loa,
(From Aloxiindor's map of IS85.) Figures show depths below the upper rim, in
foot.
difficulty mentioned. * Examples of simple sinks are illustrated in Figs.
105, lOf). 107, and 108.^
' U. A. Daly, Proc. Amer. Acad. Arts and Sciences, Vol. 47, 1911, p. 110.
> Pof*sibly the Tengger depression is better described as a sunken cmklenL
EXTRUSIVE BODIES 153
Quite recently, du Toit has described a sink so greatly eroded as to
have lost its character as a topographic depression. This is the case
with the Modder Fontein volcano of the Stormbergen, Cape Province,
South Africa. Its tuffs and agglomerates dip inward because of sub-
sidence of a normal volcanic cone during or after the growth of the
pyroclastic deposit.^
Nested sinks, — More or less concentric sinks have been mapped in
Hawaii and elsewhere (Fig. 109). Verbeek has specially emphasized
the occurrence of both simple and nested sinks in Java.^
Volcanic Rents. — The great gaping depression at the summit of
Haleakala in Maui, Hawaiian Islands, is commonly described as a
** crater," but Dutton long ago pointed out the fallacy in so doing.
He prefers to call this form also a caldera, remarking that it is '* strictly
homologous'' with the main depression (sink) at Kilauea.'
During a visit to the summit in 1909 the writer failed to find
evidence of this homology. General circumferential faulting, which is
topographically so clear at Kilauea, is not evident on the walls of the
vast depression on Haleakala. On this problem Dana writes:
•
"In my 'Exploring Expedition Report' I suggest that the mountain was
fissured across along the lines of the two discharge-ways, and the eastern
block shoved off a mile or two. But a subsidence of the masses that occupied
them into caverns below, leaving the walls as fault planes, may be more
probable. The abyss which received them in this case had been prepared
during a long period of undermining through ejections. Still there is some
reason to believe in the grander view of a subsidence of the whole eastern
block, across the cross-fracturing. The island, as is seen on the map, is
abruptly narrowed (instead of widened) at the spots where the Koolau and
Kaupo streams reach the sea; and the part to the eastward is small, as if
narrowed by such a subsidence. Moreover, the mean height of the eastern
crater-wall is lower than that of the opposite or western by five hundred to
a thousand feet. A subsidence of a thousand feet increasing in amount to the
eastward would account for the narrowing and for the very short eastern
radius of the eccentric volcano. The question merits investigation."*
The present writer is inclined to believe in Dana's first interpreta-
tion of the depression. Brigham long since spoke of it as a "rent,"
and it seems safer so to designate it rather than as a crater (popular
usage) or a caldera. No other example appears to have been recorded.
* A. L. du Toit, 16th Ann. Rep. Geol. Comm., Cape of Good Hope, 1912,
p. 132.
*R. D. M. Verbeek and R. Fennema, Description g^ologique de Java et Ma-
^oura, Amsterdam, 1896.
' C. E. Dutton, 4th Ann. Rep., U. S. Geol. Survey, 1884, p. 106.
* J. D. Dana, Characteristics of Volcanoes, New York, 1891, pp. 277-278.
PART II
CHAPTER VIII
COSMICAL ASPECTS
Principal Source of Magmatic Heat. — In a brief, necessarily general
form, the material of the igneous-rock problem has been laid before the
reader. The facts so far presented indicate the chemical and mineral-
ogical diversity of the rock species; their distribution and relative
abundance; and the endless variation in the sizes, forms, and relations
of igneous bodies. Throughout the preceding chapters the attempt has
been made to admit only such descriptions and classifications as are
direct expressions of objective facts, though here and there a choice of
rival interpretations of certain facts has been compelled before they
could be succinctly stated.
In now turning definitely to the explanatory side of the subject,
we immediately encounter its most difficult and elusive phase. What
is the origin of magmatic heat? The high temperatures actually
measurable at volcanic vents or inferred from the character of the
intrusive bodies represent a principal part of the igneous-rock problem.
Its systematic solution must be founded on a sound cosmogony. In
this respect geology must still wait on astronomy for the final word.
Nevertheless, it is well to note that the sure results of astrophysics
do not yet negative the geologist's traditional view as to a chief source
of the earth's thermal energy. That view was originally founded on
the Kant-Herschel-Laplace nebular hypothesis, which involved an
earth once molten at the surface, then encrusted through radiation,
and still intensely hot within. This conception was not essentially
disturbed by Lockyer's meteoritic hypothesis. It must still be sub-
jected to scrutiny after the new planetesimal hypothesis of Chamberlin
and Moulton has been more fully tested. At present the latter con-
ception of the origin of the solar system seems to be sounder than the
older nebular hypothesis which it is intended to supplant; but a
detwled application to the thermal problem of any one of the planets is
an extremely hazardous imdertaking. More concretely stated, the
planetesimal hypothesis does not yet give any certain indication of the
former maximum temperature of the earth's outermost shell. It is
with that part of the planet that the petrologist is most concerned, and
12 155
156 IGSEOVS ROCKS AND THEIR ORIGIN
only the relation of the planetesimal hyiK)tbesia to the comparatively
siiperfirial temperatures need lx» discussed in this place.
Planetesimal H3rpothesis in Relation to the Heat Problem. — C'ham-
iMTJiirs eoiieeption of the earth's origin may I)e summarised in hi8
own words.
" Arrordiiifj; to the planetesimal theory, the core of tho earth is made up
of planctosiinal mat tor, ix'rhaps corres|M>ndinK somewhat in compoeition to
meteorites. After a^Kref^atioii, the planetesimal matter was probably re-
erystailizHJ under the influenee of the heat and pressure which the agxreipi-
tion involved, the resulting; roek being essentially igneous in its nature. Out-
side the central core there should therefore he (1) a thick zone made up larfcdy
of planet(^<inlal matter, hut j)artly of igneous rocks erupted from below.
and partly of s(>dimentary rocks. The planetesimal matter is assumed to pre-
dominat<> in the lower and major part of this zone; igneous rock, erupti^'e and
irrupt ive, is assunu^l to have a somewhat irregular distribution within it;
while th<' s<Mlimentary rock increasi^ in importance above, though remaining
throutrhout a very subordinate constituent. This zone records the gro^h
of the earth from the initiation of volcanic and atmospheric processes to the
close of the period <»f notable growth by accretion. The central core and this
thick zone about it represent the Formative con. (2> The next zone, probably
a relatively thin one, is assume<i to l>e ma<ie up dominantly of extrusive
igneous roeks. with whieh would I>e ass(K'iat(Hi subordinate amounts of sedi*
UK'ntary matter and matter gathered in from .<ij)ace. This zone representfl
the Kxtru-ivi' eon. i'.i) ()ut>irle it lies the superficial zone in which sedi-
mentary roeks prt'dominatc. though associated with not a little rock of igneous
origin. The first two zones outside the core are assumed to be universal,
while the outermost z(»ne, being com]>os(»<i primarily of material washed
down from th<' land and de|M)>ite^l in the sea. fails to encircle the globe.'*'
ChanilK»rlin further deduces from the hypothesis that the earth's
surface* was cwA enough to support a water ocean even at the stages
when the planet was still much smaller than now; and that the surface
always thereafter retained temperatun* of the same order of maf^tude.'
In other words, the existing statement of the planetesimal hypothe-
sis, as applied to the earth's history, runs counter to the idea that the
external shell of tlu' earth, when of approximately its present siie, has
passeil through a jwriod of general fluidity.
1. Chamberlin grants the possibility that the earth in the early
"nuch»ar" stage of its development wa** hot. The causes of this high
t<*mixTature are: the (piasi-gaseous mcale of condensation in the prim-
itive nebular knot : the heat of central eompres.*5ion; and the heat de%'el-
»T. i\ C1i:iin!MTlin an<I R. D. Salisbury, (Joology, New York, Vol. 2, 1906,
pp. I'M 13.').
u)p. cit.. pp. ias-110.]
COSMIC AL ASPECTS 157
oped in molecular rearrangement.^ He assigns the chief portion of
internal heat to compression. During this and later stages of plane-
tary growth the speed of heat generation, from all causes, is unknown,
but the authors of the planetesimal hypothesis assume a comparatively
rapid solidification at the surface of the growing earth. So far, how-
ever, no proof of this vital assumption has been presented. If the
speed of heat generation reached a certain value, the nucleus would
be fluid and incandescent at the surface. In such a condition, radiation
would quickly chill the superficial layer and tend to crustify the
body. Radiation from a rock surface into free space is about 400 times
more rapid at 1100° C. than at 20° C. A temperature near 20° C.
would soon characterize the surface of a very thin crust of rock, which
is an efficient blanket. As heat is generated by internal compression
and by molecular rearrangements, it would tend either to melt the
crust intermittently or to keep the crust very thin. The actual alter-
native would be determined by the ratio of the heat generated to the
heat lost in a unit of time. The analogy of existing volcanoes suggests
that intermittent fusion of the crust would be the more probable result.
No facts have yet been adduced to show that this process would
be necessarily restricted to an early stage of the planetary growth.
The assumptions of the planetesimal hypothesis do not compel belief
in a solid earth even when accretion had first brought the planet
approximately to the present size. Only when the (constantly vary^
ing) ratio of heat generation to heat loss throughout the epoch of accre-
tion has been determined is it possible to discuss adequately the physi-
cal state of the earth's outermost shell at any stage of the planetary
development.
2. Chamberlin postulates a transfer of internal heat toward the
surface by the extrusion of molten tongues of the more fusible material
from the central region, through the otherwise solid material of the
earth. These tongues have fluxing power. No indication is given,
in the existing statements of the planetesimal hypothesis, as to the
speed of this heat transfer. It might be rapid enough to cause nearly
or quite perfect fusion of all, or of large areas, of the surface shell of
the planet.
3. With certain assumptions Lunn has calculated the earth's
mtemal temperatures expected on the planetesimal hypothesis.' The
central temperature was found to be practically 20,000** C. Halfway
to the surface the temperature would be 12,250° C. Since nothing is
» T. C. Chamberlin and R. D. Salisbury, Geology, Vol. 2, New York, 1906,
p. 100.
* A. C. Lunn, in Chamberlin and Salisbury's Geology, VoL 1, 2d ed., New York,
1906, p. 564.
158 IGNEOUS ROCKS AND THEIR ORIGIN
known about the behavior of matter at such temperatures and at the
pressures reigning in the deeper region of the earth, it is impossible to
assert that the gro\inng body would be chiefly in the solid state. Arrhe-
nius holds that at such temperatures as those quoted from Lunn all
known substances would l>e in the critical state. If so, there is a clear
probability that the heterogeneous materials of the growing planet
would have Ixeconie stratified ; the slow rise of the lighter substances and
the slow sinking of the heavier must take place unless they were infi-
nitely viscous. Such convective overturn, bringing upward the very
hot material at the center, might be rapid enough to cause comfdete
fusion of the superficial shell. Here again the existing statement of
the planeter<inial hyi)othesis is not complete enough for the needs of
igneous geology.
4. Nor does that statement take proper account of the low densities
of the outer planets. The average den.sities of the earth, Jupiter,
Saturn, Uranus, and Neptune are respectively about 5.5, 1.3, 0.7
1.2, and 1.1, if water l>e assumed to have unit density. The masses
of Jupiter, Saturn, Uranus, and Neptune are respectively about 316,
95, 15, and 17 times the mass of the earth. At the surface of each of
these planets the force of gravitation is, respectively, about 2.65|L18,
0.91, and 0.88, if gravity at the earth's surface be taken as unity.
Authorities in cosmical physics generally attribute the low densities
of the four greater planets to very high temperatures, and Jupiter is
often spoken of as a ''s<>mi-sun.'' The only alternative to assuming
high temp<*rature for each of the four greater planets is to assume that
it is entirely compos(»d of one or more very light elements; this view is
taken by C'hani)>erlin.* Such a highly sp<*cialized constitution is a de-
duction from the planetesimal hyi>othesis in its existing form of
statement, which implies that the thermal evolution of the earth was
not essentially different from that of the greater planets. Since, on
the other hand, high authorities in astronomy prefer to credit very
high temperatures for the surfaces of Jupiter and Saturn, a heavy
burden of pr(K)f rests on anyone who claims a continuously solid sur-
face shell for the earth during the latter half of the period of its growth
by planet€»simal accretion. The sun is known to be largely composed
of heavy elements (iron, calcium, barium, titanium, platinum, and
other metals); gravity at the sun's surface is nearly 28 times that at
the earth's surface, and the internal pressures of the sun are incom-
parably greater than those of the earth. Nevertheless, the mean
density of the sun (1.39) is only slightly above that of water. The
sun's low density is obviously explained by his high temperature
» T. C. Chanilwrlin and K. D. Saliabur>', OeolofO', Vol. 2, New York, 1006.
p. 58.
COSMIC AL ASPECTS 159
rather than by his chemical constitution. This fact must be con-
sidered in the problem of the outer planets.
Again, the planetesimal hypothesis as stated and applied to earth
history, does not clearly reconcile the fact that the sun has a surface
temperature of approximately 6000® C, with the assumption that the
earth remained cool enough to support a water ocean during the latter
half of its period of accretion. It is, of course, true that the condensa-
tion of the enormously greater mass represented in the sun must pro-
duce much higher temperature than that due to the condensation of
the earth mass; but, on the other hand, the total loss of heat through
geological time has been many millions of millions of times greater.
As the mechanism of heat production is assumed to be the same for
earth and sun, one may well question the deduction that the earth,
in its later history of accretion, did not pass through a stage of incan-
descence at the surface. Even the moon, with its relatively feeble
capacity of heat production by self-compression, seems to have been
in a magmatic state close to its surface, if not actually at its surface.
5. Finally, the ocean ought to be more salty than it actually is if
Chamberlin's view is correct, that the water ocean has existed from
the time when the growing earth had about half its present volume.^
Computations of the age of the ocean from salinity data are sufficiently
reliable to cause serious embarrassment to the geological student, who
may feel compelled to adjust all the myriad recorded events of post-
Eeewatin time to fit the 80,000,000 years deduced by the calculation.
If, in addition to the salt supply poured into the ocean during Keewatin
and later time, the ocean were to have received the highly soluble
chlorides during the indefinitely long epoch implied in doubling the
earth's volume by accretion, it is certain that the existing ocean must
be a much stronger brine than it is at present.
We may conclude that the planetesimal-nebular hypothesis, like
the older gas-nebular and meteoritic hypotheses, does not forbid belief
in: (a) a former molten stage for the earth's external shell; (b) the
density stratification of this planet; (c) a fairly uniform composition
for the surface shell; (d) general magmatic temperatures not more than
a few miles below the surface, throughout geological time. Important
as the planetesimal hypothesis is in cosmogony, it does not seem
to affect the traditional view of geologist^ as to the thermal con-
dition of the globe. The first and fourth of the postulates above listed
have formed the basis of most modern theories of vulcanism and of
igneous action in general. The second and third postulates are less
universally made but they are almost certainly corollaries of the first-
^ T. C. Chamberlin and R. D. Salisbury, Geology, Vol. 2, New York, 1906,
p. 109.
160 IGNEOUS ROCKS AND THEIR ORIGIN
mentioned postulate as to a former molten condition of the earth's
external shell.
The discussion of these assumptions will be continued here only
with reference to a few relevant studies of recent date.
Density Stratification of the Earth. — Daubrde and other geologista
have expressed the view that the meteorites represent, in a general
way, the average stuff constituting the earth body. Though the
total weight of the known meteorites is only a few tons, their great
numlx^r and their viide distribution over the earth have suggested that
they represc*nt qualitatively a sample of the cosmic material composing
at least the inner part of the solar system. Farrington's new average
for the more trustworthy chemical analyses, including 318 analyser
of ''iron'' meteorites and 125 analyses of stony meteorites, has
prompted him to re-state this hypothesis with approval.^ He writes:
"The large proportion of iron in the constitution of the earth indicated
by meteorites is in accord with the earth's density, rigidity, and magnetic
proportions. Assuming the density of the rocks of the earth's crust to be
2.8, which may be too high, and combining with it metal of the density of
7.8, which is an average of the density of iron meteorites, it will be found that
77.58 per cent, of metal will be required to obtain a density of 5.57, that of
the earth as a whole. This is very nearly that of the sum of the metals in the
above result after eliminating the proportions present as oxidss. Such a
proportion of iron would seem to be in accord, as has been stated, with the
earth's rigidity and magnetic properties."
The percentage of metal so calculated for the earth corresponds to
a globe with a radius less than 325 miles (520 km.) shorter than the
average radius of the earth.
This estimate is subjec*t to various corrections. A leading one will
allow for the fact that the metallic meteorites attract attention more
readily than the stony meteorites. In fact, only of late years have the
tektites l)een recognized as of extra-terrestrial origin. For this reaaon
the metallic meteorites are doubtless considerably over-emphaaiied
in the world collections. If a complete collection of meteorites were
made and analyzed, its average composition would almost certunly
l>e more salic than the average shown in Farrington's table. It
may l)e that the non-metallic portion would have a ratio of the same
order as that of the outermost 1500-kilometer shell of the earth
to the mass of the whole planet. For quite independent reasons
foumled on seismic study, Wiechert, Oldham, and others have deduced
*0. C. Farrinftton. Publiration No. 151, Field Museum of Natural Hisleryi
Chicago, 1011, p. 214.
COSMIC AL ASPECTS 161
1500 kilometers as the approximate thickness of the earth's silicate
shell, the great central mass being metallic.^
The facts of terrestrial density, the facts of seismology, and the
facts derived from meteoritic studies thus agree in suggesting a coarse
stratification for the earth as a whole, with the silicate matter chiefly
or wholly confined to a comparatively thin external shell. This view
has been adopted by Suess in the last volume of his great work.^
According to the older nebular and meteoritic hypotheses, this
general stratification is an expected feature of the terrestrial globe.
The explanation is comparatively simple if it be granted that the whole
planet was once fluid, even though highly viscous at the core. Immis-
cibility of metal and silicate, together with the influence of differential
density, must ultimately cause the metallic core to separate from the
silicate shell. This process, so commonly credited as the essential
one, has an obvious parallel in the gravitative sorting of the material
in an assaying crucible.
The planetesimal hypothesis, as stated by Chamberlin, recog-
nizes the formation of the outer silicated shell by repeated extrusion
of molten "tongues'' of the appropriate composition. As already
noted, this suggested mechanism is not only complicated; it fails to
allow for the high probability that all the important substances of the
earth's interior, if subjected to the temperatures actually calculated
on the planetesimal hypothesis, must be in the critical state. More-
over, the hypothesis fails to account for the failure of metallic alloys
with low fusion points among the known extrusive '* tongues."
Yet it is important to note that even the authors of the new plan-
etesimal hypothesis are also inclined to share the orthodox view that
the earth is coarsely stratified.
Whether the external silicated shell was all simultaneously molten
or has been formed piecemeal by successive eruptions of large ig-
neous "tongues" from the deep interior, gravity must directly aflfect the
liquid magmas involved. The important question arises concerning
the degree and kind of density stratification to be expected in the sili-
*E. Wiechert, Nachrichten der Gcscllschaft ftir Wissenschaften, Gdttingen,
1897, p. 221; and Deutsche Rundschau, 1907, p. 376. R. D. Oldham, Quart.
Jour. Geol. See., 1907, p. 347. The idea that the high pressures of the earth's
interior might suffice to give even silicate material the demonstrated high density
there existing is one often expressed but it savors of the mystical. Modem high-
preawire experiments have already suggested the improbability of this hypothesis
eren if the earth had room temperature at its center. Allowing for the actually
high internal temperatures, it seems clear that the reigning pressure at the globe's
center could not give the required density to ordinary rock matter.
' £. Suess, Das Antlitz der Erde, Bd. 3, zweite H&lfte, Vienna and Leipiig,
1909, p. 625.
162 IGNEOUS ROCKS AND THEIR ORIGIN
catcd shell itself. In C'hupter XII, page 230, will be found a collataon
of the many observations tending to show how generally even small
igneous l>odies are stratified according to density. For the present
these evidences of a general structure in the earth's interior will be
held in reserve while other facts of similar import will be discussed.
Earth's Sedimentary Shell. — The visible sedimentary shell of the
earth is relatively thin, averaging probably less than 1/2 mile in thick-
ness. It is a ragged, discontinuous ''pellicle'' on the earth.' In some
narrow geosyndinal belts it shows local thicknesses of 40,000 feet or
more. The visible material of the sedments has been largely^ if not
principally, derived from the pre-Cambrian granites and orthogneisse#.
The existence of this sedimentary shell of the earth has a very impor-
tant relation to the origin of the less abundant igneous-rock species.
(See Chapters XVI to XX.)
Earth's Acid (Granitic) Shell. — Beneath the sedimentary rocks
on every continent, the igneous complex of early pre-C'ambrian age
has been found and geologists are now disposed to regard this terrane
as composing the larger part of each continental plateau, probably
underlying at least one-third of the whole earth's surface. It is not
known to exist in the middle part of the Pacific basin, and perhaps the
terrane is actually absent in that region as well as in parts of the Indian
and South Atlantic basins.
The basement complex is typically represented in the Canadian
and Fennoscandian shields. Already enough field work has been
done to warrant a statement of the average composition of the terrane
in these great outcrops; it is that of common granite.
The depth of the pre-( 'ambrian complex is, of course, unknown.
From its uniformity in constitution at the various levek laid bare by
Cambrian and later denudation it seems likely that the average depth
is to l)e measured in miles rather than merely in thousands of feet.
We may conclude that the sediments of the continents at least
rest on a general terrane averaging a granite in composition. Though
this compl(*x may not extend under the whole of the great ocean-hasin.*,
it is fair to call it an earth shell.
Since most of the acid shc»ll s<»<»ms to 1h» of intrusive character, we
must now inquire as to whether the existing chemical composition of
the pre-Cambrian batholiths is primary. Can this granitic material
1k» explained as the result of the wholesale fusion of sciliments derived
from an ant c*c<m lent general igneous tyi>e of different composition?
Such an explanation for th(» world granites has l)een offered by a
numlxT of the older geol(»gists. It is now not seriously entertained
' St'e F. W. Clarke, T\\v Data of GeochemiHtry, 2«l edition, Bull. 491, U. 8. GeoL
Survey, 1911, p. 30.
COSMIC AL ASPECTS 163
for most of the post-Cambrian batholiths, but the evidence against
this explanation for the much more voluminous pre-Cambrian bodies
is not so directly manifest. Yet reflection will go far to convince the
reader that it fails also in this case.
The only probable rock-type which through weathering and erosion
could be conceived to form the required amount of sediments is either
basalt or andesite. Of these two, basalt has, much more clearly than
andesite, the known volume and geological relations of a general,
primary earth magma. Yet the derivation of the required sediments
by leaching and washing would be less difficult to conceive in the case
of andesite; and, in the interests of safe reasoning, we will assume that
the primary matter from which the imagined pre-Cambrian sediments
were derived was andesite. It will be further assumed that this ande-
sitic land mass would have an average composition like that of the
average andesite now exposed on the earth. This composition is
closely indicated in the calculated average of eighty-seven chemical
analyses. (See Col. 46 of Table II.) In Col. i of Table II the esti-
mated average analysis of the pre-Cambrian granites (and ortho-
gneisses) is stated. The soda percentages are seen to be nearly iden-
tical in the two averages, while potash is about twice as abundant in
the granite as in the andesite. For the present argument it will suffice
to examine the imagined process by which the potash is increased so
as to reach the proportion in the granite.
The percentage of sodium in the average andesite is 2.66; the
potassium percentage is 1.69. The corresponding percentages for
the granite are 2.40 and 3.74. Clarke's average for the river-waters
of the present day shows the ratio of sodium to potassium to be 5.79
to 2.12. It is safe to assume the ratio for the average pre-Cambrian
rivers to be not less than 4 to 1. If, then, the imagined andesitic land
lost "to the ocean 4 parts of its sodium, it would lose at least 1 part
of its potassium.
Let us further assume that the total area of the pre-Cambrian gran-
itic terrane to be derived originally covered only 50,000,000 square
miles and that it was only 2 miles in depth — certainly low estimates
in each instance. To produce this volume (100,000,000 cubic miles)
of sediments, later to be metamorphosed into granite, the weathering
of at least 250,000,000 cubic miles of the primary andesite would be
required. During that prodigious denudation all the sodium of at
least 150,000,000 cubic miles of the andesite must have gone into the
ocean. There it would remain in solution if the ruling conditions were
then the same as in the present ocean. Calculation shows that the
sodium contained in such an amount of average andesite is about
three times the entire mass of the sodium in the existing ocean. Since
164 IGNEOUS ROCKS AND THEIR ORIGIN
there is no known method by which its wat€r could be so much sweet-
ened during the intervening ages, it seems wise to conclude that the
initial assumption is fundamentally wrong. The reasoning is similar,
though yet more conclusive, in the case of the postulated basaltic
continent.
Without further amplification, the argument against a derivation
of the material of the earth's acid shell by the secular weathering of
andesitic or basaltic continents, may Ix^ regarded as sufficiently strong.
No other igneous type is found in quantity large enough that it could
l>e regarded as representing the original lands.
Clarke has calculated that ''the complete decomposition of a shell
of igneous rock 1/3 mile thick would yield all the sodium in the ocean."*
The most reasonable view seems, therefore, to be that the earth's
acid shell is ess(*ntially composed of primary igneous material, which,
for the most part, has l)een re-fused and intruded in the forms of the
pre-Cambrian Imtholiths.'
Earth's Basaltic Shell (Stratum, beneath the ''Crust").— The
acid shell is obviously underlain by at least local bodies of magma which
from time to time has traversed that shell and has crystallized as basalt
or the chemically equivalent diabase, gabbro, etc. Both chemical
and field relations show that this basic magma cannot possibly be due
to the fusion of ordinary sediments. Not so directly, but in the end
just as convincingly, those relations show that the greater basaltic
masses (fissure erupt ives) have not originated by the differentiation of
intermediate magma just l)efore their eruption. The acid pole of
such a hypothetical splitting ought to l)e on a similarly large scale and,
l)eing of lower specific gravity than the basic pole, it ought, according
to the plain common sense of the case, to l)e erupted before the basalt.
In numlHTless cases, ami particularly in the basaltic-plateau regions,
such association of acid magma entirely fails. Basalt, diabase, and
gabbro must be regarded as primary earth-magma.
As indicate<I in Chapter III, basaltic magma is the only one repre-
sented in all the larg(T divisions of the earth's surface. The visible
granite is much more voluminous than all the visible bodies belon^Dg
to the gabbro clan taken tog(*ther, but the extrusive members of thb
clan are more* c»venly spaced on the gloln?. The granites are generally, if
not always, confine<l to orogenic lK*lts; the basaltic rocks appear indiffer-
ently in mountains, plains an<l plateaus, lioth sulmerial and submarine.
> F. W. darko, Datii of C;<^>rhomi8tn', Bull. 491, U. 8. Geol. 8iir\'ey, 2d edition,
1911. p. 29.
* .\ftcr writinfc this portion the writer has found that Michel h6vy in the BuIL
80C g^l. France, Vol. 16, 18S7, p. 110, had already expressed essentially the
same conclusion.
COSMIC AL ASPECTS 165
When magma has been extruded on the largest scale and most rapidly,
through narrow fissures in which it evidently remained too brief a
time for the incorporation of foreign material (fissure eruption), that
magma has always been basaltic. We have also seen that among all
the igneous types the basalts have had the greatest persistence in
geological time, from the earliest recorded pre-Cambrian to this mo-
ment. (See Appendix B and page 56.) The gabbro clan is that
one most steadily represented in the standard eruptive sequences so far
described, though probably all such sequences for the continental
plateaus, if fully recorded, would include pre-Cambrian granite.
Finally, the basalts and most of the other members of the gabbro
clan, of whatever region or geological date, have always had striking
uniformity of chemical composition.
These facts suggest a primary origin for all or most of the world's
basaltic or gabbroid magma. (See Chapter XV.) They must have
partly furnished the motives which prompted von Cotta, long ago, to
record his belief in a continuous basaltic shell underlying the earth's
acid shell. Before him Bunsen had developed his well-known hypothe-
sis of two fundamental magmas, the **trachytic" and the "pyrox-
enic."* Von Cotta, going a step farther, conceived that the more basic
magma underlies the other and, further, that the former has long been
the only fluid magma. His own words will best state his position:
"Bunsen hat, wie gesagt, allerdings sehr zweckmassig einen allgemeinen
Unterschied zwischen trachytischen und pyroxenischen Gesteincn auch
chemisch festgestellt, nachdem ich langst geologisch sie zu unterscheiden
pAegte, in Wirklichkeit gehen aber auch diese beiden Gruppen durch Misch-
iinge ineinander fiber. Bunsen nimmt zur Erkl&rung der Thatsache an, es
best&nden «wei verschiedene vulkanische Herde im Erdinnern, ein trachy-
tischer (saurer) und pyroxenischer (basischer), das ist eine Hypothese, ein
sehr interessante, vielleicht eine sehr fnichtbare Hypothese, aber es ist doch
nor erne H3rpothe8e. Ist es denn nicht ebenso gut mdglich, dass der jetzige
Herd der vulkanischen Th&tigkeit nur pyroxenisch ist? Dass aber alle
Eniptionen eine kieselreiche Kruste durchdringen mtissen, in welcher sie
doreh allerlei Zuf&lle bis zu einem gewissen Extrem (dem normaltrachy-
^en) ungleiche Mengen von Kieselerde aufnehmcn? Und dariiber hinaus
Doch als quarzhaltige Gesteine! Mir scheint das sogar leichter denkbar, als
das Neben- oder Untereinander-Bestehen von zwei ihrer chemischen Natur
oach vefBchiedenen vulkanischen Herden, zumal da ein solches Nebenein-
aoder-fiestehen in alien geologischen Zeitr&umcn stattgefunden haben
muttte
"Wir haben es hier nur mit der heissflussigen Region und mit der festen
Kruste iiber ihr zu thun. Nehmen wir an, diese letztere sei aus vorzugsweise
kieseh^chen Substanzen gebildet worden, wahrend die fldssige Region in
' R. Bunsen, Pogg. Annalen der Physik u. Chemie, Vol. 83, 1851, No. 6, p. 197.
166 laXEOUS ROCKS ASD THEIR ORIGIN
ihrer allKcmcincn ZiLsammciiiiiictzung ungcfahr der normal pyn>3
Mcnfi^ng cnt(<pricht, so scheint mir diesc Annahmc ausiureichen, um aDe
vorhandcnc Maiinigfaltigkeit dcr Gc^tcinc im Allgcmcinen lu erkliren. Die
auM dein Iiiiicrn emporgcpresstcn hcL^flilsriigen, schr basischen Stoffgemenge
Id^tcn auf ihrem Wcgc, jc nnch den Uintft&ndeii, vici, wcnig, oder gar nichtt
von dcr vorhandencn kicselreichercn, fcsten Knistc (sowohl der eratarrten ab
der al>gelagerton) uiif und nahertcn sich dadurch mehr oder weniger den ei*
tremen trarhytisc!ieii Endglie<lern. Hierdurch erklart sich die atoflUche Ve^
schicdenhoit dor Kruptivgcstcinc, wahrend ihrc mineralogiscbe Verschiedeih
heit (die AuHbildung der einze!nen Mtncralicn) und die Un^eichhdt ihnr
Textur stets cine Folge <ler bcHondcren Umst&nde der Erstaming war, jt
nachdeni dieso schnoll oder lang^am untcr gcringem oder hohem Druck, iinUr
Zutritt von Wa-sser oder nicht, erfolgte." *
The gencTal conception of a basaltic sul>8tratum was reached,
apparently quite in<leix»n(lently, by W. L. Cin»en and briefly outlined oo
page 61 of Part II of his ** Vestiges of the Molten Glolje," publishedit
Honolulu in 1887. In 1901 the present writer was independently led
to it as the only workable hyiK)thesis for the explanation of the commoo
eruptive sequence illustrated in principle at Mount Ascutney, Vermont*
IrresfH^ctive of hyi)othesis^ the known distribution of basaltic
eruptions, both in time an<l space, (lenian<ls either a very esctenatre
series of subterranean chambers filled with basaltic material or cbe
a continuous basaltic substratum. Kven if the basaltic eniptivei
originate in separate compartments, these must have large aiie, M
shown by the vast ar(»as of country rocks which have been more or
less simultaneously pi^netrated by basaltic injection.s. So great mitft bf
the total area unclerlain by these imagined compartments of the earth'i
interior, that the whole must form an earth-shell fairly so called.
Earth's ''Peridotitic" Shell.— We have thus arrived at a mental
picture of the outer silicate<l layer of the earth which is derived dinc4f
from proved facts. That layer is formed exteriorly of a discontinnoai
sedimentary shell, a possibly discontinuous underlying "granitic
or ''acid*' shell, and below that again a basaltic shell| either con-
tinuous or discontinuous.
There are no |K)sitive fa<*ts comiH'lling a definite answer to tb
question as to what tyiM' of silicate matter umierlies the basaltic shdL
Here we must return to siwculation which can now be only ngpt
tivelv affected bv actual observations. If the known metCQlHtt
together really represent anything like a .sample of earth sufastanflti
.»<ome light on this problem is ofTered in the averages recently compM
for the chemical analyses of stonv meteorites. In Table VILCoLl
» B. Cottu, (;w»l<>gisrho Fragen, Frcilwrg. 1S5S, pp. 70-78.
< R. A. Daly, Bull. 200, U. S. Gcol. Survey, 1903, p. 110.
COSMIC AL ASPECTS
167
igton's average for 443 analyses, including those of 318 iron
Col. 2, his average for 125 analyses of stony meteorites,
jhows Merriirs average for 99 analyses of stonj' meteorites.^
^ABLE VII.— AVERAGE COMPOSITION OP METEORITES
68.43
11.07
6.33
4.55
.74
.65
.49
6.44
.44
.23
.14
.12
.11
.06
.05
.04
.04
.01
.01
.01
.01
.01
11.46
39.12
22.42
16.13
2.62
2.31
1.98
1.15
.05
.81
.04
.41
.38
.21
.20
.18
.06
}
11.61
38.98
23.03
16.54
2.75
1.77
1.85
1.32
.95
.11
.84
.33
.56
Sn,
99.98
.03
.02
.02
.20
.02
99.82
100.64
ng that the stony meteorites correspond to a rough average
itmal in the earth's silicated shell, it follows that a large
of it must be non-feldspathic and peridot it ic in composition,
iuthors have reached the same speculative result in attempt-
iw a parallel between meteoritic and terrestrial material,
no one has ever progressed much beyond the hypothetical
le inquiry. The most that can be said is that there is no
al objection to regarding the earth's silicated shell itself as
iccording to density, while that shell as a whole has the
,te average chemical composition of the stony meteorites,
n of Planetary Shells to Petrogenesis. — The foregoing part
ipter has been occupied partly with facts, partly with specu-
iceming the necessarily invisible interior of the earth.
meagre as the facts are, they unquestionably point to the
arrington, Publication 151, Field Museum of Natural HiBtory, Chicago,
pp. 211-213; G. P. Merrill, Amer. Jour. Science, Vol. 27, 1909, p. 471.
168 IGNEOUS ROCKS AND THEIR ORIGIN
existence of the Buccessive* sedimentary, ''granitic" or ''acid,'* and
basaltic shells, using the last word with the elastic meaning above
indicatc<l. The reasons are ample for the belief that only those three
earth-shells are actually concerned with the formation of the visible
igneous rocks. Each of the two overlying shells is, in part, visible at
the earth's surface. The basaltic shell, as such, is nowhere visible:
its existence is inferred. Before a fruitful attack on the general petro-
genic problem is possible, some kind of a definite idea as to the nature
of the deepest of the three shells must be obtained.
The conception that the basaltic shell is a continuous substratum
enveloping the whole earth is not refuted by any of the facts of petrol-
ogy, but it remains, and probably must remain, an undemonstrated
assumption. Many petrologists have expressed or implied their
refusal to venture so far into spei*ulation. Yet it is already clear that
without some such fundamental assumption the problem of the igneous
rocks must forever remain insoluble. If, on the other hand, the
basal hyi)othesis contains within it the, explanation of every one of the
countless millions of facts which may Ix^ recorded of the visible rocks,
and if no other basal assumption can do this, the tradition of true
science compels adhesion to the hypothesis thus successful in corrria-
tion. The earth's interior is no more removed from sensible contact
than the interior of a niolec*ule or than the hypothetical ether. As
physics and chemistry have \yovn vitalized and made practically useful
through their un(I(*monstrabIe hyi)otheses, so petrology must become
a more enriching, helpful science through its fundamental hypothesoL
It is the principal aim of this work to sketch the grounds for the bdief
that the assumption of a universal basaltic substratum may be hope-
fully regarded as the first step in a correct explanation of igneousHTOck
species, of igneous l)0(iies, and of igneous action on the earth. The
full value of the conception of a basaltic substratum is only to be known
by its fruits.
''Average Igneous Rock." — Before proceeding to the qpecific
application of this hypothesis, we may pause for a brief extenaon of
the siH'cuIative incjuiry. Of late years the question as to the chemical
nature of the ''average igneous rocks'' has been proposed by several
writers. Most of them (Harker, (Marke, Washington) have assumed
that the chemically analyzed rock si)ecimens, taken together, more or
less closely ni)resent this average. In his latest computation Clarke
states that the arithmetric mean of all the good analyses should give a
fair chemical average for the outermost ten-mile shell of the enfih.*
Menm^II and others, including the present writer, have pointed out the
overwhelming predominance of the granites in that shell down to a
> P. W. Clarke, Bulletin 491, U. S. Geol. Survey, 1911, p. 22.
COSMIC AL ASPECTS
169
probable depth of several miles.* Our hypothesis of a stratified earth,
as stated, implies that the visible igneous terranes consist partly of
that granitic material, partly of more basic material erupted from the
substratum, and partly of the more or less modified material of those
shells. Though the basaltic matter and its derivatives are exposed in
relatively small total volume, they occur in the form of distinct bodies
which are probably much more numerous than the granitic bodies.
The small average volume is largely counterbalanced by the greater
number and wider distribution of the more basic masses. For this and
other reasons the petrographers and chemically-minded geologists have
devoted special attention to the less voluminous rock types. In Osann's
compilation of 2431 of the world's analyses (1884-1900) the number
stated for members of the granitic clan (550) is nearly equal to the num-
ber stated for members of the gabbro clan (490) . These together make
nearly half the total number of analyses. Of the other half about 400
analyses belong to the diorite clan. As shown in Table VIII, the mean of
average granite and average basalt is almost identical with the average
diorite or andesite.* It is, then, not surprising that the average of the
world analyses (Col. 6) is close to the average for diorite or andesite.
TABLE VIII
I
Average
2
3
4
5
6
^ St
Average
' Mean of i
Average
1 Average
Average rock
^i;
granite
basalt
and 2
diorite
' andesite
(Washington)
236
161
Per cent.
89
Per cent.
1 87 _
Per cent.
1 1,811
Per cent.
Per cent.
Per cent.
SK),
70.47
49.65
60.06
59.19
60.35
58.96
TK),
.39
1.41
.90
.81
.78
1.05
Al/),
14.90
16.13
15.52
16.51
17.54
15.99
Fe/)j
1.63
5.47
3.55
3.02
3.37
3.37
FeO
1.68
6.45
4.06
4.17
3.17
3.92
MnO
13
30
.21
.13
.18
MgO
.98
6.14
3.56
3.93
2.78
3.88
CaO
2.17»
9.07
5.62
6.47
5.87
5.29
Ka/)
3.31
3.24
3.28
3.39
3.63
3.96
K,0
4.10
1.66
2.88
2.12
2.07
3.20
PiO,
.24
.48
.36
100.00 '
.26
.26
.38
1
100.00
100.00
100.00
100.00
100.00
» F. P. MenneU, Geol. Mag., Vol. 1, 1904, p. 263, and Vol. 6, 1909, p. 212.
' Some quarts' diorites are included in the average of Col. 4 in order to counter-
balance the influence of certain gabbroid types called '^ diorites'' and used in making
this average. Loewinson-Lessing states (Geol. Mag., Vol. 8, 1911, p. 249) that
the mean of average gabbro or basalt and average granite is a syenite. On com-
paring Table VIII with Columns 16, 17, and 24 of Table II, it will be seen that
this mean distinctly differs from average syenite, as specially observable in the
magnesia, lime, soda, and potash percentages.
* Includes .06 per cent. BaO and .02 per cent. SrO.
170 laSEOVS ROCKS AM) THEIR ORIGIN
While the average of the exposed igneous rocks is not a diorite but
a granite, the mean composition of the earth shells engaged in igneous
action may be close to a diorite. In that sense ami in that sense only
can the figures of Clarke, Washington, or Harker be regarded as nearly
representing the "average igneous rock."
Speculation as to the Primitive Differentiation of the Earth's Silicattt
Mantle. — This does not necessarily mean that the earth's silicate
shell was, in its upper part, originally of dioritic composition. Yet if
that shell were once greatly superheated, the granitic and basaltic
materials may then have l>een in fairly uniform solution; and they may
have later separated, by a kind of liquation, as the temperature slowly
fell. Such in principle is Durocher's explanation of the primitive earth
magmas which were to compose the visible rocks.'
As we shall see later, there are direct reasons for believing in an
''antagonism" or limited miscibility between granitic magma and ba-
saltic magma, whereby these tend to separate according to their densi*
ties. The facts of geolog\' directly suggest the probability of such
gravitative stratification of the earth's primitive shell. It is murh
more speculative and uncertain to hold that the surface shell of the
earth, when liquid, was originally constituted like molten andesite or
diorite. Fortunately for nearly all the practical applications of petro*
genie theor>', this point needs no decision.
A minimum degree of recency for the separation of the granitic
and basaltic shells is suggested by the eom|K)sition and structure of
the pre-( 'ambrian terranes. The earth's ** crust" must have been
solidified before the dearly extrusive basalts (greenstones) of the
Keewatin group in Canada were poured out. Similarly, the extru-
sive '*m<*tabasites" of the Finnish Bottnian and Kalevi an are among
the very oldest known rocks. The *' crust" upon which these ancient
masses were extrude<l was doubtless either granitic or dioritic. No
other composition can be assume<l for this ** crust" without involv-
ing insuperable (hfliculties in explaining the dominant, batholithic
formations of the pre-Cambrian complex. If the ** cru.st " were dioritic,
its debris should somewhere be abundant in the earlier pre-Cambrian
sediments. Yet thos<» sediments show plain evidence of having been
derived chiefly from granitic lands of great extent. • Again, on thin
assumption, the actual pre-Canibrian granites could only be explained
as the products of separation or differentiation in huge masses of the
dioritic crust wlii(*h had been remelted. It would then be necessary
to assume the |K)ssibiIity of such differentiation in the later batholithic
period, while denying its control in the earlier period when the "crust"
diorite as a whole was molten. This and other difficulties are avoided
W. Durochcr, Annates den Mines, Vol. 11, 1857, pp. 217-259.
COSMICAL ASPECTS 171
if we assume a granitic " crust *' for the earth which antedated the
eruption of the oldest greenstones of the pre-Cambrian terrancs.
It is not necessary to believe, nor is it likely, that the earth's granitic
shell is sharply marked off from the basaltic shell. From the analogy
of differentiated sills and similar bodies, we may more reasonably
imagine a transitional phase between the two shells. How important
that phase is volumetrically, and how diversified it is chemically, are
of course questions not to be definitely answered. The significant
fact is that the earliest pre-Cambrian greenstones are essentially of
the same chemical nature as the average modern basalts. The uni-
formity of composition exhibited in the great lava-floods of the pre-
C'ambrian as well as in those of later dates implies that the eruptible
material of the substratum has been basaltic throughout the period
during which the visible igneous bodies have been erupted. This
again implies that both the granitic shell and the transitional phase
l>eneath it were solidified in the pre-Keewatin time. If the outer part
of the earth has been cooling ever since, some of the basaltic substratum
has since been frozen to the overl>nng shells.
The "crust" of the earth may thus be conceived as composed of
parts which, in dow^nward succession, may be listed as follows:
1. The discontinuous sedimentary shell.
2. The continuous (?) granitic shell.
3. The continuous (?) transitional shell.
4. The continuous solid shell of basaltic composition.
Beneath that '"crust" is the basaltic substratum which is still hot
enough to flow and, in liquid form, to penetrate the crust if the pressure
is considerablv relieved.
PhjTsical Condition of the Substratum. — It is not necessary to
assume that the basaltic shell of the earth is molten as a whole, or that,
as a whole, it has been molten since the Keewatin division of pre-
Cambrian time. Where so little is known about it or can be directly
observed, the state of the substratum is subject to almost as many
hj-potheses as have been applied to the physical nature of the earth's
interior magma in general. The primary reservoirs of magma have
been conceived by different speculative writers to be :
(1) Multiple bodies of isolated liquid rock, still molten because of
the earth's primitive heat;
(2) A single substratum of rock, kept liquid by primitive heat; or
(3) Multiple bodies made temporarily molten in an otherwise
^M earth.
There are no known facts compelling belief in any of these hypotheses
for the basaltic substratum. The third view, that it is, in general,
truly solid or crystalline, involves the special difficulty of requiring
13
172 IGNEOUS ROCKS AND THEIR ORIGIN
the local introduction of an enormous amount of thermal energy.
The latent heat of fusion for crystallized basalt or gabbro is from one-
fourth to one-fifth of its total meltinp; heat measured from 0* C. To
find the source of such adclitional heat in radioactivity is to pile diflS-
culty on difficulty, for the lo<>al development of such a furnace is hard
to conceive in an earth shell which had already become cooled enough
for complete crystallization. If uranium or other suitable radioactive
matter of the ])ostuIated abundance were originally in the shell, that
shell could not have In^come crystallized without a temporary lull in
the atomic transformation; an improbable condition. On the other
hand, a mechanism whereby new radioactive matter is brought into
the shell is hard to imagine.
Tlie first hypothesis had the advantap;e of making relatively easy
the reconciliation of th(» pnnifs of a very rigid earth with the facts of
igneous geology. Kut it is cpiite ])ossible that this reconciliation is
possible for a very different reason. Several investigators, especially
Brid^man and Adams, have n>cently shown that a confined liquid, like
a soft solid, un<ler very high pressure may l>ecome extremely rigid.
Glass at 24,00() atmospheres Incomes much stronger than steel is at
atmosi)hcric presr^un*. Under j?reat pressure paraffine (like rublier)
will indent steel and tlie ordinarily soft mineral} fluorite, becomes
less plastic than steel.* Tliese exjH'riments suggest that the entire
basaltic substratum is possibly a true fluid, wliich resists tidal deforma-
tion .so well bccaus<» of iiit<rnal fri<*tion develo|>ed by great pressure.
The objection mi^^ht possilily l)e urged that a solid crust would not
be in stable ecpiilibriuni if resting on a general fluid substratum.
Since silicate matter is more dinse in the solid state than in the liquid
state and since heightened t<*mperature means lowered density, one
might assume that there would be danger of cru.stal foundering. Such,
imleed, was the basis of Kelvin*s w«*ll-known s])eculation on the origin
of the earth's internal temperature at the time of crust ification. But
the objection loses mueh of its fon*e when it is assume<l, as here, that
the globe is stratified, with its shells l»<'eoming more ferromagnesian
with incn\'ise of depth. Other vital considerations on this topic are
discus.«ed by the writer in earlier publications.^
Since the hy|N)thesis of a continuous, erupt ible substratum is also
th(» sinipl(»^t of tin* three so far conceived it may Ih» favored in one*s
attempt to seeure a UK'utal picture <»f the .substratum in action. How-
ev(T, the lack of full knowledge on this matter does not destroy the
* r. W. I<ri<Inm;in, in Vu>c. .Viiht. Acul. ArtH and •S'iciirofl, Vol. 47, 1911,
p. 53. F. D. Adams, Jour. GooIokv, Vol. IS, 1010, p. 500; ibid., Vol. 20, 1912,
p. 97.
» AiiKT. Jniir. Sri.nce, Vol. 22, 1900, p. 201; ibid., Vol. 26, 1908, p. 82.
COSMIC AL ASPECTS 173
probability that liquid parts of the basaltic substratum have partici-
pated in all igneous action. That probability is derived from many
facts in geology and has become independently appreciated.
General Conclusion. — We have thus arrived at the conception that
igneous eruption, since the beginning of Keewatin time, has been due
to the interactions of the basaltic substratum on the overlying acid
shell and sedimentary shell of the earth. The succeeding chapters
are concerned directly or indirectly with the nature of these inter-
actions. As their analysis leads to detailed comparison with the facts
of geology and petrography, the sequel may be regarded as a long
statement of the supreme test — the test of prophecy — which should
be applied to the theory of a stratified crust and a basaltic substratum.
CHAPTKH IX
ABYSSAL INJECTION
Introduction. — We have seen that rocks belonging to the gabbro
clan arc always exotic where they are found among the other visible
rocks of the earth's crust. By no conceivable process of fusion in
place can ordinary sedimentary or gneissic material be transformed
into basaltic magma. Adopting the generally accepted view, such
magma is to be regarded as due to eruption from the invbible interior
of the earth. The mean thermal gradient determined by borings
makes it eciually clear that molten magma must originate at a depth
of many miles. Assuming the average increase of temperature to be
3° (\ jKT 100 meters of descent from the earth's surface, the tempera-
ture of thoroughly fluid basalt (1200° C.) is reached at the depth of
40 kilomct<Ts (about 25 miles). There is little reason to doubt that
that depth is nearly the minimum for the basaltic magma before erup-
tion. Since the thermal gradient has nearly the same average steep-
ness in all the continents, magmaticr temi>eratures are likely to be found
everywhere at the d<'pth of 40 kilometers or more. Hence, so far as
visible rock formations are concerned, visible basaltic rock or naagma
must have b(M*n driven up through fissures in the earth's crust. Such
fissures must now be at least 30 kilometers high and for much or all
of re(M)rdcd geological time they must have been at least 15 kilometers
high. In a word, all visible basaltic eruptives, whether extrusive or
intrusive*, are Ix'st interpreted as due to abyssal injection of Bahetr^
turn material along nhyssal fissurfs in the crust.
Belief in this important principle is im])elled by facts and is inde-
pendent of hyiwthesis. But to appreciate its full bearing on petro-
genesis an explanation of abyssal injection must be found, and that
necessarily involves speculation.
In llHMi tht* writer published a paper which attempted to outline
the <'onditions for abyssal injection, on the prevailing assumption that
the earth is. and has long been, a contracting lx)dy.*
In 1909 Johnston-Lavis issued a paper emI>odying an independent
explanation of volcanic action on the same basis though it does not
«li<euss some of the fundamental features of the hypothecs. It was
' Aiiirr. Jour. Science, Vol. 22, 1906, pp. 195-216.
174
ABYSSAL INJECTION 175
an amplification of a note published in 1890, which was first dis-
covered by the present writer in 1910.^
Extracts from the writer's 1906 paper will serve to lay the elements
of the speculation before the reader. However, before entering on
those considerations, it is expedient to refer to the bearing of the re-
searches carried out on the radioactivity of rock matter during the
years elapsed since 1906. Moreover, the quite recent epoch-making
experiments of Adams on the strength of rock under conditions of cubic
compression (published in 1912) are of exceptional importance in this
connection; they will be noted in the appropriate place.
Strutt, Joly, and others have proved so much radioactivity in all
the typ^s of rocks at the earth's surface that this planet must be re-
garded as a true, very efficient furnace, if the average content of radio-
active matter also characterizes the rock matter in great depth. That
is, the earth's outer shell must be growing warmer. The fact that
traditional geology has demanded a cooling earth led Strutt to postulate
a concentration of the radioactive matter very close to the actual sur-
face, and Chamberlin has recently explained such a concentration in
terms of the planetesimal hypothesis. ^ Some geologists are wisely
withholding assent to Strutt's hypothesis. Though the break-up ,
of radium seems not to be inhibited by increase of temperature or of
pressure, or of both simultaneously (?), it is not yet proved that the
break-up of the parent uraniimi proceeds independently of pressure and
temperature. The "half-value period" of radium is only about 1300
years, while for uranium it is a billion years or more. If, then, the
subsurface conditions in the earth do prevent the spontaneous disinte-
gration of uranium, the earth cannot be a furnace ** fired" by radio-
activity. Until the physicists make clear the proof that the imagined
stream of energy is not thus dammed at the fountain-head, geologists
may well receive with caution the published statements regarding the
relation of radioaction to the earth's thermal gradient.
Contraction of the Earth. — In what follows it is assumed that the
earth is contracting, as so clearly suggested by structural geology.
Loss of terrestrial heat is only one of several causes for the contraction.
However the materials composing this plant were assembled, it is in
the highest degree improbable that they have been in chemical equi-
librium ever since pre-Cambrian time. Their slow rearrangement, to
form compounds that are stable under the conditions of high pressure
in the earth's interior, involves net decrease of volume. From this
» H. J. Johnston-Lavis, Geol. Mag., Vol. 6, 1909, pp. 433-442; ibid., Vol. 7
1890, pp. 246-249.
* T. C. Chamberlin, Jour. Geology, Vol. 19, 1911, p. 692.
176 IGNEOUS ROCKS AND THEIR ORIGIN
cause alone the glolx; may l>e Hhriiiking although its mean temperature
should I>e artually rising.
Secondly, allowance must be made for the considerable alterations
of volume involved in changes of state. Crystalluation of earth
magma represi^nts only one of these changes. Bridgman's experiments
are showing that many substances besides water have, respectively*
several polymorphic solid forms at varying high pressures. The trans-
formation of one of these forms into another, due to appropriate shifts
of pressure or temperature, is accompanied by notable volume changes.
Are there similar phase changes in the silicates and metals of the earth's
interior?
Again, actual field observations suggest that, at moderate depth,
granite, which has crystallized from magma with the usual massive
structure, may not be stable under conditions of permanent one-sided
pressure. It is then changed into micaceous orthogneiss of density
greater than that of the original rock. This transformation is familiar
as a product of dynamic metamorphism, but it may be indefinitely
more important as a product of static metamorphism, namely, that
caused by the dead weight of the rocks overlying a mass of new granite
or other crystalline mass. The quantitative value of such secular
volume changes is exemplified in the 16 per cent, of net shrinkage in-
volved when orthoclase passes over to muscovite and quarts." The
writer has come to the conclusion that, for geophysics if not also for
geology, the effects of static metamorphism may very greatly trans-
cend those of dynamic metamorphism.
It is seen that the '^contraction theory" of the earth is a quite
different thing from the ''thermal-contraction theory," which has been
so much discounted in late years. The net volume change of this planet
since the early pre-Cambrian is possibly a hundred times greater than
that due merely to the lo.ss of heat during that long era.
Shells of Compression and Tension
Whether the earth, as it cools and contracts, be solid and highly
rigid throughout, or whether it consist of a solid crust with an under-
lying fluid substratum, it is generally held by geologists that there is a
'* level of no strain" beneath the surface. The depth of this level has
been computed for a solid earth by Davison and Darwin, who have
made various assumptions w*hich are more or less reasonable provided
the fact of complete solidity is established. Their estimates for the
> Cf. R. A. Daly, Summary Report, Gcol. Surve}* of CmimU, 1011, p. 168.
ABYSSAL INJECTION 177
depth of the zero-strain level vary from 2 miles to 8 miles. ^ With
analogous assumptions Fisher has calculated that there will similarly
be a level of zero-strain in a crust overlying the fluid substratum of a
globe solidifying from the circumference inward. He found that "if
the time elapsed since a crust began to be formed has been 100 million
years, the depth of the level of no strain at the present time will be
about 4 miles."* In any case the depth increases very slowly with the
time elapsed since the crust first formed.
Rudski has pointed out that, if the earth's initial temperature were
not uniform, the level of no strain would, in a given time be deeper
than by the amount calculated on the assumptions of Davison.' It is,
in truth, probable that the initial temperature increased downward.
There are grave reasons for doubting the conclusion of Kelvin that an
initial uniform temperature was secured through the foundering of early
crusts. Like the suggestion of Lc Conte, that this thermal state might
be secured through convection currents, Kelvin's idea is not accept-
able to those who hold the very probable view that the earth's internal
density increases downward, not only because of increasing pressure
but because of differences in chemical composition as well.*
It may be noted that, in the above-mentioned calculations, no
account has been taken of the special and important contractions
characterizing the passage of lava from the liquid to the solid state
nor for polymorphic and chemical transformations, nor, except in the
case of Fisher's estimates, for the fact that, with a given fall of tempera-
ture, liquid lava (diabase) contracts about twice as much as solid
lava.*
All of these calculations have been made on the supposition that the
thennometric conductivity of the material of the earth is a constant
quantity. It is, however, practically certain that this conductivity
decreases with rise of temperature and very greatly increases on the
passage of liquid magma into solid rock.
Quantitative studies on the conductivity of rock-matter at different
temperatures, in different states of compression, and in the two states
of aggr^ation, are clearly needed. Until these are made it remains
impossible to calculate the exact position of the level of no strain in the
earth's crust. Nevertheless, in an earth composed of a crust floating
»C. Davison, Phil. Trans. Roy. Soc. London, Vol. 178A, 1887, p. 231; G. H.
IHrwin, ibid., p. 242; C. Davison, Proc. Roy. Soc. London, Vol. 55, 1894, p.
Hi. Cf. M. Reade, Origin of Mountain Ranges, London, 1886, p. 121.
*0. Fisher, Physics of the Earth's Crust, London, 2nd ed., 1891, Appendix,
p. 45.
* M. M. P. Rudski, Phil. Mag., Vol. 34, 1892, p. 299.
* Cf. J. Le Conte, Amer. Geol., Vol. 4, 1889, p. 43.
* C. Banifl, Bull 103, U. S. Geol. Survey, 1893.
178 IGNEOUS ROCKS AND THEIR ORIGIN
on :i suhstrutum whidi, because it is fluid and hot, has a lower condur-
tivity than tho solid, cool crust, wo might expect the level of no strain
to Ik' well within the crust, even if the initial temperature gradient
were comparable to that now obs<Tveil in the earth's superficial shell.
In a ixTsonal letter to the writer, the Rev. Osmond Fisher states that.
with a Ii(iuid interior, there must be a level of no strain in the crust:
and this is ai)parently true no matter what the initial temperature may
have been, lie states, further, that **the level of no strain would U'
the .same whatever the conductivity; but the time would not be the
same. The posit ion of the level would not fall so rapidly if the condur-
tivitv were less."
The shell above th(» zero-strain level is under tangential compres-
sion. The shells under that level, for a considerable distance down-
ward, are under tc'nsion. On account of the weight of all overlying
shells anv shell b<»low the zero-strain level tends to Ik? stretched or
(using Keade's term) to suffer ** compressive extension." This ten-
dency increases with depth to a maximum in a level computed by Davi-
son for a solid earth to lie 72 miles below the surface. The com»-
s]>ondiiig level for an earth with a fluid substratum has lx*en calculattni
by Fisher to lie at a ch'pth of from 30 to 55 miles, depending among
other conditions on the temperature of solidification.
Beneath the surfac(> shell of tangential compreasion, the rate of
S4>cular cooling and contraction and the consequent tension increase
from the level of no strain downward all the way to the substratum.
In his fir>t paper Davison calculates that the average rending stPWB in
the lower >liell is. after a given time (if there l>c no relief by stretching
or by cracking), four times the average compressive stress in the upper
shell. So long as folding or ov<rthrusting of the shell of compression
does not oc<'ur. the two .shells an* in physical continuity and are
strongly bound t<»gether.
Secular Accumulation of Tensions and of Cooling Cracks. — It is
generally agreed that, on the contraction theory of mountain-build-
ing, orogenic folding and crumpling is pos.sible through the secular
accumulation of compressive st rrsses in the outer shell. The crucial
question has not yet been satisfactorily answere<l as to whether there
mav be similarlv a secular accumulation of tension and of it8 effects
in the inner ^^hell of tlu' cru^t. If the earth's surface layer were a fluid
of onlv mo<lerat«'lv high viscositv, theaccumulationof tension would be
impo>sib|r to any sen>ibU> <>xt<'nt; moreover, its own weight would
nec(\<sarily closr all cavities almost as fast as formed during the slow
s(Miilar co(»Iing. Rut the average rock of the crust is a true solid
known to hav<* a very low mo< lulus of plasticity. Pfail has, indeed*
denied c>ven the smallest measure of true plasticity to the average crust-
ABYSSAL INJECTION 179
rock, and his experiments, like those of Adams, prove that massive gran-
ite, gneiss, or gabbro would, at surface temperatures, not flow under the
weight of even 25 miles of overlying rock.^ They would rupture and
shear, but the deformation would not reach the perfection of the
molecular shearing implied in true flow.
A vertical crack due to cooling contraction would thus tend to be
partly closed by shearing-in of masses from its walls. The shear-
planes would be inclined to the vertical. Each partial bridging of the
crack makes further shearing and closing of the crack more and more
diflRcult. A greater weight of crust w^ould now be required since some
support of the load is formed through the local meeting of the solid
walls. The simple vertical stress becomes partially resolved into a
complex network of oblique stresses tending to balance each other in
the loci of lateral support (the principle of the arch!). The portions
of the crack occurring between these loci of support may remain open
because of the diminished shearing stresses along the still gaping walls.
It thus appears that, though all rocks which are not laterally supported
will rupture under the weight of 6 miles of crust, yet the complete
closing of cracks at the same temperatures would not be expected even
under the weight of a much greater thickness of crust. The depth of
the shell ("zone'') of fracture has been deduced from the crushing
tests of stone and from the brilliant experiments of Adams and Nicol-
son on the deformation of marble enclosed in steel collars. The former
tests evidently do not prove anything at all definite as to the pressures
required to produce true plastic flow. The flow of marble under con-
finement has been produced under relatively low pressures, but this is
a special phenomenon, the result of movement on gliding planes. A
pen-knife and a few pounds of pressure will cause "flow" in a crystal
of calcite. It is safe to say that similar conditions are not found in
the average rock of the crust; if it flows at all the mechanism of the
flow must be something entirely different.
Deformation within the shell of tension is not to be estimated simply
by the ultimate strength of surface rock deformed in the laboratory.
The experiments of Spring, Hallock, and others show that the rigidity
of a solid increases with pressures ranging up to those about twice that
borne by our substratum. ^ This experimental law strengthens the
belief that cavities may remain open in the shell of tension. On the
other hand, the downward increase of temperature tends to lower the
internal friction and thus to promote the closing of cavities. A com-
» See F. D. Adams and J. T. Xicolson, Phil. Trans. Roy. »Soc. London, VoL
195, 1901, p. 367; F. D. Adams, Jour. Geology, Vol. 20, 1912, p. 97.
* For references see review by C. F. Tolman, Jr., Jour. Geology, Vol. 6, 1898,
p. 323.
180 raxEof's ROCKS axd their origin
parison of llu» prt'ssuro-grailiciit (1 atmosphoro to al>oui 3.7 metem
of tlcscent) with tli(» tcinixTatun' j^radu'iit (1° C ■. to about 30 metcroof
dosront) sii^K<***tsth(.'p«)ssihiliiy that rigidity actually increase!) through
the shoil of tension down to its bottom layer, where, on account of
the hifi^h temiNTature. tlie cliange of state, from solid to liquid, i»
aj)proa(hed; in that layir ravitie^^ an* doubtlr^s im|>ossihlc.
A further indication tliat cavities may remain open in the shell of
tension is indirect luit none the less noteworthy. According to the
assumption generally held by those adopting; the contraction theory
of mountain-l)iiihling. the ^heIl of tanf^ential compression, free of load
and unconfined as it is along its upjxT surface, can nevertheless for
long iK*riods of time endure without (h^formation a compressive stress
perhaps several times greater than the weight of 5 miles of rock. It
is the release of this pressun' (which was not relieved by simple radial
flow and thickening of the <hellj that has h'd to the paroxysmal growth
of a mountain rang(\ If the outer shell can long withstand such pm-
sures, it is reasonal>le to lH>li<'ve that the material of most of the shell
of tension is not perfectly j)Iastic mi<ler the weight of overlying crust —
a pressure which is great l>ut, in general, is only a fraction of the accu-
mulatecl tangential stress of conipre^>ion.
The last four paragraph> are taken from the writer's 1906 paper.
Since it wa*^ )>ubli>h<'d. iioteW4)rthy t'orroboration of the conclusion
there stated has lieen furnished by Hridgman, Adams, and King.
Telling experiments by the first two investigators mentioned have
l>een describeil in the foregoing cha)>ter (page 172). Indicating the
results of his suiK-rb ex|)eriment>. Adams writes as follows:^
'M. The ralcul:itii>n> wliich liavr Ihtm made as tu the depth Mow the
earth's surface nl which all caxitics in the earth's crust would he cloaed by
plastic flow, ha>od un the rru>hirig ^trcii^th of rocks at the surface of the
earth, arc crnmeou.-.
^'2. At ordinary tcni|H'ratur4'S l>ut under the conditions of hydroftatic
pressure or mhic conipr>>>ion whicli v\\>\ within the earth's crust, granite
will sustain a load of nearly KM) tons to the s({unre inch, that is to say, a load
rather more than seven times as ereat us that which will crush it at the sur-
face of the earth under the conditions of the usual laboratory test.
"3. rndcr the conditions of pressure and ten)i>eraturc which are believed
to obtaiti within the earth's cru-^t, empty cavitiis may cxiflt in granite to a
depth of at Ira^t 11 miles. The<«c may extend to still greater depths, and,
if fiiltMl with watfT. ga^ or vapor, will certainly <lo 50, owing to the praHOfe
exert<.il i>y :;uch fluids or ga.M's upon the inner surfaces of such cavities or
fis?urcs."
King's mathematical «Ii^cussion of these expi*riments led to the
following conclusion:
> F. D. Adams. Jour. CJcoIok}', V<»1. 20, 1912, pp. 97-118.
ABYSSAL INJECTION 181
"It is also shown that as far as hydrostatic pressure in the earth's crust
is concerned a small cavity at ordinary temperatures will remain open pro-
\'ided the depth does not exceed a value between 17.2 and 20.9 miles. At a
temperature of 550° C, supposed to exist 11 miles below the earth's surf ace,
cavities will remain open when submitted to considerably greater pressures
than are found at this depth. These values greatly exceed previous estimates
because experiment shows that a much higher value of limiting stress-differ-
ence than that usually employed must be taken in the neighborhood of small
cavities.
**The size of a cavity which can exist at a given depth depends on con-
siderations of stability and would demand a separate investigation."^
The unique value of Adams's experiments consists especially in the
proof that an elevation of temperature, even to a point approaching
dull-red heat, does not annul the effect of high pressure on the
strength of rock.
Hence it is all the more probable that discontinuous cracks produced
by secular cooling in the shell of tension should not be closed by the
dead weight of the overlying shell of compression.
This part of the argument may now be summarized. On the whole
it seems probable that a percentage of the whole tension developed in
the lower shell through secular cooling remains, at any time previous
to mountain building, unrelieved by the stretching or cracking of that
shell. The development of tensional stress and the multiplication of
cooling cracks will be at a maximum at some level near the middle of
the shell of tension. The accumulation of compressive stresses in the
outer shell will be relieved to a certain extent by recrystallization,
leading to the formation of denser minerals in the shell; but geo-
logical observation shows that, in a long period of time, enormous com-
pressive stresses are always stored until relieved by a more catastrophic
process. The accupiulation of the tensile stresses in the lower shell
will be in some direct proportion to the degree in which relief is with-
held in the shell of compression. Beneath a crust so diversely stressed,
there is a compressed elastic fluid which is ready, with relative sudden-
ness and with prodigious force, to inject itself into the shell of
tension as soon as there is any local relief of pressure or any breaking
of the continuity of the shell.
The whole system is evidently in unstable equilibrium. If each
shell were of uniform thickness and composition, and if there were no
external forces acting on the system, it would be difficult to forecast
when or where the stresses could be relieved.
Injection of Magma into the Shell ot Tension. — But the earth's
crust is not perfectly homogeneous; none of its shells is of perfectly
* L. V. King, Jour. Geology, Vol. 20, 1912, p. 137.
182 IGNEOUS ROCKS AND THEIR ORIGIN
uniform thickness; and, thirdly, there are external forces acting on the
shell of tension. Of special importance is tidal stress. Slight as may l>r
the effect of a single tidal period, for example, it will, in certain line*
appropriately ol)lic|\ie to the earth's equator, tend to \ivTench apart the
crust even <lo\vn through its viscous bottom layer. To such a powerful
fluid as that composing the substratum, this viscous layer, suddenly
sheared or hniken, is relatively a solid mass; to the searching fluid a
plane of shearing; in the viscous layer is virtually a crack. Into that
plane the tidal pulsations will pump the fluid, which instantly exerts
^
ft
/. --" >v ^ -
L^
1
I)
• ■
Vui. 110. — Swtion JiloiiK tho roiirso of the Cleveliind dike, YorkBhire. (After
A. (leikio, Aiirioiit VoIr:in*)CH <»f (Jrrut Britain, Vol. 2, 1879, p. 148; ori|dnal by (5
Burrow.) L, Liiissic sfNlimcntH; I), dike. B:iHO of Hcrtion, 400 ft. ah<i\*e sca IrvrL
Hori7.ontaUr:ilr. 1:J<),(MK).
its lateral hy<Irostatic an<l expan.Nional pressures on a shell already
I)rone to recoil Ixrau.^e of the nal though mild tension residual in the
lM)ttom of the shell. As the fluid thus works its way upward, it en-
counters nx'k which is increasingly more rigid and increasingly charged
with accumulated tension and cooling cracks. In fact, if we conceive
that the vi.^cous U)ttoin layer is once completely fx'netrated, it is eai«y
to iK'lieve that tlu* ahy.^sal dike will Vie rapidly injected toward the topof
the shell of tension. The shearing-in of the solid rock opposes the con-
J)
* Mi. 0
Ml. " . . 2 K.
Vui. HI. — Sortinn aloiiK the roiirfc of the Clovoliind dike, acrom the Crom
F«*ll cscarpiiirnt. (Same ref. n^ for Tik. HO, p. ir>0.) R, roof rock; D, dike.
tinned (»|M'ning of the potential fissure, but this shearing, as the level of
no strain isap])roach<M|, becomes .<lower and .slower and thus more and
more powcrlc.vi to check the ra])idly acting wedge of expanding fluid.
There arc some field indications that the aseensive force of erupting
magma is of the .<ame order of magnitude lui that of the weight of the
«*arth's cru>t (soli«l shrll overlying the substratum). Kven thegremtcst
lava flows arc minute when comi>arcd with the earth as a whole or with
any one of its primary shells. Very many, perhaps most, dikes have
not reached the surface, though their terminations approach it nearly.
ABYSSAL INJECTION
183
A good example, demonstrated during long-continued mining, is found
in the famous Cleveland dike of northern England (Figs. 110 and 111).
Such facts suggest that magmatic eruption is, in the first instance,
a hydrostatic phenomenon, though, as further noted in Chapter XIII,
other subsidiary factors are involved.
Relief of Tensions through Abyssal Injection. — On account of the
strong compression at the earth's surface, the magma of the abyssally-
injected wedges will not in most cases reach the surface. The act of
injection produces a great change in the conditions of equilibrium in
the shell of tension and therewith in the whole crust.
Fig. 112 represents a sectional view of the system after injection,
the earth's curvature being neglected and the wedge being shown in
Surface oc
SHELL OF
Leve/ of no strain
COM PR ESSION
a
OF : T EN S
° 'nCA,-
Fig. 112. — Diagram illustrating abyssal injection.
cross-section. The level of no strain is represented as about 5 miles
below the surface — a depth somewhat greater than the maximum
calculated by Fisher. The principle of the following argument is not
aflfected if the depth should bo a fraction of 1 mile or as much as 6
or even more miles.
A is a particle of the crust within the solid shell of tension. In the
stretching of the shell such a particle must move not only radially
toward the earth's center, but tangentially as well. If the shell is
homogeneous, the weight of the overlying crust will tend to shear the
particle indifferently toward m or m' or toward any one of an infinite
number of other points lying in the circumference of a horizontal circle
described about the vertical passing through A and with radius
Am. The shear-movement of particle A is, however, strictly con-
trolled in direction so soon as a liquid wedge is injected. At the level
of A the point o in the wall of such a wedge bears a combined hydro-
184 IGSEOUS ROCKS AND THEIR ORIGIN
static and elastic pressure from the magma. The former premire is
sensibly equal to the weight of the column of rock Ax; the maximum
elastic pressure eijuals the weight of the column Ay. The total of
these pr(*ssures, represented hy the line orif is equal to the oppositely
directed force o'n' on the wall of the wedge. On is not only a positive
force compressing the matter lx»tween o and ^4 ; it is also, and yet more
significantly, a din dive force which determines the direction in which
particle A must move as it is affecteil by the tensional pull of seeular
cooling and by sh(>ar during the compressive extension (stretching)
of the shell of tension. As long as the wedge remains fluid, particle A
will move in the direction of the arrow Am'. The condensation of
matter, which before the wedge injection had been only potential
(Ix'ing due to the accumulation of tensions and cracks in the shell),
now l>ecomes actual. As particle A is forced toward m^ a neighboring
particle, Ai, on the same level and to the right hand of A, is similarly
brought under pr(»ssure and moved in the direction of the arrow Am',
A\ communicates its motion to ^2 and so on. The pressure at o b
thus felt within the shell as far away from the wedge as the relief of
the accumulated tension and the closing of cooling cracks can take
place.
The movement of the particles A, Ai, At, etc., is analogous to that
of a railway engine pushing down a train of cars which had been
standing on a grade with each coupling pin at full length because of
the grade. BufTer meets buffer, communicating the pressure of the
engine. If the train had U'en nicely poised, just ready to move before
the pressure was applied, and if the grade were indefinitely long, a
small pr(>ssure woul<I set in motion a train of indefinite length. The
analogy is not perfect since the creep of the particles in the shell of
tension is not free but is controlled by internal friction and by the
strong at Ihesion Ix^t ween t he shells of compression and tension. Never*
theless, it is not difTicult to l>elieve that lateral creep would be set up
at a distance p<'rha])s s(*veral times the thickness of the whole crust.
Since the conditions are precisc^ly the same for particles B, B\
(to the left of /i), Ih, etc., there will be similar creep on the side of
the wedge opposite to .1 in the direction of the arrow 0' n^ The wedge
is thereby wi<lened. The continued injection of new fluid '"•g"*^
makes this new system of motions self-perpetuating until the at-
tainable reli(*f of tensions and closure of cracks is accomplished.
Thereaft<T, two possil)ilities are op<*n. The now much widened
we<lgc may have lost sufficient heat to solidify. The system of directed
creeps or latrral movements will then Ik* exchanged for an undirected
compressive extension similar to that which prevailed before the in-
jection. Or, if the wedge remains fluid, it will cause an indefinite oon-
ABYSSAL INJECTION 185
tinuance of lateral creep keeping pace with the diflferential cooling
contractions in the shell of tension. In the former case, the injection
of a second and of yet later wedges is possible, and their net effects, pro-
vided these wedges are elongated in the same general earth-zone,^
are additive to those of the first wedge. Tidal or other torsion may
locate such a zone of special igneous injection.
Downwarping of the Surface as a Result of Abyssal Injection. —
We have seen that lateral creep will be fastest somewhere near the mid-
dle level of the shell of tension, because it is there that the defect of
condensation of matter, shown in cooling crack and in residual tension,
is at a maximum. The ensuing condensation of matter in the shell is
^O f O S Y N C L J N E
*$■ 0 B S T R A T t/M
^m^^ir—m ■ i ■!
SUB STRATUM
B
Fio, 113. — A, Diagrammatic section showing the relation of abyssal inject-
ion to geoeynclinal downwarping. C, earth-shell of compression; T, earthnshell
of tenBion; broken line shows position of the surface before the downwarping.
B, Diagrammatic section showing the relation of abyssal injection to orogeny.
S3rmbo]s as in il, with addition of S-S\ shear-surf ace of the mountain building.
Igneous bodies injected and crystallized before this deformation are not shown.
New, large batholiths, of slightly different ages, are indicated by stippling.
at a maximum in the immediate vicinity of the zone of injection and
gradually decreases to each side of the zone. Since the two shells are
still solidly knit together, the enforced creep of matter to right and left
of the great wedges involves a strong downward pull exerted on the
shell of compression. A downwarp of the earth's surface is thus estab-
lished. The initial downwarp is of length, breadth, and depth de-
pendent on the magnitude of the injected body or bodies. Where the
injection is on a large scale the downwarp may be of geosynclinal
dimensions (Fig. 113, A).
The down warping implies, however, that the former nice balance
of stresses in the zone of compression is destroyed. Those stresses will
henceforth tend directly to increase the downwarp. Sedimentation
within the downwarp increases the weight on the creeping material
^ "Zone" here means a surface belt.
186 IGNEOUS ROCKS AND TUBIR ORIGIN
of the shell of tension, which is also now beginning to feel a small
(ioH-nward pressure, a component of the total thrust of the bent shell
of compression. The downwarping of the surface may thus gradually
increase even after all magmatic injections in the zone of tension have
frozen solid.
Here we may pause to apply an obvious test to the speculative
reasoning so far outlined. Though magmas must have difficulty in
forcing their way through the shell of compression, we should still
expect that molt(*n material would occasionally penetrate that shell,
so that areas of geosynrlinal .sedimentation would be zones specially
characterized by contemporary vulcanism. The following table (IX)
illustrates the actual facts determined for the better known geosyn-
clinals.
TABLE IX.— ILI.rSTHATIONS OF VOLCANIC ACTION CONTEMPCV
UANP:oUS WITH (;kosyn(xinal skdimentation
1. Loirer IlHronian^ North of Lake Huron. iVs. V,. LoKan, Geology of Canada^
1863, p. 55.)
Measured section of 18,000 fert of Ixrddcd rocks, with greenstone (often
amygclaloiilal) at seven horizons.
2. Animikie {Upper Huronian), I^ke StiiMTior District.
Thick mctar^iHiti>fl and quart zites, with interhedded extniaivea at
varioiLs horizons.
3. Keivrenafparif I^kc Superior District. (W. C. Gordon, Mon. 52, U. S. Geol.
Survey, 1911, p. 3SI.)
Thick sandstones, shales, and conglomerates, with extnunves at tUMny
horizons.
1. ShuM^cap Tcrrane (pre-Camhrinn), Southern British Columbia. (R. A. Daly,
Summary Rep. (Jcol. Sur\'ey of Canada for 1911, p. 167.)
At least 20,000 feet of limestones, alten^l argillites, quartiites, etc., with
basic volcanics at several horizons.
.'). finitid Canyon Scrirn (prr-Camhrinn)^ .Arizona. (C. D. Walcolt.)
Chu.ir se<iiments. 5,120 feet.
I .**e<limcnt> 475 feet.
Tnkar * lavas and se<iiments . . 800 feet.
se<liin<*nts 5,475 feet.
11,870 feet.
1 1 . I'tnwls nf ( 'on frm fHtni h*oum \'u lea niitm *l uri n g HrjHtn it ion of the A ppntaehian
(itthfifnelinaL'
Pcrioil llcgion
CarlM>niferon.« Nova Scotia; Boston district.
Silurian . Nova Scotia: Fox Islands, Maine.
Odi»viriiin. ( 'oIh«<iu id Mts.. Nova Scotia,
IN »st -Cambrian an«l pre-Carbonif-
cn»us Boston district.
ABYSSAL INJECTION
187
7. Rocky Mountain Geosyndinal at the ^9th Parallel. (R. A. Daly, Memoir No.
38, Geol. Survey of Canada, 1912, p. 161.)
Measured section of more than 25,000 feet of Beltian and Cambrian
sediments, with five horizons of extrusive basalt and basic andesite.
8. Main Pacific Geosyndinal {compound), Alaska to California.
Repeated extrusion of basaltic and allied magmas during prolonged sedi-
mentation in each of the Pennsylvanian, Triassic, and Jurassic periods.
In California measured sections showing 15,000 to 20,000 feet of bedded
rocks, including extrusives at five horizons.
9. Cretaceous Geosyndinal in Cascade Range, 4Qth Parallel. (R. A. Daly, Memoir
No. 38, Geol. Survey of Canada, 1912, p. 481.)
About 29,000 feet of sandstones, argillites, and conglomerates, overlying
1400 feet of andesitic breccia.
10. Cretaceous Geosyndinal of Vancouver Island. (G. M. Dawson.) Sediments
13,000 feet thick, with pyroclastics near the base.
11. Tertiary Geosyndinal of Central Washington. (G. 0. Smith.)
Period
Formation
Miocene.
Maximum
Thickness in Feet
Ellensburg sediments
Kecheelus andesite (extrusive) 4,000
Yakima basalt 2,000+
Guye sediments 3,500
Taneum andesite (extrusive) 1,000
Unconformity
Roslyn sediments 3,000
Teanaway basalt 4,000
Kachess rhyolite 2,000
Swauk sediments 5,000
1- Three British Geosyndinals:
Eocene.
Measured Thickness
in Feet
A. Upper Paleozoic:
Coal measures 8,000
Millstone grit 5,500
Carboniferous limestone series 4,500
Horizons of Contem-
porary Vulcanism
Basaltic flows.
18,000
Unconformity
B. Mid-Paleozoic:
Old Red sandstone
10,000-12,000
Wenlock and Ludlow 2,840- 3,480
Tarannon and Woolhope 1,150- 1,650
Llandovery 1,400- 2,300
Extrusive basalt, etc.,
at various horizons.
Volcanic band.
Volcanic band.
15,390-19,430
14
188
IGNEOUS ROCKS AND THEIR ORIGIN
riironformity
C. Lower PaUozoic:
Ortlovirian f*'L«w<T
Silurian"; I(MKK) +
Cambrian
12,000 +
I^VBS at thrre chirf
horitoiu.
Extrusive baaaltA and
andcsites at baae.
22,000 +
13. Wituntrn^nnui (ivosifficliual. South Afririi. iF. H. Hatch and G. S. Corvttir-
phini'. Tin- (ti-olo}:y of South Afrira, VM)\ p. 137.)
Wit\v:it4Tsran<l .scrirs stuiws 20,(KK) fri*! of thirknciM, with surface flov» of
diaba.'**' at iutiTvals.
14. O*onyurlinnh of Snr South \Vnh'<. \i\ A. SuHMiiiilchf Introduction to the
(icolo^y of New Soutli Waif's, Sydiiry, 1911.)
Age of SiTit's
Siiliiiiciit.'^
19,000
(*oi)toinporaneoufl Thick neas
Volcanim (Max.), Feet
IVrriio-CartM>nif4Tou.^. S a ti d s t o n <•. Hhalc, l^vofl and tuffs. 17,700
conKlt^niiTatf, roal.
rpper CarlMmiftTiMis. S andnt otic, shah*. Ditto.
ron ir 1 o Ml (• ra 1 1*.
liiiH*.'4tono.
rnr«»nforiiiity
Slialc. > a n d .s t o ti i*. La van and tuffs at
liiiif'>roii«>. *'\i\ many horisons.
Dirtii. Ditto.
Slati-s. •*lial«'r!.(|uart7.- TuITb.
it«'s.
IX'Vonian.
Silurian.
Ord<>virian
31,000
I0,000(?)
"Thick."
Orogenic Effects. - By the tlcvelopmont of a geosynclinal down-
warp, th(* shell of (M)iii])r('s>i()n is weakened, as experimentally iUii»>
trated hy Willis in liis memoir <m mountain-building.^ The weakening
is most felt in the two lines where the <h>wn-warpcd surface partafrooi
the s])hen»i<lal riirve <»f the earth. If s^-diment accumulates to the
depth of many thousands of feet in a ^eosyndinal, the material of the
original shell of compression is softened by the rising of the iaogei^
therms, while the stn^n^i^th of the new .^hell of compression occupied
by the seiliments is low berause of the ixM)r eonsolidation of this new
formation. For a double reason, therefore, a broad zone of weakaen
in the .<hell of <-om))n>sion is develo]M'd over the zone of igneousiiyce^
tion. Sooner or lat<T the se(*ular aeeumulation of compressive fltiwet
will express itself in the oropiiie eollapse of the shell; the buildup of
an alpine mountain ran^^e is bep;un.
Renewed Abyssal Injection during or after Moimtatai Boikliiig;
Development of Batholiths. — The extent to which shortening of the
transverse axes of the world's mountain ranges has occurred abows
that eaeh orop'nie revolution has been aeeompanied by a wbolcaak
» B. Willi?". 13th Ann. Hep., V. S. Gool. Surv., 1893, p. 217.
ABYSSAL INJECTION 189
shearing of the shell of compression over the shell of tension. The
surface of shear is probably not far from the level of no strain.
One effect of the shearing, faulting, and crumpling may be to
squeeze small bodies of magma up into the upper shell. But the grand-
est results of igneous intrusion would be felt in the shell of tension.
The instant that the two shells are sheared asunder, the tensions that
have been accumulated because of the solid continuity of the two shells,
and are still residual after the preceding injection of magma, are re-
lieved. The shell of tension is henceforth free to contract on itself.
A fluid wedge now injected into this shell, or a wedge injected previous
to the shearing but still fluid, would tend, according to the process
already described and especially because of the energetic, spontaneous
retreat of the country rock on either side, to enlarge itself. Opposed
to the active retreat and enforced creep of the solid rock of the shell
away from the middle plane of the wedge and thus to the ready contrac-
tion of the shell, is the friction developed at the surface of shear. Since
the shear is directed tangentially with respect to the curve of the earth,
the strength of the friction is measured directly by the weight of the
shell above the shear-surface. At the upper extremity of a wedge
which reaches exactly to the shear-surface, the hydrostatic pressure
exerted on the wall of the wedge is at least as great as the weight of the
shell above the shear-surface. The magma has, in addition, the live
energy of elastic expansion measured by the compression due to the
weight of the whole shell of tension. The net eflfect of these forces is
to permit of the contraction of the shell, already prone to movement on
account of the sudden relief of tension, and to cause a widening of the
wedge, w^hich may assume batholithic proportions (Fig. 113,5). It
is important to note that the recoil within the shell due to the relief
of tensions will characterize the whole of the area over which the shells
of tension and compression have been sheared apart; this area may be
several thousand miles in diameter. The piling up of the moimtain-
masB above would also cause an enhanced rapidity of lateral flow in
the shell of tension and likewise widen the magmatic chamber. Inject-
ion into the mountain rocks themselves would only be possible where
there is local relief of compression in the now heterogeneous, unequally
squeezed, and writhing mass. Since, in the nature of the case, com-
I»«S8ion generally dominates, igneous injection will, in this period,
afford but small geological bodies as constituents of the range.
At the mountain-roots below the surface of shear, there are one or
more great bodies of basaltic magma. As detailed in the following
chapters, there are excellent reasons for believing that the acid batho-
liths are the product of the inevitable reactions between the primary
basaltic wedges and the solid earth-crust. On account of the relief
190 IGNEOUS ROCKS AND THEIR ORIGIN
of compressive stresses in the superficial shell, these greatest of abyssal
injections are free to flux or stope their way toward the surface of the
earth, perhaps actually reaching it in certain cases. We shall see that
visible hatholiths and stocks arc to be explained as the less dense
gravitative different iates from such gigantic solutions of crust-rock in
the basaltic wedges.
Since it takes time for an abyssal wedge to work its way well up into
the shell of compression, the orogenic crumpling should generally
l>e nearly or quite completed before the visible batholitbic contacts
were established. In other words, batholithic intrusion to levels which
can ordinarily be exposed by erosion should lag behind the crustal de-
formation.
Here, again, we can return to observed facts for an indication of the
validity of the argument. Table VI, page 98, summarizes the leading
evidence that batholithic intrusion is actually confined to orogenic
belts and has probably always occurred in geological periods closely
following those of mountain-building disturbance.
According to the abyssal-injection hypothesis, batholiths should
be found in orogenic belts and, ideally, at or near the principal axes
of those belts. The corresponding fact is reflected in the phrase
"central granite," which carries a geographical truth though recalling
an obsoh'te notion as to the cause of mountain building. Yet one can-
not hold that all strongly folded mountain ranges have batholiths at
their roots. The Rocky Mountains of Alberta and northern Montana,
the greatly deformed western Appalachians of Georgia and Tennessee,
the northern Alps of Switzerland, and the northern Carpathians are
not characterized bv batholiths of intrusion dates connected with the
formation of these actual ranges. Each range is, in fact, celebrated
for its overthrusts, which have made it uasymmetrical. Our hypothe-
sis suggests that the batholiths connected with the deformation should
be looked for on the side from which the overthrust block came.
Matching the deduction, the Alberta Rockies arc flanked by British
(^ohimbia batholiths of Tertiary date; the Appalachian overthnut
belt, by the i)ost -Cambrian granites of Georgia and the Carolinas;
the northern Alps, by the** tonalitic zone" of Italy and the Tyrol* (Fig.
'A7), Similarly related batholiths are not visible in the Carpathians
but very large niass(»s of Tertiary acid volcanics arc developed on the
insitle of the Carpathian arc.^ One is tempted to guess that the
Tertiary granit(» of IClba locates the original site of overthrust Apen*
nines. The charriage of the huge block thrust from Norway over Sweden
1 S(*o ninp in cle Martonnc's Traits dc G^^ographie Physique, Paris, 1900^ p. 585.
' Soo V. UhHg*8 Tcktonischc KarteoFkizsc dcr Karpathen in "Bau imd Bfld
( )estorrcichfl," Vienna, 1903.
ABYSSAL INJECTION
191
was not associated with Swedish granitic intrusion but it may have
been connected with some of the greater mid-Paleozoic intrusions of
central and southern Norway.
Doubtful as some of these cases may be, the agreement of fact
and theory in the others is worthy of close attention. The repeated
discovery of highly specialized field relations in at least three principal
mountain chains must aid in establishing a final theory of the connec-
tion between magmatic movements and mountain-building.
Volcanic Action Subsequent to Mountain-building. — The larger
part of visible igneous rock is intrusive. Most of the large Paleozoic
and later injections have not extended to the surface. These facts
suggest that the upper layer of the earth's crust has long been difficult
of complete penetration by the abyssal magma. A leading cause for
this relative impenetrability is the compression of the superficial shell.
The stress characteristic of that shell is relieved by an orogenic parox-
ysm. After each paroxysm compressive stress is locally replaced or
overcome through the cooling of the rocks which had been heated by
shearing or by batholithic intrusion. For a double reason, therefore,
the penetrability of the crust should be at a maximiun in periods subse-
quent to strong mountain-building, especially after the solidification of
batholiths associated with the disturbance. (See Frontispiece.) This
expectation seems to be fairly matched by the known time relations
of the greater floods of basalt, as illustrated in the following table (X).
TABLE X.— RELATION OF FISSURE ERUPTION TO MOUNTAIN-BUILDING PERIODS
Locality
Date of fissure eruption | Preceding orogenic period
Lake Superior District . . .
Rocky Mts. at 49th
Parallel.
British Islands
British Islands
Appalachian Mts
British Columbia
Deccan, India
Great Rift, Africa
Keweenawan
Middle Cambrian (?)
Washington State
N. W. Scotland . .
Iceland
Washington State
Great Rift, Africa. . .
Great Basin, U. S. A
Snake River, Idaho. .
Hauran, Syria
Iceland
Devonian
Carboniferous
Triassic
Triassic
Cretaceous (or early
Tertiary?)
Cretaceous (Kaptian
series.)
Eocene (Teanaway basalt)
Oligocene (Lower Miocene)
Miocene
Miocene (Yakima basalt) .
Miocene (?)
Pliocene
Pliocene
Pliocene
Pleistocene and Recent . . .
Close of the Animikie.
Early Middle Cambrian (?)
Caledonian.
Hercynian.
Close of Paleozoic.
Carboniferous.
Late Triassic (also later?)
Late Triassic (also later?)
Close of Laramie.
?
?
Post-Eocene and pre-Mio-
cene.
Tertiary (Alps, etc.).
Miocene.
Late Miocene.
Earlier Tertiary.
Tertiary.
192 laXEOUS ROCKS AND THEIR ORIOIN
The effusion of a basaltic floo<l is usually ascribed to the mere
squeezing-out of the magma from beneath a cracked and sinking earth-
crust. Yet some force* may also be available from the expansion of the
substratum material as it rises to levels of enormously lessened pressure.
This expansion is of two kinds — that of the lava regarded as bubble-
free, and that of th(» gasc»s s(»parated from it in bubble form. If the
expansive energy of the licjuid proper is not all expended in driving
asunder the walls of the injected body, some of that great force is
available for extrusion. As magma nears the surface, the separation
of the dissolved gas must still further increase the volume and tend to
cause outflow at the surface. The relative importance of these three
conditions for extrusion is by no means apparent, though the writer
believes that the expansional energy of the injected liquid should have
more attention than it has had in general treatises on igneous action.
Summary
Postulates. — The assumptions on which the foregoing hypothesis
has been based are the following:
a. A contracting earth sujxTficially composed of a relatively thin
crust ov<Tlying a fluid basaltic substratum of unknown thickness.
b. The substratum so much compressed by the weight of the crust
as to be probably able to float the crust.
c. Through contraction, the (lev(»lopment of a level of no stnun in
the crust not far from the <'artirs surface.
d. The accumulation of j)r(»ssure in the shell of compression and the
simultan(K)us accumulation of cracks and of some of the powerful ten-
sion unrelieved in th(* sh(*ll Inflow th(» level of zero-i)train.
e. A stea<ly or nwurrent dislocation of the shell of tension permit-
ting of th(» forceful injection of the fluid substratum, to which'even the
viscous layer of the shell acts as a relativelv solid mass at the moment of
dislocation. This <lislo<-ation hjus Ix'cn referred to the tidal torsion of
the <*arth's crust, but tin* subecpiatorial torsion implieii in the tetra-
hedral theory of the earth, or crustal <leformation due to the play of
other cosmical forces or of forc(*s induced by the heterogeneity of the
crust, mav similarly cause dislocation in the shell of tension.
Conclusions. — 1 . The abyssal injection involves condensation of the
matter in the shell of tension. Cracks are closed and much of the ac-
cumulat(»d tension is relievcMl by an enforccMl creep of matter away from
the inject<»(l body. So long as the ImmIv n»mains fluid the stretching of
this shell due to continued contraction of th<» earth is accomplished
by creep of matter in the same directions. Th(» amount of creep is at a
maximum above the zone of injection and decrea.Hi*s to a minimum at
certain distances to right and left of the middle line of the zone.
ABYSSAL INJECTION 193
2. This lateral creep induces a downwarp of the earth's surface
immediately overlying the zone of condensation. The resulting
geosyncline may be the seat of prolonged sedimentation. If so, the
weight of the sediment itself tends to increase the lateral creep in the
shell of tension and the downwarp slowly deepens.
3. The shell of compression is already weakened at the angles of
downwarp; it is further weakened by the sedimentary blanket which,
comparatively little resistant itself, causes a softening of its basement
through a rising of the isogeotherms. When the filling of the geosyn-
clinal has sufficiently thickened, the shell of compression, owing to its
secular accumulation of stresses (which are intensified by metasomatic
changes in the shell) , begins to collapse. Mountainous forms and struc-
tures result.
4. The complete shearing-apart of the shells of compression and
tension during the orogenic revolution releases the tensions still unre-
lieved in the underlying shell. Abyssal injection on a large scale is thus
initiated or continued in the shell of tension. The relief of compressive
stresses in the act of building the mountains first occasions the possi-
bility of magmatic stoping and thus of the extensive assimilation of
schists and sediments by the primal basaltic magma. The diflferen-
tiation of the compound magmas of assimilation may explain the batho-
lithic central granites, etc., of mountain ranges, along with their satel-
litic stocks, injected bodies, and volcanic outflows.
5. The regional warpings of the earth's crust may be partly, at
least, referred to the varying strengths of abyssal injection from a fluid
substratiun.
6. The location and alignment of mountain ranges, the location and
elongation of geosynclinals, the final development of igneous batholiths
and satellitic injections, are all interdependent and related to special
zones of powerful abyssal injection from the substratum. These zones
are, in the large, located by cosmical stresses affecting the earth along
special azimuthal lines.
7. Mountain building causes relief of compressive stresses in the
superficial shell. The surface outflow of magma, either secondary or
directly derived from the substratum, may therefore be specially pro-
nounced after an orogenic revolution. In general, the theory of vul-
canism is also fundamentally affected by the doctrine of the shell of
tensions which are not entirely relieved by the compressive extension
of that shell.
CHAPTER X
MAGMATIC STOPING
Development of the Theory. — In 1893, Professor J. E. Wolff afwiKiu'd
the field problem of Mount Aseutney, Vermont, to the writer, then a
student at Harvard University. It was found that the mountain was
ess<»ntially made up of three typical stocks of successive intrusion
]M*riods ( Fig. 04) . For many y(»ars the writer was baffled in the attempt
to exphiin the mode of intrusion of these l>odies. It was not until
HK)2 that a reasonable hypothesis became disentangled from the mass
of facts (*ompil(Ml from this local study and from the literature of
plutoniir geology. The writer thus first conceived the principle of
''magmatic stoping." and "The (leology of Ascutney Mountain.
Vermont," embodying a brief statement of the hypothesis, was pul>-
lished in 19()3.* Simultam'ously, a full(*r account of it was printed in
the American Journal of Science.' The subject was further discuss^nl
by the writer in volumes U\ (1003) and 20 (1908) of the same periodical.
In the sear(*h for other statements of the hypothesis, it was found
that the central idea had in(l<>pendently impressed it«elf on Lawson
and (loodchild, thougb neither of these authors elaborated the argu-
m<'nts for and against it.' Meantime, Barrell independently deduced
a similar mechanism for the batholith at Mar>'Hville, Montana, but
did not publi.sh his n^sults until 11K)7.« Still later (1911) the maaterly
work of N. V. Ussing. on the g<'ology of the JuHanehaab region, Green-
land, was posthumously issuc^l. bearing the information that it«
author had also invented the stoping hyiK)thesis, during the year 1900.
when at work in (ireenlan<l.
The reader is reft»rred to the pajxTs mentioned for much of the
publishe<l evi<lence favoring the stoping hy|)othesis. To keep reason-
able limits of size for this book it is necessary to omit many details of
that evidence, both published and unpubli.shed. Yet the matter is so
vital to iM>trog<'nic theory that a fairly full .summary of the argument
will \h* presented. Some paragraphs of the writer's 1908 paper will
here be quoted.
» Hull. 2(>0, I*. S. ('.(M.l. Survey. 1903, pp. 0.V113.
' Anirr. Jour. S<i<npo. Vol. 15, 1903. pp. 2m 298.
'.\. (\ Lawson, Sciinrf, Vol. 3, \S\H\ p. 037; J. G. Goodchild, Geo!. Mif..
Vol. 9, 1S92, p. 147, Mui Vol. 1, 189^1, p. 22.
* J. BarrcU, Prof. Paper No. 57, T. S. GcoL Survey, 1907, pp. 151-174.
IM
MAGMATIC STOPING
195
The problem relates to the mode of intrusion for the magma of
bathohth or stock through the last few thousand feet of its uprise.
This question is independent of any theorj- as to the source of the
magma. In what follows it will be assumed, as a phase of our funda-
mental postulate, that the original magma of every post-Keewatin
subjacent body has been basaltic. Even if that postulate should be
Fia. 114. — Map or the quarts diorite stock (careU) at Maryaville, Montana.
(After J. Barrel!, Prot. Paper No. 57, U. 3. G. S., 1907, p. 74 and map in pocket.)
StippU, Empire shale; bUink, Helena limestone; solid black, diorite, microdiorit«,
Uid diorite porphyry; P, faults; dips ahown in degrees; thin lines are BtrilEa con-
tgura with 250-ft. interval. Note cross-cutting igneous contacts and general evi-
deoce of magmatic replacement. Scale, 1:62,000.
shown to be incorrect, the strength of the sloping hypothesis would
not be diminished.
A leadii^ fact concerning subjacent bodies is their replacement of
the invaded formations, as illustrated on pages 99 and 109- In &
large batholith the rock so replaced is often seen to have covered
196 laSEOVS ROCKS AND THEIR ORIGIN
hundreds of scjuarc miles and to have a volume to be estimated in
terms of hundreds of cubic miles.
Some of the replacem<'nt may lx> credited to marginal assimilation,
as advocat<*d sju^cially by the French school of petrologists. Yet
this cannot be the controlling factor. Allowing everything possible
for the solvent power of magmatic gases, the difRculty of imagtning
thermal conditions ad(*quate to p<Tmit of the known amount of replace-
ment through marginal assimilation alone has rightly helped to pre-
vent general belief in this process as the dominant one. The usual
lack of chemical sympathy betwwn a subjacent body and its country
rocks offers a difficulty no less great. Neither stirring by convection
or other currents, nor magmatic differentiation, nor both actions
together can explain these chemical contrasts, if the assimilation is
wholly marginal, that is, on the roof and walls of the magmatic body.
Only one other possibility is appan^nt. The rock matter replaced
Fig. 115. — S<M*tion alon^ tin* line X-Y in Fi^. 114. (7, Greyson diAle; E,
Einpiro^h:iU'; //. Hdona linn'stnnr: A/, mirnxliorito; li, IMmont diorite porphyry;
I), (liuritc. Not<> stopiiiK n^^nt rants in tho nnif of the Ktook; and the peripheral
po.Kition (if tho (li(>rit<'.
must have sunk into the magma to levels well l)elow those of the
visible contacts (Figs. 114 and 115).
]{erently, Cloiigh. Maufe, and Baih'y have suggested that a
batholithic clianiber may Ik* formed by the sinking of a single subter-
ranean block whos(> bright is nearly as great as the whole thickness of
the earth's crust. With the down-faulting, the chaml)er so formed is
filled with magma wlii<')i rix's along the fault planes to the level of
the chamber. An anah»gy is found in the n*markable "cauldron*
subsitlenc(»" at ( Uen ( 'oe, Srotlan<P (Fig. 1 10). This case is, however,
only an anal<»gy. It is at least iK)ssible that the sunken block at
Cilen ('(M» is merely part of the n'latively thin roof of a batholith.
Such partial su))siden<'e of a roof fragment might be expected on any
theory of batlHilithie iiitru-^ion. The existence of the "cauldron"
d(H's not c<mipel the view tiiat the great chamlnT l)eneath was opened
by similar <lown-fauIting en bloc. Tlu* authors assume that the
average d(*n.^ity of the eartl/s crust is greater than that of the mag-
K\ T. Cloiigh, II. ]i. Maiife, And K. H. Bailey, Quart. Jour. Geol. Society,
Lonaon, Vul. 65, 11KI9. p. till.
MAGMATIC STOPING
197
matic substratum. This would be true if the substratum were of
granitic or rhyolitic composition. It may not be true if, as assumed
in the present work, the substratum is basaltic. The writer has
published evidence suggesting that the earth's crust is, in fact, of
lower average density than its substratum and, since Keewatin time,
has been generally under conditions of stable equilibrium.' Without
further discussing the mechanical difficulties in the way of accepting
"cauldron-subsidence" as an explanation of batholithic chambers.
Fio. 116. — Map of tbe Glen Coe district, ScotlaDd, showing location of the
"naUron-subBidence. " (After C. T. Clough, H. B. Maufe, and E. B. Bailey,
Quvt. Jour. Geol. Soc., Vol. 65, 1909, p. 614.) US, Highland echists; V, Glen
Coe Tcilcanics; R, Rannock Moor granite; C, Cruachan granite; iS, Starav porphyri-
tic granite. The heavy curved line represents the fault rim of tbe sunken part
t' tbe batholithic roof. The intrusivee form a composite batholilh.
we may assume that the authors of this interesting su^estion accept
as probable the principle of magmatic stoping.'
The hypothesis of stoping includes the following essential points :
1. Marginal shattering of the solid rocks which form the roof and
vails of the magmatic chamber.
2. Sinking of the blocks (xenoliths) produced by the shattering.
' R. A. Daly, Amer. Jour. Sciei
■Op. cit., p. 665.'
:, Vol. 22, 1906, p. 201.
198 IGNEOUS ROCKS AND THEIR ORIGIN
3. Repetition of these processes until the chamber filled with liquid
magma is appreciably enlarged.
Marginal Shattering. — At the upper levels of an ab3rs8ally injected
wedge the magma must have a temperature many hundreds of degrees
C'entigra<le above the original temperatures of the country rock. At
the* contact the solid rock is rapidly heated to the magmatic tempera-
ture; farther away from the conta<'t it must be heated very slowly.
Rock matter is comparable to artificial glass in its thermal conductivity
or (liffusivity. Using the average vahie of the diffusivity («) for rock
at ordinary temperatures, it is possible to calculate the temperature
gradient established at the end of a given period, if the original tempera-
ture (r) of the wail-rock and the magmatic temperature (b) are assumed.
It may l)e further a.ssumed that the magma is of large volume and w
kc»pt stirred by currents.
If b \h* taken as 4(K)° F. (corresponding to an average depth of
aU)ut 24,(XK) feet), c as 2200° F., and /c as 400 (the value used by
Kelvin), the* temp<Tatures of the* wall-rock at the end of 1, 4, 16, and
100 years would have the valu<»s shown in the following table (XI)
for the n'sjM^ctive distances from the contact which are shown in the
first column of figures.*
TAliLK XI
i^iiitaiH'c in
ffH't
One year
Four yvnTH
Sixteen years
Onehui
yeai
0
22{Hr \\
22(HV v.
2200* F.
2200
W
17():{
ltM7
2074
'20
litVJ
1703
1917
10
(is:{
12ti3
17(W
sr»
lOS . 5
tiS.3
12G:)
W)
ra. KK)
537
1078
1703
W)
100
40S :>
683
2(N»
4(N)
v:i. 4(H)
537
12e3
:V2i)
4(N)
400
408.5
UN)
400
400
ra. 400
683
The tai)l(* shows that, at the end of the first year, the tempermture
(»f the rock is hut slightly afTccte<l hy the magmatic heat at a point
SO feet frf»m th(* contact, and that the temi)erature gradient for the 80-
foot shell then averag<'s nearly 23° F. p<T foot. At the end of four
years the t<*nip(Tature is hut slightly affected at a point 160 feet from
the contact and the temjKTature gradient is about 11* F. per foot.
But K cannot Ix' lu'arly m) gnat as 4(K) in the case l)efore us. The
conductivity k <lecrcases rapidly with ris<» of tem])crature in rock. The
experiments of Weher, Hart<ili, ItolnTts-Austen and KOcker, and Banu
* Sco U. A. Daly, Amor. Jour. iScicnce, Vol. 26, 1908, p. 24.
MAGMA TIC STOPING 199
show that the specific heat of rock averages about .180 at 20® C. and
increases regularly with rise of temperature, so that at 1100° C. the
specific heat averages about .280.^ It follows that thermal diflfusivity
in rock decreases with rising temperature even faster than the con-
ductivity decreases.
It seems safe to assume, first, that the diflfusivity of the gradually
heated wall-rock may vary from 275 or less to 150 or 100; secondly,
that the average diflfusivity of an 80-foot shell heated during the first
year by adjacent molten magma will be no greater than 200. If tc
be r^arded as averaging 200 for all periods greater than one year, the
four columns in the table showing temperatures will serve if the times
are, respectively, 2, 8, 32, and 200 years.
As a result of somewhat rigorous calculation, then, it appears cer-
tain that the heating of wall rock by plutonic magma must progress
^ith great slowness and that the resulting temperature gradient in
the shell adjoining the molten magma must be steep for many years
after the original establishment of the contact.
The stresses produced in the wall rock by this diflferential heating
must greatly transcend the strength of the rock as estimated by ordi-
nary breaking tests. The tendency is to reproduce underground the
shattering and exfoliation so often exhibited on the sills, columns, etc.,
of stone buildings wrapped in the flames of a city conflagration.
Again, several experimenters have shown that diflferent rocks have
different coefficients of thermal conductivity, as herewith illustrated
from a recent table of Konigsberger.^
Absolute Conductivity (XlO')
Simplon gneiss, highly feldspathic 5 . 50
Simplon gneiss, rich in biotite 6.75
Aare granite 4 . 03
Phyllite -. 6.77
Calcareous phyllite 7.30
Marble 5.20
Obsidian (Lipari) 1 . 92
Andesite (Orizaba) 3 . 06
It is well known that layered rocks conduct heat at varying rates,
depending on the direction of heat flow in relation to the layering.
The conductivity may be about twice as great along cleavage planes as
it is across the cleavage. The flow of heat must also differ in rate as
the country rock varies from point to point in its content of water.
These factors co-operate in producing great differential stresses in
the wall rocks of a magmatic wedge.
* For references see J. H. L. Vogt, Christiania Videnskabs-Selskabets Skrifter,
I. math.-naturv. lOasse, No. 1, 1904, p. 40.
* J. Kdnigsberger, Neues Jahrb, fUr Mineralogie, etc., B. B. 31, 1911, p. 141.
2no /',v/."rs i:,,< k.< i \ii riii-:n: omaix
I'iiially, a<i<iitiiiiial slri-s.-ic?. iiiust Im- iiiitii<'i'<l by the loiisiuii uf Miml-
trapiM'tl in the hcatcil coiitact-roi-k. if a cavity i» entirely fillwl with
water, the exi>an.sional force of the fluid when heated through several
hundred degrees i» indefinitely itreator than that jast Bufficient to rend
the rock even under condilionH of culiic comprotwion.
It sccm» certwn, therefore, that the contact shell of the countrj-
rock mast iKi-ome packed with tensioa->. These slowly accumulate
until the shell in a^iense "flics to pieces," like a Rupert's drop suitably
scratched.
However iniperfe<'t this explanation of contaet-ahsttering may be,
there can l>e no doubt as to the reality of the process. The typical
batholith outero[)s with an external lielt of apophyses. In many casn
their intrusion must have l)ei'n preceded by the development of intenw,
inconceivably complicated stresses in the country rock (Fig. 117).
Fti;. llT.^Arri'i'liHl utiipinK iit thi- rimf uf the Laiwili gnuiite bktboUth, Firb-
telfccliirRci (iiwrry cjiiiiwiirc. (Afti-r it. U)H<ii», GooloiDr von Deutaddkad, T«l
2, 1903, II. 191.) .1. Aniliansiti-mW rix-k (ImnifcU); «, ip-anite.
^[any liatluiliths and stocks are al»o characterized by intemal
iM'lts of incbi-sioas of the count ry rwks (Fig. 118). These Mocks
(xenolithst are entirely similar to many of the smaller masses separat-
ing the intrusive tongues of tlic outer Ix-lt, and very often no sharp
line separates the two Ik-IIs in their outcrop.' The intrusion of apophy-
ses and the euniplete enclosure of the blocks in the magma are thus
clearly parts nf the same process.
The fact that the xenoliths have not sunk or risen in the magma
far fruni their oriKtnal niches in wall or roof shows that the magmft was
extremely visctms at the time of enclosure.* Yet the xenolitha are
< K. ('<Mi<- liiui iii;t|>i>e<l iiTi umi:iunlly ituoil exuniplr in the Madoe iliiliin «f
Ontiiri..: «i- Atn.T. .Iimr S<i,.nrc, V<.t. 10. 19(B. p. 118.
: S<iiii<- ■•xt>iTiiiii-iilp liy TiiTiitiiniiii I<h1 him t» rnm-Iucle thftt the viacMitjr of ■
Rn'utiy iitiileri-oolifl li<|iiiil njiiiroaclirs that t>( the solid crystal of th« mob wab-
Mtancc. ]]i' rmiiKl i) •lisrontiniijly in the lpmprrftturt>-viBcoalty cunre. (O. Tan-
lUMin, Zoit, phya. Chemie, Vol. 2S, IS99. p. 17.)
MAGMATIC STOPING
201
characteristically angular and generally they are not arranged with
their longer axes parallel; nor, as a rule, are the xenoliths pulled out
into smears. These are considerations strongly adverse to the notion
that batholiths are merely injected bodies, intruded by a forcing
asunder of the invaded rocks along master fissures. On the other
hand, the actual facts of the field suggest that batholithic magma
has actively attacked its country rocks even in the very last stage of
the magmatic history, when the batholith was almost frozen. With
the higher temperatures and lower viscosity of the long antecedent
period, the magma must certainly have had still higher activity with
greater shattering power.
Mis.
Km.
Fig. 118. — Shatter-zone (S) at the contact of the Trail batholith, British
Columbia. (R. A. Daly, Memoir 38, Geol. Surv. Canada, 1912, p. 349, and map sheet
No. 8.) Rf Rossland volcanic series (latites, andesites, and basalts) and older forma-
tions; M, Rossland monzonite stock; GD, granodiorite batholith; G, alkaline bio-
tite-granite stocks.
Further illustration of the eflBciency of magmatic shattering is
given in the writer's second intrusion paper. ^
Relative Densities of Xenolith and Magma. — Will the shatter
blocks sink in the basaltic magma of abyssal injection? According
to experiments by Bischof , Delesse, Cossa, Joly, Douglas, and others,
the glassy phase of ordinary rock is always of lower density than the
' Amer. Jour. Science, Vol. 16, 1903, pp. 110-125. Through the displacement
of a decimal point, the writer, on page 1 13 of this paper, indicated a theoretical
pressure of more than 1,000,000 atmospheres as due to magmatic heating in the
country -rock. The calculation actually gave 10,000 H- atmospheres. The
general argument is not affected by this error.
202
IGNEOUS ROCKS AND THEIR ORIGIN
holocrystallinc phase at the same temperature.^ Bams, in a series
of famous experiments, has followed the density changes of basaltic
glass to high temperatures.'
Using the I)e8t minimum values (Douglas) for the observed de-
creases in density with change of state, and using Barus's result
for the further decrease on heating the glass, the writer has calculated
the densities (at atmospheric pressure) shown in Table XII.*
TABLE XII
Si)ooific fauvity of crystal- iSpccifir gravity of same rock when
line rock at: molten at:
•2(r C. 1000^ C. 1300^ C. 1000** C. 1100* C. 1200* C. 1300" C.
2.80
2 73
2 71
2.57
2.56
2.54
2 53
Gal>l>ro
2 90
2 8.3
2 80
2.06
2.65
2.64
2.63
and
H (X)
2 92
2 90
2.75
2.74
2.73
2.72
(lioritc
3 10
3 (r2
3 00
2.84
2.83
2.81
2 80
3.20
3.12
3 10
2 94
2.92
2.91
2 91
(juartz <ii(>-
rite and
tonalitc.
2.70
' 2 80
2.G3
2 73
2.01
2 71
2.46
2 54
2.45
2.53
2.44
2.51
2.43
2.51
2 60
2 54
2 53
2 33
2 32
2.31
2.31
Syenite ^
2 70
2 r)3
2 01
2 42
2.41
2.40
2.40
2 80
2 73
2 71
2 52
2.51
2.50
2.50
Granite
2 00
2 54
2 52
2 31
2 30
2.29
2.29
and
2 70
2 (>.3
2 01
2 40
2 39
2.39
2.38
gneiss
2.80
2 . 73
2 71
2.49
2.48
2 47
2.47
Table XIII show.s the chanR^^s in spc^cific gravity undergone by
blocks of stratified and schisto.^e rocks (common country-rocks about
batholiths), as these blocks, arbitrarily regarded as still solid, assume the
temiMTature (1300** C.) of molten ma^ma in which they are immersed.
T.\BLE XIII
Gneiss. . .
Mica schist
Sandi«tone
Ar^cillite
Liin<»i*tone
Ranfie of »\}. pr. at 20® C.
2 . r»0-2 . 80
2.75-3 10
2 20-2 75
2 10-2.80
2 05 2 SO
UanKe of sp. fcr. at 1300* C.
(solid)
2.52-2.71
2.67-3.00
2.13-2.67
2.32-2.71
2.57-2.71
i O. Bi.srhof, L. iind J. Jahrhiich fur Minoralof^ie, 1841, p. 565; cf. ibid., 1843,
p. 1: A. Deles.4e, Bull. S<>c. fciol. France. Vol. 4, 1817, p. 1380; A. Coosa, quoted in
Zirkel's Lehrbuch d<r IVtrographie, Vol. 1, 1S93. p. 681; J. Joly, Trans. Roy. See.
Dublin, Vol. ti, 1S97 OS, p. 2s3; J. A. DouKla.H, Quart. Jour. Geol. See., Vol. 63,
1907, p. 145.
- C. Baru«, Bull. 103, U . S. C.eol. Survey, 1893.
' Amer. Jour. Science, Vol. 26, 1908, p. 27.
MAOMATIC STOPING 203
It appears from these tables that nearly all xenoliths must sink
in any molten granite or syenite; most xenoliths must sink in molten
quartz diorit^, tonalite, or acid gabbro. Many xenoliths might float
on basic gabbro but the heavier schists and gneisses must sink in even
very dense gabbro magmas at 1300** C.
Giving, then, the highest permissible values to the specific grav-
ities of magmas, it is still true that blocks, such as are shattered from
the wall or roof of a batholith, must sink when immersed in most
magmas at atmospheric pressure. As shown in the first intrusion
paper, the blocks would likewise sink, though the magma enveloping
them lies at depths of 10 or 15 kilometers below the earth's surface.
Sinking of the Shattered Blocks. — It has been objected to the
stoping hypothesis that the viscosity of granitic magmas is too great
to allow of the sinking of blocks even much denser than those magmas.
This objection has, however, never been sustained by definite experi-
mental or field proofs. The xenoliths visible along batholithic contacts
have assuredly not sunk far from their former positions in wall or
roof and the reason for this must be sought in the high viscosity of the
magma. High viscosity is an essential attribute of a nearly frozen
magma. The phenomena of magmatic differentiation unquestionably
show that each plutonic magma must pass through a long period of
mobility. The most viscous of granitic magmas, the rhyolitic, issues
at the earth's surface with such fluidity that the rhyolite often covers
many square miles with a single thin sheet. The absolute viscosity of
the Yellowstone Park rhyolites must have been of a low order when
many of these persistent flows were erupted.
Even granting that the kinetic viscosity of a plutonic magma is
thousands of times that of water, it seems inevitable that it could
not support xenoliths more dense than itself. In a few days or weeks
stones will sink through, and corks will rise through, a mass of pitch,
the viscosity of which is more than a million of millions of times that
of water. Ladenburg has lately shown that small steel spheres will,
in a few minutes, sink through 20 centimeters of Venetian turpentine,
a substance 100,000 times as viscous as water. Ladenburg's experi-
ments have verified Stokes's generally accepted equation expressing
the rate of sinking of a sphere in a strongly viscous fluid:
2gr^(d-d')
where x = the velocity of the sphere when the motion is steady; g =
the acceleration of gravity; d = the density of the sphere; d'=the den-
sity of the fluid; r = the radius of the sphere; and t;=the viscosity of
the fluid.
15
204 IGNEOUS ROCKS AND THEIR ORIGIN
AAsumin^ that a ^anitc magma has viscosity even as high as that
of the* hani pitch al)ove mentioned, a sphere of a common type of
gneiss, 2 meters in <iiameter, would sink more than 10 centimeters
per day. A similar sphere 4 meters in diameter would sink more than
1.3 meters jht day. If the sphere were much larger, the Stokes for-
mula d(K*s not apply. Allen has developed the following formula for
such very large spheres:
, 1 4.T d-d'
''=k 3 ' ^' d '
in which k is a coastant for a given liquid-solid system. The terminal
velocity in this case* varies directly &s the square root of the radius.
Other things iK'ing equal, it follows that very large shatter-block.**
would sink much faster than those having diameters of only 1 or 2
meters. This d(Mluction agrees with the common observation that
along granite contacts very large xenolitlus are generally rare, though,
of cours<», roof-pi»ndants of indefinite size may there \)e found.
Subjacent bodies must cool with extreme slowness, both on ac-
count of the low conductivity of th<» country rock and of the libera-
tion of latent heat in crystallizing. Ilencc the presence of xenolithi
at th<» ol)serv(»d levels in these bodies must generally betoken for the
liquid at the time of enclosure of the block a viscosity much higher
than that of ordinary pitch. At the very clo.se of the magmatic
period the viscosity may 1m> comparable to that of the glass in a window.
Thes<» conclusions seem valid vvrn though the influence of pressure
on niagni.'iti(* viscosity is not accurately determined. The pressure
due to the weight of a batholithic roof, at the close of the magmatic
p<Tio(l. is generally to be estimated in terms of scores or hundreds of
atmosphen's, not thous:in<ls of atmospheres. Under such relatii'ely
low pressures, temperature must have <lominant control over the
mobility of tlu' magma, that is, ov<t its resi>on.se to stress-iliffcrences*.
Hy cooling, <Tystallization is induce<l and therewith the viscmcity
becomes practically infinite. Such a condition was nearly approached
when the visible xenoliths wiTe enclosed in stock or batholith.
We are driven to the conclusion that xenoliths must sink rapidly
in magmcis of such relativ<'ly low viscosity as that permitting magmatic
difTerentiation or the injection of ofT-shooting dikes. This magmatic
.stage is certainly of long duration for the subjacent masses, a stage
.so prolonged that many suc<'essiv(» shells of roof and wall may be
shattered and sto])ed away. If the original magma lie an abyssal
w«Mlge of primary ba>alt, xhv shatt(T-blocks c<mi|)08ed of the lighter
rock materials might float in the pure basalt, but the heavier type«
MAGMATIC STORING 205
would sink.^ In general, the primary magma must become somewhat
acidified and hence less dense by the solution of such blocks as well
as by marginal assimilation. Stoping thus for a time becomes more
and more rapid; after reaching a maximum of speed, the process
reaches zero activity with the crystallization of the body.
Roof-foundering. — How far have magmas thinned the roof of
their chambers by stoping? Is it possible that some batholithic
roofs have, in part, been destroyed? In the first presentation of the
stoping hypothesis the writer stated his belief that such foundering
of surface rock has not taken place during Paleozoic or later time,
but suggested it as a possibility in the early pre-Cambrian period,
when the acid shell of the earth was being developed. Further con-
sideration has caused a revision of that opinion. He has since sug-
gested that actual roof -foundering may be represented in the rhyolitic
region of the Yellowstone National Park and in the Blue Hill complex
of Massachusetts.^ Other possible examples of the process have
been mentioned in Chapter VII.
Nevertheless, it is clear that roof-foundering has rarely occurred
in post-Huronian time, and this fact must be reconciled with the
stoping hypothesis if the latter is to be finally accepted as a true and
essential part of the batholithic mechanism. Barrell speaks of this
necessity as "the greatest theoretical difficulty in the way of accepting
stoping as one method of batholithic invasion." He points out,
however, the unescapable truth that the same problem faces every
theory of such invasion.'
Since the Keewatin period, the earth's crust has remained essen-
tially coherent, and through it the primary basalt has been erupted
often and in many places. However, the irregular attitude of the
axes of the Laurentian batholiths as well as the abundance of those
bodies may possibly be explained by the repeated foundering of an
earth crust which was especially thin and weak in that early epoch.
We have seen that most, if not all, of the post-Cambrian batholiths
are confined to the sites of folded geosynclinals, and that such bodies
are generally arranged with their longer axes parallel to the respective
orogenic axes. (See pages 91 and 94.) In the last chapter this
relation is explained on a genetic basis.
Thus, the intrusion history of the globe may be conceived as
divisible into three epochs: the first being that in which the outer
primary shell was becoming stable through successive solidifications
> Metamorphism at batholithic contacts will cause even argillites, with initially
low densities, to assume specific gravities higher than that of molten baaalt.
' R. A. Daly, Proc. Amer. Acad, of Arts and Sciences, Vol. 47, 1911, p. 00.
» J. Barrell, Prof. Paper No. 67, U. S. Geol. Survey, 1907, p. 172.
206 IGNEOUS ROCKS AND THEIR ORIGIN
and foundcrings; the second being the post-Keewatin (Laurentian)
epoch of very general interaction between the fluid basaltic substratum
and acid crust, without extensive foundcrings but with development
of many large, irreguhirly occurring batholiths; the third, a period of
the localization of l)atholiths in certain mountain-built belts, where
alone there seems, in this period, to have occurred the injection of
molten magma in masses of hatholithic size, in but few cases accom-
panied by roof-foundering.
In the third pai)er on the Mechanics of Igneous Intrusion, the
writer briefly discussed the conditions leading to this contrast betwec^n
batholithic activity in the early and later stages of earth histor>'.'
The final explanation may partly lie in the highly probable fact that
the earth's sh(»Il of comprc»ssion has been notably thickened by secular
cooling during post-Laurent ian time. Moreover, a special thickeninic
of the shell above the "level of no strain" is a necessary feature of a
mountain range produced by the crumpling and overthnisting of
geosynclinal sediments. Conduction into roof and walls, magmatic
stoping, and the attenchint abyasal .solution of solid rock replace<l
in the shell of coniprci^sion, all involve the ultimate exhaustion of
heat supply in a magmatic wedge. With sufficient thickening of
the shell above the level of no strain, stoping must l>e arrested before
the batholithic roof is dangerously thinned. It should be remem-
bered that there is no danger of foundering until a part of the roof can
be wholly immersed in the batholithic magma.
In conclusion, we seem to have reason for believing that the
''problem of the cover'' will find its solution. A number of conditions
specially developed in Pali^ozoic and later times have together tendeil
to prevent the foundering of the roofs of granitic batholiths, though
IH»rhaps without succ<\<s in the cas<» of a few of these bodies.
Stoping in Sills and Laccoliths. — The writer has published a note
as to the testimony of the laccoliths on the stoping hypothesis.* It
was pointed out that these bodies generally show negative evidence
and a chii'f reason suggested is the high initial viscosity of these intru-
sive bodies. At the time of injf^ction. a sill is l>elieved to have been
relatively much less vis<*oiis t)i:in a la(*colithic mass. Hence it is
somewhat significant thai many sills enclose xenoliths, which because
of their greater densities have doubtless fallen from the respective
roofs of the sills. Cas(^< in the Pureell .sills of British Columbia, in
the Pigeon Point intrusive of Minnesota, and in the Sudbury sheet of
Ontario have been citt'd bv the writer.'
» .\mcr. Jour. Science, Vol. 26, 1908, p. 32, i
»R. A. Daly, Amer. Jour. Science, Vol. 15, 19a3, pp. 285-286.
' R. A. Daly. Amcr. Jour. Science, Vol. 20, 1905, pp. IH 201, 204, 207-308.
MAGMATIC STOPING
207
According to Weed, shale fragments may have been stoped up, in
the minettic magma at Yogo Canyon, Montana^ (Fig. 119). Lewis has
suggested, in connection with his valuable study of the huge diabase
sill of the New Jersey Palisades, the possibility of "underhand
stoping," stating that the (crystallized) diabase is 20 per cent, heavier
than the enclosing strata. ^ He makes no estimate as to the density
contrast of the molten diabase and the invaded sediment. The
difference cannot be nearly as much as 20 per cent. In any case the
dominant rocks of the earth's crust must have specific gravities higher
than that of molten diabase or basalt, so that, if ''underhand stoping"
characterized the exceptional sheet of New Jersey, "overhand''
(overhead) stoping is all the more probable in the average chamber of
fluent magma.
•? •' J ^
D
I
Fio. 119. — Section of top of sapphire-bearing minettic dike (3 to 6 ft. wide) in
wall of Yogo canyon, Montana. (After W. H. Weed, 20th Ann. Rep. U. S. G. S.,
Pt. 3, 1900, p. 456.) The blocks in the dike are composed of shale and limestone.
L, Limestone; D, dike. According to Weed this dike illustrates underhand stoping.
Abyssal Assimilation of Stoped Blocks. — A principal corollary
of stoping is the chemical change necessitated in the invading magma.
We may reasonably assume that the mass is hotter in its deep interior
than at wall or roof and that solution of the sunken xenoliths took
place during most of the magmatic period. If the stoping is pro-
longed, the amount of secondary magma formed by abyssal solution
must be considerable. Here we reach the most vital phase of t
subject, as it is now seen to imply large-scale petrogenesis. He
the matter is, for the moment, chiefly important as it suggests t
test of the hypothesis concerning intrusive mec : '^
The average wall-rock is of gneissic or granitic
primary basalt of an abyssal wedge must be aci
» W. H. Weed, 20th Ann. Rep., U. S. Geol. Survey,
* J. V. Lewis, Ann. Report, State Geologist of New i
132.
208 IGNEOUS ROCKS AND THEIR ORIGIN
of the average xenolith in depth. As we shall note in Chapters XII,
XVy XVI, and XX, there are plenty of proofs that much of the new
acid magma will rise to the roof, either as such or after the "differen-
tiation" of the xenolith-hnsalt mixtures. In general, then, great
batholiths should, at their upper levels, be of granitic composition.
If thick MHliments form a large part of the sunken material, the
differentiate at the roof will differ more or less conspicuously from
granite. A fair judgment of the facts recorded in Chapters XII and
XVI to XX must lead to the l>elief that this test of the stoping hjrpotbe-
sis, when founded on the basal premise of substratum-injection, is
fully met. In the next chapter the subject of abyssal asmmilation
will l>e more generally cHscusmmI.
CHAPTER XI
MAGMATIC ASSIMILATION
Introduction. — Few questions in geology are more important than
that as to the capacity of eruptive magmas to dissolve the country
rocks. It must evidently concern a petrogenic theory which assumes
the basaltic substratum to be the seat of igneous action since the
Keewatin lavas were extruded. The question has been and still is
answered in exactly opposite ways by able authorities. However,
many field studies of the last twenty years have given results suggest-
ing the great importance of magmatic assimilation. Iddings, Brog-
ger, Loewinson-Lessing, Harker, Clarke, Doelter, and others have
reviewed the history of opinion on the matter and it is not necessary
to repeat the statement.^ The petrographers who have opposed the
idea of magmatic assimilation as a leading factor in rock genesis in-
clude Rosenbusch, Brogger, Vogt, Harker, Cross, Iddings, Pirsson,
and Washington. Those who have favored the idea include Kjerulf,
von Cotta, Fouque, Michel Levy, Lacroix, Barrois, E. Suess, Loewin-
son-Lessing, Hibsch, Johnston-Lavis, E. C. Andrews, Coleman, Seder-
holm, Barlow, Brock, and N. H. Winchell.
Most geologists, even those specially engaged in the mapping of
igneous terranes, have either not considered the problem seriously or
have refrained from publishing the product of their thought concern-
ing it. Many geologists have mapped and discussed igneous terranes
with clear indication of their opinion that assimilation has no essen-
tial place in petrogenesis. The writings of still others show that
petrogenic theory is to them a closed book and these authors have been
content with color mapping and empirical description. The common
failure of field workers to think intensely about their rocks — and that
means in terms of origins — is a serious misfortune for geology. With-
out explanation ever in mind, vital facts are bound to escape observa-
tion and record. A signal instance is afforded in a vast number of
published papers dealing with eruptive masses. These memoirs,
» J. P. Iddings, BuU. PhU. Soc. Washington, Vol. 12, 1892, pp. 91-127; W. C
Brogger, Vidensk. Skrifter, I, Math.-Naturv. Kl. No. 7, Christiania, 1895, p. 116,
F. Loewinson-Lessing, Compte Rendu, VII*^ Congrds G^ol. Intemat., 1899, pp.
308-401; A. Barker, The Natural History of the Igneous Rocks, New York, 1909,
p. 83 ff. and 333 ff.; F. W. Clarke, Bull. 491, U. S. Geol. Survey, 1911, p. 294;
C. Doelter, Petrogenesis, Braunschweig, 1906, pp. 109-123.
209
21U laSKOUS ROCKS AND THEIR ORIGIN
often of great length, give either no account or a meagre account of
the country rocks, so that the work of such authors is not directly
useful in testing the hypothesis of magmatic assimilation"*.
Heat Supply and Magmatic Temperatures. — Our inquiry may log-
ically l>egin with a review of the facts indicating the amount of mag-
matic en<Tgy available for the solution of foreign rock. We need
not stop to consider the countless experiments, made cither specifi-
cally or incidentally in the industries, to prove the solubility of each
kind of rock in the other kinds. Even the mattes of the smelter be*
come more or less miscible with their slags if the temperatures are high
enough. The prevailing theor>' of magmatic differentiation itself
suggests the high probability that a.<«similation has often occurred on
a large scale. If, with falling temperature, a partial magma becomes
immi.scible with and separates from its magmatic complement, the
two materials, if in contact, should remix with suflScient rise of tern*
perature. Magmatic differentiation is, thus, a reversible process and
one cannot but wonder how certain advocates of the universality of
that process have l)een such uncompromising opponents of the assim-
ilation idea. The possibility of the solution of solid rock in a magma
is clearly a question of heat supply. If an ordinary magma is at all
8UiM*rheated at a contact, the ordinary country rock must be dissolved,
just as ice is dissolved in a large volume of water with a temperature
above 0° (\
The heat necessary for assimilation may be developed in several
distinct wavs; these will be successivelv considered.
1. The primary heat of the substratum, even if it be fluid, may not
now give it a tenip<Tature as high as the melting point (about llOO'
C.) of holocrystalline basalt at the earth's surface. Under the quiet
conditions of the interior, the primary liquid may be supercooled to
some extent. That influence of supercooling on the temperature is
at least partially, if not more than, offset by pressure control, for pres*
sure raises the fusion-point of ba.salt. Vogt has estimated that the
weight of 25 miles of crust rock raises the fusion-point about 60* C.
but he is careful to state the uncertain nature of the experimental
data on wliich his calculation is based.^ The actual effect (A pressure
may be still greater.
2. If the viscosity of the undisturbed substratum material is kept
high b(>c:uise of imdercooling as well as pressure, it follows that con-
siderablr heat must be tleveloped in the very act of injection into the
crust. How imi>ortant this cause may be cannot be stated but cor*
resp(»n(ling elevation of temperature in the abyssal wedge may con-
> J. H. L. Vogt, Vidonskabs-Selskabets Skrifter, I, Math.-naturr. KL. 19M*
No. 1, ChriBtiania, p. 210.
\
MAGMA TIC ASSIMILA TION ^11
i
ceivably be more than 100° C. This additional energy is, of course,
due to the conversion of the work done in the massive readjustments of
crust and substratum outside the magmatic wedge. On the other
hand, the expansion of the liquid magma in rising to levels of lessened
pressure must have a cooling effect, of uncertain magnitude.
3. Some superheat might be expected in a great abyssal wedge of
magma, part of which has been drawn from a level w^ell below the
bottom of the solid crust. It is probable that the substratum is itself
chemically stratified. Thermal-convection currents are, thus, im-
possible or else are too weak to affect essentially the normal tempera-
ture gradient, which may continue nearly unchanged from its value
in the overlying crust downward into the basaltic shell. The tap]f>ing
of the lower layers of the substratum during the injection of a very
great abyssal wedge might thus introduce some magma distinctly
hotter than that just below the crust.
4. Surface blanketing may produce some superheat in the erup-
tible part of the substratum just beneath. Such blanketing is exem-
plified in geosynclinal sedimentation, in the formation of the greater
volcanic piles, and in the local thickening of crust rock by close folding
and overthrusting in mountain systems. The inevitable rise of the
isogeotherms beneath such blankets must tend not only to fuse the
lower part of the crust but also to raise the temperature of the basaltic
layer, which is then subject to injection.
5. If radioactive matter is really concentrated in the outermost
skin of the earth, a superficial rock blanket must cause a heating, not
onlv of the crust but of the substratum as well.^
6. In any case blanketing will heat the crust rocks beneath the
blanket and so far facilitate their solution in abyssally injected magma.
7. An even more important heating of the country rock is the bi-
product of orogenic crushing and of dynamic metamorphism. Any
such independent raising of temperature in the wall rocks of an abys-
sal wedge obviously tends to postpone the freezing of the wedge and
thus to lengthen its life as an active solvent.
It is impossible to say how great may be the combined effects of
the several factors affecting magmatic temperature, but it would be
rash to assume maximum superheating of the primary basalt to an
amount greater than a few hundred degrees. We shall, for the pur-
poses of further discussion, assume a maximum average temperature
of 1300° C. in the larger magmatic wedges.
Observed Temperatures at Volcanic Vents. — ^The lava lakes of
the world cannot be expected to indicate the maximum or average
superheat developed in an abyssal wedge. No accurately measured
^ Cf. J. Joly, Radioactivity and Geology, London, 1909, p. 108.
•I
212 IGNEOUS ROCKS AND THEIR ORIGIN
volcanic temperature exceeds 1 200** ( \ As described in Chapter XIII,
such vents are kept (ijM'n l>y the fluxing of crust-roek which, by rapi<l
radiation into spa<-e, tends to a^^sume temperatures only a few degrees
alK)v<* the zero point.
N<'vertheless, some di^gree of sujKTheat is manifest even in extru-
sive l)odi<'s. 'rii(» v<TV fact that the* lif(» of volcanic vents is often to
I)e measured in 4nill(*niums strongly suggests that their feeding mag-
mat ic we<lges w<'re consideraMy sufXTheated at the time of injeetion.
It is difficult to lM>li<*ve that, at times of great activity, the lavas of
Kilauea are not considerablv hotter than is sho^iTi bv the 1911 tern-
pcTature (1010'' (\) of the 'M>Id Kaithfui*' center in that lava lake, as
measured by Sheplwrd and Perret, The incandescence of the lava
in the *'<'av<s" at \\w edge of this lake in IIHK) wa** certainly higher
than that of the lava in X\u* ''<)1<1 Faithful fountains.'* Similarly, the
testimony of several good observ<Ts, that the lava in the vastly greater
fountains in Mokuaweoweo (Mauna Loa) has often U'en "white hot*'
or 'Mazzling white,** is not to be lightly s(*t aside. The actual surface
temiMTatures reached at this greatest of active volcanoes may l)e as
much as VSiHf (*.
Observed Liquefaction of Country Rocks. — A further indication
of suiMTheat in basaltic magma is given in the facts recorded by
Lacroix an<l von John.^ These authors <lescribe instances where
blocks of gneiss, reaching the size of a cubic meter, ha%'e been "entirely
transformed'* into porous glass by immersion in l>asalt on its way to
the earth's surface. Ordinary gneiss is not a specially hydrous rock
and its t(>m}x*ratun> (»f liciuefactinn is doubtless more than 100^ C\
higher than the tem]X'rature of consoliilation of basalt. The differ-
ence niav well be mor(> than 200° (*.: and it would still be a minimum
value for the }xissible degree of su]M*rheat for the primary nuigma.
In some cases visible xenoliths have l>een softened and partly dis-
solved by acid. batholithi<* magma. The fact that such a block has
not sunk to depths far 1n>1ow the reach of erosion implies very high
visc'osity for the magma at the moment of enclosure. That viscosity
means :i relatively low temperature, probably no higher than the in-
version-! oiiit of (piartz. namely, S7()^ (\ Some assimilation is, there-
fore. po»iMe even in the greatly .^u]M'rcooled acid magma. How much
more rapiilly must it occur in magmatic wedges even no hotter than
the viMble lavas of Sirilv or Hawaii!
Fluxing by Concentrated Volatile Matter. — So far we have con-
>idered the possibility of solution orily through superheated, homo-
g«neous magma, especially tin* primary basalt. Yet to subterranean
> A. LiuToix, Les Kncluvt's dcs Hf>rh(*s V(>I('»nic|iics, Macon, 1803, p. 563-5;
C. von Juhn, Jabrb. d. k. k. IleicbsanHtalt, Vienna, Vol. 62, 1002, p. 141.
MAGMATIC ASSIMILATION 213
solution there are powerful aids other than the mere superheating of
the magmatic wedge.
First, the influence of gas concentration may be noted. It is well
established that some gases under pressure will form fluid solutions
with rock matter at temperatures far below the fusion point of ordinary
minerals and rocks. The well known experiment of Barus represents
an extreme case. He found that water and silica in a sealed tube
form a liquid at 210° C.^ Fouqu^ and Michel L^vy proved that the
materials of a very acid granite (79 per cent, of silica) mixed with
water in a sealed tube, became at least partly fluid at 1000° C; though
the dry materials did not become "even viscous *' at 1300° C.^ Other
gases must have corresponding effect on silicate melts.
Without further illustrating this universally accepted principle of
fluxing by volatile matter, we may conclude that a concentration of
either primary gases or gases derived from adjacent sediments greatly
facilitate magmatic solution. As stated in Chapter XIII, the mere
change of pressure involved in abyssal injection doubtless causes the
special concentration of the primary volatile matter at the top of the
magmatic wedge. The secondary gases, introduced from invaded
sediments or other wall-rocks, will similarly tend to assemble at the
top of the magmatic chamber. In each case the concentration takes
place in the part where the gas is most eflScient in helping to dissolve
the maximum amount of country rock in the magma.
To the French school of geologists belongs the distinction of having
long emphasized the influence of magmatic gases in the production
of secondary magmas. Those observers have not assumed the ex-
istence of a basaltic substratum as a necessary prerequisite to igneous
action in Keewatin and later times; yet their arguments in favor of
gaseous influence in assimilation are also, for the most part, good under
that assumption.
Influence of the Mixture of Rock Matter on Liquidity. — ^Again,
mere chemical contrast between two rocks tends to lower the tem-
perature of their mutual solution. The principle is exactly the same
as that just mentioned in the case of the gaseous fluxes, and there is
general agreement among physical chemists as to its truth in the
present instance. A single example will illustrate the efiiciency of this
kind of fluxing. Petrasch has experimentally shown that two parts of
limburgit^ and one part of Predazzo granite will melt together at 960°
C, and that the solution remains fluid down to 850°' C.^ Predazzo
* C. Barus, Amer. Jour, of Science, Vol. 9, 1900, p. 161.
*F. Fouqu^ and A. Michel L6vy, Comptes Rendus, Vol. 113, 1891, p.^283.
Also personal communication from Mons. Michel L^vy.
* K. Petrasch, Neues Jahrb. fur Mineralogie, etc., B. B. 17, 1903, p. 508.
214 IGNEOUS ROCKS AND THEIR ORIGIN
granite softens at 1150** C. and the limburgite at 995** C* The freei-
ing point of the mixture was, therefore, 300° C below the melting point
of the granite and 145** ( \ below that of limburgite.
Magmatic Stoping in Relation to the Low Tenqienitures of Con-
solidation for Batholithic Rocks. — Assuming the validity of the "geo-
logical thermometer*' now lK»ing constructed by the Geophsrsical Lab-
oratory of the C/arnegie Institution of Washington, each of the ordi-
nary' granites has l>een at least partly molten at a temperature no
higher than 870** (\ nor lower than 575** C* Both undercooling and
the presence of volatile matter in the batholithic magma are doubtless
responsible for this low temperature of final consolidation. At that
temperature, solution of the country' rocks along the main contacts
may be almost infinitely slow, but magmatic stoping may still con-
tinue and therewith we have the possibility, if not the necessity, of
assimilation in great depth. (See page 216.)
It seems lik(*ly, then, that assimilation would continue in a mag-
matic wedge of batholithic proportioas until its average temperature
fell to 900** C\ and ptThaps lower.
Available Heat for Assimilation in Batholithic Masses. — Since, for
the substratum-injection hy|K>thesis, the heat problem is most
serious as regards the greatest of the intrusive bodies, it will be well
to illustrate the inferred possibilities of assimilation by a quantitative
discussion of granite batholiths. The substratum-injection hypothesis
involves a secondary origin for all post-Keewatin granites. The writer
has, on several occasions, (*xplained ma<«t of these as differentiates from
solutions of the earth*s ncid shell in ba.«<altic wedges injected from the sub-
stratum; and that view is aln>ady implied in the earlier theoretical part
of the pre.«5ent l)ook. The average country rock of the upper half of
each batholith is either acid gneiss or a type chemically identical with it.
The assumed range of temperatures characterising the active
magmatic perio<l of :i major injection is from 1300** C. to 900^ C.
The energy emitted by 1 gram of TKiuid basalt cooling from 1300^ C.
to 900° (\ is about (40()X.H3j = 132 calories. The total melting-heat
of gn<Mss, if molten at 900** (\, is about 350 calories. If we neglect
heat of solution (positivt* or negative), 10 grams of the primary basalt
132
would ren<ler molten about l()X*= 3.8 grams of gneiss* if all the
surplus heat of the basalt were available for the solution of the gneiss.
So far as the assimilation is further aided by magmatic gases, the pro-
portion of gneiss capable of solution in a unit mass of the basalt
would be still higher.
» C\ Docltcr, THclicrm. Min. u. Potmg. Mitt., Vol. 20, 1901, p. 210.
> F. E. Wright and E. S. Lorsen, Amer. Jour. Science, VoL 27, 1909^ p. 121.
MAGMA TIC ASS I MIL A TION 215
We seem, therefore, not to stretch the probabilities of the case,
nor to contravene known facts, when we conclude that a limited but
important amount of assimilation in primary basalt is theoretically
possible.
The actual calculation suggests that the magnitude of permissible
assimilation may be of the order demanded if, for example, a Tertiary
granite is due to the assimilation of silicic rocks in a primary basaltic
wedge. If the wedge of liquid basalt stood originally to a height of
20 miles above the substratum level, its heat content might suffice
to melt up a mass of gneiss of the same average horizontal extent
and from 6 to 8 miles deep. In fact, such may be the order of depth
for the granite of post-Cambrian batholiths. Unfortunately, erosion
has seldom or never penetrated as much as 4 miles into such a batho-
lith, so that this rough deduction cannot be directly checked by
observation.
However, the justice of the assimilation theory becomes only fully
apparent when the chemistry and field relations of the different igneous-
rock types have been reviewed in some detail. Such a review, though
a partial one, is made in Chapters XV to XX.
We may now continue the general outline of the subject by noting
the characteristics of magmatic solution in its two leading phases.
Marginal Assimilation. — No informed worker on the igneous-rock
problem needs to be reminded of the rarity of the cases where the con-
tact of magma and country rock is marked by shells of transition
material. Generally there is no chemical "consanguinity'' directly
evident between intrusive and wall rock. A granite batholith often
contacts with gneiss, greenstone, quartzite, argillite, or limestone, and
yet the contact phase of the intrusion is everywhere of essentially the
same composition. Only very rarely have large rock bodies, like those
of the Haute Arifege (Pyrenees) or the contact phase of the Alno stock
(Fig. 184), been explained as due to direct melting-together (syntexis).^
The exceptions surely prove the rule in this case.
Such failure of transition rocks or of direct evidence of consan-
guinity have prompted many petrologists, including those listed in
an early paragraph of this chapter, to deny forthwith the efficiency of
assimilation in forming large bodies of magma.
Yet we have already seen that this suggested test of the hypothesis
cannot be conclusive. In the nature of the case notable assimilation
is to be expected only in very great bodies of magma, that is, generally
only in batholiths. Apart altogether from theory, it is plain that
«om6 stoping has occurred in practically every subjacent body. There-
> A.Lacroix, Bull.des serv. carte g6ol. de la France, Nos.64and 71, 1898, 1900;
A. G. Hdgbom, Geol. F6ren. Fdrhand., Vol. 17, 1896, p. 132.
216 IGXEOUS ROCKS AND THEIR ORIGIN
fore their main contacts, at least in part, were established after the
magma had \HHm somewhat cooled. The suspension of xenoliths in
their visible positions proves In^yond peradventure that the enclosing
magmas were nearly frozen when the blocks were broken off from
wall or roof. In such a case the block could not be digested unlesK
there were a local concentration of fluxing material, such as a consider-
able perc(>ntage of water in the block itself. A much more important
consideration is that the stoping of the earlier period, when the magma
was an c»nergetic solvent, removed from roof or wall the S3mteetic
films and shells as fast as these were formed. Roof, wall, and xeno-
lith are generally free from syntectic shells l>ecausc the period of stop*
ing must Ik* longer than the fHTiod of solution. Perhaps the stoping
hypothesis has its greatest psy<'hological value in henceforth forbid-
ding the use of this time-worn argument against assimilation. More*
over, the opponents of this doctrine have failed to allow for the highly
probable fact that magmatic <lifferentiation can occur at temperaturesi
below those of active assimilation.
For several reasons, therefore, chemical evidences for or against
marginal assimilation must generally l>e indirect.
Abyssal Assimilation. — However, we have ol>served the improl>-
ability that marginal assimilation alone is competent to explain the
observed amount of rej)laccni<>nt jK^rformed by batholithic magmas.
Xor can it explain the huge volumes of secondary magma which, on
the substratum-injection theory, have U'cn formed by injected wedges
of primary basalt. Our way is clear if we follow the stoping hypo-
thesis to its in<*vitabl(> corollary, ahysnal assimilation. In most
cas(»s nearlv all of the sunk<*n blocks fnnat l)e melte<l or dissolved or
both.
IliiL^trations of pun>-ineltingof blocks enclosed in extrusive basalt
are recorded by Ln<Toix, von John, an<l others, as above noted.
These and nianv other writers have described numerous cases of
mutual solution in lavas as w(>ll as in intrusive magmas which could
not remain licpiid for a tenth of one ])er cent, of the time allowaMe
for a typical subjacent body. It is little wonder that Brogger, though
one of the leading oppon(*nts of the marginal a.ssimiIation theory, is
willing to cnuiit magma in depth with great assimilative power.'
His statement implies that he had chiefly in mind marginal aMtmi-
lation. The present writer heartily agrees with that conclusion.
Abyssal assimilation must oeeuron themaincontactsof abyssal wedges
at grf*at depth, and it may be res]><»nsibie for a notable part oi the
secondary magma in a batholith. Hut the other t>'pe of abjrsnl
^ \V. (\ I)n)KKor, Dii' Kniptivgf^tcino des KriMtiaDiagebietes, Vol. S| MB^pw
350.
MAGMATIC ASSIMILATION 217
assimilation, namely, that represented in the solution of sunken
xenoliths, has certain advantages over pure marginal solution, which
we shall do well to review.
First, marginal assimilation is largely effective only in the earliest
parts of the magma's history, when it is absolutely and relatively very
hot . There is thus an early time-limit fixed for the gigantic work of
dissolving the thousands of cubic kilometers actually replaced in the
intrusion of a large batholith.
Secondly, on the older view the assimilation takes place primarily
on main contacts and along a relatively limited amount of surface.
For example, a cube of wall-rock 1 kilometer in diameter can offer
only about 6,000,000 square meters of surface at a time to the dis-
solving magma. If that same cube were shattered into cubes 10 meters
on the side and then engulfed, the magma would carry on the work of
solution on 600,000,000 square meters of surface. If the shatter-blocks
were cubes one meter in diameter the original surface would have been
increased one thousand times.
Thirdly, the average crust rock, being allied to gneiss, dissolves at a
lower temperature in basic magma than in acid. On the stoping hypothe-
sis, solution of the xenolith generally occurs in the lower, basic part of
the magmatic chamber; on the older view, it is the granitic magma
which must do most of the work of solution. For even if the originally
injected magma is a basalt, the products of its assimilating activity,
which are more acid and less dense than itself, must remain at the batho-
lithic roof and rapidly assume the chemical composition of mean
mountain rock. It follows that the primary magma must be very
much more superheated than is required on the stoping hypothesis
or than seems easy of explanation, in view of the difficulty of under-
standing how plutonic magma, which is capable of intrusion, can be-
come superheated more than 200° or 300** C.
Fourthly, the stoping hypothesis has the special advantage of
providing a mechanism of thorough agitation within a batholith.
Strong stirring of the mass is induced by the sinking of xenoliths
and by the necessary rising ol the magma locally acidified by their
solution. This agitation can explain the marvellous homogeneity
in each large batholith. It helps greatly to explain the manifest
evidences of magmatic differentiation within batholiths — splittings and
segregations that cannot be due to the slow process of molecular
diffusion or to mere thermal convection. The whole process of stop-
ing and the rising of syntectic magma tends to equalize the temper-
atures in the batholithic chamber and thereby we can understand the
even grain and rapid, nearly simultaneous crystallization of a batholith
throughout its visible depth.
218 IGNEOUS ROCKS AND THEIR ORIGIN
Fifthly, the engulfment of blocks of geosynclinal sediments en-
riches all parts of the batholiths with water, chlorides, etc., which so
greatly aid in solution; while, on the older view, these agents are
confined to the iippi»nnost part of the chamber.
Sixthly, as already noted, the cleansing of syntectic films from
contact of solid and liquid is much the more rapid and perfect accord-
ing to the stoping hypothesis, thus providing and renewing condi-
tions for molecular lowering of the fusion-point along contacts.
In short, the newer view has not only the advantage of better
explaining the facts of the field but it is incomparably more economical
of the heat postulated for the work of batholithic replacement than
is the theory of pure marginal assimilation. Melting and marginal
assimilation of count r>' rock takes place in the initial, superheate<l
condition of a basaltic injection, but must be regarded as alwayit
subordinate in replacement efficiency to stoping and the correspond-
ing type of abyssal assimilation.
Assimilation in Intrusive Sheets. — Probable as large-scale solution
in depth may be, the argument for assimilation has been greatly
strengthened by recent discoveries in the geology of injected bodies.
These igneous masses have floors, which can often be located in actual
outcrop. If both roof and floor as well as the body itsdf are writ
represented in out<Top, we have an approximation to the petrogenistV
ideal. He can never rest assured of his conclusions regarding the
chief chemical processes in abyssal wedges or batholiths until he has
compared the chemical dynamics plainly exhibited in the smaller
intrusions. Since injected Ixxlies are generally narrow and therefore
rapidly chilled, the conditions for their extensive assimilation of
country rock are bound to be rarely represented. Of the total number
of injections sufficiently large, only a few have both roof and floor
preserve*!. This is clearly one of the reasons why the doctrine of
a.^similation has not hitherto won the general recognition it deserved.
It has already attained new dignity in the minds of several petrologists
by ofi'ering the only f(*asibl(> explanation of the chemical facts reoorded
in the few ])ro])erly ex|K)sed. large bodies of igneous rock. One may
doubt that any other line of study will shed light so rapidly on the
ultimate problems of petrogenesis. Speculations regarding the oripn
of the vast, "bottomless'^ intru.sives arc to \ye both suggested and
controlled by ilirect observations on the larger underground "slag-
pots'* which can Ix' examined from top to bottom.
The thi<'k<T sills and interformational sheets are obviously those
injected bodies which are of most significance. They are oflFsboots
from main abyssal injections and they have a characteristic form
whi(*h shows their initial tem|)eratures to have been relatively hi^*
MAO MA TIC ASS I MIL A TION 219
The wide extension of intrusive sheets cannot be explained unless
their magmas possessed some initial superheat, notwithstanding the
partial cooling in the dike fissures or other feeding channels. In other
words, the initial temperatures of intrusive sheets approach, though
probably never reach, the maximum temperatures of the parent mag-
matic wedges. If, then, it can be shown that notable assimilation has
occurred in the sheets, we are forced to credit the possibility of as-
similation on a much larger scale in the abyssal wedges.
The writer has collected the pertinent facts regarding some of the
thicker sheets already mapped and described.^ These and additional
facts are summarized in Chapters XII, and XV to XX. The many
details need not be repeated here. The more salic magma in each one
of these sheets is simply explained by the principle of assimilation and
seems to fail of explanation if that principle be excluded. The chief
motive in reaching the conclusion is the chemical "consanguinity"
between each salic magma and its country rocks.
The assimilation in an intrusive sheet may take place at roof and
floor, or within the mass, by the solution of blocks stoped down from
the roof or, much more rarely, stoped up from the floor. As far as
secondary magma is thus formed, it is relatively easy to discern the
"consanguinity" mentioned in the last paragraph. But the proofs
of it may be obscured, in many cases, by several complications inci-
dental to intrusion of the sill type.
Generally the channels (dike fissures) through which magma is
forced on its way to the sill chamber, are relatively few and are very
narrow when compared with the thickness of the sill. The magma
must flow through the channels during a considerable time in order to
fill the great chamber. At that stage the magma is at its hottest and
is being moved rapidly past the wall rocks of the channels. The
solution of those rocks, in depth, must be stimulated by the movement
as well as compelled by the high temperature. Arriving in the
chamber, the magma is already secondary in part and that portion may
differ chemically from the roof and wall rocks of the sill itself.
Again, the character and field relations of the Moyie sills (see
page 344) suggest that a magma, after forming a thick sill in a sedi-
mentary series, sometimes breaks through the roof and forms a sill
at a higher horizon. The secondary magma developed in the first
chamber may contrast essentially with that which could be due to the
assimilation of the country rocks at the higher level.
In spite of the causes for uncertainty in individual cas^s, the
evidences that the assimilation hypothesis meets this "consanguinity"
» R. A. Daly, Amer. Jour. Science, Vol. 20, 1905, pp. 185-216; and Festschrift
lum siebzigsten Geburtstage von Harry Rosenbusch, Stuttgart, 1906, pp. 203-233.
16
220
IGSEOVS ROCKS ASD THEIR ORIGIN
tost are very Htroni^. It is safe to say that those evidences cannot be
set aside h}' tlic petrologists who l)elieve the differentiation of pure
primary m<iKmas to \h: the only important cause of the diversity in
igneous rocks.
CHAPTER XII
MAGMATIC DIFFERENTIATION
Introduction. — The kernel of the theory so far presented is the con-
clusion that late pre-Cambrian and younger magmas have originated
either in primary basalt or in syntectics of the substratum basalt and
solid rocks. At many points we have noted that this cannot be the
whole explanation of the known variety in magmas and in igneous
rocks. The inquiry as to the nature of the additional causes is clearly
the next step in working out a complete theory. All these supple-
mentary processes, so far as determined, have long been grouped
under the general caption, **magmatic diflferentiation." By this
term is meart the separation or segregation of fractional amounts of a
magma so that at least one sub-magma, chemically contrasted with
the parent magma, is produced. Some authors distinguish "mag-
matic*' diflferentiation from "Krystallizationsdiflferentiation" or
modification of a magma by the separation of the solid crystals formed
early in the crystallization of that magma. This distinction is not
strictly logical, since in the latter case the mother liquor is a true,
segregated magma. Nor is it now possible to be sure that, in many
cases, the segregation of "crystals'' has not taken place when these
were still in the liquid phase. A more valid, or at any rate, more use-
ful distinction is that between diflferentiation through liquation and
differentiation through fractional crystallization. In the first case
the segregation of two or more liquid fractions is the controlling factor.
In the second case the segregation of one or more crops of solid crystals
leaves a mother liquor of a new chemical type.
The history of opinion on this complex theme has been written
by Teall, Iddings, Zirkel, Schweig, Loewinson-Lessing, Harker,
Clarke, and others. Much of the published literature on differentia-
tion really concerns mineralogy and physical chemistry rather than
geology. Yet many biproducta of these studies have geological
bearing and an exhaustive survey of their geological applications
would, alone, occupy a work larger than the present volume.
In this chapter the writer will attempt little more than a statement
of opinion as to the relative geological importance of the various
modes of diflferentiation, and therewith a brief indication of the grounds
for special faith in those processes. For further discussion the reader
221
222 IGNEOUS ROCKS AXD THEIR ORIGIN
is specially refc»rro<I to Loowinson-Lessing's **StutIien ttber die E>up-
tivgestciiK*/' which still remuiiis one of the fullest and most suggestive
of all the g(»n(»ral nu'moirs on diffen^ntiation.*
For clearness it is well to distinguish two stages in differentiation:
first, th(» preparation of units; and, secondly, the segregation of thorn*
units.
The unit of difft nntiation is either a molecule, a solid er\'8tal, or a
small non-eoiisolut<' fluid portion of the magma. The formation of
any on(» of tlu'S<* is largely a chemical matter and to get to the bottom
of it in a given case, the principles of '* affinity," vapor pressure, sur-
face tension, and saturation must l>e applied, and the influence of
time, viscosity, superfusion, etc., must he evaluated. These important
hut endlessly complex probK»ms have already undergone attack; in
the writings of Vogt, Iddings, Harker, Klsden, Loewinson-Lessing.
and others. \\\v n-adcr will find cliscussion as to the degree of contempo-
rary su<*cess in solving them, i'or present purposes the mechanism of
mol(»<'ular devehjpnunt, of crystal growth, and of the development of
immiscihie portitms, need not he discussed in detail. The existen(*e
of su<'h units will he assumed. Their s<>gregation into large masses is
more directly a geological process and will l>e illustrated in greater
detail.
Molecular Diffusion.- Some iN'trologists have tried to rest in
the Ix'lief that difTerentiation could })e explained by the segregation
of individual molecules. A few authors have even suggested that
silicate suli-magmas hav<* he<'n derived hy th(» collection of free oxides
within a ]>rincipal magma characteriz<'d hy molecular dissociation.
However, the d<struetive criti<ism of (1. F. Becker has showed the
almost infinite im])rohal)ility that such diffusive processes have been
directly responsible for the formation of large masses of igneous rocks.'
Fractional Crystallization.- -That crystals of early formation in
a magma are units of difTerentiation is obvious. Their segregation
has been imagined to tak<* plac<' in two different wa3'»r-l)y gravitative
sinking and through convection currents. From the day when Charles
Darwin ])ul»lishe(l his classic "(MH»logi<*al Observations" to the pre-
.^ent time. th(> hyiN)thesis of magmatic differentiation through the
settling of solid crystals has not wanted adherents. Schweig ha#
* F. I^orwiiison-Lcssiiif;, CVtiiiptr Kciidu, 7c SostfioD, Codkt^ Gfokigique
Int<rnalinn:tl(>. St. I'rtt-rshiirp, isilO. p]i. ms 401. See the works listed in the foot-
n(>t<> to t)it> sf'cnnd paragraph of Oiajitcr \I. S<*e n].<(o J. V. EliKien, Principlff of
du'iiiiral (Holo^y, I^nMion. P.)](): J. J. II. Tt-all. Hriti.*«h Petrography, London.
ISSS; V. Zirkcl, Lt'lirhmit di-r Pc1ro^ra])hus \A'\\m^, 1S93, Vol. 1, pp. 711-7!
M. SrliwciK, X(Mir*s Jahrl>iir)i fiir Miiicralo^H*, etr., B. H. 17, 1903, p. 516.
' Ci. F. Becker, Auit-r. Jour. St-ieiire, Vol. 3, 1S97, p. 21.
MAGMATIC DIFFERENTIATION 223
probably made the most general application of the principle.^ At a
volcanic vent, magma may be held for many years at temperatures
above freezing but below the temperatures where minerals like the
olixanes, pyroxenes, amphiboles, etc., must crystallize. Under such
conditions it would seem necessary that diflferentiation through this
type of fractional crystallization should take place. The writer has
thus explained the derivation of p>TOxene andesite from basaltic
magma, though even here a parallel liquation may dominate.*
Becker and Pirsson consider that thermal-convection currents
have been efficient in segregating the crystals of early generation.*
The suggestion has been applied to very few cases where it can be
tested quantitatively. In fact, the only instance of the kind seems
to be that of the Shonkin Sag laccolith in the Highwood Mountains,
Montana. Pirsson has stated the argument in the following form:
"At some period crystallization would take place, and this most naturally
would begin at the outer walls. It would not begin at the top because the
material would arrive there from below at its highest temperature. Moving
off toward the sides the material begins to cool and descend and becomes
coolest as it nears the floor; there crystallization would commence. The
first substance to crystallize is the solvent, which in this case would be the
femic minerals, chiefly augite. Part of the material solidified would remain
attached to the outer wall and form a gradually increasing crust, and part
would be in the form of free crystals swimming in the liquid and carried on in
the current. Probably at first, as the liquid moved inward over the floor of
the laccolith and became reheated, these crystals would remelt, giving rise
to numerous small spots of magma of a different composition, which would
slowly diffuse. As time went on, however, there would be a constantly in-
creasing tendency for the crystals to endure; they would be carried greater
and greater distances. But as they are solid objects and of greater specific
gravity than the liquid, there might be a tendency for the crystals to drag
behind and accumulate on the floor of the chamber. Moreover, from the
heat set free at the time of their crystallization and from the resulting con-
centration of the chemically combined water vapor in the magma, the residual
liquid would tend to have its mobility kept undiminished, since these would
be factors which would tend to counteract the increase in viscosity due to
cooling. In this manner it may be possible to understand how there would
form a femic marginal crust and a great thickness of the femic material at the
bottom of the laccolith. As the cooling went on the edges of the outer crust
would rise more and more toward the top, finally spreading over it, and as a
result the crust would be thinner on the top than elsewhere, as in the Shonkin
Sag laccolith, in which the upper crust of femic rock is still preserved."*
* M. Schweig, Neues Jahrbuch fiir Mineralogie, etc., B. B. 17, 1903, p. 663.
« See page 375, and R. A. Daly, Jour. Geology, Vol. 16, 1908, pp. 401-420.
» G. F. Becker, Amer. Jour. Science, Vol. 4, 1897, p. 257; L. V. Pirsson, Bull
237, U. 8. Geol. Survey, 1905, p. 187.
* L. V. Pirsson, Bull. 237, U. S. Geol. Survey, 1905, p. 188.
224 laSEOUS ROCKS AND THEIR ORIGIN
But ono may woll question that thermal convection could be 80
efficient in distributing the products of fractional crystalliiation in
this laccolitli. Before the intrusion tlie strata now forming itn roof
and floor had practically the same temperature. There is nothing
to in<licat(* that the floor was of a higher temperature than the roof
during the magmatic {N'riod. The laccolith is only 140 feet thick, 90
that it nuist have )H>en frozen U^fore the theoretically more rapid
cooling at the roof could establish any practical difference of tempera-
ture lK»tween roof and floor rm*ks. Again, even if that difference were
100° C., th(* convection gra<lient would l>e less than one-eighth as
steep as that in an equal thickness of water similarl}- heated. Our idea
of the six»ed of convection, in an ordinary l)eaker of ice-cold water
placed over a flame must be drasticall}' mmlified in appreciating the
case of a <*olumn of water 140 Uh^X high. Moreover, the speed of con-
vection is a direct fun<*tion of viscosity. The viscosity of the Shonkin
Sag laccolith, )>eing und(*r pn*ssure, couNl hardly have l>een leas than
fifty times that of wat<T at atnu)sphere pressure, and the ratio more
probably ran into the hundreds of thousands. It is no exaggeration
to say that the sfx^MJ of currents due to pure thermal convection in
this laccolith must have than lKi*n millions of times less than that in a
hot-air furnace, and many thousands of times less than the speed of
convection in an ordinary b< akcr in the lal)orator3\ In this matter,
as so often in geological dynamics, we must think to scale.
Finally, the time available for convective distribution in the
Shonkin Sag laccolith was v<Ty limite<I. In a bod}' so thin, even if
initially sufxThcated as mu<'h as 500^ (\, conduction into the cool
rocks above and ) clow would certainly chill the mass to the point of
prohibitive viscosity within the iHTicxI of one year. If, as ia more
probabl(\ the initial sui>erheat was no more than 100^ C, thermal
convection would doubtless ))e unable to affect the composition of
the lO-foot contact phase of the laccolith after the lapse of one
W(M»k.
We <M)n<hnl(» that, from the slowness of convection and the short-
ness of the magmatie life of this laccolith, the basic contact shell of
the Shonkin Sag laccolith cannot l)e explained by the combined in*
fluence of tluTmal convection and fractional crystallization. The same
general argument and th<' same con<*lusion apply also to the Square
Butte laccolith, for which Pirsson has evolved a similar explanation.
Kvcn for large lac<-oliths one may well question the efficiency of this
t hernials on vtM't ion hyfxithesis on the quantitative side. It is also
weak on the chemii al s'ulv. In no <l(*scril>ed case is the basic contact-
pi. ase at roof or wall of an intrusive lx)dy of the composition expected
on this hypothesis.
MAGMATIC DIFFERENTIATION 225
Liquation. — Ostwakl points out that the number of liquids mis-
cible only wnthin definite limits is much greater than is the number of
those which mix in all proportions.^ Since magmas are solutions, it
is a priori wise to consider their possible differentiation through the
principle of limited miscibility at certain temperatures. Though
Vogt has held that this principle does not, in general, apply to silicate
mixtures, one of his latest publications contains the statement that
magmatic differentiation consists in the splitting of liquid phases,
which splitting is controlled by the laws of eutectics.^ Unless the
^Titer misapprehends his meaning, therefore, Vogt has come to re-
cognize limited miscibility as a general law for silicate solutions as
soon as these approach the consolidation interval of temperature.
Richards and Ostwald believe that solid crystals develop from a
transitory liquid phase in the case of substances which melt at tempera-
tures not far from their respective temperatures of crystallization — ^a
condition realized in the case of most rock-forming minerals.' From
specially designed experiments Slatowratsky and Tammann have con-
cluded that, for naphthaline, yellow phosphorus, CaCl+H20, potas-
sium, sodium, or ice, the passage from the solid phase to the liquid is
marked by ** plastic crystals,'^^ Dittler adopts this view also for the
silicate, anorthite.^ Schade's microscopic studies showed that the for-
mation of crystals of cholesterol from an alcoholic solution is preceded by
a separation of the substance in the form of liquid drops. He observed
similar indications for solutions in ethyl ether and in oils, as well as
when the pure molten substance was rapidly cooled. The freshly
formed, acicular crystals were very plastic, but this plasticity dimin-
ishes with time.* Von Weimarn holds that the liquid-crystalline
state is a general property of matter at the appropriate temperature.'
Quite recently Buchanan has described a remarkable series of ob-
servations on saturated solutions. He used a saturated solution of
calcium chloride and water as a type. It was kept at a constant
temperature and at intervals the density was very carefully deter-
mined. The result was to prove a dilatation of the solution for a
considerable time preceding the appearance of the first solid crystal
of calcium chloride. This change of volume is of the same quality
» W. Ostwald, S.>lutions, 1891, p. 39.
* J. H. L. Vogt, Videnskabs-Selskabets Skrifter, I, Math.-Naturv. Klasse,
.\o. 10, Christiania, 1908, pp. 6 and 102.
» T. W. Richards, Phil. Mag., 1901, p. 500.
* X. Slatowratsky and G. Tammann, Zeit. f. phys. Chemio, Vol. 53, 1905, p.
341.
* E. Dittler, Tscher. Min. u. Petrog. Mitt., Vol. 29, 1910, p. 273.
* H. S<^hade, Koll. Chem. Beiheftc, Vol. 1, 1910, p. 391.
^ P. P. von Weimarn, Zeit. Chem. Ind. KoUoide, Vol. 3, 1908, p. 166.
226 IGNEOUS ROCKS AND THEIR ORIGIN
as that produced in the system by actual crystallizatioii. Baehanan
conclu<Ie8 that the ''stretching" of the solution, before a solid phase
is formed, reflects a ''new creation" in that solution. One may infer
tliat he means a partial concentration of the calcium chloride about
centers.*
I'hose analogies suggest that the solidification of a very slowly
cooled intrusive magma may \ye preceded by a long period in which
one or more chemical components form small liquid globules or
crystals.
More generally, we must also allow for the possibility that a ho-
mogeneous magmatic solution may IxH^ome broken up into non-eonso-
lute portions in a certain region of temperature and pressure. Im-
mediately after the formation of these fractions, the magma would
)>e an emulsion. The often-repeated statement that the dominant
rook-forming materials are miscible in all proportions has sddom
been properly guarded. It may \)e quite true for high temperatures
and yet quite untrue for a temp(Tature just alx)ve that of crystalli-
zation of a given component. No one has yet succeeded in holding a
molten mixture of silicates within this narrow range of temperature
for a l(*ngth of time suffici(*nt to warrant any definite condurion on the
matter.
On the other hand, there are some facets suggesting that limited
miseibility ha.'^ actually characterized natural magmas just before their
final fn^'zing. Most basic segregations and probably all orbicular
granites, diorites, and gahbros are direct evidences of the emulsion
stage. The common banding of neplielite syenite, the banding of
certain gabbros, the phenomena of some differentiated dikes (Ent-
mischte (fiinge) are other illustrations of true magmatic splitting.
The constitution of the Moyie sills or of the Sudbury sheet (see
Chapter XVI) is inexplicable except on the assumption of the limited
miscibility of granitic (micropegmatitic) and basaltic (gabbroidy nor-
itic) magma under certain conditions. In this connection Bick-
stromV point that there is a lack of intermediate rocks in the liparite-
basalt field of Iceland has great significance.' The sulphidic ore of
Sudbury l>ecame non-<*onsolute with the norite just as matte is non-
consolute with its slag. Partial immiscibility is illustrated even in
the brief period of liquidity allowed to artificial glass in the factory.
HarkerV objection to this application of the principle ot inunis-
cibility at certain temix^ratures is that it involves disconUnuous
variation lu'tween <lifferent parts of a single rock-body instead oi the
actually observed continuous variation. But, in the first place, the
* J. Y. Buchanan, Trans. Roy. Soc. Edinburgh, Vol. 49, Part I, 1912» p. IM.
* U. Dackstrdm, Jour. Geology, Vol. 1, 1893, p. 773.
MAGMATIC DIFFERENTIATION 227
separation between such silicate differentiates is in many cases re-
markably sharp, especially when we consider the scale of operations in
magmatic chambers. Secondly, we could hardly expect the separation
to be as perfect between these viscous and highly complex magmatic
fractions as, for example, the separation between phenol and water.
In summary, it may be stated that a host of field and laboratory
observations favor the application of the liquation (limited misci-
bility) principle to natural silicate magmas; and that not a single
fact is known to the writer which conflicts with that assumption. The
efforts of physical chemists should be spent, not on denying its valid-
ity, but in defining the conditions under which the liquation so often
demonstrable in nature has taken place.
Loewinson-Lessing's suggestion, that the equilibrium of a homo-
geneous magma, at nearly constant temperature, may be disturbed
by the solution of a small quantity of the country rock, is worthy of
close attention.^ Linebarger, Duclaux, and others have shown that
a solution of each of many colloids can be made to coagulate or gela-
tinize by the addition of a mere trace of a certain substance.* This
analogy aids the imagination in following the process described by
Loewinson-Lessing, which so far lacks full experimental proof.
Gravitative Differentiation. — The sinking of crystals is expected
to have its maximum differentiating effect within volcanic vents
where the agitation of the magma tends to prevent undercooling and
to promote crystallization, while the magma retains relatively low
viscosity. These conditions are chiefly due to the upward passage of
gases in volcanic vents, which in this respect are contrasted with dikes,
sheets, and laccoliths. The steady or intermittent passage of hot gases
through the lava columns at surface vents is competent to keep the
column long within the temperature interval of crystallization. (See
page 288.) Nevertheless, liquation may co-operate even in this case.
In general, it is probably a much more efficient cause of differentiation
than fractional crystallization but the relative importance of the two
processes can be estimated only after the physical chemistry of magmas
becomes better understood. Meanwhile we may use the expression
^^gravitative differentiation^^ as a name for the chief mode of magmatic
separation, without implying that fractional crystallization or liquation is
the more active in a given case. However, experiments like a well known
one by Morozewicz favor the liquation hypothesis.' (See p. 363.)
' F. Loewinson-Lessing, Compte Rendu, 7e session, Cong. G^l. Intemat.,
St. Petersburg, 1899, p. 377.
* C. E. Linebarger, Jour. Amer. Chem. Sec, Vol. 20, 1898, p. 375; J. Duclaux,
Comptes Rendus, Vol. 138, 1904, p. 144.
* J. Morozewicz, Tschermak's Min. und Petrog. Mitt., VoL 18, 1898, p. 232.
228 laXEOrS ROCKS and their ORIOIN
Differentiation at Central Vents.—TIiiM'ontrol of gravity is suggested
in tilt- vi»l<-anws of Kruriicin, Hawaii, and tlic Juan Fernandei Islands.
During till- 1874 i'ru)ition in Ii(^unioii, lava flows issued simultaneously
from tlic summit of tlic octivv i-onc and from a tatoral fissure which
doul)tli>)>s i-ommuniiatcd witli Itii- main vont in depth. The sutninit
flow wsM an auKitc anilcsit<- with .*)7.41) prr cpnt. of nilica and a Bperifir
gravity of 2.7U. Tlic lava from the flanking fiwture was a basalt rich
in olivine, with a silica percentage of 48.98 and a specific gravity of
2.07.' The base of Muiina Kea, Hawaii, is rhiofly a pile of flows of
olivine hasalt. On as<-cnding the eone that type is succeeded, in
order, tiv olivine-puor hasalt, augile andmtr, and, at the i
n lit Ml. JohnMin, Quebec. r.\rtrr F. D. Ailams, Jour. GtnL.
Vitl. 11. 1(MI3. |>. :!.V>i. /..S. IjincrSiliiriiin M^limciitxi /, pulpakit4>;{, i "^ "
riM'k; S, i-A-i'xite; .1, i-iuilurt :iiirm|p in LS.
trachyiliilerilie lava with a silica [xTccntnge of alwut 51 and a s|
gravity (lioliicrystnlline pliase) of 2.7(>.' Similarly, Qucnsel has found
that the lower-lyiuK laviis of Masafuera, one of the Juan Pemandei
Istaiiils. arc liasic olivine ami plnKiocla.-«> ha.ialt8, nuccccded above by
I)asaiiitic lava, which in turn is overlain liy sfxta-trachyte with a
silica jMTcentaKe of (i;i.4;i.'
.Vdams regards Ml. Johnson, QucUr. as a volcanic neck. The
conecntrii- <level<iiimcnl of pulasktic, transition rock, and CAaexite is
I C, V.'liin. Missii.ii .ir III.- Siiint-I'mil. I'arw, \*i^S. \t. 1X1.
> It. X. n:ily. .Ii.iir. ..f <;.-.)i..B.v. Vol. 111. 1911, p. SUT.
' P. D. Quunxd, Uiill. licul. Inot. I'putltt, Vol. II, 1912, p. 288. .
MAGMATIC DIFFERENTIATION 229
explained by gravitative differentiation. The outside collar of pul-
a^kite is thought to represent the salic pole of an early magmatic
splitting in the vent. Somewhat later the heavier essexite magma,
the femie pole, was thrust upward, displacing the pulaskite in the
axial portion of the vent' (Fig. 120).
Chemical Contrast of Plutonic Rock and Corresponding Effusive
Type. — As noted in Chapter II, an effusive magma is usually richer
in silica, soda, and potash, and poorer in iron oxides, lime, and mag-
nesia than the corresponding deep-seated rock belonging to the same
clan. This important fact is illustrated in Table II. The chemical
contrasts between the respective pairs of rocks are explained by the
special conditions at a volcanic vent of the central type. As just
noted, such a vent is a place of special concentration of magmatic
gases. These fluxes lower the freezing temperature of the lava, thus
tending to lengthen the temperature interval in which the femic con-
stituents may individualize. For a second reason gravity has special
efficiency in differentiating the lava column; a central vent generally
has many alternations of dormancy and activity, often passing through
the temperature interval (just above the point of complete freezing)
where individualization of the femic minerals takes place. On
account of their higher specific gravity, the early-formed substances,
whether in the solid or liquid phase, must sink in the lava column
which, in its upper levels, becomes more salic than the original plu-
tonic magma. An actual surface flow at a surface vent generally
comes from the upper part of the lava column. The general chemical
relation of plutonic and affusive in each clan seems, in fact, to be a
strong evidence of density control in differentiation.
Differentiation in Laccoliths and Intrusive Sheets. — Many thick
laccoliths and sheets show tellingly not only the reality of large-scale
assimilation but also the nature of the differentiation processes in
primary and syntectic magmas. The following table (XIV) lists more
than seventy injections, mostly sills and laccoliths. In twenty-nine
of them the control of gravity is evident and it has probably charac-
terized six others at least. In most cases the available evidence sug-
gests that the units assembled by gravity were liquid fractions, non-
consolute at the moment of differentiation. The splitting is sugges-
tively analogous to that seen in the stratified arrangement of colloids
in water which has stood undisturbed for some months.*
Gravitative control over the differentiation is indirectly illustrated
in the upward transfer of some of these fractions, including aplitie
» F. D. Adams, Jour. Geology, Vol. 11, 1903, p. 281.
*Cf. R. S. Symmonds, "Our Artesian Waters," Sydney, N. S. W., 1912, pp.
42-50.
IGNBOUS ROCKS AND TBBIK OSIGIN
■Mi
i i
III :
-5 ^< y- M
SI
If
ill
l|l
_ « 5
■lili
3
i I
-i^S If
■ £ « » -0 5
I slil 9-e
s|i5ll3 II
9S
ifil
s iS g
I JS8
li
Wl
\i s
'3
+ +
§
I'll
' i
i 1
j
Locality
i. S u d b u r y ,
Ontario.
S. Cobalt Lake
district, On-
tario.
«. Pigeon Point,
MAGMATIC DIFFERENTIATION
231
e
a
00
X
Z.
<
X
n
O
w
<
*5 "^
+
C^
4-
o
CO
+
o
?-
CO
2 X
]
r
I
X ^
Q
o
.4
o
o
4-
o
o
B
*-•
o
4-
co
CO
o
CO
3 3
o "^
o -55
OQ
OS
>
I
a
OQ
CQ
OQ
02
o
a
OQ
-a
3
I « 03 O
> M > O
C •^ oo .2
-rt C ^ *» -O
I « ;^ .2 'C a
1 « S S ffl 3
oO
OS (h
r '♦^
OQ
o
o
o
I:
o
I
a
o
OS
43
OS &
OQ
— •-•08
fe (3 I ^ (2
a
o
O
o g
8 -8
^1
inXSOVS ROCKS ASD THEIR ORIGIK
i§ll
' I1 1
J 11
Till IsS^l
111
?1
I itllllliiillilillill llllil
^ fc.a
1 5 «i
11
3
II
3 Sf
3 4^
•■ i
5 11
:5 J ^ti
a ill
111 llii li
i ^ a li
MAGMATIC DIFFERENTIATION
233
9
e
'^
c
I
X
C
c
C5
X
c
'3
a
ec
33
2
eS
O
X
eS
O
o
O
O
•M ^
o
00
eS
03
00 O) -^
I -§ ii 1 2 a -a
o 'O CO M o ^
CO
o
o
c
OS
c3
00
O
o
G
08
43
0)
0^
0)
00
O
N
"♦J
0)
eS
a
a
o
.a
a
o
43
0)
•4^
3
0^
*-• /< d
o P O
0)
•a
X
o
o
*-•
Xi
Xi
03
tA
c
O . a>
»; gc X
X g O
•4^
c
o «) a
•*; •i>^ •»-<
03 OD L<
o| g
s -^
a
^^1
s
a s
^4^ S -4^
00 efi oQ
W -4^ W
43 -^ 4a
.!_» ^^ .<J
O pO
^ o
< ^<
o .2
^1
>
eS
•4^
>
03
o
>
>
o3
o
>
OS
o
>
%
d
I
OS
-s ^
2 .t^
oc
e! 2
o
-* 00
o g ;a
^ :3 bC
t. oo 03
o
I
I
«> -
00 -*-*
03 -4J ^
^ CO
.« C 'T3
O ^ O
N 0) •4->
o c: 03
« o -2
a
a
0)
ec
4^
a
a
.2 «5
as
I'M s
>• 0« OQ
.1 a
tA . a>
lee
am
SOQ
<4J
•55 I
^ 00
o8 ^
a cT
05 ■*£
.—4 ?3
GO Q
•4J "
^ s
^ •= .S :?!
<
o
00
CI
o
o
<
CO
o
u
1^
o
00
+
o
O
I P--
o
0 •
•-3
I
g S
^ ^ CT ^
cc
p-
;3
o
o
43
;3
o
43
o
1-^
'A
O
QQ
•J
"Si
' O tA
•*• C
X s •
fc. ^ o
O 3 Jt
CI4 cO
!e«
I
<
O
a
3
00
0;
OJ o
PL. "S
•C o
5^
a
03
»4
« o ®
3 • i i — J C S O
M gO-tJW>^Pu,HOO
c
O
2
o flS
to "d
s -s 6 =§ I « -s 3
M
CD
e«
00
M
«0 ••
; I
I
'■V
S
i
1 I
lONBOUS SOCKS AND THEIR ORIOIS
h
Sit °1
1 1 iilliHiyi!
•5.3
il
11
.5
IS
•3 2
I'll
lill
li
ill J
ii
1! Ill If
3 Si
i |i 1
i II3JJI
5 - " 1 •= »■ .
8 J 3
MAGMATIC DIFFERENTIATION
235
I
f
(£
z
u
o
o
a
o
z
H
00
oo
B
H
N4
iJ
O
o
o
Q
z
<
3
00
I
>
a
la
S
a
CO
M
O
I
o
a
s
a
08
•a
3 p « oT
03
^ a
'B ^ •
P &5
II
♦a
IS
o
t:
o
ii
3
-« -r
3 0)
Q .O
O >0 «d
3
S
;3
a
CO -»^
o8
U)
^ &.■§
» 8 S
5 g 8
'B ^ 2
S s g
CD dj C>
<h 08
2*1
C8 a>
0;5
oo
>
o
o
>
>
08
o
>
>
eS 08
4^ C- -4^
gag
o ^o
CO
a
'■13
0)
•s
a
u
>
>
08
; CO
•a
!• •=
a
3
o
* 15
JJ » -a 5
■g S s-s
§5 S ^
OQ j^ « '53
S 3
^ a
s ^ 5 ^ a
55 5i 3 o -3
5
0
a
if
'8.2
I
I
8
lO
I
o
11
o
a:
CO
" a
S /-A
Si •-
® 5 8
08
QQ "^
So
17
236 IGNEOUS ROCKS AND THEIR ORIGIN
and sodalitic magmas, which have risen in the chamber with the aid of
magmatic g&ses. Even thick laccoliths of the Henry Mountainn type
— highly viscous and relatively cool at the time of injection — ^will not
be expected to show pronounced gravitative splitting in place.
At least sixty species of igneous rocks, besides tranutional and
hybrid tyjx's, arc represented in the list. All of the plutonic familieA
quantitatively inifmrtant in the world — granite, granodiorite, diorite,
gabbro, anorthosite, sj-enite, foyaite, peridotite — are represented; and,
in addition, many of the rarer families — analcitic and leucitic types,
essexite, theralite, teschenite, urtite, ijolit^, jacupirangite, lujavrite,
shonkinite, lK)rolanito, magnetite ore, chromite ore, sulphide ore, ete.
This great range of magmatic tyjKNs is a principal indication of the
significance of these injections for petrogeific theory.'
Lkading Heferenceb
Nunil>or8 printcnl in l>oId type refer to the injected bodies listed in the table.
1. A. P. Coleman, Ann. Hep. Bureau of Minos, Ontario, Vol. 14, 1M5; Jour, of
Geol., Vol. 15; 1907, pp. 750-782.
2. \. L. Bowen, Jour. Cieol., Vol. 18, 1910, p. 658.
8. W. H. CoIIina. Econ. CSeol, Vol. 5. 1910, p. 638.
4. W. S. Hayley, Hull. 109, U. S. Gof)l. Survey, 1893; A. C. Lawaon, BulL 8, Geol.
and Xat. Hist. Survey, Minned4>t.% 1893, pp. 30, 31, 44.
6. C. H. Van Hise and C. K. Ix>ith, Mon. 52, U. S. Geol. Survey, 1911, pp. 2Q2 and
372; X. H. Winchell and othere, Final Hep. Geol. and Nat. Hitt. Sunrej of
Minneflota, Vol. 5, 1900, p. 978, and Vol. 4, 1899, Platea 66-<M), p. 302, ete.;
W. S. Bayley, Jour. Oe<»l., Vol. 2, 1891. p. 814.
6. C. H. Van liine and C. K. lx^ith, Mon. 52, U. S. Geol. Survey, 1911, p. 877.
7. H. .\. Daly, Memoir 38, C;e<>l. Survey of Canada, 1912, pp. 221--265.
8. S. J. Srhoficld, Summary Hep. (k^I. Sur\'ey of Canada, 1910, p. 181; alM
abfltract of doctorate (hesLs published by the MaaBaehuaetia IiiBlitaie of
TerhnoloRy, 1912.
9. F. C. Calkins, Bull. 384, U. S. Geol. Survey, 1909, p. 50.
10. Ibid.
11. J. T. Pardee, Bull. 470. V. S. (Seol. Survey. 1911, p. 47.
12. J. V. I^wis, Ann. Hep. State G(H>loKiHt of New Jersey, 1907, p. 90.
18. G. F. LouRhlin, Bull. 492, V. S. (;eol. Sur\'ey, 1912, p. 78.
14. A. G. HoKbom. Bull. Geol. InHt. I'niv. ri)8alA, Vol. 10, 1909, p. 9.
16. A. (;. Iloffborn. G(m)1. Form. Str>rkholm Forhand., Vol. 31, 1909, p. 809.
16. G. W. Tyrn»ll, Geol. Mag., Vol. C, 1909. p. 299.
17. J. B. Harriiw)n, Geology of the Goldfields of British Guiana, London, 1909,
pp. 22. 92.
18. G. A. F. Molengraaff, Gc^log>' of the Transvaal, Edinburgh and Johanwburg,
1904, p. 42.
19. A. L. du Toit, l.'>th Ann. Hep. (;eol. Comm. Cape of Good Hope, 101Q,p. III.
20. F. H. Hateh and (;. S. Corstorphine. The Geology of South Afriea, London,
1905, p. 172.
»See note at end of Table XIV.
MAOMATIC DIFFERENTIATION 237
21. G. T. Prior, Annals of the Natal Museum, Vol. 2, 1910, p. 150.
22. C. Viola, Bol. R. Com. geol. d'ltalia. Vol. 23, 1892, p. 105.
23. J. S. Diller, Port Orford folio, U. S. Geol. Survey, 1903.
24. L. F. Noble, Amer. Joiu". Science, Vol. 29, 1910, p. 517.
26. F. L. Ransome, Prof. Paper, 12, U. S. Geol. Survey, 1903, p. 85.
26. J. P. Iddings, Mon. 32, Pt. 2, U. S. Geol. Survey, 1899, p. 82.
27. N. L. Bowen, Ann. Rep. Bureau of Mines, Ontario, 1911, p. 127.
28. F. Di* Adams and A. E. Barlow, Memoir No. 6, Geol. Survey of Canada, 1910,
p. 153.
29. F. D. Adams, Ann. Rep. Geol. Survey of Canada, Vol. 8, Pt. J, 1898.
30. A. E. Barlow and others. Report on the Geology and Mineral Resources of
the Chibougamau Region, Quebec, 1911, p. 156.
31. C. F. Kolderup, Bergens Museums Aarbog, 1903, No. 12.
82. A. Geikie and J. J. H. Teall., Quart. Jour. Geol. Soc, Vol. 50, 1894, p. 645.
83. A. Harker, The Natural History of Igneous Rocks, New York, 1909, p. 140.
84. G. T. Prior, Annals of the Natal Museum, Vol. 2, 1910, p. 147.
85. L. V. Pirsson, Bull. 237, U. S. Geol. Survey, 1905.
36. Ibid.
87. J. A. Allan, Abstract of doctorate thesis published by the Massachusetts
Institute of Technology, 1912.
88. G. W. TyrreU, Geol. Mag., Vol. 9, 1912, p. 75.
39. R. Campbell and A. G. Stenhouse, Trans. Edinburgh Geol. Soc, Vol. 9, 1907,
p. 121.
40. G. W. Tyrrell, Geol. Mag., Vol. 9, 1912, p. 122.
41. Ibid., p. 76.
42. Ibid., p. 70.
48. S. J. Shand, Trans. Edinburgh Geol. Soc, Vol. 9, 1910, p. 376.
44. H. J. Lowe, Geol. Mag., Vol. 5, 1908, p. 344.
46. N. V. Ussing, Geology of the Country around Julianehaab, Greenland, in
Medd. om GrSnland, Vol. 38, 1911.
46. W. Ramsay and V. Hackman, Fennia, Vol. 11, No. 2, 1894.
47. H. S. Jevons, H. I. Jensen, T. G. Taylor, and C. A. Stlssmilch, Proc Roy. Soc
New South Wales, Vol. 45, p. 445, and Vol. 46, p. Ill (1912).
Some of the bodies illustrate a principle which seems certain of
increasing emphasis, namely, the influence of contact chilling in for-
bidding differentiation. The contact phase, a more or less continuous
shell, thus represents the original magma. The other phases, enclosed
by this shell, are the products of its splitting. The development of
chilled phases in dififerentiated sheets has been described by Lewis
(Palisades of New Jersey), by Jevons and others (Prospect intrusion
of New South Wales), and by the present writer (Moyie and other
sills of the Purcell mountains, British Columbia).
The writer has also suggested a similar explanation for the leucite-
basalt porphyry enclosing the syenite and shonkinite of the Shonkin Sag
laccolith.* The middle of this body shows a section described by
Pirsson as follows:
^Memoir No. 38, Geol. Survey of Canada, 1912, p. 772.
238 IGNEOUS ROCKS AND THEIR ORIGIN
ThiekneM in Fwi
a. Lcucitc-basalt porphyry 5
h. Dense shonkinite 5
r. Shonkinite 5-6
tl. Transition rock 3
e. Syenite . 25-30
/. Transition rock ... 15
g. Shonkinite 60»75
h. Leucite-basalt pon>hyry 15
Total 140 (nearly)
Pirsson ha.s calculated the approximate average composition of the
laccolith, with the result shown in Column 3 of Table XV. Column 4
gives the composition of the leucite basalt abundantly extmded in
the Highwood Mountains. Columns i and 2 respectively show the
compositions of the syenitic and shonkinitic differentiates.
I
TAHLK XV
2
3
4
SiO,
50.0
47 9
48.0
48 0
Al,(),
10 4
12.1
12 4
13 3
FojO:
:i.9
3 5
3 5
4.1
F<0
2 7
4.8
4.7
4 2
Mk<)
2 2
8 (i
8 3
7 0
CaC)
5 0
0 4
9 2
9 3
Xa:()
3 Tl
3.0
3.0
3.5
K:0
S 5
5 a)
5 8
5 0
As already stat(Ml, Pirs.son explains the various rock-tjrpes by a
combination of crystallization, thermal convection, and settling-out.
The influence of convection seems to l)e an unnecessary postulate.
An alternative conception of the differentiation is suggested by the
chemical natun* of the average rock in the laccolith. Let it be as-
sumed that a leucite-basalt magma, such as elsewhere in the region
forms volcanic mass<\s, was here injected. On all contacts of the
laccolith, though particularly at its rim, this magma froie quickly.
Th(» interior part, much longer fluid, was cooled until it reached the
temperature of liquation or the tempc^rature of initial crystallisation.
Then the settling-out of phenocrysts or (preferably) of the correspond-
ing units of liquation caused the density stratification, with syenite
overlying the other pole of differentiation, namely, shonkinite.
The small chemical differ(*n(*e In^tween shonkinite and leucite
basalt would make it very hard to prove that the "shonkinite" shells
of b and c in Pirsson^s section do not really form a granular continuation
of sh(>ll a. All three shells s^K^m, in fact, to represent the original,
magma, which has differentiated in the center of the laccolithi pving
MAGMATIC DIFFERENTIATION 239
shells dj Cj /, and g. The analyses of b and c have not been published,
but in any case their analyses would fall within the limits of variation
assignable to leucite basalt.
The same explanation of the likewise celebrated Square Butte
laccolith is feasible, though the loss of its roof by erosion makes a
final test of the hypothesis here impossible. Tyrrell has conceived a
similar mechanism for the Lugar sill of Western Scotland, in which
T
Fig. 121. — Diagrammatic, longitudinal section of the Lugar sill, Scotland.
(After G. W. Tyrrell, Geol. Mag., Vol. 9, 1912, p. 75.) TH, theralite; T, teschenite;
P, picrite. The pill illustrates gravitative differentiation and the forming of
chilled phases. Length, 3.5 miles; thickness, 140 ft.
picrite and various phases of teschenite are associated (Fig. 121).
This sill is about 140 feet thick. The rock phases are listed as under:
Specific Gravity
Top marginal phase — black, "ba-
saltic teschenite" ....
Teschenite, coarse, highly analcitic . . 2 . 64
Teschenite f normal 2 . 70
Teschenite f camptonitic 2 . 98
Teschenite, monchiquitic 2.99
5/8 of sill Picnte 3.01
Teschenite, camptonitic 2.81
Teschenite, normal 2 . 77
Basal phase — black, ''basaltic tes-
chenite"
Upper one-fourth of
sill.
Lower one-eighth of
BiU.
The original magma was here a teschenite. It was chilled at the
contacts, giving "sharp margins of basalt, both at the top and bottom."
Strongly analcitic teschenite ("theralite") and picrite are the poles of
the gravitative dififerentiation in the interior.^ Still another example,
on a great scale, seems to be found in the gabbro-anorthosite mass in
the Adirondack Mountains (Fig. 122).
As a rule the chilled phase at the roof merges into the salic phase
of gravitative dififerentiation. This relation has often been incorrectly
described as "contact basification." The diflSculties inherent in the
application of Soret's principle or of the principle of fractional crystal-
lization are seen to have given needless trouble in petrology.
» G. W. Tyrrell, Trans. Geol. Soc. Glasgow, Vol. 13, Part 3, 1909, p. 298; Geol.
Mag., Vol. 9, 1912, p. 75.
240
mxEoia ROCKS asd their okioix
In ftcncrul, the larger a bodj- the; more advanced ia gravita-
tivc splittiiif;. The liuf^ Iluslivchlt, Diilutb, Sudbur>', Ilimaumk.
Cliilioiiifaiiiaii. iiiiii Muriii I)cniip» have lx*n rospcctivoly split intu
liishly ^:lli(■ atiil highly fi-niic siilmiaKmas, licvHopwl on the larit**
s.!ih-. The thiiiniT I'uni-ll, New Jtrsi-v, Natal, Scottish, and Aiis-
Ki.i. I---'.- M.-.|i "f V-itt i.f Il>.- T^mit Ijik.- ciiiii.lriiimlp. Now Ywk. iAfl«
H, r. rii:.|,iMu. Holt U:.. N v. Slal.- Mii-.i,i,i. inoT.i (;*;, CrmviUe M-nn
)iiiii>tu[ii's :i[i.i i;iiri-.-<-.- : .1, ;iiiiirili<isii<-: Mi, Kiil>lin>iil iimtnrt phoM of .4: N,
«y<-nit.'; SH. Ii:>ir |il.:i.-.- .if tli.- ^v.iiit.-. Ml .i|.|"-;in. to l.e a rhillnl plUM of A
traliiiii iiiji'i'tiiiiH an- less thun Highly ilifTon-ntinled. ThU rule tiU|t-
Kl■^tw thi' nn-i.ii fur the falir atul liomo^-m-ous nature of anrmal
i.!itli.,lill.i.' r..rk>.
nil itii. iiili.r hiui'l, sonif uf the thiiimr sheets art' more or k-«
i]ru.-iii-:il]y ihtTrnntiati'tl. 'rhii-r at Shoiikin Saj;, Square Butte,
MAGMATIC DIFFERENTIATION 241
Ice River, Lugar, and Cnoc-na-Sroine (Loch Borolan) show that
alkaline magmas must have relatively low viscosity in spite of rapid
chilling and in face of experimental proofs that many artificial alkaline
melts are highly viscous. The contrast suggests that these natural
magmas have been specially charged with volatile fluxes — a concep-
tion supported by the mineralogy of alkaline rocks and one clearly
implied in the hypothesis that these magmas are derivatives of sedi-
mentary syntectics.
The published descriptions of most of the igneous bodies listed in
the table illustrate the "freezing-in^' or '* fixing" of small masses of
one differentiate in the crystallized equivalent of its complementary
magma. Thus, the micropegmatitic roof differentiate in a Purcell,
Sudbury, Minnesota, or South African sheet always overlies a gab-
broid or diabasic phase carrying interstitial micropegmatite or schliers
or *' veins'' of the same material. These have evidently been trapped
during the solidification of their respective hosts. A large-scale
parallel is found in the 'transition'' or V intermediate" rocks in differ-
entiated injections. Sills and laccoliths evidently throw light on the
origin of many plutonic, hypabyssal, and volcanic species which are
found only in small volumes.
The anorthosites of the world are best regarded as differentiates
of gabbroid (basaltic or diabasic) magma. The positions of the
anorthosite masses in the Duluth laccolith and the Thunder Bay
sills suggest some degree of gravitative control over the splitting.
The described field relations of the large Bergen, Morin, Glamorgan,
and Chibougamau bodies strongly indicate their laccolithic nature.
In each of them the anorthosite is specially developed at or near the
roof, while pyroxenites, hornblendites, peridotites, or iron ores are
found at or near the floor. The writer has, in fact, come to suspect
that all anorthosite occurs in injected bodies of the sill, laccolith,
chonolith, or dike type, and that even the enormous masses found in
Quebec, Labrador, New York State, Norway, etc., are similarly not
to be regarded as "bottomless" batholiths. Combining the facts
known about the anorthosites, and especially considering the bodies
listed in the foregoing table, the most promising hypothesis remains
that the larger masses are gravitative differentiates of gabbroid
(basaltic) magma. The relatively minute masses found in the banded
gabbros of Skye and other regions, and as schliers in many small
bodies of gabbro, are clearly local differentiates "frozen in" before
gravity could assemble them in thicker sheets. Sills and laccoliths
of pure anorthosite amply show that the splitting has often taken place
in depth, before injection at visible levels. (See p. 32L)
Because of their limited supply of heat, most sills and laccoliths
242 IGNEOUS ROCKS AND THEIR ORIGIN
have not assimilated important amounts of their country rocks.
Yet the flat shape and great horizontal extension of many of these
intrusives (all originally of basaltic composition) prove low magmatic
viscosity, which mcan«s some degree of superheat. Large miperbeated
injections must dissolve the invaded rocks to some extent. There is
a tendency toward the formation of a chilled phase in the magma,
whereby tin* country rock is protected against assimilation, but this
tendency may Ih' checked by the stirring of the magma during injec-
tion, by magmatic stoping, by ''two-phase convection/' and at the
roof by the rise of volatile fluxes. All of these conditions are likdy to
affect such enormous bodies as the Sudbury sheet, the Bushvddt
laccolith, and the Duluth laccolith. The chemical character of the
invaded formations is obviously important; if they are calcareous or
notably hydrous, the fluxing of the original magma at contacts is
fa<*ilitated. The available field data show that the feeders of sill or
laccolith are c()mf)aratively narrow. The passage oi the original,
hot magma through these channels must take considerable time, and
thus allowance should 1h* made for possible assimilation during as well
as after injection. In the writer's opinion, all of the non-basaltic
( non-gabbroid) rocks in the bodies tabulated have originated in syntec-
tics. Faith in this conclusion cannot be won from the study of con-
cordant injections alone, but it is significant that a considerable
numl>er of these are des(Til>ed as having been active solvents of their
country rocks. Such is the case for the following:
Sudbury .sheet (A. P. Coleman).
Pigeon Point sheef (\V. S. Bayl(\v and A. (\ Lawson).
Moyie and other Purcell sills (K. A. Daly).
Bonner's Kerry and Flathead River sills (F. (\ Calkins).
Duluth laccolith (X. II. Winchell and others).
Insizwa sheet (A. L. du Toit).
Natal sills (\V. Anderson; see (1. T. Prior in bibliography above).
Angermanland sills (A. (i. Iloglnim).
Kilsyth-Croy laccolith ((;. W. Tyrrell).
ProsjM'ct intrusion iH. S. Jevons and others).
(fowganda Lake sills (X. L. Bowen).
The grcitcst laccolith on record — that in the Bushveldt — has in
its upper levels a va.st development of ''red granite" which is a
strict honiologue of the "red rock ** of the Duluth laccolith. The argu-
ment for a s(M*ondary origin of these salic differentiates is strong, as
it so thoroughly accords with the proofs of secondary ''red rock" or
niicn»peginatite at Pigeon Point, Sudbury, and the Moyie river.
The aiialcitic phases of the sills at Teschen, which are partly dia-
base* and partly tes<*h<*nitf*, can Ix' accounted for by the interaction
MAGMATIC DIFFERENTIATION 243
of diabasic (basaltic) magma on the invaded basic hydrous sediments.^
An analogous explanation is suggested for the analcitic rocks in the
sills at Lugar, Inchcolm, Benbeoch, Castle Craigs, and Howford
Bridge ; and for the leueitie phase of the Shonkin Sag laccolith. Else-
where the writer has published the thesis that the foyaitic types of
Ice River and Cnoc-na-Sroine (Loch Borolan), as well as the alkaline
types at Square Butte, Shonkin Sag, etc., are differentiates of syntec-
tics in which the invaded limestones have played an important r61e.^
It may be noted that the induction on which that hypothesis was based
has been greatly strengthened by a more recent, more complete survey
of the world's alkaline-rock terranes. As shown in Chapters XVIII
and XIX, similar statistics go far in favoring the idea of sedi-
mentary control during the formation of syenitic and granodio-
ritic magmas.
In general, the tabulated sills and laccoliths illustrate a principal
deduction from the assimilation hypothesis: the silica content and re-
lated chemical features of the roof differentiate should vary with the
chemical nature of the rock assimilated. The salic phase of a Moyie
sill (cutting thick quartzites) is an abnormal granite; that of the Shi-
numo area (cutting shales, limestone, and sandstones) is a syenite; that
of the Port Orford intrusives (cutting dominant argillite with sand-
stone) is a dacite of granodioritic composition. Limestone control
is suggested in the leueitie and nephelitic differentiates above noted.
The special effects of resurgent water (that absorbed from sediments)
has already been found in the analcitic differentiates and in the com-
monly developed soda-aplites and albitic veins of many of these
injections.
Gravitative Differentiation in Stocks and Batholiths. — The dif-
ferentiation phenomena of concordant injections have been considered
at some length because of their supreme importance in the batholithic
problem. From their very nature the floorless chambers, in which
mast of the world's magmas have originated, can only be understood
by indirect reasoning. If gravitative splitting and gaseous transfer
have been responsible for the phases of sill or laccolith, they may
fairly be considered as controlling differentiation in subjacent bodies.
Owing to its colossal size and consequently longer magmatic life, a
batholith should contrast with any sill in showing more advanced
differentiation. Hence, hybrid rocks in batholiths are rare and the
homogeneity of their salic and femic submagmas is great; more often
than with sills is syntexis completely masked.
1 See V. Uhlig's section in "Bau und Bild Oesterreichs/' Vienna and Leipii&
1903, p. 898.
« R. A. Daly, Bull. Geol. Soc. America, Vol. 21, 1910, p. 87.
244 laSEOUS KOCA'.S AND TUSIR ORIOIN
Because of the shallouiicss of the depths reached by eroddli, the
batholithic outcrop must always exhibit the rock formed near the
C'liamVx'r roof. The femic sultmagma of gravitative splitting lies too
deep for cxpotiuro except as it is expelled upward aa dikes cutting the
country- rooks or the already solidified, salic phaae of the bathoUtfa.
The rarity of such late injections and the danger of confuBiiig tbem with
emanatiuns uf other primary abyssal wedges mean that useful obserra-
lion.s are almost wholly to l>e confined to the salic phase.
In this connection the pctrogenic theory so far outlined has aome
chief consequences which may l>e checked by field observations.
a. I'hc primary ItaHsltic wedge is increasingly affected by ayntexif,
generally Icatlinft to a mixture more silicious than basalt. Tbe new.
Erosion Surfaem
FiQ. 123.— DilkKrutniiiaticHpc'tionilliiHtrntinicthedevdopmefit offemieeonUrt
pliMi-:< in b^tholiths. BaMtd areiu, fr-niic phase; btmtk ana, namul rack o( tba
liulliolith; eTot»-tined, roof an<l wall riK-kH. Double-hmded unnra Aam din^
tiuna of niovcmrnt of salic tiS'} itnil fcmic {F) units of differeDtiation attn etjntolli-
xution <j( llio chilli-d mntiirt phase. SinKlc-hcodrd arrowB repweBt th* opwvd
iiiovrnii-nt of ealie niatfrial transferred by nisKmalic unnwi
mixed magma unilergues progressive differeDtiation. If a sueecMwa
of satellitic )>o(Iies are eruirtcd from the we<lge into levds ■< < t WJNi
through erosion, these botUes should, as a rule, be derived from tbe
upper part of the wolge; and hence formed in the order of infrraiini
acidity, i^uch is the usual observed order, as shown in the tabl» of
.'Vppendix B, and briefly c6m<idcrefl in Chapter IV. The aTcngc
succession i.-j thus a powerful argument for gravitative control in tbe
largest of intrusive Iwiiiea.
b. The salic <lifferentiate should tend to vary chemieally with tbe
character of the average country rock assimilated. We have just seen
tliat this principle affects the nature of the differentiates in sills and
laccoliths. The succin-ding chapters are largely engaged i
that it also actually affects the subjacent bodies.
MAGMATIC DIFFERENTIATION
245
c. Toward the end of its long life the magma of stock or batholith
becomes nearly or quite incapable of further assimilation. The
partially dififerentiated syntectic is chilled at roof and wall and there
solidifies. Inside this contact shell the still fluid magma continues to
liquate gravitatively. Therefore, at the levels reached by erosion the
internal part of the mass should be more salic than the chilled phase
2 Km.
Fig. 124. — SynRenetic granite and diorite in the Penobscot Bay quadrangle,
Maine. (After Penobscot Bay Folio, No. 149, U. S. G. S., 1907.) Ey Ellsworth
schist — Cambrian or pre-Cambrian; C, Castine volcanics — Cambrian?; D, diorite
and gabbro — Devonian?; G, granite — Devonian? The diorite appears to be an
older, chilled phase of the batholith in which the granite later differentiated and
invaded the diorite.
(Fig. 123). Herein we have an explanation of most basic contact
rocks in subjacent bodies. As with the sills and laccoliths, it is no
longer possible to attribute such phases to '^contact-basification,"
produced by diffusion operating on Soret^s principle.
d. The solidified contact phase has often been intruded by the cen-
tral, residual liquid. Such renewed eruptivity may be due to massive
CARN CHOIS
0
Fig. 125. — Section of the Grampian Hills stock. (After the Government
map of Scotland, 1893, Sect. 2.) Q, quartzite, etc.; P, porphyrite; A, amphibolite;
Df diorite; G, gramte. Petrogenic relations probably the same as for batholith
of Fig. 124.
readjustments in the batholithic chamber or to the corrosive power of
the now differentiated liquid magma, which is not chemically in equi-
librium with the already chilled, solid rock (Figs. 124 and 125).
The same proce33 is probably represented in wide dikes, where
the interiors are often more salic and less dense than the respective
contact phases (Fig. 126). The general theory implies, in fact, that
246
IGNEOUS ROCKS AND THEIR ORIGIN
a batholitU is a modijied dik€ of t'normuua sise, a nuuD abyiwal weditr
which was initialbj injected.
Expulsion of Residual Magma. — Hurkcr liaa suggested anolhiT
kind of gravitativc control in diiTiTPntiatioD.' He writee:
"Any different iution which ilepcnds on the sinking of cr)*8tals undrr
gravity bclongM ni'cewtarily to a Minicwhat early stage of cr}-Htalliution.
when the bulk uf the mafcnia watt still in a liquid condition. At a later stage,
when the cry^talx formc<l arr tw numerous or so large aa to touch and nippolt
one another, the condition may l>e likened to a sponge full of water; and it w
easy to picture a partial Hepuralionbcingcfroctot] by theafrainin^^or afUMi-
£ro 3 /
urface
Kill. 120, — DiaKrnminntir nection illustralintt differentiation in loinB dikr*
Aftrr rrvHlidlitLUinn of (In- <'hillwi phiuu-, V, ptravitalive splitting of lalie (S> umI
ff-mlr (f; unitM cunlimiiti in the middle of the dike.
ing-out of the reaidual fluid mttgma from the portion already crystalliini.
Thill isurh a |>ri>ross <1<hv in ftiot take place \a amply proved by the phenomen*
uf ix-Kinntitcs, which rcpn-seiit the fiMiil residual magma of plutonicintruBKMU."
The stjncpzing-out is regardorl as (ipt'cially noteworthy if thr
freczitig magmn is subject to pressure from movements of the earthV
cnist.
Gas and Vapor Differentiates.— Mugmatic gasm arc in origin partly
juvenile, piirtly rcsurgt-nt. (Sec page 249.) The veaiculatioaof llu^
fiu'e liiva-->, the explosive iictivity nt volcanic vent), the activity of
fumtiroles, and tlie piienomenii of many mineral veins, are obvious
]>r<H)f.t that magmatic difTerentiation affontii gaseous or vaporous. a>
well iL" lifiuid products. In a sense the ocean is the largetit viaible
body of "magma" or "rock." probably in greatest part antedating
the primitive crust ificat ion of the earth. The origin of atmosphfre
■ .\. IIorLcr, Natural Iliiitor)' of the Ifcncous Uocka, 1909, pp. 323-327.
MAGMATIC DIFFERENTIATION 247
and ocean is ultimately a problem in the origin of the igneous
rocks.
Gaseous Transfer. — Mere gravitative differentiation in liquid
magma cannot explain certain small, basic or ultra-basic phases in
intrusive bodies. Contact segregations of magnetite, ilmenite, mica,
hornblende, tourmaline, etc., are often found. There is growing
Ix^lief that these are generally due to upward transfer by emanating
gases. (See Chapter XXI.)
CHAPTER XIII
MECHANISM OF VOLCANIC VENTS OF THE CBNTKAL TTPB
Introduction
The larger part of volcanic literature deals with the activities at
cone and crater. Exteasive and intensive as these studies have been,
the numl)er of memoirs treating of all the essential problems of central
eruptions is very small. Yet every general theory of volcanic action
must undergo the test of such a thorough questionnaire. This applies
to the h>'pothesi.s that all vulcanism is a result of the abyssal injection
of primary ba^^alt. The following argument will be clearer if a prelim-
inary list of the specific problems relating to central eruptions be
reviewed. The list includes:
1. The localization and opening of the vent.
2. The persistence of a principal vent for many thousands of yean.
3. The intermittent character of the eniptivity, including (a) the
alternation of active and dormant phases, and (b) the pulsatory or
geyser-like quality of eruption during the active stage.
4. The origin of the heat which, by radiation in active cratefs, is
lost in stupendous quantities.
5. The normal evolution of a vent as illustrated in (a) explod
ness, and (6) the nature of the lava emitted.
6. The mechanism of lava outflow at central vents.
In the present chapter, which is essentially a reprint of part of
earlier publication, these questions vnW In? briefly considerecL^
Some Direct Consequences op Abyssal Injection
The estimate of 40 kilometers for the average depth of the surface
of the sul)stratum may he wide of the mark, but it will serve as the
numerical basis for a statement of certain immediate effects of
injection.
Firsiy basaltic magma, rising from such depth nearly to the earth's
surface, mtist undergo an average expansion ranging between 1.5 and 6
I)er cent.- A small i)art of this expaasional energy may be directly
available for oi>ening Assures in the shell of compression, with conse*
qtient extrusi(m at the surface or development of laccolithic or other
bodies within that shell.
* Cf . R. A. Daly, **The Nature of Vulcunic Action," Proc. Anicr. Aead. of
Arts and Sciences, Vol. 47, 1911, pp. 67-108 and 119-122.
> See Amer. Jour. Science, Vol. 22, 1900, p.201.
248
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 249
Secondly, abyssal magmatic injections vary in solvent power,
according to their own volume and degree of superheat, and according
to the chemical nature and temperature of their wall rocks. The
primary basaltic wedges are thus divisible into the two, assimilating
and non-assimilating, classes. Recognition of this fact is of deep
import to volcanic theory.
Thirdly, magma which has been forced from the substratum level to
levels where the pressure is 10,000 atmospheres less, must have com-
pletely altered conditions of equilibrium for the juvenile gases. These
include hydrogen, sulphur gas, carbon monoxide, carbon dioxide, chlo-
rine, nitrogen, and other gases, elementary or in combination. The
theory of physical chemistry indicates that the dissolved volatile con-
stituents must, in sych a case, slowly diffuse upward, in order to re-
establish equilibrium. There is thus a tendency to saturate and then
supersaturate the upper part of the magma with juvenile gases; if by
the mere change of pressure the magma is supersaturated with one or
more of the gases, bubbles must form and these must slowly rise. If
the injected body is tightly roofed, the gases continue to rise until the
growing gas-tension at the upper levels stops diffusion.
Genetic Classification of Volcanic Gases
It will conduce to clearness if a brief statement is here made as to
the absolute necessity of distinguishing the different classes of volatile
materials which are associated with igneous activity. These fluids are
either magmatic or phreatic.^ Phreatic fluids are of atmospheric or
oceanic origin, and include vadose waters, and also those which Lane
has called connate (contemporaneous) waters, because trapped in sedi-
ments at the time of their deposition. As indicated by Suess, explo-
sions due to the heating of phreatic fluids by intrusive magma have
occurred without the ejection of true lava, either fluent or pyroclastic.
Magmatic fluids are those actually dissolved in magma or eniaiH^
ing therefrom. Those of primary origin and reaching the earth's sur- ^
face for the first time are of the juvenile class. The magmatic fluids
of secondary origin, that is, those absorbed from country-rock forma-
tions, have been called resurgents Resurgent fluids may enter the
magma either as constituents of assimilated country-rock or by in-
dependent solution.
Although only magmatic fluids are important in the present connec-
tion, it is useful to review, in tabular form, the whole group of gases
and vapors which are engaged in volcanic and subvolcanic activities.
> Cf. E. Suess, Das Antlits der Erde, Bd. 3, 2te H&lfte, Wien and Leipsig,
1909, p. 655.
* R. A. Daly, Amer. Joum. Science, Vol. 26, 1908, p. 48.
250
IGNEOUS ROCKS AND THEIR ORIGIN
Magmatic fluids
(volcanic ; in-
ternal).
Juvenile
K<».surgent
Phreatic fluids (suhvolcanic;
external)
Emanations directly from abyaaal in-
jection.
Emanations from primary solid abys-
sal country-rock.
Vadosc and connate fluids absorbed
in the syntectic process.
Vadose fluids absorbed independently
of rock assimilation.
Vadose.
Connate .
The resurgent fluids may possibly do something toward keeping a
vent oi)en, but their volatilization means the partial lowering of tem-
perature in the magma, so that their abundance in a conduit implies
a certain 'Mamping of the flr(*s" already accomplished. In basaltic
volcano<'s assimilation of the normal, acid crust-rocks has evidently
not lx*en important; at such vents the juvenile emanations are clearly
in control from In^ginning to end of ea(*h volcano's history. This
statement does not conflict with the fact that resurgent water, either
vadosc or connate with sediments, Ls often responsible for the explo-
sions at basaltic and other volcano^'s. The clearing-out of the explo-
sion funnel, which is always shallow and superficial, is not so vital to
continued activity as the preservation of fluidity in the magma of the
conduit.
()PKXiX(i AM) Localization of the Vent
Enlarged Fissures.- -The events of 1783 at the famous Laki
fissure of Iceland illustrate the close relation l)etween some central
eruptions and the pronounced fracturing of the surface rocks of the
earth. For much or all of its length the master crack was doubtless
connected with a typical, narrow, abyssal injection. Many hills of
the cone-and-cratiT tyiK' w(Te built along the fissure, which emitted
floods of basalt on the greatest scale recorded by man. ISmcmpt of
lava from the abyssal injection w&s (^viilently much easier at some
points along the visible fissure than at others. The case is analosouf
to th(* formation of the '' Dewey craters*' (cinder cones) on the Mauaa
Loa lateral fissun* oix^ned in 1890, and of scores of similar accumula-
tions on the flanks of Mauna Loa, Ktna» etc. Dutton' gives this
" Sixth Ann. l^-p. V. S. (Jrol. Survey, 1S8,5, p. 172. Other fi
are the cdrnM^hains hiiilt in the zone of the Afriran Greal Rift (J. E. 8. MoofC,
TfinKanyika IVohleni. I^mdon, 1903, pp. 80 and 80; J. W. Grssory, TIm Gnst
Hift Vallfv, r^mdon, 18%, p. 21G and niai>9l; the lines of vents opened on tW
Ktnu fis.sun>s n)uppe<l by Silvestri; the Niearaguan cone-chain, msppsd faj K. v«B
Seebaeh,
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 251
explanation for some of the necks occurring in the well known Mount
Taylor district of New Mexico. In all such instances certain points
in the fissure-lines are favored in the eruptivity, while the remainder
of each fissure was either never opened clear to the surface, or else
was rapidly sealed up by congealing lava (Fig. 100).
The continuance of eruption at any point depends on victory in the
struggle with cold. That victory in its turn depends in part on a suffi-
cient width of vent to permit of a column of lava which is not chilled too
greatly by conduction into the wall rock. Since erupting fissures are
never more than a few meters in width at the surface, it seems neces-
sary to postulate a widening of each fissure where it carries cone and
crater of prolonged activity. The widening may be conceived to de-
pend on four different factors: solution and mechanical removal of
B
^jBHIl
FiQ. 127. — Artificial diatremes in granite cylinders. (After A. Daubr^,
Bull. 80C. g^l. France, t. 19, 1891, p. 317 flf.) A and C are sections; B is an end
Tiew of cylinder; all three after explosion.
wall rock by emanating lavas; melting and explosive abrasion of the
wall rock by magmatic gas emitted through the lava column. It is
not important here to decide on the relative efficiency of these processes
in merely enlarging the original fissure to full vent size. Their relative
efficiency becomes of fundamental significance in the problem of the
persistence of eruptivity at a central vent. In the following discussion
of this topic it is concluded that the vent is kept hot, and therefore
active, because of the emanation of free juvenile gas rising from great
depth — a process which may be styled "gas-fluxing.'' Since a great
enlargement of an original fissure, below the bottom of any possible ex-
plosion funnel, demands much time, it would follow that most of the
enlargement is due to gas-fluxing. Gaseous explosion and erosion of
the walls by emanating lava might be more effective in the widening
of smaller and more short-lived vents.
18
262
IGNEOUS ROCKS AND THEIR ORIGIN
It is an eaoy step from the o).>serve(l case where central eniptiooa
are developed on fissures of lava-flooding, to the caae of the formation
of central ventH on Kurfaoo fissureii from which no true fisBure-eniption
hati ever taken plnce. Such n crack may be too narrow to permit tbe
extrusion of fcns-fn-c lata, whii-h, through quick chilling, eeab tbe fis-
sure, and yet the crack may be wide
enough to allow passage of the juvenile
gasot from an underlying abyssal injec>
tion. Entering the crack under pmaure
and therefore at high temperature, these
gams must tend to enlarge it by alow
fuxion of the wall rock. The proccm
may or may not be supplemented by the
opening of an explosion funnel at tbe sur-
face. As the vent is enlarged by ga^-
fluxing the magma rines within it, and,
kept open by the emanating gas, permits
of further ui)ward blowpiping.
This mechanism implies that the
original surface fissure may not be di»>
cernible by the gmlt^t. It may corre-
siKjnd to no vortical or horisontal dis-
placement, and at the surface itaelf be no
wi<ler than an ordinary master joint or
fault fracture. Enlai^ng slowly down>
ward, ."uch a fis.surc might be charged
with accumulating gases so f ar aa ulti-
nial4-ly to cause an explosion. Since the
Fio. 12S.— Diiiimi... oiM-nwl gnscs must tend to accumulate about oi"-
ij a fiwiirc. iJiWH ( :istlp, Uli'-
or more points along the fissure, tbe ex-
plosion form will be that of a vertical
tul>e surmounted by a funnel.
Diatremes. — The resulting vent is a
iiRglomprste of ,//n(r< w*c. the formation of which was so
jiu<'ce(vfully imitated by Daubrfe (Fig.
127). Thb type of diatremes is, then,
liK'ate<] on a surface fissure, which may
or may not Ik- continuous with the abyssal fissure of the primar>'
injection (l-'ig. \2S). These considerations show the difficulty of dis-
)>roving the existence of throtigli-going crustal fixsures beneath central
vents. Some volcanic diatremes may Ix' formed in homogeneous, un-
tissured rwk. and n secimd iy\f, a pure explosion form, should he
recognized in a full clas-sificat ion of vents.
{Mu
of lA»\fTn rife. 19(12, ]>. 21K.I
Linm ehim' Carbon if croue ecili-
menle travcrniHl, with jniliculion
of dip. IMi,
iiamistuiK-, vhulf
pilli, frufcriionts iif olivine bitsall
toliil lilafk,litu<:tH.
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 253
A diatreme of either kind may be enlarged by the continued pas-
sage of the blowpiping gases, by the mechanical erosion of the walls by
outflowing lava, or by the piecemeal stoping of the walls by the lava
column. In some cases the last^mentioned process may be more im-
portant than pure explosion itself.
Plutonic Cupolas. — ^Lastly, a complete genetic scheme should
recognize a process of vent-opening which is neither explosion, nor the
enlargement of through^oing fissures.
The rise of batholithic magma is differential. Partly because of
gaa control its attack on the roof is most efficient at points, rather than
along lines or in large areas (Fig. 129). This deduction seems well
matched by the field fact that round intrusive bosses or small stocks
^^^.^^^nfra/
Vent
IjMm^
WliWIlii
m ..T„«x.T„ w«
W/li
Fia. 129. — Ideal section showing formation of volcanic vent through the diSer-
cDtial rise of assimilating magma.
are characteristic cupola forms on large batholiths. Some of these
bosses have been proved to have very steep contactr-surfaces and, in
shape, as in their cross-cutting relations, closely simulate volcanic
Decks. It is evident that every such cupola increases as well as
localizes the danger of true volcanic action. Blowpiping fusion or
pure explosion may destroy the relatively thin roof above the cupola.
The resulting vent is, then, of composite origin. Its upper part is
like the two simple types of central vents already described. Its
lower part is neither diatreme nor blowpiped hole, but represents the
work of all the agencies of magmatic assimilation in depth. This com-
posite type of vent illustrates the close connection between volcanic
and plutonic geology.
The original location of each first-rank vent is thus explained by
2M
laSEOVR ROCKS ASD THEIR ORIGIN
the roof topography of the underlying magRia chamlter. Somr one
of tlic cupola-likr offshoots of tlic fluid magma, where it peiietrakv the
solid rock above, must become a place for the accumulation of the
rising gases. A vi-nt once formed at the top of the cupola, it muxl
tend to persist as a vent throughout the perio<l of magmatic fluidity.
Uther vcntij from the !<amc rhamlxT may be opened, but must have
shorter live:*, Itecause of the <lrawing away of the juvenile gases toward
the more favored vent. (Fig. 133, p. 270.)
CoXTIN't'ANCE OF AcTIVITy AT CENTRAL VbNTS: ANALTSia
OF Conditions at Kilacea
Left to itself, the lavn column of a vent matt «soon freeie and activ-
ity must cea.te. Yet there arc ulmudunt proofs that the lives of many
Tin- r
.i|>ti..
— MHIish'iwinn tli''li>nKi'niitin
l.r M. lloul,-. Hull, .--■rv, ci.rU- e<-"I
< .H'i'iitii>'.l |Mrt iif MiriM-ni'iitiil I'li'x
iilranir action at the CmnlKl.
■sncc, No. 7fl, 1900. p. 29
central vciitf. have liecn pniloiiged for great periods of time. The
example of the fmitiil voUimo may U' recalled (Fig, 130), Ix>ng-
continiied activity is [inKlitioiiut on victory in the struggle with eohl.
How is the victury iittaiiicd? Hnw is the heat of the underlying
magnin <'linnil>er tr!insfi'rrc<l to the narrow vent? Hawaiian vont.-i
supply data on this fundamental question. Though Kilauea may tie
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 255
the vent of a satellitic injection (see p. 294), the mechanism is doubt-
less the same aa for a vent over a main abyssal injection.
Rate of Heat Loss through Conduction into the Walls. — It is pos-
sible to obtain a rough idea of the enormous rate at which heat is
given out, by conduction and radiation, at Kilauea (Fig. 131). Actual
calculation will show that radiation is much more responsible for the
Fio. 131. — Map of Halemauniau crater, Hawaii, id July, 1909. (After 1. M.
Lydfpkte assisted by the author.) Arrons indicate the general trend of the lava
currenU duriDp; that month. The poBitions of the "Old Faithful" lava foun-
tains and of the more important caves eroded by the liquid lava in the "Black
Lwluc" are nhown. Figures show the depth (in feet) below the rest house. Scale,
1:5,000.
loss of heat than is conduction into the wall rocks of the vent. To
make this point clear it will be assumed that the cross-section of the
conduit is throughout as large as the area of the lava lake, though very
probably the lake represents a strongly flaring part of the lava column.
(See Fig. 132.) The conditions of the lake in 1909, when it waft
studied by the writer, are assumed. The area of the lake (and there-
256
lasBova ROCKS and tusih oriqjn
with the cross-flection of the lava columu) is considered u eireulv,
with radius of 100 metcrti. Tliis ia more than the BuperBcimI eztrat
of the lake in 1909 but less than lis average extent since 1830. The
cylindric^ pipe with the uniform cross-section is assumed to extend
to a depth uf 2 kilometers, where it opens out into the great feeding
chamber.
Let the temperature of tlie magma be assumed as 1200° C; and
let the averaf^ original temperature of the rocks now forming the
conduit walls l^i assumed as 40° C. Two hundred and fifty years
after the conduit was first opened and henceforth occupied by lava at
the uniform temperature of 1200° ('., the rate of flow of beat through
Fio. 132. — DiaKraniitmtir sc^tiun of Halemaumau, illiulniting t<
vrrtion. crmiun of cavu, and vortical action. The lava "scum" ii repMKnt«d by
the iM^vy bLnck line at the lake surface. Lcnfcth of arrtion about 300 m«ten.
ThicknPKS of sruni exanci-rated.
the walls would be m-arly uniform; and 12 meters from the contact tA
the molten lava the tem)M>rature of the wall rock would be aboat
1115° C This e.-itimate is based on the assumption that the dif-
fusi\-ity for heat in rock at high temperature has the value given by
Kelvin. That value is rertainly too high, but the temperature stated
for a )H>int 12 meters fruin the contact would in any case be reached
after some ct-nturies following the establishment of the lava cdumo.
An idea of the heat loss by conduction may now be obtaioefL The
equation for heat flow is:
T-T,
where k is the coeffici<'nt of conductivity, .4 the area of the surface
traversed, x the thicknes.i of the plate traversed, ( the time, 7* and Ti
■ R. A. Daly, Amer. Jour. Science, Vol. 26, 1908, p. 23.
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 257
the steady temperatures of the two sides of the plate. In this case we
may use C. G. S. units, with k 0.005 (certainly too high a value for
these temperatures), t one second, x 1200, and A 2 7:r X 200,000. We
have
0 = .005X2X3.14+X10,000X200,000X^-^j^--Xl
= approximately 4,450,000 gram calories.
The result expresses the approximate amount of heat lost by
conduction into the wall rock during each second.
Rate of Heat Loss through Radiation at the Crater. — Siegl has
recently supplied a datum required for estimating the heat lost by
radiation from the surface of the lava lake. The general equation is
logS = logc+£logr,
in which S represents the number of calories radiated per second, T is
the absolute temperature stated in degrees Centigrade, and c and e are
constants. For basalt Siegl has found that c= (10) "^2X0.589, and
£ = 4.083.* His experiments show that the equation holds for basalt
up to 472® absolute. It is very probable that it may be applied, with
relatively small, or at least non-significant, error, to basalt at the higher
temperatures and under the conditions of radiation represented at
Kilauea. Such extrapolation gives the following results:
ec.
T° abs.
S.
450
723
0.277
727
1000
1.044
1000
1273
2.800
1200
1473
5.082
In 1909 the present writer used a Fery pyrometer to determine the
average temperature of the non-incandescent scum which regularly
covered at least two-thirds of the lava lake. The average temperature
for this part was estimated to be about 450® C. ; the corresponding heat
loss is computed to be 0.277 cal. per square centimeter per second.
At the best points of observation in 1909, the area of the hottest
lava was not large enough to cover the "black spot'' of the pyrometer
for a time long enough to give a reading for its full temperature. It
was clear, however, from the behavior of the galvanometer needle
during the brief exposures of the very hot lava in the " Old Faithful
fountains," that its temperature was well above 1000® C. From the
color the temperature of the hottest lava visible in the lake was esti-
> K. Siegl, Sitsungsber. Akad. Wissen. Wien, Math.-Naturw. Klaaae, Bd.
116, 1907, p. 1203.
258 IGNEOUS ROCKS AND THEIR ORIGIN
mated to Ik* not far from 1200'' C. The third of the lake
hoe from scum w&s estimated to have an average temperature of
1000° C\, (*orrespon(ling to a heat loss of 2.8 ealories per square centH
meter pcT s(H'on<i.
With radius of 100 meters the circular lake would loee in heat about
375,000,000 calories |kt second. The actual lake of 1909 probably
lost more than 230,000,000 calories per second.
We may conclude that heat was then Ix^ing lost by radiation more
than fifty t imes fa^^ter than by conduction into the walls of the Kilauean
pil)e, if it l>e assumed as 2 kilometers deep. It would seem that radi-
ation at th(* crater must l>e the dominant one of these two phases of
heat loss in any strongly active volcano.
Methods of Heat Transfer. — The upward transfer of heat into a
volcanic pifH^ might conceivably take place in five different wa3'8:
(1) by explosive removal of material from the upper part of the vent,
followed by the uprise of magma from the still fluid chamber; (2) by
simple overflow of magma at the lip of the crater; (3) by thermal con-
vection in the lava column: (4) by a process which may be called, for
convenience, **tw(>-phas(» convection"; and (5) by the passage of free
juvenile gas through the lava column, thus bringing abyssal heat to
the upiMT part of the vent.
The first and second processes have obviously played no essential
r61es in keeping up the heat supply in Kilauea since 1823, when de-
tailed records of its activity began.
Men* thermal (*onvection can hardly l>e regarded as an essential
factor in posti)oning the solidification of a lava colunm. In this
matter the analogy with water heated from below should be applied
only with due attention to quantitative values. The degree of super-
Ixat in the actual well-established vents is not indefinitely high; it is
doubtless no more than 200^ or SOO"* (\ If convection be lively enough
to keep the column nu)lten, the maximum thermal-density differences
within the vent itself should ciTtainly l>e less than those corresponding
to a (lifT(*rence of KK)"". A tem|KTature change of 100" C. means a
density change in nuigma of l(\ss than one-half of one per cent.^ The
density change in wat(T its it passes from 4** C. to 100* C, or vice
versa, is about 4.3 pvr cent. With a density difference about one-
tenth that of water in the same temperature interval, and with that
diiTerence distribute<l through kilometers of depth instead of throu^
decimeters, as in the ordinary convective experiment with water, the
convective |M)tential in the lava column is evidently of a very low order.
Moreover, the spetMl of the convection depends on the viscosity of the
magnia. which through chilling and through pressure is doubtless, on
^ According to Barud, as quoted Id Amer. Jour. Science, Vol. 36, 1008, p. 95.
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 259
the average, hundreds or thousands of times more viscous than water.
It follows that in resistance to be overcome, as in working potential,
heat convection must be incomparably less rapid in a volcanic conduit
than in artificially heated water.
For example, let us suppose that at the depth of 2 kilometers the
conduit passes into the feeding magma chamber; that there the tem-
perature is 1300^ C, while the temperature at the surface is 1200® C;
that the average kinetic viscosity of the conduit lava is as low as that
of a liquid 100 times more viscous than water; that the thermal con-
vection in the conduit is to be compared in rapidity with that obtain-
ing in water heated from 4® to 100° in a wide tube 1 meter high. The
maximum convective gradient for the water system may be ex-
pressed as
4 . 3 (per cent, expansion) _ ^ ^
1 (meter, thickness)
The gradient in the lava column is approximately
.5 (per cent, expansion) _ ^ww.^,^
' ~ ' 2000 (met^sl " •^^^•
The maximum speed of convection in the water of the imagined ex-
periment is, then, ("7yfj?J9KX100= ] 1,720,000 or more times greater
than that of the lava in the conduit.
It seems certain that such slow transfer of magma could not keep
the temperature of surface lava of the lake at anything like the ob-
served point. On the average, every square centimeter of the lake's
surface in 1909 radiated about one calory per second or 86,400 cals.
per day. Taking .35 as the mean specific heat of basalt (1200'*-1300'*),
this implies a daily heat loss corresponding to a temperature fall of
100® C. in a vertical column of lava more than 2400 meters deep, and
1 square centimeter in area at its upper surface. Evidently, other
and much more effective agencies must be at work to keep Kilauea
active, year in and year out.
Two-phase Convection, — However, there is a different and very
powerful kind of convection constantly illustrated in Halemaumau
when that lake is in full activity (Fig. 131). The persistent stream-
ing of the lava into the caves, characteristically developed at the shore
cliffs of the lake, is evidently due to surface gradients. In general,
the "scum" stands higher in the central part of the lake than it does
in the caves and in the channels leading to the caves. The "scum"
or thin crust of the lake prevents or retards the escape of the mag-
matic gases, which accumulate beneath it and form a kind of froth, or
emulsion of lava and gas, of relatively low density. The tendency is,
260 IGNEOUS ROCKS AND THBIR ORIGIN
thus, to raise the crust in one or more areas. In each cave, becausg of
reflection from its roof, and perhaps also because of special heating
through actual combustion of sulphur, hydrogen, and other gases, the
crust is rapidly and completely fused. The escape of the gases is
there facilitated and the surface of the lava is correspondingly lowered
The surface slopes are, therefore, steepest in the channels leading to
the caves, and streaming at the rate of 2 to 5 kilometers an hour may
l>e observed in the channels. Elsewhere the surface slopes are lower
and streaming is less rapid. The caves are not outletting tunnels,
as so often stated, but each is closed at a distance of a few meters
from its entrance. The lava which has streamed into the cave must
return to the main part of the lake. Only one way of return is possi*
ble, that by a ba(*kward sub-surface current. Having lost its dilating
gas and grown rapidly denser, the heavy lava sinks and flows toward
the center of the lake. Similarly, the ever-changing surface slopes in
other parts of the lake compel vertical currents and vortices of the
most complex design (Fig. 132). This type of magmatic movement
may !>e called ''two-phase convection.'' It depends on the presence
of a liquid "ph&se'' and a g&s "phase" in the lava.
If vesiculation of the liquid magma is possible in great depth, two-
phase* convection may cause a relatively speedy transfer of hot magma
to the surface. How effective this process can l)e is worthy of some*
what detailcnl statement. The imposing change in magmatic density,
which is effected by very slight increase in vesiculation, will first be
indicated. The speed at which individual bubbles rise will then be
estimated, and, finally, a rough quantitative idea of the conveetion
enforced by the develoj)m(*iit of g&s bubbles in depth will be obtained.
The sp<*cimens of Hawaiian paho(*hoe lava collected by the writer
contain, on the average, at le&st 200 vesicles per cubic centimeter of
the lava. The v(*sicles of the surface layers are roughly spherical and
average no more than 2 millimeters in diameter, though, of course, the
range of diain (*ters is very great . For convenience, let a spherical mav
of hydrogen, having the radius of 1 millimeter at one atmosphere of
pn^ssure and at 12(K)'' (\, ho called the "standard bubble'' for basalt
Extrapolating on .Amagat's pres.sure-volume curves for hydrogen at
liCX)'' (\, the volumes and radii for the standard bubble at high pres-
sures may be calculated within a margin of error which is probably
very small. Examples are shown in the following table:
•
Volume Radios
(cm.*) (na.)
.000115
.000025
.000014
Approximate depths in
Prcjwure in
maKTnutir roliiiiin (iiiHen*)
at mospherofl
730
200
3,050
1,000
7,300
2.000
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 261
Gas-free basalt at 1200° C. and one atmosphere has a specific
gravity of about 2.75. On account of the slight compressibility of rock-
matter, that value may be assumed as typical for bubble-free basaltic
magma at pressures up to several thousand atmospheres. If such
magma become charged with 200 standard bubbles per cubic centi-
meter, at 200, 1,000, and 2,000 atmospheres, the specific gravity falls
to the following approximate values (Col. 1) :
1
Specific gravity
xTrflsure vais.^ i
1
1
1
1
2
200
1000
2000
2.688
2.736
2.742
2 . 7495
2 . 74975
2 . 74997
A single standard bubble replacing the liquid in each cubic centi-
meter of bubble-free basalt would lower the specific gravity to the
amounts shown, again approximately, in Col. 2.
This last table illustrates possible ranges of buoyancies induced by
vesiculation at the three depths chosen. The actual buoyancy attained
may often be much higher. It will be seen that the buoyancy pro-
duced by only a small extra vesiculation of a local mass of magma
must occasion a rapid uprise of that mass.
The bubbles themselves, as independent bodies, must rise with com-
parative slowness. The experiments of H. S. Allen have shown that
small spherical bubbles, rising in a liquid, attain their terminal velo-
city according to the formula previously deduced by Stokes for the
rise of light solid spheres of very small radius.^
Let r represent the radius of a bubble; d', its density; d, the density
of the surrounding magma; v, the coefficient of viscosity of the magma;
Qy the acceleration of gravity; and x, the terminal velocity of the rising
bubble, that is, the velocity when the motion is steady. The Stokes
formula applies if the product dxr is small compared with v. This is
dearly true for the standard bubble in liquid basalt with the viscosi-
ties appropriate to pressures of 200 to 2000 atmospheres. We have,
then.
x = ^gr^
m
Computing the values of x when the magmatic viscosity is assumed
to be constant and only 100 times that of water at 15*^ C. (0.0115) or
1.15 in C. G. S. units, we have, at the three illustrative pressures:
> H. S. Alien, Phil. Mag., Vol. 50, 1900, pp. 323 and 519; G. G. Stokes, Cam-
bridge Phil. Trans., Vol. 9 (2), 1850, p. 8.
262 IGNEOUS ROCKS AND THEIR ORIGIN
Depth Pressure Terminal velocity (x)
meters) (ats.) Cm. per second Meters per hour
730 I 200 47 ! 16.9
3G50 1000 .17 ; 6.1
7300 2000 .12 I 4 2
Since experiment shou-s that the viscosity of a liquid rises rmpidly
with pressure, it is instructive to assume higher values of v for the
greater pressures. If t; he taken again arbitrarily as 500 and 10,000
times that of water for magma under the pressures of 1000 atmospheres
and 2000 atmospheres, respectively, we have for x these values:
Pressure ... ., Terminal velocity (x)
\ iHoosity ' ' ~
(at«.) Cm. per second Meters per hour
200 1.15
1000 5.75
2000 115.00
.47
034
.0012
16 9
1.2
.04
In all cases smaller bubbles would rise more slowly, x varying
directly as the scjuare of the radias.
Two important conclusioas may Ix' drawn from these computatiov.
G&s bubbles of the ''standard'' mass or of smaller mass must rise from
the deeper lev<*ls of an ab^'ssal injection with extreme slowness. In
view of the high magmatic viscosity and great pressure in depth, it '»
conciMvable that it mav take thoasands of vears for a ''standard"
bubble to rL*e from a depth of, say, 10 kilometers to the earth's surface.
This suggests one reason why gaseous emanation is so prolonged at
central vents.
Secondly, from the slowness with which bubbles rise, it is clear thai
a swarm of bub))l(>s, which for any reason have l)een aggregated locally
in .special abundance, would t)e dispersed into the surrounding, lesi
vesiculated magma with great slowness. The local mass of
thas sjK^cially vesiculated would l>e less dease than the average
and, as a unit, would rise toward the crater. It now remains to i
cate that a verv moderate amount of extra vesiculation must cause such
a two-phase niiiss to rise with comparatively great velocity.
Of course, this case has not l>een investigatetl experimentally; an
indirect method must be used in its discussion and the result can at
present hardly be othcT than qualitative.
Once again to make the mental picture clearer, it is well to assume
certain conditions arbitrarily. As an example, let the swarm-filled
mass be spherical ; let the reigning pn^sures and magmatic viscosity he
a< in the foregoing ca<(^; let the surroumiing magma have a density of
2.75; and let the extra vesiculation hv to the extent of 50 "standard"
bubbles \>vr cubic centimeter on the average. The corresponding den-
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 263
sities of the sphere are shown in the second column of the following
table:
Pressure
(ats.)
200
1000
2000
Sp. gr.
d-d'
V (assumed)
R (cm.)
2.734
.016
1.15
.52
2.746
.004
5.75
2.40
2.748
.002
115.00
22.25
For solid spheres rising in the magma we may compute the " critical
radius" (/2), that is, the radius of the largest sphere which would obey
the law of the Stokes formula. The values for R, as stated in the fifth
column, have been found with the help of Allen's formula:^
R' =
9t;2
2gd{d-'dy
The terminal velocities of the solid spheres having the critical radii
would be, for the corresponding values of iJ, v, and (d— dO» as follows:
d-^' '
V
1.15
1 R(cm.)
.52
Terminal velocity
Cm. per sec. j Meters per hour
.016 1
.82
29.5
.001 1
5.75
2.40
.87
31.4
.002
115.00
22.25
1.88
67.6
The figures show that even for small solid spheres the velocities are
considerable. With increase of radius the terminal velocities would at
first increase very fast, and then more slowly. However, since the
resistance to the motion would, for large spheres, vary with the square
of the velocity, neither the Stokes formula nor any other yet developed
can declare the actual velocity for large solid spheres moving in the
magma.
Nevertheless, Allen's formula for large spheres is of distinct help in
guiding one to a proper appreciation of the case. It reads:
,,, 1^71 d-d'
^'—k'S'^'' d '
where fc is a constant for a given liquid-solid system.* It follows that
the terminal velocity here varies directly as the square root of the
radius and as the square root of the difference of the two densities.
Referring to the table showing terminal velocities for solid spheres with
» H. S. AUen, Pha. Mag., Vol. 50, 1900, p. 324.
* H. S. Alien, ibid., p. 532.
264 IGNEOUS ROCKS AND THEIR ORIGIN
critical radii, it seems clear that, in any of the three caseB, spheras of
corresponding dea<ity and of radii of 10 or more met^s would rne at
the rate of at least 10 centimoters per second or 360 meten per
hour.
ThL^ analogy of solid spheres seems to afford some bdp in our
imagining the course of a specially vesiculated mass of liquid magma.
The rough quantitative estimate just made for large solid spheres can-
not Ix^ directly applied to this ca««e. On account of the possibility of
internal movements in the rising m&s8 of liquid magma, itd speed of
uprise \^ill not Ix; quite the .same as that of a solid mass of the same
shape, size, and deasity. Yet the correction to be applied is probably
small.
As such a ma<s approaches the surface, through a column of rapidly
decreasing viscosity and with a coastant increase of buoyancy because
of oxpaasion of the contained bubbles, the velocity must greatly in-
crea'^e. However much a given mass of magma might lose buoyancy
through the loss of its larger, more swiftly rising bubbles, the total
effect mast be to generate a powerful upward current in the rrflgmftfr
column.
In spite of the lack of the necessary, full experimental daln* our
general concla^ion seems to be as follows. Experiment does show that
the rise of indivi<iual giis bubbles in magma will be very slow. Neither
experiment nor theory- can as yet declare the actual speed of the rise
of a mass of specially vesiculated magma, but the analogy of solid
spheres moving under gravit}' in a liquid enforces the belief that the
more buoyant magma will move rapidly if it^ volume is of the oiVler of
thousands of cubic meters. Assuming such differential vesiculatkm in
great depth, and assuming also a mechanism by which the gas of risen
magma is dissipated (as in a volcanic vent), two-phase convectioB
must stir the magma column to great depth and with considerable
rapidity. Such a process must be incomparably more rapid than that
of thermal convection under volcanic conditions. The transfer of heat
may readily )h' conceived as able to supply the radiation loss in the
crater for long pcTiods of time.
The basal assumption, that vcsiculation occurs at great depth in a
volcanic conduit, is necc?<sarily difficult to test by the facts of field geol-
ogy. During its soli<iification an intrusive l>ody is likely to be rifenofd
of its bubbles, which rise, and the gas so collected at the roof is slowly
dissipated into the country- rock. This may be the explanation of the
lack of vcsiculation in most dikes, sheets, laccoliths, and hnthoHth*.
In general, the ro(*k of a lava ntM'k may l)e similarly freed from bubbles
(luring the relatively long period of cr>'stallisation« Neverthefess,
cases are not wanting where bubbles are known to have been trmipcd
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 265
in basalt at depths greater than 300 meters. The basalt of the West
Maui neck, illustrated in Fig. 136, is charged with many minute vesicles
at a depth at least 300 meters below the original top of this lava column.
Vesiculation at some depth is proved by the discovery of dikes and sills
abundantly charged with gas pores. The writer has recorded vesicular
basaltic dikes of the Okanagan mountains and in a sill of the Columbia
range, British Columbia. Du Toit has found two such porous sills
in the Stormbergen region, Cape Province, South Africa.^ Rogers
and du Toit observed that sandstone xenoliths in an intrusive dolerite
sheet of the Karroo have been rendered vesicular by the magmatic
heat in spite of the considerable pressure.^ These and other known
examples seem to strengthen the belief that bubbles may form in
magma at the depth of several kilometers.
It is important to note that two-phase convection has two distinct,
though related causes. Principal stress has hitherto been laid on dif-
ferential vesiculation in depth, whereby a m^s of magma becomes more
buoyant than the enclosing magma and rises. Just as inevitably, the
magma which is freed of gas at the crater must sink and stir the
column to great depth. Even if the column is not vesiculated at all,
this second mode of convection is likely to be effective in the vertical
transfer of the magma. As a rule, the density of a liquid is lowered by
the absorption of hydrogen, nitrogen, oxygen, or other relatively light
gas. This is very probably true of natural mixtures of juvenile gases
when dissolved in magma. As these gases stream or diffuse from all
azimuths in the feeding chamber toward the base of the narrow con-
duit, they are there concentrated. Thus, the magma in the conduit, at
its lower levels, attains a density less than that of the average magma
of the feeding chamber, and, a fortiori ^ less than that of the gas-freed
magma descending from the crater level. This is another kind of den-
sity convection depending on the relative concentration of juvenile gas.
For lack of experimental data, it is nolf impossible to estimate the
eflSciency of this species of convection. It may be a powerful ally of
two-phase convection proper. For example, it is conceivable that the
upward movement of magma is begun in the conduit because of the
concentration of gas in solution and not in bubble phase. Then, as
the magma rises to levels of smaller pressure, the gas begins to separate
out in bubbles and enforces true two-phase convection of ever-increas-
ing speed. In view of these various modes of gas control, the vertical
stirring of the magma column may, perhaps, be more safely described
* A. L. du Toit, 16th Ann. Rep. Geol. Comm. Cape of Good Hope, 1912, pp.
122-123.
' A. W. Rogers and A. L. du Toit, Ann. Rep. Geol. Comm. Cape Good Hope,
1903, p. 39.
266 IGNEOUS ROCKS AND THEIR ORIGIN
as, in general, a gas-concentration convection. Yet the actually ot^
Kcrvecl fact in that at the crater the gaseous phase does septtrmie, in
bubble form, from the liquid phase, and the writer has preferred
to emphasize this empirical fact in adopting the name "two-phase
convection/'
Lava Fountains. — Herein the writer l>elieveH that we have an eawn-
tial part of the explanation of ''Old Faithful/' the site of the greater
periodic ''fountaias'' of Kilauea. That circular area, abcmt 20
meters in diameter, has represented the true axis of the lava column for
many 3'ears, and seems to have been the main source of magmatic heat
throughout the known history of Kilauea. In 1909, at average
intervals of a))out thirty-five seconds, the surface of the lava lake in
this area was domed up to maximum heights of a few meters. These
fountaias are not due purely to the rise and explosion of great gas bubbles,
the collapse of which could have l>een readily observed. Very flODall
amounts of gas or vapor were given off at the moment of doming
or immediately afterward. The outbursts are best explained, in
part, on the principle illustrated in the upspringing of a log of light
wood freed at the )>ottom of a lake. Through its momentum the log
may jump clear out of the lake. In part, the outbursts of "Old Faith-
ful*' are due to true explosive dilatation of the gas bubbles in the '*log."
The latter process is doubtless the chief caase of the smaller "foun*
taias** playing over the surface of Halemaumau, and of those which
played over th(» surface of Dana Lake or New Lake twenty-five
ago. The draining of each of these two lakes? has shown that it
saucer-sha]>ed and very shallow over most of its area, and the writer
iN'lieves this is true of Halemaumau to-day. (Compare Fig. 132).
The depth i-i generally much too small to allow of such momentum
in mugmatic '* logs'' that they might leap to the heights actually
o))served. The |HTiodicity of ''Old Faithful" is suggestively like the
rhythmical. ])ulsat()ry action so often observed when a liquid flows
againM strong friction, as water does in a drain pipe.
The site of *'()ld Faithful" is, thus, the place where the juvenile
gases rise from the depths in two-phase mixture with liquid lava.
With the c<>lla|)se of each dome, the gas-charged magma finds its level
and runs under the semi-soli<l or solid *'scum" on the lake surface
(Fig. 132). There the ga*< is slowly freed and accumulates beneath
the ''scuin" until the tension produces a true explosion, that is, one
of the many smaller "fountains*' so coastantly appearing on the
lake.
The incessant streaming in Halemaumau, the nature of the '*01d
Faithful fountain^,*' and the ceasehss vortical motion in the lake, a^
well as the similar phenomena in the active Mokuaweoweo,
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 267
many direct evidences of two-phase convection, which calculation
shows must be rapid, provided slight variations in vesicularity occur in
the depths of the lava column. Though it is not possible to prove
absolutely that the Kilauean column is vesiculated in depth, it cer-
tainly is so at the surface to a remarkable degree. At many points,
the lower part of the wall of Halemaumau was found, in 1909, to be
covered with thin coatings of black glass which represented splashes
of lava from the adjacent lake. This lava almost instantly "froze"
to the wall. In every case it was extremely porous, so as to be quite
spongy in appearance. The vesiculation was almost if not quite
complete before the "splash" struck the wall, and it is simplest to
suppose that the surface lava of the lake is a froth. There is no
known reason why vesiculation should be the rule at one atmosphere
of pressure and non-existent at one hundred or one thousand atmos-
pheres; it is all a question of the degree of saturation with gas. The
two-phase convection hypothesis rests on this unproved assumption,
but its merit is great, as it explains the essential facts of circulation in
Halemaumau.
Cooling by Rising Juvenile Gas. — As a fifth hypothesis it might be
conceived that the heat is kept up in the lake through the rise of
bubbles of free juvenile gas from the magma chamber, the bubbles
arriving at the surface with some excess of temperature above that
required to give the lava of the lake its observed fluidity. But the
feeble explosiveness of the emanating gas at Kilauea shows that any
uiyt mass of it, arriving at the surface, is already nearly expanded to
the volume appropriate at one atmosphere of pressure, and therefore
that the gas is in nearly perfect thermal equilibrium with the enclosing
lava at the surface. Such bubbles, as they rise and expand, must
thereby tend to cool the magma.
The cooling effect is very great, as may be shown by the following
calculation. In an adiabatic expansion of a perfect gas: let T' be the
initial absolute temperature, and T the final absolute temperature;
let p' be the initial pressure, and p the final pressure; and let ^^ ( = 1.4)
be the ratio of the specific heat of the gas at constant pressm*e to its
specific heat at constant volume. Then
At about 37 meters below the lake surface the pressure is ten atmos-
pheres. If the bubble, after expanding adiabatically, is to arrive at
the surface at a temperature of 1200® C, it must have at the depth of
37 meters a temperature of about 3700® C. (assuming no dissociation
of the gas). Evidently the free-moving gases have a cooling effect on
19
268 IGNEOUS ROCKS AND THEIR ORIQIN
the upper part of the lava column.^ That this effect is actoAlly small
is, of course, due to the small mass of gas emitted in a unit of time and
to the fart that r is much less than 1.4 for the actual (not "perfect")
gases while rising through the deeper levels. Moreovefi it has been
noted that the rL<o of a bubble mast be exceedingly slow if its maas is
anything like that in the average vesicle of frozen lava. 80 slow is
the transfer that the rapid heat wastage at Halemaumau cannot pos-
sibly l>e compensated by any residual superheat in the emanating gas.
On the other hand, the thermal conditions are different in craters
floored with highly viscous lava. There the emanating gases com-
monly issue at pressures of more than one atmospherei and thqr nre
thus kept hot and endowed with some fluinng power. The amall blow-
holes in Kilauea, as in most other basaltic districts, have long been
kept open through this a<*tion. It is quite posriUe that such hot-
blasting is operative on a greater scale in larger openings like the crater
of Stroml)oli. Yet even at Stromboli that cannot be the chief method
of heat transfer from the depths, and again no other method than that
of two-phase convection seems competent to keep the lower and greater
part of the lava column fluid. At Kilauea, at the wonderful Mokua-
weowoo (the vent of a main abys.«<al injection), at Matavanu in Savaii,
we seem comi)elled to exclude all other agencies for heat transfer except
this type of convection. The same explanation seems to apply also to
Vesuvius and Strom)>oli, for their craters in times of strong activity
have boon observed at close quarters and, like Halemaumau, they show
lava "fountains'* and other features of this convection.
The Volcanic Furnace. — So far, no assumption has been made
that the heat transferred to the top of the volcanic conduit is other than
primary in origin, that is, heat due to the initial temperature of the
parent abyssal injection. Such is the orthodox view of volcanic hemt.
The rough estimate made in the discussion of thermal convection sug-
gests the (lifl[irulty of understanding how the mere primary heat suffices
to (*xplain the long life of many volcanoes.
It may well )>o questioned, however, that all the heat at a volcanic
vrnt is ])rimary.- That due to the radioactivity of magma during
> I. C. White (in the Bulletin of the Geological Society of Amsriea. YoL 94.
1913p p. 280) haa recently de(<«ril>ed the notable coolinffs of a Penafsrhraaia
by the expansion of natural gas. At a depth of (SOOD feet the temperali
found to be 100'' F. instead of US'" F., the value expected from the loeal padieat
and White attribut<'9 the anomaly to chilling by a Btrong gaa-How aatr tlie 60(X^
foot level. On the other hand, the unusually steep temperature |^adi«ita fonad
in natural-gas fieMs may possibly be explained in part by the iacreBM of pramie
as the gas is generated under a tight cover.
' Cf. C;. THchermak, 8itzungpl>er. Akad. Wins. Wien, Vol. 75, 1877, p. 1C2.
where a brief statement is given, showing a clear anticipation of thii
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 269
the fluid stage of an abyssal injection is too small in amount to affect
the rate of heat loss to any sensible degree. More promising is the
idea that heat-producing chemical reactions in the conduit may have
powerful effect. Since the day when Sir Humphry Davy renounced
his own explanation of magmatic heat as due to the oxidation of alka-
line metals contacting with water, most volcanic theories have regarded
magma as inert so far as exothermic reactions are concerned. On the
other hand, recent studies of gaseous emanations from active volcanoes
and from artificially heated rocks and meteorites clearly suggest the
possibility of such reactions.
Analysis of any perfectly fresh igneous rock shows the presence of
water to a considerable percentage by weight. This is true of in-
trusive gabbros and diabase as well as of basaltic lavas. Most of
the non-hygroscopic water determined in the analysis of quite unal-
tered gabbro or basalt may be as much a primary constituent as the
silica or the alumina. We must believe that hydrogen and oxygen, in
the proportion characteristic of water, are present in primary basaltic
magma. It does not follow that, under volcanic conditions, these ele-
ments will issue from the vent in combination as water. In his able
monograph on "The Gases in Rocks," R. T. Chamberlin indicates the
general reaction to be expected in the Kilauean or other basaltic magma
chamber. He writes:
"The effect of pressure on chemical equilibrium is to favor the formation
of that syBtem which occupies the smaller volume, but if there is no change
in volume, in passing from one system to the other, the increase of pressure
presumably has no influence on equilibrium. In the reaction
3FeO+H20t:?Fe304+H2
considered as a thermochemical equation, the number of gaseous molecules,
and hence the volume of gas, always remains the same, so that it is not likely
that this reaction will be influenced by change of pressure. A rise of tem-
perature favors the formation of that system which absorbs heat when it is
formed. A comparison of the amount of heat liberated by oxidizing three
molecules of FeO to Fe304 and one molecule of Hi to H2O shows that, in the
former case, 73,700 calories are evolved, and in the latter, 58,300; that is,
SFeO+HiO— Fej04+H2H- 15,400 calories. As heat is evolved in this process,
a rise of temperature would accelerate the reaction in this direction less
than in the reverse. In other words, the higher the temperature, the more
would the formation of ferrous oxide and water be favored as compared with
the conditions at lower temperaturas.
"Because of this, there is much reason to suppose that, at the depths
where lavas originate, hydrogen and oxygen exist combined as water, since
up to temperatures of 2000"^ C, the dissociation of water takes place only
to a limited extent. If a state of equilibrium between hydrogen, water.
270
lONEOUS ROCKS AND THEIR ORiOtS
and the iron compounds were cstabliEhed in the heated interior where m magma
originated, as bood as it commenced its vay upward and began to loae beat the
condition of c<]uilil>rium would be ilcetroyrd. With the failing temperature
tlie tendency to rerstablLsli equilibrium would favor the formation %£ that
Bj'stem which van produced with tlic liberation of heat, i.e., magnetic oxide
and free hydroKon. In ai^cniliiig lavao which arc losing heat, the tcDdenry,
therefore, ix to produce hydrogen and magnetite, or ferroso-femc compound*.
This is doubtlcKS an important source for the hydrogen which is so copioudy
exhale<l during a volcanic eruption. At the same time thia proceea aeeounU
for the nidcspread occurrence of magnetite in igneotu roeka."*
fj RESIDUAL FLUID \^
/ 1 t o.'l t .»
If ABYSSAL INJECTION \1
/t^u/y.
INJECTION
Fia. 133. — Iiiml lonKitudinul lurlion of an abyssal injectian, il
lation of vulcanixiii to thf> M-ruIar riKe (urrowK) of juvenile gas. The rnHdle vent
is active bn-nufp it oriftinntra itt the hiichml point (rupola) in the injected body.
The other venlH .^rc extinrt Im^hiiki- (i[ thtH advantage of the middle vent. Sob)
black rrprrHcntit the already rryittalliiei) maleHal of the injection. CravJilMd
area in theroiintr> fork. I.enffth of section about 100 km.
The abundant nnimni life of < 'amhrian and lator time implies ibM
tho earth's ntmosjihere ha^ lonfc had a very low content of rarboo diox-
ide. Thp amount of thi^ oxide which ha^ been locked up in tbe car-
twnate rocks .lincc the Ix-frinniiiR of the Cambrian period is so «nor-
mousthat most of il.orallof it,n)u.-<t Ik* comtidered as of juvenile ori^n.
Yet more clearly than in the case of water, carbon dioxide must be
regarded a.< a prinmry constituent of earth magma. Under tbe same
conditions as thos<' il<-scril>ed by C'liamU-rlin, ferrous iron is oxidised
to maRnetile by carl>un dioxide, yielding carlxin monoxide and 6000
calories per Rram molecule.
> R. T. Chamlxrlin, The Gasra in Itocka, I'ublication No. 108, I
laatitution of Washington, 1908, p. 00.
MECHANISM OF VOLCANIC VENTS OP THE CENTRAL TYPE 271
The list of the juveniie gases and vapors also includes nitrogen,
ehiorine, sulphur, and hydrocarbons. These and other volatile sub-
stances, including hydrogen and carbon monoxide, stream from all
azimuths in the magma chamber to the lower end of the conduit. The
pipe has always a very much smaller cross-section than the feeding
chamber, implying some concentration of the volatile matter (Fig. 133.
and 134). At conduit temperatures
this ever-varying mixture of gases
must, according to practically infinite
probability, be in unstable chemical
equilibrium; under the conditions new
equilibria are attained with the evolu-
tion of heat.
The relative proportions of each
gas must, in general, be different from
that in the primary magma before it
was injected. Concentration of the
gases means, according to the law of
mass-action, the development of new
compounds. As the pressure is less
in the conduit than in the underlying
chamber, the viscosity of the magma
b less, the gas bubbles are larger, and
the speed of possible reEictions is
thereby increased.
Of course, the actual amount of heat evolved during the chemical
rearrangements in the conduit cannot be estimated, but a glance at the
following tables (showing some examples) must assure one that the
heat product from the complex system may be of a high order,'
SUBSTRATUM
A. — Ideal crosB-section
through middle cone shown in Fig.
HEATS OP FORMATION
Calories per
Calories per
Calories per
gram-molecule
gram
[SOd~
molecule
grun-molecule
[HCl] +22,000
+71,080
[CaCld +190,300
[H,0] 68,300
[CO.]
96,960
[K,CIJ 211,220
IH,81 6,400
[SO.l
103,240
[Na,CIJ 196,380
[H.NI 11,890
|P.0,|
369,900
[PeCIJ ^,060
[H.CI 21,750
IFeSI
24,000
[CO] 29,000
ICaFJ
238.800
' The values for the heats of formation and reaction are taken from the woriu
of Thomsen, Muir and Wilaon, Neroat, and others. In some cases more reeent
ezperimeata give slightly difFerent values.
272 laWEOrS RCH^KS AXI) THEIR ORIGIN
HEATS OF REACTION
CH4+2OJ =C02+2H20+ 196,200 cab.
( Oj+Hj =CO+H2()+9,980 cals.
3FtO+H2() =Fe504+H2+ 15,400 cab.
3Im()+('0, =Fe3()4+ CO +6,000 cals.
XHs+Hri =NH4Cl+42,600cab.>
('()+() =CX)2+ 68,040 cab.
In a(l<lit ion, there is the po8.sibility that a large supply of energy was
potentialize<I at the high temperatures of the primitive earth and that
this energy becomes converted into magmatic heat under the conditions
of a volcanic vent. Becker has suggested thb in the case of uranium.-
Arrheniu8 has proposed the hypothesis that the heat of the sun is
supplied principally through the break-up of endothermic compounds.'
Warren has shown that, at high pressure and temperature, steam u
partially converted into the strongly endothermic oione and hydrogen
peroxide.^ Lines indicating cyanogen arc found in the spectra of
some stars and comets, and Arrhenius attributes the nitrogen of the
air largel}' to the dissociation of volcanic cyanogen. In the formation
of a gram-molecule of this gas, 65,700 calories are potentialiied. The
dissociation of chlorine involves the absorption of 113,000 calories.*
Dissociation of other gaseous elements meaas heat absorption of the
same order of magnitude. When ferric oxide and iron sulphide react
to form ferrous oxide and sulphur <lioxide, 80,640 calories are absorbed.
When carbon and carlx)n dioxide react to produce carbon monoxide,
38.800 calories are al)sorbe<l. When steam and C react to form carbon
monoxide and free hydrogen, 28,900 calorics are absorbed.
In addition to the heat evolved by the dissociation of endothermic
com])ounds, another source of great energy is to be found in the com-
bination of the freed, "naseenf* elements with other constituents of
the magma. The ]M)\v(Tful thermal effect of interaction between
hydrogen or oxyg(>n and the carlam or nitrogen atoms of cyanogen
hardly needs quantitative statement to show its value.
^ Thoufsh Ainmonium rhloride iiiiiv not l>c able to form within the
colli inn of a volcanic conduit, it does form at the surface, where the Iom of heat
chiefly (xTurs. Similarly, ammonia may form from itii elements in the rdativelj
cool crust of the lava lake in a crater, also producinR heat at the sone of ndiatkNi.
> G. F. Becker, Bull. Oeol. S<m\ America, Vol. 19, 1908, p. 1411. The final yield
(»f ra<iium is about 2,(K)0 millions of calorics per gram, or nearly 260,000 fimri the
thermal value of a Knim of c:irl>on burnt in oxygen.
* S. Arrhenius, WorMs in the Making, New York, 1908, p. 91.
« II. N. Warren, Chiin. News, Vol. 77, 1S9S, p. 192. When 1 ffram of oiQrgM
is convert (h1 into 1 jjram of ozone, 750 calories are al»sorhed.
» M. Pier. Zeit. fur phvfl. C^hemie, Vol. 62. 190S, p. 385. Ekholn
gt^tetl that the formation of "elements" may partly explain eolar
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 273
In this connection it may be noted that the total melting heat of
ordinary rock-matter (measured from 0® C.) is only 400 to 450 cals. per
gram, and that the latent heat is only about 90 cals. per gram.
Such examples emphasize the value of the conception that abyssal
injection, entailing a sharp change of pressure and a slower change of
temperature in primary magma, may set free a vast amount of energy
which is available for conserving the melting temperature in a lava
conduit. Very great superheat is, however, prevented by two-phase
convection, which tends to keep the volcanic furnace and the surface
lava at nearly the same temperature.
The whole system, as imagined, is somewhat analogous to a modern
hot-water plant with an almost perfectly lagged vertical pipe running
up from the furnace. Or, again, the generation of heat in the conduit
Is analogous to that in the gas-mixture of a blowpipe. In the first
case (two-phase convection) the rising gas is a passive agent in the
upward transfer of heat; in the second case (chemical changes) the
gas is a positive heater, thus itself tending to annul or diminish the
cooling effect due merely to its own expansion. For a double reason
juvenile gas has fluxing power in the vent.
Summary on the Heat Problem of an Active Central Vent. — A vol-
cano of the central-eruption type, like all others, depends on antece-
dent ab3rssal injection of magma into the earth's crust, furnishing a
magma chamber whence the vent may draw its supply of energy.
Three possibilities are open: (1) The primary magma may have been
initially saturated with juvenile gas at the original pressure of 10,000
atmospheres or more. (2) Only the upper part of the magma may be
saturated with gas because of the change of pressure resulting from
the injection. (3) Or the magma may not be saturated immediately
after injection, even at the pressure of one atmosphere.
In the first case, bubbles must form throughout the chamber at all
levels above the original depth of the magma. In the second case,
bubbles must form at all levels above the lowest one where saturation
has been developed by change of pressure. * In the third case, bubbles
will form only after other causes than mere change of pressure have
operated. At least three such causes are conceivable, (a) The upper
part of the magma chamber might become supersaturated through
the upward molecular diffusion of gases, whereby these are concen-
trated. This is a reasonable expectation on the general principles of
physical chemistry, though experimental or other proofs are lacking.
(h) The slow crystallization of the magma might be accompanied by
the ejection of gas, as it is actually seen to emanate during the crystal-
lization of artificial slags. That process might cause local supersatura-
tion in the still liquid magma, with the formation of bubbles, (c)
274 laXEOUS ROCKS AND THEIR ORIGIN
Chemical reactioas in the magma, such as the generation of hydrogen
from dissolved primary water vapor — a reaction to be expected with
a slight fall of temperature — might produce gases insoluble in the
magma at the pressure reigning at the place of the reaction*
Among so many possibilities, it seems legitimate to assume the gen-
eration of free gas in the main magma chamber. Irrespective of their
origin, the bubbles mast rise with great slowness through the magma
chaml)er, becaase, first, they are of small size; and^ secondly, because
the viscosity of magma un<ler great pressures must be relatively high.
Even in the case of suporsaturation in all parts of the new abyssal in-
jection, the entire freeing of the bubbles may occupy many thousands
of years.
As the bubbles rise, the gas tends to \)e concentrated in the volcanic
conduit. There the laws of mass-action and of the degradation of
energy sc^cm to enforce exothermic reactioas of the gaseous con-
stituents among themselves and with the elements of the liquid
magma. It is mo-^t probable that the heat so generated is very great
when compan^l to the mass of matter participating in the reactions.
The conduit is thus a furna(*e where the potential energy of the accumu-
lating gases is converted into heat energy.
Other and perhaps v(Ty im]M>rtant sources of heat prolonging the
a<'tivity of the volcano are: (a) the conversion of the potential energy of
liquid components of the magmatic* system when thrown out of chemical
equilibrium by the change of pn»ssure and sulisequent lowering of
temperature; (b) the liberation of latent heat in the slow erystalliia-
tion at the walls of the magma chaml>er; and (c) some degree of
initial superheat in the miigma, perhaps of the order of 100^ or 200^
Centigrade.
Since the loss of heat at an iU'tive vent is chiefly due to radiation
at the crat(»r, the continuance of a<'tivity Is controlle<i by the efficiency
of the mechanism i)V which the heat of the main chamber and the hc^t
chemically generate<l in the conduit are transferred to the earth's sur-
face. Field oliservatioas at Kilauea and (^Isewhere, along with a prion
deductions, have sugg<*sted th<* gemTal <lominance of two-phase con-
vection (or, more generally, <*onvection due to systematic, local changfs
in ga**-concentration) in making this transfer.
Juvenile gas is tlias conceiv<M| to act in a two-fold capacit)* — as a
|K>sitive heater (its chemical rea<'tions tending to annul the cooling due
to cxpanson) and a>« the agent enforcing convection. Its net effect
is to keep fluid the top part of the lava column during the
activity. The conce])tion a< a whole may therefore be (»dled
fluxing hyiwithesis. For vents occu]>ied by highly fluid
hy|)othesls as just stat<.Ml seems to suffice. For c flOQie<'
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 275
more viscous lava, the emanating gas issues under more or less high
pressure and may function as a melting blast, making more perfect the
analogy with an artificial blowpipe.
Revival of Activity at the End of a Dormant Period
One of the leading problems in vulcanism relates to the periodicity
of central eruptions. This also seems to find explanation on the gas-
fluxing hypothesis. We have seen that the accumulation of gas
bubbles in the conduit mast be a very slow process. So long as the
vent is open, the escape of the gas from the magma is specially facili-
tated. That is true, not because the pressure on the main part of the
lava column is less than in times of dormancy, but because of the rapid
freeing of gas into the open air, with the consequent rapid production
of heavy, gas-freed lava which sinks and thus hastens the two-phase
convection. The tendency is, therefore, sooner or later to exhaust the
gas concentrated at the lower end of the conduit. With sufficient re-
moval of the heat-producing and heat-transferring agent, the forces of
cold temporarily win in the never-ceasing struggle and the lava solidi-
fies at the surface; a plug of greater or less thickness is formed. The
crater may become temporarily so dead that even solfataric action
ceases and a forest may flourish within the crater, as has been the case
with Vesuvius.
On account of the small horizontal dimensions of the average vent,
the consolidation of such a lava plug may be completed in a few years.
This new rock is characteristically tough; when cooled, it is the
strongest rock in the average volcanic cone. In the text-books on
dynamical geology and in special vulcanological memoirs, the removal
of the plug is usually stated to be due to simple explosion of the gases
accumulating below it. Yet it is obvious that in the normal cone,
which is largely built of loose ash deposits of very low tensile strength,
the weakest place in the pile is on its flank and not at the main central
plug. By the orthodox view, therefore, the new crater, the main one
for the succeeding period of activity, should have a different location
from that of the earlier main crater. The fact is, that, in very
many cases, the main vent is located at the same place through
the many different periods of activity of the greater cones. The
beautiful symmetry of a Fujiyama or of a Mayon is the result. The
removal of the plug at the close of a dormant period is clearly not the
mere mechanical result of explosion. There must be a preliminary
weakening of the plug, and apparently the only cause for that weaken-
to be found in the fluxing by juvenile gas.
we may consider the case where the terrestrial forces keep the
"p« ipporto? '^ *^^^ conduit. With the formation of the
276 IGNEOUS ROCKS AND THBIR ORIGIN
plug, the loBs of heat ff^ts to s very low rat« u compared with that
ruling in the active period. Until the plug is removed, neatiy all the
loss is due to conduction and is very slow. Two-phaae oonveeUoo ia
slowed down, hut the rise of bubbles does not cease nor does tbe vol-
canic furnace coasc working, since a renewed concentration of juvenile
gas is begun. To that positive source of heat in the conduit is to be
added the heat developed by the compression of the gas as it accumu-
lates beneath the plug and as it is squeezed by any uptbnutiiig of the
magmatic column due to crustal movement. Gradually tbe lowest
part of the plug liecomes liquefied, preferably along its vertical axis,
where the heat inherited from the last active period preaervea the line of
maximum temperature in the whole upper part of tbe volcano. Tbe
rcliquefied lava sinks into the column, dissolving some of tbe aeeumu-
CRATER
Fig. 1^5. — St-otiun of upper part of a ilurninnt cone, shawing i
Rafl-fluxinit. Th<> broken line in the niiilille of the vent shcnra the <
the solid pluK.
lating gai'. so that heat of solution is prol>ably to be added to the other
supplies which tend to tlircatcn the existence of the plug. Hence, at
least thriH' processes co-operate in fusing the plug; these are: beat of
clii-mical n-uction, licat of gas compression, and heat of gas iolutioD.
.\s the pluR is thus weakened, the ga.s-tcnsion increases and activify is
renewed by one or more major explosions, shattering the remaining
I>art of tlif plug (Fig. 13.'3).
Thuuiili full ex|>cTimental data for the testing of these coneluooBS
an- not yet in huiid, it is not difficult to sec that the fluxing power of
even small masses of juvenile gas is great under these cooditions. If
clK>n)ical react ions supjily any large fraction of the heat lost by india-
tion in the active )N>riod, tliey must raise the temperature of tbe lava
culunm under eonditionn of dormancy. This involves a slow n
of the country rock, and cKiicciftlly the plug.
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 277
• In most cases dormancy is ended by explosions so powerful as to
show pressures under the plugs of even higher order than the pres-
sures in the greatest modern cannon at the moment of discharge.
Tliese pressures run well over 2000 atmospheres. If a given plug,
when frozen to maximum thickness, is 1000 meters deep, the initial
pressure on the gas first collecting beneath it would be about 270
atmospheres. Amagat's experiments furnish the data from which the
temperature effect of the adiabatic .compression of the typical gases,
carbon dioxide and hydrogen, may be approximately computed.
At the request of the writer Professor H. N. Davis has kindly de-
duced the thermodynamic equation for these two cases. If To is the
initial temperature; T, the final temperature; po, the initial pressure;
and p, the final pressure, we have
^-MlT-
in which n is an exponent approximately determined from Amagat's
curves. If the initial temperature and pressure are, respectively,
1273° C. absolute and 270 atmospheres, and the final pressure is 2700
atmospheres, the average value of n for carbon dioxide is about .17;
and, for hydrogen, n probably lies between .26 and .31. For carbon
dioxide the computed value of T is 1880° absolute, and for hydrogen
2300° to 2600° absolute. This adiabatic compression of carbon dioxide
would develop heat to the amount of about 200 calories per gram.
Similar compression of pure hydrogen would develop an amount of
heat ranging from 3000 to 4000 calories per gram. The compression of
the gaseous mixture actually formed under volcanic plugs would pro-
duce heat to amounts intermediate between those calculated for carbon
dioxide and for hydrogen. Since hydrogen is one of the most abundant
constituents of the mixture, it is possible that adiabatic compressior
of the mixture, under the conditions above described, would produce
at least 1000 calories per gram of gas. Since the latent heat of holo-
cr>'stalline igneous rock is about 90 calories (Vogt), this heat of com-
pression could fuse more than 10 grams of rock per gram of gas.
Calculation further shows that if the compression of a considerable
volume of gas be isothermal, other conditions being as above assumed
for the adiabatic compression, the heat produced is of the same large
order of magnitude.
As the lower part of the plug is fused, the liquid sinks through the
gas-rich part of the magma column, so that the fluxing gas is always
in immediate contact with the solid rock. Since the solid plug retains
a relatively high temperature inherited from the last active period,
and since the vertical axis of the plug is the hottest part of the vol-
278 IGNEOUS ROCKS AND THBIR ORIOiN
cunic cone at all IcvcIh al>ove the top of the lava column^ it is clear
that fluxing will ho most rapid along the axis.
Again, a local development of heat is to be expected as the re-fused
rock, which had Ix^cn largely freed of gas in the last active period,
Ix^gins to absorb the gases collecting in the conduit. Nothing is known
as to the solution heat of any juvenile gas as it is alisorbed in a nat-
ural magma. In each case it is practically certain to be positive and
it may Ik* important in amount. The data for the same gases when
dissolved in water have some value in the way of analogy. The fol-
lowing table is taken from Thomsen's Thermochemistry:
Vapor or gas
Heat of solution
dissolved in water
For 1 gram-m(
8,430 cals.
ol.
For 1 sraiB
NH,
■
496 cab.
SO,
7,700
120
CI,
4,870
60
CO,
5.880
134
H,S
4,500
134
HCl
17,315
475
When hydrogen dissolves in water, heat to the amount of about
800 cals. ]H'r gram of the gas is evolved.*
In this whole problem it must Ik* rememl)ered that hydrogen forms
a relatively large part of juvenile gas-mixtures. This gas has the
highest s|KH'irie heat of all substances y(*t measure<l, and its hettt of
solution in wat<T is also very high. Its efficiency in melting a volcsoic
plug may ]><>rhaps hv grratcT than that of the other gases and ynipm%
])Ut together.
In view of all the conditions, it seems correct to hold that the
accunuilation of gas tH*neath a solid volcanic plug develops a q)ecial
kind of local furnace. The energy here transformed into heat is both
potential and mechanical. In part, it is heat of solution; in large part,
it may be due to chemical rea(*tions: in part, it is due to the condensa-
tion of free gas constantly incn':tsing in mass, within a closed chamber.
The increase in mass is assumed to Ih' due to the exclu.sion of gas in the
crystallization at the walls of the al>yssal injection, to diffusioQ from
great (le))tli. and {Hrhaps to othcT molecular transformations within
the magma chainlxT.
If the lava colunui is not kept sup|K>rted, but withdraws for a time
from the plug, the compression-melting of the plug must awut suf-
ficient areumulation of gas from beneath or the return of the fluid hiva
'vUuanse of general strains in the earth's crust or for other reasons)
into the conduit. The mechanism is, however, the same as in the case
' G. N. Lewis, verbal ruinmunication.
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 279
just discussed, and the base of the plug is gradually melted. Piece-
meal destruction of the base of the plug would also be brought about
through overhead stoping by the lava column — a process evidently
facilitated by just such surgings of the column as are characteristic
of Kilauea and other volcanoes.
The re-fused magma must become gradually more and more
charged with gas. How much gas per unit weight of rock would be
required to fuse an average plug is obviously now impossible to declare,
but the maximum quantity of gas in solution may not need to be more
than 2 or 3 per cent, of the total weight of the magma in the actual
conduit. The astounding explosive energy of newly awakened vol-
canoes, as shown in the vast heights to which fine ejecta are thrown
and by the excessive comminution of the respective plugs, seem to
indicate concentration of the gas to an even higher degree. The
"evisceration" of some cones has possibly been due to the concentra-
tion of juvenile gases beneath plugs not yet sufficiently fluxed to per-
mit of a reopening of the former vents by more moderate explosions.
In neither case, however, is it probable that pure explosion could restore
activity to the dormant volcanoes. Here again, as in the continuance
of activity after the vent is opened, the problem is one of heat supply.
Another cause for dormancy is to be found in the sudden emptying
of a lava-filled conduit by escape through a lateral fissure, forming
satellitie intrusion, or distant surface flow, or both at once. This is a
conmion event at both Kilauea and Mauna Loa. A multiple effect is
produced. A large volume of specially concentrated juvenile gas is
taken out of the vent, just so far diminishing the motive power and
heat supply in that vent. As observed at Kilauea, the level of the
conduit lava may not be restored to its former height for months or
years. During that time the upper part of the conduit wall is cooling,
and, through decrepitation and initial weakness, large masses fall from
the wall and choke the vent. A resumption of activity at the surface
must be delayed by these processes.
In the present argument we need not dwell on the fact that, if the
volcanic mechanism is nicely balanced, a minute effect, like tidal strain,
may pull the trigger and renew activity, for which the essential con-
ditions have been long preparing. However, it seems clear that cos-
mical stresses do not seriously deform an abyssal injection during its
lifetime as the feeder of a central vent. During such a period, which
may be thousands of years in length but seldom or never millions of
years in length, crustal readjustments must be minute, for even the
greatest lava flows that could have been thus squeezed out at central
vents are always very small in relative measure. The last-mentioned
fact and the persistent recurrence of eruption at the main vent appear
280 IGNEOUS ROCKS AND THBIB OBIOIN
to forbid the hypothesis that renewal of activity at central Tenia la due
to reneical of injection along new abyssal fissures. It would be highly
improliable that the vont of a seoontl injection would coincide with that
of the fir»t injection; and on the other hand, the great cruatal disturb-
ance accompanying the second injection should aonnolly cause firat-
i-lass lava floods at .the initial vent, instead of the comparfttively
insignificant flows actually oljscrved at central vents. IXfficult as
the problem is, the change from dormancy to activity does not, in
general, seem to call for anything so drastic as a strong deformation
of the earth's cru»t in its entire thickness.
In conclusion, the gan-fluxing h>'pothc8is appears to be worthy of a
leading place among thone which can be constructed to account for the
stubborn persistence in the revival of activity at a vent like Mokua-
weoweo, Vesuvius, or Ktna.
Small Size of Central Vests
The gad-fluxing hypothesis accounts for other general features of
central eruptions. The small cross-sections of the vents at Kilauea,
Hualalai, Maunn Ix>a, and even at Mokuaweoweo, as everywhere else
in the world, arc all of the order of size expected if the fluidity of each
lava column is due to the slow passage of relatively minute masse* of
gas through those vents.
The writer is not aide to agree with J. D. Dana, that the conduits
lK>neath Kilauea and Mauna I,<>a arc nearly equivalent in horisontal
section to the great sinks (" calderas '') in which the lava lakes are sit-
uated. Each of tliosc sinks measures roughly 5 kilometers by 3
kilometers. The periodic rise and fall of the floor of the Kilauean
sink (the only one carefully studie<l) can be explained on the aamimption
that its conduit has a much smaller cross-section. The "New Lake,"
after Ave years of activity, was emptie<l in 1886, and was proved to
liJivc had a tlcpth of only a few meters. It was a Baucer-6hi^)ed dieet of
lava resting on solid ruck. When the present Halemaumau is emptied,
the lava runs out through a very narrow hole apparently leas than
;iO meters wiile, and leaves a broad, funnel-shape<l cavity. The action
is like (hat of water running out of a domestic sink with centrally
placed discharge; in l>uth casts vortical motion is observed in the rap-
idly escaping liquid.' Similarly, the \-ast Kilauean lake o( 1890 to
IStK) is Ix'st interpreted as a true lake with solid floor, exDC|it for tt>
narrow pipe which has always supplied the heat at this i
That pipe is probably the same pipe into which Hi
discharges its lava and from which the j
■ C. H. Ililchcook, Hawaii nnd its Volcanoes,
, exec|« for tt*
t this vdm^M
MECHANISM OP VOLCANIC VENTS OP THE CENTRAL TYPE 281
tains of "Old Faithful." All the Kilauean lakes have represented over-
fiowa from that vent or from a few, more temporary, narrow pipes.
The fluidity of the lake has, in each case, been preserved for years by
the process above outlined for the existing lake.
Whatever adverse criticism of this conclusion regarding Kilauea
may succeed, it is certain that the whole area of either of the Hawaiian
sinks cannot be directly taken to represent the size of the conduits.
The surface areas of other lava columns active in historic time are all
Fia. 136. — Section and plan of basalt-lava neck in a lateral gorge of the lao
valley in Weet Maui, illuBtrating the cylindrical form due to gas-fluxing. This WM
oae of the subsidiary vents on the ftanlca of the great West Maui cone. There is no
trace of faulting in the well-exposed ash-and-flow series and it is possible that this
neck Tepreeents the local enlargement (by gas-fiuxing) of one of the dike fisBUrCB
now visible in the canyon. Nearly natural scale; major diameter of the neck about
50 meters. Sketched in the field by the author.
nrj much smaller. It is doubtful that any one of them, just below
the floor of the Saring crater, has been as much as 1 kilometer.
1. 1). DaQA romputed the volume of the 1852 floor from Mauna Loa,
*likli uppears to have emptied the conduit to a depth of 2500 feet,
UKtimatcd itf"\ t**" d'''''rence of level of the sunmiit lake and of the
Pf^qfdifieb >«ult was 10,560,000,000 cubic feet.' This
7<ricKnoM, New York, 1891, p. 240.
282 IGNEOUS ROCKS AND THEIR ORIGIN
corresponds to the volume in a cylindrical conduit about 2300 feet or
700 meters in diameter. Similar calculations from other lateral out-
flows seem to give a mean diameter for the conduit of the same order
of magnitude. Such a lateral fissure once opened, it would seem highly
probable that the conduit would l)e emptied almost entirely by the
simple outflow of the lava through the fissure; discharge into ''8ul>-
terranean cavities, '* would Ix> unlikely. Moreover, it is possible that
some of the 1852 lava represents a temporary rise of magma in the
conduit, so that only part of the estimated volume of the flow can l»e
used in calculating the average diameter of the Mauna Loa conduit.
Thus, the calculation made according to the method outlined, strength-
ens the suspi(*ion that tin* lava column of the world's vastest volcano is
but a comparatively narrow pifM*, perhaps much less than GOO meters
in average diameter.
All of the ancient central vents now exjwsed as "necks'* after pnn
hniged denudation, are relativ(>ly small. (Compare Fig. 136 and
also page 127 ff). The average diameter of the pipes recorded in geo-
logical literature is well under 3(K) meters.
We may conclude that the conduits of central eruptions are always
small and of the order of magnitu<le appropriate to the gas-fluxing
hypothesis. On any other hypoth(*sis it is hard to explain the fart
that the pi|H>s of modcTately large cones are alx)ut as large as those of
the very greatest cones. In all ca.ses there seems to be a UmiUdgizt
and that is controlled by the available heat supply along the axis of
the vent. The size is small l)ecause the (indirect) fusing power of ema-
nating gas must Ih» strictly limited. Moreover, the cylindrical shape
of each typiral pipe is a solutional or fluxing form (Fig. 136).
Kxi'Lo.six K TvpKs: Ma(;mati(; and Piikeatic
The f(»regoing genetic statement for the Hawaiian vents has lieen
sketched in terms of a cjuite g<'neral process and it is necessary to glanre
at th(» relation of the hypothesis to the explosive tyi)e of central eruption.
Volatile matter occurring in the nn'ks of the contact-shell about any
intru>ive magma mu^t show increased tension. If the intniaion is
large an<l near enough to the earth's surface, this tension may lead to
explo>ion in the rcN)f of the igneous body. In ca-^e no incandescent
matter is extrud(*d. the explosion is not strictly volcanic. FoUomng
Suess, it may be calletl phnntic. A similar explosicm may
result of the slow eoiKhietion of heat from the c(mduit of
mant volcanic cone. Such a cone is normally porous,
snow-water, or sea- water is trapjM'd in i** ^-esicular vn-
tuffs, as these are in turn burie<l durin
MECHANISM OP VOLCANIC VENTS OF THE CENTRAL TYPE 283
The circulation of vadose water is also facilitated by this special
porosity.
Fic. 137. — Schematic plan and section ot the Rieskesscl, with its zones of frac-
ture (faults). (After W. Branco and E. Fraas, Abhand. k. preuse. Akad. Wiss.,
1901, p. 39.) The Oat floor of the depression is about 10 miles (16 km.) in diameter.
The suggestion of Suess that the remarkable explosion at the well-
known Ricskessel was of phreatic origin has been supported by the
F^K- 138. — Section of the Rieskessel, showing inferred laccolith beneath.
(Adapted from W. Branco and E. Fraas, Taf. 1 in ref. of Fig. 137.) 0, Grundge-
birge; GB, brccciatcd granite of the Grundgcbirge; BJ, Brown Jura; WJ,
White Jura; B, breccia; T, Tertiary and Quaternary aedimenta; LT, Uparite tuff
of Branco and others.' Purely geological studies bad
vce of a large laccolithic mass beneath the great
VOn. preuse. Akad. Wisa. Berlin, 1902, p. 14.
384
IQNBOVS ROCKS AND THBIB OBtOTN
Ries depression (Figs. 137 and 138). That concliuion has beeo
brilliantly supported by the ma^etic studies of Hauaimanii in the
region. The local diHturbancefl of the needle in dip and anmuth can
))c explained, aeconling to Haussmann, only by the aBBumptkm of
one or more large nubterranoan bodies of basic rock (Fig 139).' In
the RieskpMHcl itself the upper surface of the bawc rock is calculated
to l)c no more than 2 kilometers deep. Outside the depremm, its
average depth was eHtimated at 5 kilometers. Since the visible floor
of the Hies is the granite of the " Gnindgebtrge," the basie man or
masses can only lie interpreted as due to injection. Braooo dates
Ftu. 139.~Mat(ncti(r ixofionpH for IQOI, Rin district, Gcmuy. (Aft«r K.
Haussmann, At)hand. k. preuM. Akad. Wiss., Phys.-msth. CI., 1904.) N, SttA-
linftcn; .S, 8ti?jnhpin) liniiin. The niBRnetir abnortnalilin an esqiUiiMii bjr pcalu-
lating one or niurt- larxc fcmir larcoliths not far from the aurfaee. Beala, 1 •.StOJBOO.
the intrusion of the itip» "laccolith" in the mid-Miocene. The
lying granite and its Menozoic xedimentary veneer were domi
the injection and the top of the dome was largely destrojred
phreatic explosion. It wa.s followed by the appearance of a
lipurilic tuff erupted ut it few point.-* in the newly foimed basil
the explosion itself wo.-* non-volcanic.
llrancu and Fraas have concluded that the St«nbeini bMB
sniull-seale equivalent of the Kieskessel, again showing the
succc»xion of events: laecolithir doming, phreatic explosioil, •■
' K. ItnuMmann. Abhanil. kun, preuss. Akad. Wiaa. BeiUn, 19IVI, p 137.
Itaa aiiRRCHtrd that th<- lipuritr of the Kjakeawl luffs I if BMdtin
intru'lod fiTanite >>y the banir inamna. Roe W. Bnaeo
kitn. preuM. Akad. Viias. Berlin, ISOl, p. H.
OVfT-
d by
by a
Uttle
I, but
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 285
sidence od peripheral faults. No m^matic material was here erupted. '
(Fig. 140).
According to Sekiya and Kikuchi, the great explosion of 188S at
Bandai-San was absolutely unaccompanied by the extrusion of lava.'
A priest living on the mountain survived the explosion. He reported
the vapors surrounding him to have been respirable, and the Japanese
geologists conclude from all available data that the catastrophe was a
steam explosion. There were no signs that juvenile gases formed an
important part of the volatile mixture. This "eruption" of Bandai-
San seems, therefore, to be an excellent example of a phreatic explo-
sion on a true volcanic cone (Fig HI).
Phreatic eruption means steam-explosion without magmatic extru-
sion. Kilauea represents magmatic extrusion without steam-explo-
KGROTHAU
Fia. 140. — Section of the cryptovolcanic dome of theSteioheim Baain, Germany.
(After W, Branco and E. Fraas, Abhand. k. preuBS. Akad. Wiss., 1905, p. 21.)
B, Brown Jura; W, White Jura; G, Gries formation; L, calcareous sinter. The
doming is explained by laccolithic intrusion; the depression formed by phreatic
explosion followed by erosion.
sion. Between these two extremes of terrestrial activity stands the
type representing the vast majority of active and extinct central erup-
tions. In the non-volcanic or pseudo-volcanic activity of Bandai-San
in 1888, as in a Kilauea or a Vesuvius, true igneous injection is a pre-
requisite. The gases given off at Kilauea form a nearly pure juvenile
mixture with characteristic high temperature. The gases given off at
Vesuvius form a mixture of juvenile, resurgent, and vadose volatile
matter. A type of the resurgent gas is the carbon dioxide set free in
the demonstrable assimilation of Mesozoic limestone and dolomite in
the Vesuvian lava column. The gas and vapor given off at Bandai-
San in 1888 was apparently almost purely vadose or meteoric in origin.
True volcanoes of the central-eruption type mvist vary enormously
in the relative and absolute proportions of juvenile, resurgent, and
vadose fluids composing their emanations. As the resurgent and
■T. Branoo and E. Fraas, Abhand. preuss. Akad, Wisa., 1906, p. 21.
■ A. f>ckiya ar ' I. Klkuohi, Jour. Coll. Science, Tokio, Vol. 3, 1889, p. 106.
k V. T. Lea {Ba A ioa. Vol. 18, 1907, p. 218) explaina the Kikiunt
286 IGNEOUS ROCKS AND THEIR OttlOIN
vadose flui<b arc volatilized, hoat ia lost and the viscority of the lava
column riitofl. AfMimilation of foreign rook in depth must lower the
temperature, an<l in the end, increase the viscosity and aim the avenge
violence of pxploRionfl. In ad<lition, magmatic differeatiation generally
hringf! the more (iilicioux and more viscous pole to the upper part
of the lava column, and aids in the preparation of explouve o
Km. 141 — t'Inn iiikI mi'litin iif ihv rnlilera furmeit ptircktirally a
(Kolinmlaif , J:>i>;>ii, in l>i.s.^, :iftcr S Sikiyo unci V. Kikiirlii. Jour. Coll. 8d. JafMa,
Vol. :t. 1KS<I, \'\. 2:t. Ht>iKht!> ill mi'liTS.
For tlicsc and other reasons, volrani«'s of the ecntral-eruptioD type
have »lwayF< liail & kd-hI variety in dynamic lialtil and in the chi
of their ejei'tanu-nta. Yet, in every one of them, thee Lialfi
is the same; it refers to the mechanism liy which
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 287
to the narrow, thread-like vents for long periods. To that problem, the
questions as to how sea-water or vadose water is absorbed by under-
ground magma, as to the dominance or subdominance of steam-explo-
sion at individual vents, and as to the physical differences in the
emanating lavas, are subsidiary. The problem of the Hawaiian vents
is, from this point of view, the problem of all vulcanism reduced to its
lowest terms. Here the gas-fluxing hypothesis seems satisfactory. In
most other volcanic regions, where thick sediments are cut by the feed-
ing magma or where heavy snows or rains wet the cones and, through
s^Pflge, cause steam-explosions, the control by juvenile gas maj'^ be
obscured to the eye of the observer, but it still remains in every case
the true cause of continued activity. Kilauea and Mokuaweoweo, like
Matavanu and the vents in Reunion, teach us that steam-explosion is
an adventitious feature of vulcanism. Except abyssal injection itself,
the only indispensable process in central vents is quiet exhalation.
Neither explosive drilling of the vent, nor ejection of lava, nor the
contacting of meteoric or marine water with hot lava is indispensable.
Each of these three processes is an expected effect of the slow emana-
tion of juvenile gas from main abyssal injections or from their satel-
litic offshoots.
Magmatic Differentiation at Central Vents
The chemical variation exhibited in the lavas or pyroclastic
materials successively ejected at the normal vent offers a problem of
special importance. Volume for volume, this variability is much more
striking than it is in the average large intrusive body — stock, batholith,
laccolith, or sheet. At present many petrologists favor the pure-
differentiation theory, which regards the splitting magma as primar}^
and finds no place for notable assimilation of wall rocks by the primary
magma. The writer believes that this question can only be cleared up
by an attentive study of the world's plutonic masses, and that, in the
nature of the case, its answer is not to be found at central vents. Field
and chemical relations point indubitably to the fact that wholesale as-
similation has occurred in the subjacent bodies classed as stocks and
batholiths. Because of assimilation these masses generally have not
the basaltic composition of the primary abyssal injection. The visible
granite, diorite, or syenite represents the frozen top of an abyssal in-
jection which is there a more or less differentiated syntectic. The
lower part of each injection, approaching the substratum level, is prob-
ably basaltic and little modified in composition from its original con-
dition. The syntectic-differentiation theory is so strong that the
wnUx is disposed to prophesy its ultimate victory in the competition
ezpli lations of the igneous magmas and rocks.
288 IGSEOUS ROCKS AND THEIR ORIGIN
Since the lava column of every volcanic vent is an offshoot from an
abyssal injection, the lava may represent either the pnmKty basalt,
or one of its difTerentiates, or syntectic material, or a differentiate
from syntectic material. The rapid chemical variations in the extru-
sive magma at the average central vent shows that the eonditionB are
here specially favorable for differentiation. Two of thece coiiditioni
are implied in the essential mechanism of central eruption. The up-
ward passage of juvenile and resurgent gases in great relative abun-
dance lowers the ''point'' of solidification of the magma, increases the
fluidity, and probably in still other ways aids in magmatic splitting.
Secondly, the alternation of active and dormant periods means that
the top of the lava column passes repeatedly through the narrow range
of temperature (just alx)vc the cr^'stallization point), where differentia*
tion is most likely to take place.' High superheat is opposed to mag-
matic splitting.
Ea(*h of these conditions affects only a small volume of magma at
any one time; if lava representing either pole of the differentiation is
alone extruded, the volume of that flow must be relatively small.
New magma rises in the vent. It may be mixed with that represent-
ing the other pole of the differentiation just accomplished. The mix-
ture may be extruded, or it may itself be differentiated before the next
outflow. Through absorption of foreign rock the new lava may have
a compix^ition unlike that originally differentiated.
It s(^»nis inevitable, therefore, that, at the restless volcanic vent,
the ever-changing conditions mast make a cone which is chemically
heterogeneous to an extent not matched in the usual plutonic mass.
Standard example's liave l>een described at Electric Peak and Sepul-
chre Mountain in the Yellowstone Park; at the Lipari Island vents; at
Tonopah, Nevada; at the Kal'erstuhl in Baden, etc.
Progress in Explosiveness at the Greater Vmrre
The explosive effect at central vents is a function of the magmatic
viscosity and of giis-tension, which means ga*<-eoncentration.
Though the presence of much gas tends to lower the viseositj,
temperature is obviously in dominant control over that property of
nuigma. Th<> initial store of heat in the abyssal injection is normality
lost through radiation in the crater, through conduction at the roof
and walls of the whole magma chamber, through assimilatioa of
country rock, and possi!>ly through the alisorption of vadose wstcr.
As the whole ma^< cools, the juvenile ga«< emanates with ever lowcfiog
^ InjfM^tod ina»<o!9 normally poivt through this temperature range OB|]r
before solidification.
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 289
temperature and the lava of the volcanic conduit must have a slow
decrea&e of average temperature.
Magmatic differentiation must tend to affect the viscosity of the
upper zone of lava, the exploding zone, in the same sense. The more
acid differentiate usually rises toward the top of the vent. Though
differentiation may be roughly cyclical, the successive splittings tend
to mal^e a secular increase of acidity in the upper zone of the conduit
magma. Hence, irrespective of temperature, there is an increase of
viscosity in the magma zone where explosions originate. The case of
Mauna Kea, a basaltic lava-dome capped by cinder cones of andesite
and trachydolerite, is an example of the partial control by magmatic
differentiation over explosiveness. As the viscosity rises, the escape
of magmatic gases is more difficult; the resulting tension is periodically
relieved by explosions. Here also the action is cyclical, but there is,
on the average, a slow increase in the amount of gas trapped before
each explosion.
Again, the amount of volatile matter entering the magma column,
either through assimilation of sediments or through the direct absorp-
tion of meteoric water, tends to increase with process of time. And,
with the growth of a great, generally porous cone, the chance for
phreatic explosion at or near the crater is favored.
All of these factors work together to produce maximum explosive-
ness at central vents which are long-lived because fed from great abys-
sal injections. The maximum normally appears in an advanced stage
in the evolution of a first-class volcanic cone, though necessarily some
time before complete extinction. Short-lived vents, opened above
satellitic and therefore relatively small injections, will, of coiu*se, have
no such tendency to great systematic change in explosiveness.
Lava Outflow at Central Vents
A noteworthy feature of all central eruptions is the relatively insig-
nificant size of their individual lava flows. Thoroddsen has estimated
the volume of the celebrated fissure eruption of Skaptar Jokull in
Iceland at 12,320 millions of cubic meters. He gives the volume of
one prehistoric flow as 43,160 millions of cubic meters; of a second,
23,250 millions of cubic meters. In striking contrast are the following
examples of the larger recorded flows at volcanic cones. (See p. 290.)
Geological investigation shows that the flows from the central vents
of Paleozoic and later periods have been of the same order of magni-
tude as the flows of the human period. With very few exceptions or
none at all, these larger flows have issued from lateral fissures in the
cones, and a large part of the volume of each flow is readily explained
290
lONEOlS ROCKsS AM) THEIR ORIGIN
as tlie lava drained out, hydrostiitically, from the upper part of each
conduit. Without rocordiMl exception all overflows at the main cra-
ters are incomparahly smaller than those noted in the following table.
Therefore, the as<'(»nsive force in central conduits is either slight, or,
if powerful, is applied for short periods.
LcM'ality of flow
Date
Vuliunc*
of nil
in millions
»ir nietors
Sfincnx*, Java .
1S.S.J
:joo
Ktna. .
1(»G»
9S0
Ktna
isr)2
4'20
Ktna... .
1W)5
92
Ktna
1S79
57
Matina Ix)a, Hawaii
isr)2
299
Ma una I^oa, Hawaii
1 S.').")
4.').i
Mauna Loa, Hawaii
isso-1
413
Maiina Ixia, Hawaii.
11H)7
l.W
Authority
De Lapparent.
von Walterahaufcn.
von Waltenhausm
von Waltenhaum.
von Waltetihansen.
J. D. Dana.
C. H. Hitchcock.
C. H. Hitchcock.
E. D. Baldwin.
As above noted, the smallness of individual overflows suggests that
the magma chambers which continue to feed central vents are very sel-
dom deformed by important movements of the earth's crust. If the
magma in the chamlxT were <liastrophically pinched, we should expect,
at times, rehitivelv enormous iavti-fioods from central vents. Some
authors hohl. on the contrary, that the growth of a great cone some-
times occasions subsidence, so that crustal movement may be a conse-
quence ratluT than a direct cause of lava overflow at central vents.
Without entering further into this subje<*t, it will here sufllce to
mention the priii(*ipMl causes for lava outflow as deduced from the
abyssal-injection premise. Th(»v are:
1. V<Ty minute deformation of the feeding magma chamber.
2. The efTcrvcsccnce of lava, due to the p<*rio4lic accumulation of
magmatic gases in tin* vent. These gases may Ik* juvenile or resurgent.
W. The assimihition of (*ountry rock in depth, leading, probably, to
increase of vohime.
4. Theincreas(M>f vt>hinie through heating in the conduit "furnace**
— a process specially Hkcly to o<*cur during the dormant period when
the vent is temporarily plugge<l.
These <'auses may co-operate, but at basaltic volcanoes the third is
clearlv >ubor(linate.
TiiK Two TvPKs OF Lava Flows
A preliminary study of the* Hawaiian lavas, with respect to their
field habit, has led the writer to sus|H>ct significant gas-oontrol even in
this detail of vulcanism. On the average the vesiculation of pahoehoe
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 291
or ropy lava was found to be more evenly developed than in the aa or
block lava. The rather uniformly disseminated vesicles of pahoehoe
are of relatively small and relatively uniform size, and tend to have
spherical form. The irregularly distributed vesicles of aa lava are
generally larger, though more variable in size; much fewer in number,
and of less total volume per unit volume of rock; and more irregular
in shape. These facts indicate a more uniform distribution of gas
in the pahoehoe than in the aa type. The aa vesicle, which is often
thousands of time bigger than the average pahoehoe vesicle, has
undoubtedly grown through the coalescence of many bubbles of gas.
Such growth must in very high degree (see page 262) favor the escape
of the gas into the air, and we may regard these large vesicles as
representing so much gas trapped in the freezing lava. Before solidi-
fication had set in, gas must have escaped from every aa flow in large
volume. In fact, observers of the two types in actual movement agree
that the gas emanation from flowing aa lava is much more abundant
than that from flowing pahoehoe.^
The difiference of field habit in fluent lava and block lava is thus
explained, with some show of probability, by the relative abundance of
volatile matter and, still more, by the evenness of its distribution.
For both reasons pahoehoe lava is certain to be less viscous than is aa
lava, other conditions being the same; the pahoehoe moves, as it were*
on molecular and vesicular ^'ball-bearings."
The fact that many flows, from the very points of emission, are al-
together of the one type, while others are throughout of the other type,
shows that the differences of gas-distribution are developed in the vent.
The problem as to exactly what circumstances there control the gas-
distribution has not yet been solved. Slight differences in temperature,
or differences in the advance toward solidification (with gas expulsion)
may be the effective cause. The writer has observed a tendency for
the phenocrysts of aa lava to be of larger average size than those
in pahoehoe lava which gives practically the same oxide proportions
in ordinary chemical analysis (volatile matter other than water neg-
lected) ; but he is as yet not prepared to regard this as an established
rule.
VuLCANisM Originating in Satellitic Injections
We have so far considered central vents as, in general, direct off-
shoots of main abyssal injections. The latter have been described:
»Cf. J. D. Dana, Characteristics of Volcanoes, New York, 1891, p. 242.
Judge Hitchcock describes a typiced Hawaiian aa flow as advancing ''with no
explosions, but a tremendous roaring, like ten thousand blast-furnaces all at work
at once."
292 IGNEOUS ROCKS AND THEIR ORIGIN
<as dikc-Iikc, though often of great widths ; as extending upward from the
primary substratum, nearly or quite to the earth's surface for some
such vertical distance as 40 kilometers. Batholiths have been in-
terpreteil as chemically modified abyssal injections of the primary
basalt. Plutonic stocks and Ixxsses represent cupolas in batholithic
roofs. Stocks, l>osses, and batholiths compose the group of "sub-
jacent'' intrusive l)odies. I^Accoliths, sheets, and ordinary dikes are
individualized bodies, satellitic with respect to their feeding abyssal
injections and, like the latter, owe their intrusion to a umple parting
of the invaded rock-formations. Irregular bodies intruded in the
same fashion have been called ''chonoliths"; they form a fourth ebss
of "satellitic injections."
All satellitic injections soon lose thermal and hydrostatic connection
with their respective abyssal injections. All laccoliths and chonoliths,
like most sheets and some dikes, have solid floors during most of their
magmatic activity. If a satellitic injection is of large sise, its content
of heat energy and of gas may suffice to open one or more venta to the
earth's surface, according to the methods already described. Volcanic
action is thus initiated which differs in some respects from that due to
direct emanations from a main abyssal injection.
The importance of this fact is manifold. Its recognition aids in our
understanding: the short life of many volcanoes of the central tjrpe;
the lack of lava flows at many of them ; the independent activity of
neighl)oring vents; the chemical dissimilarity of the lavas from nagh*
boring vents; the quite common clustering of many small vents in a
region which shows no trace, or but few traces, of the alignment of its
volcanoes; and the frequent evidence of surface deformation in such
regions. The evidences for this type of vulcanism are indirect, but
they are numerous; taken together, they form a combination of no
mean strength.
In the first place, an excellent analogy to the vents from saleOitie
injections can l)e observed in nature. The blow-holes and driblet cones
formed on the surface of the deep lava flows of Etna, Rtenion, Hawaii,
Savaii, etc., are continued in their brief activity because of the thermo-
gaseous energy- of lava quite removed from the parent vent. The blow*
holes occasionally opened in the dome-shaped "bulges'* or ''tumuli*'
formed on the paho<4)oe of Hawaii or Reunion are particulariy instmc-
tive, for such tumuli, when just formed, represent small laccoliths of
still fluid lava capped by recently frozen lava-crust.
To the weight of analogy is to be added that of o priori reasoning
According to almost any of the extant theories of igneous aetioii, vul-
canism originating in magmatic satellites should be sxpseledL Maiqr
satellitic injections of great size have been exposed by tfosion; it would
AtfISM OF VOLCANIC VENTS OF THE CENTRAL TYPE '.
294 lONEOVS ROCKS AND THEIR ORJQIN
lie a matter for distinct surprise if none of them ever paforated
its roof.
Fielit observation mu-st naturally make the compelling test of the
principle. Have ne any active example? Can we God traces of it in
denuded rcKJons where eroxion enables us to study the anatomy of vol-
canoes? Kach method of applying the field test has its own difficulty.
In the first case the satellitic injection is inaccessible and can only bi:
loeatcil through inference; in the second case it is but rarely that de
nu<lation could ex^Mise the injected mass without destroying the con-
duit alntve. Yet the wTiter l>elieves that the field inferences seem to
ftupjKirt t he prini'iple.
The ease of Kiluuea as an illustration (Fig. 142) was detailed in
the VfTiter's ori);inid paper on "The Natureof Volcanic Action" (1911).
Considering the fiehl rehitions of the Hafvaiian vents it seems justi-
I of till' Ilr'MKitborK rulcano, Irt-hind. (After H. ^tA,
1. <;.-.., Vol. m. 1910, p. 31<t.} P, palftsonile; D, doknW:
HiiH ix :in Krli'-liunKHkralpr, tormetl by PspkiaioD of rmm
I'l-ii'il. Lii'i'iilithii- H\a.m of lava, and thus illuiitrmles "tutMrdi-
fiiil>l<- to class Kilaueii tcntiitivily as the living vent of a still liquid
s:it<-lliti<- injection. We may also citm-lude that it is unsafe to deny,
simply liociiuse of the hydrostatic independence of the two active
llawaiiiin 1hv:i columns, that a primary fluid substratum of basaltic
comjxisition uiulirlii-s the whole island.
An Icelandic Example.-- Keck has sngKc^tcd that the crater of the
Ilros.saliorfc in central Icelaml is a Ki^-<'Xploded opening (diatreme) in
the roof of ii laccolith. The injection is of comparatively late dale,
presumably Hecent or ijuatt-rnary' (Fig. 143). A volcanic pipe at
Oorli^'s I'oort, CaiH- Province, South .Africa, is described as having
been fed from a .-'ill.'
We may now )<rietly note other prul>able examples of the opraiog of
vents iiliovi' siiti'llitic injections.
Tertiary and Older Vents from Satellitic Ittjectioiu. Swabiaa
and Scottish Examples. .Vs a result of his extraordinarily thorough
' It. Iti'vk, M.iTutv-lir. •Utiit. it'-<>l. (ii-ri. No. 4, 1910, p. 293.
' A. L. <lu Toil, Ulth Ann. ilcp. (iral. Comm., Cape of Good Hops, 191^ p. Ul.
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 295
study of the mid-Miocene eruptions in Swabia, Branco concluded that
the Urach region is underlain by a ** kuchenformige Masse," or laccolith.
In ground-plan its estimated diameters range between 30 and 45 kilo-
meters. Its position coincides with that of a very low, but broad
KIRCHHEIM
r
e
10
2.
Km.
5
10
MIt.
Fio. 144. — Map of part of Swabia, showing positions (black spots) of " Vulkan-
Embryonen." (After W. Branco, Schwabens 125 Vulkan-Embryonen, 1894.)
The dotted line is the edge of the Swabian Alb escarpment.
doming of the Jurassic strata in the Bavarian Alb, as determined
by Regelmann. Through secular erosion, the frontal escarpment of
the Alb has retreated at least 25 kilometers since the volcanic epoch
(Fig. 144).
296
IGNEOUS ROCKS AND THEIR ORIGIN
On the top of the Alb plateau are thirty-eight tuff venti; in thr
eacarpment arc thirty-five more, and in the "Vorland," or repon
traversed by the escarpment in its southward retreat, there are fifty-
four tuff vents and five basaltic vents {¥\%. 145). No lava flmn oc-
curred on the Alb, and the few lava necks have become viable because
of denudation in the "Vorland." Though the exploeion funndB are
Ktill more or Ie»H Jntac-t on the Alb, the largest of them, the Randerk
" Moar," iloeti not exceed 1 kilometer in diameter. The average diam-
eter of the 132 ventx is far less. The evidence is clear as to tbe short
life of ea<-h of thettc vontn, which Branco has made the world type of
"volcanic embryos." Their brief, almost wholly explosive activitin.
Fii;. \Ah. — Planii iiml Rortionn of flwabian necks ('"
{Sumr rrf. OB Tor Fiff. Ui, pp. 202. 367, 376.) /, Kft-k south of BcaacB. f.Kcefca
nt the Hohbohl anil ritjtenhrUhl. S, Plan of the G«ienbr11hl neck, i, Nedi cast
of SccburR. J, Jurafwic Mrata: T. tulT: B, baaalt.
their distribution in a cluster nithout reference to roaster fraetum,
and tile dome-like wnrpinf; uf the Jurassic beds in this region, all de-
clare the juotice of Branco'« larcolithic hypotheeia. Furtbennorr,
his discutisiun of MandeNloh'tt :)40-meter boring at Neuffen sbowi that
the temperature gradient in at least one part of the Uracb rc^lM is
abnormally high, alxiut 10 meters per d^rce Centigrade. Braaeo re-
gards the abnurmul gradient as due to the wave of heat rtill being
conducteil' from the mid-Miocene injection. This suggestion is by no
meanH extrem<- and it clearly tends to support his IrHMillllili tijUNri liesii '
Tbe peculiar abundance of small tuff-necks of Pomian age in parts
> W. Branco, Srhwubcn'N 12.> Vulkan-Embrj-onen, Stottptt, I8M. Cf. E.
Suesa, Das Antlitt dor Erdo, Bd. 3, 2l« HAlfte, 1909, p. SU.
MECHANISM OF VOLCANIC VENTS OP THE CENTRAL TYPE 297
Oj^ FOffTfi
Fio. 146. — Map of pari of Fifeshire, Scotland. (After A. Geikie, Geology of
EMtern Fife, 1902, map.) OA, Old Red (aDdesiteaad dacite); OS, Old Red (eaad-
■tooe); C, Carboniferous; D, dolerite sille; T, tufTs (Permian); loXid Uaek, necks
(Pennian). The close association of sills and necka suggesU the possibility that
Ibe latter represent "subordinate" vulcaniam.
Fi<i. 147. — Section of the coast of Fifcflhire, between St. Monaos and Elie,
•howing volcanic necka in folded Carboniferous sediments. (Same ref. as for Fig.
IM, p. 112.) T, Agglomvate and tuff; B, basalt. Scale approximate.
298
IGNEOUS ROCKS AND THEIR ORIOIN
of ScotlaiKl \» siitijcct to u similar tentative explanation. Various
memoirs of A. (icikip Iihvc inado tlicsp vpnts famouB m tjiw* of true
necks. In Fifeshire, eighty of them have l»eei» countnl in an area
nioasuriuK 18 kihmieters by 10 kitomet^-rs (Fig. 146). In western
Ayrshire, sixty rents ar<- roiiinl In an area measuring 60 by 30 lulo-
mt'ters, anil, of thom- rents, twenty are necks oecurring within ao area
of ^!i stgiiure kilometers. In the great majority of caws, Geikie and
his eolliiliomtorK have iH'en unahic to find any connection between
tlie iiositions of the necks and lines of dislocation. The Carbonif-
erous strata have suffered sieve-hkc perforation like that of ibe Juraa-
Vui. us.' -1 1 1 Va»U fii'c of the New M<iiintain ut I'du-Hsn, J
in-ciiiii t'>ii: tiiihckrt.rrutirh'iM foniiiil iliirinK the periodoT rierklkm. (AfM
)-:. It. It^kilrv. (Ii-.>1. M:it:.. Vol- '1. ntl2. p. -J4H, un<l F. Omori, Bull Inp. Evtk-
<,ti:ikr liivi'~Ni::ili.>n (-..nirii.. Vol. ■'>, No. 1.)
<2< l^■ll-^:lll n-illi the N<-w Moimriiiii us sith from wrow lh« Lftkeaf T«)r»-
iS:ilii.-r.-r.)
siclu'dsinSwaliiiiiFifE. 147^ In each of I lie Scottish districts, the lower
part iif l)ie very thick Carboniferous sedimentary series carries nu-
iiicrous ttiick sills of dolerite. 'rhes<' sills are mapped as chiefly Car-
)HinifcriiL]< in il:it<-. Imt Ceikie thinks that some of the Fifeshire rills
at Ic.-tst :iTv rcniiian. The steady tussociat inn of tuff-neck and sll in
t)ic S.'<>lti>li .-hires scarcely looks accidental.
(ki- ctn;iiiali<ins rroin the inai;nia forming these actual intnisiTM
or .-iiinilar oiic^ ociurririi; in the unclcrlying pro^arboniferous for-
mations, together with the possililt- emanation of gas from the heated
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 299
5]i
ill
hi' •
Is I 38 a -|
300 IGNEOUS ROCKS AND THEIR ORIOIN
country-rock, would seem to be competent to explain most of the tuff-
necks. Explosive drilling (diatremes) and gas-fluxing might in turn
dominate in the opening of the vents. The total activity must be
small, because each gas-emitting or lava-emitting chamber was small.
The list of distri(*ts where the ^Titer suspects secondary vulcanism
includes also the area of necks in Noss Sound, Shetland.' These are
small and the volcanic throats are filled with a coarse agglomerate
of sandstone and shale. Peach and Home infer that the vents never
emitted any streams of lava.* The eruptivity is referred to the Lower
Old Red Sandstone period. The date may be nearly the same as that
of the injection or the thick sills and dikes which abound in the Noss
Sound region.
Finally, the forty-five new craterlets ranging along the foot of
the ''New Mountain" at Usu-San, Japan, afford other probable
examples (Figs. 148, 148a). This remarkable deformation of the
land surface is clearly due to magmatic injection not far below, and it
seems likely that the craterlets have been opened through the inde-
pendent activity of the injected mass of magma. •
A Necessary Division of Central Vents. — It is obviously difficult
to devise field criteria which shall infallibly distinguish centnd erup-
tions respectively originating in main abyssal injections and in satel-
litic injections. Long and strong activity, large outflow of lava, and
alignment in chains will generally characterize the centnd vents of
abyssal injections. Brief activity, small output of lava, cluster group-
ing, and traces of surface deformation in the region are the generally
expected features of the central vents of satellitic injections. As one
or more of these features is absent or is ol>scured, the classification
is hard to apply.
The chief object of the foregoing discu.»<sion has been, howevefi not
to propose a division directly u.seful in field work, so much as to erect
a fence ovct which speculation al>out the earth's interior cannot pass.
The lK»st way to check niiscliievous speculation is to advance benefi-
cent spe(*uIation. founded on all the known facts. For example^ the
bold statement that there can be no magmatic sulistratum beneath
a district }K\iriiig two simultaneously active lava columns of dilFering
heights can no longer be made without an investigation of their nature
as *• principal" vents (abyssal injections) or '* subordinate" vents
(satellitic injections). The formal classification is of positive use in
recognizing a mechanism by which the petrographic contrast of the
* .\. ("irikir, (jtiiirt. Jour. Cicol. i^i^. Ix)n<ion. Presidential Address^ VoL 48*
1S92, p. 9o.
> B. X. Poach aii<i J. Hornc, Trans. Roy. Soc. (xlinburgh, Vol. 92, UM^ puSMk
and Vol. 28, 1878, p. 418.
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 301
lavas from neighboring vents may have originated. If two of these are
opened above different satellitic injections, the chances are good that
the magmatic histories of the injections will be different; their emanat-
ing lavas would diverge in chemical type according to the progress
respectively made in the formation of syntectic and differentiated
magmas.
Some "subordinate" vents are monogenetic in Stubel's sense, and
there are a few analogies between the system of vulcanism here sug-
gested and that elaborated by the illustrious German. But the writer
cannot agree with Stubel's principal conclusion as to the motor power
in vulcanism, and entirely fails to find geologic or petrologic evidence
for the existence of his "Panzerdecke."
General Summary
The general hypothesis briefly outlined assumes that the earth is
exteriorly composed of successive shells of density increasing with
depth. Beneath the interrupted sedimentary shell is a continuous
solid "granitic" shell, and still deeper, an eruptible basaltic shell or
substratum. All igneous action, since an early pre-Cambrian period,
is the result of the mechanical intrusion of the substratum basalt
into the overlying shell. This fundamental process is specifically
called ^* abyssal injection J' It is not a hypothetical process, but one
which is clearly apparent in the chemistry and field relations of igneous
rocks. The conditions leading to abyssal injection form a subject of
great theoretical difficulty, but the discovery of the exact mechanism
is not essential to the presented explanation of volcanoes. Nor is it
necessary to decide on the degree of viscosity characterizing the basaltic
substratum, although it is pointed out, once again, that the observed
small amount of deformation of this planet under cosmical stresses
does not prove that the substratum is crystallized. The central thesis
of this chapter is that all vulcanism is a consequence of abyssal injec-
tion, or in other words, that from the date of the oldest known pre-
Camhrian lavas, every volcanic vent has been opened because of a
preliminary mechanical intrusion of molten basalt into the acid earth-
shell.
Emphasis is laid on the absolute necessity of classifying the gases
and vapors which do important work at volcanic vents. These are
either magmatic or phreatic in origin. The magmatic volatile fluids
are subdivided into juvenile and **resurgent'\' the phreaiic fluids into
vadose and connate. Each of these classes may be important in the
dynamics of volcanic explosion; on the other hand, the juvenile mag-
matic gases are assuredly the most important in keeping a volcanic
rent alive.
302 IGNEOUS ROCKS AND THEIR ORIGIN
Central eruptions are of two main classes. In the " principal^' class
each vent represents emanations from a main abyssal injectioii; in the
other, ''subordinate,'' class each vent originates over amagmatic body
(laccolith, sheet, etc.) which is satellitic with respect to a main ab^'SNiI
injection. Of these two cla.sses, "principal" volcanoes must, on the
average, he the more intense in activity, of longer life, more productive
of lava-flows, and more clearly related to crustal fissures. The facts
of the field suggest that Kilauea isa''sul)ordinate" volcano. Terti-
ary and Paleozoic examples are probably represented in Swabia and
Scotland. The localization of central vents and their very common
alignment are explained by the principle of abyssal injection. Lack
of alignment in a group of vents is suggestive of their "subordinate"
origin. In the nature of the case, "suiiordinate" vents must, in their
activities, show a high degree of independents of one another and of
neighboring "principal" vents.
Continued eruption at a central vent is a heat problem. The pri-
mary heat of its abyssal injection is not the only source of thermal
supply. A leading place in the theory should be kept for the supply
<lue to chemical reactions among the primary constituents of the in-
jected magma. Abyssal injection means an enormous change in the
pressure conditions of the magma. As a result, the juvenile gases rise
toward the top of the magma chamber. They are concentrated in
the actual volcanic pipe and, according to the law of mass-ttction.
exothermic reactions on a large scale are to be expected. The pos^i-
l>ility that energy was potentializcd in the primitive basaltic Bubstra-
tum by the formation of dissolved endothermic compounds, like cyan-
ogen, ozone, hydrogf'n peroxide, etc., is indicated. In consequence
of changes in pressure and temperature, due to injection, the disMMria-
tion of those compounds and the formation of new, stable compounds
formed i>artly or wliolly from their elements, give a double source of
heat in the magmatic colunm. The heats of formation for the probaMe
reactions aro so great that small masses of juvenile gas might fumt^h
a relatively large supply of heat. Moreover, it seems likely that
exotherniic reactions oeeur when the liquid components of magma
attain eliernieal equili})riuni uniier the new conditions induced by the
injection of an abyssal niagniatie wedge. Though it is at present
impossible to estimate the fraction (»f the total volcanic heat due to
eheinieal reactions, a working i)hiloso])hy of vulcanism should giv^
ilue regard to the hyi)othe.sis that a central vent is a true furnace.
I'sing SiegPs recent results from experiments, the writer has calcul-
ated the approximate rate of heat loss through radiation at an active
crater. The loss by radiation occurs, in general, at a much faster rate
than the heat loss l)y conduction into the walls of the vent for a depth
MECHANISM OF VOLCANIC VENTS OF THE CENTRAL TYPE 303
of many kilometers. The methods of the transfer of heat from the
depths are discussed.
The principle of ^Uwo-phase convection*^ is concluded to be essential
to the maintenance of prolonged activity at central vents. This con-
ception is illustrated by the analogy of solid spheres moving, under
gravity, in viscous fluids.
Explanation is offered for the dormancy and related periodicity of
certain vents; for the typical shape of such a vent, and for its com-
paratively small size. These features, when coupled with long persist-
ence in activity, are chiefly dependent on two-phase convection.
Since the latter process is, in its turn, dependent on the rise of gases
in the magma chamber, this general conception of central eruptions is
called the gas-fluxing hypothesis-
The petrographic variety of lavas is to be largely explained on prin-
ciples which have been demonstrated in plutonic geology. The lavas
emitted at central vents may be: primary basalt; differentiates of pure
primary' basalt; syntectic magmas, i,e., those produced by the solution
of foreign rock in primary basalt; or differentiates of syntectic mag-
mas. The petrographic diversity in the lavas of neighboring vol-
canoes becomes better understood through the recognition of the two
(*' principal" and '* subordinate") classes of central vents.
The explosiveness of volcanoes is a necessary step in the march of
events following abyssal injection. The inciting cause is to be sought
in the tension of resurgent gas as well as of juvenile gas. Progress in
magmatic differentiation tends to favor explosiveness to an important
degree. Magmatic and phreatic explosions must be distinguished if
the tangle of vulcanological facts is to be unravelled. Though the
rise of hot magma into rocks charged with vadose or connate water
does often cause explosion, the steam-pressure produced by such
volatilized water can no more be regarded as the cause of vulcanism
than is the boiling of a kettle the cause of the heat in the stove. The
formation of the magma column, extending through the earth's
'* granitic" and sedimentary shells to the surface, is the crucial prob-
lem. It is obviously a mere matter of detail whether or not the coun-
try rocks at the upper end of the magmatic column are wet and there-
fore explosive.
The facts of volcanic geology seem, therefore, to co-operate with
the facts of plutonic geology in showing that the essential process in
igneous action on this planet is the rise of basaltic magma from the
universal substratum along abyssal fissures in the earth's acid shell.
CHAPTER XIV
ECLECTIC THEORY OF THE IGNEOUS ROCKS
SUMMAIIY OF THE ECLECTIC ThEORT
Tho genetic scheme outlined in the last five chapters is the result
of an attemi)t to combine the soundest ideas of petrology in a general
working theory. The value of the scheme cannot fully appear until
it is applied in detail to the different rock clans and rock bodies of the
world. B<'fore entering on such concrete studies it will be expedient
to summarize the proposed theory.
1. It is practically certain that the earth is stratified, at least
roughly, according to density. This is the conclusion of cosmogonist^,
like I^iplace, I^^gendre, Koche, and others; of mathematical ph>*sii-
cists, like (i. II. Darwin and Fisher; of seismologists, like Oldham;
of gi'ologists, like A. (leikie and many others. The deeper layers of
the globe are not known ever to have furnished material for the visible
lK)dies of igneous rocks.
2. To explain the igneous rocks it is only necessary to deal with
three ext(»rior earth-shells. The thin sedimentary^ and the thicker
acid, crystallinv shell iM^neath are both visible in part and may be
reasonably infiTred to extend much more widely than their respective
total outcrops. The sedimentary shell is known to be discontinuous:
the underlying acid >hell is known to form most of the area of each con-
tinental plateau, but it may not underlie all of each ocean basin.
From the ])henomena of igneous action the existence of a basic shell,
still de(>{MT, is inferred. It may be discontinuous; if so, ita part5
underlie every large area of the eartlfs surface, continental or oceanic.
It may be s(>li<i, though preference is given to the view that it is fluid,
perhaps a very rigi<l fluid. In either case it is known to be compowii
of eruptible matter. Many facts compel l>elief that this third shell
is basaltic in composition. In accordance with the ascertained fact5
of p(*trology and geology, the txisaltic substratum is held to be the
only shell of the earth which, since an early pre-Cambrian period
(,tyi)ifie(l in the Keewatin), has been hot enough for spontaneous enip*
tion. Probable as this relation may be, it is to be clearly recogniied
as an assumption, the )»a>al assumption of the whole theory to l>e
outlined. Such is the writer's conce])tion of the exterior shells 61 the
earth, as deduced from the facts of modem geology. Such amiears
304
ECLECTIC THEORY OF THE IGNEOUS ROCKS 305
to have been the general conception of von Cotta who generalized from
the knowledge of the earth gained up to the year 1858.
3. It is known that the basaltic magma so abundantly erupted
since the Keewatin period cannot be due to the liquefaction of either
the sedimentary shell or the acid shell of the earth (together called
the "crust")- Hence, basaltic eruption implies the injection of ba-
saltic magma along abyssal fissures in the crust. Each abyssal
fissure must usually grow narrower toward the top, and hence the
primary basalt filling it is called an abyssal magrtiatic wedge. The
mechanism of this injection can be discussed only hypothetically.
The writer's general conception of it has been independently reached
by Johnston-Lavis, and Fisher has deduced some of its essential
elements.*
4. Abyssal injection involves the production of some superheat in
the basalt raised to, or nearly to, the earth's surface. If the magmatic
wedge is of large size, a limited amount of marginal assimilation of
the wall rock is a necessary consequence of the injection. This deduc-
tion from observed fact agrees, in principle, with the views of many of
the advocates of the hypothesis of marginal assimilation in depth,
including von Cotta, de Beaumont, Fouqu^, Michel L6vy, Lacroix,
Barrois, Loewinson-Lessing, Brogger, Lawson, and many others.
5. The larger basaltic wedges perform magmatic sloping to a
variable, but often great, extent. The process has been described by
Goodchild, Lawson, Barrell, Ussing, and the writer. However, the
four authors first mentioned do not associate the process with primary
basalt.
6. Stoping involves abyssal assimilation of the country rocks.
This consequence of stoping has been specially emphasized by the
writer and (as a recent letter from Professor Cole shows) by Goodchild.
7. Both the primary basalt and each of its solutions with crust-
rock is, under certain conditions, subject to magmaiic differentiation.
This principle is now accepted by practically all workers in modern
petrology, though many of them deny the importance of syntectics
and are content to regard almost all igneous rocks as derived from
either primary magmas or from liquid differentiates of these.
8. The essential facts known concerning batholiths, injected bodies,
and volcanic vents appear to find explanation on the hypothesis of the
primary basaltic wedge. The batholith is here interpreted as an abyssal
wedge or gigantic dike which has been physically enlarged and chem-
ically modified at its summit.
The petrogenic scheme outlined is thus seen to be an eclectic theory.
Not merely for the interest of an academic study pursued in the
^ O. Fisher, Physics of the Earth's Crust, London, 2nd edition, 1891.
300 W.S'EOCS ROCKS A\D THEIR OHJGIS
lilirary, but becjiiisc of tho ijn^ssurc of facts encountered in the field,
tlic writcT lijis *\sol('ct(»(l and ai)i)n)priat(Ml'' whatever seems best in all
tlio earlier theories of the igneous rorks. He has found reason for
helieviiiK in (*ss«*ntial tenets of the Freneh petroh)Kists and in tha-v of
the (ifTUian an<l other pet rolojjjists who refused to follow de Beaumont
and his sueeessors. Of late years the national l>oundaries, so curi-
ously p<Tsistent between the o])]>osing schools of petrogenic thought,
have been breaking <lown. Mi<*hel Levy an<l Lacroix have retained
their Ix'lief in assimilation while learning to grant the great importance
of ditTerentiation. On the other hand, an increasing numl)er of
(S(Tn)an. Austrian, Scandinavian, Kritish, and American petrologists
are heaving the extreme position of the pure diiTerentiationistj«. hi
h>ng led by Kosenbusch. Hrogger, Teall, Iddings. Pirsson, Washington.
ami others. Th<'s<' t<*ndencies of themselves suggest the justice of the
ecle<-tic position.
It is significant that some of those petrologists who still adhere Xu
pure ditTerentiation as tin* only important cause of the diversity of ign«-
ous rocks, liave had <'omparatively little experi<'nce with igneous
rocks in Ow fit hi. < )n the other hand, steadv contact of other workers
with a siK'cession <»f large ignrous intrusi<»ns luis often, if not generally.
h'd them to st'c tlie ne<'('><ity of large-scale assimilation. Among
the writers of such ripe fifld expc-rience are Barrois, Michel ly*vy.
Kjerulf. Scdrrholm. Ilogbom. ]]. (\ Andrews. Kmerson, Coleman,
Lawson, Hnx-k. ('am>ell. and many others.
The writ IT c.anu" to brlirf in a-similation as a iH'trogenie factor of
first rank thnnigh two >t;igr<* of reasoning. His early studies on th«-
batholiths of the " porphyritii' granite" of New Hampshire, on thr
^tock< of Mount AMiitncy. \'ermont. and on the great hathoUth at
Marre. \'ermout. Ird him to conclude that at l<>ast some of the hatho-
lithiv magma< had •fturpn'titttl thrir rtiuntry rocks. The al>sence of
('hrmi<-al "consanguinity*' i)('tw<-tn magma and invaded formation in
prai'ti*'allv rvrrv va^r <«'eme«l. liowrvtr. ti» f<»rbid the notion that
iiu't>rporatit»n nUo mi-ant -»ohiti«»ii of tin- country rock. The difficulty
wa** rfnh>vrd tliroimh tijf dt'Vrlopmiut nf the st4>ping hypothesis at
Mount A^rutnrv. Tliat h\ ]>ot!i, -i«* not onlv involved the neee#ciitv
of al'Ns'^al as'-imilaiion; it al-^o tAj>Iainrd the general al>sence of
■■ tran-^grr^-ivr jun* tioii--" bitwitu I-atliolith an*! wall-rock and ac-
I'ouiitcil for t!:»' i:» nrral faiiurr i^i tlnir ciitniiral consanguinity. Inci-
iir!itall\. iiir .ir\t iopmriii K^i tl:at liypothi«*i<. while involving the
ol'vi*v.i>; prim ;:•!«• of inac'iiaiir -ii:Vj rfii! iaii«>n. also showtMl the fallacy
of ti'.e prr\ ailiTi;: ari::;!iit u'-- ac^iiii-^t marginal :is>iniilation in plutonic
luactna^. Sjiv i ;>;;■• :-!.:iii: i.> tir^t out lint* ^^i the hypothesis, the
writiT lia< stUviif.l in i:rratrr or it «h< dt'tail more than two score of
ECLECTIC THEORY OF THE IGNEOUS ROCKS 307
stocks and batholiths in British Columbia, Idaho, Montana, Cali-
fornia, New England, and Eastern Canada. In every case the general
field conditions affecting the hypothesis were found to be similar to
those at Mount Ascutney and of the other localities where the results
of actual field work had prompted this conception of intrusive mechan-
ism. With the strengthening of its proofs the writer's belief in the
great efficiency of assimilation has strengthened. He is not prepared
to state the relative importance of the assimilation at main contacts
as compared with that of stoped blocks in depth, although, as indicated
in Chapter XI, the latter type has certain plutonic conditions in its
special favor.
Loewinson-Lessing's General Theory
On referring to the literature of petrology, the writer found that
Loewinson-Lessing had, on independent grounds, also arrived at an
eclectic theory, embodying both doctrines of differentiation and as-
similation. Loewinson-Lessing called his petrogenic scheme a "syn-
tectic-liquation theory of differentiation.''^ Since it is the only
published, systematically developed scheme at all comparable to
that sketched in this book, it will be discussed in some detail. A
summary of Loewinson-Lessing's position may be given in his own
words, occurring in a recent paper.
"L The method hitherto adopted of calculating the average compositioi]
of the terrestrial magma must be considered as erroneous in principle.
'*2. Nevertheless these calculations give a fairly good result, as it corre-
sponds approximately to the mean between gabbro and granite, if we admit
that these two magmas enter into the composition of the external part of the
earth's crust nearly in equal quantities.
"3. Two original independent magmas exist which predominate in the
composition of the earth's crust, the granitic and the gabbroidal (basaltic);
all the other igneous rocks are derivates from these two and are subordinate
to tJiem in their occurrence.
"4. Differentiation is produced in two ways: during the crystallization,
differentiation by crystallization; and before crystallization, in the liquid
magma — magmatic differentiation.
"5. The differentiation b} crystallization consists in the sinking or rising
of the newly formed minerals according to their specific gravity, and in the
solution in one part of the magma (generally a deeper-seated one) of minerals
formed in another part.
"6. Magmatic differentiation consists in the formation of derived magmas
( Spallungen) y and is governed by the tendency to form eutectic and mono-
» Compte Rendu, Congrds Gdol. Internal., 7e Session, St. Petersburg, 1899,
p. .379.
308 I0NE0U8 ROCKS AND THEIR ORIGIN
mineral (or bimincral) magmas. ThiH differentiation is induced by the fusion
and assimilation of foreign mineral masses, both igneous and sedimeiiiary.
"7. Differentiation finds its l)est explanation in the Bjmtectic-Uquational
hypothesis (fusion, assimilation, differentiation).
"8. All igneous rocks belong to three tsrpes: (1) primordial magmas,
(2) rocks due to differentiation, (3) rocks produced by minting of two
magmas.
"9. The igneous rocks of all geological periods, presumably from the
Archaean, originated principally by the refusion of different parta of the cmrth*$
crust. On account of this wc meet in successive periods always the same typ»
of rocks. The pre-Archsan igneous rocks (perhaps also a part of those of
the Archffian and younger periods) were formed from primordial magmatic
masses of granitic and gabbroidal composition."^
The starting-point of Loewim^on-Lessing's argument is the assump-
tion of two different magmas, independently eniptible during the whole
of recorded geological time. "The mutual relations of the eruptive
rocks may be explained most satisfactorily by the admission of two
primordial magmas." In this he arrives at the same concluaionfl as
that previously reached by Bunson in 1851 and by Michel Uvy in
1897.
On page 169 of this book we have seen numerical reason for the
approximate identity of the arithmetic mean of all the individual
rock analyses ^ith the mean composition of average granite and the
average basalt (or gabbro). This identity is certainly referable to
the predominance of these two typos in the world's igneous tcrranes.
Yet it cannot, of itself, prove the existence of independently eruptiMe
granite magma during the post-Keewatin period of the earth*a history.
The identity of the two means can equally well be explained on the
assumption that, during that long interval, the basaltic magma alone
has been eruptible because of intrinsic high temperature.
Though several very extensive masses of anorthosite and a few
others of gabbroid nature are on record, it is not certain that a single
one of them is a true, 'M)ottomless** batholith. Most of them may
be merely injected or laccolithic bodies. In any case the relative
rarity of large basic batholiths is an assured fact. This feature of
the world-map is hard to explain if, as assumeii by Loewinson-Lessing,
the two primordial magmas co-<'xist in the earth's ''crust'* in about
equal volumes, unless we also assume that the granitic rocks dominate
in the earth's exterior shell and have l)een generally re-fused in the
batholithic ba««alt. We thus approach one of the leading condosions
on which the prest'nt writer's theory has In^en formu ( •
Loewinson-Ix'ssing has not made clrar the way
> F. LoewinsoD-Lcssing, Gcol. Mag., Vol. 8. """' f-
ECLECTIC THEORY OF THE IGNEOUS ROCKS 309
m^rdial magmas are arranged within the earth, but he does state that
each is due to the re-fusion of parts of the earth's solid crust.
'* Different geological data and theoretical considerations lead to the
conclusion that the identity of the eruptive rocks of all geological periods
can be most satisfactorily explained by the assumption that from the Archsean
up to the present the eruptive rocks represent nearly the same material,
which has been subjected several times to weathering, metamorphism, refu-
sion, and regeneration. The assumption that we find or can find anjrwhere
the primordial solid crust of the globe must be definitely abandoned. The
occurrence of clastic and sedimentary rocks in the oldest Archsean formations
and numerous examples of refused and recrystallized rocks in these forma-
tions eloquently sustain and complete the theoretical considerations which
lead us to the conception that the primordial crust has been re-melted long
ago, and probably more than once. Such a fusion of parts of the solid crust
is sometimes directly attributable to a rising of the isogeotherms at the places
in question. In reality the process may be a more complicated one, as can
be shown by the following considerations. A series of sedimentary rocks,
30,000 feet thick (the greatest thickness we can admit) would simply by the
rising of the isogeotherms acquire in the lowest beds a temperature of 300® C,
which is quite insufficient for melting these materials. But the geosyncline,
where sedimentation of our 30,000 feet of sediments has taken place, may itself
consist of an older series of sedimentary material. At a depth of 60,000 feet
under the bottom of our geosynchnal the temperature would be 600® C.
before sedimentation, and would rise to 900° C. after the deposition of 30,000
feet of sediments. Under the weight of this new sedimentary sheet of 30,000
feet the area in question would probably bend and subside; this would give
for a presumable subsidence of 10,000 feet a furthfer increase of 100®, and so
the primarj' temperature would in this way rise from 600® to 1,000® C. All
these figures are, of course, hypothetical. But we must bear in mind that
epeirogenetic and orogenetic movements can occasion a far greater subsidence
of parts of the crust than the process of sedimentation by itself. It is also not
to be forgotten that in these great depths the magma is probably rich in water
and different gases, and consequently fusible at a lower temperature than the
*dry* magmas generally considered. And, lastly, we must also infer with
Suess that the fusion of certain parts of the earth's crust may be produced
partly by the rising from below of hot plutonic gases. In short, there are
sufficient factors for sustaining the hypothesis that in successive geological
periods different parts of the soUd crust have been caused to melt, and that
by this process have been generated the plutonic and volcanic rocks.
" When such a refusion or 'anatexis,' as it is called by Sederholm, embraces
a portion of the crust, which consists of definite eruptive rocks, e.g,f granite,
gabbro, basalt, the resulting magma will again be after consolidation the same
rock, perhaps only slightly modified by the assimilation of other material
during the pj age of the magma to the place where it consolidates. But
iHm ilM fem ^ted portion of the crust is composed of different rocks, eruptive
->? both together, the process is rather a 'syntexis,' as I have
310 IGNEOUS ROCKS ASD THEIR ORIGIS
called it, an a>siniilatiun which is followed by liquation and differentimtion;
the same process of 'syntexis' or assimilation must take place at the margins
of such renieltcd jx)rti(>rs where they arc in contact with non-melted portions.
Althouf>h I am an advocate of the hyi>othehis of a fluid nucleus, I do not
U'lii've that this nucleus would have l>een (|MTha|)8 with a few except ions)
the source of the Archa'an and post-Archa»an intrusive iKnlies and superficial
volcanic masses; hut certainly it mu^t l>e admitted that different portions of
the solid or anatectic crust can l>o mixed during the process of refu>ion with
fluid material coming from ])eri])heric magmatic hasins/*'
From the relative emphasis noted in this passage it is fair to sup-
p()s(> that the author is most inclined to explain local re-fusion of the
solid crust by sedimentary blanketing. However, his caleulat ion itself
shows the straits into which the advocates of this time-honored sug-
gestion are driven. It is needless to rehearse all the obvious geological
facts which show sedimentary blanketing to be merely a neeondary
c'ause of the heat in (Tuptivo magma. The hypothesis entirely fail-
to account for the vast numl>er of liquid ma*«sc»s which have been
erupt<'d in regions of little or no sedimentation. Kxamplesat first hand
nmy be .*<een in the fissure eruj)ti()ns of the Deccan, the long voleanic
chain of Kast Africa, the vulcanism of central France, and that of
(Greenland. The long chains of enormous volcanoes built on the floor
of the Pacific basin can hardly represent fusion under geosynelinal
prisms of notable thickness. In many cases geosynclinal down-
warping may hi(jiu with magnuitic intrusion (Sec* Table IX, page 186.«.
so often, indeed, that it seems likely that subterranean movement of
nuigiiia is largely responsible for the formation of geosynelinals.
The ajuxal to fluxing by plutonic gases is little safer. It must Im*
remembered that a gas, n^aching a r(*gion of lesseneti pressure. i.<*
bdund to ex])and and cool. (Calculation shows that to supply only the
hiit'iit heat recpiired to melt KK) cubic miles of cry.st alii zed basalt
>\(»uld demand the locali/ati(m of a >tU|><>ndous quantity of juvenile
^•a- uiuh-r great j)res>ur<». If the magma so prepared were to break
ihiuu^li to the earth's surface, the accompanying gas prc^*«ure mu^st
|ii\iduce e\ph>sions boide which that at Krakatoa would l)e in.^ignifi-
i.ihi \e! nM»>t u{ the world's basalt has Ih'cu extruded quietly.
h Mciii^ nt( essary to conclude that the imagined meehanism for
\\\\ "Upplx Ml niagniatic heat is in>uf!icient for the requirements of
twMtiMu- ^cMln^y; yet the ])etrogeinc firoblem is, at bottom, a beat
I wn k^iMhtini: tin- j)os>'ibility of loral r(»-fu.<ion8 of flolid
\\u\ HI imlr .jMihajKs with the aid of *^ i* l"»«t evolvi
^^\ UN ^ \'»» Hie Male*h'inan(i "
* \ I «>\*t\iitmtit t r«'Niiiiii;, C
ECLECTIC THEORY OF THE IGNEOUS ROCKS 311
rocks and the observed sequences in their eruption cannot be ade-
quately explained. We have here a major difficulty like that faced by
the theory of differentiation as the sole explanation of igneous rocks.
No one has yet conceived the arrangement of local reservoirs which
could erupt, with repeated alternations and in one small region,
magmas belonging to the granite and gabbro (basalt) clans. Yet
such magmatic successions have often been recorded. If a reservoir
of molten basalt underlies one of molten granite, it is inconceivable
how the pure basalt could reach the earth's surface. The same ele-
mentary difficulty applies if the order of the reservoirs was reversed;
and we are in about as much trouble if they are side by side, though
separated by a necessarily thin partition. This argument is per-
fectly general. It applies even to the case of the Keewatin and other
early pre-Cambrian basalts. Is there any escape from the conclusion
that these basic extrusives traversed a solid, fissile earth's crust?
The theory outlined in this book is thus believed to have some
advantages over that proposed by Loewinson-Lessing, whose notable
and inspiring statement represents the only other published attempt
to form a complete explanation of the diversity of igneous rocks.
Von Cotta's early suggestion, that a basaltic substratum includes all the
independently eruptible matter in the earth, has the merit of simplicity.
One can imagine the mechanism. Its working can be deductively
studied and inductively tested. It is economical in its demands on
terrestrial heat. No further theory can dispense with at least one
assumption as fundamental. No future theory can stand unless the
deductions from its basal postulates are likewise matched by the
observable facts.
Genetic Classification of Igneous Rocks
Before passing to a further testing of the writer's eclectic theory, it
will be well to summarize it once again in terms of the origin of the
individual clans of igneous rocks. The following table names those
clans whose genesis has been stated or implied in the preceding pages;
other clans are credited with origins which are not directly implied
by the basal premises of the theory. After the reader has mastered
the facts recorded in the next eight chapters, he may be better able to
tolerate the boldness of attempting a genetic classification of igneous
rocks even in this day of confessedly limited knowledge of the earth.
lONEOUS ROCKS AND THEIR ORIOIN
MAGMAS MAY BE CLASSinED AS:
1. Primary bualtic. .
2. Primary granitic. .
3. Direct diffcrcntiatw uf primary ba-
■alt.
4. Syntcctics, three cUhmm: —
A. Chiefly compoaed of basalt and
earth's acid shell.
B. Chiefly composed of basalt and
sediments.
C. Composed of basalt, sediments,
and arid shell.
6. DilTerentiatcs of syntcctics of class A.
6. Differentiates of syntecticB of class B.
7. Differentiates of syntcctics of class C.
I
8. Mixtures of two or more uf above-
mentioned types of liquvl^.
9. Tranaition maicma markinft incom-
plete differentiation.
ReprMenUtiT» roelf
Gabbro clan in guMnL
? Perhapa only rcprwsntod is tbc
"an«t«ctic" (refuaed) form of Laufeit-
tian and other pro^^ambrian batho-
litfaa.
Pyroxene andesite, anorthoaite: ecftain
members of the peridotit« elan; mttot
iron ores and sulphides.
Some members of the dioril« claa.
Rare hybrid typea.
Rare hybrid types.
Most rocks of the pwiite elan. Some
rocks due to gaeeoue tnaMtr-
Abnormal Kranitea; moat of the Bcphp*
lite and leueite rocks; bodm eertmd-
iferous typea; etc. &Iany rocka due
to gaseous transfer.
Granodiorite elan for the moat part;
some members of the syenite elan; etc.
Many rocks due to gaaeoua tiaa^o'.
flome hybrid rocks?
Many "intermediate" typea.
PART III
CHAPTER XV
GABBRO CLAN
Included Species
he gabbro clan includes the gabbro family, the gabbro-por-
ite family ; and the family of the basalts, melaphyres and diabases,
rding to the last (1908) edition of Rosenbusch's ''Mikroskopische
?iographie der Massigen Gesteine," these families are made up of
» than 50 types. The species may be listed as follows:
nal Types:
nrant Types:
Plvionic Types
Gabbros,
Olivine gabbro, some hyperites,
Olivine-free gabbro.
Norites,
Olivine norite, some hyperites.
Olivine-free norite, labradorite norite,
bronzite gabbro.
Anorthosites,
Anorthosite proper.
Olivine anorthosite.
Quartz gabbro, quartz norite.
Hornblende gabbro.
Mica gabbro.
Orthoclase gabbro, perthitophyre, gabbro-syenite
Troctolite, krageroite.
Oligoclasite.
Andesinfels.
Kyschtymite.
Dike Types
Gabbro porphyrite, microgabbro
Some augite porphyrites.
313
314 JONBOUS ROCKS AND THStB OBMtN
EffuMive Types
Normal Types:
Basalts,
Olivine basalt.
Olivine-free basalt.
Hypers thone basalt.
Enstatite basalt.
Hyalobasalt.
Tachylite.
Palagonit«.
Schalstein.
Melaphyres,
Olivine ntrlaphyrea, olivine tboldite
Tholeiite.
Diabases,
Olivine diabase.
OIivinc>free diabases.
Palatinite.
Ix'ucophyte.
Kpidiorite.
Proterobase. ■
Salitc dial)ase.
Hunne diabase.
Ophite.
DiAba»e porpbyrite.
Spilite.
Hyalodiabase.
Variolite.
Kinnc diabase.
Hollefors diabase.
Aasby diabase.
Siirna diabase.
Ottfjall dial>ase.
Abi-rratil Typrit:
Iruii hitMilt.
Qiinrtz husalr.
Quartz ilialiH.-u-.
Konga <lialia.-ic.
Mere difTcrcnoes in Rt'ulufili'al a^o ur minor differcneci !■ rack
structure, in dt-frrre of eryftnlliuity, and in -'— ee "i[ alteration bj
wfatlwrinn or iiu'tanitirphi:?ni, have- prompted
loast half thp names in this long li^'
1^
GABBRO CLAN 315
we shall lay especial emphasis on the important chemical contrasts
and in a few cases on mineralogical peculiarities.
Primary Basaltic Magma
On reference to Cols. 51-62 of Table II, it will be seen that the
respective average analyses of basalt, diabase, olivine diabase, mela-
phyre, gabbro, olivine gabbro, norite, and olivine norite are very
similar. Average extrusive basalt is slightly higher in silica, soda,
and potash, and lower in magnesia and lime, than average gabbre.
These contrasts are analogous to those generally seen between a plu-
tonic type and the corresponding extrusive type. In fact, the many
basalts actually averaged include some types verging on augite andesite,
which seems to be best interpreted as a differentiate of basalt. (See
Chapter XVII.) In order to estimate more nearly the composition
of primitive basalt, the writer has calculated the average of the 198
extrusives of Col. 53, Table 11, and the 17 olivine gabbros of Col. 60.
Even after the inclusion of these gabbros, the resulting average may
.^till show a slight ejccess of the alkalies above the proportions char-
acterizing the quite undifferentiated basaltic magma.
The computed average is as follows:
Water-free
SiO,
48.84
49.65
TiO,
1.35
1.37
A1,0,
15.90
16.16
Fe,0.
5.23
5.31
FeO
6.30
6.40
MnO
.29
.29
MgO
6.38
6.48
CaO
9,16
9.30
Nb,0
3.05
3.10
K,0
1,46
1.48
H,0
1.60
P.0,
.45
.46
100.00 100.00
The calculated composition of the primary basalt will be used
in the discussion of the other igneous clans as well as of certain types
of the gabbro clan. Deviations in the percentages for individual
oziites from the true values for the substratum magma must be
^relatively air ,d they are not likely to affect the foUowii^ applica-
B||MM^|ta| principles.
316 laSEOVS ROCKS AM) THEIR ORIGIS
Normal Olivine-free Species
A moderate amount of olivine would normally crystalliEC from a
magma of the composition above stat(>d, giving olivine-bcaring basalt,
diabase, and gabbro. Slight gravitative differentiatioD, whereby the
olivine substance (with other f(>mic con^stitucnttf) settles in the mag-
matic column, must tend to produce a corresponding olivine-free
basalt, diabase, or gabbro in the upper part of each column, and a
corresponding olivin(»-rich pole in the lower part. There arc good
grounds for iK'lic^ving that this simple meehanitsm has been partly
responsible for the range of mineralogical (and chemical) composition
found in the gabbro clan.
Lewis has well illustrated the case in his study of the intrusive
Palisade sheet of New Jersev in which olivine-free dialmse overlies
oli vine-rich diabase J I)u Toit has found a similar relation between
olivine-free gabbro and olivine gabbro in the great Insizwa intrufiive
sheet of Kast (Iriqualand.^
It is highly probable that simpl(> gravity' has caused the simul-
taneous dev(>lopment of olivine-free basalt and oli\nne-rich basalt in
volcanic vents.' The problem of origins is here evidently much more
difficult of solution than it is in the case of well-exposed intrusive sheets;
yet the constantly recurring association of both magmatic tJTX*s in
the basaltic fields clearly implies incipient differentiation of a primitive
homogeneous magma in each field.
Automatic changes in bodies of the primary basalt are competent
to explain these half dozen species in the gabbro clan.
A brief discussion of some other tyiMS in this clan is especially
demanded since the j)rocesses of their formati<m directly concern the
theory of the other nxk <'lans, as more fully stated in the following
chapters.
Qi ARIZ Diabases and Tiikir .\llie8
At intervals during the last few years several authors have empha-
siz<'<l the wi<lesj)n*a<l distributi<»n of rocks which are chemically near
typical basalt, an<l yet contain more or less free quartz along with
minerals characteristii* of basalt (»r its intrusive equivalents. These
tyiK's include the (piartz diabases, quartz dolerites, quartz gabbros,
some (|uariz-bearing jjorphyrites, and cpiartz ba^^lts. Well known
examples are: the Konga diabase of Fc^nnoscandia; the Svir diabases
' J. V. Lo\\if», Ann H'port, State (wMiloaist of Now Jersey, for 1907, p. 131-
' A. J. «lij Toit. l")tli Aim. K«p., (Irol. C'omin, (*npenf Good Hope, 1910, p. 111.
» H. A. Diily, Jour. Geoloto', Vol. 19, 1911. p. 305.
GABBRO CLAN 317
of Russian Karelia; the quartz diabases of many areas in the British
Islands; the quartz dolerites of the Karroo and other districts of South
Africa; the quartz diabases, quartz porphyrites, and quartz basalts
of California; the quartz diabases and quartz dolerites of Antarctica,
Central Australia, and Western Australia; the quartz diabases of Cutch
and Southern India, of Northern Siberia, of the Congo Free State and
British Guiana; the quartz diabases of Arran, Greenland, Sumatra, and
other large islands; the quartz diabases and quartz gabbros of
British Columbia, Minnesota, Michigan, Ontario, and Quebec; etc.^
Other types of similar, or but slightly different, magmatic character
are: quartz norite, some orthoclase gabbros, and some mica gabbros.
These rocks are unlike normal basalt, diabase, or gabbro, in containing
free silica; many of them have a decided tendency to crystallize with
micropegmatite interstitially developed. Some authorities have
attributed certain of the types to the slight acidification of ordinary
basaltic (gabbroid) magma through the solution of silicious country
rocks. The present wTiter believes that this hypothesis may be
fruitfully applied to all of the rocks above listed.
The dominant hypothesis has long held that the quartz diabases
and their magmatic allies have crystallized directly from a primeval
magma. Recently, Wahl and Thomson have stated their adherence
to this view. 2 Their evidence is, however, not conclusive and there
are several strong grounds for belief in a secondary origin for the whole
group of species.
It is noteworthy that the quartz diabases and their allies are not
kno\\Ti in the deep-sea islands, that is, in the multitude of volcanoes
which have been built up in areas outside the continental plateaus or
the more or less submerged parts of those plateaus. In the same,
truly oceanic region of the globe there is likewise an entire absence of
outcrops of acid sediments, acid crystalline schists, and granitic ter-
ranes, although normal basalt is represented in most of the oceanic
volcanoes. These contrasts of distribution are difiicult to explain if
the quartz-diabase magma is an independent, primitive constituent
of the earth.
Secondly, these quartz-bearing rocks are almost always in intimate
field association with normal basaltic, diabasic, or gabbroid rocks.
The relation is often one of gradual transition, or otherwise such as to
render the hypothesis of mutual independence thoroughly improbable;
* See the many references given in Zirkers Lehrbuch der Petrographie; in Rosen-
buflch's Mikroskopische Physiographie der Massigen Gesteine; in the papers by
Tyrrell and Thomson noted below; and in the next chapter.
* W. Wahl, Fennia, Vol. 24, No. 3, 1908, p. 69; J. A. Thomson, Froc.Roy.
Soc., New South Wales, Vol. 45, 1912, pp. 311-315.
318 IGNEOUS ROCKS AND THEIR ORIGIN
the mechanism by which the two primitive magmas could be kept
separate until the time of actual eruption has never been described.
Thirdly, not only are all the known bodies of the quarts diaba^H*:!
and their allies erupted through or into terranes more silicious than
primary basalt; many, if not most, of the intrusives are of compara-
tively largo size. This fact is explicable and expected on the syntectir
theory; it fails of explanation on the alternative, ohler view.*
But, finally, the compelling argument is to be found in an analysis
of those cases where large lx)dies of micropegmatitic granite have
clearly l)een formed by the solution of acid terranes in normal basaltic
magma. This subject is di.scussed at length in the next chapter.
where it will be seen that many quartz diabases, quartz gabbros, quartz
dolerites, etc., are transitional into secondary' granites which an*
chemically and mineralogically con.sanguineous with their respective
count rv rocks.
The solution of the acid wall-rock, forming a syntectic of the quartz-
diabase tj'pe, may tak(» place during the steady or intermittent rix-
of the substratum material in the act of eruption, or, in part, it may
take place after the magma has found its final position in the earth's
crust, as sill, laccolith, or dike. There is reason to believe that thf
former method of acidification has often operated so that it is impossible
to test the measure of consanguinity between syntectic and countr>*
ro<'ks, which must in most cases be them.selvesof varied composition.
Xcvertheless. cases favorable for this test are alrea<ly known in num-
ber sufTieient to warrant iM'lief in a syntectic origin for quartz diaha.<«*-i
and their chemical equivah'nts.
XORITES
Rosenbusch points out that the norites and olivine norites are
generally found only in close geological asscK'iation unth gabbro-^:
that the norites are transitional into gabbros and olivine norite on the
one hantl, and into niiea-hypersthene diorite and quart z-mica-hyp«-r-
sthene diorite on the other; and that the noriti»s are somewhat more
acid than the gabi>n»s.' ( 'oliimn 51 of Table II shows that the average
of the available analyses <>f norite (calculated as water-free) has about
1 per cent. mon» silica than average ** primary ba-salt." Quarts anil
biotite, or hornblende, ar(» conunon accessories and at Sudbury, in ihr
Insizwa Mountains, ('aj)e Province (see page 232), and at other well
known localities, norite j)asses into quart zose or granitic types.
» Cf. R. A. D.aly, Arnor. Jour. Scicnro, Vtil. 20. 1905, p. 215; G. W. Tyrrrll.
i;iM.l. Mar., Vol. 0, unni p. :Ui3.
• \l. Ho.sonhiisoh. Mikroskopwrlio I'hyftioKraphio Her MawiKen Gestciiie. 4th
ihJ., 1907, p. 34s.
GABBRO CLAN 319
Prior has suggested that the norites or coarse-grained enstatite-
dolerites of the Umqueme Range, Zululand, "might very well be more
deeply seated rocks derived from the same magma which supplies
the dolerit€s'' of the region. In the next chapter (see page 350) will
be found a statement of the probability that these dolerites have ab-
sorbed the invaded sandstones on a comparatively large scale. ^ From
the facts in hand we may fairly entertain the hypothesis that most
of the norites have crystallized from basaltic magma which has dis-
solved minute proportions of acid country rock. This explanation
will not suffice, however, for certain so-called norites very low in alka-
lies or otherwise peculiar. In both chemical and quantitative mineral-
ogical composition these types must have originated differently from
the average norite.
Hypersthene Basalts and Enstatite Diabases
These types are respectively connected with normal basalt and dia-
base by transitional, often oli vine-bearing, rocks. In part they clearly
represent differentiates of primary basaltic magma, not essentially
affected by the solution of foreign material, and thus mark the way to
the pyroxene andesites. But in part they seem to match the theoret-
ical deduction that acidification of primary basalt should cause the
generation of orthorhombic pyroxene in place of olivine. Rosenbusch
has listed some of the occurrences of hypersthene-bearing and ensta-
tite-bearing diabases which also carry granophyric (micropegmatite)
material. These rocks are thus true quartz diabases or else merge
into them. So far as recorded, the field relations justify the hypothesis
that these particular hypersthenic and enstatitic species, like their
near relatives the acid norites, illustrate syntectics of basaltic magma
and acid country rocks.
Hornblende Gabbros
This group is not largely represented in igneous terranes, but it
offers a special problem which arises also in a detailed genetic study of
some other members of the gabbro clan, e.gf., certain norites. It is char-
acteristic of these rocks that the alkalies are decidedly less abundant
than in primary basalt, while other oxides are present in nearly the
same proportions as in the basalt. The writer will hazard a suggestion
as to the genesis of one type of the hornblende gabbros, a type which he
has studied in some detail.
The dominant rock in the Purcell sills of British Columbia is one of
these abnop gabbros, with composition as follows (Col. i) :
« O. T. • . Natal Museum, Vol. 2, 1910, p. 147.
320 IGNEOUS ROCKS AND THEIR ORIGIN
I
2
SiO,
52 94
48 . 81
TiO,
73
1 . 35
A1,0,
14 22
15 90
Fo,0,
2. OS
5 23
•
FoO
8.11
0 30
Hornblende.
MnO
.35
.29
Labradorite.
MiK)
6 99
0 38
Quartz —
CaO
10 92
9.15
Titanile...
\a,0
1.40
3.05
Magnetite. .
K,0
.49
1.46
Chlorite.
n,o-
.12
'■'■ 1.60
11.04-
1 50
PaO*
OS
.45
Mode of t
Percent.
54.8
25.6
6.3
2.0
3
11.0
100 0
99 99 100 00
Column 2 gives the composition of the theoretical "primary
l)asalt '* of the substratum. The sills ami their feeding dikes cut «ilic-
ious sediments of enormous thickness. As noted in the next chapter.
the basic magma has absorbed some of the quartzitic country rock.
It must, therefore, have ab.^orl)ed water and other volatile Hubstances
originally enclosed in the sediments. Many hatbolitbic contacts
show that magmatic water has the power of transferring part of the
alkalies out of the magma chambtT. Granting that this has occurred
to a considerable extent in the Purcell magma, its chemical contrast
\nth, and derivation from, the primary basalt can Ik? understood. The
specially great abun<lance of microperthite in the thick homfelses at
the roofs of these .<ills is in j)art exi>licable on this hypothesis of gaseou.^
transfer, though in fiart the alkaline feldspar is indigenous to these
quart zites. The presence of considerable water in the magma is sug-
gested In' the presence of hornblende it.self, the cr>''stallization of
which, as experinn'iits show, seems to demand that water lye in the
solid amphibol<> solution.^ Excepting the silica, the residue of the
(primary) ba.^altic magina has the proper composition after subtraction
is mtulv of the fel<lsi)ar molecules to the required amount. The defi-
ciency of silica in the residue has. on the initial hypothesis, l>een supplier 1
by a slight assimilation of the invach'd quartzite. (See Fig. 158.)
This explanation is not contra venetl by the facts now known about
the Purcell rocks, but it is advi.^^edly described as a preliminary gueM at
the solution of a difficult question. It has, Iniwever. the value of facil-
itating belief in a secondary <lerivation of the abnormal gahhro from
the normal ba.<altic magma. In fact. Schofif'M reiwrts the oecuirence
of more normal gabbro among the Purcell sills which he has specially
studied.
> £. T. Allen and J. K. Clement, .\iner. Jour. Science, Vol. 26, 1006, |i. IIS.
OABBROlCLAN 321
An analogy is to be found in many quartz diabases which are char-
acteristically low in alkalies and tend to be low in alumina. An ex-
ample is seen in the analyses of the diabases of Kusjkin Island, Siberia,
by Backlund. ^ As shown in the next chapter, the evidences are weighty
that quartz diabases are generally developed as syntectics of basaltic
magma and wet, acid sediments. The chemical contrast of these
igneous rocks with primary basalt is also in part explicable by the in-
fluence of absorbed water.
Iron Basalt
In his handbook Rosenbusch devotes much space to a discussion of
the celebrated Greenland basalts carrying segregations charged with me-
tallic iron. He notes their special importance for his general h3rpothe-
sis that igneous rocks have been derived from metallic alloys (Kerne) .
On the other hand, Zirkel favors Tornebohm's suggestion that this iron
is due to reduction of the iron oxides in an originally normal basalt.^
The reduction is ascribed to the absorption of carbonaceous matter from
bituminous sediments traversed by the basalt.' In his more recent
study of the problem, Schwantke has indicated the probability of the
derived origin through the reducing action of foreign carbon.* The
close field association of "graphite basalt" with the Greenland iron-
bearing rock distinctly favors this explanation. Quite recently Bene-
dicks has shown that the Ovif ak iron is a true natural steel, thus power-
fully sustaining Tornebohm's view.*
Anorthosites
Many bodies of anorthosite are so enormous that several of their
special students have been led to describe them as batholiths, thereby
implying that their mechanism of intrusion is like that of the greater
masses of granite. If this were proved to be true, the basal conceptions
of the eclectic theory would need drastic revision. Batholithic granite
has been interpreted as a differentiate of crust materials dissolved in
large abyssal wedges injected from the basaltic substratum. The
emplacement of batholiths at visible levels is credited to this assimila-
tion. Since the anorthosites of the world regularly cut rocks much more
acid than themselves (gneisses, acid schists, etc.) it would seem inevi-
table that, if they were emplaced by assimilation, their chemical com-
> H. Backlund, Mem. Acad. Imp. Sciences, St. Petersburg, Vol. 21, 1910, p. 25.
* A. £. Tdrnebohm, Bihang, Svenska Vet. Akad. Handlingar, Vol. 5, 1878, p. 16.
» F. Zirkel, Lehrbuch der Petrographie, 2nd ed., Vol. 2, 1894, p. 894.
* A. Schwantke, Sitzungsber., Akad. Wiss., Berlin, Vol. 50, 1906, p. 853.
* C. Benedicks, Compte Rendu, Congrds G4ol. Intemat., 11th seBsion, Stock-
holm, 1910 (1912), p. 885.
322 IGNEOUS ROCKS AND THEIR ORIGIN
position would not l^e that of a labradorite-rock. It is, therefore, a
matter of importance that many field evidences indicate an injected,
laccolithic origin for most or all knowTi bodies of this rock. On account
of the critical nature of the subject, in relation to petrogenic theor>',
these evidences will be noted in some detail.
Abnormal Features of the Anorthosites. — As already obser^'ed,
most plutonic or granular igneous rocks are represented in pre-Cam-
brian eruptions, as well as in eruptions of post-('ambrian time, includ-
ing the C'enozoic era. The larger anorthosite masses are unique in
having eruptive dates almost entirely, if not quite, in the pre-Cambrian.
This fact is illustrated in the following table (XVI) showing the princi-
pal outcrops of the rock.
The only large bodies not assigned definitely to the pre-Cambrian
are those of Norway. Kolderup states that the latter cut the fofisil-
iferous "Silurian'' schists near Bergen, but it is not made clear in his
memoirs that the labradorite-rock itself is in this relation. The pub-
lished map of the liergen masses show them to l)e entirely enclosed in
the pre-Cambrian terrane. (labbroid rocks do cut the Silurian bedn
but it is not rendered certain to the reader of Kolderup's papeni that
these gabl)ros are contemporaneous with the anorthosites.
However this may l)e, no wry large Inxly of anorthosite dating from
the later Paleozoic, Mesozoi<*, or Cenozoic time has yet been reported.
It M^enis as if the generation of the greater ma.»<sifs of this rock were due
to the special conditions obtaining in an early stage of the earth's his-
tory. This stagr is, however, not the earliest, for every descrilied body
of anorthosite s<'enis to l)e youngcT than the oldest greenstones and
associated granit<'s of the pre-( 'anibrian basement complexes. The
greatest and most numerous bodies of all have been assigned by Adams
to the *'rj)per Laurentiaii."
A second point of contrast with the other plutonic types is seen in
the fact that the anorthosites do not s<H'm to Iw represented by effusivt?
equivalents. This characteristic is undoubtedly significant. In view
of the large* size of the anorthosite bo<lies one might expect their magmas
occasionally to jienctrate the roofs and issue as flows or breccias. The
absence of these cannot Ix- attributed to removal by erosion, for that
would imply a ])crfcct denudation n(»t matched in the normal bfttho-
lithic-vohanic a»ociations of the j)re-( 'ambrian. On the contrary,
at thi (inn wht ri mch atmrthnsitir mOijina as such originaUd, it appears to
have lacked the [>ower to reach the surface through fissure or diatreme.
This siigj;e<ts some spe<ial profwrtv of the magma, such as high viscosity
or a niiniinal charge of the jjaseous constituents neces8ar>' for continued
volcanic action. As we shall see. there is good evidence of high
viscosity in several of the greater masses at the time of intrumon.
GABBRO CLAN
323
TABLE XVI
Region
Approximate area,
Square miles
Date of intrusion
Labrador peninsula, 17 large bodies (total
area about 50,000 sq. miles.)
Saguenay district, Quebec
St. Urbain district, Quebec (body 70 miles
long).
Morin district, Quebec
Ten smaller areas between Three Rivers
and Montreal, Quebec.
Kildare, Cathcart, and Brandon townships,
Quebec, 5 small bodies.
Chibougamau district, Quebec
Moisie River, Quebec
St. George's Bay, Newfoundland (area 60
miles long.)
St. John, New Brunswick (small body)
Adirondacks, New York
Northern New Jersey (small body)
North Carolina (small body)
Glamorgan township, Ontario (small body)
Georgian Bay, Ontario
Thunder Bay district, Ontario (small body)
Rainy Lake district, Ontario
Sherman Quadrangle, Wyoming
Upper St. Joe River, Idaho
Southern Vancouver Island
Bergen district, Norway
Bergen district, Norway
100
Bergen district, Norway,
Ekersund-Soggendal district, Norway
Voea-Sogn district, Norway
25
52
5 +
10+
18 +
60+
4
400
800+
Lofoten Islands
Angermanland, Sweden
Volh3mia, Podolia and Cherson districts,
Russia ("large massifs '0
Island of Skye, local differentiates
Island of Rum, local differentiates
Raniganj coal-field district, India
Xatal-Zululand
Egypt
Pre-Cambrian
Pre-Cambrian
Pre-Cambrian
Pre-Cambrian
Pre-Cambrian
Pre-Cambrian
Pre-Cambrian
Pre-Cambrian
?
Pre-Cambrian
?
?
Pre-Cambrian
Pre-Cambrian
Pre-Cambrian
Pre-Cambrian
Pre-Cambrian
Pre-Cambrian (?)
Late Mesozoic
Silurian or post-
Silurian (?)
Silurian or post-
SUurian (?)
Silurian or post-
Silurian (?)
Silurian or poet-
Silurian (?)
Silurian or post-
Silurian (?)
?
Pre-Cambrian
Pre-Cambrian
Tertiary
?
Pre-Cambrian
Phases of sills
cutting Permian
strata.
Probably pre-
Cambrian.
324 I0NE0U8 ROCKS AND THBIR ORIGIN
A third abnormality is the very coarse grain generally characteristic
of these rocks. The grain has evidently been controlled in part by
the existence of a cover in each case; in other part it is due to the mono-
mineralic nature of the magma, with its almost infinite concentration
of the feldspar molecule.
Anorthosite a Differentiate of Gabbro. — Most petrograpbera agree
that anorthosite is a direct derivative of gahbroid (basaltic) magma.
The full story of mineralogical and chemical resemblances, of transi-
tional phenomena, and of intimate field association, need not be told
here, but certain field illustrations will be recalled in order to indicate
the probable modes of differentiation.
That the splitting of anorthosite from originally gabbroid noagma
occurs in true injected bodies of the sheet type is abundantly prove*!.
When these masses are fairly thick, the gabbro shows a general ten-
dency to become banded, with the separation of highly feldspathic
(labradoritic) layers from more femic layers. This segregation is some-
times accomplished in place. On the other hand, Geikie and Teall
have given reason for the belief that the well kno^'n sills of the Cuillin
Hills, Island of Skye, are banded because of the intrusion of gabbro
magma differentiated at lower levels and then injected as a heterogene-
ous mixture of salic (anorthositir) and femic magmas.' Similarly, Bar-
ker has interpreted the alternating sheets of peridotite and "allivalite''
(anorthositic rock) of the island of Rum a*< distinct intrusions of for-
merly gabbroid magma, made heterogeneous by spontaneous splitting
in depth. He notes that even these partial magmas were themaelvet
heterogeneous, so that an individual sheet is conspicuously banded with
more feldspathic and morr olivinitic layers. From his description it
appears that the differentiation was progres^sing during the steady or in-
termittent rise of the original magma. It continue<l after the hetero-
geneous magma has rea(*hod its final position, for salic and femic
materials are still further segregated in "concretionary Btructurttn
traversing the various bands.** The magmatic splitting was here, then,
partly in place, and partly preliminary to injection into the visible
sill ehambtrit,*
The vast laeeolith of Diiluth gabbro is locally banded, again with
anorthositic faeies. but it also displays very large segregations of anor-
thosite with in<lividual outerop<. eneh measuring nearly or quite a
> A. Geikif* ami J. .1. H. Tt^ill, Quart. .Four. Gml. Soc., Vol. 50, 18M, p. 645. An
analoio' is found in the nMiiarkablp n>ck eroup including the recently denribed
''ornoitcs** of Sw«>don. iStv A. O. iI(ti;hom. BuU. Geol. Inst., UpMlay VoL 10,
1910, p. 149.>
* A. Ilarker. GtHilosy oi tho Small IitK's. Mcmoim, GeoL S wf^ Groat IMtaia,
1908, pp. 69-77: The Natural History of the Igneoua
p. 140.
GABBRO CLAN
325
square mile in area." In part at least these are regarded as differen-
tiates in place and N. H. Winchell suggests that the splitting was in-
Huenced by gravity, which tended to move the feldapathic pole toward
the roof {southeastern contact)' (Fig, 149), Bowen has recently dis-
covered anorthosite as a differentiate of diabase magma in the sills of
the Thunder Bay district, Ontario, and he points to the possibility that
Fio. 149.— Map of the DiUuth laccolith. {After C. R, Van Hise and C. K.
Leiih, Mon. 52, U. S. G. S., 1911.) A, Animikie slates, etc.; KBE, Keweenawan
"basic extrusive;" /f A/, Keweenawan "acidic intrusive^" C, Cambriftn and later
sedimenta.
large bodies of anorthosite may be formed by similar splitting.* Prior
,states that the dolerite sills cutting Permian strata of Natal and Zulu-
land have split into anorthositic and pyroxenitic phases.'
Very little systematic work has been done to test the hypothesis
that this differentiation in gabbro or diabase sheets and laccoliths has
l>een controlled by gravity, but the writer believes that it is a well
■ N. H. Winchell, Final Report, Geo!, and Nat. Hist. Survey of Minnesota,
Vol. 4, 1809, p. 302 ; N. H. WincheU and H. V. Winchell, Bull. 6 of same survey,
1S91, p. 126; C. R. Van Hise and C. K. Leith, Monogreph S2, U. S. Geol Survey,
1911, p. 374.
' N. H. Winchell, Final Report, Geol. and Nat. Hist. Survey of Minnesota,
V6L 5, 1900, p. 66.
*N.L. Bowen, 20th Annual Report, Bureau of Mines, Ontario, 1911, p. 127.
*<}. T. FlkH Annals, NaUl Mueeum, Vol. 2, 1910, p. 147.
32f)
IGSEOUS ROCKS A\li THEIR ORIGIK
jti;itilip(l cuiK'fption. This doos not mean that all of the feltbpaihi''
pfiti- s)ioiilil Ih' rollprtrd nt the roof, for some of its material muM If
"frozen in" ut docpiT levels iluriiif; the splittinK and simuitanooii<
(Tystullixiition of caili muss, t'niler tlie eiroumslanci-s, gravity woul<l
merely Uiid to raise the uiiortliositic material toward the roof and In
sink the mafrmiitic complemcntK — iwridotiti'S, pyroxomtes, iron nn-*.
et<'.- ' townnl llie floor. A eompilution of nil the available publishtil
data »liows tliat tliese (ccneral tenilenr-ies are illuMrated in the Diiluili
liKTolith. At intervals ulunf; its fl<K>r, ultra-fcmic gahhros, titanie iron
Vir.. l.M).- Mi.p ofilirl
.4. :ui.|>hi1>.>lit.-; (J. <iii:>ri/i
II f.'.V,
II. I I..
Ill t.Ml>l>ri>, C)iir:irio. l.Vtcr F. D. Adiunn anl
M-.l .-iiirv. Cnio-la. IflKt.) /.., Grctiville limatmif.
:ini| i>:ir:iKii<-i-'s: <i\. icrsnite Knrin (intruriv* .
, ii.|.l..lit.> syriiito: (t.\, Ralilirn. Strike lin«a in.l>
(ins, iieriilotili-, iiml pymxi iiiti'sare well ilevHoped; according to Bay-
liv :nnl others. ttiiM- :ire ilitTereiiiiates from the larcolilhir galiliro.'
In I lie r|iilioiii;:iiii;i'i di-triit of tjiii-hee. a Iiody of aiiorthosite cov-
ering Miijii-uhai more tlmri llill siiuaie miles prosenis analogous phe-
luimeuji. Ttiis ma— i-^ over L"< niile> in lengt li. with a width of from 2 to
."i iiiile>. Ii li.-i- l.e. 1. .1.-, liLi.i :,^ !i -Laiholith," but it has th* ear-
mark.'- i<i' :i hK'i'cNJtli. of iinii'li ~iiialli-r >\/.f tlian the Duluthbody. Along
its iionlierii ii:iri. \\\,[.], is maile with Keewatin mtbaam, the
' \V .<, Ilujl.y, .lour. Vol.2, ISM, p. HI*.
OABBRO CLAN 327
anorthosite has gabbro, basic norite, pyroxenite, and iron-ore facies.
Unfortunately, the original roof is not exposed; in its place is a contin-
uous mass of granite which cuts the anorthosite,'
Another case where one may suspect gravitative differentiation is
found in the intrusive gabbro of Glamorgan township, Haliburton
County, Ontario (Fig. 150). This body measures 8 miles in length by
2.5 miles in extreme width. It has been intruded at or near the plane
of contact of the great Grenville limestone and a group of amphibolites.
The igneous contacts are mapped as concordant with the schistosity,
and presumably with the bedding, of these rocks. The relations are
apparently those of a laccolith. Along the northern contact, the gabbro
passes into pyroxenite, hornblendite, and iron-ore phases; feldspathic
phases approaching anorthosite are found near the southern contact.
These facts agree with the conception that the mass is a tilted and
eroded, differentiated laccolith, but obviously little stress can be laid
on this idea until further structural work supplements the reconnais-
sance of Adams and Barlow.^
The greater bodies of anorthosite are exposed in such complicated
relation to country rocks and to the present erosion surface that it is
impossible to locate roof or floor with precision; but, where the dimen-
sions arc large in all directions, it is a priori just to consider the out-
cropping rock as roughly representing the actual area of the roof, which
has been eroded away. If gravitative spHtting has occurred in these
huge bodies, we should expect the differentiation to take place in a
manner like that so often seen in true batholiths. (See pages 240-244.)
As a matter of fact, the greater anorthosite bodies usually have a well-
developed contact phase of more or less typical gabbro or norite. In
the interior each mass becomes rapidly more feldspathic and, for most
of the outcrop, the rock is monotonous anorthosite. The femic pole
has settled to the floor and remains invisible unless fas often happens)
a later eruptive effort sends injections of ultra-basic material through
the anorthosite or its country rocks.
Illustrations of such gabbroid contact phases as indicators of gravi-
tative differentiation have been noted in the following areas:
1. Morin district, Quebec (Fig. 153).
2. Saguenay district, Quebec (Fig. 156),
3. Rainy Lake district, Ontario.
4. Adirondacks, New York (Figs. 122 and 157).
5. Ekersund-Soggendal district, Norway.
» A. E. BmIow, J. C. Gwillim. and E. R. Faribault, Report on the Geology and
KineralReaourci'softhe Chibougamau Region, Quebec, Quebec City, 1911, p. 156.
' ' F. D. Adams and A. E. Barlow, Memoir No. 6, Geol. Survey of Canads, 1910,
p. 1S3.
328 lONKOUS ROCKS AND TBBIR ORIOIN
It n-mains to note the relation of thin splitting to that whereby thr
pyroxene andesites have also l»een derived from l>asalt. Id Chapter
XVII will l>e found a brief statement of the eonditions under whirh
aufcite ftndesite is thus formed. They are the eonditioni* at the vents
of volcanic eruptions of the ecntral type, where magmatie ((ases an-
specially concent rat (hI, with resultinp; effect on the chcmiral proresit uf
differentiation. There is no similar merhanUm for int«nM> gu am-
eentration in laccoliths, where the xalic pole of l>aKaltic B|:rfittiii|{ may
well differ from that at central vents. Other contrast* — different
pressures, <iifferent lengths of magmatie life, etc. — may also lie in
control. Where so little is known of the physical conditions of m»ti-
matic differentiation it is idle to attempt full analysis of the problem:
we may safely rely on the facts of the field as shoning that liaaalt litien
split, more or Ies8 spontaneously, into these two very different, more
salic submagmas. The recorded field relations seem to be best ex<
plained on the theory of gravitative differentiation, where the feliK
pathie units (probably non-consolute submagmas) have tended to
collect at the upp<T part of each chaml)er.
Mode of iDtrusioo. — We have seen that anorthosite haa split off
from normal gabbroid magma within chambers of the sill and laccolith
types. All the clcari'st illustrations of this segregation have been found
in injected )>odies of magma, that is, those with soIi<l roek ftooia. la it
possible to extend that rule to all of the larger anorthosite mnnnrr of
the globe? Obviously the iguestion cannot be answered in full but
there are grounds for lieheving it to lie worthy of serious consideration.
A conceivable diiliculty in the way of crnliting a laccolithir
character for such bodies as the Sagiienay, Murin, or Atliromlaek
masses is the vast size of each, an area out of all proportion to thone
ordinarily characterizing laccoliths. Yet it luu<)x>eome gradually clearer
that theprc-Cambrian was a time when very large l>asic injections were
comparatively frequent, once more illustrating the fallacy of the strict
uniformitarian principle in geoiog.v.
In 18!)n Orant stated the view that the Ouluth gabbro of Minneaota
is a laccolith, aii<l he is supported therein by Van Hise and Leith.* It
was injected nhmg the contact of the .Animikie and Keweenawan form-
ations (Fig- 14!))' It.s length is aljout 12.5 miles, n-it ha maximum width
of about 2.'> mites and an area of alNnit 2400 square miles. Hmh^^
locally cro.-is-euttinK. it.*: contact planes generally paralld the t^^'Mi'^g
of the Animikie, giving a concordant relation locally oK^^-rv.-.! in aetu;
outcrops by the geologists of thr> Minnesota Survey, and suggested i
'U.S.liritnt, Final U('|H>rl.<icol.i i ■■iwrta, Yrt,
l-igg, p. 320; C. K. Van Hisi' jiml <'. K
1911, p. 202.
OABBRO CLAN 329
a map of the whole body. With the averse dip of 10"+ estimated from
the associated sediments, the thickness of the laccolith at its center is
calculated to be over 20,000 feet.
Fio. 151.~Mnp of the Bad River laccolith, WisconBin. (Same ref. as for Fig.
149.) O, pre-Huronian schists, granite, eU,; H, Aaimikie slates, quartiites,
etc.; fir, Kewecnawan basic volcanics; JIT , Keweenawan conBlomerate, sandstoDe,
nud slate; GA, gabbro laccolith; R, "red rock" (granite, etc.).
A similar but less extensive gabbro laccolith has been mapped in
the Bad River region of Wisconsin. It is stated to have an outcrop
FiQ. 153. — Sections at lower contact of Bad River laccolith. (After R. D.
Irving and C. R. Van Hise, Mon. 18, U. 8. G. 3., 1892, PI. 2.) G, granite, gneiss,
sod schist; L, Huronian limestone; Q, Huronian slate (quartioee); /, Huronian
iron-bearing member; S,Animikie slate; G^, gabbro of laccolith. Scale, 1:100,000.
lengtli of 60 miles, a maximum width of 5 miles and a thickness esU-
, mated a» between 9500 and 25,000 feet.^ This body also follows the
• C. tt. Var "'— — 1 C. K. Leith, op. eit., p. 377. Is the Bad RWof body
' ' laccolith repeated in outcrop by the great Lake Super-
330
IGNEOUS IK)CKS AND THBIR ORIOIN
basal contact of tlie Kewcenawan seriea (Figs. 151-2). The structural
relations and form are, lion- again, indications of laccolitbic or sheet
intruHiun and thi»< conclusion !» backed up by the local occurrence of
nunil>erlcss dearly understood, basic Hills already famous in I^e
Superior gcologj-.
The laccolithic or ttlicct nature of the Sudbury norite is specially
well demonHtratrd, afTordinR still another illustration of the immense
scale on which galjbroid magma was erupted into Jloornf ebambers dur-
ing the late pre-C'ambrian time (Fig. 160. p. 348).
Flu. l.Vt.—MHii (>r tli>- Morin ilixirUt i>f fknortliOHitf, Quebec. (Aft« F. D.
.\.h.m^. Ann. H.|.. Col. S.irv. l\n!x<\:i. V..1, S, 1895.) (!N, taam; L {mU
lii'irk), liiiK^lono: 0, k"'''""": -1. :in<>Tt)ior>irf: /', CambriaD uid Ordorkwa
Htriitii (<)vcrlii|i;.
Some of the (Eiibbri) masses uf the .\dirondack8 are definitely de-
siribwi as tnci-olttlis.'
Many oilier gabbroid ami auorthosite n
have not been rrjiar<li-d or dcsrribeii a--< of Iru
lliey are a<'cum|ianii'[l by striiciuj
t'igs. I't'A und 154 Jllu:
' Rcpun of ihp In*
Vul. l.^, IWir. |i. Jlis
. by -tnii'tnf^ fpatiir. -
illustrat' ^^"""^^W*-"
GABBRO CLAN
331
Morin area of anorthosite in Quebec. The concordant relation of the
igneous contacts to the bedding of paragneisses and limestone is clearly
shown, both for the satellitic sheets or laccoliths and for the main body
itself.^ Adams notes that the Canadian anorthosite intrusions
'* frequently" take "a line of least resistance" and lie between bands
or strata of the Grenville series. ^ The recurrent parallelism of the
mapped bands of upturned anorthosite sheets with limestone beds is
significant (Fig. 153). Hogbom has noted that the gabbro-anorthosite
mass of Angermanland, Sweden, has a **laccolithic" relation to the
Archean rocks on the west.^ The anorthosite areas of the Bergen
district, Norway, are very long bands with concordant relations to the
enclosing paragneisses and other schists of the pre-Cambrian; the
:: 1
0
: Mis.
Km.
FiQ. 154. — Sections 1, 2, and 3 of area mapped in Fig. 153.
ground-plans of these intrusives are thus like those of upturned and
eroded laccoliths (Fig. 155).*
Less obvious modes of intrusion have characterized a considerable
number of gabbroid bodies discovered by Adams, Barlow, Lawson, Low,
and others in the Canadian Archean.
In the great Haliburton-Hastings area a dozen masses of gabbro,
**diorite," or anorthosite have been mapped. In every case there is,
marked concordance between igneous contact and the schistosity of the
country rocks (limestones, basic eruptives, etc.). In large part this
schistosity is parallel to true bedding in the invaded sediments and it is
difficult to believe that dynamic metamorphism should have superposed
theMneordance, to the degree mentioned, on a series of igneous contacts
Ann. Rep. Geol. Survey Canada, Vol. 8, Pt. J (map), 1895.
^-*-». Geology, Vol. 1, 1893, p. 334.
»4»L Foren. Stockholm Forhand., Vol. 31, 1909, p. 366.
Museums Aarbog, 1903, No. 12, map.
IGNEOUS ROCKS AND TUEIH OHIOIN
Fid. LVi. -MiiiKif thr:kii<M't1i<»ii<'<u-Ffui. BnKra dktriet, Hot
F. K..l.!rni|.. n.Tit.-nH MiH.iiiii-Aiirh.iK, 1903, No 12.> 0, |
Siliirion nxki; .1. HTii>rlli'>sii.-: .1/. tj
f..r«trikP:m.iai|..
GABBRO CLAN
333
which were initially cross-cutting.^ The field relations so far as
described seem, therefore, to indicate that these intrusives belong
in the injected, laccolithic class, rather than in the subjacent, batho-
lithic class. On this assumption it has been above suggested that the
differentiation of the Glamorgan gabbro CFig. 150) into anorthositic
and peridotitic phases has occurred in situ. The anorthosite body of
of the Rainy Lake region similarly shows almost perfect concordant re-
250
Mis
250
Km
Fig. 156. — Map of anorthosite areas in eastern Canada. (After Atlas of Can-
aria , Interior Dep't of Canada, 1906, PI. 5.) PC^ pro-Cambrian gneisses, etc.; P,
Paleozoic formations; doUedy anorthosite.
lations to the surrounding gneisses.^ We have already noted that the
visible original contacts of the Chibougamau anorthosite are concordant
with the Keewatin green schists and it is significant that the authors of
the report on that district note the occurrence of small laccolithic basic
intrusives in the Keewatin terrane.'
In fact, it seems to be a general rule that the great basic intru-
* F. D. Adams and A. E. Barlow, Memoir No. 6, Geol. Surv. Canada, 1910..
.See maps and page 31, where these authors state that the banding of the thick Gren-
ville limestones represents bedding.
' A. C. Lawson, Ann. Rep. Geol. Surv. Canada, 1887, Pt. F, map and section
"K-L" at page 43.
* A. E. Barlow and others, Report on the Geology and Mineral Resources of
the Chibougamau Region, Quebec, Quebec City, 1911, p. 164.
334 IGNEOUS ROCKS AND THEIR ORIGIN
aivos of the prc-Cftmhrian arc acrompsnied by large and small
Bills and laccoliths of similar or clowly allied compoution. Are the
extensive masses not simply largc-sralo equivalents of these satellitef,
the intrusive mechanism of which has lieen provedf In view of the
structural complexity and metamorphosed character of these pre-
Cambrian areas, one must bo imprt>Hse<l with the large Dumber of
instances in which the pre-Cambrian gabbroid masees have been dis-
tinctly deRcril>e(i as intrusive sheets or laccoliths; and with the other
considerable group of cases where the recorded facts point to the same
mo<le of origin.
Fiii. l'i~' — M'M> "f iin(irthc>sit(> anil ityonitc in Iho Adirondmeka, New Yerl
State. (After llio wjill-riiii)i iif the Slate Sutvp)-.) ON, gBOM and pBuitc: L.
Grcnvillc limcMtonr iiml Hi'histii; .-1, unortbooite and itabbro; S, lyenitc; P, Cun-
brian anil Siliirinn iMilinicntH IhvtUii).
Critical field data are luckini; for the Saguenay anortborites (Fig.
156); for the St. rrimin and Moi.-iie Iliver IxKiiea of Quebec; for ihf
many bodies of this rock in t)ie Labra<lor peninsula, of which about
50,000 square miles, tieconlinKto .\diims 's estimate, arecovered by anur-
thosites; for tlie Adiriindaek mftsM (Fig. 157); for the long Newfound-
land Ix)dy; for the juHirthuf^ites of Wyoming and Idaho; for the Egpr-
sund-Soggendal, Vii.-is-Sogn iind Lofoten l>odies of Norway; for the
Volbynia-Podolia-Clierson anorthtisites of Hussia; and for the body or
IxKlics in %ypt. Ncvertlielcss, the writer believes it best to enter-
GADBRO CLAN 335
tain the laccolithic hypotliesis for all of these and for the other, not
listed, smaller bodies of anorthosites and associated gabbros.
Mere size is no barrier to this conception. The greatest known area
of anorthosite, the Saguenay mass, is no broader than the Duluth
gabbro, to which a laccolithic origin has been ascribed by its latest
students. The area of the Saguenay body (5800 square miles) is
greater than that of the Duluth body (2400 square miles), but that con-
trast may be due to the different attitudes of the two masses with re-
spect to the present erosion surface. It is important to note, also,
that many well-exposed sills have ground-plan areas of the same order
of magnitude. (See page 67.) These well-understood bodies clearly
indicate the possibility of horizontal injections characterized by much
greater thickness.
Special Conditions for the Formation of Anorthosite. — Yet it seems
necessary to postulate unusual conditions for the differentiation of
anorthosite from gabbroid (basaltic, substratum) magma.
In the first place, the original magma seems always to have occurred
in a very large volume wherever anorthosite is well developed. This
is true for all the instances of differentiation in place, as above de-
scribed. Adams, Kolderup, and others have emphasized the common
protoclastic structure in various anorthosites; a feature of the rock in-
dicating a low temperature at the time of the latest magmatic move-
ments, if not at the time of actual injection.^ It is thus possible that,
in these cases, the material of the basaltic substratum was differen-
tiated in depth before the laccoliths were formed. This possibility is
matched by the numerous instances where satellitic sheets of pure
anorthosite have issued from the larger, visible bodies. But here
again, the original gabbroid (basaltic) magma must obviously have
ha<i large volumes to supply so great masses of the feldspathic rock by
differentiation.
Without attempting the impossible task of evaluating the relative
importance of differentiation in place and differentiation in depth, we
may conclude that the substratum basalt was erupted into closed
chaml)ers during pre-Cambrian time on a scale unmatched in more
recent periods. The cause of this contrast evidently furnishes a large
question affecting the physical geology of the earth as a whole. The
fact suggests speculation as to the existence of a possibly thinner earth
crust during the pre-Cambrian and prompts also speculative correla-
tion with the other principal fact that the pre-Cambrian was also a time
of the general development of granitic batholiths. These speculations
will not be pursued in this place, but it should be noted that the peculiar
,Ay« D. Adams, Ann. Rep. Geol. Surv. Canada, Vol. 8, Pt. J, 1895, p. 115, and
G. F. Kolderup, Bergens Museums Aarbog, No. 12, 1903, p. 46.
336 lONEOVS SOCKS AND THEIR ORIOIS
conditions of later pre-C'am)>rian timo are those allowed for on the
hypothesis of a bsnaltic subHtratum and on the related aRsumptioiu of
the eclectic theory hrrc presented.
Rock Types Syngenetic with Aaorthosites. — The theor>- demancLt
that if a large Kheet or laccolith of superheated gabbro magma ia in>
jectod into the earth's crust, some assimilation of roof or floor, or ai the
walls of feeding dikes, must take place. As detailed in the next
chapter, this expectation is abundantly matched by field facts. The
quest ion arises as to its lM>aring on the present connection. Other things
lM>int; equal, the huge pre-( 'ambrian laccoliths should show proofx of
a.sdimilative power rorn>spondingly greater than that illustrated in
the Purcell sills of British Columbia, the Minnesota sills, or the South
African sills. (Se<' i)agcs 3-14, 340, and 349.)
i^ome of the large ma.s.-<es composed in greater or less part of anor-
thu.iite do in fact <-arry nx'k pha.s4>s which are lx»t interpreted aii
hybrids or else ilifferentiates of synterlic magma. The Dululh gabbro
pa.-«Hes not only locally into true aitortliosite and also its ultra-femic
or ferric e<im[>tenients, but also inlu the " red rwk.t," graniten, syenite*.
etc., whi<-li have long l>een corn-ctly referred by N*. H. Winchell to
.■solution of country rocks in the gabbro magma. These acid types hare
lieen differentiated from the syiitecticunr) are chiefly represented at the
roof of the huge intrusion. The case is strictly homologous to that at
Pigeon Point, Minnesota, admirably described by Bayley, «itb whom
Winchell, I.awson. and others are in agreement. In Chapters XIX and
\X areoutlineil the reasons underlying the nTiter'sIielief that syenites
and other alkaline ru<'ksaredifTereiiliatesof hyltnd magmas formed by
the solution of basic se<liiNents, etc., in l««lies of primary bawalt. The
aiiorttiosites of the Ailirondaek Moiiiilaiiis, New York, like thotte erf
Norway, are, in fact, intinuitely .is.s<iciate(I. and syngenetic with syen-
itic riK-ks in A[>pni|iriHte cliemical relations to the respective countr>-
rocks. In New York. 1 he syenites, with local granitic, monzonitir, and
slioukinitic facies. jinss gruduully. at certain places, into anorthosite
anil also into the conteniiKiraneous gabbro.' .\t other places the syen-
ite cuts the cnl<'ic riK'ks. The double relation shows the poaubility that
the more "siliciuusniagmii )ins lieeu first generatetl in a commoa cbanH
ber with the gabbro-anortliusite and has then been moved i
ally so as to cut locally tlieulrently <rystallii!ed femica
The complex held relations and esjs'cially inUiUHe defonnalioo I
metamorphisni have cnnspired to obscure the exact proceM by whi
this a-venililing of igiienus ty[>es lias Ixrcn n.iuniplinhed. Thw
Gushing U-iieves the .V.lirundack syeni *« "^^ '■ I a
' il. 1>. Cushinn. Bull, 115, N. Y. Stft*
GABBRO CLAN 337
at their mutual contact, he also suggests that the syenite is a differ-
entiate of the anorthositic magma.'
A close parallel is found in the group of syenitic types, anorthosites,
norites, quartz norites, monzonites, adamellites, and banatites in the
Ekersund-Soggendal district of Norway. Another likewise found by
Kolderup is the consanguineous group of anorthosites, norites, gab-
bros, mangeritcs, monzonites, soda-syenites, and soda-granites of the
Bergen district and he has also described monzonitic, banatitic, and
adamellitic phases in the anorthosite-gabbro area of the Lofoten
Islands.'
Adams describes contact phases of the Morin anorthosite as charged
with quartz and hornblende and notes the sporadic occurrence of or-
thoclase and biotite. One may suspect here a slight acidification of the
original magma, but it seems clear from Adams's account that granitic,
syenitic, or other alkaline differentiates of syntectics are here not de-
veloped to any notable extent. The same appears to be true for many
of the anorthositic areas of Eastern Canada. These failures to show
evidence of assimilation may possibly be related to the low temperatures
which Adams credits to some of the anorthositic magmas at the time
of their intrusion; they were then not superheated but were crys-
tallizing, so that solution of foreign material was almost entirely in-
hibited. The heat supply wa.s, in these cases, just sufficient to permit
differentiation of the original substratum material. The cause of this
relatively low temperature was perhaps another phenomenon special to
the later pre-Cambrian period. But even then, basaltic magma was
often superheated at the time of eruption, and that has been the rule
for the whole post-Cambrian era.
Conclusions. — The geological facts known about the anorthosites
thus appear to sanction the conclusions: f 1) that they are gravitative
differentiates of basaltic magma of the same composition as that here
attributed to the substratum of the earth's crust; (2) that this differ-
entiation generally occurred in large chambers of the laccolithic (or
chonolithic) type, though possibly in part below the levels where the
visible masses arc situated; (3) that the anorthosite bodies, huge as
many of them are, are not subjacent or bathoUthic in character; (4)
that the anorthositic laccoliths were developed almost wholly in pre-
Cambrian time and that, in general, the world's greatest laccoliths of
gabbroid magma were injected during the same period; (5) that some
of the anorthoaitic magmas were generated in chambers hot enough to
> H. P. CusKinK, BnH Geol. Soc. Amcr., Vol. 10, 1899, p. 188; BuU, 9S, N. Y.
, Statu Miweun
& unu Aarbog, 1890, No. S; 1903, No. 12; 18B8,
338 lOXEOUS ROCKS AND THEIR ORIGIN
permit of the solution of large volumes of foreign rock, while othera
8e<*m not to have lM*en sufficiently superheated to perform ootAble
assiniilution; an<l ((>) that no vulcanic phase of anorthosite as such has
In^en recognized, though, of cours<% great laccoliths of basalt may have
originated and fed volcano<*s alM>ve their roofs during the initial, hot,
magmatic stage l)ef()re the anorthositic phase could be differentiated.
Pillow (Ellipsoidal) Basalts and the "Spilitic Suitb"
Of late years the ''pillowy lavas'' have been discovered in many
parts of the earth, in terranes ranging in age from the early pre-
Canihrian to the twenti(*th century flows from the Matavanu volcano
of Savaii. These luvas are either normal basalts or else types doariy
allii^l to them. lYi most cases their special students have concluded
that they rcpn^^nt subaqueous flows. The reason for the balling-up of
the lava into relatively small, completely separated pillows or ellip*
soids is a physical problem of fascinating difficulty; the structure ap-
IM»ars to Im» connected with the development of the "spheroidal state'*
at the contact of water and sufx^rheated basic lava, but no one has yet
made the matter clear.
Thes<* pillow lavas have recently won renewed attention since Dewey
and Flett stated that " the pillow-lavas are meml>er8 of a natural fam-
ily of igneous nx'ks. the spilitic suite, that can lx> clearly distinguished
from the Atlantic and Pacific suites.*'* Marker himself, the originator
of the concept that Atlantic and Pacific suites exist and are related to
two diff'erent kinds of cnistnl movements, has acknowledged that *Mt is
at least manifest that the distribution of different groups of igneous
rocks in Britain cannot be explained by any initial want of uniformity
in the composition of the earth's cru.st (including the primary magma
chamlKT or chanilKTs) in this tract.*'* According to Harker*s hy-
I>othes(^s, the two suit(*s have Ix^en differentiated from an originally
homogeneous, primordial magma underlying continent and ocean alike.
and the caus(> of that differentiation is found speculatively in the
dynamical conditions respectively repri-sentcd by normal faults and
by overt hru^ts. Several writers have noted reasons for doubting the
possibility of distinguishing the two suites, either on a geographical
basis or by the ty[H\s of asso(*iated crustai movements. Harker has
also admitted that the two suites are {X'trographically identical at their
corresponding acid and basic ends.' It is certainly imposmUe to
» n. Dewey and J. S. Flett. (let)]. Mag., Vol. 8, 1911, p. 245.
* A. Harker, The Natural History of the Iriicoiib Rocks, New York. 1909^ p.
109.
' A. Harker, op. cit., p. 90.
GABBRO CLAN 339
tinguish many basalts syngenetic with the typical alkaline rocks of
Atlantic suites from basalts so often found in Pacific suites. Since
the "spilitic suite*' is throughout of basaltic habit, it is obviously a dif-
ficulty piled on difficulty to distinguish its members from the basic
types of the Atlantic suite as distinguished from the Pacific suite and
vice versa. The offered definition of the new suite thus suffers from the
indeterminate nature of the rock classes with which Dewey and Flett
have attempted to bring it into contrast.
The independence of the spilitic suite is no more evident if it be com-
pared with the basaltic and diabasic rocks without regard to the assign-
ment of these rocks to the hypothetical suites of Harker, Becke, and
Prior. The typical spilites of Germany are most intimately associated
with ordinary basalts or diabases and Dewey and Flett admit that all
these rocks belong to one eruptive period.^ The albitic character of
the feldspar in spilite is interpreted by those authors as generally not
primary but as due to pneumatolytic, " post- volcanic or juvenile
changes of rock-masses." They do not discuss the obvious suggestion
that the abundant soda of a spilite has been concentrated from an
underlying mass of normal basaltic magma; yet there are well-ascer-
tained facts supporting that view. Space here fails for their full pre-
sentation but, as usual, special emphasis must be laid on the testimony
of the sill, that magmatic form which, when exposed from floor to roof,
offers the maximum of certainty as to what actually happens in an erup-
tive magma. Bowen and Collins have studied instructive examples
in Ontario. Many diabase sills of the Nipissing and Timiskaming dis-
tricts cut Huronian argillites and show albitic facies with veins or dikes
of sodic aplit€ or granophyre. The salic differentiate has tended to
accumulate at the roof and its albitic constituent has gone out into the
roof argillites, forming typical adinole, essentially like that generated in
the slates contacting with the spilites of Europe. These diabases of
Ontario are themselves either quite normal or carry small amounts of
quartz and micropegmatite. After comparative studies, Bowen con-
cludes that "the literature of albite-rich igneous rocks shows their gen-
eral association with gabbros intrusive into argillites," and suggests that
the *' water originally contained in the sediment and, in this class, in
large amount, takes an important part in the transfer" of the albite
molecule from the normal basaltic magma. ^
* H. Dewey and J. S. Flett, op. cit., p. 205.
« W. H. Collins, Econ. Geology, Vol. 5, 1910, p. 538; N. L. Bowen, Jour. Geol-
ogy, Vol. 18, 1910, p. 658. Similar instances of gaseous transfer and of the
development of albite-rich rocks or mineral aggregates by water-gas acting on
normal basaltic magma are described by B. K. Emerson (Jour. Geology, Vol.10,
1912, p. 608, and Bull. Geol. Soc. America, Vol. 16, 1905, p. 91) and C. N. Fennei
(Annals New York Acad. Sciences, Vol. 20, 1910, p. 93).
340 IGNEOUS ROCKS AND THEIR ORIGIN
The Rubmarine origin of the pillow lavau implies that their maicniA
pas.se(l through wet seilimentH of greater or k*8s thickness. Under those
conditions water-gas must play an important rdle in modifying the
magma in the vents and it secerns impossible to doubt that occasionally
the upper part of the magma eolumn and also some of the extruded lava
will l>ecome "albitized/' Meanwhile, the general iKxly of the igneous
rock must often l)e profoundly altcTed by the al)sorl)ed water-gas or hot
water, exactly as descrilx^d b}' the many authors writing of the spilitic
masses.
In conclusion, the writer believes that the spilitic rocks are pneu*
matolytic derivatives of normal basaltic magmas and that the modify-
ing gas is chiefly water of rets^urgent, not juvenile^ origin.
Transitions to Other Clans
The eclectic theory requires that there l)e close fiehl association of
basaltic r<M*ks or their chemical equivalent with rocks of all the other
clans. This deduction is clearly matcheii by the facts of distribution
for distinct massc^s, as abundantly illustrat(*<i in Table XXI and
throughout the pag(*s of this or any other general work on petroIug.v-
The intimacy of the associations is again shown in the gradational
phases so often observed betwcK'n basalt, dia1)ase, basic porph>Tite, or
gabbro, and leading represc»ntatives of every one of the other clans.
It is not necesi^ary to give complete illustration of this fact, which is
known to every informed {K^trographer. The only doubt possibly
arising is that as to the transitions lK*tween memliers of the gabbru
clan and memlKTs of the alkaline <*lans. Hence, in Chapter XX thi-
|K>int will be siKH'iaily not(>d as one of the indications of a srconilar}*
origin for most, if not all, of the so-called alkaline rocks.
CHAPTER XVI
GRANITE CLAN
Included Species
The complexity of the granite problem is indicated by the number
of principal species in the granite clan. Again Rosenbusch's hand-book
has been used in the preparation of the ligt of species, though the name
** granite*' here includes his '*granitite." This difference of usage
seems to the writer advisable since the biotite-bearing types are by
long odds the most abundant rocks of this clan. There is no good
practical reason for calling the two-mica species "granite proper."
Milch has made the same suggestion, showing that the type "granite
proper," as originally distinguished by G. Rose, is merely an altered
form of the Riesengcbirge biotite granite.^
Plutonic Types
A. Suhalkaline Granites,
1. Muscovite-biotite granite.
2. Lithionite granite, luxullianite.
3. Alaskite, alaskite porphyry, tordrillite.
4. Some biotite granites.
5. Amphibole granite, andengranite, Rapakivi granite.
6. Pyroxene granite, diopside granite, uralite granite, hyper-
sthene granite, enstatite granite, charnockite.
7. Tourmaline granite.
B. Alkaline Granites (soda granites),
1. Some biotite granites, some "quartz monzonites."
2. Arfvedsonite granite, ekerite:
3. Riebeckite granite.
4. Hastingsite granite.
5. Aegerite granite.
6. Acmite granite.
Dike Types
A. Subalkaline Types,
1. Granite porphyry, alsbachite.
2. Granophyre.
^ L. Milch, Neues Jahrb. fur Miner, etc., B. B. 15, 1002, p. 203; Festachrift
sum siebxigsten Geburtstage von Harry Rosenbusch, Stuttgart, 1906, p. 130.
341
342 JGXEOrS ROCKS AM) THEIR ORIGIN
3. Some Inotite aplites.
4. Pyroxene aplites, bronzite nplitei<.
5. Hornblende aplites.
T). Alaskite aplites.
7. Some p^'j^matites.
B. Alkaline Types.
1. Soda granophyre.
2. Ek<»rite porphyry.
3. Alkaline quartz-syenite porphyry.
4. Some biotite aplites.
5. Paisanite.
G. Some |M*gmatites:
Kffimve Types
\. Subalkaline Types,
Khyolite, liparite, quartz porphyry, some felsitofl, quartz
trachytes, microfelsite, nevaditc\ mierogranitc, some grano-
phyres, obsidian, pumice, pitchstone, pitchstone porph>T>%
feLsophyre, vit rophy re.
B. Alkaline Types.
Soda rhyolit(\ soda liparite, krablite, comendite, quartz
keratophyre, quartz pant(»llerite, pantellerit^.
(Iknkk.vl Statement
The eclectic theory assum(»s, for th(» rea.sonH assigned in Chapter
VIII, that granitic material composcMl the outermost original shell
of the earth. The thin, rhyolitic, scoriaceou!<, surface skin theoretically
ex|)ected would obviously Ik? an ephemeral feature in the prinuti\e
areas subj<»(*t to mo<l(Tate erosion. Beneath this phase the material
would Ix* granitic in texture as well as in chemical composition. ^>f
late y(»ars som(» geoloj^ists have In^en accastomed to deny the esdstencc
of any remnant of a primitive crust among the visible terranes. Thi^
conclusion is hazanlous. sinc(* it is based on an induction which cannot
be regarded as complete until all the pre-( *ambrian formations ha\*e
IxTU examined in d«*tail. Many generations of geologists will be re*
quired before these rocks nn^ well map[KMl and their structural relatione
determined with sufficient certaintv to allow of final l)elief in the com-
plete invisibility of the primitive crust. That its remnants mu.<t
cover but small areas is now practically certain: but no one can yet l>e
sure that some of the granitic rocks aln»ady <Iiseovcred are not partj>
of that primitive (*nist, which has n'maine<l unfused since the day of its
original consolidation. The question n^mains still insistent for those
GRANITE CLAN 343
geologists who, for example, arc working in the basement complexes
of Fennoscandia and Canada.
On the other hand, the vast majority of the rocks referable to the
granite clan are definitely eruptive. The problem of their origin is so
broad in its scope that it may almost be regarded as equivalent to the
petrogenic problem as a whole. The eclectic theory stands or falls
according to its ability not only to explain these eruptives but also to
forecast future discoveries concerning the granites and their allies.
Obviously, neither of these tests can yet be applied in detail. An
extremely great development of the theory is still necessary before the
exact conditions under which a pyroxene granite rather than a biotite
granite, or a comendite rather than an ordinary rhyolite,has formed at
a given place. Only a few of the more salient points on the genesis of
these granitic and allied eruptives can now be discussed with profit.
Some of these have been selected as affording illustrations of the more
important principles bearing on the granite problem.
The origin of post-Keewatin granitic magma in general has already
been discussed in Chapters VI, VIII, X, XI, and XII. In general it is
primitive material of the earth's crust which has separated from syn-
tectics, periodically and locally formed between primary basalt and the
primary acid earth-shell. The silicious differentiate has largely
crj-stallized in batholiths and stocks, which have no visible floors. It
is, therefore, supremely important that true granites have been found
in the appropriate relation to the basaltic magma of thick sheets with
exposed floors and roofs. In the visible rocks of such injections all
transitions between pure basaltic (gabbroid) magma into the syntectic
and thence into its typical granitic differentiate, are represented.
Though the known number of thick sills and allied intrusive sheets of
basic magma is small, their magmatic behavior merits special attention.
Each of them in its results is like a gigantic experiment in petrogenesis;
each well-exposed chamber is a crucible which can be examined from
top to bottom. The batholith or stock can be examined only at levels
near its roof and, therefore, obviously fails in every case to furnish all
the data required in the granite problem. For this reason the writer
has laid particular stress on those basic sheets in which granites have
been differentiated.^ In his opinion they should be given right of way
in a discussion of granites in general. The reader is referred for details
to the wTitings listed in the accompanying footnote, but an abstract
of the essential facts there described, together with important addi-
tional illustrations, will be useful.
» R. A. Daly, Memoir No. 38, Geol. Surv., Canada, 1912, pp. 221-255; Amer.
Jour. Science, Vol. 20, 1905, pp. 185-216; Festschrift zum siebzigsten Geburtstage
von Harry Rosenbusch, Stuttgart, 1906, pp. 203-233.
IGNEOUS ROCKS AND TUSIR ORJGJN
Species Derived from Kyntectics of Sediments amd Babalttc
(CiABBHOin) Magma
Purcell Sills.— AmoDR tlio hpHt-oxpoHod cxamplea are the Puredl
silla, which out Cntnhrian or older srilimrnta in 80UtheaHt«m Britiiih
Columbia. ThoHc studied in K"'a(<'>4( detail form a group invadio|t
Fui. 1.->S.— .-Jn-ticn. to wi.li-. of thi-
ilifTirrriliiittil nillx mt th« Moyw Ritw,
r :tS (;...l. .-iHr^-. CmiukU, I»12. p. 248.)
y, quurttiti-; /, bi'ititi- Kr:iiiil<': -'. ho
iil.1i-Ti.l<^l>icitiip icniule; 3. intcmcdutc
rofk; i. p.hl.m. Fiv-- ^ilL- sh..«i,.
vrry tliick, liiiniup'iirous. f('|ils{i:it)iii' lunl niicaccoua quartiites at the
point wliiTc tin- -Miivif rivi-r inisscs tin- Intomstional Boundary.
Tlic Ihirkiit-ssof tht'st-lioilit-s vurii.-' from lllOf(K-t or Ins to About IflOO
fret. Tlii'v illii>triit(' tiruvittitivi- itifTiTcntiAtiun with p
(Fig. 158) In oach of wvi-ral iii^lanc ' ' s et coi
GRANITE CLAN 345
a true biotite granite (rarely hornblendic) passing downward into horn-
blende gabbro. Especially at the upper contact of one of the thicker
sills the quartzite is more or less intensely metamorphosed. While
preserving its bedding, the now massive sediment has a field habit and
microscopic character markedly like those of the sill granites. In
both rocks micropegmatite is abundantly generated. The field im-
pression that the granitic layers are due to the assimilation of the
quartzites is corroborated by chemical analysis.
Similar cases discovered further north in the Purcell mountain
system are reported by Schofield and, in the present writer's
opinion, the granitic phases there are differentiates from the same type
of quartzite-gabbro syntectics.^ South of the International Boundary,
Calkins found still other examples and has briefly described two of the
gravitatively differentiated sills. He is of opinion that here also the
materials differentiated were syntectics of basic magma and silicious
sediments.^
In these sills of the Purcell mountains occasional quartzitic in-
clusions are seen to be surrounded with shells of mixed material which
can only be interpreted as the actual syntectic. Such material was
evidently *' frozen in'* before it could rise away from the xenolith, itself
enclosed in the magma when the viscosity was very great. In the
earlier, hotter state of the magma the dissolved silicious matter diffused
into the original gabbroid melt and, when the magmatic state was suf-
ficiently prolonged, largely collected at the top of the sill. If, on the
other hand, the sill was thin and cooled with relative quickness, this
secondary magma remained diffused through the original gabbro and
there crystallized. Such appears to be the best explanation of the
abundant interstitial micropegmatite and quartz found by the writer
and by Schofield, Calkins, MacDonald, and Pardee in the numerous
sills and dikes of this extensive region.'
A chief ground for belief that these acid phases originated in assimi-
lation is found in the " consanguinity *' between the somewhat ab-
normal granite and the invaded sediments. The relative homogeneity
of the latter through thousands of feet of thickness permits a fairly
close estimate of their average chemical composition. This principal
<latum is very seldom afforded in areas containing thick basic sills;
hence special emphasis must be laid on the facts determined in the
Purcell mountains.
The field evidence is clear that the assimilation was not confined
to the sill chambers actually occupied by the secondary granite. The
* Cf. S. J. Schofield, Summary Rept., Geol. Survey of Canada for 1910, p. 131.
« F. C. Calkins, Bull. 384, U. S. Geol. Survey, 1909, pp. 48-50.
» CSr. J. T. Pardee, BuU. 470, U. S. Geol. Survey, 1911, p. 47.
346 IGNEOUS ROCKS AND THBIR ORIGIN
Kolution of quartzitc at their main contacts and of sedimentary blocks
Ktoped down or up from those contacts has unquestionably taken
place: hut it is highly probable that assimilation occurred during the
upward passage of the gabbroid magma through the sediments under-
lying each of the thicker sheets. Ai)ophysaI dikes are occasionally sent
off from the granitic upper layers. It is reasonable to believe that,
after the differentiation of these layers, their acid magma (secondare'
granite) would sometimes Im* inje(*ted, as sills, into the overl3ring sedi-
ments. Such intrusions, late in the history of the magmas, would
theoretically l>e ai<i(*d by the great tension of the resurgent gases. Such
is one explanation which has 1)een offered for a few of the purely granitic
sills found in the Purcell mountains.^
Marysville Sill. — In tlu* Marysville mining district of Montana the
shales and sandstones of the Belt series are cut by thin and thick sills
of gabbro. This ro(*k seems to be generally of normal composition
but at the r(K)f of one of the sheets the intrusive is a musco\ntic micn>*
granite, carrying 40 per cent, of quartz and 40 per cent, of alkaline
feldspar. The dense granite is nearly of the same color as the over-
lying hornstone, which is itself a quart z-muscovite rock, yet the sili-
cious phase of the shec^t encloses " many small, sharp chips of the horn-
stone.*' The evident chemiral and mineralogical similarity between
granite and hornstone and the geologi<*al relations of each to the nor-
mal gabbro strongly sugg(*st that we have here a homology to the Movie
.sills.*
Minnesota Cases. — The well-known sheet or dike at Pigeon Point.
Minnesota, is another striking parallel (Figs. 149 and 159). At its hang-
ing wall the gabbro is ovcTlain by true granite, which Bayley, basing hin
(conclusion on unusually thorough field, chemical, and microscopic
study, has interpreted as due to the solution of slate and quartsite in
the original gabbroid magma. After independent field study, Lavson
states his full agreement with this conclusion.*
Similar relations obtain in the Duluth laccolith: its maximnm
thickness is to be measured in miles (Fig. 149). The laccolith cut*
rpp<T Huronian slates and more acid sediments, as well as the Lover
Huronian and pre-Huronian complex. Along its upper contact are
huge masses of **red rocks.** granites, syenites, etc., which have long
l>e<'n explained by Xorwoo<l and N. H. Winchell as products of the
» R. A. D:ily, M««moir No. ;W, Oih)!. Survey of Canada, 1912, p. 219. Thi^
incinoir contains ii dftaili'«l tiisnission of tlio PurcoU silU.
' J. Hurnll. IVof P:i|MT Tu, W S. (mm,! Surwy, 1907, p. 48.
' W. S. Haylcy. Hull. 1(H.). l'. S. (h>oL Sarv«-y. 1S93. A. C. Lswsoa. Bal. S.
(.leol :nul Nat. Hi^t. Snrv.-y, Minnt-:*nta. ls«U. pp. 30, 31. 44.
GRANITE CLAN
347
solution of the sediments by the gabbroid magma. ^ The lavas over-
lying the "red rocks" are closely allied to the laccolithic rocks.*
The Bad River laccolith of Wisconsin is 60 miles long and for most
of its lei^h is from 2 to 5 miles wide (Fig. 151). The maximum thick-
ness has been estimated at 25,000 feet.* Along the roof is a long out-
crop of acidic "red rock," which resembles the secondary granite of the
Pigeon Point intrusive. Van Rise and Leith regard this "red rock" of
the Bad River laccolith as a dike cutting the gabbro, but the relation
of the two types is suspiciously like that in the Pigeon Point, Sudbury,
Fio. 159. — Map of Pigeon Point, Minnesota, (After W. S. Bayley in Mon. 52,
U. S. G, 8,, 1911, PI. 12.) M, A nimikie sediments metamorphosed b^ the gabbro;
A, Animikie slates and qusrtzites; (i,olivine gabbro; /, intermediate rock; K,
quart! keratophyre {"granular red rock"). The gabbro sill dips with the sedi-
ments, to the aouth.
and Purcell intrusives. Further field work seems to be necessary to
show whether or not this spatial relation is purely accidental.
Sudbury Sheet. — The celebrated interformational sheet of the
Sudbury district furnishes a standard ca.se of gravitative differentiation,
with granite and granodiorite (micropegmatite) passing downward,
through intermediate rock, to norite at the lower part of the intrusive.
During a visit to the field the writer was impressed with the great re-
semblance of this body to the differentiated sills of the Purcell moun-
' N. H. Winchell, Final Report, Geol. and Nat. Hist. Survey, Minnesota,
Vol. 6, 1900, p. 978.
' See Monograph 62, U. S. Geol. Survey, 191!, pp. 202 and 377, and large map
in pocket; also Final Report, Geol. and Nat. Hist. Survey, Minnesota, Vol. 4,
1899, plates 66, 67, 68, 69.
'C. R. Van Hise and C. K. Leith, Monograph 52, U. S.Geol.Survey,1911,p.
377, aad large map in pocket; see abo plate 22 in R. D. Irving's "Copper-bearing
BoAa of L«ke Superior," Mon. 5, V. S, Geol. Survey, 1883.
348 IGNEOUS ROCKS AND THEIR ORIGIN
tuitiM: uuil with the likeness of Koneral field haliit between the Sudbury
mirroiM'Kniatite anil tlie overlyiiift. intensely metamorphowd setli-
ments. Ctiicman has, in fact, found field evidence of thefltopiii|[ snd
iliKcstion uf quartzitc lilwks in the noritie magma.' Here, again, it
woiihl S4f'm proliahle that all of the assimilation of acid rocks ia not to
Ik' rr(-<hted to the- original magma after injection; some of it may bsve
Ih'cii a<-(-onii)li^heil during the injection of that magma through the
underlying qiiEirtzitir and "Luun-nlian" terranes (Figs. 160 and 161).
I-'ii;. ir>(l. M:iii (ir th>' Sii'Diiir.v ili'trii). iihitwinK nirkH-bMrinit iatcrfomwticMiBl
><hi'i-l. (AfliT A. )'. Col'iiiiui. lt<'|). Itur. Mini-s. Ontsriii, Vol. 14, lg05.) L. Uum-
ti:m KiKii^** •>■)'! K>''i"i'''' "' lonvr tliirnniiiii xUle, (|uartiite, (Enywsckr, gnat-
ft'iiii'. Hi.: ('7'. Tniul l.:>k<- ronKloiiiiTittr ami OnapinK tuff; SL, Onwktin abtr;
.V.S, ('Mm^4ll^•l xiiiKUlom-: .V, nurile: M. mirrnpcfmutitic fnuul« and tf»ao-
iliuriK-. Tin- cliiM'l hits » liiisin ntnii-tiirp.
It shiiidd )m' noted that the outcrop areas of the rocks cannot be af
sunied toshiiw the exact ratio of volumes I>e1 ween acid and basic typm
in (his sheet. Coleninii holds that it was warped into a spoon shape
during or iniiuedintciv after intrusion. If such is the case, thv
ai-id dirfirititiiili' should have eolleeted in greatoxt thickness OD the
outiT. ui)per rim of the "spoon." The initial superheat of the basic
in-igma ne<'d not. therefore, he assumed to l>e so great as that de-
niandeil if assiniilatioii had taken place in prftportion to the avenge
niilths of the aetual uutemp-i of granite and norite.
Many other bodies, illustrating the loniuion development of micro-
> A. 1\ (.'i>leniiui, J»ur. CU-uluKy, \'ol. IS, 1907, pp. 774 and 77a.
GRANITE CLAN
340
ppgmfttite, cither disseminated or segregated, in inbiisive diabase,
gabbro, or norite cutting acid rocks, have been found in Minnesota,
Ontario, Quebec, Connecticut, Scotland, Sweden, Finland, Russia,
India, South Africa, Australia, Antarctica, etc. In some of these,
gravity has clearly controlled the segregation of the granitic magma or
Wa analogue. A few of such additional cases may be mentioned.
Insizwa Intrusion. — A notable parallel to the Sudbury sheet is fuT'
nished by the great Insizwa intrusion of East Griqualand. ' It is a sheet
2000 or 3000 feet thick, intrusive into the shales and sandstones of the
Beaufort series CKarroo system). The dominant igneous types are
gabbro and norite, which merge into each other. At the bottom is
Fio. lei.^Section of the sheet mapped in Fig. 160. (After A. F. Coleman,
Jour. Geol., Vol. 15, 1007, p. 703.) L, Laurentian; Q, quartiite; QR, graywack^
etc.; S, greenstone schist; O, older norite; 0, granite; C, IVout Lake conglomer-
ate; T, Onaping tuff; 8L, Onwatin elate; SS, Chelmsford aandctone; N, noritfl:
M, micropegmatitic granite and granodiorite.
" augite picrite ", locally underlain by masses of sulphidea — pyirhotite,
chalcopyrite, and pentlandJte — of the same species as those at Sud-
bury. Above comes a compound phase, partly olivine gabbro and
partly olivine norite. This ia overlain by olivine-free gabbro, bearing
much interstitial microp^matite and cut by veins of coarse pegmatite.
Both gabbro and adjacent homfels are cut by small dikes of "micro-
granite." The sediments are intensely metamorphosed for a distuice of
more than 400 feet from the upper contact and for a distance of 200
feet from the lower contact. No individualized layer of micrope^Ema-
tite or gramte appears at the upper contact, as at Sudbury. For some
reason, perhaps for lack of sufficient superheat, this South African
magma has not digested the acid sediments to the same degree as that
manifested in the Ontario, Minnesota, and British Columbia cases.
■ A. L. du Toit, 16th Ann. Rep. Geol. Conun., Cape of Oood Hop«t 1010, p. 111.
350 IGNEOUS ROCKS AND THEIR ORIGIN
Other South African Cases. — The thick intrusive sheets of diabase
or dolerite, cuttinf^ the Pretoria shales and quartzitic sandstones, are
known to contain acid (''felsitic*') micropegmatitic phases which merge
into the basic rock, but details of their precise structural relations are
yet lacking.' Hatch and C'orstorphine state that this micropegmatitic
type has ''much resemblance to, and is perhaps genetically comiected
with, the 'red granite' " of the district. Mellor has described this red
granite as forming the roof phases of laccoliths which are not exposed
to depth sufficient to show floors or their own lower levels.'
Rogers and du Toit remark on the abundance of small veins and
masses of quartz-orthoclase rock in the rock of the numerous dolerite
sheets of the Karroo; and state that it is difficult to distinguish thin
acid rock from the (metamorphose<l) sediments cut by the sheets.'
From the Natal dolerites Prior has descril)ed a "hybrid rock" collected
by Anderson who labelled it "basalt which has absorbed granite*';
its essential constituents include biotite, augite, and micropegmatite.*
Prior regards the effusive rhyolites of the Lebombo Range in the
Natal-Zululand doleritic region, as well as the granophyres of the intru-
sive bodies, as probably differentiation products "of the same magma
which supplied the dolerites.'' There is thus something to be said for
the view that the acid rocks, both intrusive and volcanic, associated with
dolerites in the vast South African field, are of secondary origin.
But the most remarkable intrusive l>ody in South Africa is the
Bushvehlt laccolith of the Transvaal (Fig. 162). It is 250 miles in
length and 75 mil(*s in width. It is <'ompo.se<i of red, micropegmatitic
or granophyric granite, gabbro, noritcs pyroxenite, and iron-chromium
ores. The ores and ultra-femic rocks and ores occur at the floor
contact, the granite at the roof, and the norite-gabbro phases in an
intermediate position. Their arrangement is strictly analogous to
that shown in the Sudburv sheet and there can be little doubt that
gravity has controlh^d the difTcrentiation of this Transvaal intrusion.
the largest laccolith yet describe<l by critical and competent geologists.*
No din'ct evidence as to the original nature of the differentiated magma
' F. H. Hatch and O. 8. Corstiirphinc, The Geology of South Africa, London,
1905, p. 172.
» E. T. Mellor, Trans. CuhA. Soc. South Africa, Vol. 7, 1904. p. 45.
> A. W. Ho^crri and .\. L. du Toit, 1.3th Ann. Rep. Geo!. Comm., Cape of Good
Hope, 1908, p. 10."). (' f. A. W. Hop-n*, l.'Mh Ann. Report of same, 1910, p. 9.
A. L. du Toit, KUh Ann. U«'p., IIMJ, p. 102.
* G. T. Prior, Annals, Natal Mu.-*«ufn, Vol. 2, 1910, p. 150.
* G. A. F. MolcnuraatT, Bull, poc ^M. France, Vol. 1, 1901, p. 13; Geology ol
the Tran.<<vaal, Johannrshur^, 1901, p. 49. H. A. Hmuwer, Oorsprong en Sameiif-
tellinR der TransvaaL><'he Ncphclicn-syenieten, 'h (Sravcnbage, 1910, p. 9. Theie
unusually valuahle memoir:! iUustrate the richneiu of the South African field
in the speaking fiicts of igneouri Ktroloi^y.
GRANITE CLAN
IGNEOUS ROCKS ASD THBIR ORIGIN
U\> nuA y<>< tiiins .AH aii.l lt>^ .>( the iwhl.ro UccoUth Rt F
('unmH-IJ.-ur. .Ahir li (■'. I^uiLihlin. Hull VJ-2. V. S CI. S.. 191-^.) QS, quart*
M-hJHt: (///. i|iiurt« :iii.l lii>r[ilil*-n<i>- xi-hiMp: US. hurnMcndr H-hiil; aoU btafk.
Slcrlini: i^uiiiti- k"'*''^'': ''< |i»l>l>rD: //<•', quHrli-lnirntilrnde nabbro. G Mul l/f<
:iri' jibiBi-- .i( (hi- Kiiin- l;i Iith; k.-iiit;iI ili-rniim' of ilciwity, from lb* Bt-or
ii|iw:irii, »lu>mi )>y vHryiiiiE I'lxM'iitiw of JiiIh.
GRANITE CLAN 353
has been published but the detailed similarities to the intrusive sheets
and laccoliths of British Columbia, Minnesota, etc., warrant the hy-
pothesis that the original magma was gabbroid, modified by the solu-
tion of the intruded acid rocks. The assimilation in this case may have
been largely accomplished at levels below the horizon of intrusion
(base of the Waterberg system of sediments).
Preston Laccolith. — The Preston gabbro of Connecticut is described
as a laccolith outcropping in an area measuring 3.5 miles by 6 miles
(Fig. 163). It has a low dip to the west and both roof and floor phases
are exposed. The body cuts quartzites and acid, sedimentary schists.
Tlie lower and greater part of the intrusive is an ordinary diallage
gabbro; the capping phase is a quartz-hornblende gabbro, carrying
accessory biotite (2-5 per cent.), with andesine accompanying labra-
dorite. The quartz often occurs in micropegmatitic intergrowths. The
two phases grade into each other and the quartz-hornblende gabbro
locally grades into an oligoclase granite, which forms segregations as
well as dikes cutting the more acid, main phase. Other, smaller
gabbro bodies in the vicinity likewise show free quartz at their upper
contacts. Loughlin explains the relations by gravitative differentia-
tion but offers no suggestion as to the cause of the abnormal composi-
tion of the gabbro as a whole. The hypothesis that assimilation of
the country rocks was responsible should be entertained.^
Medford Dike. — Many cases of obvious assimilation (corrosion)
of quartzose xenoliths by diabasic or basaltic magma have been re-
corded. Jaggar has made a detailed study of the marked corrosion of
quartz and quartzite xenoliths as exerted by the magma of the wide
Medford dike of diabase at Boston, Massachusetts. The mineral-
ogical effects are here special but again the interaction of magma and
inclusion has led to the crystallization of a micropegmatite composed
of quartz and microcline. The dike also contains irregular lenses of
quartz-microcline pegmatite which merges gradually into the normal
diabase. The facts presented by Jaggar show that there has not
been merely a solution of acid foreign rock in situ; on the other hand,
it is clear that the corrosion aureoles and the pegmatites have attained
their composition through a kind of osmotic transfer aided by the
gaseous constituents. Thus even here differentiation has tended to
mask the proofs of assimilation, yet the fact of corrosion is evident.
It is further important to note that the rock of this dike, while generally
a normal diabase, has a (basic) hornblende-augite diorite phase and a
hornblende-biotite diorite phase, suggesting analogous local changes
in the original magma.^
» G. F. Loughlin, Bull. 492, U. S. Geol. Survey, 1912, p. 78.
«T. A. Jaggar, Amer. Geol., Vol. 21, 1898, p. 203.
354 IGXEOUS ROCKS AND THEIR ORIGIN
Globe District Intrusions. — The intrusive diabase of the Globe
district, Arizona, contaias quartzose inclusions which are "conspicu-
ously corrodod and ombayod/'^ The diabase forms thick sills, chono-
liths, and dikes cutting Pah^ozoic and older quartzites, thick limestone,
grit, conglomerate, and acid and basic schists. The differentiate of
the average syntectic in this instance should not be expected to be a
highly quart zose rock or granite. As a matter of fact, the larger
diabase masses carry true syenitic phases interpreted in the field as
local facies of the diabasic magma.^
Swedish Cases. — "In the southernmost part of Sweden one
Jotnian area only occurs, the so-called ' Almes&kra group/ S. E. from
the southern end of Lake Wettern. This group is composed of white
and red quartzites, felspar-l>earing sandstones and arkoses, chocolate-
brown shales and, more subordinately, conglomeratic layers and red
calcareous sandstones. Dikes and beds of diabase are very abundant.
They are remarkable by the intense contact influence exercised on
the quartz-rocks, many times resulting in micrographic quarti-
diabases and other rock-varieties of abnormal composition, as is de-
scribed by Iledstrom. Fragments of the intruded rocks have also
l)een mon* or less affected by the diabase magma."'
Hogboin describes modifications of olivine diabase intruded, as
sills, etc., into sandstone and granite in Angermanland. Near its
contacts the diabase loses its olivine and becomes a quarts diabase
with well -developed micrographic structure. Hybrid rocks are
produ(M'd at the contact of the diabase and the older granite and
veins of granite (red in color as usaal in these circumstances) cutting
the diabase are int(Tprete<l by Ilogbom as salic segregations of the
acidified diabase.*
Scottish Intrusions. — (1. W. Tyrrell describes laccoliths and feeding
dikes of quartz diabases in the Kilsyth-Croy district of Scotland and
suggests that these rocks, like micropegmatitic diabases and gabbros
in general, ''owe their origin to the interaction of a normal basalt-
magma with a highly silicious country rock." He further notes that
*'the mode of occurence of this rock (micropegmatitic diabase or
gabbro) is also distinctive. It always occurs in thick massive, ver-
tical-sidtnl <likes. which sometimes continue for many score inilc;^
across country, and also as thick laccolitic protrusions from such
» V. L. K:insoin<\ (ilohc Tolio. U. S. ('.col. Survey, 1904, p. 8.
- F. L. Uansoino. Prof. Paper 12. V. S. Geol. Survey, 1903, p. 85. See psffe
30.') of the present hook.
» A. (;. Hod)om. Hull, r.eol. In.«t.. Up.sala, Vol. 10, 1909. p. 9; cf. H. Hcdstrftm.
Blad 5, ^er. Ala, ine<i HeskrifninR, Sveri^es Geol. Unders., 1906.
* A. G. Hoghoin, Cieol. Foren. Stockholm Fdrh., Vol. 31, 1909, pp. 361^-370.
GRANITE CLAN 355
dykes."^ Tyrreirs generalization as to the large size of each of these
intrusive bodies is one also forced on the present writer as a result of
compiling the geological literature, and it must be regarded as full of
meaning for the petrogenic problem. Only in the larger injections of
basaltic magma should notable amounts of silicious country rocks be
absorbed.
Stecher had long before stated the probability that the quartz
diabases of the Firth of Forth region owe their free silica to solution
of acid country rocks in normal diabase. He described the corrosion
of quartzose inclusions and quotes A. Geikie's statement of belief,
founded on field evidence, that solution of silicious xenoliths actually
took place in these intrusive bodies. Stecher fiu*ther notes Schroder's
conclusion that diabase has dissolved granitic material in the area
covered by the Falkenberg sheet of the Saxon Geological Survey.^
Intrusions of British Guiana. — British Guiana has notable illus-
trations of syngenetic diabase and granophyre. The dominant diabase
occurs in dikes, thick sills, and ''laccoliths," cutting very thick sand-
stones and conglomerates. It is generally normal in type (specific
gravity 2.93-3.17), but is in places transitional into micropegmatitic
diabase and true augite granophyre (sp. gr. 2.93-2.77). Free quartz,
microcline, and biotite, are developed. The "laccolith" at Mount
Roraima is about 2000 feet thick.' The types and conditions of
intrusion are much like those in the other extensive fields already
noted. Although Harrison expressed no opinion as to origins, it seems
necessary to postulate the same mode of formation for the diflFerent
igneous types. Harrison notes the rarity of olivine diabase in the
British Guiana bodies, a fact finding ready explanation if the primary
basaltic magma was here even slightly acidified by the solution of
sandstone.
Species Derived from the Syntexis of Non-sedimentart, Acid
Rocks.
In nearly all of the cases so far cited the granitic dififerentiate
is found in basic injections cutting acid sediments. Several authors
have independently suggested that the acid rock is ultimately due to
assimilation of the sediments. The solution and differentiation are
both aided by the presence of water in the intruded formations.
On the other hand, Gavelin has recently described ''magnificent
» G. W. Tyrrell, Geol. Mag., Vol. 6, 1909, pp. 363-365.
* E. Stecher, Tschermak's Min. und Petr. Mitt., Vol. 9, 1888, p. 193.
* J. B. Harrison, Geology of the Goldfields of British Guiana, London, 1909, pp.
22, 24, 92.
356 IGNEOUS ROCKS AND THEIR ORIGIN
remolting and ajwimilation phenomena" at the contact of gabbro
cutting the lioftahammar granite of Sweden. Quarts gabbro, diorite.
and regenerated mieropeginatitie granite have been there developed.^
One is reminded of the intimate association of granitic and
magmas in the groat Brefven dike, south of Lake Hjalmaren,
which cuts acid crystalline rocks (Fig. 24).* Mennell has found
convincing evidence that the material of the Matopo granite was
dissolved in the wide, doleritic dikes of Rhodesia. This author upeaki*
of ''splendid examples" of the melting of granite by the dolerite, and
of there Wing ''every gradation" between the granophyrea of the
region and an " obvious mixed rock." He suggests that, in gmeral,
the common association of granophyre with gabbro or dolerite ia due
to "admixture of acid and basic materials lieforc intniflion/' a deduc-
tion very similar to that reache<l by the present writer in the study of
the Moyie sills.' Molx^rg descrilx^d the development of micropcK*
matite, hiotite, and orthoclase in the olivine dial^ase of the Blekinge
district, Sweden, and explained this modification of the diabase by the
absorption of the country rock, gneiss.*
Syntexis in Fekders of Fissure Eruptions
The feeding channels for great fissure eruptions are character-
istically narrow l)ut each of their walls must have become inten.<«e!y
heated during the passage of large volumes of the erupted basalt.
The dike material left in the fissure should, therefore, show some
evidence of contact assimilation. In the writer's belief this expecta-
tion is matched l)y the discovery of almndant free quarts and micro-
[M'ginatite in the dikes (and asscK'iated gabbro sheets) which fed the
Eocene fissure eruptions of basalt in Central Washington. Solution
of the sandstone or otherwise acid walls of the feeding channels may
also l>e res[K)nsil)lc for the quartz ba.<^lt, exceptionally fotmd in the
Eocene (Teanaway) extrusive sheets of this region* (Fig, 70). Thr
feeders of the C'uddapah trap flows, in southeastern India, are lil
* A. (iavclin, Gool. Foren. Stockholm, Forh., 1010, p. 099.
« P. J. Holmqaist, Bull. Gool Inst., riiiv. Tiwalu, Vol. 7. 1906, p. 107; A. G
Hdgl>om, SvcrigoB Oeol. rndrrs.. Sf-r. (\ No. 1S2, 1899, p. II, and Bull. Grol
Inst, of rp{«ala, Vol. 10, 1010, p. IS; K. Winpr, Cieol. Fdrrn. Stockhobn, F6rii .
Vol. IS, 1SIM», p. 187.
> J. P. Mennell, Geol. Maf?., Vol. S, 1011, p. 10; Quart. Jour. Gcol. 8oe., Vol
66, 1910. p. 372; Cf. Amor. Jour. Scienco. Vol. 20. 19a5, p. 215.
* Soc rcfiTcnrt* in RiKst^iliusrh's han(l-h(x>k, 4th ed., 1008, p. 1250.
* (;. (). Smith, Mcmnt Stuart f<»lio, 1004, and G. O. Smith and F. C. Caflkifi*,
Snoquahnie folio, 1000, U. S. Gcol. Survey.
GRANITE CLAN 357
more acid than the extrusive traps and vary from norite, through
diorite, to micropegmatitic granite.^
In both these great fields, many, if not most, of the masses of
extrusive basalts are normal. They do not carry free quartz or
micropegmatite, nor are they more silicious than the average basalt
of the world. Since this common magmatic type is interleaved with the
more exceptional quartzose type, and since both types were erupted by
the same mechanism, in the same region and in the same geological
period, it seems in the highest degree improbable that the quartz-bear-
ing or more silicious basalts can represent a primary magma independ-
ent of the normal basalt. Between quartz diabases and normal
diabase, and between quartz gabbro and normal gabbro, there are
very often recorded the closest associations, in structural relations,
in mode of eruption, and in time of eruption. Here again must be
felt the great difficulty of crediting an independent origin for the re-
spective rock species.
Conclusions. — Without further multiplying illustrations, it is clear
that the phenomena of the larger injected masses, especially sills, are
of great value in any genetic theory of the granites. In each of
several large fields, where basaltic (gabbroid or diabasic) sheets have
been thrust into sediments, the thinner sheets and associated dikes
preserve throughout theil* original composition; while the thicker
sheets roughly, though far from absolutely, in proportion to their
thickness, have been chemically modified. This modification is always
in the direction toward a syntectic with the country rock — generally
quartzose sediments. As a rule, quartz diabases (Konga-diabase
type, etc.), quartz gabbros, or quartz norites, are developed. Only
occasionally has the sheet been large enough, that is, liquid for a
suflficient period, to permit of marked differentiation of these syntectics.
Then granitic layers are formed at the roofs of the magmatic chamber.
The differentiates are seldom or never precisely equivalent, in chemical
composition, to the average sediment assimilated; yet there is often an
unmistakable "blood relationship" between the igneous and stratified
rock. No other process than assimilation seems capable of explaining
this consanguinity and many field evidences positively favor this
explanation. Further, gravitative differentiation is obvious in the
larger sheets. Hence, in the writer's opinion the eclectic theory is
» T. H. Holland, Quart. Jour. Geol. Soc, Vol. 53, 1897, p. 406. Fermor states
that the rhyolitic lavas, locally found in the Deccan traps, are consanguineous with
the dominant basalt, regarding the two types as differentiates of the same magma
(L. L. Fermor, Records, Geol. Survey, India, Vol. 34, 1906, p. 148). Was this
magma a syntectic of primary basalt and acid crystalline rock such as forms the
walla of the feeding dikes?
358 laSEOUS ROCKS AND THEIR ORIGIN
woil matched by the foots of nature in the case of the only natural
bodies which can be examined fully ^ from top to bottom.
Transitions to Batholiths
Most of the instances of the differentiation of acid magmas from
basaltic syntectics have been selected from the list of sheet injections
which illustrate tlie phenomena. It is obvious that the same proce*i#»s
should control originally supiTheated basaltic magma in the case al<o
where it is intru(lc<l as a wide dike into silicious rocks. However.
we have already s(»en that very wide dikes are rare. Except where
dikes are the feeders of fissure eruptions, or of great laccoliths or sheet?*.
they must six*edily be chilled to temperatures too low for significant
assimilation. This seems to \)q a leading reason why dike diabase?*.
Kabl)ros, por[)hyrites, etc., so seldom show direct evidence of aci<li-
fication by the solution of wall rocks. In any case, since vertical
dikes have no true floors an<l, as ex[K>sed, have usually lost their sum-
mits l)v erosion, the mechanism of their magmatic differentiation mu?it
be incomparably less clear than in the case of well-exposed sheet.s.
Hecause of their indefinite depth, wide dikes are affected by the rela-
tively hipli tenifHTatures of the country rocks in great depth, thus
facilitating assimilation in the l)ody as a whole. For the same reason.
wide dikes ten<l to be charged with specially great amounts of magmatic
gases rising from deep levels. The.se gases promote both assimilation
and <lifferentiati()n. The facts as(>ertained in the sheets show that
pravitative difTerentiation should be ex[K»cted in these dikes. Ac-
con linp to the level at whi<h the erosion surface cuts the dike, the
arrang(»inent of rock typ<\s will vary. Near its summit an expecttnl
arrangement will be as follows: at the walls a chilled phase of either
the primary basalt or a more or le-ss acidified representative of it; in
the mid<lle of the <like, a silicious pha*<e corresponding to that at the
roof of a (lifTerentiated sheet or laccolith. This phase should tend
gradually to increase in acidity toward the middle of the l)ody and to
be transitional into the more femi<' marginal facies. The middle phae>e
will tend towanl an aplitic (Composition since the emanating magmatic
gases must be sperially concent rate<l in the middle of the dike.
In spite of certain complications of history, the gabbroid intru.<ion
at the Carrock Fell of the Miijjiislj Lake District seems to l)e an Ulu:^-
tration in point. Though Harker describes this body as a laccolith.
liis scM'tiun shows its <'ontacts to be steep and he regards it as prolmble
tiiat the mass has ncarlv \hv same attitude as that at the time of
crystallization.^ The body is 4 miles or more in length and averagi^
' .V. IIurkiT, Quart. Jour. (hm>1. Soc, Vol. 00, 1894, p. 329, and Vol. 51, 1805.
p 12G; The Natural Ilistury of Igneous Ro<'ks, New York, 1909, p. 133.
GRANITE CLAN 359
about 1/2 mile in width. However it is to be classified, it doubtless
extends to great depth and has the essential features of a dike so far
as these affect the present discussion. Harker has clearly shown
the orderly succession of the dike phases, from gabbro at the walls
with specific gravity greater than 2.95, through an intermediate zone
of gabbroid rock with accessory quartz and a specific gravity between
2.85 and 2.95, to a middle phase ^quartz gabbro) strongly charged
vdih micropegmatite and free quartz and characterized by a specific
gravity less than 2.85. Segregations of iron ore near the margins
represent local differentiates in the contact phase of the gabbro.
Harker has discussed the mechanism of the differentiation which he
regarded as the result of diffusion, but he makes no reference to the
control of gravity. The present writer is inclined to question whether
the essential features of this Carrock Fell body are not best explained
by the combination of assimilation and gravitative differentiation in
an original gabbro (basaltic) magma.
That intrusive body is described as only slightly older than the
neighboring intrusions of granophyre (micropegmatitic granite).
The consanguinity between the granophyre and the middle phase of
the gabbroid intrusive is obvious and Harker suggests the possibility
that all the rocks have been ''derived from different portions of one
deep-seated reservoir."^ The facts certainly suggest that the grano-
phyre is a gravitative differentiate formed in the deeper, less rapidly
chilled, and hence more differentiated part of the same dike fissure
occupied by the gabbro, or else in a neighboring wide dike. The
mineralogy and chemistry of all these rocks so perfectly match those
of many of the gabbro-granite sheets above described that the writer
has been led to postulate an origin for the Carrock Fell rocks in terms
of the syntectic-diflferentiation theory.
Similar associations of gabbroid or diabasic intrusives with in-
dependently injected micropegmatitic granites are recorded in Skye
and other parts of the British Islands, as in Ontario, Minnesota, South
Africa, etc., and their development is in none of these cases directly
seen to be due to the activity of superheated sheet intrusions. One
may suspect, however, that in some cases the syntectic process affected
^ide dikes, that is, injected bodies virtually like miniature batholiths.
Satellitic eruption from the dike chambers has generally complicated
the structural geology and petrology of these respective areas.
Origin of Normal Batholithic Granites
The secondary granites found in the differentiated sheets and
laccoliths are usually somewhat abnormal in chemical composition;
» A Harker, Quart. Jour. Geol. Soc, Vol. 50, 1894, p. 330.
360 IGNEOUS ROCKS AND THEIR ORIGIN
their fine grain and micropegmatitic habit is doubtiesB related to the
eonditions of comparatively rapid cooling in the presence of resurgent
water. The granite of a typical batholith has wall rocks chiefly c<Hn-
posed, not of water-laden sediments, but of pre-Cambrian gneiss and
other constituents of the earth's acid shell. The batholith must cool
with extreme slowness and a granular habit is therefore characteristic
of its visible rock. Nevertheless, the eclectic theory, as outlined in
Chapters IX to XII and XIV, holds that the normal granite of a
batholith is a true homologue of the secondary granite in any one of
the intrusive sheets above descril)cd. In those chapters will be found
an abstract of the argument favoring the truth of this homology, and
miditional facts Ix^aring on it are noted in the writer's previous writ*
ings. A new statement is not necessar>' in this place.
It should be noted that the eclectic theory specially relates to the
late pre-Cambrian and still younger granites. The older batholitha
of the Laurent ian type may have been formed by the same mechanism
as that illustrated in the younger batholiths; the field evidence that
stoping, on a large scale, has occurred in the former is clear. Yet the
universality and immense scale of the early pre-Cambrian intrusion and
its apparent independence of geosynclinal zones suggest that the
magmas of the oldest granites may not l)e differentiates of syntectics in
abyssal wedges of the basaltic substratum, but simply parts of the
earth's acid shell which was locally fuseii. Possibly the heat necesswy
for remelting was of radioactive origin.
This uncertainty as to the genesis of the oldest liatholiths illus-
trates the fact that the granite problem still lacks a complete solution.
An approach to it is made when it is recognised that the structural.
chemical, and time relations of the rocks in a post-Cambrian batbo>
lithic area apparently all accord with the postulates and inferences
of the eclectic theory as outlined in the foregoing chapters.
Granitic Magmas Diffkhkntiated from Magmas Belonouio to
Other Clans
Vogt states that the magmas of such rocks as monsonite, pulaskite.
and most diorites are anchi-iHitectics, implying that they are incapable
of undergoing notal)le differentiation. * But the field evidence clearly
sugge^^ts that these and some other magmas tend to split, with a
granitic type as the acid jM)le. The eclectic theory carries the dedue*
tion that dioritic and syenitic magmas are either syntectics or differ-
entiates of syntectics. It regards granite as the final acid pole of
the earth*s primitive different iat ion, whereby the acid shell and the
1 J. II. L. VoRt, Norsk Gool. Tidsskrift, Bind 1, No. 2, 1905, p.
GRANITE CLAN 361
basaltic substratum became separated. It considers the final acid
pole of splitting in large post-Keewatin batholiths as granite, which
has preserved its ancient "antagonism" to basaltic magma. In
the smaller abyssal wedges or in chambers satellitic to great batho-
liths, where relatively quick cooling shortens the period of liquidity,
the differentiation of syntectics is often arrested midway and rocks
of intermediate composition result. Under favorable local conditions
a part of such an intermediate magma is allowed to split further and
the final, granitic pole may be represented in phases of the parent
chamber or in apophysal injections from it.
This theoretical conclusion seems to be amply supported by the
facts of nature. Very commonly, diorite, quartz diorite, grano-
diorite, or syenite passes insensibly into granite in such a way as to
suggest that the granitic magma was a late differentiate from the other
magma. That the separation may be controlled by gravity is shown by
the superior position of the granitic rocks in the great British Co-
lumbia, Minnesota, and Ontario sheets and laccoliths just described.
Beneath the granites are phases of dioritic and granodioritic composi-
tion. An analogy is found in the Cnoc-na-Sroine laccolith at Loch
Borolan, Scotland, where quartz syenites (sp. gr. 2.625-2.635) with 12
per cent, of free quartz, overlie melanite syenites (sp. gr. 2.65-2.78).^
Influence of Resurgent Gases. — The local assimilation of special
sediments, notably those carrying large quantities of volatile matter,
may produce conditions for extreme differentiation in a part of a
plutonic chamber which in general is represented by crystallized rock
of intermediate composition. Thus, the extensive Bayonne batholith
of southern British Columbia is made up chiefly of granodiorite, while
several of its small satellitic stocks are made up of the more salic
biotite granite^ (Fig. 164). This contrast may be in part explained
by the fact that the existing outcrops of the granite are located near
the roof level in each stock, while the batholith has been eroded to
greater depth; but the stocks in question cut sediments of exceptional
thickness and the differentiation of true granite may be due to the
influence of water and other materials absorbed locally from the walls.
On this view, the stock masses are interpreted as nearly (hypidio-
morphic-granular) equivalents of aplite, which is generally agreed to
have been segregated with the special aid of magmatic gases. It
needs no emphasis that many aplitcs and pegmatites of granitic com-
position have been derived from bodies of dioritic, granodioritic, and
syenitic composition.
Clements describes an instructive parallel from a 4-foot dike in
1 S. J. Shand, Trans. Edinburgh Geol. Soc, Vol. 9, 1910, p. 376.
* R. A. Daly, Memoir No. 38, Geol. Surv. Canada, 1912, p. 301.
362
IGNEOUS ROCKS AND THEIR ORWIN
tlie C'ryntal Falls dulrict of Miclijgan. At both contaeti the dike
is a true <liorito, whicli mcrffpfl, toward the middle, iDto a tuotite
granite. TIic Knmll size an(] (-onscqucnt short mapnatic life of tbr
diko ext'ludos any possibility that the femic phase was here caused bj*
dilTusioii of ciTtain constituonta to the cooling walls. The magma
seems to have Ixt-n splitting during the act of injection bo that the
i<alic pule was tlinist centrally into the opening fissure already bearing
chilled, undifferentiated diuritc. The rapidity of the chemical change
is explicable on the supposition that magmatic gases aided io the
(upward) transfer of the granitic magma.'
Eruptive Sequence.— In composite batholiths and stocks tbe order
of intrusion, for the bathoUihic elements, is verj- generally from femic
to salic. (See Chapter IV.) The cr>-8tallization of one of these great
_2P Hta.
_20k.
Kni. Itil.— Se('ti.m lit tlio liuyniiiK- hutholilh, Britiiih Cnlumbw, il
lion lietwii'ii till- iliiiiiiiiuiit KTiiiioilionti- imil the true (cranile ileTFloped io tmlti-
litic ruiMibs. (It. A. Diily. Mi>iiiuir :W, <i(t>l. Surv. Canada. 1912, p. 302.? P.
Prir^t Kivtr tcmine; .V. Mutik iiietiiririllilP, etc.; ir, Wolf icrit; D, I>«<lnr>'
qiiartiile: /(, lli-chive qiiartjiili': /,, I^ino Star iirhbit.
IxKlies as ii wliole is elciirly not a continuous process. In eftch of very
many cases a more fciiiic phase siilidifies in large volume and is then
attacked and re|)luceil by more sulic magma, evidently rontained in the
same a)■ys^al eliaiiiber ns that in which the solid phaxe had formed.
itecause of the quite limited depth of erosion, outrrops showingeitbrT
]>Ii,ise must be relatively near the batholithic roof. It aeema likely,
ihen-fore. thai the suliditie<l ft luie phase wait originally a roof phaw
and that lieiieatli ii (he more salie magma iKK-ame differentiated by
gravity. Tbe fuel of extensive replacemont of the femic phaar is
fully sliowii in many fields and its mechanism is probably that of Bwr-
ginal corrosion euupled with mngnmtir ntoiung. In othv mdl tha
:K)lid feinic shell ai llie roof is not in "|iiilibriiim with ibc ^lill fluiAj
jwrtion of the abyssal W^jCv As diff- n nliulinrt eontinuw in diptl^
Ihe more salie |i[>le r ^^^glcoB ac'h-.i \- ^iU- and rraursrni
. M. C'U'inenlM, Jv ^f^J'*'-'^, P- >-
GRANITE CLAN 363
gases concentrated within it, has power to remelt part of the solid
hut still hot roof-phase. The classic experiment of Morozewicz gives
proof that gravitative differentiation follow such remelting. He melted
two pounds of granite (with 68.9 per cent, of silica) and left the super-
heated melt in a hot part of an active glass furnace for five days. It
was then cooled to a glass. The
lower part of the melt was found
to have 59 2 per cent, of silica;
the upper part, 73.65 per cent.'
Continued assimilation of the
country rocks in depth may com-
plicate the process but there can
he little doubt that a normal
bat holithic sequence is due chiefly
to differentiation within one
abyssal chamber. No other sup-
position can explain the detailed
(onaaaguinity in the successive
rock types. Several large-scale
illustrations of such relationship
have I>een studied by the writer;
a few of these may be mentioned.
Differentiation from Dioritic
Magmas. — The dioritic shells
commonly found along the con-
tacts of granit« batholiths have
already been explained as chilled
phases representing the mag-
matic state before the granite
itself was differentiated. (See
p. 244 and Figs. 114, 123, 124.) p-,o les.— Map of intruBive Btocka in
The femic shell is not necessarily the Crazy Mts., Montana. (After Little
continuous; among other irregu- Belt Mts. Folio, No. 56, U. 8, G. S,, 1899.)
larities it may show that due ^' Cretaceous, Livmgaton formation; 1,
, I ,. , , , Eocene dionte; z. Eocene quarts dionte;
to local caustic replacement by c, Eocene granite. Dikes shown by Un«,;
tbe gramtic differentiate. contact aureoles stippled.
Sometimes rocks of the gran-
itic clan are known to have been intruded in the same general
cmptiTe period as neighboring, more femic rocks, but no evidence
tatg be tX band that the two types are in any definite succession
0 rpckfl of a given region may, in fact, be quite contem-
3ret there is a possibility that the acid rock is a
hermak's Min. andPetrog. Mitt., Vol. 18, 1898, p. 232,
364
IGNEOUS ROCKS AND THEIR ORIOIN
differentiate of the more basic magma. An example appean in the
late Cretaceous or parly Tertiary intrusions found in the Clifton
quadrangle of Arizona.' Some of these are composed of diorite
porphyrite, which hero characteristically forms sills and laceoliths.
Others are composed of Kmnit'c porphjTy and quarti-monioDite por-
phyry, devclo|)0(I in Htockn or in apophyses of stocks. Compared to
sill or laccolith, the Imttomleiui stock has a greater supply of heat u
well as of magmatic gases risen from the depths. Both conditiotu
Fia. 166,— Mop of intnuive rUwVf. in the CsBtle Mta., MonUaa. {
tiH for Fiff. IC).) K, Mi-sninic fmliinenta; Jlf , metammphie BiiKola; D, Ktortat
(KobJiuon) iliorilc; (1, Neorcnc (Caatte) granUe; P, NeoMOC rhyiditc pocpbrry;
R, Npoccnp rhyoUtc.
tend to promote <lirrcrentiation and it is conceivable that tbe arid
porphyries arc tlius dcriveil from the original dioritic magnuL
Inmany other regions granite stocks and batholiths are surrounded
by satellitir liikcd, sheety, laccoliths, or irregular injections ot ibnite
in such relation that the smaller bodies are only to be interpreted ai
apophyses from the larger bodies. The latter, with longer magmatic
life, have lieen ilifTercntiated under the combined control of pAvity
and gaseous transfer, with granite as the salic pole near tbe loofa of
the great chomliers.
Weed h&s described striking ca.ocs in Montana. 71w Loco
■ W. LindgrcD, Clifton Folie U. S. Geol. Surrey, 1905.
GRANITE CLAN 365
diorite lai^ely composes the small Loco Mount^n stock aod the
much larger Conical Peak stock of the Crazy Mountains (Fig, 165).
In the smaller stock, covering only 4 square miles, the prevailing rock
is an augite-biotite diorite with no hornblende and very little quartz.
The Conical Peak stock, covering 20 square miles, conflists largely of a
quartz diorite, containing hornblende, biotite, augite, labradorite,
Fifi. 167.— Mopof the Cheviot district, England-Scotland. (After H. Kynss-
ton. Trans. Edin. Geol. Soc, Vol. 7, 1899, p. 390.) C, Carboniferous; A, andeeitce;
G, f^anite. Dikes shown by lines. The granite is int«iireted u a late differen-
tiate of the andeaitic magma (Kynoston.)
ortboclase, and quartz. The contrast in composition is of the kiod
expected if, as theory directs, the larger mass, with longer magmatic
life, had been more differentiated In each instance the erosion sur-
face is not far below the roof level and the more salic differentiate Is
there to be expected. The splitting went still further in the lai^r
mass, in which the quartz diorite is centrally penetrated by the slightly
300 laxEors kocks asd their origin
younger ( 'razy Mountain hornblonde f^anito. Weed states that tbL<
granite is '*apj)an'ntly tlu* aplitic phase* of the diorite."'
In tli(' same (|ua(lraiiKl(\ the Neocene monzc»nitic diorite of the
Castie Mountains, eovering ahout 3 square miles, lies close beside
the nearly, or (piite, e()ntenii)oraneous Castle i^ranitc covering 23
s^iuare miles. In spatial relation at least the small diorite maK^ is
satellitie to the nnieh -larger body of granite.* (Fig 166. See als«»
Figs. ."iO. 1 14, and pag(»s 105 and 384.)
The >yngene^is of granite and effusive andesite in the Cheviot
Hills of the Scottish Ixinler seems clear from their |i;eoIogical and
petrographic n^ations. The granite form^ a large stock cutting the
ande-^ite which (»vi<h»ntly n^presents an earlier phase of the magma.
before the granite was differentiated. (Fig. 167.)
Dififerentiation from Granodioritic Magmas. — In the Okanagan
composite hatholith of the (*ascade Mountains, the Similkami^en
graiUMliorite (with granitic, nionzonitic and <lioritie phases) is cut and
])artly replaced l>y the ('athe<lral biotite granite. The heart of the
main Cathedral IhmIv is itself cut 1)V a less femic but rloselv allied
• • •
granite in the form of a huge dike. Microscopic and chemical study
<'orroborates tin* already comp<'lling field evi<lence that these iKKlies
are all phases of nnv batholithi<* mass. The granite:? are, therefon*.
to be regarded as differentiates of a nuigma which was grano<lioritir
at an earlier stage. ^ Parallel cas<*s have lH»en recorded at intervals
in the belt of composite batholiths extending from Alaska to Mexico.
'Vhr intimate field :isso<'iation of typical granotliorite, quartz diorit«'.
and uraiiite in the Skagit division of the Citscade mountains has lieen
described bv tht» writer.'
«
The derivation of true granite from typical grano<liorite is many
tinn's ilhi>trate(| in the batholiths of tin* Western Cnitf^l State's,
r-spi'ci.Mlly in thox* of the Californian Sierra Nevada.*
An :irtii,Ml example of .^uch tliffen'utiation in place seems to In-
npn-^inted at the conta<-t of the great Trail batholith, near Rosslanii
ill >nnthi'Mi Hriti^^h ( *olund)ia. There the intru.sive rock is a granite
purpliyry. stmngly charged with l»asic .M'gregations reaching a fwit or
mure in diameter. The segn'gat ions have the composition of vogesite
^ W . II \\ I. l.inl. H.li Mnuijtain> Fi.lio. V. S. Ciol. Survey. 1809, p. 4.
Si v« T.!l i:immI I xifiiplj- (if prriplnTnl ilioritc foriiiiiiK the chilliwi rontart phaM*-
• •:" i:r;i!iit" nud j:rMiii»i|iMi iir I'MTlmliths ;irr He.srrilM'il by the writer in Memoir Ni»
;;n. < I«-m1. Smiv ( -iiiailM. P.M-. |>:ii;«-«- "S."). ftr.
II A. I);il\. Mnrmir Nm ;;n. (\vn\. Siirv. CiinAdu, 1912, pp. 455-164 ami
570 17s.
K. A l):ily, ii.i.l.. i»p. :.;u :)ii».
S«'--. fi.r « x:iniph'. tin- (*nlf:i\. Hiilwrll liar, Tnirke**. JarkJWfi. Nr\*ada City
nfnl Pvraniiil I*r;ik Fijlios nf tin- \' . S. (w«nl. Survrv.
GRANITE CLAN
367
and appear to be the femic masses '* frozen in'' by the sahc magma
simultaneously developed in the local splitting of granodioritic magma. ^
Differentiation from Syenitic Magmas. — The syngenetic character
of many true granites and syenites is obvious to the informed petrolo-
gist. The sequence of their intrusion is quite analogous to that ex-
hibited in granite-granodiorite fields. The composite stock at Mount
Ascutney, Vermont, is a specially clear case, studied rather minutely^
(Fig. 64, p. 113).
The march of differentiation in the Kiruna district of Sweden,
through a syenitic phase to a '* quartz porphyry" (granite porphyry)
phase of still later eruption, is particularly clear, as shown in the
writings of Geijer and Lundbohm^ (See page 397 and Fig. 204).
Other instances of the intrusion of granitic magma after, but
closely associated with, syenitic magmas of the sajne petrogenic cycle
may be summarized in the form of Table XVII:
Region
Monsoni .
Earlier
intrusion
Monsonitc.
TABLE XVII
Later eruption
Authority
Predasso Monsonitc
Aar mamif Syenite .
Christiania Region
£keraund-Sog-
gendal dintrict,
Norway.
Bergen district, '
Norway. !
Thousand Islands, i
N. Y.
Adirondacks, N. Y.
Alkaline syen-
ites.
Monxonite and
banatite.
Mangerite,
monxonite,
soda-syenite.
Alkaline syen-
ite.
Syenite
Granite ; O. von Uuber, Jahrb. K. K. Geol. Rcich-
i sanst.. Vol. 50, 1901, p. 395.
Granite ', W. C. Brdgger, Videnskabs. Skrifter,
' Christiana, I. Math.-naturv. Kl., No. 7,
1895. p. 163.
J. Kdnigsberger, ErlAut. lur Geol. u.
Miner., Karte dos 58tl. Aare-massive,
i 1910.
Granite ' W. C. Brdgger.
Hornblende granite
and biotito granite.
Granite.
Granite.
C. F. Kolderup, Bergens Museums Aar-
bog, 189G, No. 5, p. 183.
C. F. Kolderup, Ibid., 1903, No. 12, p.
118.
Alkaline granite
H. P. Gushing and others, Bull. N. Y.
State Museum, No. 145, 1910, pp.39-41.
Granite (Morris I H. P. Gushing, Ibid.. No. 95, 1905, p.
Port Cold well Syenites.
type).
Granite
32G.
Belknap Mt« . ,. . .
New Hampshire
Tripyramid Mt... .
.Vcw Hampshire
Syenite.
Monsomte and
syenite
H. L. Kerr, 19th Ann. Rept. Bureau of
, Mines Ontario. Vol. 19, 1910.
Aplite I L. V. Pirsson, Amer. Jour. Science, Vol.
22, 1906. p. 507.
Aplite L. V. Pirsson, and W. N. Rice, ibid.,
Vol. 31. 1911. p. 288.
Here also, continued assimilation of acid country rocks in depth
must affect the composition of the main body of magma, so that local
^ R. A. Daly, Memoir No. 38, Geol. Surv. Canada, 1912, p. 348.
« R. A. Daly, Bull. 209, U. S. Geol. Survey, 1903, pp. 48-85.
* P. Geijer, Igneous Rocks and Iron Ores of Kiirunavaara, etc., Stockholm,
1910; H. Lundbohm, Guides des Excursions en SuMe, No. 5, 1910, 11' Cong. G6ol.
Internationale.
368 IGNEOUS SOCKS AND THEIR ORIGIN
granitic phases in any one of tlic ulwvc-namcd maaaes
differentiates of a preliminary syenitic phase.
Granites formed by the further splitting of ayeniUc
almo8t always rich in alkalies.
Granitic Aplites and Peouatites
It is unnecessary to dwell in detail on the theory of common ^ite
and pegmatite. The two are known to be syngenetic at thoiuands of
localities and may often be seen forming parts uf the same dike or sill.
They have crystallized from gas-charged magma, which may commonlr
be regarded as the residual mother-liquor of batholiths. Barker has
Fia. ItiS.— Smtiunof N'i<'k<>iriut«.Mt.,IlM]ley(li<t.,Briti^C<dumbia. (After
C. CiiniBoll. M<-iiimT2,(ii'iil-Siirv. Canada, 1910, p. 101.) 5, PolMwoie UmaMom.
arfcilUtr, anilqiiartcitc; U.iliorite tH>Tph)Tyaheeta;G,gnnodiarita; Ai^idhUtk),
aplitp. IllustralinK thr HefErcKution uf apbtic material at the roof of a bathoGtfc.
outlined the process by which this always Bubordioate phaae of a
bathohthic mngma is probably developed.' fSee page 246.) The
abundant water and other volatile materials in this aalie solution are
not merely " mineralizcrs" ; they also facilitate the upward tranter of
the quartz-fcldspnr cutcctic.
Aplites approximating that eutectic in compoaition bftve beea
formed in large intrusive IxHlics which have cryatalllied as gimaitr,
granodiorite. quuriz diorite, diorite, syenite, or even as quarta gabbro
or quartz diabase; and these aplites are recorded at the roofa of iUl»,
laccoliths, ehonoliths, stock.s, and batholiths, especially the laiL One
> A. Harki-T, The Xaliirnl Ilintory of the iKncoui Rocks, New Yock, WM, pp.
293 and 323.
GRANITE CLAN
369
of the best exposed segregations of this sort yet described is that
found by Camsell in the Hedley district of British Columbia' (Fig.
ItiS). His section illustrates the upward and outward expulsion of
the aplite from a granodiorite batholith. Weed and Barrell found
another strikii^ case in the Elkhorn district of Montana (Fig. 169).*
Similarly, the roof apophyses of the Monzoni granite carry over
76 per cent, of silica while the normal rock carries but 70 to 71 per
cent.* Most of the hundreds of sills and dikes, seen to cut the fissile
roof rocks of a pre-Cambrian batholith in British Columbia, are
Fio. 169— Sections io the Elkhom Mioing district, Montana. (Aftw W. H.
Weed, 22nd Ann. Rep., U. S. G. S,, 1901, Part 2, p. 444.) C, country rocks; M,
quarts monsonite and granite; A, aplite. Aplite collected at roof of tutholith.
either aplitic or pegmatitic, while the main mass is common biotite
granite (or orthogneiss).* It seems to be a fact that the pre-Cambrian
batbolitbs were more prone to give off these satellitic magmas than the
later intrusions have been.
However, one may seriously question the view tbat all aplites
I C. CamseU, Memoir No. 2, Geol. Surv. Canada, 1910, p. 101.
■ W. H. Weed and J. Barrel!, 22d Ann. Kept., Fart 2, U. 8. Geol. Surrey, 1001,
Plate 48.
• O. von Huber, Jahrb. k. k. Geol. Heicheanat., Vol. 60, 1901, p. 396.
* R. A. Daly, Guide-book No. 8, 12' Cong. G£oL Internat., Ottawa, 1913, pp.
128 and 222.
370 KiXEOrS ROCKS AND THEIR ORIGIS
and iK'gmatitos arc derivatives from definite magma chambers. Then*
is much to Im» said for the hypothesis that some of these salic rocks arc
due to what Lane has called ** selective solution." During intense re-
gional nu^tamorphism, (»specially of the dynamic kind, deep-seated
rocks, charged with much interstitial water, may reach the relatively Ion*
t(*miK*rature at which minerals corresponding to the quarts-feldicpar
eutectic go into solution with the water and other volatile fluxe^.
Such small, locally generated pockets, lenses, or tongues of fluid may
l)e driven through the solid country rock for an indefinite distance:
subsequently to crystallize ^^ith the composition and habit of the tru»-
hatholithic derivatives. It is thus quite possible that these particular
ro<*ks, though truly magmatic. have had no direct connection with
abyssal injections.
()hi(;ix of the Hhyolitic Typeh
Chemically the average* rhyolite, liparite, and quartz porphyry
are nearly identical. Kach differs from average granite in the sy.-'t*^
niatic way usual for volcanics and corresponding plutonics. (Set-
page 229 and Cols. 4-12 in Table II.) This contrast is most strikinjt
in the case of the granite clan and explanation is probably to U-
sought in the specially larg(» averag(» size of granitic magma chamlnT*
The length of the magmatic jw»riod for a batholith must favor it*
gravitative differentiation. As usual, the extrusive lava is drawn
from the upjM'r magmatic phase in the plutonic chamber.
Yet there are very numerous localities where rhyolitic (liparitir
lavas have not been directlv connected with batholiths but, on the
other hand, have issued from much narrower abyssal we<lges of xhv
type of shcM'ts, dikes, or central volcanic vents.
The s(»con(lary granite formed at the top of the Duluth laccohtb
has been extru<le<l to the surface as true rhyolite and a similar origin
in smaller bodies of injectetl, somewhat superheated l)asaltic (galv-
broid) magma is highly probable for other extrusive **red rocks" m
Minnesota. A thin rhyolite flow in the Purcell mountains seems to \k
a surface exi)ression of th(* secondary magma generated in the Moyit*
and other sills of the range.*
The soda liparites, comenditc^s, and quartz keratophyres aw
similarly contrasted with "alkaline'* granite, but these effusive type*
are seldom from large batholiths, or visibly transitional into granite.
It is probable that magmatic gas(^s (both juvenile and resurgent
have played an essential part in the formation of these alkaline
«'ffusiv(*s.
Our understaiKJiiig of the rhyolites evidently depends in part
» K. A. Daly, Memoir Xo. 3S, C.ih)1. Sun-.. Canada, 1912. pp. 211 and 219.
GRANITE CLAN
371
Oil the facts bearing on the origin of the granites actually seen to
he differentiated in thick intrusive sheets. If these granites are
accepted as derivatives of syntectics of basic magma and acid country-
rocks, it is easy to credit a secondary origin for some rhyolites. Since
superheat has obviously characterized the lavas of volcanic vents
both active and extinct, small amounts of wall rock should, under
certain circumstances, be dissolved. Two different possibilities are
open.
If the dissolved country -rock is acid, gravitative splitting may form
a rhyolite composed chieflj- of the oxides of that foreign material,
as in the differentiation of a Moyie sill.
L
.
1
r.
ii
J
-rrJr'T'r'A
T
a
1
M.
Fig. 170. — Plan of composite dike, Cir Mohr, Island of Arran. (After J. W.
Judd, Quart. Jour. Geol. Soc, Vol, 49, 1893, p. 545.) G, granite; A, sndeeite;
F, quartz febtite; P, pitchatone porphyry (like F, a quartz pantellerite). A, F,
und P are succnssive intrusions following the same fissure.
Or the dissolved material, (resurgent water, etc.) may act chiefly
as an incentive to gravitative splitting in the basic mi^ma. In
C'hapter XVII will be found a digest of the argument that augite
andesite is a differentiate of basalt. The ground-mass of augite an-
desite is rhyolitic in chemical composition. The question arises as
to whether the absorption of water or other foreign material may
stimulate the extreme splitting whereby rhyolite is generated from
andesite, itself a differentiate of the primary basalt. The same
result may be imagined in the case of a mica andesite, a dacite, or a
trachyte; all the more readily since granites in so many instances have
been differentiated from dioritic, granodioritic, and syenitic magmas.
In all cases gravity would affect the differentiates more or less em-
372 IGNEOUS ROCKS AND THEIR ORIGIN
phatically. Inasmuch as central vents are always small (seepage 120),
the volume of rhyolite formed in this way must be relatively smaU. A
moderately prolonged extrusion of the upper, acid magma must
cleanse it from the vent, whence may now issue flows of basic andesite
or regenerated basalt, representing the femic pole of the diflTerentiation.
More prolonged extrusion finally brings the primary basalt, from still
greater depth, to the surface.
Assimilation and differentiation in ever varying relative importance
may thus be responsible for the extremely common alternation of
rhyolitic, dacitic, andesitic, trachytic and basaltic flows at oentral
venia. Examples are noted in the table of Appendix B. Some
given in special detail are summarized under captions .-"Berkeley Hills,
California,'' '^Cioldfield District, Nevada," "CUfton Quadrangle,
Arizona," '^Rosita Hills, Colorado," "Island of Skye," and "Eolian
Islands."
The syngenetic relation between rhyolites and andesites is further
shown in the features of certain dikes and other small bodies which
have been studied in detail.
Judd's account of the composite dikes in Arran is in point.' Figure
170 is a copy of one of his drawings. The dike is exteriorly composed
of two sheets of augite andesite; interiorly, of two rhyolitie plukses
nearly identical in chemical com|)osition. Analyses are given in the
following table.
ANALYSKS OF PHASKS OV THK CIR MHOR DIKE. ISLAND OF ARRAN
; Augite andesite Quartz felsite hvrv
feiO, ~ 55 79 " 75.31 ~ 72.37 ~
A1,0, 15.97 13.62 11.64
Fe,0, 12. ,50 2.31 1.42
FeO 1.08
MkO 2 22 .20 .62
CaO I 7.06 .97 1.30
Na,0 2 21 3.02 4.15
K,0 1.86 4 07 3.98
IIsO and loss on !
ignition | 2.43 1.48 4.88
100 (M 100.98 101.32
Other dikes of the region are similar but have dacitic and also ande-
sitic pitchstones between the wall sheets of augite andesite! **In
some cases the more acid rock (quart z-felsite and pitcbstone) was the
first ejected; but, (juite as frequently, the basic material (auglie-an-
> J. W. Judd, Quart. Jour. Geol. Soc., Vol. 49, 1893, p. 545.
GRANITE CLAN 373
desite) was the earliest to be intruded into the opening fissure. The
relative ages of the two rocks in the dike are shown, not only by the
positions which they occupy, but by the circumstance that derived
minerals from the older rock are found included in the younger one."^
Judd concludes that these magmatic types were syngenetic and became
separated in depth independently of fractional crystallization. It
would be difficult to find more telling illustrations of the march of
differentiation among these types and it is clearly from femic to salic
in the rocks actually mapped. Harker holds that the augite andesite
and pitchstone "can be explained only on the supposition that the two
are complementary products of differentiation of one magma."* The
present writer believes that the andesitic, dacitic, and rhyolitic phases
may be yet more readily considered to represent the successive, more
acid poles of differentiation, the corresponding basic pole or poles
being in the depths and invisible. On this view, the andesitic magma
was not complementary to dacite or rhyolite but was the parent
of both. The hydrous condition of the middle phase (pitchstone) of
each dike suggests that water-gas actively co-operated in the formation
and upward transfer of the salic magmas.
* J. W. Judd, op. cit., p. 561.
* A. Harker, The Natural History of Igneous Rocks, New York, 1909, p. 324.
CHAPTKH XVII
DIORITE CLAN
iNCLrDKD Species
The species recognized by Hosenbusch b» composing the di<Kitf
family and its dike and extrusive equivalentis include the foUowinft:
Plutonic Tyf}es
Mica diorite, mica-hypersthene diorite, yentnite.
Quartz-mica diorite, tonaiite.
Hornblende diorite, ornoite.
Mica-hornblende diorite.
Augite diorite.
Quartz-augite diorite, augite tonaiite, quartz-hyperethene diorito.
anden<liorite, banatite.
Dike Typen
niorit(» porphyrite. augite diorite porphyrite, hornblende-mica dio-
rite i)orphyrite, mica diorite porphyrite, hornblende dioritf
porphyrite, vinllite, i)aleoandesite, esterellite, microdiorite.
Tonaiite porphyrite.
( larnet porphyrite.
Quartz diorite porphyrite, c|uartz-mica diorite porphyrite.
Paleophyrite, ortlerite, suldenite.
Diorite aplite.
Tonaiite aplite.
Effusive Tyjpcs
Mica andcsite. mica porphyrite.
Hornblende an<lesite, horiiblen<le porphyrite, asperite.
Pvroxenc-biotite andcsite.
Pyro-xeiic andcsite, augite andcsite. augite porphyrite, Garmeldite.
mijakilc, wci.sclbcrgite, olivine weiselbergite, hypenthene
andcsite. enstatitc porphyrite.
Hvaloand<»sitc.
Labradorite porphyrite navit(».
Propylites (in part).
Trachyandesitc (?), ciminite (?j.
374
DIORITE CLAN 375
Typical granite magma cannot well be considered as a direct
mixture of any other known magmatic type with foreign material,
either liquid or solid. Granite is often quite clearly seen to be a diflfer-
entiate*of more basic magmas. The greater part of the visible gabbro
or basalt is interpreted as '* primary^' material; if its magma is a dif-
ferentiate, the splitting took place before the earth's crust was com-
pleted. The species of the diorite clan are of intermediate composi-
tion and, a priori, they may be expected to include both differentiates
and syntectics. That this judgment is probably correct will appear
after a review of certain cases has been made. The genesis of some of
the species presents complex problems which, because of the lack of
observational data, no one has yet attacked in detail. It is impossible
to list the unmodified dioritic or andesitic syntectics as contrasted
with the differentiates also belonging to this clan, but very probably
the great majority of its members represent differentiates of primary
basalt or of its syntectics.
Andesites
Augite Andesite. — The writer has published a quantitative study
of the obvious and long recognized hypothesis that augite andesite is
a differentiate from basalt.^ Some paragraphs quoted from that paper
will serve to lay the case before the reader.
Petrographers are in general agreement as to the existence of
many close mineralogical and chemical similarities between augite
anrlesite and basalt. It has, in fact, been found to be impossible to
draw any sharp line between the two species. Nevertheless, the olivine
basalts, volumetrically the most important class of lavas on the globe,
are distinctly characterized by the great abundance of the basic
phenocrysts, augite and olivine, with which basic plagioclase and much
magnetite are regularly associated as minerals of early generation.
The list of phenocrysts in augite andesite normally includes the py-
roxene and an average plagioclase which is more acid than that in the
olivine basalts; olivine is absent and magnetite is less abundant than
in the basalt.
As a result of numerous experiments on artificial basic melts and
on natural lavas, as observed under the microscope, Doelter has
proved that olivine, augite, magnetite, and plagioclase crystallize in
the order which has been deduced from the microscopic study of basalt
by Rosenbusch, Zirkel, and other systematic petrographers.
According to Doelter, both magnetite and phenocrystic olivine
crystidlize from artificial basic melts at temperatures ranging between
* Jour. Geology, Vol. 16, 1908, pp. 401-420.
376 laXEOrS ROCKS AND THEIR ORIGIN
1200° and 1030° (\ The olivine largely crystallizes l)etweon 1200** an.l
1 135° ( \ ; the magnetite, largely between 1 195° and 1 100° C. The miiz*'
for ])heno<Tystic augit(* is 1100-9(K)° (\, with tlie most abundant cr>"-
tallization i)etween 1190° and 1100° (\ The range for labradorite
is 1125'^ 1075° ('. He observed augite phenoerysts devclope<l in
molten basalt at the range, 1085°- 920° (\; in molten limburgite at
1 150° ( '. Magnetite formed abundantly in molten basalt at 1095* < '.
and in nujlten limburgite at various temperatures ranging from 1170'
to 1005° (\ For roek-melts he records only one determination fur
olivine, which "probably" crystallized out at 1085° C. in moltoD
basalt.
Throughout most of the period of phenocrystic development, that
is, through a fall of temp<»rature from 1200° to about 1080° C, basaltic
lava is still notably fluid. Other experiments b}' Doelter have shown
that strong fluidity characterizes various basic lavas at the folloi^ine
respective temperatures:
Etna biis.ilt. . 1010* C.
RomiiRon b;u«;ilt 10f>()
Vosuvi:in l:iva. lOSO
LinihnrKitt* . 1050
It is fair to conclude that at the temperature of 1050° ('. the avenurt-
olivine basalt is fluid, and at 1 KH)'^ ('. quite thinly fluid. At the latter
temp«Tatun* its kinetic viscosity is probably comparal>le to that of
the Hawaiian basaltic flow which Hccker has calculat<Ml to have hail,
at the time of its emission, a viscosity about fifty times that of wat*T.
That olivine and aupitt* phenoerysts are already formeil in highly
fluid basalt is suggested by an experiment rejKjrtecl (verlially. 1911
by V. A. Perret and K. S. Shepherd from Kilauea, Hawaii. By m<-an-
of a cabl<' and trolley, there instalh'<l by Profess^>r Jaggarof the Massa-
chusetts Instittite of Tcc'hnology, these observers ladled out of a
characteristic ''Old Faithful" fountain a mass of the molten Imsalt.
After simple chilling in the air the rock was found to 1)6 a black glas.«
bearing plienocrysts of olivine, augite, ami basic plagioclase. Fmni
the <M)ii>iderable size of the crystals it seems probable that they
existed in the lava lake and were not initiateii during the chilling
pnxess which however, may have allowed further growth.
A (|uantitative study i>r()ves that the phenocrystic olivine, augitr.
and magnetite of a crystallizing basalt fnnsl sink, provided the molten
li()U()r remains fluid. Observation shows that the lava in volcanic
pipes is ke]>t molt<'n through long periods. The fn^quently great length
of the crystallization interval is explaineil by the conditions ruling at
central vents, es])ecially Iwo-phnse convection. In the lower part of
DIORITE CLAN 377
the lava column, the juvenile gases tend to decrease the viscosity
and facilitate the settling of phenocrysts. However, as long as the
magma column as a whole is superheated fas at Kilauea), such crys-
tals must be remelted in depth and two-phase convection not only
remixes their material with the mother liquor but causes the repeated
return of all the magma to the surface. The differentiation of andesite
is thus only possible when the volcanic temperature approaches the
freezing temperature of the mother liquor.
In an active volcano the time allowed for the growth and sinking
of phenocrysts may be long enough for a complete differentiation, or
it may suffice only to remove some of the olivine and magnetite from
the cooling surface layer of the column, or it may be so short as to
forbid the growth of phenocrysts in the vent. Eruption will neces-
sarily arrest or greatly retard the process. Where the outflow is
rapid and continuous, the original olivine basalt appears at the earth's
surface. There, of course, the rapid cooling generally prevents
recognizable differentiation in the way possible, and apparently neces-
sary', in the vent itself where the basalt stands for a considerable time.
We have, then, to expect in nature a continuously graded series
of lavas from pure olivine basalt, through olivine-free basalt, to those
phases of the mother liquor which must approximate a basic augite
andesite and then an acid augite andesite. The last rock would thus
represent the one phase, the more voluminous phase, of this kind of
differentiation. In view of the notably uniform composition of olivine
basalts throughout the world, we must further expect that, in all cases
where the fractional crystallization has run a complete course, the more
acid pha.se should be relatively uniform in chemical composition. Its
phenocrysts form when the magma's viscosity is relatively high and
sinking is very slow.
The other products of the differentiation must also show a very
'great variation in composition. According to the special thermal
conditions and shape of each lava column, the phenocrysts must sink
to different depths and be segregated or dissolved in highly different
proportions in different levels of the lava column. From the original
olivine basalt many types of ultra-basic basaltic magma and of perido-
titic magma might be developed in the same conduit. During energetic
eruption or intrusion into the walls of the conduit these might become
mixed with each other and the resulting rocks present just such great
variation as is actually observed in the peridotite family. Many
peridotites, the picrites, limburgites (magma basalts), and abnormal
olivinitic basalts are, in this view, the rocks derived from the fractional
crystallization of olivine basalt, while augite andesite or allied types
represent the other pole of the differentiation.
378
lONEOVS HOCKS AND THEIR ORWJN
The prelimiDar>- paper contains a discussion of this hjrpotbem.
It appears to be subHtuntiutcd by the following facta, (a) The chem-
ical rcscnihlancp in rIoiH- tx-twccn typical or average augite uidesite
anil the glunsy base of an onlinary quenched olivine basalt, 'hi
Augiti? andenite and some rociis of the [jeridolitc-picrite group are
chemically reciprocal, their composition and volumes being exactly
those expected if they are gravitatively derived from basalt, {d
Various authons have observed the settling of the fenuc pbenocr>'!'t.s
of basalt in flow and sheet form. fSee Chapter XII.) (d) V^ain
de8cril)e« an actual instance of such differentiation shown when augite
s.
Fici. 171.— Miip <if Siroiiiboli lxlnii<l. (Aftrr A. llergnil, Abhand. k. bayrr
Akiiil. Wis;^., NEiUh-pliys. Kl . Vol. 20, \vm. Tnf- 0.) /, younxnt UmII; .'. .
li'uriiohiiAaiiiti' iwli<Tf<li'iii<>nstritt»l); .J. iildiT hiHitlt : !,. andesitir Uvaa and tull>
ut thr (.rii[in:il v<.l<-:.iiii> r..nf. C, rraKr. S.iil.v 1 ;liK,00».
undesite tloned fniiii the summit of the Iteumon volcano, while ultn-
fcmic btis.itts simullanenusly is.iiied from a fissure well down on it!*
tlimk.
These various arKUitieiil> Imld fcoiHl if the differentiaticHi is purriy
maKmalic. that is, if iis units :tre mil pheiKx-rysts but their fluid M|uiva-
lents. As inilicat< -d in ilir (iri>riii;d jiajHT. no one has yet proTecl that
thesplitlin^ tiikt-s plair l>y this (y}>e of liquation or by tnw fnotiaaal
crystallization. For tl»' purpose of making t ' dear aad alto
subject in sumo dctrni' to mat liemiit leal i eu > writ«r rbt>*)-
to omphn-^izi' tin' sinking of solid crystals wl '
DlOniTE CLAN
379
evitable. The exact mode of the difEerentiatioa is not critical in the
present connection. It can be safely left as an open question.
Numerous, appropriate associations of basalt and andesite on the
lai^e scale have been described by the writer in The Geology of the North
Fio 172 — Sect I \ 1 of F g 171 (Same ref p. 27.) 1, youngeet
baa<a 3 tuffs of nl Icr basalt c phase 4 andca tea and tufTa of origiiial cone.
f crater Scale 1 37 000
r
£^
<
y
v.
fz.
X^f
T8
t\
^1^ .
:^^
0
— -•«" ,r
Fia. 173. — Map of part of the EUensbui^ quadrangle, Washington. (After
EUensburg Folio, No. 86, U. 8. G. S., 1903.) TS, Neocene and Pleistocene aands
and silts; JVB, Neocene basalt; PA, Pleistocene andesite. Illustrating the close
li«ld connection between basalt and andesite.
Fio. 174.— Section along the line XK in Fig. 173. Scale, about 1 : 260,000.
American Cordillera at the Forty-ninth Parallel.* Other examples of
intimate field association between basalt and pyroxene andesite are
leeidled by Hfp. 171, 172, 173, 174, and 175. (See Frontispiece.)
ir Nl . Oeol. Surv. Canada, 1912, page 782, and other pages to be
e of contents of that work.
380 IGNEOUS ROCKS AM) THEIR ORIGIN
In addition to i\w mutt<*r ahstracted from the preliniiiiar>' papor,
an additional ])oint should Ik' notrd. Ry the proposed explanation.
aufz;ite andcsit<* is a lo\v-t('iniM*rature submagma generated, usually if
not ahvavs, under conditions like those at volcanoes of the eentral
type. II<*nce careful note shouM Ik» taken of the faet that, while
andcsite forms plentiful, generally viscous flows and yet more abundant
pyrodastic phases, it never composes great fields of plateau lava lik«*
that so oft<'n devel()j)e<l in the basaltic fissure eruptions. Andesite is
specially viscous not merely l^'cause of its chemieal eonstitution. In
all known <»ruptions it seems to have lacked superheat and the pro|M-r
explanation of that contrast is of manifest imfxirtanee.
Hjrpersthene Andesite. — In total volume and in geographical ex-
tension the hyiHTSthene andesites se(»m to have a much great<»r devel-
opment than the augite an<lesites. As shown in (ols. 47 and 48 of
Tabh» II, the two magmati<' types are almost identical ami the fore-
going argument may be applied also to t hese more voluminous ande^ite«.
Kosenbusch emphasizes the jMTfect transition iK-twwn the two tyj*-^
when studied ]>etrograpliically and it is often illustratetl in their clos^*
association in the field. Like the augite andesites, the hypersthen»*
andesites are sometimes transitional, mineralogically, chemically, an«l
geologically, into basalt with and without phenocrystic olivine. Wh»n
we further reflect that the eruption of hyjxTsthene andesites has l»e«'n
almost entirely restricted to vents of tlu' central type, the probal>ility
that they are g^'uerally due to the gravitativ(» diflferentiation of ba*>alt
becomes nearly as clear as in the case of tlH» slightly more basic augiti*
andesit(\ Th<» mineralogical ditTerence between th«» types is cxplicaMt-
on the assum))tion that the difTerentiation has usually proeee<Ieil a little
farther in the case of hyjMTsthene and(*sit<'. H3*|)ersthene rather
than augite would be expected in the more silicious different iate.
Mica Andesites and Hornblende Andesites.— Many workers in
an<h'sitic regions have illu>trate*l th<' very common transitions sul»-
sisting between the pyroxene andesites on the one hand and all the
other types of andesites on the other. Id<lings hits forcibly pres^entetl
the case for the lavas at SejMilchre Mountain in the Yellowstone N.v
tional Park, a region which he ha.*^ done much to make famous in |)e-
trography.* Some hornbh*nde amh'sites. like some mica-l>earinK anilt-
sites, are ch<'mically almost identi<'al with a typical augite andesite.
The average hornblmde ande<ite. like the average mica andesite, is
slightly more salic and less ferromagnesian than the average p>Tuxene
an<lesite. Thesi* diflVrenees, though systt'matic, are small and their
gene>is is obviously a problem of extreme delicacy. The contrasts may
be due in some ca<es to an advance in the difTerentiation of normal
» J. P. Mdiiigs, 12th Ann. Rep., V. S, (leol. Survey, 1892, p. 647.
DIORITE CLAN
881
basalt beyond the stage registered in hypersthene andesite. The
special addition of juvenile or resurgent water or other gasea to the
volcanic lava column may be the incentive not only to the separation of
a more salic submagma, but also to the crystallization of hornblende
and mica phenocrysts. The composition of the more silicious horn-
blende and mica andesites and their field association with acid diorite,
dacite, rhyolite, or other rocks, suggest that the former extrusives are
differentiates of magmas more silicious than basalt. The evidence
Fio. 176. — Sketch map of region embracing the Yellowstone Park, shown by
rectangle. (After J. P. Iddings, Quart. Jour. Geol Soo., Vol. 52, 1896.) A, ande-
ritic pyroclaatics and flows of Eocene and Miocene age (central eruptions); A,
Pliocene rhyolite oC the Park; B, Snake River basalt, fissure eruptions. Scale,
1:S,000,000. Illustrating rock associations expected on the eclectic theory.
as to the origin of those antecedent magmas can seldom, if ever, be won
from even the most intensive study of the volcanic piles themselves.
Their secret is to be found in the plutonic rocks syngenetic with the
andesites. As shown by Iddings in the memoir just cited, the actual
plutonic chamber can occasionally be located with reasonable certainty.
The genetic problem of the more acid andesites is identical with that
of some diorites. It will be noted in connection with the dioritea,
wUch, in the writer's opinion, are largely either syntectics of basalt
ind acid country rocks or differentiates of those magtnatic mixtures.
382 IGNEOVfl mCKS AND THEIR ORIGIN
DioaiTEs
Home ftUKitc <liciritcs may n-proscnt the bolocrystalliae phaw of a
I>yrux(-ii(''Uiiilfsit(.- miiKRia frozen in the Kfnprating volcanic vent or
injected into the country rocks .surrounding that vent. On the other
hiiiiii, tin- ficl<l rcliitiuii of many other dioritic IxMliefl do not bear uul
I)m> iile:i tluit the hitter reprcs4>nt direct differentiatPH of basalt. The
ei-leetie theory dcmiinds tlrnt the solution of the pre-C'ambrian granite
or ortlHiKneiss in [)rininry basalt shall afford the most important of all
synteet ics. This particular srnteetic sliould be of dioritic composition.
During solution at the main contact or at the contacts of xenolithx, the
vapor pressure is nearly the sjime in basalt and granitr (ortho)^cis.s).
TIh' syntectic film should have alnmt the mean composition of Uqui'l
and solid. The average comiHisitiun of the pri?-('ambrian terrane
down to a depth of several miles, is approximately given in the avcrac*-
calculated for granites of all jK-riods or that calculated for the granix"
of Sweden. The former average is entered in f'ol. i of Table VIII.
page 169, ("ohmm 2 shows tlie average composition assume<l for ibf
I)rimary basalt. The mean of those averages appean in Col. 3, whii*>
< "ol. 4 shows t he average diorile quite nearly. IxH-ninson-Leiwinff ol>-
jeets to the last average as not typical because some quartz diorite-^
were iiulucled in making the ealeulation,' The writer includetl these
analy.-^es, partly because s((me of the "diorite" analyses U9«l in the
ealeulation wire alinDniuilly basic and femic and lie on or over tin-
burder-line wilh the grili^rns. Seeondly, for purposes of geological
reasoning, it seinini f:iir tn iiiclnile the quarts diorites, since thene dii-
rites are so ufirii indi^MihiMy connected in tlw field. It really make-
little effect on I lie lalenlateil mean if the (|uartz dinrites are excluili-<l:
this will be -cen by eoinpariug (^ils. 44 and 45 of Table II.
The very el-.se eorn-^pomhwe of Cols. 3 and 4 in Table VIII
<.|)ows that the mutual Milutiim of etpial masses of primary l>8:<all ami
pre-CaMibrian granite or orthogneiss nuist produce an essentially dio-
ritic magma. I,oewiIl^cln•L^s^ing has claimed tliat themeanof fcraiiitr
and basalt is a syniitc. but this cannot Ix' true on account of the murh
higher eonlent of alkalirs and much lower content of hme, magnesia.
and inm oxides in nveragr -yeiiite. (See ( "ol. 17 of TaHe li.)
Since the ni'tabti' a.--iinil:ititiii nf nrthiigneiss and allied rockn l-^
piosiMe Kiily in larg<' itia-t--. of primary basalt, it is dear that ^orite
of synl.-etii- origin -liould l-e ihietly devi'lojM-d in aubjacent bodin, that
i-. in slocks .-mdliatholiih-'. Such is undoubtedly thfibct. Yol large
liritlioHth-iif tru'ilioriiiai'i-uiikiiowiiandthoseof '' > DriteaRrarr.
Tlie explanation lia- air-- ' ' -u implietl. In
' V. I^«'wiiis.,n.|,.s-.i..i-. . IMI.
DIORITE CLAN 383
gravitative differentiation, leading to the development of granitic types
at the accessible levels, will follow the period of active assimilation.
The syntectic itself should be expected to occur in two situations. It
might form satellitic bodies, like dikes, sheets, laccoliths, and small
stocks, all of which would chill too rapidly for differentiation. Secondly,
syntectic diorite might be looked for in the contact phases of a granite
batholith. (See pages 195, 365, and 384.)
Both deductions from the eclectic theory agree with the observed
field relations of the diorites.
That many of the larger masses of quartz diorite and diorite have
batholithic or stock form and mode of intrusion seems as clear as in the
case of the granites. A notable example is found in the gigantic
Coast Range body of Alaska, as described by F. E. Wright, C. W.
Wright, and others.^ Examples of dioritic stocks occur: at Mount
Ascutney, Vermont; in the Crazy Mountains, Montana; in the Brad-
shaw Mountains, Arizona; in the Globe quadrangle, Arizona; in the
Telluride quadrangle, Colorado; etc.
The evidences of intrusion by replacement, implying the assimila-
tion of silicious country rocks, are as striking for some dioritic stocks
and batholiths as they are for granitic intrusions. Abyssal solution
seems inevitable in both cases and the argument for a secondary
origin of the acid diorites has all the strength of that already detailed
for the granites. The spatial relations, so often observed between the
two species (diorites being either satellites or contact phases of sub-
jacent granite masses), are briefly described in the last chapter; no
informed petrographer needs a formal statement of the many known
instances in order to agree that the theory is here matched by the
facts of the field.
As above noted (page 356), Gavelin and Hogbom believe that cer-
tain small dioritic bodies in Sweden are of direct syntectic origin,
gabbroid magma having dissolved granite in place. A dioritic facies
in the trap-dike feeders of the Cuddapah lava floods has been interpreted
as a syntectic (page 356) . A similar explanation is possible for the mica-
diorite dikes found by Rogers among the dolerite sills of South Africa,
which have clearly dissolved acid sediments, etc.* (See page 360.)
Quite recently Miller has explained the diorites in the North Creek
quadrangle of the Adirondack region, New York, as syntectics of gab-
bro and its acid country rocks (granite, gneiss, etc.).'
Barker has pointed out that hybrid rocks are generally abnormal
products; "they cannot be correctly designated by names, such as
* F. JL W4| and C. W. Wright, BuU 347, U. S. Geol. Survey, 1908.
Greology of Cape Colony, London, 1905, p. 265.
#< . Geology, Vol. 21, 1913, pp. 177-179.
384 IGNEOUS ROCKS AND THEIR ORIGIN
quart z-<Iiorite, which belong to products of magmatic differentiation.'**
With this view the pres<»nt writer is in accord, but it is none the less
true* tliat some* intermediate pliases of the Moyie, Sudbury, Palisades,
and ot her intruded shei^ts, showing indubitable evidence of assimilation,
ai>pn)xiinate diorites in chemical composition. Of course these inter-
mcnliate rocks an* not din»ct synte<'tics but are phases arrested in th«*
process of differentiation and thus true analogues to acid diorites. Tlk-
typical diorites differ from them in just the sense demanded if the pan*nt
magma is a solution of (pre-Cambrian "acid shell") gneiss instead of
water-charg(Ml s<»diments. The considerable diversity of composition
in the diorite family is, however, explained by: (a) the variable nature
of the syntectic, depending on the different proportions of gneiss,
granite, schist, or sediment absorlw^d; (b) and the different degrws of
differentiation exhibited in individual tx)dies. The larger the abyssal
wedge, other things U'ing equal, the more of the "granitic " earth-shell
can it assimilate and the long(T is its magmatic life and period of
differentiation. Hence, for two reasons, the larger dioritic stocks
and batholiths should generally be more acid than the small stocks.
This (^xjK'ctation from the thecjry is fully matche<i by observed fact.
The correspondence is one of the major difficulties facing the pure-
differentiation hyiwthesis of igneous rocks.
Thus recognizing at least two quite different origins for the magnias
of the diorite clan, a direct chemical comi)arison of the corresponding
plutonic and extrusive |>hases is more <lifficult tlian it is in the gabbn)
clan or granite clan. It is true that average andesite is more saiic and
less cafemic than average diorite. and the dominant cause may again
1m' found in sjM'cial gravitative differentiation at volcanic vent«. Y<*t
the comparison between syngenetic meml)ers, plutonic and extru-^ive,
still ne<Mls to be made. The fact that the lavas of this clan have py-
roxene as their commonest femic constituent, while biotite is the com-
monest femic constituent in the plutonic types, has some definite
meaning in the gemTal problem of origins. It would seem that the
direct differentiates from primary biisalt predominate in the extrusivrs
of this clan, while the synttM'tics predominate among the intnisives.
This contrast is explicable by the root princi|)le of the eclectic theory
that the magmatic wedges f(>eding basaltic (and most ■ndfwitir)
volcanoes are initially narrow and hence t(K) little provided wHll heat
energy to dissolve much of Xhv earth's acid shell.
' A. Harkor, The Xatunil History of iKnoouH KockB, New York,
CHAPTER XVIII
GRANODIORITE CLAN
Included Species
Rosenbusch regards the granodiorites as being merely varieties of
quartz dio'rite and hence as part of the diorite family. Several writers,
particularly some in America, have advocated the recognition of a
separate granodiorite family, showing characters intermediate between
those of typical granite and typical diorite. The great importance of
granodiorite in the enormous batholiths of the North and South Ameri-
can Cordilleras justifies the latter view and it is a leading reason for the
^Titer's recognition of a distinct granodiorite clan.
As the North American Cordillera is studied, the granodiorites are
found to have the most intimate field association with the type called
"quartz monzonite" by Lindgren and other members of the United
States Geological Survey. It is probable that the rocks of the typea
called " quartz monzonites " by different authors working in the various
continents are represented in at least as many different bodies, though
not in as great total area, as the granodiorites proper. The published
chemical analyses of "quartz monzonite" are about as numerous as
those of granodiorite. But some of the so-called quartz monzonites *
(banatites, etc.) are clearly granodiorites and are often syngenetic with
them; in other part, the quartz monzonites are either granites, accord-
mg to the prevailing (Uosenbusch) definition of that term, or true mon-
fonite bearing accessory or quite subordinate quartz. Many tonalites
and some so-called quartz-mica diorites are chemically identical with
grutodlorite.
TaWe XVIII shows the close similarity of average granodiorite to
the nuui between average granite and average diorite, as well as to
avcragp to- 'it^ Columns i and 4 show how close quarts monsonite
I "f^^^y Js mite, but the average in Col. 4 implies the in-
' cJnsiooof
IGNEOUS ROCKS AND THEIR ORIGIN
TABLK .will
I
1
9
4
5
«
7
Aver-
Average
Moan of
Average
quarts
1 Average
graoodi-
orite
Averafc
ATcrac*
gran-
itfl
diorile
land 2
monso-
Dite
tonaUto
dadte
Number
of
236
70
20
12
S
30
analyses
averafpid
&iO.
70 47
57.50
64.02
67.41
65.82
67.20 ~
~67.«7
TiO,
.39
.85
.62
.61
.55
.S4
.33
Mfi,
14.00
16.90
15.90
15-76
15.99
1 14.71
1 16.81
Ferf)
1.63
3-20
2.42
1.93
1.66
1 2,38
2.47
FeO
1.88
4.46
3.07
1.96
2.69
1 4.16
1.35
MqO
.13
.13
.13
.06
.05
1 .01
1 01
MgO
.98
4 23
2.60
1.43
2.19
1.74
< 1.23
CaO
2.17
6.83
4.50
: 3.54
4.71
3.30
3.31
Na^
3.31
3 44
3,38
3.45
3.86
3.26
, 4.18
K/)
4.10
2.15
3.12
3.76
2-32
2.24
2.S3
P.O.
.24
.25
.24
.19
-16
.47
.08
100.00
100.00
100.00
100 00
100-00
100.00
■ 100.00
Column 7 shoWH the averagr dacito, the chief effuuve member of
this ciaa.
The granodioritc clan thu.'* incluclrs the fullowtng Bpedes:
Plutonic Types
Grenodiorite:^ of the provniliog definitioD.
Some "quartz monzoniten."
Tonalite.
•Some quartz diorites.
Dike Types
Granodiorite jKirphyry.
Some "quartz monzonitc" porphyries.
Tonalite porphjTitc.
Some quartz diorito porphyritca.
Dacite porphyrilos.
Effusive Types
Dacite, hyalodacitc, fclsodacite, biotito dltotla^ amptutiole daaU.
plagioliparite.
Quartz porphyrite, quartz-biotitc p
porphyrite, vitrophj'rite.
GRANODIORITE CLAN
Origin
In seeking a clue to the origin of these rocks, we note, first, that the
members of the clan have been reported only rarely from the greatest
of all igneous terranes, the pre-Cambrian complex. Occasionally a
tonalitic orthogneiss or a local body of granodiorite is mentioned, but,
wherever the workers in the pre-Cambrian have stated their petrog-
raphy in detail, the predominance of ordinary granites is generally
clear. Nor is it fair to believe that this conclusion is essentially
vitiated by the special failure of pre-Cambrian geologists to make re-
fined petrographic distinctions. As a matter of fact, those geologists
have long included a high proportion of the most ably trained and
enthusiastic petrographers.
We have already seen that many, perhaps most, of the pre-Cam-
brian batholiths have been intruded outside of areas of heavy (geosyn-
clinal) sedimentation. On the other hand, Paleozoic and later batho-
liths are almost entirely confined to such areas. The eclectic theory
implies that, where sediments are batholitiucally replaced on a large
scale, the chemical composition of both the batholitluc syntectic and
its more acid differentiate must be affected more or less strongly by the
solution of the sediments or of the thick basaltic or andesitic beds so
often laid down with the sediments. The suggestion is close at hand
that rocks of the granodiorite clan are differentiates from syntectics
containing considerable amounts of subsiUcic sedimentary material;
that these eruptives are, therefore, largely of Paleozoic or later date.
The writer was first drawn to that conception of the origin of the
granodiorite clan by compiling the facts recorded concerning the field
relations of the granodiorites, "quartz monzonites," quartz-mica
diorites, and dacitcs in the North American Cordillera. These erup-
tives regularly cut argillaceous rocks of great initial tiiickness. Through
close folding the mass of slates or other argillites has been notably
thickened in the zones affected later by batholithic intrusion. Like
tht argillites, occasional limestone formations and bodies of basic lavas
have been assimilated in abyssal wedges of first-class dimensions.
The evidences of replacement and assimilation are clear for the
Cordilleran granodiorites. Obviously, one cannot declare the exact
composition of the syntectics, which in depth will consist partly of the
pre-Cambrian gneisses, as well as of the sediments or basic volcanics
from wall and roof. Neither is the differentiation of these complex
wlutions to be traced in detail. In general, however, the abnormally
E syntectic should have an abnormally basic differentiate.
'Gvnodior te and its close associate, quartz diortte, do show
Ht] 1 which must be rated aB abnormal for batholiths
888 IGNEOUS ROCKS AND THBIB OBTOtM
aa a whole and dcviatini; from the normal (gruiite) in tbe direetion
throrctically drmandtHl. In nhort, fp-anodiorites seem to show aedi-
montar>' control on the larf^cst Hcalc
Nevprt holcKtt, even the C'ordillcran batholttha often indieate tint
granodiorite or quartz-mica <lioritc is not the final term in thrir differ-
entiation. In many places they patw inaensibly into, or are cut br,
granitr-s or salic "quartz monzoniten." Cliaptcr XVI conUuu refcr-
enro to the repratcd »ufu;eHtioa<i that these acid types are themselrai
difTerrntiates of granodioritic magma. Evidently the splitting has not
gone to the extreme more often, because the magmatic life oC each
hatholith wan limited. That life is dependent on heat supply. Is
these cases enormous quantities of heat must have been abeorfaed in
digesting the rocks replaced in great volumes. The same was true of
each hatholith composed of normal granite, but in it the syntcctie ni
initially more acid and the magmatic period needed not to be ao pro-
longe<l to afford a granitic differentiate. Thus, for two reasons, the
abundance of quartz diorites and mica diorites and the eompvatiTe
rarity of true granite in California, Waslungton State, BrittshColunibia,
Alaska, etc., are not surprising features of the great granodioritc field
of North .America.
The writer has not attcmptc<l to compile a full list of the reeordnl
oceurn'uep.1 of granodiorites and their allies, but a partial review of
maps and memoirs shows abundant field data supporting the idea tl
seilimentary control. So far a.<t the nature of the country rocks ii
conecrned, this support i^ evident in the regions bearing granodioritM
or quartz diorites, here named:
CoH-^t Hange hatholith (and satellites) of Alaska, British Colunbit.
Many ot Iht smulirr ma.-<ses in Ala.ska and Yukon.
Vaneonver Island.
Interior lianges and Interior Plateaus of British Columbia.
Many l>0ilics in Washington, Oregon, Idaho, and Montana.
Absaroka Kange, Wyoming.
Black Hills, South Dakota.
Several areas in Utah.
Many districts in ColoracKt, Nevada, New Mexieo, ■
Sierra Nevada and sateltitc.-;, California.
Coast Ilange of <'alifornia
Lower California,
Many dis*"^' ■■■ 'a Mexi
Southerr •"••mis.
>ra&sacl-
Chilwuf-
Larder
GRANODlOniTE CLAN 389
Sudbury, Ontario.
GleQ Coe, Scotland.
Adamello, Tyrol-
Urtini Highlands, Siberia.
Many districts in New South Wales, Queeasland, Victoria, and
South Australia.
These regions include nearly all the known volume of granodioritie
types and many of them contain the corresponding effusive, dacite.
The relation of dacite eruptions to country rocks fso largely covered by
the volcanics) is obviously often more obscure, but it is certain that
most of the dacites described from the American Cordilleras have been
erupted through relatively basic terranes. This is true, for example,
for many localities in the Northern Andes, where granodiorites do not
crop out.
The authors of the many memoirs relating to the Cordilleran gran-
odiorites have verj' rarely raised the point as to how the granodiorites
were intruded or that as to the origin of the magma. Ransome states
that fusion and assimilation of the invaded schists in the Mother Lode
<listrict have taken place "at least to some extent." More recently
Clapp ha.s concluded that the granodiorites of Vancouver Island (cut-
ting thick ba,sic sediments and greenstones) have been emplaced by
iMgmatic stoping and also show some evidences of assimilation in
place,' Caimea explains the intrusion of the Wheaton District
^anodiorite fYukon Territory) in the same way.*
Emerson saw very clearly the evidences of batholithic replacement
by tonalitic (monzonitic) magma in Old Hampshire Coimty, Massa-
cusetts. He con.siders the tonalite stocks of the region to be partially
denuded domes of great granitic batholitha, "which have melted so
much of the gneiss and hornblende-schist into their mass that their
composition has been greatly changed, but which, penetrated more
deeply, would change to ordinary granite."* Though this "tonalite"
bas the chemical composition of mouzonlte, the genetic principle
stated by Emerson might conceivably also be applied to the Californian
granodiorites which are in similar cross-cutting relations to the basic
Kdtmenta and schistfl. Is it possible that the Cordilleran granodiorites
m syntectics (^ these country rocks with an initially granitic magma?
The writer is ioclined to answer the question with an emphatic negative.
KbeaiB were correct, we should expect the Jurassic and older
tm Nevada to be cut by satellitic dikes or other in-
I of the initial granite before the granodiorite magma
>fo. 13, Geol. Surv. Canada, 1912, p. 110.
>. 81, G«oL Surv. Canada, 1912, p. 76.
. 39, U. S. Geol. Survey, 1898, p. 310.
300 IGNEOUS ROCKS AND THEIR ORIQIS
liad, through asaimilation, reaclwd the observed levels in the euth'p
rrust. It m hardly possible that such vast intrusioiu of Acid ntagiu
affected the late Jurasitic Sierra Nevada without an eariy fiuuring o(
t he country rock$<. Ah a matter of fact, the late MeaoiMc eniptiont is
the Bicrra Nevada arc in the order of increasing acidity, from basalli,
f^bbros, and the closely related basic andesites, through granodiortte, to
granite and aplite. The satellites and the chilled contact phases of tbr
Fto. 17*1. — Mii|i i>n<) e(H-iion of part of the RoMburg quadraogls, On^
(After H.w.-lmr)( Kolio. No. 49, I". S, tJ, S., 1S99.) K, CretuMNM, HjrrtlsfofM-
tion (i-iinKltinirnitp, MunilstHnc nnd iihiilcK S, aerpentjiw; G, meUfaUn; A
liiicite; R. rhyoliti".
{Cranodiuritc." arc usually gabbros, dioriteii, or porphyritcs. In nrarlr
all of the many SiiTra Xcvadn folios of the United States Geola|kal
Survey tho sinlciucnt is repeated that the granodioritc locally nop*
into quiirtz diorite nr dioHte, less often into true gablvo. Tbs pjni-
one nndesite of Mtiunt Raker, u-ith tho aflsociated banlte, rate ■*
ft ternine rich in gramnliorito, whieli itself cuts ol ler bMilti of th*
roftion. These rehitions are typical of much of'
tru^tiun. (See Frontispiece.)
a C'ordiUerw 'M
GRANODIORITE CLAN
391
Similar relations as to geological structure^ order of eruption, and
phasal variation are characteristic of the granodiorites of Washington
State, British Columbia, and Alaska. In all these regions the field
facts seem to be best explained on the assumption that in every in-
stance the magma initiating the petrogenic cycle was of basaltic com-
position. That magma must have had great volume, of the order of
the hypothetical abyssal wedge.
There are few recorded cases where granodiorite has been differ-
entiated in sill or laccolith — yet rock of this composition is a principal
phase of the Sudbury sheet. Its origin offers essentially the same
problem as that of the likewise micropegmatitic granite in the same
body. (See page 347.)
In the Port Orford quadrangle of Oregon, Diller has mapped
large gabbroid intrusions which seem to have laccolithic or chonolithic
relations to the thick Cretaceous sandstones and shales and pre-
Cretaceous argillites and phyllites of the region. The gabbro has
CAMCCS HUMP (3300 Fi)
10
Mis.
_i Km,
Fig. 177. — Section of Mt. Macedon, Victoria. (After E. W. Skeats and
H. S. Summers, Bull. 24, Gcol. Surv. Victoriai 1912.) 0, Ordovician shale and
sandstone; G, granodiorite; D, dacite; 8f solvsbergite; T, anorthoclase trachyte;
B, normal basalt; AB, andesitic basalt. Illustrating field relations between grano-
diorite and argillites; also its intimate association with dacite and alkaline rocks.
dacitic phases. In the field Diller interpreted certain dikes in the
sediments as apophyses from the gabbroid masses. These dikes are
composed of dacite porphyry or of granodiorite.^
A similar association was found by Diller in mapping a very large
mass of metagabbro in the adjacent Roseburg quadrangle (Fig. 176).
It has concordant (laccolithic?) contacts with the Cretaceous beds,
here largely shales with sandstone, conglomerate, and limestone inter-
beds. One patch of Cretaceous strata, lying within the gabbro and
probably part of the roof, is cut by dacite porphyrite. The gabbro
itself is said to be cut by dacite dikes. ^ One is reminded of the analo-
gous positions of the red secondary granite in, and apophysal from, the
diabasic sheets of Minnesota, Cape Colony, etc. The acid rocks are
clearly differentiates and they have the relation to gabbro and sedi-
ments which is appropriate to the explanation here favored. Indeed,
* J. 8. Diller, Port Orford Folio, U. S. Geol. Survey, 1903, pp. 3-4.
•J. 8. Diller, Roroburg Folio, U. S. Geol. Survey, 1898.
302
IGNEOUS ROCKS AND THEIR ORIGIN
thpflc Oregon bcnlies seem to represent on a small scale the grand petro-
genie events signalized in the California or British Columbia batholiths.
An instructive parallel is found at Mount Macedon* Victoria,
Australia (P'ig. 177). There the syngenesis of dacite and granodiorito
is very clear and both are in appropriate association with thick medio-
silicic sediments. The alkaline eruptives of Mount Macedon are tjrpeo
explained, in the next chapter, as also due to sedimentary control.
CHAPTER XIX
SYENITE CLAN
Included Species.
st petrographers are following the lead of Rosenbusch in recog-
within the syenite family a "lime-alkali" or subalkaline series
s and also an alkaline series. In the present chapter we shall
elude the monzonites and their extrusive equivalents in the
! clan, without necessarily taking sides on the question whether
monzonite is a member of the syenite family. The aim of
)rk is i\ot to present a systematic classification of rocks but to
size genetic relationships. From this point of view the in-
1 petrologist will hardly take exception to the plan of connecting
tely the monzonites and latites on the one hand with the syenites
ichytes on the other.
3 following list of the clan members has been compiled from
)usch's work.
Plutonic Types
lalkaline Clime-alkali) syenites.
Mica syenite, durbachite.
Hornblende syenite.
Pyroxene syenite, diopside syenite, uralite syenite,
caline syenites.
Nordmarkite.
Pulaskite.
Hedrumite.
Umptekite.
Sodalite syenite.
Riebeckite syenite.
Arfvedsonite syenite.
Analcite syenite.
Nosean syenite.
Alkaline pyro: e syenite, akerite, hype
hombl< 3 akerite, aegerite-augite si
syenite, laurvikite.
boiiomtes.
>mte proper.
804 lONEOVS ROCKS AND THEIR ORIGIN
Olivine monsonite.
KcntHllenitf.
Dike Types
A. SubalkaliDC (limp-alkali) Tyi>eit.
Syenite porptiyry, hornblende syenite porphyry, aiigit« (d
opside) porphyry.
B. Alkaline Types.
Xordmarkitc porphyry.
Pulaskite porphyry,
L'mptekitc porphyry.
Akerite porphyry.
Aegerite syenite porphyry.
Ltturvikitc porphyry.
TonnJ)crgite, t6n.sbcrgito porphyry.
C Monzonitic Types.
Monzonitc pori>hyr}'.
Monzonite aptite.
Effusive Types
A. Subalkaline (lime-alkali) Types.
Orthophyre, biotite orthophyre, hornblende orthopbyn, p!
roxene orllmphyrc.
Ilypersthrno trachyte, biotitc-hypersthcnc trachyte, towcftniti
Some hyalotrachyti's.
B. Alkaline Types.
Traihyte proper.
Phonolitic traehyte.
Sotialile trachyte,
Hniiynite (nofean) trachyte.
Catophorito trachyte.
Kiiimekite.
.\rfvc<l:-onite trachyte.
Aegerite trachyte.
Ithomb-porphyry, kenyitc.
Keratophyre, Btntsi'hitc.
Some hy a lo trachytes.
Sanidinite.
C. I-atitic TyiM-s.
Latite. augite latite, tnotite latilB
Some trachyandes
Ciminite.
Vulsinite.
SYENITE CLAN 395
General Statement of Origin
The eclectic theory impHes that the members of the syenite clan
are differentiates of secondary solutions. The theory undergoes a
specially rigorous test when it is applied to the explanation of these
numerous species. In making it we shall first review the principal
deductions from the basal assumptions and then inquire as to the degree
in which those expectations are satisfied by the facts of nature.
1. Since basaltic magma is the primary solvent, we should expect
members of the gabbro clan to be often visibly associated with syenites
and trachytes.
2. Granting that the normal differentiate from the syntectic
of basalt and the earth's acid shell is a granite bearing much free
quartz, it follows that a syenite (like a granodiorite) is a differentiate
from a syntectic of contrasted composition. That syntectic must be
(lesilicated relatively to that from which a granite is produced. The
only abundant crust-rocks available for such a syntectic are basic
sediments and, in a ^ar less degree, basic igneous rocks. The theory
seems, therefore, to demand that generally the members of the syenite
clan shall show close field association with basic sediments or with
basic igneous rocks, or with both classes of material. It is not essential
that these rocks shall crop out at the same level as the erupted differ-
entiate; on the contrary, evidence should be sought that such country
rocks were once in contact with the primary basaltic wedge in depth
where their solution was possible.
3. Since a very large abyssal wedge, with average initial tempera-
ture, is likely to dissolve much of the earth's acid shell as well as the
basic material with which it may contact, its syntectic magma will,
in general, yield a granite or granodiorite by differentiation. The
eclectic theory, therefore, implies that the abyssal wedge from which
pure syenitic magma is differentiated is usually small. The rock-
bodies belonging to the syenite clan should be comparatively small in
both area and volume.
In Appendix C, the reader will find a table of the principal localities
where the syenite family and the family of trachytes and quartz-
free porphyries are known to be represented. The species of this clan
and many of the syngenetic species are listed in the second column
of the table. The third column lists the sedimentary formations
traversed by the respective syenite, trachyte, or porphyry. This
column is incomplete, partly because of the failure of record by the
observers who have described some syenitic or trachytic bodies.
Association with the Gabbro Clan. — A fairly complete review of
the literature regarding the syenite clan shows that its members are
27
396 IGNEOUS ROCKS AND THBfR ORIGIN
very ofU^n, if not generally, in intimate association with basaltic or
gabbroid types. In hundreds of areas trachyte and basalt comporv
volcanie piles and often form alternating flows. In scores of rcfioa<
syenites and monzonites t)elong to the same petrogenic eyclcs &<
adjacent gabbros, diabases, basic porphyrites, norites, or similar cbem-
ieal species. Not only are many known rocks intermediate between
gabbro and monzonite, lietween gabbro and syenite, between bault
and latite, In^tween basalt and trachyte; within a single body syenite
or monzonite is often transitional into gabbro, diabase, or anortbosite.'
WluTe gabbro and syenite of the same petrogenic cycle are intruded
at different times, the gabbro is generally the earlier intrusion. The
se(]uence is thus analogous to that characterizing granite or grsno-
diorite and gabbro, in like circumstances. The succession of plutonic
types is, here again, the same as that deduced from the eclectic theon*.*
In a considerable numln^r of instances syenitic intrusions have been
immediately preceded by syngenetic essexite intrusions. As will b*
more specially noted in the next chapter, essexite is best interpreted
as a modified gabbro. Like monzonite it passes insensibly into typical
gabbro. The oldest memtx^r of the composite forming Mount Ascut-
ney, Vermont, is thus composed of dominant basic diorite, often e^
si*.\itic in habit and merging into true gabbro. This stock was cut
by the nordmarkite-umptekite and pulaskite stocks of Ascutney and
Little .\scutney .Mountains, .\dam8 states that essexite is repre-
sented in all eight of the Monteregian Hills of Quebec (Fig. 178V
In at least six of them (perhaps all eight) the essexite is associated
with syenitic magma. Adams notes also that the larger ("more
easterly") masses contain proportionately more syenite and the
smaller (** western") masses a greater proportion of the essexite.'
This distributit)n of tyfM's is explained by the eclectic theory which
implies that assimilation and differentiation will both, as a rule
progress farther in a larger l>ody than in a smaller. The theory alM>
implies that there should Ik* no simple order of extrusion where basalt
and trachyte or hit it e have emanated from volcanoes of the central
type: just as there is no fixed sequence for rhyolite and basalt. The
reason that these extrusions are in more complex time relations than
* Kxainplf's :in' fouiwl in the Tcllnriili* (|ua<lrangle, Colorado, mapped by the
V. S. CH'olo^ical Siirvoy (p. 7 of the folio); in the* anorthottt« districts of Xotwat.
in th(* (li;ih:isir intrusions of the Shiniimo and (ilubc districtB of Arisoaa, as df^
M*rihni on pam^ 'J:{:<. 2 1:{. :VM'k and l(H».
•Coinpan- sr<|urnr«'s (Appendix H) at Mount Ascutney, VemMMilp TVipjrr-
aniid M<iuntain. Ni'w llanipshire; in the Adirondark Mountains, New York; Nortli
(Vntral Wisconsin: Southern British Columbia; PredassOp Tyrol;
region, Norway: Kkersund-SofCf^endal diKtrict, Norway; Kinma diatiielfe
' F. D. Adams, Jour. Oool Vol. 11, 1903, p. 251.
SYENITE CLAN
397
ihe corresponding intrusions has already been noted on page 288. So
ar as the records go, the trachytes of France, Germany, Western
United States, Australia, etc., do show caprice in their times of erup-
:ion in comparison with the basalts of the same cycle.
The Kiruna district of Sweden affords a remarkably systematic
sequence among rocks which have been interpreted as partly extrusive
md partly intrusive. The table on page 398 states average analyses
calculated for the Kiruna rocks as described by Lundbohm and Geijer.
^See page 453.). Column i shows the computed composition of pri-
nary basaltic m&gma (water-free). Columns 2 to 7 show the averages
i^
so-
so
Mis \
Km.
Orforo
I.-*.
\T^
Shefporo
i.a.a.e.
^Yamaska
GCHQNT
and?)
Bromc
1.3.9
NaON
$e->'^
en
Fig. 178.— Map of the Monteregian Hills, Quebec. (After F. D. Adams,
Jour. GeoL, Vol. 11, 1903, p. 241.) Showing rock occurring in each intrusive
body. /, nordmarkite; 2, pulaskite; 5, nephelite syenite and laurvikitic syenite;
^, monzonite and akerite; 5, essexite; 6y theralite; 7, yamaskite (jacupirangite).
^f the Kiruna rocks, arranged in the order of eruption except that
Cols. 4, 5, and 6 refer to the same great body.
The eclectic theory implies an indirect field association of the
syenite clan with basalt. If post-Keewatin granites, diorites, grano-
diorites, nephelite syenites, as well as syenites, etc., are generally
ciiflFerentiates of syntectics, syenites should sometimes be clearly
derivatives from the same abyssal (initially basaltic) wedge in which
any one of the other plutonic types has been generated. According to
the local variations in the proportions of basic sediments, basic vol-
eanics, or granitic fgneissic shell) material assimilated by the basaltic
wedge, the syntectic and therefore its polar submagmas should vary.
398
laSBOUS ROCKS AND THEIR ORIQIN
Syenites arc often transitional into granite, diorite, granodiorite, or
nppbclltp pyenite. In earh case independent bodies of these rock*
Khow evidence of cons^anguinity with adjacent masses of qrenilr.
Monzonite is not only interme<liate in cotDposition between gkbbro and
syenite; its intrusion is sometimes intermediate in time between Ihow
of Rabbro anci syenite iN-longinK to one petroftenic cycle.
liiiliuw fEiren-
.Syenite ^>-'^"' iJS" '^'
m.U , 60.60 SS.9* ' W 41
TiO,.
1 :i7
AVI
i.:i7
l.Sl '
1.22
I.4S
,38
Al/),.
Itt.lR
wm
14 20
14, n7
16.21
IS-AQ :
13 92
iw>,.
5.;(i
•I m
:t.l->
K 21
3.79
5 50
3 33
FcO..
a. 411
T, m
K IKI
2 '3o
2 26
2 51
I M
MnO.
w
.in
.14
.21
.22
.22
W
MrO.
6 4X
B 27
3,W
2 21
2.03
2-10
64
CqO.. .
!) ;t(i
in m
7 :«
.1 «»
3 SI
3.00
M
XmO.
:t.i()
■A fA\
-.72
» 20
6.28
e 25
5 49
K,0. .
l,4s
■ tW
.7(1
2 73
2,87
2.81
anh
H^.
tin
IKi
..M
.54
.56
69
P,0.
.4B
.11*
07
.22
.19
.20
u
CO,.
.fi4
33
l<K> IK) HK> :t!l UNI 01 100 11 100 12 100,12 »>
A Tew conrri'te illustrations will siiffien to point (he general truth
of the deductions.
In the I,ii Plata quadranftli', <'oIorado, moDZOoite is wihiy
transitional inio ■liii
1 n) Tiiiiir. Ttiih, (Knim Titilio SpccUl Folia, No. 6S.
SlKiwiiic MiiiTunir-- iinil iiionsonilo id reUtioaa whidi uw
M'tii- ilK'ory. (^('iiiiihriiin quartsile; L, (
,. <-\iru-lvi' iiii.lrsirf :iTi<l Litite; .V, ■
■fnl
Tlie olivine syiiiitc .if ( ripjilr < riM-k nierRes into
jw well as into pyroxi-iii-syt-nili.' :ind pyroxene g)
' I,. (.'. Griitiin, I'mf. i'iijHT N*
> of the U. 8. r«-ologieoi Sajwy.l
oiivinrgmbbro
Tb«*BniU|lit
SYENITE CLAN 399
intrusives'' of northwestern Wyoming are composed of quartz
syenite, syenite, diorite, and gabbro, all merging into one another.^
The Stinkingwater Peak body of the Crandall quadrangle is composed
of syenite accompanied by smaller bodies of monzonites and diorites.
The contemporaneous Crandall Basin intrusion is composed largely
of diorites and orthoclase gabbros. In the Elkhorn district, Montana,
diorite is transitional into shonkinite and quartz diorite.^ The gran-
odiorite batholith of southern Vancouver Island becomes monzonitic
or dioritic where it contacts with thick Paleozoic limestone.' The
monzonite of Tintic, Utah, has replaced basic sediments and is syn-
genetic with andesite as well as latite fFig. 179).
Sedimentary Control. — For the purpose of estimating the kind of
influence exerted on syntectics containing a considerable proportion
of sedimentary material, we may use darkens composite analyses.*
These are entered in Table XIX, where also the averages of eighteen
analyses of phyllites and of fifteen analyses of metasedimentary mica
scliists are entered.^
As compared with the average pre-Cambrian rock (the acid earth
shell), all these sedimentary types except the sandstones must exert a
desilicating effect on a batholithic syntectic magma. Other things
being equal, the salic differentiate from a mixture of primary basalt,
gneiss, and shale, phyllite, or limestone must be lower in silica than the
salic differentiate from a basalt-gneiss syntectic. An analogous
effect would be wrought by the incorporation of greenstones, chloritic
schists, or basic traps in a batholithic magma.
The possible variety of such mixtures is infinite. The conditions
for differentiation and its ultimate product must be infinitely varied,
at least within the chemical limits set by each mixture. The immense
complexity of the whole process clearly forbids its systematic analysis.
At present no more can be profitably undertaken than the untangling
of a few rules governing the differentiation. The matter is more
capable of discussion in the case of the limestone-basalt syntectic, and
the next chapter describes the writer^s attempt to trace the effect
of the solution of carbonate rocks on subalkaline magmas. That
example will be found to strengthen belief in the reality of a secondary
^Absaroka Folio, U. S. Gcol. Survey, 1899, p. 6.
* W. H. Weed and J. Barroll, 22d Ann. Kept., U. S. Geol. Survey, Ft. 2, 1901,
p. 446.
» C. H. Clapp, Summary Report, Geol. Surv., Canada, 1909, p. 89. In
Memoir No. 13 of the same Survey, 1912, p. 105, Clapp suggests a syntectic origin
for these contact phases.
* F. W. ClarVe, Bull. 491, U. S. Geol. Survey, 1911, p. 28.
*Tlioadyhit malyses averaged are to be found in Rosenbusch's ''Elemente der
O' ' -Hkvir' ; tion, 1910, pp. 561 and 630.
• t
400
IGNEOUS ROCKS AND THBIR ORIGIN
origin for the species of the syenite clan, however complicated may be
their chemical evolution.
Compoftitff
ana^yniaf
luMatopM
6.19
TAHKK XIX
(\>fIl|N>HitC
CoiIl|>08itG
Average
1 Average
\o. of
analvHirt of
•
analysis of
analysis of
amUyakof
sporiinem*
HamlHtoncH
shales
phyllites
mica schiata
averaged
2.W
78
18
16
SiO,
78 . m
58.38
67.1
59.7
TiO,
.25
.65
1.0
1.3
A1,0,
4.78
15.47
20.6
18.8
Fe,0,
1.08
4.03 !
5.5
2.9 \
FeO
30
2 46
4.6
4.0/
MnO
Tr.
Tr.
.1
.1
MrO
1 17
2 45
19
1.9
CM
5 52
3 12
.6
2.4
Xa,0
.45
1 31
16
2.9
K,0
1.32
3 . 25
3 7
3.3
H,()
1 <>4»
5.02
3 3>
2.6«
PiO*
.08
.17
* ■ > . . • •
.1
CO,
S
SO,
5 04
2 64
■ - •
« • ■ ■
.07
.65
CI
Tr. .
.. . ..
C, organic
81
l(N) 3(>
HN) 41
KM) 0
100 0
.81
.54
.05
7 90
42.61
.05
.33
.77»
.04
41.58
.05
.02
• • p ■ • ■
too 09
The sedimentary svntectic should illustrate the stubborn tendenry
of pota^ih and so<ia to unite with the maximum amount of alumina and
silica. In general, both oxides are present in amount sufficient for
the formation of the orthoclase and albite molecules. Hence the
greater volume of the differentiates from these syntectics are fehb-
pathic and free from feldspathoids (nephelite, leucite, sodalite, etc.).
The syenite clan is, in fact, volumetrieally much more important
than the nephelitic and leucitic clans combined.
Many species of the syenite clan illustrate a concentration of
feldspathic and other alkaline material. The common acgrcgatiop
of such material in pegmatite and aplite dikes, its observed interstitial
transfer into the roof rocks of intrusive bodies (e.(7.,adinole8) and into
miaroles, vugs, true veins, etc., are good grounds for belief that the
concentration of alkalies in many syenites and trachytes is partly
due to gaseous transfer. Quite recently Brauns has offered a ma*v
of proofs that the alkalies in combination with alumina have been
abundantly transfcTrcd upward in the magma chambers which fur-
' Organic matter inrliidfd.
' Loss on ignition inrludcHi.
SYENITE CLAN
401
lis bed the celebrated projectiles in the Laacber See volcanic breccias.*
Martins notes the excessive amount of gas contained in tbe tracbytic
nagma furnishing the trass and pumice of this region.'
Tbe eclectic theory assumes the justice of this principle wbicb
las been so particularly discerned by the Frencb petrologists. It
s a clear advantage of the proposed theory of the syenitic species
bat tbe volatile matter required for the segregation of tbe alkalies
3 necessarily abundant in the average sedimentary syntectic. Tbe
ollowing table illustrates the point.
Per cent.
Vater and loss on
ignition
Total
Composite
sandstone
1.64»
5.04
.07
Tr.
Compo-
site
shale
5.02
2.64
.65
Average
phyllite
Average
mica
schist
Composite
limestone
.77*
41.58
.05
.
.02
3.3
~3T3'
2.6
8.31
6.75
2.6
42.42
These relatively abundant "resurgent" gases must greatly affect
he chemical equilibrium of the syntectic, which already contaias
he juvenile gases of the primary basalt. Whatever effect tbe gases
nay have, it accentuates the direct influence of gravity on the mag-
natic solution. The transfer of the alkalies should generally be
ipward. Several kinds of evidence tend to corroborate tbese tbeo-
etical views.
As noted below, trachyte and its effusive allies are slightly more
:alic than the corresponding plutonic species. (See Table II.)
The dominance of hornblende and mica among the femic constitu-
ents of the clan shows, in general, the presence of magmatic water in
quantity at least ample enough to ensure the crystallization of these
ninerals. Miller has lately suggested that the alternating shells of
nonzonite, syenite, hornblendite, and ''bi-)tite schist'' developed at
the contact of a gabbro stock in Warren County, New York, are due
to tbe action of magmatic gases. This gabbro locally cuts granite
but is younger than the thick Grenville limestones and other pre-
C ambrian basic formations of the region. Are the gases responsible
1 R. Brauns, Neues Jahrb. fUr Miner., etc., B. B. 34, 1912, pp. 169-175; ibid.,
B.B. 35, 1912, pp. 211-218.
'S. Martius, Verhand. Naturhist. Ver. preuss. Rheinlande und Wettfaknu,
)8 Jahrgang, 1911, pp. 381-463.
' Organic matter included.^
402
IGNEOUS SOCKS AND THEIR ORIOIN
for these remarkable contact effects, of resurgeDt migiiiT^ Geijcr
concludes that the Xa^lx^rKct magnctitc-ore body u a pnoiunfttolytk
excretion of <liu))asic mnfrniu. The diabase occuni in dikes pwmllelnl
by iMTl liile-lH'uriiif; monzonitic and syenitic rocks, into which thr
<liuba.><e merges. The syenites and the ore are ittated to be two phaws
of the same process of differentiation. The pneumatolytie oiipD U
specially indicated by the development of tourmaline in the on-*
Heim has interpreted the kaersutite syenite of Karsuarsuk, Greenlmnd,
as formed under con<litions largely pneumatolytie* fFig. 180).
Further, the eclectic theory implica that the differentiation of a
syenitic magma itself m progressive. The upper phase in its chamber
Fio. ISO.— Sn-tioii (if iliki' »t Kiu-Huumuk, ClrrraUnd. (After A. Heuu.
Mnld. om (Inmland, Vu). 47. l»lii, p. 213.) 1, Sill of peridotite btmriac ^»rf^
AUfcite cryMlikU; 3. k.^rrHUlilc rork: .?, fine-grained feldipM' (orthoclaae?) nek.
should Iweome incresisingly salic. Since the upper phase is more
capable of ('rupti.>n into the levels reached by an erosion Burfsrr,
syngt'iictic intrusives of the syenite cliin ought generally to follow
each other in tin- order of increasing acidity.
\ few of the ni:iny examples matching this deduction will suffice,
Pirssun found that the monzonite of Yogo Peak stork, Montana, u>
cut by many iHkes of banatitii- syenite and quartz syenite porpbjT}-.'
The syenite of Higaud Mountain, Qucliec, like that at Grenville.
(Jueliec, is cut by more salic syr-nite porphyry and by quarts syenite
|H)rpIiyry.'' The monxunite jiorphyry of the Ilreckenridge lUstrirt.
I W. J. MilliT. Scii-in<'. Viil. :ifl. VSVi. p. <90.
! ['. (iHJ.r, (;.-..l, Kiir, .<t<.c-ktiMlni Ff.rh . V.il. 3.3, 1911, p. 28.
■ .\- ll.'ini. M.'.lrl.'l.l.-o-r .itr, ilnxikinl, Vol. 47, 1910, p. 219.
• I.. V. )>i^^s<ln. -JDili A[iii. [{.p.. V. S. (li-i.l. Sur^-cy, Pt. 3, 1000. pp. «H. SOt
» O. E. Lc Hoy, Bull. Ceol. tvx-. .\mcri«-a, Vol. 12, 1901, p. 377.
SYENITE CLAN
403
Colorado, is cut by quartz monzonite porphyry.' The syenite stock
of Mount Ascutney, Vermont, is cut by a stock of consanguineous
alkaline granite as well as by its own quartzose aplites CFig. 64).
The nordmarkite of ShefFord Mountain, Quebec, cuts pulaskite and
still older essexite (Fig. 181).
Statistics of Field Associations. — In genend, a sedimentary for-
mation is more extensive than the intrusive mass by which it is cut.
Hence the preferred explanation implies that most of the igneous bodies
from which these rocks have been differentiated shotdd still be in
contact with sediments of the right composition. The long table in
Fio. 181.— Map or Mt. ShefTord, Quebec. (After J. A. DreaBer, Ana. Rep. .
Geol. Surv. Canada, Vol. 13, 1902, Ft. L.) S, Lover PaLeoioic sedimenta (u^l-
titefl, etc.); E, essexite; P, pulaskite; N, nordmarkite.
Appendix C contains a summary statement of the way in wluch the
records tend to substantiate this corollary of the eclectic theory.
The table is not absolutely complete, but it includes referenoeB to
the best known rock bodies belonging to the syenite clan. These
bodies are grouped by districts of very different areas. Some of the
regions cited contain only one igneous mass. Others contun many dU-
tinct bodies which, however, for each region usually show similar relai-
tions to sediments. The third column shows the nature of the invaded
sediments, either directly observed in contact with the igneous maeaes,
or inferred from the structural geology as occurring in depth. In
> P. L. Ransome, Prof. Paper No. 75, U. 8. GooL Survey, 1911, p. 71.
404 IGNEOUS ROCKS AND THEIR ORIGIN
mnny, pprhaps) tnont, instunccx, the lint of sediments in contact could
Im- nutubly oxtondcil lij- more (Iftaiitnl study of each r^ion. Sonw-
timpH I>a.sir iKn<K>us formations are ulxo listed, since their aolution alM
miftht exert <leKilU-iiting effect on batholithic magma.
In 328 rexions or 77 per cent, of all those listed in the table (436),
the rule connertinK liosic sediments with species of the syenite dan
'\^ followed. Sufficiently complete, published data have not been
found for 92 regions. Special importance must obviously be aadgned
to the 6 regions (1.4 per cent, of the total number) where syemtie or
trachytic magma seems never to have contacted with older "deailicat-
ing" formation.^.
At firHt sight, the last-mentionctt cases would appear to negative
the proposed hypothetais of origins. It must be remembered, however,
that Reveral poKxi bit i ties must be excluded before the negative ii
final. (I) The stoping of sedimentary roef rocks may have mp-
plicd the appropriate material and afterward prolonged erooon mar
have removed the corresponding roof formations, leaving granite,
orthogncisif, or other acid wall-rocks in the existing outcrop. '2)
It K conceivable that in the depth-t of a solidified granite bathotith
arc large masses of sediments stoped down during the last monwots
of its mngmatic life when the magma was incapable of arMiiTnilating
the foreign material. An abyssal wedge later injected through tbe
sunken masses and the );rnnite alike may be chemically affected by
l>oth kindri of material. (3) In each case the poesibility aboold be
initially considered that the acid country-rock is merely a aU or
laccolith resting on sediments. This relation is very often illustrated
in the well exposetl, pre-( 'amlman, typical granites of Boutbem British
Columbia. (4) Lastly, the visible sills and laccoliths prove that
magmas nitgrule horizontally, often for di-ttancea of scores of miles.
This process shonUI control syntvctics as well as pHmary basalt or
different iates. I'nclcr certain con<litions, a sedimentary S3^teetie
may be moved a long distance laterally from the site of ita par«it
abyssal wedge and injected through a terrane containing no basic
sediments or basic volcanic materials. There is herein a poanble
explanation of the alkaline syenites of the Cripple Creek district of
Colorado, wliicb were intruded after the vast geanticline al the Colo-
ratio Front Ranges was ui>domcd, with necessary disturbanee of
subterranean magniiLs. It is not incredible that the Cripple C^eefc
rocks arc different iat<-s of sedimentary syntectics generated in whjwmX
wedges even as distant as those whore the Tertiary "iginM vf tbt
Lendville, Bni-kenridge, and Georgetown districts o ated.
Seeing, therefore, tliiit the eelirtic theory ^Vtod by Ibe
facts known alH)Ut the gri-al -■ ni the sy
SYENITE CLAN 405
and that there are several possible explanations of the apparent
failure of a few regions to support the theory, the writer is inclined
to believe in a sedimentary control as dominant in the making of
syenites and their chemical allies. The differentiation of these and
other magmas rich in alkalies may conceivably be due, in a few
instances, to transfer by juvenile gases locally and exceptionally
concentrated in subalkaline magma; but it is in general simplest to
regard resurgent ga<^es as more efficient in producing the actual results.
A blanket objection to the whole conception of sedimentary
control over magmas is conceivable. All of the ocean basins and
most of the continental plateaus are veneered with sediments. Hence
the table of Appendix C might be interpreted as involving no greater
degree of association between sediments and syenites, etc., than that
necessitated by the fact that these eruptives must usually penetrate
sedimentary rocks. This view, plausible on the surface, is certainly
not a satisfactory explanation. There are broad non-sedimentary
areas of the continents showing very numerous igneous eruptions
on the small scale and on the grand scale; yet, with extremely rare
exceptions, these igneous bodies belong neither to the syenite clan
nor to the other alkaline clans. The older pre-Cambrian eruptives
of each continent — bodies to be numbered by the million — are charac-
teristically subalkaline and they occur in extensive regions almost
devoid of basic sediments of greater age than the eruptives.
The world's granite and granodiorite batboliths generally are
cut by subalkaline dikes, stocks, etc., but rarely by syenite or other
alkaline bodies. The acid crystalline-schist terranes, of whatever
origin, are rarely cut by bodies of the alkaline type, though literally
countless dikes and other intrusives of diabase, conmion granite, por-
phyrite, etc., are to be found.
Plentiful examples are in hand, showing adjacent sediment-free
and sediment-laden areas, which respectively carry subalkaline and
syenitic or other alkaline eruptives of nearly the same age.
The consecutive Montana quadrangles, mapped in the Fort Ben-
ton, Little Belt Mountains, Livingston, and Yellowstone Park
folios of the United States Geological Survey, cover two such provinces.
In the Fort Benton and Little Belt Mountains quadrangles and
northern half of the Livingston quadrangle, the Tertiary eruptives
fstocks, laccoliths, sills and dikes) include monzonite, theralite, shon-
kinite, syenite, etc., cutting thick Paleozoic argillites and limestones.
South of Livingston, over a very extensive area, the Paleozoic cover
had been largely denuded before most of the Tertiary eruptions there
I; ih« afforded rock species of the subalkaline type — ^basalts,
es, rhyolites, gabbros, diorites, etc.
406 lUSEOUS HOCKS AND THEIR ORIGIN
Tht> many Tertiary eruptions in the California Kerrm Nevada
urp characteristically itubalkalinc (baaaltH, andeaites or rhjrolitc,
porphyrites, etr.) where they cut the great granodiorite batholitlH; thr
mappeil latitCH of tlic range nil seem to have insued in areas uoderiain
by thick baHie .se<liment^ of Mcsozoic or Paleozoic ace.
The great "Laurentian" l)Btholith.4 of Canada are mostly subal-
kahne, but lliey often Iiecome alkaline (syenites or nepbelite sjretutea)
where they cut thick masws of limestone.
Fennoscantlia is another extensive licld poor in seiUments but
dominated by subalkalinc eruptives. These arc replaced only loeallr
by alkaline types, which are generally a.-MOcist«d with likewiM loeal.
thick mo.-'ses of older, basic sediments or volcanics.
Differentiation of Syotectics in Place. — Again special attmtioB
must be given to the thick intrusive sh(>ets or laccoliths which, having
large volume and heat supi>ly, should oeco^ionally be capable at
moderate assimilation of intru<leil ba^-ie sediments. Such eases are
rare but they are unique in furnishing floored and roofed chamben
where magmatie processes can lie best understood. It is, therefore,
highly Hignificant that the differentiates of a few sheets, cutting doott-
nant argillitcs and other relatively basic terranes, differ from those
of the Pun-ell, Minne^'ota, and Sudbury sheets in exactly the Mttie
demanded by the eclectic theory.
\oble lias described, unfortunately without all desired details, a
great .sill, cutting the argillitcs, sandstones, and limestones of the
I'nkar group ^Shinumu ana) at the Grand Canyon, Arisona.' The
sill varies from t>50 to 1).')<> feit in thickness. It is chiefly composnl
of a typical olivine diabase.
"Fur utxiut n Imlf-inilc cnst and ucst of the Shinumo there ocniis in tbc
upper iMiri iif tlie ilialm^'c i-ill nliine llic upper contart a pink holocryrtaDiM
nick of nic<liuiii groin. The cimtucf uf this rcM-k with the o\-eriyinf Uw
sintes is sharp an<l wi-llilcliiinl. lh>wnwnril it appears to grsde into ite
niirnml diubasc, iitiil iiu ilfritiiic line uf contact
CBD anj-where be obKmil.
L'nfurtutiiitcly the writer diil not collect tmni^iti
one s|M'ciiiic(i fniai the middle uf llic pink r
^,._ _,:_■_ , •-'..pm.
men when cxainiiKil under the micni-copc
,-r«i
of mc<liiim texuire. lou^L-tiiiR uf nillicr fr
..-.Witt
sulionlinatc ipniDz ;iiid ;i ^iiiicwliiit ultcrc^l iV
munanij-.-iaii iiuutfBi «h>ft
was ina<lc out t-. li;ivc 1 n ■irii:in;dly a liomblcndi-. ><<ini« li the qu»rt'
dLipIaycI :t nii.r..j;r;iiihic arnmii.nient within
UwftM^^taMkiM
tyj'ical hcinililcndi-^yciiiii', ;ii.d is npiinrenUy
diffcr.iiii;.ii..n in pl.-in- witlii- ♦^•4iabaso sflt
siHJchiiciis across the iq.pu ^^^BdMgl
SYENITE CLAN 407
Noble refers to a parallel described by Ransome in the Globe
district of Arizona. The diabase here occurs as thick sills and large
irregular masses cutting slaty and sericitic schists, thick shales, quartz-
ites, and limestone.
"Within most of the larger diabase areas occur occasional masses of a
reddish, usually rather coarsely crystalline rock, consisting of red feldspar,
ragged prisms of ragged amphibole, and a little iron ore. The rock disinteg-
rates readily, and its field relation to the normal diabase is not easily made
out, although it seems to occur as segregations from the diabasic magma.
Under the microscope the rock is rather decomposed, but it is seen that the
dominant feldspar is turbid orthoclase or microline, associated with a smaller
amount of plagioclase, partly chloritized and epidotized amphibole, and a
little quartz in micropegmatitic intergroAvihs with the orthoclase. The rock
is in fact a hornblende-sj-cnite, carrying a little quartz, and its composi-
tion casts some doubt upon the hypothesis entertained in the field that it
is merely a local facios of the diabase.''*
Similarly, when the intrusive sheets of Minnesota cut dominant
argillit^s they are sometimes charged with syenitic "red rock." As
in the Shinumo sill and in the Duluth gabbro laccolith, these syenites
are specially developed at or near the upper contacts of the intrusive
bodies.
Since the syenites and their allies, like the granites, are obvious
differentiates, and since differentiation almost always follows assimila-
tion, we cannot often expect to see direct proofs of the solution of
countr^"^ rocks at visible contacts. However, Bastin and Hill hold
that the monzonitic magma of Gilpin County, Colorado, has absorbed
its (calcareous) wall rock.^ Quensel has recently suggested that
alkaline rocks studied in southern Patagonia may have been derived
from sedimentary syntectics.' Barrell postulates an assimilation of
basic sediments by andesitic magma in the Elkhorn district, Montana,
where syenite and quartz monzonite have been actually differentiated
after andesitic eruption.* It is significant that the andesites are
latitic! Miller concludes that certain syenites of the Adirondack
region, New York, are due to a synte^is of gabbroid magma with the
i Gienville gneiss-sediment series.^
k Conridering the rarity of the conditions, we must regard these
B, Mii'iHiiifc Bininmr Geology of the Globe Copper District, Arizona, Prof. Paper
iW'dHHlKGeol. Survey, 1903, p. 85.
AW ^^^Hilii and J. M. Hill, Econ. Geol., Vol. 6, 1911, p. 465.
11. Geol. Inst. Upsala, Vol. 11, 1911, p. 112.
Rep., U. S. Geol. Survey, Pt. 2, 1901, p. 525-526.
"•»logy, Vol. 21, 1913, p. 178.
408 tGNEOUS ROCKS AND THEIR ORIGIN
instances as highly significant. The repeated associations of basaltic
magma, basic sediments, and syenitic differentiates cannot well be
accidental.
Though the Kiruna syenites of Sweden are not in visible reUtion
to basic sediments Ccovered by igneous rocks or eroded away?), they
seem to illustrate the gravitative differentiation of syenitie magma
in place. The more ferric syenite is overlain by the more salic syenite
porphyry forming part of the same eruptive body. ^See pages 397
and 453.) The scanty data regarding the Beaver Creek laccolith of
the Bearpaw Mountains, Montana, lead to the suspicion that the
quartz syenite, monzonite, and shonkinice composing it are diffff^
entiates in place, but the mechanism of the splitting is not yet
worked out.*
Small Size of Bodies Belonging to the Stenitb Clan
The eclectic theory implies that every body of rock bdonging
to the syenite clan should be of comparatively small volume. When
formed in a volcanic vent, which is alwa^'s narrow, a sedimentarr
syntectic (and a fortiori its differentiate) cannot attain very largf
volume. First, the initial heat supply is small ; secondly, sediments fonn
the wall rocks of abyssal wedges only for a limited depth from the
earth^s surface. If, on the other hand, the assimilation takes pla^
in a very wide abyssal wedge, there may l>e a much larger supply of
heat, but the mixture of ''acid-shell'' material in depth is there likely
to he carried further. Hence, flows of trachyte or latite should always
be relatively small and syenitic batholiths rare.
How closely these deductions are matched by the facts is evident
in the statistics of Chapter III. Extensive fissure-eruption 6riJ*
of trachyte or latite are unkno\^Ti. The larger known bodies of
syenite, or of monzonite, while very few in number, are insignificaat
when compared with a first-class granite batholith.
A good case is to be found in the contrast between the huge quarts
monzonite-granite Houlder batholith of Montana and the very small
(probably satellitic) IxkHcs of syenite in the adjacent Elkhom district-
The batholith cuts basic se<liinents and pre-Cambrian acid goeis^
as well. Its differentiation has developed a quartsose rock. The
relatively minute volumes of magma splitting with the syenitic
IK)le w(Te largt'Iy enclosed in argillaceous limestones which could
locally dominate a svnt(M'tir in a small cliaml^er.*
' W. II. Weod iiikI L. V. Pirfi.*H>n, .\iiier. Jour. Science, Vol. 1, 1896, p. 361.
« \V. II. \Vco<i aiitl J. nuTTvM 2-2d Ann. Hep., U. S. Geol. Survey, Pt. 2, 1901.
p. olS.
SYENITE CLAN 409
Chemical Contrast of Plutonics and Effusives of the Clan
The various members of the syenite elan illustrate again and
a^ain the usual contrasts between the plutonic and volcanic phases
of the same magma. Columns i6, 17, i8y ig, 24, 25, 28, 29, 30y and
31 in Table II give average analyses for the corresponding pairs.
(See pages 229 and 401.)
The different pairs, excepting that calculated for the rare laur-
vikites, tell the usual story. Gravitative differentiation and gaseous
transfer together must be held responsible for the more salic nature
of the effusive types. The comparison thus strengthens faith in the
mechanism whereby these highly feldspathic, alkaline rocks have been
derived from basaltic syntectics.
CHAl'TKRXX
ALKALINE CLANS
Incmokd Species
Of the t<-n faniiliod of plutonic rockH recognized by RosenbuM-h.
fivf will l>c discu»>cil toRothfr, in their relation to the eclectic thruo'
of oriKins. They are: (I) the family of the nephelite feleolite) 8>*enit«^
and leueito syenites; (2j the family of the eiwexitra; (3) the funilr
of the shonkinites and tlierulites; (4) the family of the mi)tHouritr!< and
fergusiles: and (ii) the f:iniily of the ijolites and hekinkiniteo. lo
general these nicks, and their dike and extrusive equivalenta or differ-
entiates, arc rich in alkalies or luive been immo<liatoly derived from
magmas rich in alkalies. The whole group, comprising nearly 3tW
named species or aliout ime-third of the species named in Rowd-
buseh's Iiand-book, may be conveniently called the "alkaline" clan*.
R<B-enbas<h and other leaders in ix'trography have of course, u«J
the term 'alkaline suite" not as a chemical description of every roct
species included in this suite, liut rather to emphasiie the syngenM^
of the whole group in which leading typos are literally rich in aliulie^
Other types, like limburgite. nepheljte luisalt, melilite basalt, bBM-
nite. or tephrile. are not al>solutely rich in soda and potash, but tW
are so often and si> doscly associated with phonolitefl, etc. that our
cannot doubt a genetic cunneition. \ few authors have recently
criticized the ii'^e nf the term and have indicated the danger of coo-
fusion in the mind-i of immature students who do not realiie that M
is u-ied syndM.lically by Itosenbii-ich and others. However, its uw i*
likely to continue until a Ix-ttcr one is in^Tnted to empbasiie tbf
s>iigenesis mentioned, which is one of the most \-ital facta in petroloi?-
The name "alkalint- <'l!ins" is not altogether happy for preaent u».
since some highly alkaline ty[)es have lieen grouped with the granitH-
Tlie syenite clan itself may lie regarded ai; IxOonging to the "alkaline"
group, but on acmunt of its impitrtance, it has been aepantely
treated. Though iiira|i:ible of precise definition, the dewgnitiftD.
"alkaline elans." will ilo no harm if it l»e n>membered tbatit iili*^_
UM'd inirely as the ni<Ht available, brief name for a gn»l«;
group of rock faniilir-;.
On the other hand, it is likewi>e clear t
line sjx'cies are very <iften associated in origin
ALKALINE CLANS 411
obvious implication of the eclectic theory. Illustrations from the
petrographic record are given in this chapter.
The five clans now to be reviewed include the species named in
Rosenbusch's hand-book, as follows:
Plutonic Types
1. Nephelite syenite, eleolite syenite.
Foyaite, pyroxene foyaite, biotite foyaite, amphibole foyaite,
miascite, litchfieldite, mariupolite.
Cancrinite syenite.
Catapleiite syenite.
Eudialyte syenite, lujavrite, chibinite, kakortokite.
Urtite, monmouthite.
Sodalite syenite, tavite, naujaite.
Leucite syenite, borolanite.
2. Essexite.
3. Shonkinite, leucite shonkinite, mica shonkinite.
Malignite.
Theralite, analcite diabase.
4. Missourite.
Fergusite.
5. Ijolite.
Bekinkinite.
Dike Types
Foyaite porphyries, eleolite-garnet porphyry, eleolite felsite,
cancrinite porphyry, liebnerite porphyry, lujavrite porphyry,
leucite porphyry, borolanite porphyry.
Foyaite aplite, essexite aplite, alkali aplite, paisanite, dahamite,
lestivarite, heumite.
Bostonite, bostonite porphyry, lindoite, gauteite, maenaite, sodalite
bostonite, sodalite gauteite.
Tinguaite, quartz tinguaite, grorudites, leucite tinguaite, solvs-
bergite, allochetite.
Camptonite.
Monchiquite, fourchite, ouachitite, farrisite, mondhaldeite,
Nephelite minette, soda minette.
Alndite.
Effusive Types
hyalophonolite, apachite, leucite phonolite.
412 IGNEOUS ROCKS AND THEIR ORIOIN
Tracliydoleritc, kulaite, mugcaritc, casexite melaphyre.
Absarokite, shashonitc, banakite.
IVphritp, leucitc tpphrito, haUynite t«phrite, sodalite tephrite.
Basanitr, It-uritc l^asaniU', haUynite basanite, buebonite, b«saii-
toid, tephritoid.
Lcucititf, Icucitr bunalt.
Xi-phflinitp, ncphclite clolcrite, leucit« nephelinite, metUitc nrpb-
elinitc.
Haflynophyre.
Ncphclite basalt, hailynitc basalt, ncpbclinitoid, eudialjrte-nepli-
elite basalt.
Mclilitc bajjalt.
Limburgite, magma basalt, rizzonitc.
AiiRititc.
Vcritp, fortunit-o, jumiltite.
Orrndito, n->-omingitc, madupitc, promrsite.
Euktolite, poppaplite.
Sanukitc (boninite).
General Statement of Origin
Ah notetl in Chapter III, the alkaline rocks have been regardfd
by some as oonxtituting an "Atlantic branch." It is beeonunt
increasingly clear, however, that alkaline rocks are not confined t«
the Atlantic region of the globe, nor to the "Atlantic type of raut-
line." In 189C Ilarker suggested that alkaline rocks are kirally
dilTerentialctl where the earth's crust has licen subjected to one type
of mee)ianicul stn-ss, namely, tliat which specially affected the North
Atlantic basin in Tertiar>' time. In order to account fw alkaline
rm-ks in the Pacific basin, Harker holds that "an Atlantic as well u
a raeifie element of structure enters into some parts of the Pacific
basin." He further admits that even in Great Britain, one of tbr
two Atlantic provinces where the hypothesis was first imagined, the
Paeific branch is reprt-st-ntetl in all six of the known eruptive epoch*
il're-Cambrian to Tertiary), while the Atlantic branch is represented
in only four. In Great Britain, Atlantic types are of Ordovician.
Lower ('arlHiniferous, l'p[XT ( 'arlwniferous, Permian, and posnbly
IVvonian age: but Paeihe tyiN-s of these dates are also plentiful.'
Since this n-gion is taken by ."^m'ss to typify the Atlantic ^rpe of
erust.Hl movenunt Tertiary fault-bloeking and foundering — it is
sigiiitieiuit that the Tertiary Itritish eruptives are referred br ]~
• A. ll:>Tkrr. Th<- Nntimil Eli^txry i>f ihf lini«HU Rorka, Xe* S
ALKALINE CLASS 418
to the Pacific branch. Similarly, the Tertiary eniptivee associated
with the very extensive regions of nonnal faulting and subsidence, in
Iceland, the West Indies, the plateaus and Great Basin of Western
United States, etc., are chiefly or wholly <rf subalkaline, "Pacific"
types.
Hence, neither in Tertiary time nor in pre-Tertiary time has the
Atlantic region been characterized by dominant alkaline types. On
the other hand, the "Atlantic" type of cnistal structure is generally
accompanied by dominant subalkaline eniptives. Used even in a
metaphorical sense, the new terms are misleading. Perhaps owing
to a recognition of this fact, Harker prefers, in lus latest publication
on the subject, to describe the two branches as "alkaline"and "calcic."*
On a following page will be found an attempt at explaining the
pbenomena which have prompted the Harker-Becke speculation.
Rosenbusch has concluded that monzonite and essexite (both
commonly regarded as alkaline types) represent the purest undiffer-
entiated form of the telluric magma.* His published statement
probably refers only to specially basic monzonite, since average mon-
lonite shows strong chemical contrast with average or typical essexite.
(See Table II, Cols. 30 and 75.) Neither of these averages nor their
mean corresponds to the highly probable average for all igneous rocks,
reckoned quantitatively. This must approximate a true dioritt; or a
yet more acid type, if the calculation is to include only that shell of
the earth of which we have actual knowledge. (See page 168.) No
other quantitative criterion has been applied by Rosenbusch and the
brevity of his statement makes it difficult to see the grounds for his
conclusion, which is still lacking in cogent geological evidence. If it
were true, this conclusion and the associated pur^-differenttation
theory, would seem to imply a differentiation of granite, granodiorite,
diorite, gabbro, basalt, etc., from an original magma which itaelf 13
" alkaline" and, where visible, has been usually associated with magma
very rich in alkalies. The logical deduction would be that subalkaline
and alkaline rocks are not derivatives ot two magmas distinct "from
the foundation of the world," but are f^ngenetic.
Chapter III (page 49) states the primary fact that the alkaline
rocks have together but a very minute volume when compared with
the species belonging to the gabbro, granodiorite, or granite dans.
It ia likewise significant that the largest known body of alkaline rock
IB very much smaller than the lai^est known body of basalt, grsnodio-
*A. Harker, Presidential Address, Geol. Sect., Britiali Asweiation for the
AdvuuaciMit of Science, 1011, p. 4 (reprint paginAtion).
' n. 1lr>M'iibi Mikroskopische Physiographie der Mataigen Geetone, 4te
AufL, SiuttRurt K». 142 and 396.
414
ICXEOrS ROCKS AXl) THEIR ORIGIN
^
*Cr
"H. r
''J
* - •
'•
P** N
,-
t.
•2
.— zn
w^\'
3f
J*
p- \-
\ V
•••
^
t'j;
• *
^
':?^
^
71
^^ ••
^'it
■ •
^ ^"
«
c «•
. J
V«r
^■!
-
X
•
^« ■ M'
^
-«■
•
c*
t •
■
•^
/ J<
■I:
1-
»
Jj
CI
e*-
."
*~«
■ w
•/.
"U
ff«
w
.31
/;
■ •
•^.
^
^
^M
^Am
^^m
^
^^
• *«
^■^
U
*■•
^
*
«»
^
•*
^ ■
^
^
«
■
^^
^^
^^
■ ^^
^0^
ti
«
—
■ ■
^
*tr
^
>.
M
A
"•
~
-w«
x
•/.
z
h>
N —
x\\i\ or granite. Fig. 194 shows the compara-
tively niiniito vulumos of classic phonolit(-s.
TIk-sc farts clinTtly suggost the wisdom of
closely (*\aniiiiing the hyjMithosis that the al-
kaline rocks are derivatives of t he spectacularly
ahiindaiit magmas. The evidence summar-
ized in this chapter is additional to that d«^
tailfMl in Chapter XIX; its strength is t^uch
that the writer favors the hypothesis. For
t his and ot her reasons he prefers *' sul)alkaline "
to "calcic" as a general designation for the
rocks relatively pcMir in alkalies. The former
tenii signali/es the fact tliat the common
magmas carry alkalies which by concent rat i< in
rise to the profuirtions characterizing a foyait* .
an urtite, or a leucitite.
The very frecjuent association of nephelii«-
>yenites. leiicitic HK'ks, essexites, shonkinite^.
etc., with ordinarv svenite or monzonite in-
dicates a common mode of origin for all th*-M-
families (Figs. 1S2— 1). The question whetli#*r
the alkaline clans have also l)een different iate<l
from sediment a rv >vnt<*<'tics has already U^i^n
attacked l>v the writer and the result has U'c-n
|)iiMi>lied.' It was concludeil that most nf
the "alkaline rock" InMlies originated in syn-
teciicsnf i»a>alt with carl>onate-lH*aring rork*.
limestone*', dolomites, marls, or other calci-
reoii>^ or ma^noian sediments. The lin«* nf
^ =^ 7 ;j tlnm^ht is, on tin* whole, parallel to that
= ;7^^- ailoptfd in tin* last chapter. Some of th«-
^ 'i '' prinripli"* mi-d not Im* fully restated.
= ^ ? ,- A ulaiui' at a few ti priori arguments will
f £ "H. "^ ;iid in (piMHtitatively estimating the proMem.
' - i-1 In till- lirst place. flui<l haj»altic magma.
^, "^ z ^ with >inMThrat f)f the order actually obser\'cii
— I 7* I .it KihiUt-a, Matavanu. or Ktna. muH al>><irl<
£ " !I 3 liiiH-^ioiii- in din'<-t contact with it. Kot*nig^-
"^ \L I 1 I'triiiT aiitl Miiller have sliown that in water
" ~ " NuhniuTi- ^ilii'ji- ariil i^ >tronger than carbonir
acid at as low a t< mi'» r;ttiiri' :i- litiO® ( ' -'. Thi'se aci«ls certainly havi'
' n A Daly. Bull r,...\ >...■ \:!.. n. :». V..|. L»l, lOIO. pp. S7-118.
-.1 Kixiii^^luTccr :i!i>i I M'iil'i. ('•-n!r:ilM fur MintT. etc., 1906, pp. S39aad
T . •
<:
SI - -
X\
;A
\
i
'V.'.
» .1
I . {
• *
a, .V-J% '
\ :
.>.i.t.
ALKALINE CLANS 415
the same relation in natural magmas, as shown by the abundant
development oi lime silicates at contact of limestone and intrusive
rock. According to Cobb, dry lime and silica begin to react at SOO^.C,
a temperature far below that of most basaltic magmas at eruption.^
The igneous rocks of Colonsay and Oronsay (Hebrides islands) include
both alkaline and subalkaline types. The intrusives carry xenoliths
of quartzite, which have suffered partial assimilation.^ Is it possible
to doubt that the same magmas could also absorb the Torridonian
limestones and phyllites likewise invaded? Reinisch holds that the
high silica of the leucite basalt of the Gaussberg, Antarctica, is due
to the assimilation of granite and gneiss xenoliths.* It is logical to
suppose that the basalt, at the same temperature, could have assimi-
lated carbonate rocks.
Hibsch concluded that the great diversity of magmas in the
"alkaline" province of Bohemia is best explained by the magmatic
assimilation of dififerent rock materials; but he has offered no discus-
sion of the chemical processes involved.*
Carbonate rocks are evidently more prone to upset chemical
equilibrium in basaltic magma than is any other large-scale formation
which can be absorbed by it.
This implies, secondly, that differentiates of the basalt-limestone
syntectic will be more varied in composition than the differentiates
of other sjmtectics. The great diversity of types in the alkaline clans
is of the order of magnitude demanded on the proposed hypothesis.
Thirdly, the conditions under which carbonate rocks can be easily
assimilated by eruptive magmas are special and local, and those con-
ditions under which the influence of the absorbed limestone is not
masked by that of gneisses or other acid rocks simultaneously dis-
solved are yet more exceptional. The comparative rarity of alka-
line rocks is also of the same order as that characterizing the geological
conditions which favor the absorption of dominant limestone in sub-
alkaline magma. The points of eruption on the earth for basaltic or
gabbroid magma should be incomparably more abundant than those
for alkaline magmas — an obvious fact.
Association with Carbonate Rocks. — Rocks belonging to the al-
kaline clans are recorded in the regions listed in Appendix D, where
will be found the names of nearly all the important localities where
^ J. W. Cobb, Jour. Soc. Chem. Industry, Yorkshire Section, Vol. 29, 1910,
Nos. 2, 5, 6, 7, and 10.
* Craig, Wright, Bailey, Clough and Flett, Memoir 35, Geol. Survey of Scotland,
1911.
» Cf. E. Philippi, Deuti?che Siidpolar Expedition, 1901-1903, Berlin, 1906, Bd.
2, Heft 1, p. 54.
« J. K Hibsch, Tsehermak's Min. Petr. Mitt., Vol. 12, 1892, p. 405.
416
lONSOUS ROCKS AND THEIR ORIGIN
alkaline rocks have been diKcovprcd. Partly becaiwe oT the purr-
ilifferpntiation thoory of ikiioouh rocks which has bo flagrantly ignoml
the potwibility of a«i>imi]ation, many authors have fumisbed little
or no information regarding the terraneit invaded by the "iDtererting"
alkaline rocks. It needs no emphaub that the distiictB named are
exceetlinRly variable in area. One may be listed because it carrin
a ninglc dike of alkaline rock ; another is named as of the same eatcgor)'
though it really represents an extensive province covering hundml>
Fi(i. 183. — Mii|t or thr .Miinrhiqiic intTimion, PorlURnl. (Ari«r K. von Kruti
andV. Il&ckmim. TtKhiT. Min. I'ptr Milt., Vol IG, 1896, and GavenuBant sup n'
Portugnl.) (.', C'orhonifiTouti xliktp and fcritj'wurkc; y, Juruiic limntoM (eUeflr::
A, mctnmorphic aiirrolc; F, foyaitc; /', pubuikite.
of nlkaltno bodies. To l>e thoroughly useful, such a table shoold he
quantitiitive nnd imlii-uto the volumetric importance of each spwin
in eacli district. The i<lGal riinnot be closely approached until much
mure fii'ld work is done, but the tables, as they tttand, have conndcfable
value in pc-trogenic theory. A more cletailetl statistic would yet tnorr
clearly substantiate the following conclusions.
In all, 234 di.>ttri<'ts are known tu contain alkaline eruptivea. Of
thcM>, Ut3 districts, or 70 per cent., siiow tliese types to be aaaocialcd
with carbonate rocks in the way required on tbe propoaed hypothesis.
ALKALINE CLANS 417
In constructing the tables, it was often necessary to search for infor-
mation as to the sedimentary terranes underlying the present land sur-
face in the districts. Care was taken to assume limestone below the
surface only when the facts pointed strongly in that direction. It
was sometimes clear, too, that limestones were present in the roofs
of alkaline intrusives of the stock or batholith order, and it was then
legitimate to consider such limestone as subject to stoping and abyasal
assimilation during intrusion. For example, the large body of foyaite
in the Foya district of Portugal is mapped as in direct contact with
Devonian graywacke, slate, and sandstone. The region has been
greatly denuded since it was wholly covered by thick Jurassic lime-
stone and other sediments^ (Fig. 183). Is it possible that the
original roof-rock of the Foya intrusive was partly limestone, which
affected the magma through stoping and deep-feeated solution? The
co-operation of pre-Devonian limestones is, of course, not excluded.
Information as to the nature of the country rocks is lacking in 63
districts, largely volcanic islands, which, however, are likely to have
limestones in their foundations.
In only 8, or 3.4 per cent., of the xiistricts are alkaline rocks appar-
ently in such relations that the local absorption of carbonate rocks
is impossible. Few as these regions are, they are of obvious concern
to the hypothesis, and it is expedient to survey the facts in detail.
Though an advocate of large-scale assimilation by magmas, Ussing
considered that the nephelite syenites of Julianehaab, Greenland,
cannot be explained as dififerentiates of carbonate syntectics. He
wrote: "The complete absence of carbonate-rocks in the whole country
around the very large nepheline-syenite areas of south Greenland is
inconsistent with the hypothesis recently set forth by Daly relating
to a genetic connection between carbonate-rocks and nepheline-
syenites (Origin of the Alkaline Rocks, Bull. Geol. Soc. of America,
XXI, p. 87, 1910). The only carbonate-bearing formation of South
Greenland is the Arsuk group (p. 9). The possibility that dolomites
of this group have once existed at Julianehaab cannot be denied.
But the late-Algonkian igneous series (Julianehaab granite, etc.)
which is later than the Arsuk group consists of sub-alkaline rocks.
Thus, actual observations tell against the hypothesis in every way."*
The Greenland case somewhat resembles that at Cripple Creek,
Colorado, where nephelite syenite and phonolite have been erupted
through batholithic granite.
Since the general statistics of the Appendix tables so strongly
favor the idea of sedimentary control, the writer is impelled to risk
» Cf. C. P. Sheibner, Quart. Jour. Geol. Soc, Vol. 35, 1879, p. 42.
« N. V. Ussing, Meddelelser om Gr6nland, Vol. 38, 1911, p. 297, footnote.
418 laSEOVS ROCKS AND THEIR ORIGIN
the charge of special pleading and to point out several poflubiltties m
these and similar instances.
As an independent author of the stoping h3rpothe8i8, Usbiiik
presumably would have applied it as explaining the emplacenient of
the Julianehaab granite, through which these Greenland foyaites, etc.,
were intruded. Since the granite is younger than the calcareous
Arsuk group, it is possible that carbonatic matter was sunk into the
granite batholith in the late-magmatic condition when it was unable
to absorb all of the foreign material, or to permit advanced diffusion
of that material after its assimilation. A later superheated basaltie
weiige, penetrating such a mass abnormally rich in lime or carbon
dioxide, would necessarily be chemically affected.
Again, these few cases are subject to the possibility that the
alkaline rocks represent magmas which have migrated laterally from
their original '^ hearths" — a conception emphasised, for other pur-
poses, by Harker, who applies it on a wholesale scale.^ (See page 4(M.)
While carbonate-rocks are unknown in the immediate vicinity of
Julianehaab, the country rocks are completely hidden by the great
ice-cap east of the alkaline Igaliko mass. The Arsuk group or other
calcareous formatioas may be liberally represented in that region:
they may have entered into a syntectic at a considerable distance
and then have migrated horizontally to the Ilimaa«ak and Igaliko
chambers. These Julianehaab bodies have several features sug-
gesting that they are not batholiths but irregular laccoliths or
chonoliths.
De<'p-sHited migration of scKlimentary syntectics is not to he
excluded as a possible i)hasc of the Cripple Creek eruptives. The
Kico, Leadville, Georgetown, Breckenridge, and other districts of
western Colorado (all carrying syenitic, trachytic, or monionitic
rocks) show conditions favorable to the abyssal assimilation of thick
limestones. The Tertiary ujwloming of the Front Range of the Rocky
Mountains necessarily involved sulxrrustal movement of magma Xf^
ward the axis of the dome. (See also page 404.) Is it rash to postu-
late that this magma was alkaline? Probably field proof of this and
of the syntectic origin of the transported magma can never be found.
but the speculation has value in forbidding dogmatic assertion that
the Cripple Creek magmas are not of sj'ntectic origin. As noted in
the preliminary paper, the deeper mine workings of the Cripple Creek
district are specially, seriously affected by the abundant emanation of
carbonic acid gas from the igneous roi'ks. The gas may be of syntectic
origin, or it may suggest that the foyaitic magma has been generated
» A. Hjirkor, Prt's. Ailil., Hrit. Assoc. Adv. Science, Section C, 1011, p. 6,
reprint.
ALKALINE CLANS
419
because of an unusual concentration of juvenile (not resui^ent) carbon
dioxide beneath the Front Range dome.
Pirsson has described the nephelite syenite stock at Red Hill,
Xew Hampshire.' The mass is intrusive into a granitic gneiss which
surrounds it on all sides. No sediments occur in the immediate vicin-
ity but it is known that thick calciferous formations underlie much of
New Hampshire as well as Maine and Vermont. Two possibilities
are open. The roof of the Red Hill stock, now eroded away, may
have been partly composed of these calcareous members; or it may re-
present syntectie magma which has migrated from a distance. It may
be noted that the trachyte, syenite, and teachenite, cutting Paleozoic
Fig. 184.— Map of part of AlnQ Island (After A G Hflgbom, Geol. FfirMi.
Forhand., Vol. 17, 1895, PI ■• O &ne ss A me amorphic aureole in gneiBs;
lolid black, limestone; kea y do a hmeatone n h nephe e syenite; S, syenite;
.V, nephelite syenite.
limestones in Aroostook County, Maine, are on the (Appalachian)
strike from Red Hill.'
Several other areas of alkaline rocks, e.g., those at Port Coldwell,
Ontario, in Brazil, in Lappland, in Eastern Africa, etc., need more
complete mappii^ before their failures to show positive evidence can
be regarded as of telling moment.
On the other hand, any number of such difficult cases cannot
>>hake one's belief, with Adams and Barlow, that the nephelite syenites
of the Hastings-Haliburton area, Ontario, are genetically connected
with the Grenviile limestones alongside. Whatever the origin of the
' L. V. Pirsson, Amer. Jour, Science, Vol. 23, 1907, p. 257.
• See H. E. Gregory, Bull. 165, U. S. Geol. Survey, 1900, p. 93.
420 IGNEOUS ROCKS AND THEIR ORIGIN
limeHtone mansrs at Alno, their partial solution in nephelite iiyeiiitf
mafcma, or in that from which nephelite syenites have been derived
is an ohjcH'tive fact (Fig 184). Is it even conceivable that the Italian
lavas have traversed the thick Mesozoic limestones and ddomites,
now covered I)y flows and bre<*cias, without dissolving some of tk
carI)onate? Could the feeders of the Pilandsberg foyaitic eruptive
fail to dissolve the Great Dolomite as they rose through its tbousaiid*
of feet of thickness? Can one ignore the fact that the alkaline enip-
tives of Montana must have penetrated thick masses of calcareous
sediments I>efore reaching the oI)served levels? Was the ArkaMM
magma inert when it intruded the very thick limestones beneath tk
Magnet Cove region? No other question in petrogeny can be re-
garded as more necessary or more insistent than these and tk
many similar ones suggested I>y the list in Appendix D.
With his usual keenness, Teall saw the unavoidable nature of the
question. In a paper on the alkaline eruptivcs of Loch Borobn.
Scotland, he wrote:
''To what extent the different rocks represent successive intnvioitf.
difTerontiation in W/ti. or the result of a modification in the composttkm of
the original niagina by the absorption of adjacent limestones, has not bees
rlcarlv made out."
He remarke<l that all thr(»e processes were possibly engaged.'
H6glM)m states that the acid contact phase (alkaline syenite) of the
Alno stock l< due to marginal assimilation of the invaded gneifls bf
nephelite-syenite magma.' Is it then possible to doubt that the lime-
stone mas-c/^ en(*losed in the same magma must have exerted the
contnm', desilicating efTe<.'t on the original magniu of the Alnd stock?
The actuiil solution of the carlx)natc on a large scale has befS
thoroughly demoa<trated by Hogbom (Fig. 184).
The possible objet^tion that alkaline rocks are generally associated
with (*aicareous sediments merely because sediments of that character
seldom fail on continental plateau or in ocean basin, can be answerrd
in the same way as the analogous objcrtion for the syenite clan hs'
Ihh'h answered. Ini|>ortant calcareous formations are truly larkioft
over vast areas of the continents, where igneous eruptions have been
numerous and yet have afTonled no meml>ers of the alkaline clans. A
few exampU»s will illustrate the fallacy of the objection.
The Livingston (Montana) folio of the United States GcologiesI
Survey shows that the Tertiary theralites and allied rocks of this^
extensive quadrangle are found only in the area underlain by thick
» J. J. H Tonll. (;f*.l Mhk. Vol. 7. VMM\ p. 390.
'.\. Ci. HoglMMu, Ci4H>l. Kori^n. Furband., Vul. 17, 1895, p. 132.
ALKALINE CLANS 421
limestones. Elsewhere the limestones had been eroded away before
the Tertiary eruptions began and these furnished only subalkaline
species.
In Java and Madura the early Tertiary lavas (all subalkaline)
penetrated a terrane devoid of important calcareous beds. After
the thick Miocene limestone was deposited along the northern coast
of Java, it was traversed by new eruptives, largely alkaline.^
The Paleozoic eruptives of Bohemia are all, so far as known,
subalkaline — diabases, diorites, granites, etc. Only after the Tertiary
limestones were deposited over the region and igneous activity renewed,
w^ere the phonolites and their allies erupted. Bohemia contains an
alkaline province now but it contained none in the Paleozoic era.
Field Association with the Gabbro Clan. — The eclectic theory
predicts two modes of association between alkaline rocks and the
substratum material. The connection may be direct, where members
K PB
KB
Fig. 185. — Section in the Uvalde quadrangle, Texas. (After Uvalde Folio,
No. 64, U. S. G. S., 1900.) K, Cretaceous limestone and clay; PB, intrusive
plagioclase basalt; N^ intrusive nephclite-melilite basalt; jYB, nephelite basalt.
Illustrating syngenesis of ordinary basalt and "alkaline basalts/' Horizontal
scale, 1:95,000; vertical scale, 1:12,000.
of the gabbroid clan are clearly syngenetic with alkaline types; or
indirect, where alkaline types are syngenetic with differentiates of
basalt or of basaltic syntectics.
The dired association may be in the form of: (a) transitions;
or (b) adjacent separate eruptions accomplished in the same petrogenic
cycle.
Transitions from an alkaline phase to a subalkaline phase in tho
same rock-body seem to be very uncommon. An example is recorded
in the San Luis quadrangle of California, where olivine diabase merges
into augite teschenite within the limits of an intrusive sheet.* fSee
also page 339.)
The rarity of such cases is not a valid objection to the sediment-
control hypothesis. The absorption of carbonate has a specially vio-
lent chemical effect on primary magma, as detailed in the following
pages. Even a small amount of this absorption must flux the magma
and specially induce its differentiation into submagmas varying
^ R. D. M. Verbeek and R. Fcnnema, Description g^ologique de Java at
Madoura, Amsterdam, 1896, pp. 38, etc.
« San Luis Folio, U. S. Geol. Survey, 1904.
422
IGNEOUS ROCKS AND TBBIR ORIQIN
from melilite basalts, nephelite basalts, and limburptM to p
Hence we should not expect transition to normal basalt, d
gabbro as often as in the caw of members of the diorite, granodiorite,
and granite clan, when derived from Bilicious syntecties. Melilile.
leucite, and nephelite basalts are generally regarded as bdoo^ag fai
the alkaline series, but many of them, like quarti basalt, an ody
slightly modified plagioclase basalt.
Much more frequently the syngenesis c& separate alkalipc and
basaltic or gabbroid bodies is evident. Fig. 185 shows the intimacr
Fio. ise.— I'liin of island of Vulcano. (After A. B«rgMt, Abhsad. k. bapr.
Aktul. WiM., nmlh.-|>hy8. KI-. Vol. 20, 1R99. Taf. 6.) C, CnUn; /, l^dK
uml basaltic andoHilnx; 2, bualtic agKlomerates; 5, liparitM; i, luffs sad MBibi'
thn Kosna ili Vulcano; J, leuuitr liasaaitc. lUuslraliiiK close usodatkiB of tiiM*"
aiulcditi<s. aDil an alkaliiit- tyi>c.
existing Itetwern the pliigioc-litsr basalt and nephelit«-iiidiUte ba«H
of a very lociil arciiin the I'viililc Mountains, Texas.'
Figure 171 isadiaKrummaticmapof Stromboli,8bowingtliegaMfS'
tion of leucite biwiiiiite in the throat of a very young voleana whieh
had erupted ilominuiit basiUt and pyroxene andesitc.* Bcrgcat f,\\^
the unler of eruption for Volcano n.«: 1. Basaltic ande«t«. 3.1^)ant»
3. Leucite basanitc (of the Vuli-anello) (Fig. 186).
' UvaKie Folio. I'. S. CpuI. Survpy, 1900. p. 5. F
■ A. Borgcftt, Abhuid. k. baycr. Akad.Wias., Zi i
t«. :f.i.^iant» i
ALKALINE CLANS
423
as of Etna are dominaatly basaltic but they also inclode leucito-
Tes.' Fig. 187 illustrates aDother occurrence of basalts and
:itic rocks in the same cone (Roccamoofina).* Many analogous
relations are known in Hawaii, Samoa, Tahiti, Possesuon Island,
w Zealand (Dunedin, Figs. 18S-9), Juan Fernandez Islands, Heard
ind, Kerguelen, 8t, Helena, Ascension, Sao Thomfi, lUuiuon,
dagascar CFig. 190), Canary Islands, Cape Verde Islands, Azores,
deira, Fantelleria, etc. In this connection the assemblage of lavas
PiQ. 187 — Map of Roccamonfina volcano Italy (After W Krans in PetCT.
>g. Mitt \ol 58 M&TE Heft IS12) CrtMiiAimng MesoMio and TertUry
ement rocks 1 leuc t tes aad leuc te tephnt«e of the Itrt eruptive penod;
lu^te andes tes of the 2d erupt ve penod T l^achytes &nd leucitopbyree of
2d eruptive per od S basalts of the 3d eruptive penod Lateral eruptioiiB
■m by croaaes; altitudes in meters. Illustrating intimate asKxtiation of banlt,
esite, and alkaline types. Scale, 1:215,000.
Germany is significant. In the Lower Rhine region, plagioclase
alts (olivine- bearing and olivine-free), augite andeaite, hornblende
lesite, nephelite basalt, trachyte, and phonolite have been erupted
ing the Miocene. In the Upper Rhine region, feldspar basalts,
lilite basalt, nephelite basalt, leucite-nephelite basalt, tephrite,
:burgite, and phonolite date from the same period. Very BimHar
tes of lava were extruded during the Quaternary in the Lower
*H. J. Johnston-Lavis, Boll. soc. Ital. Microscopisti, Acireale, VoL 1, 1889,
• W.Ktmu, Petennann's Mitt., Vol. 58, 1912, p. 131, PI. 26; Cf. 1
% n* South Italian Volcanoes, Naples, 1891, p. 26.
424
IGNEOVS SOCKS AND THBIR ORIGIN
Rhine region.' The assemblages at the Mont Dora, Vday, Cutal
(Fig 191),Liinagnc, and le Livrndois (Fig. 193) volcanic centeiv of
France, and in the Rohcmian Mittelgebirge, tell the Bame story. In
Flu. 1K8. — Map of part of the Dunodin diatrict, New Zaslaad. (After f-
Marshall, Quart. Jour. Gcol. Soc., Vol. 62, 1906, PI. 36.) T, TrMh]rte; BS.hnrt»:
/', phonolitci TD, trachydolcritc; B, baoalt; D, doUntc; aolul blaet, ncpMx
basunitc; L, leucitophyrc; MB, mclilitc basanite; blank, sand aad aOuTiW'
IltiistratiiiK close amoriation of alkaline and aubalkaUne types.
each district alkaline and su))a)kalinc rocks are not only close togrthtt;
thoir eruptions fall within the limits of a short geological period.
Fio. 189.— Scrlion at North OUro Head in area of Rg. 188. (Samewl.
p. 418.) B, basalt; T, trurhytc; BA,ba8aiutc;P,phonolit«MPt ■
Such rt'peati'd close asswintions, lx>th in apace and time, mw*
it inrrwIibU- that each set of alkaline lavas hat* an origin ilidep«ftd«*
> R. Lepriiu, Gcologie von DcutwUanil, 2er Toil, Lapug, 1887-ltUk.
ALKALINE CLANS
5. 190. — Map of northern Madagaacar. (After R, Baron, QuarL Jour.
Soc,, Vol. 51, 1S95, and P. Lemoine, Etudes g^ologjques dans te nord de
^ascar, 1900.) CiS, crystalline 8«hista and marbles; M, Jurassic and Cre-
s (largely limestone); S, sycaitc; F, foyaite; P, phonotite; T, trachyte; NB,
lie basalt; B, basalt, irhieily olivine-b earing; D, dolerito.
1. 191. — Section oC Ihe Cantul volcano France (A(ter M. Boule, Guide
ng. Gdol. latemal , Pt 10, p 9 ) 1 gneiss aad mica schist; 2, Oligocene
nts; S, Miocene basalt; i, Miocene trachyte and phonolitc; S, Pliocene brec
1 tuS; S, Pliocene andeaite flows; 7, Pliocene phonolite; 8, Plateau basalt.
426 IGNEOUS ROCKS AND THEIR ORIGIN
of that to be assigned to the aeeompan3ring basalts. Rosenbiuck
himself has concluded that the keratophyres, long held to be tjriNcal
representatives of the alkaline series, are aplitic differentiates of sub-
alkaline magmas.^ He states that the keratophyric rocks are the onlj
ones which break the rule that alkaline and subalkaline tsrpes alvsTi
occur in separated areas (Verbreitungsgebiete).* The foregoiog
examples and a host of others, such as those listed in Appendix B.
clearly oppose this view. It cannot l>e substantiated by the remark
that the listed basalts, diabases, gabbros, etc., have been wroogij
diagnosed. Their diagnosis is the work of many of the worid's aUnt
petrographers. For example, Hosiwal long ago listed the many
species of alkaline rocks and closely associated common feldspar basalti
and pyroxene andesites found h}' the various expeditions which hid
then investigated Abyssinia and the adjacent region of East Africa.'
Subsequent work has not invalidated but has substantiated the fact
of this association of tj'pes.
The indirect association with basaltic magma is not less strikiog.
The Mauna Kea center in Hawaii has recently erupted basic andesito
and trachydolerites which are almost indistinguishable in the fidd
and nearl}' i<lentical in their chemical analyses. The field relatiom
and mineralogical and chemical composition of the Hawaiian tiariqr-
dolerites strongly suggest that, like the andesites, they are diffem-
tiates of the overwhelmingly dominant oli\nne basalt of the island.
The rare phonolites of Hawaii occur in just such limited volume ai
that expected if the basalt had absorl>ed small quantities of limestone
in lateral vents. In any case, these more salic types are most aimpif
explained as special difTen»ntiates of basalt. The extremdy clo»
intimacy between andesite and an alkaline type on Mauna Kea '»
paralleled in many volcanic areas. Verbeek and Fennema found that
the hornblende andesite of the Lourous volcano in Java had bffS
intruded through leucitic olivine basalt. A similar basalt, togethrr
with leucitite, tephrite, an<l nephelinite, forms the Ringguit voksDO.
A few other areas of alkaline rocks are distributed among the prevailiiV
andesites and pIagio('las(> basalts emitted at the long Javanese fissures'
Hack has recent I v shown that the small volcanic island, Soembava,
ft '
in the same province, is built up of dacit<*s, andt^sites of several typ».
and leiuitic tei)hrites an<l basanites.^ Short-lived, local eniptivitT
* H. H(is<n))uscli. Mikrosknpischr I'hysiofcraphie <lcf Maaaigm Gcatfliae. 4U
Aufl.. Stuttpart, 11X)S, p. 1493.
= 0p. Ht.. p. 1402.
' A. Hosiwal. Drnkschr. .Aknd. Wicn, Math. Nat. Kl., Vol. 9^ 1891, p. 87.
* II. n. M. V(>rl)(H.'k and H. Frnncina, DrHoriptinn K^logiqus da Jsts A
.Madnura, .Amstcnlam, 1S9«>.
- (;. Rack, Noil. Jahrb. fflr Minor, rtr., B.B. 34, 1912, p. 42.
ALKALINE CLANS 427
luring the Cretaceous period formed the layer of volcanic breccia
isible at the railroad track near Blairmore, Alberta. The breccia
ragment^ include true andesite, augite trachyte, analcite trachyte,
nd tinguaite. It is highly probable that all four types were generated
learly contemporaneously and in identical or neighboring vents. ^
similarly, Lacroix found andesites, basaltic types, and leucitites in
he same breccia series from Trebizond.^
Injected bodies give remarkable proofs that alkaline and sub-
Ikaline rocks are not confined, respectively, to separate provinces.
\othing could be more spectacular than the rock associations in the
iushveldt laccolith of the Transvaal. Molengraaff, Brouwer, Hum-
)hrey and others have shown that this colossal granite-norite-gabbro-
)>Toxenit*e injection, marvellously dififerentiated, is traversed by
ntnisive foyaites and syenites with extrusive, phonolitic and trachytic
)hases. Though the younger complex is 20 miles in diameter, it lies
ompletely surrounded by the subalkaline rocks of its mighty associate.*
Fig. 162, page 351).
Turning to the subjacent intrusive bodies, the associations are
igain found to be of the kind required by the eclectic theory. Accord-
ng to conditions, a great abyssal wedge should locally absorb very
lifferent amounts of carbonate, argillaceous, or acid rock. The respec-
:ive syntectic phases of the wedge should, by the theory, furnish
lifferentiates of nephelite syenite, syenite, granodiorite, diorite, or
p-anite. All these types should be developed in the same batholith
)r in the members of a composite, one-cycle batholith. A few examples
dll show how this prediction is fulfilled.
In the Yenisei district of Siberia, Meister has found a highly varied
jroup of syenites, nephelite syenites, soda-granites, banatites, diabases,
^abbros, peridotites, picrites, etc., cutting a thick series of basic
schists, slates, dolomites, and limestones, the last-named having an
'immense development." He writes:
"It is quite impossible to suppose that the eruption of these rocks, partly
3ertaining to granito-dioritic, gabbro-dioritic, and gabbro-peridotitic magmas
ind partly to foyaito-theralitic magma, belongs to two distinct periods. On
the contrary, the foyaito-theralitic rocks alternate in age with those derived
from granito-dioritic magma."*
» C. W. Knight, Canadian Record of Science, Vol. 19, 1905, p. 265.
* A. Lacroix, Bull. soc. g^ol. France, Vol. 19, 1891, p. 732.
* G. A. F. Molengraaff, Bull. soc. g6ol. France, Vol. 1, 1901, p. 13 (map); H. A.
Brouwer, Oorsprong en Samcnstelling der Transvaalsche Nephelien-syenieten, 'a
Gravenhage, 1910 (map); W. A. Humphrey, Trans. Geol. Soc. South Africa, Vol.
15, 1912, p. 100.
* A. Meister, Sur les Roches ct les Gisements d'Or dans la Partie Sud du Dia-
trict d'Jenissei, St. Petersburg, 1910, p. 593.
29
428
IGNEOUS ROCKS AND TBBIR OHIQtN
The composite batholith of the Okanagan mountaiiu in Wadung-
ton and British Columltia afTorda an exceptionally wdl erpoeed,
largc-Rcale illustration (Fig. 65, p. 1 15). The ayenitcfl, "*»''g"''*«, ud
nephelite 8venit4>s of the Krugcr body arc clearly eyngenetie, not «il7
with each other but also with the granodiorite, quarti diorite ud
monzonitr of the .Similkameen intrusive, and with the alkaline granite
of the Htill younger Cathedral Ixjdy.'
In hilt important monograph on the geology (rf nortb-ocntnl
Fin. 192.— Mnji of the BHnrron dititrirl, Ontario. (Aft«T F. D.
A. K. Barlow, Memoir 0, Ocol Surv. CanaiU, 1910.) P, pKragMHi; L, Oivnfc
limentonp, etc; A, &m|ihiholite;(?jl, gabbroiG, granite utd nrthntiniw; IT. ifc
lito Hycnitc.
Wisconsin, Weidman points out that there the nephdite lyenitc.
.■syenite, and granite are not only consanguineous but are genetieillj
connected with somewhat older galibros and dioritce as well h with
(iomcwhat older rhyolit<>s and andesitea. He writes:
"Tliis close corroKpoiulence in the pronounced chemical featom of Ike
three mDRnia^' (rhyulitic, gabbro-dioritic, and Kranitc-Bj-enitic] ia not bdien^
to l)c nci'idcntnl or liy chiince, but tuwignabic to definite eausea, althon^ tkt
cati-'^es may not he undcrstooi). As additional evidence of rdatkufaqi ■■
origin may lie ntc<l tlie fact of n.-^MM-iatetl occurrence and eruptioD in theMK
geological age."*
> Cr. It. A. Duly, Memoir Xo. 39, Op«l. Surv. Canada, 1912, p, «2S, md Uif
ShcctuNoH. 12 and l:i.
' S. Weiilruan, Rull. le, Wisconiin Geol. ud Nat. Hirt. Snmj, ItOI, p. SSI
ALKALINE CLANS 429
The little mass at Predazzo, containing nephelite syenite, essexite,
onkinite, theralite, syenite, quartz syenite, quartz monzonite, py-
xenite, and granite — all cutting melaphjrre and olivine-bearing
igite andesite — is a vivid example of the syngenesis of types belong-
g to all or nearly all of the igneous clans. ^
The Haliburton-Bancroft region in Ontario offers most striking
idence favoring sedimentary control in the formation of nephelite
enites (Fig. 192). Adams and Barlow have mapped about thirty
stinct areas of these rocks, occurring in fourteen townships. The
ithors write:
"The nepheline and associated alkali syenites are found either along the
tual contact of the granite and the limestone, or in the limestone itself
*ar the granite contact. There is only a single exception to this in the area
ider discussion, namely, the nepheline syenite mass in the township of
ethuen, which occurs between a great granite intrusion and the body of
aphibolite, containing a few small bands of limestone.
"They are intruded into the crystalline limestones and associated sedi-
entary rocks of the Grenville series on the one hand, and, as far as can be
itermined at the several points where they are well exposed, they pass over
to the fundamental gneiss on the other hand. Elsewhere, however, dikes
the nepheline syenite or associated alkali syenites can be seen to cut the
ndamental gneiss. A careful study of the whole area shows that the nephe-
le syenite and its associated alkali syenites represent a peripheral differen-
ation phase of the granite (fundamental gneiss), and that in the few cases
here these rocks are seen to cut the fundamental gneite they are of the nature
dikes of differentiated material intruded into a more acid phase of the same
agma, which was already consolidated very much in the same way as in the
ise of ordinary granite pegmatite, dikes are found representing the last
roduct of consolidation of a common magma
"The nepheline syenite occurs almost invariably along the border of the
'anite intrusions where these are protruded through the limestone. When
le actual contact of the nepheline syenite and limestone can be seen, masses
: the limestone, great and small, are found scattered through the nepheline
renite along the contact. These masses are in course of replacement by
le magma, and at a distance from the contact are seen to be greatly reduced
i size and often disintegrated. Still further from the contact they are
ipresented by irregularly rounded grains of calcite lying between the perfectly
'csh individuals of the several constituent minerals of the nepheline syenite,
r in some cases actually as inclusions in these minerals
"The origin of the nepheline syenite so extensively developed in the Ban-
roft region is also, as has been shown, in some way connected with the granite
itrusions. It is a differentiation phase, or a product of the magma in ques-
ion, and is almost invariably associated with limestones, which are in some
1 Compare W. Penck, NeueH Jahrb. fOr Miner, etc., B. B. 32, 1911, p. 341; J.
lomberg, Sitzungsber. Akad. Wiss. Berlin, 1902, pp. 673 and 731, and 1003, p. 43.
430 IGNEOUS ROCKS AND THEIR ORIGIN
way genetically connected with it. It is worthy of note that in PhifeHc^r
Lacroix's area in the Pyrenees, nepheline syenite also occurs, though here ia
smaller amount, and under such conditions that it in iiiiiKMsible to detenniiie
its actual genetic relations.''^
Th<*?*(» foyaitic rocks occur over a very extensive region chara^
terized by well-<Iistril)utod, large injections of ordinary gabhro and
by many other eruptive bodies of basaltic composition. These are
generally older than the batholiths and prove the local existence of
subcTustal basaltic magma before the great masses of granite were
emplac('<i. The field relations, therefore, correspond to the baol
assumptions of the eclectic theory. If the granite itself (generally a
common type not specially alkaline) is a differentiate of a basaltic
syntectic, the nephelite syenitqs of the region must have an indirect
connection with the primary basalt of the substratum. In any case,
the masterly investigation of Adams and Barlow illustrates the fallacj
of regarding alkaline and subalkaline magmas as primordially sad
always separate and it most clearly suggests that large bodies of
foyaitic magma are genetically connected with carbonate rocks.
It is unnecessarv to detail manv other illustrations, such Bis tbit
of the Christiania Region or others listed in Appendix D. The readff
will, of course, note that the present argument favoring the indirect
and direct association of foyaitic rocks with the gabbro clan is basrd
on the same principle as that used in the discussion of the syenite
clan. The relevant facts described in the last chapter thus tend to
strengthen the reasoning of this one. Since foyaites and syenites trv
so often syngenetic, the j)ro()fs of sedimentary control for the one ds»
of rocks are likelv to affect l)elief in sedimentarv control fortheothrr
class.
General Chemical Effects of the Absorption of Carbonate Rocks.
— No experimental attempt has yet l)een made to show in detaO thf
influence of the assimilation of limestone or dolomite on subalkaliiK
magma. Obviously dolomite would not have the same influence on
the s\iitectic or its differentiates as that exerted by pure calcium csr-
l)onate: t his contrast offers one of the special problems not jret soIvrA
Nevertheless, the scattered facts already known concerning tins subject
clearly favor the hypothesis Iktc outlined.
Smelting operation*^ agree with physico-chemical theory in shoiriiilt
that carbonate rock is a powerful flux for basaltic or other subalkalia^
magma. The effect niu>t be specially' great in plutonic chambers
where the carbon di(»\i(le is not allowed to escape as it is in the inditf-
trial uu'lt. With incn-aso of fluidity, any initial tendency to differ*
1 F. I). Adarn.s and A. K. Harlow, Geology of the Hsiifa ton and BsaeroA
Ana*). Mimnir No. i\ Oool. Surv. Canada, 1910, pp. 227, 22t>, ^32, and 408.
ALKALINE CLANS 431
entiation must be strengthened. In general the magmatic stability
itself is disturbed by the entrance of the new ions, so greatly contrasted
in quality or quantity with those of the original solution. For two
reasons, therefore, magmatic splitting is incited. As above noted,
we have ready explanation for the usual failure of the unchanged
syntactic phase, and of transitions to subalkaline types, to appear with
the exposed alkaline rocks. The almost astonishing facility for dif-
ferentiation in alkaline magma will be illustrated in a succeeding
statement regarding several thin sills and laccoliths (page 437; see
also page 237 fif).
The introduction of foreign calcium oxide into a basaltic magma
has two chief effects. The new material inoculates the solution and
caases an early separation of augitic or other lime-bearing molecules
already potentially present. Because of its strong affinity for silica,
the lime forms new molecules of more or less similar kinds. The
gravitative settling of these units, either liquid or solid, must increase
the alkalinity of the remaining solution.
If the new molecules are augitic — ^a common case — the foreign
lime binds at least 2.5 times its own weight of silica. Other molecules
would have the same desilicating effect on the original basalt. The
eclectic theory predicts, therefore, that the feldspars of the substratum
magma should be partly replaced by feldspathoids, nephelite, leucite,
etc.
Since magnesia and iron oxide also enter, with the foreign lime, into
the augite and other heavy molecules, the mother-liquor should be less
ferromagnesian as well as less calcic than the original basalt. These
are, in fact, the kinds of chemical contrasts existing between phonolite
and basalt or between foyaite and gabbr o. The desilication and allied
processes are exemplified in xenoliths and external contact shells
where pure carbonate rocks have so often been converted into am-
phibolites, garnet rocks, or other lime-silicate masses.
For lack of appropriate experimental work the exact influence of
absorbed carbon dioxide cannot be stated. This compound, like
the resurgent water, sulphur oxides, etc., cannot fail to affect the
differentiation of the basaltic syntectic. Probably these magmatic
gases are largely responsible for the upward transfer and local con-
centration of the alkalies. In what form the alkalies migrate, whether
as carbonates, hydrates, aluminates, or silicates, is yet unknown;
but the fact of their upward transfer on the large scale is proved by
the formation of adinoles, by the feldspathization of roof rocks, and,
it may be, by the existence of certain hot soda springs. The writer
has been much impressed with the pneumatolytic habit of the many
bodies of nephelitic and sodalitic rocks in the Hastings-Hali burton
432
IGNEOUS ROCKS AND THBJR ORWIN
rrgion, Ontario. These typen often form the roof phases of the con-
cordant, sill-like intrusioHH so characteristic of the diBtriet~« poai-
tion exprctc-d on the gascoua-transfer hypothecs. (See pagei 246
and 339.)
Giorgia and Gallo have described an experiment of i^vsent inteCfL
They immersed three analyzed samplea of Vemiviiu lava in water aod
for two months kept a current of COi paaung through the mizturf.
At the end of that time it was found that the lavas had lost fntn 30
to 40 per cent, of their soda, the other constituents h&ag but littJr
altered in amount.'
In a volcanic vent occupied by molten basalt, which is actuiUy
absorbing limestone or dolomite, the chemical conditions must be
very complex. Moderate concentration of the alluUies by gasKW
transfer may affect a portion of the lava column which has also ben
somewhat desilicated, giving a mclilite basalt, a nephelite basalt, s
Fia. 193. — East-wcat section acroMleUTTmdow an) leCotnU.FrMWS. (Afur
P. GlnnRFaud, Bull. scrr. carte gtel. France, No. 123, 1900, p. 17.) O, fniitic
bBsement; 1, ^, S, Oligoccae; B, boealt and tephnte; P, pbaauilita.
ba!«amte, or a nephetinite, which may be "fixed" by extniaioQ ui
thuK, through chilling, be prevented from further differentiatioo.
That splitting will normally occur in such a femic magma, if it bH
prolonged life, is suf^estcil by the failure of melilite basalt to haves
plutonic equivalent. Nephelite basalt and leucite basalt magmtt,
relatively abundant themselves, have extremely rare refHvaentativef
among the granular rocks which have crystallised slowly in Isige
subterranean chamlK-rfl. Only with long life can the qmteetie separate
into the extreme .lubmagmas. During most of its life the difft
tion must be affected by: (a) the settling of cafemic i
(b) the ritie of alkaline constituents; (c) the addition erf new natoiil
from the wall rocks, which themselves are usually heterotencooi-
Under i^uch conditions we should expect the differentiates to he higUj
varied, twth chemically and mincralogically. The yariability of
individual alkaline bodies and the great number of named alkalise
ai>ecies (out of nil proportion to their volumetric importance) an both
of the order expected on the syntectic hypothesis.
> G. CiorRu and O. Callo, GupKb, Vol. 36, 1906 ()}. P- 137.
ALKALINE CLANS
433
Further, the eclectic theory explains the preponderance of femic
species among the extrusives of the alkaline clan, and the preponder-
ance of salic species among its intrusives. During its initial, hot
stage, even after some carbonate rock has been absorbed, the primary
basaltic magma is eruptible with comparative readiness. As S3mtexis
progresses, viscosity must increase and tend to prevent extrusion.
Large-scale assimilation is possible only in large chambers. These
Fio. 194. — Map of phonolites {dotted) of the Velay. (After M. Boule, Bull.
serv. carte gtol. France, No. 28, 1892, p. 147.)
are generally capable of magmatic splitting after notable assimilation
has ceased. The salic pole of the intrusive body, here as in the case
of the other clans, will usually be the only one exposed in large quantity
by rosion. The world's maps provide facts amply sufficient to show
the parallel between the deductions and the actual distribution of
igneous rocks. The nephelite basalts of Germany overwhelm the
phonolites. The phonolites of France were erupted at many points
434 laSEOVS ROCKS AND THEIR ORIGIS
but always in small volume (Figs. 193-4). In Italy leucite tephritcsand
basanites abound, while phonolitic lavas are rare. Among the intra*
sives, foyaitc clearly dominates over the more femic nepbelite syenite^
over ijolites and their allies.
Evidence from the Mineralogy of Alkaline Rocks. — ^The offered
explanation for most of the alkaline rocks involves a special mineralogy
for them and one of its strongest merits is that it gives the key to a
genetic problem which has hitherto had no adequate solution whatever.
(1) No reasonable doubt now remains as to the primary nature
of the calcite enclosed in many eruptive bodies. A few granites
contain it as a rare accessory but most of the calcite crystallised from
igneous melts has been found in nephelite syenites and their clo^
allies. Well-known examples are described in the memoirs on Alno,
Sweden; the Ilastings-Haliburton area, Ontario; and the Ice River
intrusive, British Columbia.
Cancrinite was discovered and first named at Miask ^Urals).
where, in company with scapolite and corundum, it occurs as a primar}'
constituent of nephelite syenite in contact with thick limestone.
According to Thugutt, its complex formula contains five molecules
of C'aCOa, one molecule of Na2C'0j, and, also suggestively, three mole-
cules of sodium aluminate.* Preobrajensky reports that the nephe-
lite syenite of upper Zarafshan (Turkestan) contains cancrinite, idio-
morphic calcite, titanite, and sodalite. The cancrinite forms strips
parallel to the contact with limestone which is intruded by the syenite.^
Such facts inevitably suggest a syntectic origin for these minerals.
(2) Nephelite, loucite, socialite, noselite (nosean), haOynite.
analcite, corundum, spinel, and probably muscoxite are characterL*ttic
constituents of alkaline rocks. Most of them do not occur in rocks
which have been referred to the subalkaline clans. All are compounds
such as might crystallize in subalkaline magma which has been de-
silicated in the manner described. The combined water of analcite
and the hydroxy 1 molecule of muscovitc are expected ingredients of s
sedimentary svntectic. The liberation of alumina from silica and
alkalies to form corundum and spinel is again a predictable result
of the absorption of carbonates by subalkaline magmas. The recur-
rence of corundum in nephelite syenites, etc., cutting limestones, in
Ontario. India, and elsewhere is not accidental. It is worthy of notethat
the Montana sapphires are found in minette dikes cutting thick lim^
stone' <^rip. 119, p. 207). That these aredue to "desilieation" by the
* S. J. Thuputt. Ncurs Jahrh. fiir Minor, ptc, 1911 (•), p. 45.
* P. PnM»!ir:ijonsky. AnnuU dv I'lnstitiit Polyterhnique Pierre le Grasd, 9i.
IVtorHhurir. Vol. l."», 1911, p. 293.
' W. H. Wml. JOth .Vim. Rep., V. S. Gc>ol. Survey, Pt. 3. 1900, p. 4W.
ALKALINE CLANS 435
intruded limestone, rather than to the crystallization of the alumina
of accompanying shales, is suggested by the fact that the minettes of
Montana, where cutting limestone, sometimes carry nephelite.'
Moreover, Jensen has found a mineral with the properties of corundum
in the melilite basalts of the Warrumbungle Mountains, New South
Wales.' The oligoclaaeJcorundum rock, plumasite, of California is
another type, showing the tendency of alkaline types to carry free
alumina. Its liberation finds general explanation on the carbonate-
syntectic hypothesis.'
The picritic sill of Inchcolm island, Scotland, is bordered by a
J . ,v'
J^
%
fl-JPV^
kif
^
IP'
^z^r~
^^^W
/^"'^
0
500 M
0
2090 F.
Fio. 195.— Map ot Inchcolm Island, Scottish coast. (Alter R. Campbell and
\. G. Stenhouae, Trans. Edjn. Geo), Soc, Vol. 9, 1907, p. 121.) S, sandstones,
etc., underlying sill; Sli, shale, aah, etc., overlying sill; B, oli vine-basalt lava;
toarte dots, teechenite of aill; P, picrite of hiH; QD, quartE-dolerite dike.
teschenitic phase, rich in primary analcite (Fig. 195). Campbell and
Stenhouse are convinced that the magma has absorbed a considerable
amount of the invaded calcareous sandstones and argillites.* They
state that it is impossible to find a sharp limit between sediments and
intrusive. Is this an actual case of desilicatiou in place?
According to Lowe, the Lurcombe sill in Devonshire is vertically
(gravitatively?) stratified, with a camptonitic phase at the roof,
> W. H. Weed, Little Belt Mountains Folio, U. S. Geol. Survey, 1899.
' H. I. Jensen, Proc. Linn. Soc, N. S. W., Vol. 32, 1907, p. 816.
» A. C. Lawson, Bull. Dept. Geol. Univ. California, Vol. 3, 1903, p. 219.
• R. CampbeU and A. G. Stenhouse, Trans. Edinburgh Geol. Soc., Vol. 9, 1907,
p. 133.
436 IGNEOUS ROCKS AND THEIR ORIOIN
toschenite in the middle, and an augitic phase of the tesehenite at
the l>ottom. This intrusive Ixnly is enclosed in Culm sediments and
has also evidently cut the underlying massive Devonian limestone.*
The classic teschenitc of Austria occurs in the same sill with picrite
and a hornblende-free, diabasic type. The sill cuts the Teschen
limestone and associated marly sediments. Here no decisive evidence
has been found that the differentiation occurred in place.*
(3) The local excess of lime in "alkaline" magmas is indicated
not only by primary calcite but by the common devdopment of
melilite, scapolite, wollastonite, lime-garnets, and perhaps anortbo^
clase itself. Though the solution of carbonate rock in basalt incites
differentiation so that the salic sub-magma shall be poor in lime, it
is obvious that, toward the close of the assimilating period, all of the
foreign lime cannot be cleansed from the cooling, viscous, alkaline
submagma. The hybrid rocks will thus vary greatly in compoatioo,
both according to the proportion of lime absorbed and according to
the advance of differentiation registered in the submagma. The
melilite basalts represent only moderate differentiation; they vary in
alkaline content and some varieties approach the postulated cafcmie
pole of splitting in a basalt-limestone or basalt-dolomite qmtectk.
Becker has concluded that the melilite basalt of the Wartenbeig,
Germany, is an actual product of magmatic absorption of the invaded
limestones.' Starabba reports that melilite was produced by the
partial solution of limestone inclusions in the 1883| 1886, 1892, and
1910 lavas of Etna.^ In these rocks and in the closely related melilite
basalts, melilite is sometimes accompanied by wollastonite, lia(k>nite.
garnet, and perovskite, as well as by augite. Primary scapolite and
calcite are associated in the nephelite syenites of the Hastings-Hali*
burton area of Ontario. The tinguaite of Spotted Fawn Creek, Yukon,
contains essential leucite, with scapolite in its ground-mass.*
The lime- bearing minerals, titanite and perovskite, are also sus-
piciously abundant in many species of alkaline rocks. The more salic
species, foyaite, phonolite, etc., generally tend to be correspondin^y
poor in ilmenite or titaniferous magnetite. Both facts are explic-
ablo on the view that these differentiates have been derived fron
a limestono-syntectic. Some ferrous iron should enter the augite
and other cafemic molecules forme<l by inoculation with foreign
lime. Titanic oxide is thus free to combine with lime, forming stable
» H. J. Lowe, Geol. Mag., Vol. 5, 1908, p. 344.
' V. Uhlig, Bail und Bild Oestorrcirhfl, Vienna and Leipgig, 1908, p. 880.
' K. BiH^kor. Zcit. d. <lout. Gotil. (;m<ll., Vol. 59. 1907, pp. 244 and 401.
' Stella Stnrabha. Ucnd. R. Arc ad., Lincei. Vol. 19, 1910, p. 756.
* C. W. Knight, Amer. Jour. Science, Vol. 21, 1906, p.
ALKALINE CLANS 437
compounds. These seem to be moderately "volatile" and, as in the
Turkestan intrusives above noted, may be driven out into the sur-
rounding limestone formation.^ There are, in fact, field evidences
that titanite tends to accompany the alkalies as they are magmatieally
concentrated. Associated in origin with the aegerite syenite of
Bowral, New South Wales, are pegmatite veins charged with abundant
titanite, ilmenite, and perovskite.^ Benson notes that the alkaline
(sodic) rocks of southern Australia and of eastern Australia are
characteristically rich in titanic oxide.'
(4) The hypothesis predicts that free carbon should enter a syn-
tectic if the absorbed limestone is carbonaceous. Graphite has, in
fact, been described as among the magmatic products in the nephelite
syenites of India and of Ontario.
The mineralogy of the alkaline rocks, though extensive, is highly
specialized. At least sixteen species of minerals, either essential
or accessory, are just those to be expected on the carbonate-syntectic
hj'pothesis. Many of them, and particularly their actual associations
in rocks, are most difficult of explanation on any other postulate.
Synthetic experiments give full permission for belief in the preferred
explanation. The reader will find a convenient sunmiary of these
experiments in the parts of Clarke's "Data of Geochemistry" dealing
with analcite, cancrinite, corundum, garnet, graphite, hatlynite, leucite,
melilite, nephelite, scapolite, titanite, and woUastonite.*
It hardly needs emphasis that water as well as carbon dioxide must
enter a limestone syntectic. The source of the water is either the
carbonate rock itself or the usually associated shales or other silicious
sediments. Resurgent water must co-operate in the segr^ation of
alkaline submagmas and in the crystallization of minerals, like riebec-
kit«, which form only in the presence of "mineralizers."* Like many
syenites, soda-granites, albitites, oligoclasites, veins of potash feldspar,
etc., are normally differentiated from syntectics of hydrous silicious
sediments, rather than from carbonate syntectics. In general, the
thesis that the segregation of alkalies in the alkaline rocks proper is
largely caused by transfer in volatile solutions is greatly strengthened
by the known facts concerning the aplites from subalkaline magmas.
Differentiation of Alkaline Rocks in Place. — ^The eclectic theory
assumes the active control of gravity in magmatic splitting. This
matter is so vital that it merits attention in the problem of each
^ P. Preobrajensky, Annals de ITnstitut Polytechnique Pierre le Grand, St.
Petersburg. Vol. 15, 1911, p. 293.
« D. Mawson, Proc. Linn. Soc, N. S. Wales, Vol. 31, 1906, p. 579.
» W. N. Benson, Trans. Roy. Soc. South Australia, Vol. 33, 1909, p. 137.
♦ BuU. 491, U. S. Geol. Survey, 1911.
* G. M. Murgoci, Amer. Jour. Science, Vol. 20, 1905, p. 133.
438 IGNEOUS ROCKS AND THEIR ORIGIN
if^neous clan in turn. As with the gabbro, granite, granodioritc,
(liorite, and syenite clans, so here, the facts concerning sills and lac-
roliths are of the highest importance.
As note<i in Chapter XII, the gravitative differentiation of the
Shonkin Sag laccolith, the Square Butte laccolith, and the Lugar sill
is obviously dear. In the first, nt^phelite syenite and shonkinite arf
polar differentiates of leucite basalt. In the second, sodalite syenite
and shonkinite are similarly derived from leucite basalt. In the third,
theralite and picrite are gravitative differentiates of "basaltic" tescbe>
nite magma.
Tyrrell has explained a picritic phase of the Benbeoch sill, Wert
Scotland, as a gra\ntative differentiate of "kylite" after injection.*
He notes that the Castle Craigs picrite-tcschenite sill of AyrBhire
shows magmatic splitting very similar to that at Lugar; and abo
details facts suggesting that, in the Howford Bridge sill of Ayrshire,
analcite syenite is a gravitative differentiate of "essexiteKlolerite."
The Prospect intrusion near Sydney, New South Wales, is a lacc-
olithic sheet at least 300 fe<>t thick, cutting Tria.ssic shales and neces-
sarily intruded through the underlying Paleozoic limestones and other
sediments. The roof phase represents the magma as originally in-
jected at this horizon; it is an essexite showing the phenomeoa of
rapid chilling. It is underlain by a (k*cidedly more salic, feldspathic
(»ssexite, succeeded below by an essexite layer Hith the nuudmuai pn^
portions of augite (38 per cent.) and ilmenite (16 per cent.). Thf
floor of the sheet is not ex|)osed. The splitting is not far advanced
but it has l)een governed by the same general laws as those affecting
the Montana and Scottish intrusives just mentioned.*
The (irosspriesen laccolith of the Bohemian Mittelgebirge exhibits
sodalite syenite in outcrop. Hibsch's map shows that the cover
of Tertiary sediments still largely remains. Hence this syenite phase
occurs near the laccolithic roof, that is, in a situation suggestively
like that of the sodalite syenite of Square Butte. The Grosspriesen
syenite (sp. gr. 2.r>31) is closely as<()ciat(>d with essexite (sp. gr. 2.8S3j.
Their genetic relation is not decipherable from the facts described.*
» (i. W. Tyrnll, (mm.L Map., Vol. 0, 1912, pp. 73 and 123; Tnms. Geol. Sor
(U.Msgow. Vol. i:{, ItHK^ p. :i(K). This authority regards kylite as the pluUMiic
equiviilrnt of iit-phrlito h:i.<Mlt.
» H. S. Jfvons. H. I. Jons«>n. T. G. Tjivlor and C. A. SCkMRiikh, Proc. Roy.
Sor. N. S. Wiilrs. Vol. •«.->. 1011. p. AW :in«l V<il. 46, 1912, p. 111. Taylor and
Jwons tlisniss (p:mr 4.V.M tho (lossihility of ovcrhc.'ad stoping in this cnedtc-
Their fstiiiiatrs of rock di'iisitit-i >))ow th.it uD<lerhand fttoping would be the only
kind i>().4siLlr in this rast'. Obviou.Nly siioh a thin Hheot could not be expected to
illustrate stopin^; <>\(vpt on a most insifinifioant scale.
' J. E. Hihsch, Tsrhorniak's Min. ii. Pot. Mitt.. Vol. 21, 1902, pp. 157 and MS.
ALKALINE CLANS 439
The Borolan laccolith at Cnoc-na-Sroine, Scotland, measures
4 miles by 2.5 miles in outcrop, with a probable original thickness
of about 0.25 mile. The roof has been eroded away. Shand's sec-
tion is reproduced in Fig. 196. The intrusion is stratified in the follow-
ing order:
Approx. spec, gravity
' Erosion surface
Phase 1 1 Quartz syenites
Phase 2 1 Transition zone of quartz-free syenite
Phase 3 1 Melanite-nephelite syenite and ledmorite. . .
Base concealed
2.625-2.635
2.65
2.74 -2.78
Shand favors the hypothesis that the concealed floor phase is
composed of melanite pyroxenite, like that observed as locally intru-
sive into Phase 3.^
The advanced character of this clearly gravitative splitting, in a
sheet of rather moderate thickness, is one more illustration of the
4 3
^ 3 ^.^?l!iPlS5^"^
* ■ 1 __i Mis ' Km.
Fig. 196. — Sections of Cnoc-na-Sroine laccolith, Scotland. (After S. J. Shand,
Trans. Edin. Geol. Soc, Vol. 9, 1910, p. 379.) C, Cambrian; 1, quartz syenites;
Bj transition rocks; 5, melanite-nephelite syenites, augite-nephelite syenites, eto.;
4. hypothetical ultra-basic layer. Upper section is N.W.-S.E.; lower section is
S.W.-N.E.
decided fluidity of alkaline magmas, however viscous many artificial
melts may be. This case throws light on the syngenesis of soda-
granites, nordmarkites, foyaites, etc. It throws suggestive light on
the problems of origin in connection with the much greater rock assem-
blages in the Christiania Region and at Predazzo, etc.
The very able memoir by Ussing, on the Julianehaab district of
Greenland, describes a remarkable group of alkaline rocks which
have crystallized from magmas differentiated, at least in part, under
the dominating influence of gravity.^ The sodalite foyaite, "naujaite,"
lujavrite, and "kakortokite" of the Ilimausak intrusion — all related
1 S. J. Shand, Trans. Edinburgh Geol. Soc, Vol. 9, 1910, p. 376.
' N. V. Ussing, Geology of the Country around Julianehaab, Greenland,
Copenhagen, 1911, pp. 318, 348, etc.
440 IGNSOUS BOCKS AND THBfB OBtGtN
to foyaite — are so interpreted by UsMog (Fig. 197-6). ThcM aiiil
the other roclu of the mass occur as successive, neariy boriioiiti]
shecti!, named in order, from alcove downward:
Arfvedaonitc iiriiniti'
Quu-U lycnitr
Pulukit«
Foyait*..
Sodalite foyftite ...
Naujaitc
LujavritPS and kAkortokitnt
Tliirkni'M (meters) Spec, psntj
I50-4U0 I 2.6^3.73
0-20 (7)
10-30 3. 72
0-10 (T) 3.«
2-150 (aveniRe, 100 m.) 3 «S
200-600(avera«e,300iii.) , 3.53
600+ 3.76-3.13
The naujaitic layer (highly sodalitic) is explained by an up'
transfer of the sodalite molecule from the deeper part of the I
Fki. 197.— Map of the Ilimaiwak intruaion, Groenland. (After N. V. UwiR.
Mnia. om nronlaml. Vul. 38, 1011, PI. 3.} G, granite; S, ModatOM; D, dialaw
und poriih.vri<«: E, rawnle; A, augile tyemte; L, lujavrit*; K, kakartokila; S.
nniijnitc; F, sodality foyaitp; \d, nonlinarkite; AG, arfvadacmite paoit*.
The more ferrou.s and less aluminous lujnvrito-kakortokite n
explained ns the roHilunl li(|iior left after the gravitative removal of the
sodalite. LVing cunsidcrotl it likely that thin sulmtance waa tnuirf cRcd
in the form of m>lid crytitals. Fur present purposes it is not neeeHUT
ALKALINE CLANS 441
to decide between that hypothesis and the alternative one of liquation.
The sodalite foyaite above the naujaite represents nearly the chemical
composition of the magma before this differentiation took place. Uss-
ing did not attribute the foyaitic phase to the relatively quick chilling
of the original magma near the roof of the intrusion but gave a hypo-
thetical explanation detailed on page 354 of his memoir. That hypothe-
sis, involving the rise of magmatic gases, also implies gravitative control.
Ussing explained the arfvedsonite granite phase as probably due
to the assimilation of the sandstone intruded by the foyaitic magma.
It is shown that the clearly stoped-down blocks of sandstone are sur-
rounded with thick shells of this granite merging outwardly into alka-
line, subsilicic syenite. The syenite in its turn merges into the nephe-
lite-bearing rocks. If this view is correct, the assimilation must have
Fig. 198. — Schematic section of the Ilimausak intrusion. (Same ref. as for
Fig. 197, p. 322.) S, sandstone, diabase, etc.; LK, lujavrite and kakortokite;
iV, naujaite; F, sodalite foyaite; P, pulaskite; G, arfvedsonite granite.
partly occurred in place. In this connection the discovery of a layer
of augite syenite at the lower contact and below the denser lujayrites
of the neighboring Igaliko intrusion is significant.^ Nevertheless
it is possible that the granitic magma was largely differentiated from
the sandstone syntectic and solidified before the main foyaitic magma
was differentiated. The granitic magma was clearly less dense than
the undifferentiated foyaite, so that this initial splitting was controlled
by gravity. Afterward the deeper-lying, still fluid foyaite split into
the strongly contrasted naujaite and lujavrite sub-magmas. This dif-
ferentiation affords a noteworthy illustration of the upward transfer
of alkali, the overlying naujaite being extremely rich in sodalite. Allan
has observed a similar segregation of sodalitic rock near the roof of the
alkaline intrusion at Ice River, British Columbia.* (See also p. 235.)
1 See sections in Ussing's memoir on pp. 252, 253, and also those on pp. 38, 39,
42, and 61.
* J. A. Allan, Geology of the Ice River District, a thesis abstract published by
the Massachusetts Institute of Technology, 1912, p. 11.
442 IGNEOUS ROCKS AND THEIR ORIGIN
The problem is much too delicate for dogmatism but one cannot
but ag^ree with Ussing that the Ilimausak intrusion, complicated as it ii,
does illustrate gravitative differentiation on a large scale, as shown in
the relation of the highly sodalitic layer to its more ferric associate.
In passing, it may l>e noted that the Ilimausak and still largfr
Igaliko ''batholiths" often have concordant or roughly concordant re-
lations to the invaded sandstone. The contacts of the Ilimau^
intrusion are like those at the roof of a chonolith or an irregular, partly
cross-cutting laccolith. The sections on pages 252 and 253 of Ussing's
book, when compared with the map in Plate IV, strongly suggest that
the Igaliko body is an irregular laccolith with its base exposed.
The memoir further descril)es an unusually perfect and full illu.*-
tration of primary banding. The kakortokites of the lowest viflble
phase are arranged in laj'ers of l)lack, white, and red colors, corrf-
Fio. 199. — Artual scMrtion of the IlimaiiMik intrusion from W. S. W. to E. N. E
(Same ref. as for Fitc. 197, I'l. 0.) S, sandstone; D, diabase and porphyry (iIimU::
L, ban(ie<l lujavrite, with leiL^es ami fragments of naujaitc; N, naujaite; F, sodabu
foyoite; AG, arfve<i.Monite ^anite.
sponding to great difTerences in s|)ecific gravity and mineral composi-
tion (Fig. 199).
(('
The peculiar kind of htnitificntiun characterizing the kakortokitk com-
plex will appear from the foHouing Ii:^t of a number of consecutive sheets:
Tliirkness Speeifie gravitjr
Black kakortokite ... ra. 23 meters ca. 3.12
White kakortokite va. (V-9 meters ca. 2.76
Red kakortokite. . . ra. 1 2 meters ca. 2.8ft
Black kakortokite ca. 2 3 meters ca. 3. 12
White kakortokite. ca. G-9 meters ca. 2.76
UimI kakortokite. ca. 1 2 meters ca. 2.8ft
Black kakortokite . ca. 2 3 meters ca. 3.12
"The .Mirce^ision as fiiveii in this table continues through a total thkluieff
of about 400 meters, the nuitiber of individual sheets amounting to more tk*n
a hundred, while the number of repetitions of c(»lor sets is about forty. It i»
worth mentioning that the red .•sheets in many places are badly developed or
even wanting* but even in Mich cases the lowennost portions of the white
sheets or tlie uppermost portions of the black ones are relativdy rfch in
eudialvt*' .
ALKALINE CLANS 443
"That gra\dtative separation is able to account for the differentiation
will appear from the following consideration. The black kakortokite sheets
(sp. gr. 3.12) are characterized by the abundance of arfvedsonite (sp. gr. 3.4);
the red sheets (2.85) which overlie the black ones abound in eudialyte (2.9) ;
and the white rock (2.76) which covers the red sheets has alkali felspar as
its dominant mineral (2.6). The arrangement thus agrees with what should
be expected if it were due to gravitation
''For the banded kakortokite of the Ilimausak complex the simplest
supposition is perhaps that the recurrent layers have originated inconsequence
of repeated variations in pressure. Each reduction in the pressure may have
caused the dissociation of a certain quantity of volatile matter from the magma,
and this process in its turn may have caused the crystallization of a certain
quantity of the magma."*
The variation of pressure is supposed to be due to volcanic outbursts
from the magma chamber. The present writer may also suggest the in-
quiry as to whether a principle which has been experimentally illus-
trated may apply to this and many other cases of banding in igneous
rocks. If silver nitrate is allowed to dijBFuse into a gelatin film contain-
ing potassium dichromate, there ensues an intermittent precipitation
of silver dichromate, shown by alternating bands of color,* Is it
possible that, under certain conditions, liquation is intermittently,
rhythmically produced in magmas?
Similar primary banding is seen in the great f oyaitic laccolith of the
Kola peninsula, in the malignite-nephelite syenite body of Kruger
mountain in the Cascade range (Canada-United States boundary line),
and in many other plutonic alkaline masses. Multiplied examples
show the importance of active attack on the problem of explaining
these differentiations in place.
Chemical Contrast of Alkaline Volcanic Species and the Corre-
sponding Plutonic Species. — As in all the other clans so far described,
gravitative control should be manifested in the chemical relation be-
tween each alkaline volcanic type and its equivalent plutonic type.
For this comparison averages of composition are obviously more sig-
nificant than single analyses. Table II shows the situation for the
more important pairs of averages. See Columns 35, 361 75i 76, 80| 81 1
88, and 89. The comparisons show, in every case, the more salic
nature of the extrusive phase, doubtless corresponding to lower density
for its magma.
Eruptive Sequence in Alkaline Provinces. — The demonstrated ease
with which alkaline magmas differentiate in the many sheets and lae-
1 Geology of the Country around Julianehaab, Greenland, Meddelelaer om
Gronland, Vol. 38, 1911, pp. 356, 358, 360.
« Cf . R. E. Liesegang, Zeit. phys. Chemie, Vol. 59, 1907, p. 444.
444 IGXEOUS R(JCKS AND THEIR ORIGIN
eoliths (lcscril)O(l leads to the dediietion that the order of eruption in
alkaline provinces will he somewhat variable. Yet, according to the
eelertir theory, the sequence for the larger intrusive masses should gen-
erally be from femic to salic. Such, in fact, is the eruptive order al-
ready determined for many provinces, e.g., the Okanagan composite
batholith; Midway district, British ('oluml)ia; Red Hill, New Hamp-
shire; Predazzo, Tyrol; C'hristiania Region; Julianehaab, Greenland:
Kola peninsula, etc. A typical illustration is found in the "stock"
in th(» OlxTwiesenthal, Erzgebirge. It is chiefly composes 1 of nepheliti*
basalt locally transitional, through sanidine phonolites and leucitc
phonolites, into true phonolite, and cut by numerous dikes and small
"stocks" of phonolite. Sauer concluded that the basalt was not ytt
cold when the phonolite was erupt e<l.*
The eclectic theory further implies that the eruptives from vol-
canoes of the central tyjx* should often alternate from femic to salic anil
conversely. The n^ason for this d(>duction is the name as that given
on page 372 for the analogous relation of trachytes, dacit<fs, etc. to
basalt. Several illustrations are given in the table of Appendix B.
especially the sequence for: Mont Dore, le Velayand le M^zenc, Can-
tal, Limagne, France: Rhine provinc<»s, (Jermany; Bohemia; Lipari
(Kolian) Islands; New South Wales an<l Victoria, Australia.
In general, the fcrromagnesian or cafemic differentiate of a limt^
stone syntectic must be in depth too great for ready eruption. ThL<
deduction from the theory partly explains the cause of the relative
rarity and small individual volume of the visible shonkinitic, ijolitir.
bekinkinitic, limburgitic, and augititic eruptives. Unless these rock*
are differentiates in phu-e. they should, as a rule, be erupted after the
corresponding overlying salic submagmas have solidified. According
to Shand, this has actually occurred at Loch Rorolan, where the nephe-
litic syenites are distinctly cut by the syngeneticmelanite pyroxenite.
(See page AM).) Similarly the jacupirangitic type, yamaskite, form.«
dikes cutting the esscxit(» of Yamaska Mountain, Quel)ec. Other dike*
of salic n(»phelite symite. nUo cutting the essexite, have the appearance
of being complementary to the yamaskite.'
Complementary Dikes of the Alkaline Clans.— Dia.<«(*histic dike^
have had a disproportionate share of attention in many discu-ssions on
magmatic diflfrrentiation. As a rule they are peculiar in chemical
composition and their magmas are not repn^sented in stock, batholith.
great laccolith or lava Hood. The so-called com piemen tar>' dike^ are
the relatively minutr proiluct< of magmatic splitting which ha«« doubt-
» R. Lipsins. ('.inl.mi.- • * "\tflchl:in(i. Part '2, 1903, p. 61; A. Sauer, ErUut .
Srktion Kiii>f«Tlitrc. Ivs'J ^'ioficnthal. 18H4 (Cieol. Sun-oy of SuoDj'i-
' (i. A. Vo» ^ Ada. Vol. Irt. 1 t H. 1906, p. 36.
ALKALINE CLANS 445
less taken place according to methods different from those responsible
for the larger rock-bodies. The complementary dikes of the alkaline
clans, as of all the others, seem to find best explanation in two different
principles: gaseous transfer, and the "squeezing-out of residual
magma" in the freezing stage of an abyssal wedge or of its satellite.
CHAPTER XXI
PERIDOTITE CLAN AND MAGMATIC ORBS
Included Species
Most of the ''ultra-basic/' fomic species of igneous rocks niAy, for
I)resent purposes, be grouped in a "peridotite clan." It will include
the eruptive pyroxenites as well as peridotites proper. Many large
lx)die.s of iron, chromium, copper, nickel, and sulphur ores are magmatic
in origin and syngenotic with members of the pcridotite clan. Such
ores will be briefly discussed in the present chapter, though a host of
details must be left to the standard works on ore deposits.
The pcridotite clan includes the rock species named as follows:
Plutonic Types
Kssential C'omponent
Olivine.
Olivine + rhombic pyroxene.
Olivine + diallage.
Olivine + diallage + rhombic
pyroxene.
Olivine + amphibole.
Olivine + amphilmle +
pyrox(»ne.
Olivine + biotite.
Pyroxene (+ amphibole).
Hornblende.
Species
Dunite.
Harzburgite (saxonite).
Wehrlite.
Lherzolite.
Amphil>ole pcridotite; rar.
scvelite.
Cortlandite.
Mica pcridotite, var. kimberl-
ite.
Pyroxenite; vars. websterite,
bronzitite, hypersthenite,
diallagite, kosvite.
Ilornblendite.
Kffufiivc Types
Picrite.
Picrite jwrphyrite.
CIkneral Statement of Origin
The eclectic tlieory inii)lies that these rocks arc extreme differ-
entiates of the primary basaltic magma or of basaltic syntcetira*
PERIDOTITE CLAN AND MAGMATIC ORES
447
The facts of the field can only be explained by recognizing two distinct
modes of differentiation. In the one case, the unit of differentiation is
a small, "dry" mass (either liquid or solid) of the ultra-femic silicate;
in the other case, the unit of differentiation is a solution of such silicate
in volatile matter which acts as a transferring agent. Segregation of
the units of the first type has clearly been responsible for the larger
bulk of the rock species listed. It is a process inevitably expected on
the eclectic theory, whether the primary basalt spontaneously differ-
entiates or whether it splits after the solution of foreign rock. Most
of the material actually present in a peridotite should, however, origi-
inate in the primary basalt.
Relation to the Gabbro Clan. — Rosenbusch and others have for-
cibly indicated the close genetic connection of many dunites, wehrlites.
Fig. 200. — Section of composite sill in Greenland. (After A. Heim, Medd. cm
Gronland, Vol. 47, 1911, p. 203.) S, Cretaceous sandstone; P, peridotite sill;
By basalt sill. Illustrating close association between basalt and peridotite.
amphibole peridotit€s, hornblendites, Iherzolites, p3rroxenites, and
picrites with basalt, gabbro, or diabase (Fig. 200 and 201). Wehrlite
is a "feldspar-free olivine gabbro." Harzburgite is a "feldspar-free
olivine norite,'' and is intimately associated with norite in the Hars-
burg region itself. All or nearly all these types are locally charged with
basic plagioclase and thus show actual transition into gabbro, norite,
or diabase.
The gravitative differentiation of anorthosite, as described in
Chapters XII and XV, implies the segregation, in depth, of the mono-
mineralic dunite magma, the bimineralic wehrlite or harzburgitic
448 ir.NBOVS ROCKS AND THSIR ORIOIN
Diugmas, or the triminrralic IhorsolitJc maffma. The c
this view is supported by several principal facts additioiial to thoae
mentioned in the last paragraph,
1. Obvious difTcrentiation in place has caused the dunitie, hypcr-
sthenitic, wehrlitic, and Iherzolitic nodulcH, schliers, or layers in banlta,
pyroxene andesites, gabbros, and anorthosites.
2. In recent years a number of cases have been recorded, showing
the gravitative assemblage of such units. Some ti these may here
be recalled. Dunito, diallagit«, and other pyroxenites, are found at
the floor of the Duluth laccolith.* Pjrnn-
enite and femic norite occur at the northen
contact (probably the floor) of the Chi-
Iwugaroau anorthositic intrusion. (See
page 326.) Homblendite and pjrroienite
are developed in the Glamorgan gabbro
along its northern contaet, which appears
to be its floor. (See page 327.) Pjrroxrn-
itcs and ultrnfemic noritea are floor phases
of the Bushveldt laccolith, Transvaal, and
of the Preston gabbro laccolith, Connecti-
Fio. 201.— PlHnofur<K-k- cut. (See pages 350 and 353.) Toward
(croup in Scotlamf. (Afirr J. the Iwttom of the thick Insiswa sheet of
W. Judd (juart. Jour. Geol. ^^^ Gnquoland, the olivine gabbro and
Soc, Vol. W. 1M8.1, Tf. 35B.) -, , ■ • i * - j . i ^
/>. dunite; F. pURiorU- «■«- "''"'^ '«<^o'»« mcreas.ngly femic and nt last
ri-Kaiion in the dunitr; <1, peridotitic* Wehrlite and olivine-«Dck
icabliro vein ruttinf; thr dun- (serpentine) scem to be floor differentiates
itr; PD, jwrphyritic dunite in at least one intrusive sheet <rf the Sinni
outtinn the ipibbro. liiustra- vallev, Italy (Fig. 202). These Italian
tinK the rioac ossooialion of . ,' .... , i ■ i. ■. j
e.1,1,,0, dunite, and Md.p» pcndotitic rocks .rc overiain by nonte ud
ro^]( gabbro passmg mto plagioclasite at the top
of the intrusive.* Duparc and Wyvsotsky
emphasize the density stratification so wonderfully repeated in sem
different areas of peridutites in the Ural mountains. In each dis-
trict dunite passes upward into p>Toxcnitcs, which in turn are over-
Iain by gabbroid types of inprt'a.-iinK acidity.*
Other floored injections illustrate gravitative differentiation oo
the way to producing species t hat iK'long to the peridotiteclan. Iddings
' W. S. Baylcy, Jour. Gcol,, Vol. 2. 1894, p.814; Vol. 3, 18B5.p.l.
' A. L, du Toit, 15th Ann, Rept. Geol. Comm., C&pe of Good Hope. 1910^ p.
111.
>C. VioU, Boll. R, Cum. rpoI. d'ltnlia, 1892, p. 105. Cf. Ronabuack'l
.Mikroakapiwhfl Phj-sioicraphic dcr Mw i Gnteine, 4te Aafl., 1907, p. MS.
<!.. Duparc, Archivra dm Sricff Nat., Vol 31, 1911 (OoMVa): K.
WysMtsky, M^m. O t* f-*^. dr livr. 83, lOlS.
PERIDOTITB CLAN AND MAGMATIC ORBS
Fio. 202.— Section of Sinni VaUey, Italy. (After C. Viola, BoU. R. Com. geol.
d'Itali&, Vol. 23, 1892, p. 105.) L, Eocene limestone; A, Eocene arfpllite; S,
peridotite (serpentinized) ; G, gabbro passing above into plagioclasite; N, norite;
tolid black, granite and aplite; P, Pliocene conglomerate, etc. Horiiontal scale,
1:30,000.
1. — Map of the Palisadea sheet, New Jersey. /, oryatalline roekn;
'iTiiuncsedimentB; 5, intrusive diabase (Triassic); ^, extrusive basalt (lYiuHo);
'i (^etaceoiu and later sediments.
450 IGNEOUS ROCKS AND THEIR ORIGIN
found a decided enrichment of augitic material at the bottom of an
intrusive sheet at Electric Peak, Yellowstone National Park.^ A
phase of the Palisades (New Jersey) diabase specially rich in oUvine
haA l)een d(^cribe<I by Le^vis as due to settling of the olivine substanM
in this great intrusive sheet ^Fig. 203).' Some of the floor phaaee in the
Purcell sills of British Columbia (page 344) approach homUemfite in
comj)osition.
3. The argument for the spontaneous derivation of pyroxene and»-
ite and pyroxene porphyrite from primary basalt (page 375) is in mani-
fest relation to the problem of the peridotites. That explanation is
amply supported by the close field association of the andesites with
various [xTidotites as well as with undifferentiated basalt. Instances
of such association' are so abundant that it is unnecessary to go into
details in this book.
In the nature of the case volcanic pipes can but rarely demonstrate
the advanced spontaneous splitting of basalt under the control of grsT-
ity. Yet the simultaneous eruption of augite andesite and an oltra-
feniic basalt at the active vent of Reunion has much value in illostra-
tion of the general process. (See page 228.)
Ultra-ferromagnesian and Ultra-cafemic DifFerentfates of Syntac-
tics. The eclectic theory holds that many rock. bodies belonging to
the iHTidotite clan are gravitative differentiates of syntectics. The
precipitation of magmatic units rich in iron, magnesium, and lime is
to be expected, partly t>ecause of the mere fluxing of basalt by bane
stHliinents, partly because of inoculation with and supersaturation by
the foreign lime and other substances. The character of the precipi*
tnte sluuild vary with that of the absorbed material. It is, therefore,
iinpi>rtant to observe that the ultra-basic differentiates of the alka-
line magmas are those appropriate to the limestone^^yntectie
hypothesis.
Hoseiibuseh has pointed out the syngenesis of pyroxenites with
iiienibers (»f the gabbro-<liorite series and with members of the rilf^1wfc»
nri ies, such as essexites, monzonites, shonkinites, and theralites. The
**nlk»iine*' pyr(»x(*nites are exactly the rock types which should be
u^\^'^^ readily formed in the depths of a syntectic of carbonate rock with
hu-tnlt or with one of its many differentiates. Other members of the
pel idotite rinn sluuiId be far less commonly developed in the syntectiCt
)iiul Mw\\ i^ t he faet.^ Hosenbusch gives examples of the differentiation
' ^ V liMtiiKH. Monograph 32. Part 2, U. S. Geol. Survey, 1890, p. 88.
* J \ U^w w, Ann. Hrp. State OeoloRist of New Jersey, 1007, p. 131.
* \ \iiiip:ti^ MiMiioir No. 3S, Cool. Surv., Canada, 1012, p. 782.
* S«Hs U l\iiit«*nl)iiHrli, Mikroflkopischc Phyfiiofcr&phie der MsMigen
4l«i \k\\\ . ltHl7. p 4.V2.
PERIDOTITE CLAN AND MAGMATIC ORES 451
of pyroxenite from alkaline magmas. Here we need note only a few
of the more or less celebrated associations. This cafemic type appears
with: the Duppau theralite of Bohemia; the Monteregian essexite of
Quebec; the Gran essexite of Norway; the monzonite of Monzoni, and
the syenites of Predazzo, Tyrol; the phonolites, etc., of the Sundance
quadrangle, Wyoming-South Dakota; the phonolites of Southern
Abyssinia; the tephrite etc. of the Cape Verde Islands; and the nephe-
lite syenites of Ontario.
The relation of melanite pyroxenite to the syenites of the Loch
Borolan laccolith is a striking case in point. Shand suggests that their
Jerivation has taken place within the laccolithic chamber itself. (See
page 439 and Fig. 196.)
Olivine-rock segregations in place are fairly common in trachy-
lolerites and have been found in the leucite basalt of the Gaussberg,
Antarctica,^ and in the famous nephelite syenite of Alno.* True peri-
iotite fpicrite) and olivine gabbro accompany the nephelite syenites of
the Port Coldwell region, Ontario, but it is not known whether the
peridotite is a differentiate of the alkaline magma rather than of the
mbalkaline.' The syngenesis is more evident in the intrusive sheets
carrying both "picrite" and teschenite, near Teschen, Austria; at
[nchcolm Island, Lugar, Barnton, Ardrossan, Lethan Hill, and other
localities in Scotland.* True (effusive) picrites accompany the many
ilkaline species at the La Palma "caldera" of the Canary Islands, but
feldspar basalts also compose part of this volcano and again doubt must
Bxist as to the immediate affinities of the picrites.* Exactly the same
problem faces the petrogenist dealing with the great igneous suites
in Tahiti and Reunion. ^ The delicacy of his decision illustrates again
and again the exceeding closeness of the genetic bond between the sub-
alkaline and alkaline rocks. Any general theory must recognize the
fact emphasized by Rosenbusch, that the true picrites are generally
derivatives of diabasic or basaltic magma. Why, then, are they so
often found in many typical alkaline provinces? For the eclectic
theory this question carries no mystery.
As noted in the fourth edition of Rosenbusch's hand-book (p. 457),
Carvill Lewis and Lacroix have remarked the "alkaline" affinities of
South African kimberlite, which, according to Rogers, is linked geo-
^ R. Reinisch, Deutsche Stidpolar Expedition, 1901-1903, Berlin, 1906, Bd. 2,
Heft 1, p. 75.
»A. G. Hogbom, Geol. Foren. Stockholm Forhand., Vol. 31, 1909, p. 356.
* H. L. Kerr, 19th Ann. Kept. Bur. Mines Ontario, Vol. 19, Pt. 1, 1910.
*See pages 243 and .435; and 0. W. Tyrrell, Trans. Geol. Soc. Glasgow, Vol.
13, Pt. 3, 1909, p. 298.
' C. Gagel, Zeit. Ges. fur Erdkunde, Berlin, 1908, pp. 168 and 222.
* A. Lacroix, Comptes Rendus, Vol. 151, 1910, p. 121; Vol. 156, 1912, p. 538.
452 IGNEOUS ROCKS AND THEIR ORIGIN
logically with the abundant mclilitc b&salt filling many volcaiue venU
in the region.^
Mast wehrlites are differentiates of gabbroid magma; but acme are
more or less certainly derivatives of alkaline magnuuk RoMnbusrh
notes a wehrlitic dike associated with the alkaline types of MonaonL
The present writer has found wehrlitic segregations in the Hawaiian
trachydolerite; Boulton finds them in the monchiquite of Golden Hill,
Monmouthshire, England.'
Hornblendite is a large-scale differentiate in the Glamorgan gabbro
intrusion of Ontario and may have been segregated by gravity. (See
page 327.) The anorthosite of Chibougamau, Quebec, is accompanied
by dikes and sills of hornblendite, pyroxenite, and dunite, in such re-
lations that they appear to represent types actually complementary to
the feldspar rock.' The hornblendite of the Sierra Nevada, California,
is transitional into olivine gabbro.^
On the other hand, hornblendite forms segregations in alkaline
magmas; for example, in the Norrbotten syenite of Sweden.* It u
one of the femic differentiates of the essexitic magma at Gran, Norway.'
Reviewing the ground this rapidly traversed, we observe that ei-
pectation of gravitative control in the differentiation of peridotites it
fully met by the facts concerning sills and laccoliths, which again most
eclipse all other types of igneous Ixnlies in giving required informataoo.
The principle of gravitative control explains: (a) the comparatiTC
rarity of peridotitic intrusions, for the eruption of an abyssal phase if
manifestly less readily accomplished than that of the salic, overlying
phase; (h) the relatively small size of every mass belonging to the peri-
dotite clan; (c) the prevalence of the dike form of intrusion for peri-
dotites and pyroxenites; (d) the general absence of vesicular stnittiin
in the picrites, since the volatile substances in a differentiated msgnt
rise away from the ultra-femic phase.
Species Formed by Gaseous Transfer. — Some of the homUenditi*
seem to have been eniplaced under conditions analogous to those of
ordinary pegmatites. The meta-<liorite of the Mother Lode distrirt,
California, parses peripherally into ver>' coarse-grained homblendiJe
carr>*ing accessor}' epidote, muscovite, and quartz.^ Both the griis
of the mass and the character of its accessory minerals suggest thi*
* .\. W. Rofcers, GooIofO' of Cape Colony, London, 1905, p. 346.
« W. S. Boulton, Quart. Jour. G«>1. Soc., Vol. 67, 1911, p. 460.
' A. E. Barlow. J. C. Ciwilljm, and E. R. Faribault, Report on tbs CUbo«-
gamau Region, Quehor. 1911, p. WA.
* Jackson Folio. V. S. Geol. Surwy, 1S94. p. 4.
» P. C.eijtr. Cn^A, For. StiMkht»liii Fi.rhand.. Vol. 34. 1912, p. 183.
* W. C. BhijKgor. Quart. Jour. Ci«h»1. Soc., Vol. 50, 1894, p. 15.
' Mother Lode District Folio, U. S. Geol. Surreyp 1900, p. 4.
PERIDOTITE CLAN AND MAGMATIC ORES 463
"' W4, — Map of the Kiruna district. (After Lundbobm and Geijer.) Scale,
1; 120,000.
454
IGNEOUS ROCKS AND TUSIR ORIOtX
analog^'. It may be recalled that the experimental fonnatioD of ui-
philtolc ha-s never succeeded if water is absent from the mdL TV
writer has often observed masses or lenses of homblendite segregatfd
in tlic Hhear zones of dynamically metamorphosed granodioritn ud
fCiiWIiros, in the British Columbia mountains. Clapp has receotly givcB
a i<imilar explanation of the homblendite found in the Sooke gabbco of
Vancouver Island.'
There can be little doubt that these segr^ations are due to gaae-
ouit or vapor transfer, and that such cases tend to corroborate the view
that watcr-goA and other volatile substances have co-operated in Ike
formation of certain hornblende rocks which have been more direetly
derived from magmas. In this connection one may recall Heun'i dmt
illustration of gaseous transfer as responsible for the amphibole (kan^
sutite) segregations at Karsuarsuk, Greenland. (See page 402.)
Hagmatic Ores. — Many small masses of titaniferous iron ona an
obviouiily local differcnliates of gabbroid magma or of ita own derha-
Fia. 20.5. — Socliunii throufch iron ore depoeiu in the Ea^e Mta., CUift
(Aftn E. C. Harder, Itiil). .'.03, V. S. G. S., 1912, PI. 1.) VQ, vitraoiu
D, dolomite; Q, quartzilp; .U, quorti nioDionitc. Scale, newly 1:6,000,
tive.", anortliosite, etc. In his admirable statement for the Adiroa-
(lark ores, Kemp expresties the view that the differentiated oxides, like
the silicate t:ul)-ni.iginas, separated while still in the liquid state.'
Vogt, another sixTiali.-it on this problem, agrees with Kemp.*
That gravity h:u> often eontrolle<l the segregation of the laiger
l>udie.4 of iiiagniatic iron ore is already clear. These form ngnificsB^
lluor phases of the Bushvelilt and Duluth laccoliths. Other maaie*
have unilateral <listribution in the great (prot>ably laccolithic) injection^
of the Iterfim ili^Irict: of (.ilumorgnn ton-nship, Ontario; and of tb'
■C. 11. ri:ipi>, Mi-iiioir No. 1.^ r.pol. Sun-. Cantula, 1912, p. 123.
' J. F. Kom)>, lyth Ann. Itcp-. 1'. S. Ctol. Survey, Pt. 3. UBB, p. 417.
• J. II. L. Vogt, Norak lifol. Tidukrift, Bd. I, No. 2, 190S, p. «.
PERIDOT IT E CLAN AND MAGMATIC ORES 455
Chibougamau district, Quebec. (See pages 326 and 332.) The justice
of the conclusion reached by Coleman and Barlow, that the sulphide ores
of Sudbury are magmatic differentiates segregated by gravity, is shown
not only by the plain facts of the local field but by the recent discovery
of a perfect parallel in the Insizwa sheet of East Griqualand. (See
page 349.) Ores of copper and of chromite also occiu- along the lower
contact of the Bush veldt laccolith.^
The relation of the very large iron-ore deposits of Kiruna, Sweden,
to the neighboring quartz porphyry is strikingly similar to the rela-
tion illustrated in the gravitatively differentiated laccoliths just men-
tioned (Fig. 204). Geijer considers that the Kiruna ore and quartz
porphyry represent independent eruptions, probably extrusions.* On
the other hand, none of the facts yet published is incompatible with the
view that these Swedish ore-bodies are concentrations, in place, from
the quartz porphyry body, whether that body is intrusive or extrusive.
This conception should certainly be tested in the field.
It is not yet possible sharply to distinguish some magmatic ores
from those which have been concentrated by aqueous solutions at low
temperatures. Nevertheless, certain ores have clearly been segre-
gated by gaseous transfer in molten magmas. Geijer describes an
instructive example at Nasberget, Sweden. (See page 402.) Some
of the magnetite bodies developed at the contact of limestone and in-
trusive rock have features suggesting pneumatolytic origin (Fig. 205).
Further illustration of this topic may well be left to the standard works
on ore-deposits.
^ H. A. Brouwer, Oorsprong en Samenstelling der Transvaalsche Nephelien-
syenieten, 's Gravenhage, 1910, pp. 9 and 28.
« P. Geijer, Geology of the Kiruna District (2), Stockholm, 1910, p. 269.
CHAPTER XXII
ECLECTIC THEORY APPLIED TO THE NORTH AMEBICAH
CORDILLERA
A review of facts detailed in the last seven chapters appean to
give a certain sanction to the general theory. Those facta have beta
selected from a countless number concerning sin^e igneous
families, and clans throughout the world. Imperfect original
tions and imperfect interpretation of map and printed page are obvioai
dangers to the investigator striving to find his way throu^ the msse.
Yet the wTitor feels the value of a mental scheme toward whUtt Ae
explanation of each of the rock clans in turn seems to point. He
believe that so many converging rays end in a kind of wiU o' the
leading one away from the main road to a correct geologleal phi*
losophy . The honesty of the light is proven when the g^ieral theory
matches all the interlocking observations to be made in a eoi
sive eruptive area. That area must be large enough to include
of the so-called ''petrographical provinces." Its close sdaatifa
analysis will reveal any vital weakness in the structure of the theory and
will determine its prophetic value.
Even if the ^xiter wore able to make so full a comparison, hk re-
sults could not be stated in one volume or two. HoweveTi qiace in Ae
present one may l)e taken for an outline of some salient consideratioof
affecting one of the ideal regions.
The motive which has impelled the writer to develop the geMrsl
theory was supplied during nine years of work on the geologj of Ae
Cordilleran belt at the lK>undary between Canada and the United
States. The correspondence Ix^tween the theory and the maiqr fatti
then discovered is considered in Memoir No. 38 of the Geological Sa^
vey of Canada (1012). The l)elt crosses a half-doien large piufineci,
each of which Invars intrusive and extrusive formations. CompiliBg
the information derived from the l)oundary survey with that won fron
field work and laboratory studies in areas to the north and to the sooth,
the writer will now apply the ecle(*tic theory, point by point* to the
North American C'oniillera as a whole. Insufficient knowledge of the
field must naturally involve some uncertainty in the result but^ on the
other hand, any systematic theory of petrogeny has its chief value in
being a guide to the future increase of knowledge. The proposed sys-
4A0
THE NORTH AMERICAN CORDILLERA 457
tern is so complex that a final illustrated summary, parallel to that in
Chapter 14, may be helpful to the reader.
The igneous geology of the Cordillera is essentially like that of the
general continental surface of the earth. It is the largest well-defined
region of post-Huronian eruptivity, containing the greatest exposed
batholith, the most extensive area of fissure eruption yet mapped, and
one of the world's longest stretches of volcanic vents. All igneous-
rock families and clans are represented. Excepting the granodiorites,
the species are found in individual and total volumes roughly like
those of similar areas in the other continents. In Washington's
'* Chemical Analyses of Igneous Rocks published from 1884 to 1900,"
573 superior analyses of Cordilleran rocks are recorded. Adding to
them 44 analyses made for the memoir on the geology of the inter-
national boundary, the writer has calculated the average silica per-
centage to be 59.97 (unreduced computation). This value is nearly
identical with Clarke's average silica percentage calculated from the
analyses of rocks of the whole world, and with that calculated by Harker
from 536 analyses of British rocks. ^ Calculations of the other oxides
would show corresponding similarity. Though containing several
unknown geological and psychological factors, the Cordilleran, British,
and World averages tend to strengthen belief that the earth is homo-
geneous in the very heterogeneity of its igneous products.
The existence of the three principal earth-shells below the Cordilleran
surface is to be inferred from the structural geology. Beneath the
geosynclinal sediments of Paleozoic and later dates is an acid pre-
Cambrian terrane. Today, this outcrops in many places where erosion
has denuded the tops of local upwarps. The former exposure of much
larger areas of the acid pre-Cambrian terrane is proved by the composi-
tion of the '^Beltian" and Paleozoic sediments in the Rocky Mountain
geosynclinal, which sweeps from Mexico to Bering Sea. The quartzose
character of the oldest pre-Cambrian sediments themselves, as exposed
in British Columbia, Montana, Colorado, and Arizona, shows that the
unknown lands from which these sediments were derived were of quartz-
ose (probably granitic or gneissic) composition. Thus, as far back as
we can yet penetrate into geological time, the surface formations of the
Cordilleran region were highly silicious. Most of the visible pre-
Cambrian basement is constituted of intrusive granite, exactly as in the
Canadian and Fennoscandian shields. Evidently, therefore, the ex-
isting complex cannot be regarded as an original earth-shell. However,
the argument of Chapter VIII, that the visible pre-Cambrian granites
must essentially be the products of a remelted acid earth-^heU, is as
valid for the Cordilleran pre-Cambrian as it is for any other region of the
1 See F. W. Clarke, Bull. 491, U. S. Geol. Survey, 1911, p. 25.
458 IGNEOUS ROCKS AND THEIR ORIGIN
globe. The compelling reason for this belief is found in the composition
of the ocean. That the conclusion applies to the Cordiileimn section
is suggested by the general similarity of the pre-Cambrian gjtoiogj here
and in the more extensive pre-Cambrian areas of the world.
The existence of a continuous layer of erupiibU ba9aU beneaih Ae
Cordillera is to Ix^ inferred from facts of the kind noted in Chaptrr
VIII. Those specifically applying in the present instance may be sum*
marized under headings ris follows:
1. The clearly exotic nature of the Cordilleran basaUs of all agei.
They have characteristically penetrated the pre-Cambrian basement
and the overlying geosynclinals through narrow fissures in which im-
portant solution of the visible terranes was impossible.
2. The equal impossibility of believing thai these baaaUM are pod^
**Archean " differentiates of primary magma. Nowhere in the Cordiilen
is there exposure of the complementary submagma expected on thi«
hypothesis.
3. The persistence and uniform composition of the baaatie erupied
during late pre-Cambrian and subsequent time. There are no demon-
strable chemical differences in the basalts which were actually erupted
during the early pre-Cambrian (Shuswap series of British Columbia),
and during the ^'Beltian/' Cambrian, Carboniferous, Triassiet Jur-
assic, Eocene, Oligocene, Miocene, Pliocene, Pleistocene, and Recent
periods.
4. The general distribution of basaltic (gabbroid or duUnaie) bodies
throughout the Cordilleran region. They are numerous in most of the
states, provinces, and territories, from Alaska to Southern Mexico.
5. The enormous volumes of the basalts issuing as fieaure empHons.
These are of many dates, including an early pre-Cambrian period
(Shuswap series of British Columbia), the "Beltian" period (Unkar^
Chuar group of Arizona), the Middle Cambrian? (40th Parallel), the
IVnnsylvnnian (Southern British Columbia, etc.)i the Triassic (Soatk-
ern British (*olumbia, etc.), the Eocene (Washington), the (Migoccnr
(British Columbia), the Miocene (Washington), and the PUocenf
(Idaho, etc.). The mere size of those bodies does not, of oourse, prore
the existence of a continuous substratum, but it makes any other pub-
lished hypothesis of their origin extremely difficult of acceptance.
So far as it can l)e checked by field observations, the principle of
abyssal injection seems to be verified in the Cordilleran ge<dogy. Tk
thousands of basaltic, diabasic, or gabbroid dikes are the actoil
fillings of fissures which must extend to great depth, since the vsrt
basaltic floods just noted have issued from fissures averaging lesi thtf
50 feet in %\ndth.* Temperature gradients in the region are generilij
* See, for example, the Mount Stuart folio of the U. S. QeoL Buifiy,
THE NORTH AMERICAN CORDILLERA 459
normal in quality, implying a minimum depth of about 25 miles for the
substratum.
That the lower part of this thick "crust" is subject totensional
stresses or else is condensihle under magmatic pressure is quite clear.
The proof is seen in the fact that these deep-reaching Cordilleran dikes
are composed of material which cannot be essentially due to fluxing
of the intruded (chiefly acid) rocks. No alternative explanation seems
tenable.
Abyssal injection as stated in the eclectic theory involves a tendency
to dovmwarping in the invaded zone. In spite of the relative impene-
trability of the shell of compression, the magmatic material should
occasionally work its way, through fissures in the downwarped area, to
the earth's surface. Contemporaneous vulcanism is thus expected as
an occasional event during the thickening of geosynclinal prisms.
On page 186 will be found the table containing many actual examples
which are taken from the Cordilleran geology. A few of these may be
specially recalled. In the Hozomeen division of the Cascade range
(see No. 9 in the table) the downwarp affected a Lower Cretaceous-
Jurassic land-surface, which was deeply covered with volcanic breccias
before the thick Cretaceous geosynclinal was deposited. Similarly,
the erosion surface limiting the Pennsylvanian limestones near Kam-
loops, British Columbia, is almost directly covered by the thick basaltic
extrusions of the Triassic (Nicola group) . The thick Tertiary basalts of
Oregon rest on an eroded mountain range, the surface of which, at
the Columbia river, has been downwarped well below sealevel. Is it
possible to believe that the crustal deformation is thus again and again
associated with vulcanism as a result of pure accident?
Many facts derived from the Cordilleran field prove superheat in
the primary basalt. The narrowness and great height of the basaltic
(and diabasic) dikes; the thinness, imiformity in thickness, and wide
expanse of multitudes of basaltic lava flows, are all familiar to the
geologist working in the "Belt terrane" of Montana, Idaho, and Brit-
ish Columbia, and in the lava fields of the northwestern United States
or of Mexico.
The passage of superheated basalt through fissures must soon tend
to cause limited solution of the wall rocks. Such is the preferred ex-
planation of the quartz-bearing diabase found in many of the feeders of
the Eocene (Teanaway) basalt eruptions in Washington State. Some
of the Eocene flows themselves, like others of Miocene age (Yakima
basalt), are similarly acidified.
Certain intrusive sheets of gabbro, injected from the main Tean-
away fissures, carry interstitial micropegmatite. Similar material
is found in dozens of the thick gabbroid sills in the silicious Belt terrane
81
460 IGNEOUS ROCKS AND THEIR ORIGIN
of the Purcell mountains and the neighboring ranges of Idaho and
Montana. In general the acidification is proportionate to the thiek-
ness of each sill. In some of the thickest sills this silicious ingredient,
so foreign to normal basaltic or gabbroid rock, has been assembled bjr
gravity into distinct layers. Its mean chemical compositkm is then
determinable with some nicety. In the Purcell sills it is nearly iden^
tical with that of the invaded feldspathic quartxites, which are rda-
tively uniform through a great vertical range and may be ehsmically
averaged. The similarity between acid igneous phase and quariiite
also extends to mineralogical composition. In very few other re-
gions are the conditions so favorable for testing the hypothesis that
thick injections of basalt normally assimilate th&r country roeks.
The Purcell sills clearly confirm that view and other Cordilleran sills,
also discussed in Chapter XVI, illustrate its truth.
The horizontal extension of an individual sill in the CcMxlilleran re-
gion seldom, if ever, approaches that of one of the greater South
African sills, but, as in those cases, proves the abiliiy of magmA to
migrate laterally from the feeding fissure for long distances. The great
areas of these sills in ground plan go far to strengthen bdief that the
wide gabhro masses in western Oregon (Roseburg-Port Orford region)
and the still vaster ones in Minnesota, Wisconsin, South Africa, etc.,
are of laccolithic origin and are definitely floored. Such bodies are
liable to break through their roofs and initiate "subordinaie** soleafioes,
contrasted with the **principaV' volcanoes which are situated on
the main abyssal wedges. The principle of horiaontal migratioB
of magma is also of great importance in the question as to the origin
of certain eruptive rocks.
Though not the longest on record, the CordUleran basaltie dikee
include many which are in length of the same order as ordinary bofk^
liths. Hence, so far as that dimension is concerned, the ^™iffM^ of
dikes — the visible magmatic .wedges — tends to corroborate the idea
that batholiths are thick magmatic wedges, chemically and physieaQy
modified because of their own great supply of heat.
The eclectic theory holds that the greatest wedges are imjeeki
from the substratum during or immediately after energeUe mevMlais
building. The Cordilleran batholiths have, in fact, been intruded at
such epochs; the \^Titer knows of no exception to the rule. (See Table
VI, page 98.)
Usually the batholiths should be intruded along or near the ani
of the mountain chain, there to solidify as "central'' granite. Strong
overthrusting ("charriage") may, however, transfer a large part of a
folded geosynclinal to a new position outside the axial aone; in that
case batholithic masses are not necessarily expected in the overthrust
THE NORTH AMERICAN CORDILLERA 461
block. This principle may possibly explain the absence of batholiths
in the Rocky Mountains (proper) of Canada and northern Montana.
An abyssal wedge composed of superheated basalt and of batho-
lithic size must stope down the roof and wall rocks and dissolve them.
Stoping is admirably illustrated in practically all the batholithic prov-
inces of the Cordillera — the Alaska-British Columbia Coast Range,
the West Kootenay province, the Idaho batholith, the Boulder batho-
lith of Montana, the large stocks of Colorado, the Sierra Nevada, etc.
It was in the Cordilleran region, at Marysville, Montana, that Barrell
independently invented the stoping theory of magmatic emplacement
and there also that the writer, working along the 49th Parallel, be-
came convinced of its value.
Its most important corollary — abyssal assimilation and the large-
scale development of secondary magmas — cannot, by its very nature,
be directly proved in the field. The principle of inference is here
paramount. Cordilleran geology is amply charged with facts which
seem to enforce belief in a positive inference, for most of the non-
basaltic rocks in the mountain chain. A full statement is now clearly
impossible, since it ought to be quantitative and should consider the
relative and absolute volumes of the different bodies of rock. Never-
theless, the folios of the United States Geological Survey furnish con-
venient samples of the Cordilleran igneous geology with a fair ap-
proach to quantitative description for as many local areas. In the
writer's opinion, their synthetic study tends to corroborate the as-
similation theory.
Confidence in the described explanation of the Cordilleran batho-
liths is notably strengthened by the eruptive sequences already demon-
strated in these mountains. The batholithic sequences are orderly,
as shown in Appendix B. Where field work has been appropriate and
suflSciently extended, the order of eruption is found to pass from basic
to acid. The initial product is generally rock of basaltic composition,
either in volcanic masses or intrusive as dikes, sheets, chonoliths, etc.
Then follow the granitic rocks of acidity that normally increases dur-
ing successive intrusions, with aplitic types as the final term of the
series. The eruptive sequence in a given area may, of course, be
complicated by the independence of activity in neighboring magmatic
wedges. These may assimilate contrasting kinds of country rocks, at
contrasting rates, and in contrasting amounts; in addition, the syn-
tectic magmas may undergo differentiation with contrasting results.
In view of so many disturbing factors, it is highly significant that the
general rules above mentioned are well observed in apparently all the
greatest batholithic fields of the Cordillera — ^in Alaska, British Colum-
bia, Washington, Idaho, Montana, and California. The rules also
462 IGNEOUS ROCKS AND THEIR ORIOIN
apply to the different petrogenic cycles registered in a an^e area, as
in southern British Columbia, where pre-Cambrian, Joraane, and
Tertiary cycles are all represented on a large scale.
Again, the assimilation theory demands that the batboUthie magma
should vary chemically with the nature of the country rocki. This princi-
pal test has I)een discussed in Chapter XI and succeeding cbaplcn,
where illustrations are largely taken from the Cordillera of North
America. The facts need not be repeated. The reader may refer to
the pages dealing with the quartz diabases, quarti gabbros, komblende
gabbros, quartz basalts, granites (normal and abnormal), dioritea, quarti
diorites, andesites, granodiorites, dacites, quartz momonitea, monaon-
ites, latites, syenites, trachytes, sodalite syeniteSi foyaitea, and the
nephelitic, analcitic, and leucitic rocks in general. Not all of thaw
Cordilleran types were developed in batholithic chamben; maiqr of
them are formed in smaller ones, including sills and other iigeetioitf
from abyssal magmatic wedges.
In the Cordillera, as elsewhere in the world, inirusive akcaCi katt
extraordinary meaning for petrogenic theory. We have just noted that
these lend indirect but powerful support to the assimilation doetriar
as applied to the primary wedges.
Thus, four chief kinds of evidence — magmatic replacement (h
poration) by hatholiths, the eruptive sequence in batholithic
the systematic chemical variation in batholiths, and the teatimooy of
floored injections — seem to show that abyssal assimilation ia one of the
three fundamental processes by which batholiths and atocka have ben
formetl.
After abyssal basaltic injection and assimilation, the remaining
principle, magmatic diferentiationf is logically considered. Thia does
not mean, of course, that the two secondary processes are ao aeparated
in time. In general, the solution of foreign material in magma ineri*
tably tends to produce some immediate differentiation. However,
many facts derived from the Cordilleran batholiths and atocka afaow
that differentiation contimtcs long after significant asrimHaHon of the
country rock is impossil)le. For example, the wide stoping breccia
bordering the batholith at Trail, British Columbia, ia compoaed cf
sharp-angled, basic xenoliths, which suffered no essential amoimt cf
corrosion by the including granodiorite magma. Thia "*ir"^ ia a
differentiate and has itself continued to split, locally, into aplitie and
vogesitic submagnias. (See page 366.)
More than a score of the Cordilleran sills and laccolitha ahow
graiitativc differentiation. (See Table XIV, page 230.) Bodiea of thii
type have the right of way in the discussion of magmatic aiilltUng;
their significanf' of all proportion to thf im r of aaoh j^jactioaa
THE NORTH AMERICAN CORDILLERA 463
actually recorded. More clearly than any other eruptives they are
witness to the certain control of gravity in the differentiation of batho-
lithic syntectics. The extent of that control in batholiths is, however,
chiefly to be inferred from the observed eruptive sequence. Another
indirect evidence is found in the frequently observed contrast between
a plutonic rock in the Cordillera and its corresponding effusive phase.
On the average, the former is the more femic and magmatically was of
higher specific gravity. This contrast is found, for example, between
the effusive dacites and the syngenetic granodiorites of the western
half of the mountain chain. Though clearly of batholithic origin, the
Yellowstone Park rhyolite is more salic than the average granite or
quartz monzonite of the neighboring Boulder batholith which, in its
comparatively recent date, structural relations, and chemical habit,
is so suggestively allied to the Park rhyolite.
The more subordinate control of gaseous transfer in differentiation is
abundantly illustrated by many intrusive bodies in British Columbia,
Montana, California, etc. (See pages 320, 361, 368, 452, and 455.)
No other mountain chains more tellingly illustrate the relation be-
txveen size of the magmatic chamber and the advance of differentiation.
Examples are described in Chapters XVI and XX, pages 344, 363-6, and
428. In spite of many complications, the facts observed in Montana,
British Columbia, New Mexico, etc. point in the direction indicated
by the eclectic theory, which also accounts for the existence of basic
contact phases occurring in numerous intrusive bodies of the western
mountains. (See pages 237, 366, and 390.) The laccoliths of the
Highwood Mountains, Montana, furnish standard illustrations of
gravitative splitting and the formation of peripheral basic phases by
contact chilling; they also prove the high liquidity and strong
tendency to differentiation characteristic of at least some alkaline
magmas.
The eclectic theory recognizes two kinds of differentiation of pri-
mary basalt which has not been specially affected by syntexis.
Under certain little-understood conditions, intrusions of this magma
split gravitatively, mth anorthosite as the salic pole, Cordilleran
geology has, so far. little to offer on this question, but it has abund-
ant illustrations of the other mode of splitting, that at volcanic vents
of the central type.
The more salic and more voluminous submagma is here a pyroxene
andesite. Theory demands that this andesite shall often be transitional
into normal basalt, as so frequently observed in the Cordilleran field.
Theory expects that this andesite, while very abundant throughout
the world, shall not constitute great fissure eruptions like the basaltic
floods. Our Cordillera clearly matches this deduction as it does the
464
IGNEOUS ROCKS AND THEIR ORIGIN
related deducUon that pyroxene andesite should show evidence of
low temperature at eruption. Under norma! conditjona, intrutiTe
ba.<Rltic magma is rather xtubbom in refusing to differentiate. The
homogeneity of dial>asi> and galibro in mont dikes and ailla and in many
lacroliths and chonolith», independently of size or geological age, is ex-
actly a feature to be expected if basalt is itself a primitive differentiate
of earth magma. Such a. solution should early be brought to ebemieal
equilibrium uniler subsurface conditions. Evidently those conditions
are alt<'red in a volcanic pipe through which concentrated gas is stream-
ing. The ba-^alt there differentiates at a relatively low temperature.
The andesitic submagma may issue in the form d a true flow but it*
high visco^<ity, coupled with periodic freezing of the vent, should more
generally lead to explosion and the formation of pyroclastic deposit*.
The C'ordilleran andesites seem, in fact, to be chiefly pyroclastic, in
formations that date from the pre-C'ambrian to the Pleistocene. They
outcrop at intervals, throughout the whole length of the Cordillera.
Structural geology and existing topc^raphy show that the andesitic
volcanoes of the Cordilleran Ix-lt have l)een habitually ereeUd in lintM
rimghly parallel lo geonyndinal antt orogenic axts. Their arrangement a
intelligiijle on the a.'wuraption that they have been local vents from
primary basaltic wedges in theoretically appropriate relation to down-
warp and mountain range.
The eclectic theory is thus capable of explaining the dominance of
pyroxene andesites among the products of the Cordilleran volcanoes of
the central type, as well as the dominance of basalt in the spectacular
fissure eruptions of the northwestern United States.
In the Cordillera, as in any of the major subdivisions of the earth's
surface, boKnltic lypfi lUiminate among extrugireii, and add diffentUiaUM
dominate among the rixi'Mc inlnmire maase» — a fundamental fact which
it«'lf goes far toward corrol>orating the general theory.
In Cordilleran IkmUcs, lx>th extrusive and intrusive, ottoftM aptatM
are of very umall volume as <:om{>arfii with the mfxiUaline species, whether
individual masses or total quantity l)e considered. This is yet an-
other of the geological truths which ha%-e been too little KglRled by
petrologists.
The fon^ing brief sketch of the huge western field c
mention of many important points which are
ters. However, it is already e%-ideDt that tl
dillera is rich in illuf •=-' "**" * ' TF"? 1
The continued stud*
and admirable exr*
eomplete petngeib
THE NORTH AMERICAN CORDILLERA
465
the coining century of petrological research. Those compiled to test
the eclectic theory appear to support it in general, but the writer's chief
purpose in elaborating the theory is to break ground for a better state-
ment of igneous-rock philosophy for our Cordillera and for the earth.
APPKNDIX A
rABLK \X -SHOU'INii Tin: N'CMIlKlt UK UKI'AltATK DKTEKMINATIOMi USED
IS' COMPL'TINU Till-; AVKRAQl-: (JUANTITY OF EACH OXIDE IN THB KOCK-
TVPK4 L14TBD IN TABLK II. (S» pm> IV)
1 3 94 56 ra »» 11 uuu
SKI.
47
lU
184
236
64
24
40
60
20
TiO,
22
74
60
87
40
10
30
20
12
AM>i
47
lU
180
232
63
23
40
49
20
Ke/),
;w
101
118
158
01
22
39
32
11
FK)
as
101
US
158
42
6
30
32
11
MnO
24
S(i
Hi
93
32
4
28
20
5
.Mr<)
47
114
1S4
230
03
24
39
49
20
i-^O
47
114
IH4
•£W
61
24
40
49
20
S^^i
47
lOS
182
234
63
24
39
49
20
K,()
47
11)8
182
234
oa
24
30
49
20
II:()
:tK
40
41
41
17
15
17
39
14
I'^.
15
:h
7a
81
27
4
23
17
5
16
16
17
18
1»
SO
ai
M
St
SiO,
2
11
50
48
10
7
«
TiO,
1
9
as
20
4
5
5
Al^),
2
11
49
48
10
H
F,-:(l.
2
U)
43
38
0
3
F.^)
2
10
43
38
0
3
MnC)
1
s
38
34
4
5
MbC
11
50
48
10
8
CiiO
2
11
50
48
10
8
N:i-i)
2
11
50
48
10
8
K:<»
2
n
50
48
10
8
n,o
1
8
41
44
10
3
8
P.Ot
1
S
34
25
a
3
39
30
31 33 33
84
36
36
37 SB Sfl
1 4C
Si( ).. 7
12
HI 10 3
3
43
2.)
4 :
8 5
20
Til ). a
11)
10
. 3
30
16
3 3
IS
10 3 10
S4 U M IT M
23 19 7 12 3
23
19
IS
15
14
15
14
7
22
18
22
19
22
19
21 19 7 13
«1 «l tt M 41
12 30 20 n 70
12 15 16 71 ST
II- 10 10 3 3 43 25 4 8 5 20 12 30 20 80 70
12 10 6 3 3 30 18 2 2 4 20 12 21 IK M M
KH) 6 12 10 6 3 2
Milt) 3 10 HI 4 3 2
18 2 2 4 19 12 24 18 86 W
15 1 3 1 20 12 14 II 04 a
12 lU 10 3 a 41
12 10 10 3 3 43
12 10 10 3 a 43
4 8 5 20 12
9 K) 10 a 1 2ti 23
PtOi 3 12 10 4
30 aol Milt J
APPENDIX A
U 47 <a <9 60 61 52 63 U 66 66 67
U 1 U 1 « 1 01 1 63
SiO, (37 33 20 24 10
7
41
198 10120
17
11
9
24
17
6
2
TiO, 51 16 13 13 9
fi
26
132 113 13
6
fl
8
16
10
4
2
A1,0, 87 33 20 24 10
B
41
197 160:20
17
11
9
24
17
4
2
Fe,0,71 25 18 ,18 10
36
174 146'lS
14
h
»
21
16
fi
•1
FeO 71 25 18 :18 ;iO
;jfi
173 ,14618
14
5
V,
21
If,
fi
2
MnO 44 16 14 1 8 1 6
2S
108 1 9613
fl
2
4
Ifi
13
4
2
MgO ;87 ,33 20 24 10
4rt
IflT ;]60|20
17
11
9
24
16
5
2
CaO 187 33 '20 '24 ,10
7
41
198 i 161120
17
11
9
24
17
5
2
NftiO 84 32 '20 22 |lO
40
190 154 20
10
11
9
24
16
fi
2
K,0 ,84 32 ,20 '22 10
3t)
190 154 20
16
11
9
23
16
5
2
H,0 |57 ' 5 lis |24 10
fi
17
56 27 16
1
S
2
12
b
4
2
P,0. ]47 ,14 |13 ,11 , 9
6
27
135 11614
5
4
9
16
11
4
2
lei
S4|SIi>S<l!T,e6
»(w>fi
n
n
74 1 76 1 76 1 77 1 7B 1 79
80
SiO,
12
31715
3
A ^
ti
4
IS
9 m
34
14
6
6
6
•no,
1
3
« 113
11
fl
3
20
26
11
3
2
Alrf).
12
3
fi
4:20
14
10
3
19
34
14
6
6
F„0.
110
3
2
4 I28
14
10
3
19
31
11
6
6
FeO
10
3
2
6 ,29
14
10
3
19
31
n
6
«
MdO
3
1
2
2 l21
7
8
3
13
14
6
4
M,0
12
3
S
R 31
U
in
3
20
•KA
14
fi
fi
ao
12 i 3
J>
4 ,29
14
in
3
20
34
14
6
fi
NiiO
|l2 , 3
3 1...
2 ,17
13
7
3
20
34
13
6
6
Krf)
,12 ' 3
3 ..
1 I15
13
e
3
20
34
13
6
fi
H,0
|7'3
»
4 B
3
6 j29
14
9
3 ,20
22
11
4
6
P*.
i 1 ! 1
3
3 !...
2 ,12
«
4
Sllfi
29
6
2
3
{ai)83ISS{ M 1 85 l«|87|U
H
M H
M|«
M
M
M
97
SiO,
24
20 t 4
20
Ifi
4 1
4
7
2
9
26
5
«
TiO,
8
14 1 4
4
11
3
2
fi
2
S
23
S
AW.
21
20
4
20
IB
4
2
7
2
9
26
fi
Fe,0.
22
19
3
19
15
4
2
B
2
3
26
6
FeO
22
IS
3
19
15
4
2
fi
2
3
25
fi
MnO
,14
7
1
13
fi
2
2
4
4
15
4
MeD
24
20
4
20
16
4
2
7
2
9
26
6
CbO
24 120
,24 bo
4
20
16
4
2
7
2
«
26
fi
X.rf)
4
20
1R
4
2
7
2
9
26
Ji
a
Kfl
i",l
lA
4
2
7
2
e
2ff
A
3
I tUk..
Jm
2
3
4
2
jt
2S
4
s
■■■
m
2
S
s
a
«
23
J
3
^^
P
IGNEOUS ROCKS AND THBIR ORIGIN
98
«9
100
101
IDS
toi
IM
m
IM
tan
8iO,
10
10
20
IS
TiO,
10
9
20
9
Al,0.
10
10
20
IS
Ferf),
10
10
20
11
3
FeO
10
10
20
10
MdO
10
10
19
8
MrO
10
10
20
IS
CaO
10
10
20
IS
Na,0
10
10
20
IS
Krf>
10
10
20
IS
H,0
10
10
10
4
3
It
P^.
10
B
16
3
7
IM
109
SiO,
. . 6
8
•no,
,,, 3
AW,
Fe,0.
F«0
MnO ..
MbO
CaO
X..0
K,0
Hrf)
...! 3
rju.
110 111 lU . US I 114 I us : lU
: 15
10
20
4 IS
16 «
9
8
Ifl
4 10
12 S
15
10
19
4 IS
16 i 6
10
8
10
4 10
14 ' 5
9
S
16
4 10
14 . S
B
7
17
1 8
9 2
15
10
20
4 IS
16 1 6
■ 15
10
20
4 IS
U •
15
10
20
4 IS
IS 3
15
10
20
4 tS
IS ; s
10
10
IB
4 10
16 ; 3
5
7
18
4 6
tl 3
APPENDIX B
(if«rifonla7 Una rcpraenJ unanformHiu or other Imff A'
Prb-Caubrian Series
juenay Dutriel, Qitebee (F. D. Adams, Neuea Jabrbuch fflr Minu«logie, et«.
B. B. 8, 1893, p. 464):
1. GreenstoDc, amptubolite (bftsic lavas); ortbc^neiM.
2. Gabbro, anortbosite, norite.
3. Pegmatite.
4. Diabase, augite porpbyrite.
n^ Lake QuadrangU, Adirondaeks (and vieinity} (H. P. Ciuhing, Bull. 116, N«w
York State Museum, 1907):
1. Gabbro, amptubolite.
2. OrthogneisB.
3. Gabbro, anortboaite.
4. Syenite.
5. Granite.
S. Gabbro.
7. Diabase.
iriA Central Wieeontin (S. Weidman, Bull. 16, Geol. and Nat. Hirt. Sarver ot
Wisconsin, 1907):
1. Greenstone (diabase, gabbro, etc.).
2. Granite, quarti syenite.
3. Rhyolite, rhyolite-andesite.
4. Gabbro, diorite, troctolite.
5. Granite, quarts syenite, nephelite syenite, miu tytadtt, pegmatite
ike Superior Dutritt (C. R. Van Hise and C. K. Ldth, Monognqdi £2, U. B. G«o).
Survey, 1911):
A. KeewUin: Greenstone (basalt and andeeite), qoarta porphriT,
rhyolite, felsite, diorite.
B. Lauretdian: Granite, ayemte, orthogneiaeM.
C. Poet-Lower-Middle Hwonian: Granite, syenite, nephdhc iTmita^
gabbro, diorite, rhyolite.
Vfper HMFonian: Gabbro, diabase, basaltf dioritp, fdlite.
'PP«'' Hvronian: Granite.
470
IGNEOUS ROCKS AND THEIR ORIGIN
F. Keweenawan: Olivine gabbroi gabbro, norite, diTUie norile, quwti
norite, quarts g^abbro, diorite, quarts diorite, diab—f, orthodaae
gabbro, granite, soda granite, syenite, monsonite, tnehsrte, rhyoUtc,
quarts keratophyre, quarts diabase, augite andedte, tMndt, hjrpcn-
thene diabase, troctolite, pyroxenite, peridotite, pligjoeliite.
Sweden in general (A. G. Hogbom, Bull. Geol. Inst. Univ. UpuU, VoL 10, 1910) :
A. Lower Pre-Cambrian:
1. Orthogneiss, granite, leptite, amphibolite, unJite porph3rrit«,
dioritc, gabbro, syenite, etc.
2. Diabase.
3. Granite.
B. MidtUe Pre^amhrian: Greenstone ("bamc").
C. Subjolnian: Granite, syenite, g^abbro, anorthotite, etc.
D. Upper Pre-Cambrian (Jolnian): Diabase, olivine diabaie.
Kiruna DUtrict and Vicinity, Stcetlen (P. Geijcr, Geology of the KtrunA DHtrirt,
Part 2, Stockholm, 1910; H. Lundbohm, Geol. Fdr. Stockholm Fdrhand..
March, 1010):
1. Diabase.
2. Soda greenstone.
3. Alkaline syenite and syenite porphyry.
4. Alkaline quartz iM)rphyry, magnetite ore (eruptive).
Finland (J. J. Scdcrholm, Bull. Commission G^l. de Finlande, No. 23, 1907; for
revi8e<l statements see No. 21, 1910, and No. 28. 1911.)
A. Katarchean: Granite, "metal>asite'' (altered diabtte, etc.)-
B. lAulogian: ** Motabasito.*'
C. PoBi-I^oifiau: Granite, diorite, amphibolite, etc.
D. BoUnian: I'ralite porphyrite, plagioclase porphjrrite, ete.
E. PoBt'Bottnian: Granite.
F. Lower Kalevian: "Mctabasite."
G. Upper Kalevian: '•Mctahasite."
II. PoilrKaUinan: Granite.
I. Lower JaliUian: ** Metabasite."
J. Upper Jaiylian: Augite porphyrite, " metabaaile/'
K. Joinian: Diabase, ««v ' -H«at^" granite (RapakiTi)
APPENDIX B
B. Cambrian (7): Gabbro, basalt, abnoruiKl Bill pairita.
C. Carhoniferoua ( Pen nsy Ivan ian) :
1. Basalt, augite andeaite, basic greenstone.
2. D unite, Berpenltne.
D. TriattU: BaaalC, basaltic andente.
E. /unune:
1. Gabbro, diorite.
2. Granodiorife, quartz
3. Biotite granite, apliti
F. Cr^aewaa (KnoxviUe) : Au^te andente.
G. Eocene (?): Basalt and augite andedte of Selkirk range; nephelite
syenite and malignite of Okanagan range.
H. OiiffocCTie (Midway district):
1. Basalt, au^itc andcaite.
2. Mica andesile, hornblende andesite.
I. Miocene:
1. Granodiorite and diorite of Selkirk, Okanagan, Hozomeen
and Skagit ranges; pulaekite of liossland mountains; monaon-
ite of Selkirk range.
2. Syenite porphyry of Rossland mountains, and of Columbia,
Hoiomt-en and Skagit ranges; rhomb-porphyry and "shack-
anite" (an analeitic lava) of Columbia range; alkaline
granite of Okanagan range.
J. PluHxne: Basalt dikes of Okanagan range.
K- Ptgutocene: Pyroxene andesite of Mount Baker (Waafa.).
lem Mcmeo (W. Lindgren, L. C. Graton and C. H. Gordon: Prof. Paper 68, D. 8.
Geol, Survey, 1910, p. 26 and sequel):
A. Pre-Cambrian:
1. Greenstone, amphibolite, rhyolite.
2. Granite.
3. Aplite, pegmatite, diorite.
B. Terliary:
1. Monzonite, quarts monionite, granodiorite, diorite, and por-
phyrire of similar composition.
. Basalt, rhyolite.
. AndcBite, basaltic andeaite, lalite, traehyte.
4. Rhyoiite.
T/-Reeent: Olivine basalt, nepheb'te basalt, phonolite
tt Tertiary?).
472 IGNEOUS ROCKS AND THEIR ORIGIN
Globe Quadrangle, Aritona (F. L. RaoBome, Prof. Paper 12, U. 8. G«oL Snnrcy,
1903):
A. Pr&^ambrian:
1. Quarti-mica diorite.
2. Two-mica granite and muaeovite granite.
3. Biotite granite (Diabaae older than No. 8).
B. Mesozoic:
1. Diabase.
2. Diorite porphyrite.
C. Tertiary: Dacite.
D. Quaternary: Basalt.
British Isles (A. Geikie, Quart. Jour. Geol. Soc. London^ Prea. Addren, VoL 47,
1891, p. 63; T. C. Cantrill and H. H. Thomas, Ibid., Vol. 62, 1906, p.
250):
A. Pre-4Jambrian:
a. Lewisian: Granite, diorite, gabbro, pyroienite, horablcfid-
ite, peridotitc, picrite, syenite.
b. Post'Lewisian of Scotland, in the following ocder:
1. Dolerite.
2. Pcridotite, picrite.
3. Granite, syenite, pegmatite.
c. Vriconian of W(des and Shropshire: Rhyolite, felsite,
mirrogranite, diabase, greenstones.
B. Cambrian:
Bangor district:
1. Rhyolite, quarts porphyry, felaite.
2. Andesites, rhyolite.
St. David's area: Diabase, olivine diabase, rhyolite (ffelsite).
C. Silurian:
a. Arenig:
1. Augitc andesite.
2. Rhyolite.
3. Augite andesite, hornblende andedte.
4. Diabase, porphyry.
b. IJandeilo and Bala:
Caernarvonshire :
1. Rhyolite, andesite.
2. Diabase.
Anglesey: Rhyolite (fclsite), dolerite.
Lake District:
1 . Andesite, basaltic andesite, gabbro, Hiahaee, paaite.
2. Rhvolite.
Scotland: Diabase, fclsite, andesite (7).
Ireland: Feltite, andesite, diorite, quarts diorite, mil
granite, diabase, dolerite.
APPENDIX B
D. Old Red Sandstone:
a. Lower OUl Red Sandstone: Olivinii dirkbaae, diabase, augite
andeaite, trachyte, granite, quartx diorite, minette, Togesite.
b. Upper Old Red Sandstone: Diabase, porphyrite.
Devonian: Doleritc, diabaae.
E. Carfeom/erou* (Scotland);
a. Plateaus: DoleriCe, olivine basalt, andcsite, picrite, lim-
burgite, trachyte, phonoljte, febite.
b. Piiys: Olivine b»aalt, basalt, dolcritc, andeaite, picrite, lim-
burgite, felsite, quarti porphyry.
F. Permian: Basalt, picrite.
G. Tertiary:
1. Basalt, olivine dolerite, gabbro.
2. Rhyolite, felsite, dacite, granophyre, granite, pitchatone.
rmany (R. Lepsius, Geologic von DeutacWand, 2 Liet., Stuttgart, 1887-1910):
A. GrU7idgebirge (all pre-Cambrian?): Granite, orthogneisBes, pegmat-
ite, diorite, gabbro, amphibolite, granite porphyry, mioette, kersant-
ite, alsbachite, malchite.
B. Cambrian (Lower Rhine Region}: Diabaae, quuts porphyry, "por-
phytoid."
C. Silurian (Lower Rhine Region): " Porphyroid."
D. Devonian (Lower Rhine Region):
1. Ivower Devonian ; "Porphyroid."
2. Middle Devonian: Diabaae, diabaae porphyrite.
3. Upper Devonian: Diabase, diabase porphyrite.
E. Carboniferous to Permian (indusive) : Melaphyrc, diabase, olivine
diabaae, augite porphyrite, gabbro, granite, granite porphyry, quartE
porphyry, syenite.
F. Tertiary (Miocene) :
Lower Rhine Region: Olivine basalt, basalt, trachyte, born-
blende basalt, hornblende andeaite, augite andcsite.nephel-
ite basalt, phonolito-
llpper Rhine Region: Olivine basalt, basalt, trachyte, bom-
blende basalt, tephrite, limburgite, melilite baaalt, nephel-
ite basalt, leucite-nephelite basalt, phonolite.
G. Quaternary (Lower Rhine Regjon); Baaalt, leucite' basalt, nephclite
basalt, trachyte, phonolite, leucite phonolite.
apan (Compiled by the officials of the Imperial Geol. Survey of Japan, Tokio,
IW2):
A. Pre-Cambrian: Granite, granulite, amphiboUte, serpentine.
B. Paleozoic: Gabbro, diabase, porphyrite, ampbibolite, peridotite,
pyroxenite, serpentine.
474 IGNEOUS ROCKS AND THEIR ORIGIN
C. Triawie: Porphyrite.
D. Jurnasic: Porphyrite.
E. Cretaceous: Gabbro, diabase, diorite, porphyrite, peridoUte, ir«nite,
quarts porphyry.
F. Tertiary: Basalt, pyroxene andeaite, inie»*honiblende aiidcnC^.
dacite, liparite.
G. QiuUemary: Basalt, basaltic andesite, aufcite andesile.
Po6T-Cambrian Serieb
Bdknap Mountains^ New Hampehire (L. V. Pirsson, Amer. Jour. Science, VoL 22.
1906, p. 507) :
1. Granite.
2. Syenite.
S. Aplite, oamptonito, spcssartite, esscxite.
4. Camptonite, aplite.
Red Hillf New Hampshire (L. V. Pirsson and H. S. Waahington, Amer. Jour
Science, Vol. 23, 1907, p. 446) :
1. Granite.
2. Nephelite syenite.
3. Aplite, paisanite, bostonite, syenite porph>Ty, camptonite.
Tripyramid Mountain, New Hampshire (L. V. Pirmon and W. N. Raee, Amv. Jow
Science, Vol. 31, 1911, p. 288):
1. Granite.
2. Gabbro.
3. Monsonite.
4. Syenite.
5. Aplite.
Aseutney Mountain, Vermont (R. A. Daly, Bull. 209, U. 8. Oeol. Survey, 1901, p.
36):
1. Gabbro, diorite, acid cssoxite.
2. Nordmarkite, umptekite, monionite.
3. Camptonite, paisanite.
4. Alkaline biotite granite.
5. Diabase.
Penohscot Bay Quadrangle, Maine (G. O. Smith, E. 8. Baitin, and C. W. Bnfn,
Folio No. 149, U. S. Geol. Survey, 1907) :
A. Cambrian (7): Diabase, trachyte, s>'emte, riiyolite,
B. Silurian:
1. Pyroxene andesite.
2. Basaltic andesite.
3. Hornblende andesite.
4. Rhyolite.
stfttcment of
C. Zfdle Silurian or Deuonian: ' ■' ■-'■ ;-:i^ii4i9 fwhn-'kfr-iiiiiO
1. Gabbro, diabase, diorite. .'^,im ■^uii'aij
2. Granite. -, ..- >--..-,^ t
Bttex County, MaasachvselU (C. H. Clapp, personal cominiIiiie>^(^; "
A. Posl-Ordovieian (?) :
1. Gabbro-diorite, granodiorite, quartz diorite, granite.
2. ApUte,
3. Diabase.
B. Devonian (?) to post-Lower Carboniferoue: Quarts kwatophyre, fJa-
chyte, dacite, andesite, bostonite.
C. Post-Lower Carboniferom:
1. Pulaekite, umptekito, nordinarkjte, nephelite Eyecile, alka-
line granite.
I 2. Olivine diabase.
ProTimonal ^' ^°'^^*'"P'^' tinguajtp, vogoaite, niinette, cftniptonitc,
I keraantite.
I 4. DiabsEe, with and without biotite.
5. Quart* porphyry and paissiute. ' . , | -, ,
[ 6. Diabase porphyrite. ■■.■■..■i..».iF..'j <•
D. Triaaeic (?); Diabase. - - -
Crandalt and Iihaviooa Quadrangles, Wyoming tAbsaroka folio, U. S. Geol. Survey,
1899):
A. Eocene: Hornblende andeaite, hornblende-micft audesite, dacite,
pyroxene-hornblende andesite.
B. Neocene:
1. Hornblende-pyroxene andesite, jiyroxene andesite, olivine-
free basalt, basalt.
2. BaSalt, Icueite-bearing haealt, ortlioclase-beaiing basalt.
3. Hornblende-mica andesite, hornblende andeeile, pyroxene
andesite, basalt.
4. Hornblende-pyroxene andesite, pyroxone andeaite, olivine
6. Basalt, with quarts phenocrysta.
6. Hliyolite,
SmOo ffaif, Colorado (W. Cross, 17th Ann. Hop. V. S. Geol.Survey, Part2, 1896,
p. 274tf.):
A. Pro-volcanic Period:
1. Granite.
2. Diabase, peridotite, syenite.
B. Volcanic Period:
1. Hornblende-mica andesite.
2. Augite-biotite-hornblende ande«it«.
3. Diorite.
4. Dacite.
5. Rhyolite.
6. Biotite-augite andceilfc
7. Trachyte.
476 IGNEOUS ROCKS AND THEIR ORIGIN
Ouray-SiloerUm Quadranf^f Colorado (U. S. Geol. Sunrey foliot, 1906 and 1007):
(Tertiary sequence).
1. Pyroxene andesite, latite. (San Juan tuff.)
2. SUverton series (in order) :
a. Pyroxene andesite.
b. Rhyolite.
c. Latite.
d. Pyroxene andesite.
3. Potosi series: Quarts latite, rhyolite.
(Diabase, cuts Jurassic; relation to volcanies not known.)
Livingston Quadrangle ^ Montana (Livingston folio, U. 8. Geol. Surrey, 1804):
1. Hornblende-mica andesite, other andesites.
2. Pyroxene andesite, hornblende-pyroxene andesite, basaltic andentc,
basalt, trachytic rhyolite.
3. Rhyolite.
4. Basalt.
Elkhom Distnct, Montana ( W. H. Weed and J. Barrell, 22nd Ann. Rep. U. 8. Geol.
Survey, Pt. 2, 1901, p. 420):
1. Gabbro and diorite.
2. Quarts-dioritc porphyry.
3. Granite.
4. Aplitic granite.
YeUow8tone Park^nake Rircr DistHd (J. P. Iddings, Quart. Jour. Geol. 8oe^
London, Vol. 52, 1896, p. 606):
A. ^octf n€ (Absaroka ranice) :
1 . Hornblende andesite, homblcnde-mica andente, dadle.
2. Hornblende-pyroxene andesite, pyroxene anderite.
3. Andesitic basalt.
4. Great Breccia, including types as in 1-3, with iDtruMOe of
dacite, gabbro, diorite, granite.
B. Pliocene:
1. Basalt, rhyolite.
2. Rhyolite of Yellowstone Plateau.
3. Basalt of Snake River region.
Midway District, Southern British Columbia (R. A. Daly, Mem. 38, Ged.
Canada, 1912):
A. Carboniferous (7):
1. Basalt, pyroxene andesite.
2. Dunite.
B. Late Jurassic (7):
1. Gabbro, diorite.
2. Granodiorite.
C. Oligocene:
1. Olivine basalt, gabbro, augite andesite, ausKe porp^yiHc.
2. Homblende-augite andesite, biotite-augite aaderite^ biolile
andesite.
APPENDIX B 477
D. Miocene (?):
1. Alkaline trachyte, pulaskite porphyry.
2. Rhomb-porphyry, ''shackanite" (an analcitic lava).
Okanogan Range, BriHsh Colurnbia-Waahington (R. A. Daly, Bull. Geol. Soc.
America, Vol. 17, 1906, p. 363) :
A. Carboniferous (7): Gabbro, amphibolite, dunite.
B. Late Jturaesic: Granodiorite, gabbro.
C. Eocene (?): Nephelite syenite, alkaline syenite, malignite.
D. Miocene (?):
1. Granodiorite.
2. Alkaline biotite granite.
£. Pliocene: Olivine basalt.
Mount Stuart Quadrangle, Washington (Folio No. 106, U. S. Geol. Survey, 1904):
A. Carboniferous (?): Diabase.
B. Pre-Tertiary:
1. Peridotite.
2. Granodiorite.
C. Eocene: Basalt, gabbro.
D. Miocene:
1. Hypersthene andesite.
2. Basalt.
E. Pliocene (?): Rhyolite.
John Day Basin, Oregon (F. C. Calkins, Bull. 'Dept. Geol., Univ. California, Vol.
3, 1902, p. 170) :
A. Clamo Eocene:
1. Hornblende andesite.
2. Basic p3rroxene andesite.
3. Quarts basalt.
4. Rhyolite.
B. John Day Miocene:
1. Trachyte (?).
2. Andesite.
3. Rhyolite.
4. Andesite.
C : Basalt.
D. Mascail Miocene:
1. Rhyolite.
2. Basalt.
3. Rhyolite.
E. Rattlesnake Pliocene:
Rhyolite.
San LuisQuadrangHe, California (Folio No. 101, U. S. Geol. Surv^, 1904):
A. Pre-JurorTrias: Granite.
B. Jura-Trais (?): Olivine diabase, diabase, basalt, peridotitei pyroz-
enite.
478 IGNEOUS ROCKS AND THEIR ORIGIN
C. Crelaceous:
1. Dftcite, andesite.
2. Diabase.
3. Gabbro, prridotite, pyroxcnitc.
D. Neocene:
1. Rhyolite.
2. Pyroxene andesitc.
3. Quarts basalt.
4. Olivine diabase.
5. Augite ioschenite.
Sierra Nevada of Caiifomia (Folios of the U. S. Geol. Survoy, and H. W. Tume
Jour. Geol., Vol. 3, 1895, p. 385) :
A. Devonian or older:
1. Pyroxene andesite.
2. Rhyolite.
B. Carboniferotu: Diabase, diabase porph3rrit«t Migite aadcsCc, hon
blende andesite, peridotitc, pyroxenite, rhyolite.
C. Jura-Trias: In general, same types as in Carix>iiiferoiis.
Redding quadrangle gives:
Triassic:
1. Ophitic basic andesite, augite andesite.
2. Rhyolite.
Jurassic: Pyroxene andesite.
D. Jurassic or Early Cretaceous:
1. Granodiorite, quarts diorite, gabbro, qrenltey dacite poi
phyry, homblendite.
2. Granite, aplitc.
E. Neocene:
1. Rhyolite.
2. Basalt.
3. Hornblende-augite andesite.
4. Pyroxene andesite, latite.
5. Basalt.
F. Quaiemary: Pyroxene and(«ite, basalt.
Berkeley Hills, California {Tertiary sequnce). (A. C. Lawson and C. Palacbc
Bull. Department of Geol., Vniv. of Cal., Vol. 2, 1902, p. 438):
A. Lower Berkeleyan:
1. Andmto.
2. Basalt.
3. Rhyolite.
4. And<*site.
5. Bi&salt.
6. Rhyolite.
B. Upper Berkeleyan:
1. Andcflite.
2 Basalt.
3. Rhyolite.
APPENDIX B 479
4. Andesite.
5. Basalt.
C. Campan:
1. Andesite.
2. Rhyolite.
3. Basalt.
GMfidd District, Nevada (F. L. Ransome, Prof. Paper 66, U. S. Qeol. Survey, 1909,
pp. 90-91 and PI. IV) :
A. Mesozoic: Granite, quartz monzonite, syenite.
B. Eocene (?):
1. Rhyolite.
2. Latite.
3. Rhyolite.
4. Olivine basalt.
5. Biotite andesite, hornblende andesite, pyroxene andesite,
dacite.
6. Dacite, andesite, vitrophjrre.
7. Rhyolite.
8. Andesite.
9. Olivine basalt (quarts-bearing)-
10. Rhyolite.
11. Olivine basalt.
Eureka District, Nevada]{Mon. 20, U. S. GeoL Survey, 1892, p. 290):
Late Tertiary and Quaternary:
1. Hornblende andesite.
2. Hornblende-mica andesite.
3. Dacite.
4. Rhyolite.
5. Pyroxene andesite.
6. Basalt.
aiftan Qwidrangle, Arizona ( W. landgren, Clifton folio, U. 8. QeoL Sorviyi 1905) :
A. LaU Cretaceous or Early Tertiary:
1. Qranite porphyry, quarts nMnzonite porphyty/ diorite
porph3rry.
2. Diabase.
B. Tertiary:
1. Rhyolite.
2. Basalt.
3. Pyroxene andesite.
4. Basalt.
5. Rhyolite.
6. Basalt.
7. Rhyolite.
Idand oj Skye (A. Harker, and C. T. Cloug^, The Tertiiiy Igneous Rooks ol fflgre,
Glasgow, 1904, p. 433) : ^. . . .
A. Pre^Tertiary: Gabbro, granite.
480 IGNEOUS ROCKS AND THEIR ORIGIN
B. VcHeanic Phase (Eocene) :
1. Basalt, olivine basalt, hypcnthcnc basalt, augite andcrite.
2. Trachyte, rhyolito, andesite, felsite.
3. Basalt.
C. Plutonic Phase (in general younger than B.) :
1. Peridotite, "picritc," troctolite.
2. Gabhro.
3. Granite, granophyrc, syenite, marscoite.
D. Phckse of Minor Intrusions (in general younger ihaii C):
1. Basalt, granophyre, felsite, porphyry, trachjrte.
2. Dolerite, olivine dolerite.
3. Peridotite, "picrite."
4. Trachyte, augite andcsite.
Glen Coe, Scotland (C. T. Clough, H. B. Maufe and E. B. Bailey, Quart. Jour.
Geol. Soc., Vol. 65, 1009, p. 615):
Old Red Sandstone Period:
1. Augite andesitcfl.
2. Rhyolites and andesites.
3. Hornblende andesites.
4. Rhyoliie.
5. Andesites and rhyolites.
Mont Dore, France (A. Michol I^vy, Bull. soc. gdol. France, Vol. 18, 1800, p. 743):
(Sequence from mid-Pliocime to Recent) :
1. Phonolite, phonolitic trachyte.
2. Rhyolite.
3. Basalt.
4. Andcsite, basalt.
5. Acid tuffs.
6. Acid andesite, trachyte.
7. Augite andesite, tephrite.
8. Phonolite.
9. Basalt of plateaus.
10. Basalt.
Le Velay and le M^zenc, France (P. Termier, Bull, des serv. carle gfoL France,
No. 13, Vol. 2, 1800 and M. Boule, Ibid. No. 28, Vol. 4, 1802):
(Miocene sequence) :
1. Basalt.
2. Trachyte, phonolite.
3. Augite andesite, trachydolerite.
4. Basalt.
5. Phonolite.
6. Basalt.
Cantal, France (M. Boule, Bull, des serv. carte gM. France, No. 70^ VoL 11, 1900).
A. Miocene:
1. Basalt, olivine basalt.
2. Trachyte, phonolite.
3. Augite andesite.
APPENDIX B 481
B. Plioetne:
1. Andesite, trachyte.
2. Augite andeette ("labradorite"), olivine basalt.
3. Hornblende andeaite, augite andesite, phonolite.
4. Basalt of the plateaus, olivine basalt.
La Linagne, France (P. Gtangcaud, Bull, des serv. carte gfal France, No. 123,
Vol. 19, 1909)1
A. Loioer Miocene: Basalt, teechenite, t«pbrite, ncphelinite, phoDolit«.
B. Middle Miocene: Olivine basalt, limburgite.
C. Upper Miocene: Olivine basalt.
D. Lower Pliocene: Basalt.
E. MiddU Pliocene: Basalt.
F. Upper Pliocene: Basalt, tepbrite. i ,
G. PUislocene: Basalt.
WeMennald, Germany (G. Angelbis, Jahrbuch preius, geol. Londraanst., Vol. 3,
18S3, p. kIv,):
1. Basalt.
2. Trachyte.
3. HornbleDdc andesite, augite andesite.
4. Basalt.
SiAena^rge, Germany (H. von Decben, GcognostischeT Fllhrer in das Siebenge-
birgc am Rhdn, Bonn, 1861):
1. Trachyte.
2. Nephelite (?) baaalt, trachyte.
3. Olivine basalt.
Bohemian Uittelgebirge (J. E. Hibech, Tschermok'e Min. u. Petrog, Mitt., Vol. IS.
1900, p. 493. and FOhrer fflr die geol. Exkursioncn, Internal. Geul.
Cong. 1903, Part 2; F. E. Sueas, Bau und Bttd Oesterreicfaa,
190):
A. Cambrian: Diabase, diorite.
B. CarboniferoM-Pemian: Melaphyre, Kr&nite, granite porphyry,
quartz porphyry, aplite, lampropfayre.
C. Upper Otigocene:
1. Basalt, phonolttc.
2. Feldspar basalt, nephelite basalt, leucite basalt, liDiburfsitc.
3. Trachydolerite, hatlynite tephrite, sodalite eyenite, Bodnlitc
gautcite, sodalite porphyry.
4. Essexite, camptonite, gautcite, bostonite, nephelite tepliritv,
Icucite tephrite, nephelite basanile.
D. Miocene {?).■
1. Basalt.
2. Trachyte.
3. Phonolite.
4. Tinguaite, eieolite porphyry.
Carpathiant (V. Uhlig, Bau und Bild Ocslerrcirha 1903, p. 894) :
A. Upper Eocene: Traehyt*
B. Fim Mediterranean Stage: Rhyoiite,
482 IGNEOUS ROCKB AND TEEIR ORIGIN
C. After firtt Medilerranean Stage: Basic fnyrojtne
D. Somewhat later. Biotite-amphibole andflttte.
R. SamuUie epoch: Rhyolite.
Predazzo (W. Pcnck, Xcuea Jahrb. fttr Mineralogie, etc., B. B. 32, 191 1» p. 341):
1. Porphyrite (augitc andesite, often bearing olivine).
2. Melaphyrc.
3. Monzonitc.
4. Pyroxenite.
5. Quartz monzonite.
6. Monzonitc aplite.
7. Syenite.
8. Quartz syenite.
9. Syenite aplite.
10. Bostonite.
11. Nephelite syenite.
12. Tinguaite porphyry.
13. Granite. (Probably follows the nephelite syenite.)
14. Aplite.
15. Camptonite.
Momoni, Tyrol (O. von Huber, Jahrb. k. k. geol. Reichsanst.i VoL 50^ 1901, p.
395):
1. Pyroxenite.
2. Monzonite.
3. Melaphyrc, augititc.
4. Plagioclase porphyrite.
5. Granite.
6. Camptonite.
7. Liebnerite porphyry, orthoclase porphyry.
Eolian Island$ (A. Bergeat. Abhand. k. bayer. Akad. der Wiss., K1. 3, VoL 90,
1899, p. 1):
Islands in general:
A. Middle or laU Tertiary to Middle Quaiemanf:
1. Basalt.
2. Andesite.
B. Late Quaternary to Present: Acid andetitey Upftritc^ dadte,
basanite.
Lipari:
1. Basalt.
2. Andcsite.
3. Cordicrite andositc.
4. Lipari tc.
Vulcano:
1. Basaltic andcsite.
2. Lipari tc.
3. YounRcr liparitc.
4. Leucitc basanite (Vulrancllo).
Filicwli and Alicudi:
1. Basalt.
2. Augite andesite. - "•
APPENDIX B 483
ChriaHania Reffian, Norway (W. C. Brfigger^ Zdt. ftkr EryBtallographiei Vol. 16,
1800):
1. Diabase, diabase porphyritei augite porphyrite.
2. Laurvikite, mica syenite, laurdalite, rhomb-porphyryi ditroite, foyaite,
tinguaite, minette.
3. Akerite.
4. Nordmarkite.
5. Soda-granite, hornblende granite, arfvedsonite granite, CBgerite granite.
6. Biotite granite.
7. Diabase, diabase porphyrite.
EkerBundrSoggendal Districtf Norway (K. F. Kolderup, Bergens Museums Aarbog,
Vol 6, 1896, p. 183) :
1. Norite, anorthosite.
2. Norite, gabbro-norite, quarts norite.
3. Monzonite, banatite.
4. Ilmenite norite, ilm^tite (perhaps older than some of the banatiteB).
5. Augite granite, aplite.
6. Diabase, olivine diabase.
JuUanehaah District, OreerUand (N. V. XJssing, Geology of the Country around
Julianehaab, Greenland, Copenhagen, 1911, p. 318):
1. Diabase.
2. Essexite.
3. Alkaline syenite. \ , . .
>f All 1* -^ > almost contemporaneous.
4. Alkahne gramte. j '^
Martinique (A. Lacroix, La Montague Pel6e et ses eruptions, Paris, 1904, p. 22):
1. Basalt, augite andesite ("labradorite")-
2. Augite andesite C'labradorite''), hypersthene andedte, dadte.
Krakatoa (R. D. M. Verbeek, Krakatau, Batavia, 1886):
1. Hypersthene andesite.
2. Basalt.
3. Hypersthene andesite.
Java and Madura (R. D. M. Verbeek and R. Fennema, Description gMogique de
Java et Madoiura, Amsterdam, 1896, p. 38 ff.):
A. Cretaceous: Diabase, gabbro, quartaose potphyrj.
B. Eocene: Diabase, andesite (with dioritio phases).
C. Base of Miocene series: Basalt, diabase, gabbro^ pyroixene andesite.
D. Lower Miocene: Gabbro, pyroxene andesite, hornblende andesite,
quarts-mica-homblende andesite (dacite).
£. Middle Neo-Tertiary: Pyroxene andesite.
F. Post^Tertiary: Basalt, pyroxene andesite, leueitio rodn and phooo-
lites.
New South Wales (C. A. SOssmilch, Geology <tf New South Wales, Qydn^, 1911, p.
165):
Ordomeian: Andesites. *^ '
Silurian: Rhyolites, andesites.
Devonian: Basalt, rhyolite, granite, tonaUte, qiuurts-mioa diorite^ grano-
diorite, quarts porphyry, serpentiiie (peiiddlite).
484 IGNEOUS ROCKS AND THEIR ORIGIN
Carboniferoua: Granite, rhyolite, hypentlieDe anderite, gruute porpliyiry.
feldspar porphyry.
PemuhCarboniferoua: Basalt, andesite, trachsrte.
Eocene (?): Basalt, olivine basalt.
Miocene or Lower Pliocene: Basalt.
Pliocene:
1. Comenditcs and quarts trachytes.
2. Alkaline trachytes.
3. Phonolitic trachytes.
4. Andesites.
CanoMae Mis., NewSoiUk Wake (C. A. SOssmilch and H. L Jenaeiiy Proe. Linana
Soc. New South Wales, Vol. 34, 1909, p. 170.):
1. Comenditee, panteUeritcs, quarts trachytes.
2. Trachytes, phonolitic trachytes.
3. Basic andesites, augite andesite, alkaline basalt.
4. Melilite basalt.
New England Plateau^ New South Wales (E. C. Andrews, Records GcoL Surrey,
Xew South Wales, Vol. 8, 1905, p. 20):
A. Carboniferous: Dolerite, trachyte, andesite, riiyolite.
B. Permo-Carboniferous:
1. "Lavas."
2. Granite porphyry.
3. Granite.
4. "Eurite" (aplite).
5. Rhyolite, acid porphyry.
6. Diorite, hornblende and mica lamporphyres.
C. Tertiary: Basalt, pitchstone.
Victoria Slate, Australia (E. W. Skeats, Pres. Address, Austr. Assoe. Adv. Scienee,
Vol. 12, 1909, p. 173):
A. Ordovician (7): Diabase, diabase porphyrile, diorite, gruiopliyrev
microgranite.
B. Silurian: Andesitic tufT.
C. Lower Devonian (?): Quarts porphyry, syenite porphyry, >alfsb«^
Rite, bostonite, alkaline trachyte, quarts kermtophyre, grmaodiorite,
dacite, quarts porphyrite, icranite porphyry.
D. Middle Devonian: Diabase, diabase porpbjrrite, rlqrolite, biotite
andesite, augite andesite, hornblende andesite.
E. Upper Devonian, or Lower Carboniferous: Basalt, rhyolite, quarts
porph>Ty.
F. Lower Cenozoic: Ba8Hlt, augite andesite.
G. Middle Cenozoic:
1. Solvsbcrgite.
2. Alkaline trachyte.
3. Basalt ^ith anorthoclase.
4. Olivine trachyte.
5. Olivine-anorthoclase tradiyto.
APPENDIX B 485
6. Limburgite.
7. Basalt.
H. Upper Cenozoic to Recent: Basalt, olivine basalt, limburgjte, haOyn-
ite-bearing dolerite.
East Moreton and Wide Bay Dietridi, Queenttand (H. I. Jensen, Proc. Linn. 8oc.
New South Wales, 1906, Part 1, p. 166) :
A. Pre^arboniferoua: Diabase, greenstone.
B. Carbaniferoua'Permian: Tonalite, quarts diorite, granite.
C. Cretaceotu (?): Porph3nite, tonalite, monzonite, soda-andesite,
quartz andesite.
D. Eocene (?): Trachyte, rhyolite, keratophjnre, comendite.
E. Late Eocene (?): Dacite, andesite.
F. Pliocene (?): Basalt.
APPENDIX C
TABLE XXII.— LISTS OF DISTRICTS CHARACTERISED BT ICBHBBRS OP
SYENITE CLAN. WITH NOTES ON THE NATURE OP COUNTRT-VOCKI
(See page 395; many references to authors found in RoMobaseh's haadbook-)
FIELD ASSOCIATIONS OF THE SYENITE CLAN
Region
r
Representatives of syenite clan
and other alkaline clans _
NORTH AMERICA
Sedimentscot bj cmplivw*
Quebec.
Mt. ShefTonl.
Pulaski te, nonlmaricitc, essex-
, ito, thcralitG, camptonitc,
j boetonito, trachyte.
Pakosoic ih., eg.» li.| m.
Mt. Johnson.
Puloskite, cffsexite, campton-
ite, solvsbcrgite (?).
Mt. Bromc.
Nordmurkitc, laurvikitic syen-
ite, essexite.
Ditto.
Paleoaoic ih. and it.
Mt. Orford.
Nordmarkite, monsonite,
camptonite.
Grenville.
Hornblende 8>*enite, quarti-
syenilc pori>hyr>*.
Mt. Rigaud.
Hornblende syenite; alkaline
quartz porphyry.
Ottawa (^ounty.
Syenite orthogneiss.
Keekwk and Ko-
wagama lakes.
Syenite.
Metagami Lake.
Nova Scx)tia.
Syenite porphyry.
Ditto.
Pre-Cambrian IL and
/Vrisaig-Antigonish Monzonite
dintrirt. i
Pre-CamlMriaii li., and
ite.
Pre-Cambrian bear
Keewatin greenaloBes
Huroniaa iL, eg., arkoK.
Cambrian el., grits* aad gr*y
I wackes.
^Abbn'viaiion$: Arg. — argillitc; eg. — conglomerate; li.*
si. — slate; es. — sandstone.
486
APPENDIX C
487
TABLE XXII.—FIELD ASSOCIATIONS OF THE SYENITE CLAN.-Co«ltmi«ii
Region
Representatives of syenite clan
and other alkaline clans
Sediments cut hy eruptives
Ontario.
•
Larder Lake Dist.
Syenite.
Eeewatin green sehiste, si.,
and dolomite.
Abitibi Lake.
Homblende-albite syenite,
quartz-albite syenite.
Eeewatin green schists; bL,
dolomite.
Monmouth Town-
ship.
Alkaline syenite, alk. granite,
nephelite syenite, monmouth-
ite.
Grenville li.; schists.
Glamorgan Tp.
Albite syenite, nephelite syen-
ite.
Ditto.
Harcourt Tp.
Corundum syenite, nephelite
syenite.
Ditto
Methuen Tp.
Ditto.
Amphibolite, li., gneiss.
Faraday Tp.
Alkaline syenite, nephelite
syenite.
?
Monteagle Tp.
Ditto.
Li.
Raglan Tp.
Alkahne syenite, craigmontite,
nephelite syenite, corundum
syenite.
Ditto.
Nipissing-Timifl-
kaming Dist.
Syenite orthogneiss.
Huronian si. and grajsrwacke.
(Grenville li.?).
Pigeon Lake (Mon-
treal river).
Hornblende syenite.
?
Lutterworth Tp.
•
Corundum syenite.
Grenville li., etc.
Port Coldwell.
Hornblende syenite, nephelite
syenite, quartz syenite, es-
sezite, camptonite, alkaline
granite.
KeewaUn chlorite sothisl and
greenstone; Lauresitiaa
schists.
Rainy Lake.
•
Hornblende and mica syenite.
Keewatin green BohMl; eg.,
sl.| etc.
Sturgeon Lake.
Hornblende syenite.
Ditto.
488
IGNEOUS ROCKS AND THEIR ORIGIN
TABLE XXII.—FIELD ASSOCIATIONS OF THE SYENITE CLAN.
Region
Rtiprcscntativefl of syenite clan a J* ^ a i.
I J Ai. II 1- I Sediments cut mr cniptj
and other alkaline clans "^
Gunflint Dist.
Alberta.
Hornblende syenite.
Keewatin green sehist; ck..
si., etc.
i
Blairmore.
Analcite trachyte.
Paleosok li., sh., eie.
British Coluiibia
I
Camp Hedley. i Monsonite, keratophyre.
Paleoioie arg., li., eCe.
Roche River.
Syenite.
Ditto.
Tulameen Dist. Augitc syenite.
Edwards Creek. Porphyritic syenite.
Triaasie arg., li., and
volcanics; proiiably-]
soic li. and arg.
Paleosoie li., sh.,
freenstone.
Bonaparte Lake. : Auicite syenite.
Franklin Camp.
Phoenix Dist.
Nelson Dist.
' Monxonite, syenite, pulaskite . Paleoioie li., arg., and
; porphyry. stone.
Syenite, syenite porphyry, tra- Ditto.
chyte.
RoAsland Dist.
Monzonite.
Ditto.
Monzonite, latites, missourite. Paleosoie li., arg., eie.
Salmon River.
Monzonite.
Ditto.
Christina Lake. Pulaskite (Coryell batholith). Ditto.
Kettle River (Mid- 1 Pulaskite porphyry, alkaline Ditto,
way). ! trachyte, rhomb-porphyry,!
sharkanite. <
Skagit Range. ! Monzonite.
Ditto.
Si wash Creek Dis- Syenite porphyry,
trict.
Pakonie (7) •!., seliMi, IL
I
Shuswap Lake Hornblende syenite.
(Salmon Arm).
Pre-Cambrian U^ phyUitca.
etc.
TABLE XXII— i
APPENDIX C
-FIELD AS80CIATI0NS OF THE SYENITE CLAN.-
Region
Represcntfttivea ot syenite 1 q .. , , .
AtASKA.
1
Naknck Lake.
Syenite.
MesoBoic sh., li., and chert.
Copper Mt. und
Moira Sound.
Syenite, trachyte.
Paleozoic (7) sh., li., and
Kluane River.
Syenite.
Paleozoic Bl.,ii.
Tre.-idweU Mine.
Sodium syenite.
Thick si.
Chii-hagof Cove.
Alkali-ayenite porphyry, latit«.
GenoKoic sh., es., li., grit.
Kichatna Valley.
Olivine monionite.
Jurassic et.
Rampart ReKion.
Monzonite,
Paleosoic arg., li., as.
Lynx Mt.
MoDionite.
Paleoaoie arg., 11., es.
Glenn Creek.
Monsonite,
PaleoBoicarg,, li., as.
Copper River.
Monzonite.
Probably eula arg. and li,,
etc., of Valdeii and Paleosoio
scries.
Yentna River. | Hornblende syenite.
?
Swentna River. 1 Quart* syenite.
Thick sh.
Kasnan Peninsula. \ SyeniU. Paleozoic ss., eg., li., and grecn-
1 { stone.
Matanuaka Valley.' Trachyte.
Cenoioic and Mceoioic sh.,
SB., eg., U., and basic igneous
rocks.
Yv.oN.
ffheaton River Syenite, syenite porphyry,
Diat. trachyte.
i
Paleozoic;?) li, green schists
and quartiitra; also prab-
ably-Mesozoic arg., as., eg.
Lake Ubarge.
Syenite porphyry.
Carboniferous li.; alsoss.^sh,,
tuffs.
HTiife Horse Cop- Syenite porphyry, boetonite.
ppT Belt.
Paleosoic li.
490
IGNEOUS ROCKS AND TEEIR OBIO!N
TABLE ZXI1.--FIELD ASSOCIATIONS OF THE SYENITE CLAN.
^^&on I , ^ ^^ , ., ^^ .iu.i:«^ ^u«. ScdimenU cut l^ cffvpUvet
clan and other alkaline dans
WASHINaTON.
Myen Creek Dist. Hornblende syenite.
Paleoioie •!., IL,
and quartsHe.
Idaho.
Vermilion Creek. Syenite.
i
Pre-Cambrian db^ ■.
sibly
Upper St. Joe Ri-
ver.
Monzonite, camptonite.
Pre-Cambrian rii.» U^ a.
Coeur D'Alene j Monsonite, monsonite por- : Pr»-Cainbriaii dL« IL,
Dist. ! phyry, syenite. '
Montana.
I
Rlkhom Dist.
Helena Dist.
i Syenite, shonkinite, boetonite. Cambrian and oUar
i and quartsite.
.fc.
Three Forks Quad-
ranfcle.
Bozeman.
Latite, monzonite.
Paleoioie and
li., quartsite.
•nt
Syenite.
Pre-Cambrian IL, ■., cr.
Corundum syenite.
Sweet Grass Hills. Quartz syenite porphyry, min- Cretooeoua sIl, ale.
ette. '
Castle Mts.
' Syenite porphyry, acmite tra- i Paleoioie and
I chyte, theralite, monchiquite. ■
oUflf M.y S.} M.
Little Belt Mts. Syenite, monzonite, shonkinite, Paleoioio and
trachyte, syenite porphyry,
etc.
oUflf M.y fi.t ii*
Crazy Mts.
Acmite trachyte, theralite.
Paleoioie and oldflr rit.. Mn »
High wood Mts. Syenite, monzonite, shonki* Paleoioie and
nitc, etc.
APPENDIX C
491
TABLE XXII.— FIELD ASSOCIATIONS OF THE SYENITE CLAN.— Conttnued
Region
Representatives of syenite
clan and other alkaline clans
Sediments cut by eruptivea
Judith Mts. Syenite, tinguaite, etc.
Paleozoic and older sh., li., ss.
Bearpaw Mts.
Mica trachyte, monzonite,
shonkinite, augite syenite,
nephelite basalt, leucite ting-
uaite, leucite basalt.
Paleozoic and Mesozoic li.,
sh., 88., etc*
Bannock, Beaver-
head Co.
Syenite.
"Paleozoic sediments."
Wyoming.
Crandall Quad-
rangle.
Monzonite, syenite, shoshonite.
?
Sundance Quad- Monzonite porphyry, syenite
ranglc. porphyry, nephelite syenite,
bostonite, alk. lamprophyres.
Paleozoic li., sh., etc.
Absaroka Quad-
rangle.
Monzonite, syenite, quartz
syenite.
Paleozoic li., sh., and quart-
zite.
Laramie Mts.
Syenite.
Pre-Cambrian mica schist,
•
hornblende schist, and per-
haps li.
Aladdin Quad-
rangle.
Syenite porphyry, phonolite,
pseudo-leucite porphyry, etc.
Paleozoic li., ss., etc.; Plre-
Cambrian schists.
Colorado.
Rico Dist.
Monzonite, monzonite por-
phyry.
Paleozoic and Mesozoic ah.,
li., ss.
1a Plata Quad.
Monzonite, monzonite porph.,
augite syenite, syenite porph.
Paleozoic and MesoBoic ah.,
li., ss.
Tcllnride Quad.
Monzonite.
Paleozoic and Mesoioic ah.,
li., 88.
Spanish Peaks.
Monzonite porphyry.
Paleozoic and Mesoioic ah.,
li., 88.
North Peak.
Trachyte.
Paleoioic and Meaosoio ah.,
li., 88.
Cripple Creek.
Olivine syenite, syenite, phon-
olite, trachydolerite, etc.
7 (Pre-Cambrian granitet)
33
492
IGNEOUS ROCKS AND THEIR ORIGIN
TABLK XXII.— FIELD ASSOCIATIONS OF THE SYENITE CLAN.
Region
R€pr€flentativet of syenite
! clan and other alkaline clsns
SedimeDts cut 1^ cnipihrcs
Georgetown Quad.
Syenite porphyry, latite, bos-
tonite, etc.
Pre-CambriBD IL
Evcri^rccn Mine,
Gilpin Co.
Monzonite.
MetssedimenlAfsr biolite
schists, wHh MisSROUi
phases.
Silver ClifT and
UosiU Hills.
Syenite, trachyte.
Monzonite, monsoniie porph.,
monchiquite, camptonite.
Syenitic lamprohyres.
Trachyte.
Augite syenite.
Trachyte, nephelite basalt.
Monzonite porphyry.
!
(Pre-CambriftB pBaitc>
gneMsO.
Engineer Mt.
Quad.
Psleosok sh., li., ss.
Two Buttes.
TrissHc sedimeDU (sod f).
Leadville.
Li., grsoite.
Denver Basin.
(Pre^Csmbrisii pieisBO*
Elkhead MU.
Cretseeowf
Durango Quad.
•
Pennisn (7) MkaiWMS aivl-
Ute,ss.,e8.
Sangre de Criflto
Ilange.
'Syenite orthognciss.
T
Breckenridge Dist.
Monzonite porphyry, quarts-
monzonite porphyry.
Plre-Osinbffisii miflft sckisls^
Mesosoie sh., IL, cc ss.
Silvcrton Quad.
Ditto, with latite. . !
Pre-Csmbrisii w\Mn mkI
si.; Pdeosoie sh.p i^ Si-.
quartsite.
Ouray Quad.
I.Atite, quart z-monzonite por-
phyry.
Paleosoic sh., IL, ss.. etc.
Twin Butte.
Syenite porphyry.
1
Mesosoie sh.^ as.
Gray back Dist.,
Costilla Co.
Monzonite.
1
Paleosoic sedimsBU.
California Dist.,
La Plata Co.
Monzonite porphyry.
"Red Beds.**
En^t Mancos Dist.
.Montezuma Co.
Monzonite.
"Red Beds."
APPENDIX C
493
TABLE XXII.— FIELD ASSOCIATIONS OF THE SYENITE CLAN.— Con«»nt««d
Region
Representatives of syenite
clan and other alkaline clans
Sediments cut by eniptives
Tarryall Dist., ' Monzonitc.
Park Co. !
Monarch and Tom- Monzonite, quartz monzonite,
ichi Dist., Chaf- i latite.
fee Co. '
Paleozoic and Mesozoic sed-
iments.
Paleozoic li., sh., quartzite.
Utah.
Tintic Dist.
Monzonite, monzonite porph.,
latite.
Paleozoic li., chert, quartzite.
Bingham Dist.
High Plateaus.
Monzonite, monzonite porph.
Paleozoic li., chert, quartzite.
Many trachyte masses.
Paleozoic li., sh., quartzite,
etc.
Cactus Mine (San ' Monzonite.
Francisco Mts.) !
Li. and quartzite, probably
Paleozoic.
Downtown Dist. ' Monzonite porphyry., quartz
monzonite porphyry
Paleozoic li. and quartzite.
•
Big Cottonwood
Canyon.
Syenite porphyry.
Paleozoic li., sh., etc.
Twin Peak.
Trachyte.
Cambrian sL
Western Uinta
Range.
Trachyte.
Paleozoic li., quartnie; also
probably Meaoioie li., 88.
City Creek and
Easy Canyon Cr.
Trachyte.
Paleozoic and Meaoioie li.,
sh., quartzite.
Fountain Head
Trachyte.
Paleozoic li., and quartzite.
niijs ^utanr;.
Oquirrh Range.
Trachyte.
Carboniferous li.
Stansbury Range.
Trachyte.
"Paleoidc strata."
Cedar Range.
Trachyte.
?
Thomas Range.
Trachyte.
Paleoioicli.
Picacho Range.
Trachyte.
?
494
IGNEOUS ROCKS AND THEIR ORIGIN
TABLK XXII.— FIELD ABflOCIATIONS OF THE SYENITE CLKH.—CmmtimmM
Marysvale, TuHh- Monxonitc, quarts monionite, Mcaoioic and Palcome li.
ar Range. quarts latite. And quartsite.
Beaver Lake Dist. Monxonitc.
'* PaleoBoic ■cdimoita.'*
North Star Dist. Monzonitc.
<4
Paleosoie •edimeDta.'*
Preuss Diflt.
Monsonite.
14
Paleoioic aediiiieDta.
fi
Rocky Dist
Monsonite.
<*
PaleoBoie wdimenta.
»i
Frisco Dist.
Monsonite.
14
Paleofoic Mdinienta.**
Miner's Basin, Monsonite poqihyry.
Grand Co.
Ophir Dist., Too- Monsonite.
ele Dist.
Paleosoic iedimenta.
Nevada.
Cactus Range. I^tite.
PaleoioiG li., eg., and quart-
site.
Stonewall Mt.
Monsonite porphyr>', quarts Thick Cambrian IL, ate.
svenite.
CiuMfield Hills. Latite, quarts latite.
Southern Klon- (j^iartz latite.
(like Hills.
Silver Peak Range. latite, monsonite.
Kawirh Range. Monsonite poq)hyry.
Fielted Range.
Monionitf*.
Pahute Range.
Monsonite.
Yucca Mt.
Latite.
Aniargosa Rangt>. L:it it e
Probably cuta Gaabriaa b
and ah.
Cambrian li. and rii.
Cambrian li. and ih.
Paleofoic li., rii.« ai.
Probably cula Caaibriaa li
and ih.
Probably cula Ganbrian b.
andih.
Paleoioie li. and quartaitr.
Paleoaoic li. and quartaile.
APPENDIX C
495
TABLE XXII— FIELD ASSOCIATIONS OF THE SYENITE CLAff.—Conlinued
Region
Representatives of syenite L ,. . ^ ,
1 J Ai_ n 1- 1 Sediments cut by eruptives.
clan and other alkaline clans "^ ^
Panamint Range.
Soda syenite, quartz monzonite
Paleozoic li. and quartsite.
Cortex Range.
1
Syenite, syenite porph., trach-
yte.
Paleozoic li. and quartzite.
Black Butte and
Eugene Mts.
Syenite.
Paleozoic sL, li., ss.
Montezuma Range.
Syenite.
Jurassic si., li., ss.
Winnemuca Peak.
Syenite.
Jurassic (7) si., li., ss.
Pyramid Tiake.
Trachyte.
?
Virginia Range.
Trachyte.
7
Seetoya Mts.
Trachyte.
Paleozoic li. and quartzite.
Pinon Range.
Trachyte.
Paleozoic li. and quartzite.
East Humboldt
Range.
Trachyte.
Paleozoic li. and quiirtzite.
Wah-weah Range.
Hauynite-bearing trachyte.
Paleozoic sediments.
River Range.
Trachyte.
Paleozoic li. and quartzite.
Shoshone Range.
Trachyte.
Paleozoic li. and quartsite.
Kawsoh Range.
Trachyte.
7
Pahroc Range.
Trachyte.
7
Robinson Mining
Camp.
Monzonite, monzom'te por-
phyry, uiinette.
Paleozoic li. and ah.
Contact Mining.
Syenite.
Paleozoic li. and ah.
Bare Mt. Dist.
Monzonite porphyry.
Paleozoic sediments.
Ward Dist.
Monzonite porphyry.
Paleozoic sediments.
White Pine Dist.
Monzonite.
Paleozoic sediments.
496
IGNEOUS ROCKS AND THEIR ORIGIN
TABLE XXIL— FIELD ASSOCIATIONS OF THE SYENITE CLAN.
Region
Representativet of syenite < o .. . . . ..
1 J Ai. II. 1* I SedimenU cut by enipta
clan and other alkaline clans : ' "^
C'aufornia.
Big Trees Quad Syenitef latite.
Paleoioic arg., li., i|uartaite.
Nevada City Quad. Augite syenite, monsonite.
I
Sonora Quad.
Soda 8>'enite.
Paleosoie arg., li., quartsite.
Mesoioio and PaleoaoM arg..
li., and at.
Spanish Peaks Soda syenite, plumasite.
(Bid well Bar
Quad.)
Downieville Quad. Syenite porphyry.
Clay d..
Inyo County.
Hornblende syenite.
Silurian li.
Darwin Dist. (Inyo. Monsonite.
Co.)
I Paleoioic sediments.
Goldbclt Dist. Monsonite.
(Inyo Co.)
Paleoioic sediroents-
Skidoo Dist.
Monsonite, syenite.
Virginia Dale Dist. Syenite.
I
14
Sediments''
Mother Lode Dist. Latite.
Paleoioic li., d.
Eagle Mts.
Monsonite, syenite, granodio- Thick ddomite and qusrtsite.
rite, quarts monsonite. ,
Arizona.
I
Bradshaw Mts. Monsonite porphyry, trachy- PhyUite, miea sduslSt horn-
dolerite.
Bisbee Quad.
Monsonite porfhyry.
bleiide schist, li., eg., qnart-
nte.
Paleoioic li., qusrtsile, etc.
01<)l>e Quad.
Monsonite (quarts monsonite).; Paleoioic sediments.
Sierra FI.srudillo. Trachyte.
Siorra Caluiro.
Trachvtc.
.\8!)o«tos Canyon. Hornblende syenite.
Unkar ti., ih.; Vishna seUsU.
APPENDIX C
TABLE XXII.~FIELD ASSOCIATIONS OP THE flTENITB CLAN.-
o Repreaentatives of syenite i „ ,. . , .
«^«""' dans and other alkaline ckns ! S^i-menU cut by empt.ves
Dripping Springa i Monsonite porphyry.
Di.,.
Paleojoic Bedimento,
Rivrraide Dist.
Monzonite porphyry.
Silver King Dist
Syenite.
Paleoioio sedimenta; pre-
Owtle DomelHat,' Mo nzonite porphyry.
Pre-Cambrian schirt.
. (Draeoon UU.)
Paleoioic ah., li., quartsite.
New Mexico.
WalaenburgQuad.
MonKmite porphyry.
Paleoioic li. and other aedi-
naents.
C*mTlo«HUIs.
McsoEoio arg. and as. (also
Paleozoic eediments?)
Deming.
Paleozoic U., eh., as.
Sierra Luera.
Trachyte.
t
Cook'B Peak.
Syenite.
CuboniferouB U.
Burro Mt«.
Latite, trachyte, quarti mon-
sonite.
Probably out Cretaceous ah.
andli.
PtnoaAltofl.
Monxonite, syenite, trachyte,
7
CimarTonrntoDisl.
Monaonite.
Carbonileroua and Creta-
Moreno Dist.
MoMonite porphyry.
Paleocoio and Cretoceoua
sedimentB.
Ute Creek Dirt.
Paleoioic and Cretaceous
eediroento.
Black Mt. Diet.
Monionite.
Palooaoic »edimentB.
Nogal Dirt.
MoDKDite porphyry.
t
498
IGNEOUS ROCKS AND THEIR ORIGIN
TABLE XXII— FIELD ASSOCIATIONS OF THE SYENITE CLAN.-
^ Reprcsontativc Of syenite cUni g^i„^„^ ^ ^ «ttp|iv«
I Ann /\tlior nllrnlino f*l«na
and Other alkaline rlanii
Whiteoaks Dist. Monzonitc.
Tres Hermanas Quartz syenite.
Dist.
Jarilla Dist.
Monzonite porphyry.
Cretaceous sedimcnU.
Paleozoic aedimwiU.
Paleosoie ■edimeDU.
Cochiti Dist.
Monzonite.
HiUahoro Dizt.
Monzonite.
Ticrra Blanca Dist. Monzonite*.
Pakoioie •edimenU.
Paleozoic ■edimwito.
Jones Camp Dist. Monzonite.
Paleozoic li.
Magdalena Dist. Monzonite.
Red River Dist. Monzonite porphyry.
Paleozoic •edimcnta.
Pre-Cambrian acfaist and
nice.
Texas.
Brewster County. Trachyte, phonolite.
Franklin Mts.
Apache Mt^.
Monzonite porphyr>'
Thick Cretaceoiu li. and 9h
Pre-Cambrian al., very Ibick
Paleozoic li.
Syenites, nephelite syenite, Paleozoic and M
phonolite, l>ostonito, tinguaite,
paisanite.
ic li.
Kl Paso Quad.
Arkansas.
Magnet Cove.
Syenite porphyry.
Paleozoic li., aa.
Shonkinitf-, nephelite syenite. Paleozoic li.
etc. I
VlROIXIA.
Luray Dist.
I
Syenite.
"Blue Uidge Hog- Hypersthone akerite.
ion
»•
ittji
Southwest Vir- Hornblemle syenite.
ft
ginia."
Basic schists And diorite.
APPENDIX C
TABLE XXII.-
FIELD ASSOCIATIONS OP THE SYENITE CLAN.— C<mK-Mil
Region
and other alkaline clana
Sediments cut by eruptives
Wisconsin.
1
North Central Wis.
Mica syenite, quartz syenite,
syenite.
Pre-Cambrian b1., paywacke,
calcareous arg., quarteite.
MicmoAN.
PeooVee Dial.
Syenite.
Mica schists and gneisses.
Marquette Dist.
Hornblende syenite.
Thick pre-Cambrian si. and
greenstone.
Minnesota.
Kekequabic Lake.
Nordmarkite.
Pre-Cambrian afg., gray-
wacke, grit, green schists.
Meaabi Range.
Soda syenite.
Pre-Cambrian »l.
Basswood Lake.
Syenite.
SI., miea schist, quortNte.
Sauk Center.
Syenite.
Mica schists.
WhiU Iron Lake.
Syenite.
Arg., micft schist, quart*it«.
Other localities.
Trachyte.
SI.
Red Rock areas.
Syenitic phases.
Animikie si.
New York.
ThouBJind Islands.
Various syenites.
Grenville li., quartsite, et«.
Little Falls Quad.
Syenite.
Grenville li., quartiitc, etc,
Elisabethtown-
Port Henry Quad.
Syenite.
Grenville li,, quartiite, etc.
Paradox Lake
Quad.
Syenite.
Grenville li., qnsrtiite, eto.
LoDg Uke Quad.
Syenite.
GrenTille li., quartsite, etc.
North Creek Quad.
Syenite.
Grenville It., quartsite, etc.
Lyon ML
Trachyte, bostonite.
(LHiroitian R&eisal).
5(K)
IGNEOUS ROCKS AND THBIR ORIGIN
TABLE XXII.— FIELD A6SOCIATION8 OP THE 8YENITB CLAN^— «
Region
IjiKtii hake.
RepreaenUtivet of tiyeniie eUn
and other alkaline cImm
Augite syenite.
SediiDMiU out by flrapCmf
QrenTille U., quArtate,
Peekskill.
Syenite, monsonite.
Cortlandt Dist.
Syenite, trachyte, aodalite qre-
nite.
Brewster Dist. Syenite.
Vermont.
Mt. Ascutnev.
Cuttingsville.
Xew Hampshire.
RedHiU.
Jackson.
Columbia.
Sandwich.
Stark.
Nordmarkite, pulaakite, mon-
xonite» etc.
Syenite, essexite, etc.
Umptekite, foyaite, paisanite,
bostonitc, camptonite.
Augite syenite.
Hornblende syenite.
Mica and grapUtie achislf.
(li.?).
Bmm achiaUy IL, qoarUite.
Mica adikt, li.,
Calfliferoiit miea e^iat, K.
8chirta,lL
Hornblende syenite.
Hornblende syenite
Albany.
Hornblende syenite.
Belknap Mts.
Syenite, essexite, camptonite.
Bant mica adiitU (Monlal-
ban and RoddnghMB)*
Tripyramid Mt. Monsonite, syenite.
Main'e.
.\roo8tookCoiinty. Syenite, trachyte, teschenite. Paleofoio li.^ dk,
Penobstock Bay Syenite, trachyte.
Quadrangle.
SI., li., quartnte;
APPENDIX C
501
TABLE XXII.— FIELD ASSOCIATIONS OF THE SYENITE CLAN.— Conitntied
Region
MASSACHrSETTS.
Representatives of syenite clan
and other alkaline clans
Sediments cut by eruptives
1
Neponset Valley. Trachyte porphyry.
Cambrian arg.
Essex County. ' Nordmarkite, akcrite, etc.
Paleozoic li., atg,, quartzite,
etc.
Connecticut.
Fair Haven. Keratophyre.
TriassioBB.
Mexico.
AltAr Diet.
Syenite.
7
Cerro de Muleros. Syenite porphyry.
Cretaceous li., ss., and marl.
Ferreria San Este- i Trachyte.
ban. 1
1
7
*
Santa Catarina. j Trachyte.
?
Sierra de las Cruces.i Trachyte.
?
Mazapil Valley.
Syenite.
Jurassic and Cistaceous as.,
li., sh.
Cananea Dist. | Syenite, syenite porphyry.
Thick Tertiary li. and vol-
canics.
San Jose Dist. Syenite, nephelite syenite,
camptonite, tinguaite.
Thick CretadeouB li.
1 SOUTH AMERICA
•
Colombia. Syenite, syenite porphyry.
i
PhylliteB, amphibolites, quart-
sites.
Rio Magdalena
(Colombia).
I^tite, quartz syenite, quarts
monzonit€.
Cretaceous and older "Sehi-
chten."
Sao Paulo, Brazil.- Augite syenite, laurvikite, f oy-
aite, etc.
Paleosoic li., si. (ako pre-
Cambrian basie sohiil and
Matto Grosso, Bra-
zil.
Augite syenite.
Pakosoie H., sL, m.
502
IGNEOUS ROCKS AND THEIR ORIOIN
TABLE XXII.— FIELD ASSOCIATIONS OF THE SYENITE CLAN —Ctmham^
RoKion RcpreHcntativw of -yenite clan g^j^^^ ^^ ^y «rupti
and other alkaline clanfl I '
ITff
Madeira River, Augite Hvenite.
Braiil.
Cabo Frio, Braiil. Pulaskite.
Ccara, Braiil. ! Hornblende syenite.
Pre-Cambruui
and li.
Sao Thom^, Brazil.' Trachyte.
Rio Payne.
Akeritc.
? (probably euU
sedimenU).
Aconcag^ua.
Trachyte.
French Guiana.
Me0osoic li., etc.
Syenite.
? (probably cuts
and li. adjaeent)
Cerro Balmaceda, I Monzonites, nordni*irkite, py- Mesotoie iL, li., nwrL
Patagonia. | roxene syenite.
British Guiana.
Great Britain*
Island of Skve.
Syenite.
T
KUROPE
Trachyte, trachyandesite, aye-
Meaosoic and Palfoaok ik^
nite, alkaline granite.
Kiloran Bay, Scot- SyenitCi^ kentallenite, monchi'
land. quite.
Arg\ib*hire.
Kentallenite.
Loch Borolan.
Nordmarkite, borolanite, ne-
phelite syenite, etc.
Colonsay and Or- Quartz syenite, kentallenite,
onsay. monchiquite, etc.
Pembrokeshire. Trachytes, keratophyre, etc
U.
II
Hif^bbnd
CambffiMi IL
TomdoniMi
CambilM tL
ILT)
APPENDIX C , 503
TABLE XXII.— FIELD ASSOCIATIONS OF THE SYENITE CLAN— CaiXiBMrf
Regioi.
Representatives at syenite clan
and other alkaline clana
Sediments cut by eruptives
Fkascb.
Mont Dore.
Trachyte, etc.
Pre-Cambrian mica Hchista,
phyllite; Tertiary li.
Savoy.
Oithophyre.
aa., eg.
UiK
Syenite.
?
M¥ne.
Syenite.
?
Velay,
Trachyte, phonolitc.
Tertiary marls, li.
Morvan.
Trachyte, 8>-enite.
Devonian si., bb., li. and pre-
Devonian Bchists.
lA Sioule VaUey
(Puy-de-Dome
Dist.).
Trachyte.
Spain, PoRTfOAi..
Fori una, Spain.
Trachyte.
Cretaceous li. .Tertiary marls.
Sem de Monchi-
que.
PiiUskite, etc.
Paleoioic li.
Italy.
Biella, Italy.
Syenite.
Mica schists, gneiss.
£ugtinean Hilk.
Trachyte.
Eocene marls.
iBChia.
Trachyte.
Tertiary and Mesosoie claya,
marls, li.
Many localitiee
«taewhere in Italy
Trachyte, etc.
Tertiary and Mesosoie clayfl^
marls, li, ..^
SWITIEBLAND.
^
Aw MaMif, Swit-
aerland.
Syenite.
ment&
504
IGNEOUS ROCKS AND THEIR ORIGIN
TABLE XXII.— FIELD ARSOCIATION8 OF THE SYENITE n IT* riiiffiiiiil
„ Representatives of syenite clan i « .. * ^-
I^<*«">n ««^^*i,.,-ii-i;:..i Sediment cut by eruptivfs
Balkanh.
Montenegro.
and other alkaline clans
Trachyte.
Triaone li., older tedimwiu.
Dobrogea.
Nordmarkite, paisanite, etc. Paleotoie sb., li., m., etc.
Austrian Empire.
Monzoni.
Monzonite, etc.
TriMsic li.; m-j etc.
Predazio.
Monzonite, shonkinite, nepbe- TriMnc li., older
lite syenite, etc.
Duppau Hills, Augite syenite, phonolite, etc. Tertiary mmrk and eakar^
Bohemia. ecus tuffs.
Mittelgebirge,
Bohemia.
Trachyte, phonolite, etc.
Vihorlat-Gutin Trachyte.
MU.
sod Pteleosok
li., mmrly etc.; pro-Csmbnaa
mica mnd hornblende ■rbisls.
n8.| si. I li.
Bulza Mts.
Trachyte.
Brunn.
Syenite.
Limestone.
Blansko, Moravia. Syenite.
Gleichenbcrg
Trachyte, ncphclinite, etc.
Holbak, Sicben- Aegcrite trachyte,
btirgon.
Bukowina.
Quartz keratoph\Te.
"CiysUlliiie sebists.'
81., li., ss., mailsr sL
Germany.
Lower Silesia.
Pluuen Gnind.
Syenite.
Mica aebist, b.
Syenite.
Krzgebirge, Sax- Syenite porphyry,
ony.
Mica sebist, li.,
wacke.
»»>
Dohlen Coal Di.«»t. Syenite.
Silurian pbyOHe.
Ober Lausitz.
Westerwald.
Syenite porphyry.
Pideoioic ebj sL
Trach3rt6, trachyandasite.
Derooian d^
APPENDIX C
505
TABLE XXII.— FIELD ASSOCIATIONS OF THE SYENITE CLAN.— C<mh'ni««(l
Region
Representatives of syenite clan
and other alkaline clans
Sediments out by eniptives
RothflchOnberg.
•
Mica syenite.
PhyUites.
Han.
Keratophyre.
Probably Paleosoio si., li.,
graywacke (?).
Katzenbuckel.
Shonkinite, theralite, etc.
Mesosoio li. and dolomite.
Northern Oden-
wald.
Trachytes.
?
Rhdn.
Trachjrte, phonolite, etc
Mesosoio li. and Paleozoic
calcareous sediments.
Vogelsberg.
Trachyte, phonolite, etc.
Mesosoio li. and Paleosoic
calcareous sediments.
Siebengebirge.
Trachyte, essexite, etc.
Devonian calcareous gray-
wacke and older sediments.
Laacher See.
Sanidinite, etc.
Devonian calcareous gray-
wacke and older sediments.
EifeL
Trachyte, phonolite, etc.
Devonian arg., etc.
RUSHIA.
Magnetberg, Urals.
Syenite, syenite porphyry, kera-
tophyre.
Paleosoio li.
Zalas, near Cra-
cow.
Syenite porphyry.
Mesosoio li.; Carboniferous
and older sediments.
Miask, Russia.
Corundum syenite, nephelite
syenite, etc.
Crystallme li.
Asof, Russia.
Orthophyre, grorudite, etc.
Paleosdo li. ?
Piatigorsk (north-
ern Caucasus).
Trachyte, alkaline liparite.
Eocene and Meoosoio sedi-
ments.
Ah venaara, Finland
Syenite, essexite, etc.
T
Kola peninsula. Umptekite, nephelite syenite,
etc.
Mariy day slate, etc.
506
IGSEOVS ROCKS ASD THBIR ORIGIS
TABLE XXII. FIELD A.-WlCIATloXrt OF THE SYEXITE CIAS.
Kf^on
Repreflentativefl of syenite cUn
and other alkaline clans
Sediroenta cot by wpiUfes
Sweden*.
Northern Smaland.
Orthophyrp. ratapleiite syen-
ite
? (Granite, effiMnre <|iiaru
porphyiy).
del li van? Diai.
Syenite, quarts syenite, alka-
line granite.
Mica schiata, ampjiibolite.
Kafninda Dint.
Nordmarkite.
SchisU.
EkfftrdmflbcrK Dirt.
Soda syenite, keratoph>Te.
Chlorite schisl, li.. quartaite,
etf.
I^iikkujar%'i.
Syenite granulite.
Li. and greeostooe.
Svoppavaara.
Alkaline svenite^
Ditto.
Mf*rtaim*n.
Syenite.
*
Painirova.
Syenite fwrphyrj'.
•
Kinina Di-it.
Ditto.
Greenstone; also probably li.
and other basic aediractiU.
Tome Trask,
Lappland.
Syenit«\
Pre-Cambrian greoi ■rhists
and dolomita.
Norway.
BiTgcn, Norway.
Monzonite, Hoda syenite, man-
gerite.
Pre-Cambrian mica ackisU;
Silurian "aedimenta."
('hristinnia Reg-
ion.
Monzonite, nonlmarkite, pu-
lankite, etc.
Paleosoic li. and argiilitc.
ASIA
Tsang and C pro-
vinces, TiJ>et.
Syenite.
Paleosoic and Meaoaoie
si. and li.
Western Shan-
tung, China.
Syonite porphyry, quarts sye-
nite porphyry.
Paleosoic ah., ss., li.
Korea (Park-tsrh*-
hoii).
Symiti- porphyry.
PhyUite.
TABLE XXII.-
APPENDIX C 507
J ASSOCIATIONS OF THE 8VENITB CLAN.— C<Hi«iHi«f
_ . Representatives of syenite clan
**^^°^ 1 *nd other alkaline cluna
Sediments cut by eruptives
Urtioi Highland
(Lena River).
MoMonite, syenite.
Schists, 8l., thick IJ.
Yeniaei.
Monionite, umptekite, etc.
81., dolomite, thick li., phyl-
lite, mica schist.
Siberia.
Syenite.
Paleosoic sediments, includ-
ing U.
Wyi*-Teich (Ural)
ayenit*^iiorite.
7
Nijni-Tagilsk.
Syenite.
Li., 8l., tuffs.
Dlagodat, Ural
Ditto.
Thick Devonian li., etc.
^■agfttapatam,
India.
Charnockite, syenite.
?
Madras, India.
Augite syenite, corundum sye-
nite, ncphelite syenite.
?
Coimbatore Diat.,
India.
Corundum syenite.
"CrystaUineBohiatB."
Charnockite.
Pre-Cambrian li.
Iflhan Peninsula,
Arabia.
Alkaline trachyte.
7
PaUndottan Pla-
teau, Armenia.
Trachyte.
Probably cuts Mesoaoio ti.
adjacent.
Demavend Vol-
cano, Persia.
Trachyte.
Paleosoic and Meeowio U.
Kfldi-Kale, Smyr-
Syenite.
PhyUite, eg., li. (7).
Kimituria Mine.
Trachyte.
Probably cuts li., etc.
Dcvelikoi.
Ditto.
PhyUites, eg. and li. (T).
Sary-BoulalcGor^,
Turkestan.
Syenite.
Li. and si.
Borio River, Turk-
eaUn.
Ditto.
Li.
508
IGNEOUS ROCKS AND THEIR ORIGIN
TABLE XXII.— FIELD ASSOCIATIONS OF THE SYENITE CLAN.-
„ . Representatives of syenite cUn ^ .. . . . ..
^K'"" and other Ikalme clan. i S««UDent. cut by «n.pUT»
Kirai-Ruir River, Ditto.
Turkestan.
U.
AFRICA
Abyssinian Pla- Trachyte, microsyenite.
teau.
li., M., phyllite,
Cireat Rift Valley. Trachyte, phonolite, etc.
I
Air, Central Africa. Trachyte, phonolite.
Sflurimn li.
iSocotra.
Syenite porphyry.
I Amphilxrfitff,
Adamana, Kame- Trachyte, phonolite, nephelite
run. s>'enito.
Phyllite, green aehist, mark
li., amphibolile,
Cape Verde Pcnin- Trachyte,
sula.
Fontaine du G^nie, Monzonite.
Algeria.
Doomberg, Cape Syenite.
Colony.
Sh.y se.| li.f
Potchofstroom Syenite, nephelite syenite.
Dist.
Dolomite, m.. thiek d.
quartiite.
Dushveldt.
Monionite, nephelite syenite, - SI., doloinite, qoMttahm.
bostonite, camptonite, mon-
chiquite.
Los Islands.
Pulaskite, monzonite, essexite,
shonkinite, nephelite syenite,
etc.
XEwSorxH Wales
AUSTRALIA
Warrumbunj^Ie. Trachyte, phonolite, etc.
Mts.
TriMnc ealcAreoiit A., Filro-
soie li. uideli.
Nandcwnr M ts. Akerito, trachyte, nephelite Paleoioie H., sh., cte.
syenite, etc.
Canohdljw Mts. Trachvto
Ditto.
APPENDIX C 509
TABLE XXII.— FIELD ASSOrlATlONS OF THE SYENITE CLAN.— CoiKinwd
_ Representatives otayenite clan o,jj_ , . .
■^K"'" .nd .Ihe, .iMi.. .1... '■ ^"^^ "' I-? <>■■?'■"»
Bowral. ' Aegerite Bycaite. j PaleoHoic ]j,, sh., etc.
Mittagong-
Syenite, trachyte.
Ditto.
5 other districts.
Ditto.
Ditto.
QUEENBI^ND 1
Sdistricu.
Ditto.
?
Victoria.
Mt. Mftcedon.
Traohyteo.
South Aostraua
HouBhton.
Syenite,
Calcareous schist and li.
Aldgate.
, Ditto.
Ditto.
YH>kaJllU,
Ditto.
?
ISLANDS
Port Cygnet, TaB-
monia.
Faleoioic U., etc.
Duncdin, New
Zoalanil.
Trachyte, foyaite, etc.
Tertiary li. and calcareous
■»., etc.
Trachyte, kenyite, etc.
Tahiti.
Trachyte.
MonEonite, etc.
New Pomeranift.
Monionite.
T
Solomon Ida.
Trachyte.
Juan Fcrnandec.
Ditto.
510
IGNEOUS ROCKS AND THEIR ORIGIN
TABLE XXII.— FIKLI> A8HOCIATIONH OF THE SYENITE CLAN.-
Region
Representatives of syenite clan ^ .. . ..
1 .* n I- I Sediments cut by erupt iTM
and other alkaline clans
Oki Ids. (Japan). Quartz syenite.
Binangonan Pen- Trachyte,
insula, Philip-
pines.
Tertiary sh., ss., etc.
Eocene li., etc.
Liu Rawas, Sum- Monzonite.
atra.
Sh., si., li.
Madagascar (10 Monionite, syenite, trachyte, }
districts). etc.
ic arg., li., maiit etc.
Reunion.
Syenite, trachyte, etc.
Seychelles.
Syenite.
Clay slate.
Kerguelen.
Trachyte.
Ascension.
.\lkaline trachyte.
Canary Ids.
Madeira.
Monionite, etc
Trachyte.
Cape Verde Ids. Syenite, foyaite, etc
.\ lores.
Trachyte, etc.
Miocene li., etc.
Los Ids. (See p. 50S. ■ Pulai<kite. munzonite, etc.
Samos.
Cyprus.
Trachyte.
Trachyte.
Li. and bane cruptivea, etc
Li., marla, basic cnipUvea.
Columbretes.
Ditto.
Lofoten Id.**.
Monzonite. Kyenite.
Dolomite,
J u 1 i a n e h a a l> , Nordiiiarkite, pulaskite, foya- (Granite and ia.T)
GnM'iiland. ite, etc.
APPENDIX D
TABLE XXIII.— LIST OF DISTRICTS CHARACTERIZED BY ALKALINE ROCK-
TYPES: WITH NOTES ON THE NATURE OF COUNTRY ROCKS
(See page 415 and also Appendix C)
Region
Alkaline eruptivcs
Carbonate rocks cut by
alkaline eruptives ^
NORTH AMERICA
Quebec.
Mt. Brome. Essexite, laurvikitic syenite,
I nordmarkite, etc.
Paleozoic li. and sh.
Mt. Johnson. Essexite, camptonite, sdlvs-
% bergite, pulaskite.
1
Ditto.
Mt. Royal.
Essexite, nephelite syenite,
bos tonite, tinguaite,sdlvsberg-
ite, camptonite, fourchite,
monchiqiiite, aln5ite.
Ditto.
Mt. ShefTord. Essexite, theralite, bostonite,
' camptonite, pulaskite, etc.
Ditto.
1
Montarville. ' Essexite.
1
Ditto.
Rougemont.
Essexite.
Ditto.
Mt. St. Hilaire
(Beloeil).
Nephelite syenite, sodalite
syenite, essexite, pulaskite.
Ditto.
Mt. Yamaska.
Essexite, akerite, yamaskite.
Ditto.
Ontario.
Dungannon Tp.
Nephelite syenite.
Grenville li.
Faraday Tp.
Ditto. .
Ditto.
Glamorgan Tp.
Ditto.
Ditto.
1
Harconrt Tp. Ditto.
Ditto.
Methuen Tp. j Ditto.
Ditto.
^ Abbreoiatians :ATg. — argillite; eg.— conglomerate; eh. — shale; sL— «late; m.
— sandstone. Many references to authors found in Rosenbuseh's haadbook.
511
512
IGNEOUS ROCKS AND THEIR ORIGIN
TABLK XXIII.-LI8T OF DISTRICTS CHARACTERIZED BY ALKALINE ROCK-
TVPK8: WITH NOTKS OS THE NATURE OF COUNTRY ROCK.-CMlMMrf
Region
Monteagle Tp.
Alkaline eruptives
Ditto.
Monmouth Tp. Ditto.
CarboiiAte roelu cut by
alkaline eruptiTea
Ditto.
Ditto.
Raglan Tp.
Wollaston Tp.
.1.
Ditto.
; Ditto.
Ditto.
In diorite cutting GicnTiDe li.
Pooh-bah Lake. Malignite.
Port Coldwcll.
Nephelite syenite, essexite, ? (Keewatin chlgrite aeliiala^
laurvikite, camptonite, etc. etc.)
Alhf.rta.
HIairmorr.
Analcitc trachyte
Paleoioic li., ah., et«.
Hriti.sh
CoLrMBIA.
Ice Uivcr.
Xrphelite syenite, ijolite, can- Thick pre-Ordorieiaa dolo*
rrinite syenite, urtite, tingua- mitea and li.
ite, etc.
Ros.slan<l
Missourite, monzonite, latites. Carboniferous li.
Kettle River.
Rhonih-porphyr>', analeitic Paleoioic li., arg.
rhoinl>-porphyr>', alkaline
trachvtc.
Camp Iledli'V.
KriiptT Mt.
MONT.WA.
Bearp.'iw Mt.*^.
Keratoi>hyre, monzonite.
Ditto.
Xophelite syenite, malignite. Ditto.
Leiiriteb.'uialt, nephelite basalt. Paleozoic and Mmomdt li.
leurititi>. tinguaite, trachytes, and dolomite.
syenites.
High wood Mti*.
IxMieite l>asalt , scnlalite syenite. Ditto.
slKinkinitf, mis-sourite, anal-
rite basalt, svenites. etc.
Judith Mt^.
TinKuait^•^», phonolit«'s, sven- Ditto.
iU'.
APPENDIX D
513
TABLE XXIII.-LIST OF DISTRICTS CHARACTERIZED BY ALKALINE ROCK-
TYPES: WITH NOTES ON THE NATURE OF COUNTRY ROCKS.— Continued
Region
Alkaline eruptives
Carbonate rocks cut by
alkaline eruptives
Little Belt Mts.
Analcite basalt, shonkinite,
syenite, monzonite.
Ditto; also Beltian li.
Castle Mts.
'I'heralite, acmite trachjrte.
Ditto.
Crazy Mts.
Ditto.
Ditto.
Elkhorn.
Shonkinite, bostonite, syenite.
Paleozoic argillaceous li.
Wyoming.
Sundance Quad-
rangle.
Nephelite syenite, ijolite, leuc-
ite porphyry, bostonite, neph-
elinitc, camptonite, monzon-
ite, etc.
Paleozoic li. and arg.
Aladdin Quad.
Ditto.
Ditto.
Absaroka Quad.
Leucitic syenite, etc.
Ditto.
Leucite Hills.
Wyomingite, orendite, madup-
ite.
Paleozoic and Mesozoic li.
South Dakota.
1
1
Black Hills. .Phonolite, grorudite, tingua-
: ite.
1
Pre-Cambrian calcareous sed-
] iments; also Qocally) Paleo-
zoic and Mesozoic li.
1
Colorado.
•
Denver Basin. Trachydolerite.
Mesozoic li.
Cripple Creek. , Nephelite syenite, phonolite,
i syenite, trachydolerite, etc.
1
(? Intruded into Pre-Cam-
brian granite!).
•
Georgetown. Bostonite, latite, etc.
Li., calcareous sh. and ss.
Engineer Mt. Monchiquite, camptonite, mon-
zonite.
1
Paleozoic li., sh., ss.
Elkhead Mts. Nephelite basalt, trachyte.
1
?
514
lONSOUS ROCKS AND THEIR ORIGIN
TABLE XXIII.-LIST OF DISTRICTS CHARACTERISED BY ALKAUNB ROCE-
TYPES: WITH NOTES ON THE NATURE OF COUNTRY ROCK8. -€*•«*'•« W
Region
Nevada.
Alkaline eruptives
Cftrbonmtc rocks cut by
alkaline eniptiTCf
BullfroR.
Leucitc bawilt.
Li.
New Mexico.
Las Vegas.
California.
.\nalcitic camptonite.
Paleoioic li. mod Mctosoir
sedimenU.
Point Sal.
Teschenite.
Tertiary li. and clays; CreU*
ceous shalcii.
San Luis Obispo
County.
Ditto.
Ditto.
Texas.
Brewster County.
Phonolite, trachyte.
Thick Cr?t«eeoiiB li. and sh.
Rio Grande Plain
(Pilot Knob).
Nepholite basalt, melilite ba-
salt, orthochise basalt, phonol-
ite, lifnburgite.
Cretaceous (and older?; li.
Apache Mts.
Nephelite syenite, phonolite,
tinguait«*, bostonite, paisan-
ite, syenite.
Paleosoic and McwMoic b
and marl.
Uvalde County.
Nephelite basalt and basanite,
melilite basalt, phonolite,
limburgite.
Mesoioic (and older?) U. and
maris.
•
1
.\RKANAAft.
1
Magnet Cove. : Ne|)hi'lit<» Hvpnito. ijolite,
ffhonkiiiite. jarupirangite,
ItMirito ptirphyry, monrhi-
quite, tinguaito.
Paleosoic mainwsisn li.
WwcONrtlN.
North-central
Nephelite syimite, syenit<*«.
Calcareous arg.
part.
APPENDIX D
515
TABLE XXm.-LIST OF DISTRICTS CHARACTERIZED BY ALKALINE ROCK-
TYPES: WITH NOTES ON THE NATURE OF COUNTRY ROCKB.—Continumi
Region
Alkaline eruptives
Carbonate rocks cut by
alkaline eruptives
New Jersey. 1
1
.
1
BeemerviUe. Nephelite syenite.
Thick limestone.
Brookville.
Ditto.
Li. adjacent.
Franklin Furnace.
Ouachitite.
Crystalline 11.
Virginia.
Augusta County.
Nephelite syenite, teschenite.
Paleozoic li. and sh.
New York.
Cortlandt.
Sodalite syenite, trachyte, syen-
ite.
Thick li.; basic schists.
Tiake Champlain.
Bostonite, monchiquite, four-
chite, camptonite.
Pre-Cambrian and Paleosoic
li.
Lyon Mt.
Bostonite, trachyte.
(Laurentian gneiss!)
Massachusetts.
'
Essex County.
Foyaite, essexite, syenite, etc.
Lower Paleozoic li. and cal-
careous arg.
New Hampshire.
Red Hill.
Foyaitc, bostonite, campton*
it^, paisanite, umptekite, etc.
(Gneiss and granite!)
Belknap Mts.
Essexite, syenite, camptonite.
(Basic schists.)
Vermont.
Ascutney Mt.
Essexite, alkaline syenites,
etc.
Paleozoic li. and caleareous
schists.
Maine.
Aroostook County.
Teschenite, syenite, trachyte.
Paleosoie li., sh., m.
516
IGNEOUS ROCKS AND THEIR ORIGIN
TABLE XXIII.-LIST OF DISTRICTS CHARACTERISED BY ALKAUNB BOCK.
TYPES: WITH NOTES ON THE NATURE OF COUNTRY ROCKB.*
Rogion
Mexico.
Alkaline eruptivcs
Carbonftte rocks cut by
Alkaline cmptiYea
San Jose.
Nicaragua.
Uani.
Costa Rica.
"Eastern part".
Culebra Bay.
Nephclite syenite, camptonite,
eyenite, tinguaite.
Thick CretaeecNis tt.
Phonolite.
T
Theralite.
Tertiai7 (Bad oMarT) B.
Phonolite.
Nicoya Peninsula. Toschenite
Avanfi;are8 Dist. | Limburgite.
SOUTH AMERICA
Thick U.
Tertiary H.
Brazil.
Sao Paulo.
Foyaito, jacupirangitr, teschen- Patooamc U. in aL;
; ite, nopholinitc, augitite, lim- brian li. T).
burgite, syenite, etc.
Caldas.
Foyaite, phonolite, leucito- ,
phyre.
Paraguay.
(Locality?)
Argkntine.
Phonolite, limburgite.
Salta Province. tijwxite, trarhy«lolerite.
San Juan Province. Ksscxito, nephelite basalt, ; Paleosmc (T) and Ji
ielL
limburgite, trachyte-tephritc. I
Patagonia.
Sul^andine region. K.ssexite, tniehydolerite, anal- ; Li., calcareous arB«i phjrDitc.
citioessexite, camptonite. j
APPENDIX D
517
TABLE XXIII.-LI
TYPES: WITH ]
Region
ST OF DISTRICTS CHARACTERIZED BY ALKALINE ROCK-
NOTES ON THE NATURE OF COUNTRY ROCKS. -Continued
Alkaline eruptives
Carbonate rocks cut by
alkaline eruptives
EUROPE
ScOTT*AND.
Loch Borolan.
Nephelite syenite, borolanite,
pulaskite.
Paleozoic li. and dolomite.
Inchcolm Island, i Teschenite, picrite.
Paleozoic calcareous sedi-
ments.
Lugar. Teschenite.
Paleozoic li., etc.
Hebrides. Monchiquite, crinanite, etc.
1
Torridonian li., etc.
Eildon and Garlton Phonolite.
Hills. 1
1
Paleozoic li. (probably).
Kilpatrick Hills. Mugearite, trachydolerite, lim-
burgite, etc.
Paleozoic li., etc.
England and !
Wales. j
Lureombe, Devon. Teschenite.
Thick Devonian li.
Golden Hill, Mon-| Monchiquite.
mouth. 1
Old Red marls, etc.
Ireland. 1
Rath Jordan. Analcite basalt.
7
France. j
1
1
Velay. ! Phonolite, trachyte.
Tertiary marls, li.
MontDore. Phonolite, tephrite, trachyte.
1
? (Crystalline complex).
1
Cantal.
Phonolite, trachyte.
Li. in crystalline basement
complex; Oligocene li., marl.
Pouzac.
Nephelite syenite, bostonite, | Mesozoic li.
etc.
Fitou (Aude).
Nephelite syenite.
((
Mesozoic sediments."
518
IGSEOVS ROCKS ASD THKIR ORIGIS
TABLE XXIII.-LI8T OF DISTRICTS 'CHARACTERIIED BT AtKAXJlSm BOCK.
TYPES: WITH NOTES ON THE NATURE OF COUNTRY ROCBR-Ci
Region
Alkalinr erupt ives
CmiiMMiate rodw c«t bj
Spain.
Fortuna.
Fortunite, jumillitc, trachyte. Tertiary
Catalonia.
Nephelite basanite, limburgite. Tertiary and older M.
PoirrroAL.
Serra de M o n-
chique.
; NVphelite syenite, monchi- , IVe^ulm li.; Ji
, quite, camptonite, boetonite, roof(?).
tinguaite, pulaakite.
Cezimba, Fonte da Teschenite.
Bica, etc. ;
Cevadaefl.
Nephelite gneiss.
Creiaeeoua and older K.
Sehista, iocludiag
caterocka.
Italy.
Thirteen dijitrirtfl Leucite tephrite, trachyte. Thick MflMsoie and Tcrtiaiy
as under: phonolite, leucite basanite, li. and doloinite.
V u 1 8 i n i a n , leucitite, mclilitic leucitite,
Ciminian, Saba- ■ latite, etc.
tinian, Latian, ;
Hernican, Au-
runcan, C a m -
panian, Vesbian, ■
Phlegrean, Mte. |
Vulture, Tuscan.
Venetian, Apu-
lian (namod by
H. S. Washing.'
ton).
Germany.
Kaiserstuhl, ; Phonolite, nephelite basalt, MeMMoic and Tertiary li ,
Baden. ! limburgite, leucite basalt, ■ dofemitea, and
' tophrite, leucite, phonolite, |
' etc. i
Haanlt Mount- Lirnbuffcite
ain9. Els ass:
Mainz basin, j
etc. I
MuMshelkalk and
youiifCTli.
oUs and
APPENDIX D
519
TABLE XXIII.-LIST OF DISTRICTS CHARACTERIZED BY ALKALINE ROCK-
TYPES: WITH NOTES ON THE NATURE OF COUNTRY R0CK8.— Continued
Region
Alkaline cruptivcs
Carbonate rocks cut by
alkaline eruptivee
Katzenbuckel, Nephelite basalt, shonkinite,
Baden.
theralite, etc.
Hegau, Baden.
Nephelite basalt, m e 1 i 1 i t e -
nephelite basalt, melilite ba-
salt, phonolite.
Swabian Alp.
Siebengebirge.
Melilite basalt.
Mesozoic li. and dolomites.
Ditto.
Ditto.
Essexite, trachydolerite, mon-
chiquite, trachyte, etc.
Eifel.
Vordereifel.
Westerwald.
Leucite basalt, nephelite basalt
(bearing melilite), phonolite.
Ditto.
Phonolite.
Weser-Werra
Fulda District.
Rhon.
Nephelite basalt, leucite basalt,
melilitic nephelite basalts.
Vogelsberg.
Phonolite, trachyte, nephelite
basalt (often leucitic), lim-
burgites, basanites.
Nephelite basalt, phonolite,
trachyte.
Devonian calcareous gray-
wacke and older sediments.
Ditto.
Ditto.
Devonian li. adjacent.
Mesozoic li. and marls and
Paleozoic calcareous sedi-
ments.
Ditto.
Ditto.
Saxony, many Phonolite, nephelite basalt,
localities.
Austria.
Predazzo, Tyrol.
Monzoni, Tyrol.
Nephelite syenite, essexite,
shonkinite, theralite, mon-
zonite, tinguaite, camptonite,
monchiquite, etc.
Monzonite, essexite, alkaline
syenites, tinguaites, etc.
(Lausitz granite.)
Triassic dolomite. (Also
older H.?)
Duppau Hills, Bo-
hemia.
Phonolite, leucite basalt, leuci-
tite, leucite tephrite, leucite
basanite, nephelite basalt,
Ditto.
Tertiary marls and caloare-
ous tufifs.
520
IGNEOUS ROCKS AND THEIR ORIGIN
TABLE XXIII.-LiaT OP DIdTRICTS CHARACTERISBD BY ALKAUNB BOCK-
TYPES: WITH NOTE8 ON THE NATURE OF COUNTRY BOCKS.-
Region
Diippau Hills, Bo-
hemia.— Con.
Alkaline cruptives
nephelinite.nepheiitetcphrite, :
limburgite, augititc, oeph- >
elite Byenite, augite syenite,
theralite.
CarbooAta roein cut by
alkaline erupUTM
Mittelgebirge, Phonolite, tcphrite, essexite, Meaosoio and
Bohemia. sodalitc syenite, trachyte, ' and maria.
nephelite basalt, limburgite,
melilite basalt.
ie b
Steiermiirk, Aus- ' Nephelite basanite, nephelin- Tertiary Meaoioie (and old-
tria. ite, nephelite basalt. ; er7) li. and marb.
Ditro, Hungary. Nephelite syenite (many' Li. in pt^rllitie tetmne. (Alfo
phases). younfer limftonf T)
Teschen.
Teschenite.
Li. and ma^ "Scliirfi
Meclves Mts.
Nephelite basanite.
Hn4.siA.
(^luoasus.
Teschenite.
MeiQfoie li. and arfiOite.
Azof.
Mariupolite, orthophyre, ' 7 Paleoioie (Deronian and
groruditc, pyroxenite, etc. Carboniferoua) IL ?
Mi:isk and Kiiss.i. Nephelite syenite, cancrinite Crystalline li. adjacent .
syenite, corundum syenite.
Kuusamo Parish, Nephelite syenite, ijolite.
P'inland.
Kiiohijiirvi, F i n- Cancrinite syenite,
land.
Limestone.
Kola Peninsula.
N( phelite syenite, u r t i t e, Devonian (?) mmAj day-
lujavrite, umptekite, chibin- slates. (Also li. 7)
itr, etc. !
Ahnrvaara.
I-'ssexite, syenite.
SWKDKN.
.\lnn.
Foyaito, ijolite, ja(Mipirangitic Crjrstalline li.
phases, hoxtonite, aln6ite, etc.
APPENDIX D
521
TABLE XXIII.-LIST OF DISTRICTS CHARACTERIZED BY ALKALINE ROCK-
TYPES: WITH NOTES ON THE NATURE OF COUNTRY ROCKS.— CoiUinued
Region
Alkaline eruptives
Carbonate rocks cut by
alkaline eruptives
Elfdalen.
Cancrinite tinguaite.
Paleozoic 11.
Norway.
Gran.
Essexite, bostonite, camp ton-
ite.
Paleozoic 11., including Etage
lto3.
Christ iania.
Essexites, akerites, laurvikites,
monzonites, laurdalite, pulas-
kite, nordmarkite, ekerite,
camptonite, rhomb-porphyry,
etc.
Paleozoic li. (and arg.).
1 ASIA
Smyrna. ' Leucite tephrite.
1
Li., marl.
Tschamly Bei. j Leucite basanite.
Li.
1
Kesmek-Kopni. | Nephelite dolerite.
1
Marls.
1
TrebiionA Leucotephrite.
1
1
Eocene li. and Cretaceous
sediments.
Kula.
Kulaite (trachydolerite), leuc-
ite kulaite.
Tertiary (and older?) li.
Troad.
Nephelite basalt.
Thick Cretaceous and Ter-
tiary li.
Northern Syria.
Nephelite basanite, limburg-
ite.
Cretaceous li.
Aden PeninHula. Phonolite, trachyte.
Cretaceous IL 7
Cah6tie Mts., Geor-
gia.
Teschenite, dacite, etc.
Cretaceous li. ; Tertiary marls
and clays.
Upper Zarafshan | Nephelite syenite, sodalite By&-
(Turkcstan). nite.
1
Li. and calcareous schist.
Madras, India.
Nephelite syenite, syenite, cor-
undum syenite.
7 (Li. in complex?).
Raj pu tana, I ndia. Nepheli te and sodalite syenites
1
? (Li. in complex?).
Coimbatorc, I ndia. I Nephelite syenite.
? (Li. in complex?).
522
IGNEOUS ROCKS AND THEIR ORIGIN
TABLE XXIII.-LI8T OF DISTRICTS CHARACTERISED BY ALKAUNB ROCE-
TYPES: WITH NOTES ON THE NATURE OF COUNTRY HOCKS.-
Region
Alkaline eniptives
Mount Girnar, Xcpholitc syenite, monchi-
India. quite.
Manchuria.
Nephelite baflalt.
Carbonate roein cut by
alkaline enipth
T
? (Cambrian li. in reiion).
Yenisei, Siberia. Nephelite syenite, Icucite sye- Thick, eztenaiTe li., with dol-
nite porphyry, camptonitc, etc. omite and alate.
(t
Ostsibirien."
Tesohenite.
Li. and dolomite.
Southern China. Nephelite syenite.
? (Thick,
calcareous
. gion.)
Kamenin, Africa.
AFRICA
Nephelite syenite, phonolitc, Cretaceous (and older?) li.
monzonite, keratophyre, boe- ;
tonite, camptonite, vogesite,
leucitite, etc.
Niger- Bcnu6 Area, P'oyaite
West Africa.
Li. of unknown
AhaggDr, Central , Phonolite.
Africa . i
Air, Central Africa. Phonolite. alkaline trachyte. Silurian li. (Cretaceous OMris
?; other li.t)
Great Hift Valley, Phonolitcs, coinendites, trachy-
Eoflt Africa. te8,kenyites,nephelinitcs(beAr-
ing melilite), borolanite, neph-
elite basalt, limburgite, teph-
rite, nephelite basanite, leu-
cite hasanite.
? Oyvtalline li. of
plateau; cakarm
stone, conglomente and
shales of PaleoaoJeT
Ahv88inia.
Phonolitf^, tinguaites, groru- (Thick crystalline li. and di>
(lite, pairianite, solvsbergite, ! lomitcs in gneiasae pbtean;
etc. alsoJuraasieandyouaivli.)
Pot chefst room. Nephelite syenite.
Dolomite, si., etc.
Bushveldt, Trans- Nephelite syenite, monchiquite. The Great Dolomite.
v:uil. camptonite, Inrntonite, etc.
.Marifo, Transvjial .Xephelite syenites.
Ditto.
APPENDIX D
523
TABLE XXIIL— LIST OF DISTRICTS CHARACTERIZED BY ALKALINE ROCK-
TYPES: WITH NOTES ON THE NATURE OF COUNTRY ROCKS.— Continued
Region
Alkaline eruptives
Carbonate rocks cut by
alkaline eruptives
Spiegel River, Cape Melilite basalt.
Province.
I
Sutherland, Cape Ditto.
I*rovince.
Xamqualand.
Nephelite basalt, melilite- neph-
elite basalt.
NewSouth Wales
AUSTRALIA
Warrumbungle
Mts.
Phonolite, trachydolerite, com-
enditc, melilite basalt (bear-
ing corundum), trachyte, etc.
Nandewar Mts.
Canobolas Mts.
Nephelite syenite, bostonite,
syenite, trachyte.
Li., calcareous sh. and schist.
Ditto.
Li. in gneissic terrane and in
Malmesbury beds.
Mesozoic calcareous sh.;
Paleozoic li., etc.
Paleozoic li. and calcareous
sh., etc.
Melilite basalt, comendite,
phonolitic trachyte.
Dubbo.
Mittagong.
Nephelite syenite, phonolite,
trachyte.
Essexite, trachyte, syenite.
Ditto.
Ditto (probably).
Mt. Prospect.
Essexite.
Kiama-Jamberoo.
Barrigan.
Nephelite syenite, tinguaite,
monchiquite, etc.
Paleozoic calcareous sh. (and
li. probably).
Triassic sh. (probably Paleo-
zoic li.).
Permo-CarboniferouB sb.
(probably Paleozoic 11.).
Tinguaite.
Capertee Valley. ■ Nephelite basalt.
Permo-CarboniferouB sh.
(probably Paleozoic li.).
Kosciusko.
I Phonolite.
Queensland.
Glass House Mts.
Pantellerite, comendite, alka- ! Mesozoic sh., etc. (Paleo-
line trachyte, keratophyre, j zoic li.?)
bostonite. I
Maroochy-Cooran i Ditto.
Ditto.
35
524
IGNEOUS ROCKS AND THBIB ORIGIN
TABLE XXIII.-LI8T OF DISTRICTS CHARACTERISED BT ALKAUNB ROCK-
TYPE8: WITH NOTES ON THE NATURE OF COUNTRY ROCK&— CmImmW
Urffion
Alkaline eniptiTes
Carbonale rocka cut by
alkaline
Mt. Flinders*- FaH- Phonolite, pantellerite, comeiH
sifom. <litc.
Highly
■edimenta.
Kiist Moreton.
Aniilcite dolerite, comendite, ; Metoioie and
solvsborKite, pantcllcrite, moD- 1 (and liT).
zonite, trachyte.
•nt.
Vlf-TORIA.
Mt. Macctloii.
Sdlv»borRito, alkaline trach3rtCy - (Ordovician ah.)
inacedonitOy limburgite, anor-
thoclasc basalt, etc.
SoiTH ArSTRALIA.
Harden.
Leucite basalt.
(Silurian al.)
ANTARCTICA
HoHri Archi|)elago.
Phonolite, traohy dolerite, ke- ? (Thick li.
nyite, leiirite kenyite, limbur- terrane.)
gite, camptonite, trachyte.
in
ISLANDS
Refloat ta Point
(Taf^niania.)
Xopholite Hvenite, eflsexite, jao- ■ Paleoaoic IL
upirangite, sdlvHbergite, tin-
ffuaite. nephelinite, syenite,
melilite b:u<alt, liniburgite,
etc.
liohart (T:i«mania^ Xephelite basalt.
Shannon TiiT (Tat*- Melilitt*-nephelite basalt, eudi-
maniaj. alyte-nephelite basalt.
Diine<iin i New
Zealand'.
Foyaite, trachy dolerite, tea- Tertiary" li. and
chenite, phonolite, tinguaite, aedimcnta. ((Xdarfi?)
leucitophyre. nephelite baa- I
anite, melilite basanite, tra- =
chyte, etc.
Campbell I.^land. IMionolite, melilite basalt.
MiooenelL
.\urkland Islands. .Mkaline trachvte.
Pfissrssion Inland. IMionolite.
APPENDIX D
525
TABLE XXIII.— LIST OF DISTRICTS CHARACTERIZED BY ALKALINE ROCK-
TYPES: WITH NOTES ON THE NATURE OF COUNTRY ROCKS.— ContiniMd
Region
Alkaline eruptives
Carbonate rocks cut by
alkaline eruptives
Juan Fernandez.
Phonolitic trachyte.
Ponapt Island
(Carolines).
Nephelite basalt.
Raiatea Island j Phonolite.
(Society Group).
Viti (Fiji).
Foyaite.
Tahiti.
I Nephelite syenite, tinguaite,
! monchiquite, phonolite, ess-
? (Crystalline li. in island
basement.)
exite, monzonite, picrite.
Savaii, Upolu, and
other islands (Sa-
moan Group).
Trachydolerite, phonolite, ne-
phelite basanite.
?
Java.
Tephrite, leucite basalt, leuci-
tite.
Li., marl.
Loh oclo (Java) .
'I'heralite-d iabase.
Calcareous Eocene and Cre-
taceous sediments.
Western Celebes.
Leucite basalt.
?
Northern Celebes.
Phonolite.
?
Saleyer (Moluccas)
Nephelite tephrite.
?
Bum Island (Indo-
Australian Archi-
pelago).
Melilite basalt.
?
Timor.
Foyaite.
Tertiary and Paleozoic li.
Hawaii.
Phonolitic trachyte, trachy-
dolerite, etc.
?
Maui.
Melilite-nephelite basalt.
?
Oahu.
Nephelite basalt, melilite-ne-
phelite basalt.
?
•
526
IGNEOUS ROCKS AND THEIR ORIGIN
TABLB XXIII.-LIST OP DISTRICTS CHARACTERIZED BY ALKALINE ROCK-
TYPES: WITH NOTES ON THE NATURE OP COUNTRY ROCKS.-
Region
Alkaline eniptives
Dependent Isles A nalcite basalt,
of Taiwan (off
Japan).
Carbonate roeks cut by
alkaline eniptr
T
Reunion.
Essexitic gabbro, mugearite,
phonolitic trachyte, syenite.
Madagascar.
Nephelite syenite, nephelite Jurassic li. and marb.
basalt, phonolite, trachyte, li.t)
limburgite, camptonite, mcli-
lite basalt, laurvikite, monso-
nite, augitite, essexite.
(Oldrr
Trinidad, South Phonolite, nophelinite, limbur- ? Coral reefs?
Atlantic. gite.
Heard.
NVpholitc basalt, limburgite. 7 (Li. blocks in crater).
Kergiielen.
Nightingale.
Phonolite, trachyte, limbur- 7 (Roth found dokmiite
gite. here.)
Phonolite.
Fernando Noronha Phonolite.
Cabo Frio ( Brazil). Foyaite, puhuikite, essexite,
tingiiaito, monchiquite.
Saint Helena.
Phonolite.
Ascension.
Alkaline trachvte.
Sao Thom^.
Phonolite, trachyte, limbur-
gite.
Los LMandi* (Af- Nrpholite syenite,
rica).
Selvagem Oanvle Phonolite. ni^phelinite, limbur- 7 (Li. dikes" in island.)
(Salvages Isl:iii«ls gite.
Madeira.
l'>stxitc. trarhytl<»lt'rite, alka-
line trachvte, etc.
AtOH'tt.
Plumolitc, trachyte, etc.
APPENDIX D
527
TABLli XXIII.— LIST OF DISTRICTS CHARACTERIZED BY ALKALINE ROCK-
TYPES: WITH NOTES ON THE NATURE OF COUNTRY ROCKS.— ConlmiiW
Region
Alkaline eniptiveB
Carbonate rocks cut by
alkaline eruptiTes
Canary Islands.
Phonolite, trachydolerite, bas-
anite, nephelinite, Umburgitei
tephrite, essexite, nephelite
syenite, monzonite, campto-
nite, bostonite, gaut^te, lim-
burgite, nordmarkite, pulas-
kite, akerite.
Cape Verde Is- | Foyaite, syenite, phonolite, leu-
la nds. citi te, tephrite, basani te, neph-
elinite, nephelite basalt, lim-
burgite, etc.
Julianehaab,
Greenland.
Foyaite, sodalite syenite, luja-
vrite, etc.
Columbretes (Med- Phonolite, trachyte,
iterranean).
Monte Femi, Sar- j Phonolite, trachydolerite, leu-
dinia. < cite basanite, leucite basalt.
? Mesoioic and Paleoimc IL
beneath volcanoesT li. on
Fuerteventura. (C o n t i n-
uation of Atlas Mountaina
li. formations?)
? Li. of Mayo, 8. Thiago, and
Fraya islands?
7 (Pre-Cambrian granite and
Paleoioic sandstone.)
Pantelleria.
Phonolite, trachydolerite.
Lipari Islands.
Etna, Sicily.
Leucite basanite, trachydole-
rite, etc.
Leucitophyre, trachydoleritio
basalt.
Tertiary li. and marls (older
U.?).
? (Tertiary and .Mesosoio
U.?).
? (Tertiary and Mesosoie
U.?).
Tertiary and Mesosoie IL
INDEX
Aa lava, 133, 291
Aar massif, 367, 503
Aasby diabase, 314
Abich, H., 144
Abitibi Lake, 487
Absaroka quadrangle, 491, 513
Range, 388, 475
Absarokite, 412
Abyssal injection, in Cordilleran region,
458
in eclectic theory, 305
relation to vulcanism, 248, 270,
279, 300
theory of, 174, 192
Abyssinia, rocks of, 426, 451, 508, 522
Acid sheU, earth's, 162 if, 170 if, 304,
360, 457
Acidification of basaltic magma, see As-
similation
Aconcagua, 502
Adamana, 508
Adamellite, origin of, 337
Adamello, 389
Adams, F. D., 237
on anorthosite, 322, 330-331, 333,
335, 337
on eruptive sequence, 469
on gabbro, 326-327
on Monteregian Hills, 396-397
on Mount Johnson, 228-229
on nephelite syenite, 419, 428-430
on strength of rocks, 172, 175,
179-180
Aden, 521
Adinole, 339, 431
Adirondack Mountains, 54, 59, 98
anorthosite of, 323, 327-328, 334
(map), 336
basic contact-rock in, 239
diorites in, 383
eruptive sequence in, 396, 469
laccoliths of, 330
syngenesis in, 239, 367
syntexis in, 407
Adventive craters, 142, 144, 150
iEob'an Islands, see Eolian
Africa, alkaline rocks of, 48, 508, 522
Ahaggar, 522
Ahvenaara, 505, 520
A!r, Africa, 508, 522
Ak^rite, 393, 397
Aladdin quadrangle, 191, 513
Alaska, bathohth of, 98, 116, 388, 391
quartz diorite in, 53
syenite clan in, 489
Alaskite, 341
Albany, New Hampshire, 500
Alberta, alkaline rocks of, 427, 488, 512
Albitic rocks, origin of, 243, 339-40,
437
Albitite, 437
Aldgate, Australia, 509
Alexander, W. D., 152
Alicudi Island, 482
Alkalies, affinity for silica, 400
concentration of, 4C0, 431 ff, 437
Alkaline clans, great diversity of types
in, 410, 415, 431-432
in geological time, 58, 60
in relation to qyenite dan, 414,
511
origin of, 410 ff, 414
extrusives, dominantly femic, 433
intrusives, doxninantly aalic, 433
magmas, viscosity of, 241, 439
provinces, 46, 415, 486, 511
provinces, eruptive sequences in,
443
rocks in dose aasociatioii with
subalkaline, 378, 410, lia-414,
421, 424, 426-428, 430, 451
in relation to sedimentary qrn*
teotics, 414 ff, 420 ff, 430 ff,
486,511
mineralogy of, 434 ff, 437
relative abimdanoe (k, 43,46-40,
52, 53, 413, 464
series of rooks, 18, 393
suite, use oi tenn, 410
Allan, J. A., 237, 441
529
530
INDEX
Allegheny Mountains, al>sence of fq^n-
ites in, 94
AUen, E. T., 320
H. 8, 204, 261,263
Allivalite, origin of, 324
Allochotite, 411
Alno Island, alkaline rocks of, 520
basic segregations of, 451
diaschistic dikes in, 59
map of, 419
primary calcite in rocks of, 434
syntexis in, 215, 420
Alnoite, 411
Alps, granites in, 94, 98, 190
tonalitic zone of, 190
Alsbachite, 341
Altar district, 501
Amagat, E. H., 260, 277
Amargosa Range, 494
Ammonia, volcanic, 272
AmHtcrdam Island, 142 (map)
Analcite, 437
diabaf<c, 411
syenite, 438
Analcitic rocks, origin of, 243, 434-
435
Analysis, composite, of sediments,
400-401
Anatexis, 309, 312
Anchi-eiitectics, 360
Andcndiorite, 374
Andcngranite, 341
Anderson, \V., 242
Andcsinfels, 313
Andcijiite, compared with diorite, 3S4
origin of, 223, 22S, 312, 375
relation to basalt, 377
Andesites, chiefly pyroclastic, 4r>4
Andesitic magma, viscosity of, 464
Andrews, E. C, S7, 115, 201>, 'MM\ 4R4
Angelbw, G., 481
Angermanlantl, unorthosite of, 323, 331
sills of, 232, 212, 354
Anortlioclasc. 436
tra<hyto. 391
Anorthositc, 313
abnormal fcaturt'.s of, 5s, 322 fT. 335
rhiefly pre-Cambrian, 5S, 60. 322.
335
coarM' grain of, 324
dilTen'iit iation of. 230 -211, 32 1, 325,
335
AnorthoHite, different iat km of, con-
traHte<l with that of uideate,
3-28
mapped, 240, 330, 332-334
mo<ie of intrusion of, 328 If, 333.
335, 337
not represented amooff the estru-
sives, 322
origin of. 241, 306, 312, 321 If, 324-
325, 330, 335, 337, 448, 463
rock-types syngenetie with, 336
Anorthositic magma, high Tiwoaty of,
322, 335, 337
Antagonism between granite and basal-
tic magmaa, 170, 361
Antarctica, alkaline rocks of, 415, 524
quarts diabaae of, 317
Antillean mountain system, 91 (map)
Apache Mountains, 408, 514
Apachite, 411
Ai>cnnines, overthnists in, 100
Aplite, 342
origin of , 361 , 368, 403, 426, 462
AiM>physe8, 83
belt of, 200
Appalachian Mountains, fissure erup-
tions in, 191
geosynclinal of, 186
igneous-rock areas of, 44, 01
(map), 98, 100, 388
Apulian district, 518
Ardrossan, 451
Areas of igneous rocks, 43 If
Arfve<lsonite granite, 440-441
Argentine, 516
Argillaceous rocks desilicate their sya-
tectics, 387, 305
Arg>'llshire, 502
Arisaig-Antigonish district, 486
Arizfma, granodioritic rocks of, 388
syenites of, 306, 407, 406
Arkansas, alkaline rocks of, 420, 408, 51 4
Arnoux fergusite stock, 53
Aroostook county, alkaline rocks of.
419, 500, 515
Arran, dikes of, 80, 371^72
quarts diabase of, 317
Arrhenius, 8., 158, 272
As)>ostos canyon, 406
Ascension Island, 423, 510, 526
Ascensive force of magma, 182, 180, 102,
290
INDEX
531
Aschistic dikes, 14, 17, 39, 58
Asciitney Mountain, 54
alkaline rocks of, 515
assimilation at, 306
composite stock of, 110, 113 (map)
diorite of, 383
eruptive sequence at, 166, 474
granite of, 367, 403
stoping at, 194
syenites of, 367, 396, 500
Asia, alkaline rocks of, 48, 506, 521
Asperite, 374
Assimilation, abyssal, 207 ff, 216-217,
305, 383, 461
conditions for, 210 ff, 214 ff, 217-
218,354,357,408,415,427
in feeders of fissure eruptions,
356
in feeders of sills and laccoliths, 219,
242, 318, 345, 348, 459
in intrusive sheets, 218, 435
in relation to stoping, 217
marginal, 196, 215, 217, 305-306,
336, 420, 435
of limestone, 436
of quartzite, 344 ff,348, 414
possible extent of, 214-215
precedes differentiation, 216, 245,
383, 407, 433, 462
Atatschite, 394
Atlantic branch (suite) of igneous rocks,
42, 54, 338, 412-413
region, eruptives not chiefly alka-
line, 413
Atlas Mountains, granites of, 96
Atmosphere, origin of, 246
Atrio, 145
Auckland Islands, 524
Augite andesite, 80
origin of, 375, 463
viscosity of, 380
diorites, origin of, 382
granophyre, origin of, 355
porphyrite, 313, 374
Augitite, 412, 444
Augusta county, Virginia, 515
Aureoles, contact, 103, 106-108,363-364,
416, 419, 431
Australia, alkaline rocks of, 437, 508-509,
523
quartz diabases of, 317
trachytes of, 397
Austria, alkaline rocks of, 504, 520
Auvergne, plug-dome of, 131
Avangares district, 516
Average analysis of primary basalt, 315
igneous rock, 168 ff, 308, 413
of Ck)rdillera, 457
Averages, chemical, value of, 16
Ayrshire, necks of, 128, 130, 298
Azof, 505, 520
Azores, 54, 145, 423, 510, 526
Backlund, H., 321
Backstrom, H., 226
Bad River laccolith, 71, 231, 329 (map),
347
Bailey, E. B., 85, 122, 196, 298, 416,
480
Baldwin, E. D., 290
Balkan peninsula, 504
Banakite, 412
Banatite, 374, 427
origin of, 337, 402
Bancroft district, 428-429 (map)
Bandai-San, 147, 285, 286 (map)
Banding, primary, 226, 324, 442
Bannock, Montana, 491
Bare Mountain, Nevada, 495
Barker Mountain, Montana, 73
Barkly East district, necks of, 130
Barlow, A. E., on anorthosite, 333
on assimilation, 209
on Chibougamau, 237, 327, 452
on gabbro, 326-327, 331
on nephelite syenite, 419, 428-
430
on Sudbury sheet, 69, 45$
Bamton, 451
Baron, R., 425
Barre, Vermont, granite of, 306
Barrell, J., on aplite, 369
on assimilation, 407
on Boulder batholith, 121
on Elkhom district, 408, 476
on injection of dikes, 80
on Marysville sill, 346
on stoping, 194, 205, 305, 461
on transition rocks, 399
Barrigan, 523
Barrois, C, 111, 209, 305-306
Barrow, G., 182
Bartoli, A., 198
Bams, C, 177, 198, 202, 213, 258
532
JXDEX
Basalt, average analysis of, 315
chenncally compared with gabbro,
315
dominant in fixsure erupt ions, 121,
458
origin of, 312, 315 ff, nee Substra-
tum
pillow, 3:j« flf
Basaltic magma, mc»st wiiiespread, 53,
164, 415, 458, 4t>4
I>er8istenre of in past eniptivity,
56, 458
primar>'. 315, 45S
substratum, sec Substratum
Basanite, 22S, 410, 412. 426, 432, 434
Basantoid, 412
Bascom, F., 87
Basic contact-phases, see Contact basi-
fication
Basswoo<i lake, 499
Bastin. E. S., 407. 474
Batholithic intrusion, dates of, 59. 198
Batholiths, 90, 103, 253
bottomless character of. 109
composite, 115. 116, 197
cross-cutting chara<>ter. 99, 100-
101, 107
development of, 1S8 flf, 19:^. 198 ff.
358.462
differentiation in, 243
«iownwtird enlargement of. 103 ff
elongation of, 94
feature.** of, 100
homogeneity of. 89, 217, 240. 243
length of, comparison with dikes.
location of. 91. 94, KN)
inodi6e<i dikes. 24<), 305, 35S
multiple, 115
of diorite unknown, 382
progressive unroofing, 108
rarity of basic, 113, 308
relation to t»rogeny, 92, 94. 96. 9S,
190. 205, 40*0*
repla(*ement by, 109. 112
roofs of. 103-ioS. 205
Haylcy. W. S.. 236, 242. 326. .336.
346 347, 44S
H:iy«inncbathoIith. 107 (map), 1 1 1, 361-
362
IWaumimt, K. de, 305-306
Bear Lodge Mountains, 72
Bearpaw MountAini, 406» 401, 512
Beaver Creek Uccolith, 408
lake, 494
Becke, F., 38, 330, 413
Becker, E., 436
Becker, G. F., 222, 223, 272, 376
Beemerville, 616
Bekinkina MountAint, bekinkiiiile of.
53
Bekinkinite, 410, 444
total are* of, 60
Belknap Mountains, 367, 474, 600, 515
Bell, J. M., 146
Beloeil Mountain, 611
Belted Range, 404
Benbeoch nil, 234, 243, 438
Benedicks, C, 321
Benson, W. N., 437
Bergeat, A., 378, 422, 482
Bergen district, anorthoaite of, 233,
241, 322-323, 331-332 (oMp)
granite in, 367
iron ores of, 454
syenite clan in, 337, 367, 606
Berkeley Hills, volcanic sequeace at.
372; 478
Bi<lwell Bar quadrangle, bfttbolitfas of.
102 (map), 366
syenite clan in, 406
Big Cottonwood canyon, 403
Trees quadrangle, 406
Biella, 503
Bimineralic magma, 447
Binangonan peninsula, 610
Bingham district, 403
liisbee quadrangle, 406
Bischof, G., 201
Bitterroot Range batholith, 53
Black Butte, Montana, 73
Black Butte, Nevada, 405
Black Buttes, Wyoming, 73
Hills, alkaline rocks oT, 613
laccoliths of, siibjaeeni bodies in.
08,388
Mountain, New Meodeo, 407
Blagmiat, 507
Blairmore, Alberta, 427, 488, 512
Blake, J. F., 75
Blanketing in relatioo to magmatir
temperatures, 211, 310
Blansko, 604
Blckinge district, diabaM fai, 366
INDEX
533
Block lava, 133, 291
Blow-holes, 144, 292
Blue Hills, Massachusetts, 205
Ridge, 498
Bogoslof, plug-dome of, 131-132
Bohemia, alkaline rocks of, 519
assimilation in, 415
eruptive sequence in, 444, 481
laccolith in, 438
petrographic province of, 54,. 421
rock associations in, 42 i
syenite clan in, 504
Bonaparte Lake, 488
Boninite, 412
Bonner's Ferry, sills near, 231, 242
Borlo River, 507
Borolan, see Loch
Borolanite, 411
Boss, intrusive, 90
Bostonite, 411
Boulder batholith, 121, 408, 461-462
Boule, M., 137, 254, 425, 433, 480
Boulton, W. S., 452
Bourdariat, A., 140
Bowen, N. L., 236-237, 242, 325, 339
Bowral, N. S. W., 437, 509 -
Bozeman, 490
Bradshaw Mountains, 383, 496
Branco (Branca), W., 144, 283-285,
295
Branner, J. C, 99
Brauns, R., 400-401
Brazil, 501, 516
alkaline rocks of, 419, 501, 516
Breached cones, 139
Breccia, stoping, 462
Breckenridge district, 402-404, 418, 492
Brefven dike, 78 (map), 81, 356
Bresson, A., 112
Brewster county, Texas, alkaline rocks
of, 498, 514
New York, 500
Bridgman, P. W., 172, 176, 180
Brigham, W. T., 135, 153
British Columbia, alkaline rocks of,
488,512
batholiths of, 98, 307, 388, 391
eruptive sequences in, 396, 470,
476-477
fissure eruptions in, 191, 458
geosynclinals of, 186-187
quarts diabase in, 317
British Columbia, sills in, (See Purcell,
Columbia.) ; syenite clan in, 488
Guiana, quartz diabase of, 317, 355
sills of, 232, 336
Isles, see Great Britain
Brittany, subjacent bodies in, 93 (map),
95, 98, 111 (map)
Brock, R. W., 83, 209, 306
Brogger, W. C, on assimilation, 209,
216,305
on Christiania Region, 116, 483
on differentiation, 306
on dikes, 39
on Gran, 452
on Predazzo, 367
Bronzite gabbro,.313
Bronzitite, 446
Brookville, New Jersey, 515
Brouwer, 8., 350-351, 427. 455
Brown, C. W., 474
Brunn, 504
Buch, L. von, 145
Buchanan, J. Y., 225
Buchonit-e, 412
Bucking, H., 77
Bukowina, 504
Bullfrog, Nevada, 514
Bulza Mountains, 504
Bunsen, R., 52, 165, 308
Burckhardt, C.^ 88
Burnett Creek laccolith, 71
Burro Mountains, 497
Bum Island, 525
Bushveldt, alkaline rocks of, 427, 508,
522
assimilation in, 242
differentiation in, 232, 240, 350-
351 (map)
granite of, 53, 76, 350
laccolith of, 68, 70, 75, 114, 232
ores of, 454-455
pyroxenite of, 448
Bysmaliths, 84
Cabezon volcanic neck, 127
Cabo Frio, 502, 526
Cactus Mine, 493
Range, 494
Cafemic (lime-iron-magnesium) oom-
ponenta, 384, 432, 436 ff
Cah6tie Mountains, 521
Cairnes, D. D., 389
534
rS'DEX
Cal'-i*- l»r.in*'li of iKn^^mi?* n>f^k<-. Ii;$ 414
(':ih'H4'. primary. 4:il. A'M't
C'al'la-. •'»l*i
f ':iM»*ira 'Li* f^'U- ("u\:%h*-. 1 1.'» 1 U\
iiiap
«!«• San'a M:irJ ar:i. I t.'i
f 'al'^'ira* 'la- Furria^. 1 1."»
r:iM«.ra of La I'aliiia. 1 14 14.'.. 4'»1
Talhra-. 14*>. 141. l.VK l.VJ. 2Hii
rrit*Tion for. 147-14**
m-.-T#**i. 147. \r^)
s<irik«-n. 14!l I.V)
C'alifnriiia. alkaline* nx-ks in. 'A t
hatholifh in, :ii»7. :{sn
<li>trirf, (*olMra'l*>. V.r2
quartz <Iiaba.-«' of. 'il7
.*y<'nite i-Ian in, 4^mJ
Talking. V (' . 2'M\. 212. Hi:i. '^'^f^, 477
f "alviiiia. *ill- rif. t»*i
(''aniparii:in •ii-tri''!. .'ils
Canipl»ell. K.. IS5
Man.l. o.M
(^^aniptnpito. 41 1
CaiiiM-ll. r., :uH). :{«is :{♦,*.♦
('ana la. annrrh-i-iN- of fa*t«'rn. 'V.V.\
I map . '.V\7
hatholiTh* nf f-a^f.-rn. *»!. f»^. :{07
Caiianca. .VM
('a nary I-lainN. alkahnr* r»<k- in. r»in.
.VJ7
li<»rnitn> in. !!.'>
pM'k a-.-'Hiation in. .M. IJi. l.M
('anrrinit*'. i;M. M7
.•*yi*nit«'. ni
( 'anolMila*^ Mn!intain<i. alkaline rocks of.
:»n^. :>2\
(•niptiv«* MMjit'iH'o in. 1**1
(/antal vnli-ano. alkalin<> mrks of. .'»17
tTii|)iivr .MMjiit-ntf in. \\\. 4*^0
lone life of, 2'i\ rnap
ri'lation to rrn>*. frai't'in*-*. l.■4^
unap
pM'k a--«H'iation at. IJl \2't
Cantnll. T. ('. 472
('ai>e Provnu'c ('apf <'«il.iny , uranst*'-
of. '.»s
-ilU of. lit'., .;l^
vi'-iiMiliir ^ilU Ui. ■_••"•'»
voli-anii- ni'.-k- of. l.;o. •_»'.»!
Vrr.li' l>lan.U. IJ:;. i:.l. ."ilo. .->27
poninMila. .'lOs
CrtiHjrtee valley, iiJ^i
CarUm liioxiile. effect no maitin&s
431 ff
r»rifrin of. 270. 2S5
< 'arlxiiiatOii. influenre of in tyntcrtic*.
414 ff. 421 ff. 410 ff. 436. 4oO
r*arVK»nir ariil ami nilirir arid ecNnpAred
in strength, 414
in Cripple Creek minai, 418
Carmeliiite, 374
(^aniline IsUnda, 525
Carpathian MounUuns. M, 190
eruptive sequence in, 481
Carrock Fell intrusive. 358
CVsoade Mountains, balholitlis of. 9S.
366
fceosynclinal in. 187, 459
Ca.stle Craifts sill. 234. 243, 438
I>>mc. 497
fcranite. 364. 366
Mountains. Montana, alkaline rocks
in. 513
stfMrks of. 364 (map)
Peak ^itock. 99 fmapi. 109. 110. Ill
( 'atalonia. 518
Ca*apl(*iite syenite. 411
Cathciral icranite. 366, 428
Cau('a.<-u.t. alkaline rocks in, 505, 520
(n'anite.*< in, 94, 98
Caiil'lron-Kuhflidenre. 122, 196
Canst ir replacement. 243, 353, 362-363
( *a: It «'rt:ts granite, 112 (map)
( 'aves alH»ut lava lakes, 255-256, 259-
2tiO
< '**ar:i, Brazil, syenite of, 53, 502
( Velar liange, 493
(VIcIhm, 525
( Vnter-(Miintrt, chemical, 16
(Vntral eruptions, 117, 124, are Vents
(icpreMi(ion forms connected vith.
125
rock l>odies associated with, 125
granite.<«, 53, 90
(Vrrillo.-^ Hills, 497
CVrro Halmaceda, 502
(ic Mulcnw, 501
(Vvadaos, 518
(Vzimha. '}18
Chamhorlin, R. T., 269-270
T. C. Iho ff, 158-159, 175
( *liarn(»okite, 341
Charriaffe in relation to Vethi?lith*. 91.
94, 190. 206
INDEX
535
Chemical equilibrium and magmatic
temperature, 157, 177
Cheviot district, granite in, 365-366
(map)
Chibinite, 411
Chibougamau, anorthosit^ of, 323
differentiation at, 233, 240
granodiorite of, 388
iron ore of, 455
laccolith of, 241, 326, 333
pjrroxenite of, 448, 452
Chichagof Cove, 489
Chilling, contact, checks differentia-
tion, 237
explains basic phases, 240, 245-246,
327, 363, 366, 383, 390, 441
prevents assimilation, 242, 358
China, alkaline rocks in, 522
Chonoliths, 84, 86, 87, 442
Christiania Region, alkaline rocks of,
430, 521
bathohth of, 98, 116
eruptive sequence in, 396, 444, 483
granite of, 367, 439
petrographic province, 54, 439
syenite clan in, 506
Christina Lake, 488
Chromite ores, 455
Cicatrix, batholithic, 122
Cimarroncito district, 497
Ciminian district, 518
Ciminite, 374, 394
Cir Mohr dike, 371-372 (map)
City Creek, 493
Clans, alkaline, 41
igneous-rock, 40, 311-312
more important, 40
Clapp, C. H., 389, 399, 454, 475
Clarke, F. W., 209, 221, 437
on average rock, 17-18, 168, 457
on composite analyses, 399
on sedimentary shell, 162
on sodium in the ocean, 164
Classification, genetic, of igneous rocks,
311
Clements, J. M., 320, 361-362
Cleveland dike, 81
upward termination of, 182
Cliften quadrangle, intrusions of, 364
volcanic sequence in, 372, 479
Clough, C. T., 79, 85, 122, 196, 415,
47&-480
Cnoc-na-Sroine laccolith, 439, see Loch
Borolan
Coast Range, batholiths of, 98, 116,
383, 388, 461
Cobalt Lake sills, 230
Cobb, J. W., 415
Cochiti district, 498
Coeur D'Alene district, 490
Coimbatore district, 507, 521
Coleman, A. P., on assimilation, 209,
242, 306
on stoping, 348
on Sudbury sheet, 69, 236, 348-349,
455
Colfax quadrangle, 366
Collins, W. H., 236, 339
Colloids, stratification of, 229
Colombia, syenite clan in, 501
Colonsay and Oronsay, 415, 502
Colorado, alkaline rocks of, 513
plutenic rocks of, 98, 388, 418
syenite clan in, 491
Columbia, New Hampshire, 500
Mountains, silb of, 66
vesicular sill in, 265
Columbretes, 510, 527
Comendite, 342
origin of, 370
Complementary magmas, 241, 373,
444, see Diaschistic
Compression as a source of mag-
matic heat, 156-157, 276-
277
orogenic, 180
shell of, 176 ff, 183, 185, 192
Comt6, le, 432
Concordant injections, 63, 242
Conductivity, thermal, 177, 198 ff,
256
Conduit, volcanic. Bee Vent
Cone chains, 140, 250, 209, 300
clusters, 140, 292, 300
Cones, breached, 137, 139
volcanic, 135
Congo Free State, quartz diabase of,
317
Conical Peak stock, 363 (map), 365
Connate waters, 249
Connecticut, keratophyre in, 501
Consanguinity, chemical, 215, 219,
243-244, 306, 318, 345, 357, 363,
462
530
INDEX
Contact basification explained, 237,
239, 240, 244-247, 327, 363,
306.383.390,441,463
metamorphiflm, see Metamorphic
mining district, 495
Contraction of earth, 174 ff, 192
thermal, 177
Convection, failure of, 211
fniulient, 224
in magmas, 158, 177, 258
in relation to differentiation, 223 ff,
238
strength of, 224
through gas-concentration. 266
two-phase, 256, 259 ff , 264 ff, 273
Cook's Peak, 497
Cooling by decrease of pressure, 211.
267-268
Coppaelite. 412
Copper Mountain, 489
ores. 455
River, 489
Cordillera, basaltic substratum of, 45H
earth-shells in, 457
eruptive sequences in. 461
igneous-rock areas of, 13, 43, 91
(map). in>. 95, 116, 190, 385,
387.456
vulcanism in, 458-464
Coniilleran and world rocks compared,
456
Comdon. Shropshire, 77
Cornwall, granites of, 96 (map), 9S
Corrosion, magmatic, 245.35^^,362-36.3
Corsica, dike system of, 83 (map)
fissure eruptions of, 120
Corstorphine, O. S., 188, 236. 350
Cortlandt district, 500, 515
Cortlandite, 446
Corundum in alkaline rocks, 434-4ii5,
437
Cor>-ell batholith, 53
Cosmogonies, 155, 159
Cossa, A.. 201
Costa Rica, alkaline rocks of. 516
Costc, E., 200
Cotta, B. von, 52. 165, 209. 30.'). 311
Covers, batholithic. 100 fT, 121
Cowal. multiple dike in. 79
Cracks, closing of. in earth's crust,
178 ff
Craig, E. H. C, 415
Cranbrook diftrici, ailk of, 331
Crandall quadrangle, 3M, 476, 491
Crater Lake, Orefon, 150
Craters, 140
criterion for, 147-148
nested, 143-146
smaU sue of,141
Crawford, R. D., 80
Craxy Mountaiiis, alkalina roeks of, 513
laccolitba in, 53, 70
stocks in, 363 (map), 305, 385
syenite clan in, 400
(ripple Creek, 398, 404, 417-118,401.
513
Critical state, rock-matter in, 101
Cross, W., 10, 75, 209, 475
Crust, earth's, 170-171, 314, 808-309
flotation of, 177, 192, 190
remnants of primithpo, 809, 342
Cryptovolcanic dome, 283, 285
( O'stal Falls district, diffcrcnlkted
dike in, 302
CrysUUisation, fractional, 331 ff, 238
interval, 370
of batholiths not eontiDiMNM» 882
order of, 375
temperature of, 214, 375-^76
Oystals in Kilauean lavn lnk% 370
liquid, 225
mixed, 9
plastic, 225
Cuddapah trape, 350, 383
Cuillin Hills, Skye, 71, 72, 324
Culebra Bay, 510
Cumulo-volcanoes, 131
( uiiolas, batholithic, 102, 105, 258
( ushing, H. P., Ill, 240, 880, 357.
469
Cutch, Runn of, laeeolitlii of, 75
quarts diabase of, 317
(^ittingsville, 500
Cyanogen, energy potentinliaed in, 272
Cycles, petrogenic, 50, 110, 890, 308
Cyprus, 510
Dacite, 80, 386, 403
origin of, 243, 389-301
Dahamite, 411
Dalmer, K., 100
Dana, J. D., 131, 153,280-281,390-291
lake in H
Darton, N. H., 74
INDEX
537
Darwin, C, 222
district, 496
G. H., 176-177, 304
Daubrde, A., 160, 251-252
Davis, H. N., 277
Davison, C, 176-178
Davy, H., 269
Dawson, G. M., 187
Deccan traps, 118 (map), 119, 191, 310
Dechen, H. von, 481
Deformation, crustal, associated with
valcanism, 292 ff, 295, 300
Delesse, A., 201
Demavend, 507
Deming, New Mexico, 497
Densities, rock, 201 ff
Density, earth's internal, 161
stratification according to, 161,
167, 170, 304
Denver Basin, 492, 513
Dependent Isles of Taiwan, 526
Deprat, J., 83
Depression forms, volcanic, 125, 140
De-roofing eruptions, 117, 121
Desilication, 387, 395, 399, 404, 420,
431, 434-435
Develikoi, 507
Devon, granites of, 98
Dewey craters, 250
Dewey, H., 338-339
Diabase, 314
quartz, 316 ff
Diallagite, 446
Diaschistio dikes, 15, 39, 58
origin of, 82, 444-445
Diatremes, 251-252, 294, 300
Differentiates, gas and vapor, 246
Differentiation, affected by two-phase
convection, 377
a reversible process, 210
at central vents, 223, 227 ff, 287 ff,
288
follows assimilation, 216, 245, 383,
433, 462
in abyssal wedges, 208
in place, 76, 229, 324, 335, 391, 406,
437, 451, 455, 462
in sills and laccoliths, 76, 229, 324,
437, 462
intermittent in batholiths, 362
magmatic, 221, 287, 305, 378
primitive, 170, 360
Differentiation, progresses according to
size of magma chamber 396-397,
432, 463
specially manifest in alkaline rocks,
431
temperature of, 229, 288
units of, 222, 447
gravitative, 227 ff, 359, 462
in abyssal wedges, 190
in batholiths, 243-244
in dikes, 246
in gabbroid magma, 241, 316,
325 ff
in laccoliths, 325 ff, 350 ff, 439 ff,
448 ff, 452, 454
in sheets, 238 ff, 344 ff, 347 ff,
407, 448 ff, 452, 454
in the earth, 161
in volcanic vents, 228, 316, 370,
378, 409, 443, 450
Diffusion, molecular, 222, 265, 273
Diffusivity, thermal, 198 ff, 256
Dike injection, rapidity of, 120
system, 82, 83
Dike-rocks in geological time, 58, 60
Dikes, 78
aschistic, 14, 17, 39, 58
composite, 79, 80
diaschistic, 15, 39, 58, 82, 444-445
differentiated, 77, 78, 226, 245
great lengths of, 81, 460
multiple, 78, 79
widths of, 81 , 120
DUler, J. S., 113, 150, 237, 391
Diorite, 82, 113, 368
aplite, 374
average, 169, 382, 413
batholiths unknown, 382
in chilled phases of batholiths, 105,
196, 245, 363-364
origin of, 287, 312, 353, 356, 361,
381-382, 384
peripheral to granite, 105, 196, 363
clan, 374
includes both differentiatefl and
syntectica, 375 ff, 384
in geological time, 68
family, variability olF, 384
Discordant injections, 63
Displacement, magmatic, 62, 64, 69, 84,
88, 109
Ditro, 520
538
IXDEX
Dittler. K.. 22',
DobroRpa. 5()4
DrxiKO, F. S.. l.'il
I)r»eltcr. C. 2CH». 214, :{7'j
Dohlen distrirt, .'>0l
Domes, lava. oniloKPiioiw, 130 ff
exogenous, 13.>
D<H»rnl)erjr. .V>H
Dr»rmanry of voIr*an(»es, 275. 27y-2M)
DoiiKlas. J. A., 201, '202
Downicville qiiarlrunKle, 49* i
Downtown district. 403
Down warping due to ahyssal injection,
1R5. 193, 4.yj
Dragoon Mountains, 407
Drakenslierg. dike in, *^1
Dresser. J. .\.. 403
Drililet ctmes. 135-13r). 202
Dripping Springs, 407
Duhho. 523
Diirlaux. J., 227
Duluth laccolith, tW, 113, 325 map;
anorthosite of. 230. 211. 324. 32h.
335
as.<iiiiil:ttioii in. 212. :>3ti
difTcrcntiution of. 210
dinien>ioii>< of, .'»;». 75
gahhni «.f. .324
granite of. 230. 33ti. 34ii. 370
pcritlotitc clan in. 230. 32«i, 44n,
454
relation t«i Had Kivcr laf^nlith.
32t»
svenitc t-hiu in. 33t». 3 Mi. 407
Dunuroyn urrk. 12s .map-. 12'.»
Dunedin .livtrirt. 421 121 niap'. 50l»,
524
Dun^aniioii town'^liip. 511
Dunite, '\U\ tT. 452
relation to an<»rtho>itf, 117 1 is
Duparc. L.. 44S
Duppau. alkaline rocks «»f. 451, 501. 510
Duran^i) (piadrangle. 402
I>iirl»a«ljite, 303
Durorhrr. .1.. 52, 170
Du Toit. A. L.. on ihkc-. S!» si
on intru<ivr >hcets. »i7. 111. 23ii.
212. 2ri5. 31t). 310 350. \\s
• •n Mudilrr roiitcin volcaim. 153
on vulcaiiir pipe. 201
Dutton, ('. K.. 144. i:>0, 153, 2.''>0
Dyke, 9€e Dike
Eagle Mountains, California, 454, 496
Earth, degree of heterogeneity, 457
dennity Ptratifiration of, 160 ff. 171
Hhelli^. 159. 167. 170. 304, 360.
4.')7
Karth'rt enifft, Htability of, 172, 205
Eai«t Africa, alkaline rocks of, 522
voleanneA of, 310. 419, 426
Mancofl Distrirt, 492
Moreton. 485. 524
F>lectie theor>' of igneoiu rodu. 304-
305, 307, 343, 360, 456
K«lwardii Creek. 488
KfTuj«ive rocks, are Extniaire
Egypt, anorthoHite in, 323, 334
Eifel. 5a5, 519
Eigg, Itfle of. map, IS5
Eildon Hills, 517
Ekerite, 341
Ekersund-iSoggendal, anortbonte of.
323. 327, :m, 337, 367, 396
eniptive sequence. 483
Ekholm. N.. 272
Ekstromsberg. 506
Elba, granite of, 190
EMgjd lava flow. 120
Electric Peak, sill at. 233, 450
lavas of, 2HS
El<H»lite syenite. 410-411
Elfdalen.V)21
Elizaliethtown. 499
Elkheaii Mountains. 492, 513
Elk horn difitrict, alkaline rocka of, 513
aplite of. :i«)9
aviimilation in, 407
difTorcntiaticm in, 408
eruptive sequence in, 476
svenitc clan in. 490
transit i(»n rocks of, 399
EUensburg qua«lrangle. lavas of. 37V
(map)
Elliott district, necks of. 130
EIli|>s<>idal basalt, 338 ff
Elnienthal, Thuringia, 78
El Taso quadrangle, 498
Elsden. J. v.. 222
lOincrson. H. K., 79. 97, 306, 339. 3K9
Eniplat-enient of magmas, 62 ff, 109 ff
Emulsion Mage of a magma, 226
Enclos of Reunion, 150
Endogenous growth of lava hums, 131
Endothennic compouiuli, 273
INDEX
539
Engineer Mountain quadrangle, 492,
513
England, batholiths in, 95, 98
Enlargement, downward, 103 ff
Enstatite basalt, 314
diabase, origin of, 319
porphyrite, 374
Eolian (Lipari) Islands, alkaline rocks
of, 527
breached cone in, 137
differentiation at, 288
eruptive sequence in, 372, 444, 482
Epidiorite, 314
Equilibrium, chemical, in basaltic
magma, 415 ff, 430
in relation to temperature, 157,
177
of batholithic phases, changes
in conditions of, 362, ,401, 415,
430
of magmatic gases, 249, 260,
269 ff, 271, 274
Erhebungskrater, 294
Eruption, ascensive force in, 182, 192
in geological time, 59, 60
Eruptions, fundamental cause of, 173
Eruptive sequences, 55, 244, 311, 362,
390, 396-397, 402, 443-444, 461
Erzgebirge, 504
Essex county, alkaline rocks of, 500, 515
eruptive sequence in, 475
Essexite, 228, 411, 429, 440, 444, 450
and the telluric magma, 413
aplite, 411
differentiation of, 396, 403, 438,
452
-dolerite, 438
family, 410
total area of, 50
Esterellite, 374
Ethmolith, 87
Etna, alkaline rocks of, 423, 527
assimilation at, 436
craters of, 143
flows of, 290, 292
lateral vents of, 250
periodicity of, 280
superheat of lava, 414
Eudialyte-nephelite basalt, 412
syenite, 411
Euganean Hills, 503
Eugene Mountains, 495
e
Euktolite, 412
Eureka district, eruptive sequence in,
479
Europe, alkaline provinces in, 517
syenite clan in, 502
Eutectics, 225, 368
Evergreen Mine, 492
Evisceration of volcanic cones, 279
Exogenous growth of lava mass, 136
Expansion, cooling in, 211, 267-268
of magma during injection, 182, 248
Experiments in mineral ssmthesis, 437
Explosions, magmatic and pbreatic,
282
volcanic, 277, 282
Explosiveness of central vents, changes
in, 288 ff, 303
Expulsion of residual magma, 246, 445
Extension, compressive, 178, 189, 192
Extrusive bodies, classification of, 117
rocks, chemically compared with
plutonics, 19 ff, 39, 229, 315,
370, 384, 443
dominantly basic, 50, 53, 57, 464
mapped areas of, 45
special variability of, 287, 292,
372,444
Fair Haven, 501
Falkenberg, syntexis near, 355
Faraday township, 487, 511
Faribault, E. R., 327, 452
Farrington, O. C, 160, 167
Farrisite, 411
Faroe Islands, fissure eruptioiu of, 118
neck in, 126
Fassifem, 524
Feeders of sills and laccolithB, 219, 242,
318, 345, 348, 459
Feldspathization, 339, 431
Felsite, 342, 371
Fennema, R., 138, 144, 159, 431, 426,
483
Fenner, C. N., 339
Fennoscandia, batholiths of, 02, 08, 406
quarts diabaaes of, 814, 010
Fergusite, 410
total area of, 50
Fermor, L. L., 367
Fernando Noronha, 526
Ferreria San Estebfto, 501
Fichtelgebirio, batfaoUth of, 106» 300
»0
ISDEX
Fife»hire. explosion fiiwure in, 127
neclu of. 130. 2.52. 297 ff mAp;
Fiji I.-Unds. o2o
Filiriidi Inland. 4S2
Finiantl, batholith.- of. W. 9>
eruptive -equence in. 470
Fi-her. O.. 177- 17S. IW, 3CM-:ja5
Fi.«-iire eruption, a 9U<Men act. 120
erupt ioni?. 117
dat*-!! of, 59
fw^iers of. S2. :*.%
in relation to orogeny. 191
thirkne?tff accumulated, 119
'•ruptivr-ji. romfKj«ition of. 120. Itio.
171. :tsO
Fuseure!*, ahvival. 171 ff. 192
voUamx-?* locate*! on. 250, 2>1
Fitou, France. 517
Flaiienlava. V-^-i
Hanng of crater. 127 ff, 140. 141. '255
Flatheail Kivcr. nil at. 2:n. 242
Fletl. J. .S . :«»y-3:j9. 415
Flow of ri-K-k>, 179
Flow.', lava. cla.-eiticatifin of. \'M
thickne^-of. IIH. \M\
voltimi? of. 1-20. 1>2, 27*«. 2>1.
2^9
Fluidity, maiunatic. 213. 241. 2**1.
37G
Fluxe... 212-213. 21tV-217. 430, 4.-iO
FVintainc du Genie. 50>
Fonte da Bica. 51S
Fort lienton quadranfcle. 4n5
Fort una. 50:5. 51 >
Fort unite. 412
Founilcring of r»:»of. 122. 206
F>>untain Ilea*! HilU. 4<.«3
Fountain.-*, lava. 250. 2ri«i ff. 20r>
Fouqiic. F.. 14**. 2<K*. 2i:{. :i05
Fourchite. 411
Fox Islaniis. granite nf. y\
River. Wi:icon.*in. 122
Foya district. 417
Foyaite. 411. 414, 4U5. 41>. 425. 427.
434
origin itf. 439
Fraas, K . 2<5-2S:>
Framhnini hiva flow. 12i»
Fntm-e. alkaline n>cks of. 4>. 517
>yenite clan in. 397. 54 W
volc:im»esi of. 13> map . J54. 310,
424
Frmnklin Canp, 488
Fumftce, 515
MounuiiM. T(
F'reeiing-in of
326. U5, 307,
Friedlaender. I.. 134
Frifleo distnct, 4M
FujiyAiDA Tokaao, 275
FiunAce, Toleanic, 260,
302
FuaibiUty, jee Melting
Fuaagranit,
241.
ff, 274, T.^,
Gabbro, abnomuJ, 310
chemicmlly cootrMtad with bwah,
315
clan, esotic, 174
olivine-free flpeciea of, 316
oriffia of, 312 ff
rerurrence of types bdoogiiig bi.
56. 165
Gabbrofl. banded, 226
Gagel, C, 144*145, 451
Gallo. G., 432
Garlton HUk 517
Ciamet in alkAline rocka, 436-437
porphyrite, 374
C;a0. a flusing agent, 213. 218, 229. 241.
273, 274-276
bubbles, Tery riov mm of, in
depth, 2M ff
concentration of, 213, 229, 241.
249, 254, 271, 274, 279, 32S.
358,362
cooling by rieing, 267 ff
expelled during eryvlnlliintion, 273.
278
tension, an aid to injeetaoB, 346
(.taf<eou9 transfer, diffeicataatioa by.
247, 368, 409, 432, 445, 4£{.
455
in batboUths, 244, 361
in (likes, 373, 454
in sills, 320, 339
in volcanic venta, 400
of alkalies, 400-1, 406, 482
(ias-fluxing. 251-252, 280-282, 303.
310
Ga^es. classification of voleaaie, 249
in magma diambcr, coodition of .
273
in rocks, 269
INDEX
541
Gas-well, cooling by expansion of gas
in, 268
Gaussberg, 415, 451
Gauteite, 411
Gavelin, A., 355, 383
Geijer, P., 367, 397, 402, 452, 455, 470
Geikie, A., on assimilation, 355
on dikes, 7^81, 182
on eruptive sequence, 472
on fissure eruptions, 119
on necks, 126 ff, 252, 297 ff, 300
on sUb, 64, 237, 324
on stratification of the earth, 304
on veins, 83
Gelliv&re, 506
Georgetown, Colorado, 404, 418, 492,
513
Geosynclines, caused by magmatic in-
jection, 185 flf, 193, 459
in relation to batholiths, 94, 310
in relation to volcanic action, 185 ff,
310
Germany, alkaline rocks of, 433, 518
batholiths of, 98
eruptive sequences in, 473, 481
syenite clan in, 397, 504
Geyer-Ehrenfriedersdorf Sektion, stocks
of, 106 (map)
GhAts, fissure eruptions of the, 119
Gilbert, G. K., 69, 70, 75, 84, 86, 129
Gilpin county,. Colorado, monzonite in,
407
Giorgis, G., 432
Glamorgan township, 487, 511
anorthosite in, 323
differentiation in, 327, 333
laccolith in, 233, 241, 326 (map)
peridotite clan in, 448, 452, 454
Glangeaud, P., 432, 481
Glass House Mountains, 523
Gleichenberg, 504
Glen Coe, Scotland, intrusion at, 85,
122, 196-197 (map), 389, 480
Glenn Creek, Alaska, 489
Globe quadrangle, eruptive sequence
in, 472
district, intrusions, 354, 383, 396,
407
sUls of, 233
syenite clan in, 496
Goldbelt district, 496
Golden HiU, 452, 517
Goldfield district, volcanic sequence
in, 372, 479
Hills, 494
Goodchild, J. G., 194, 305
Gordon, C. H., 471
Mrs. Ogilvie, 85
W. C, 186
Gowganda Lake sills, 230, 242
GdzenbrQhl, neck at, 296
Gradient, thermal, 174, 198, 268, 296
Grampian Hills stock, 245
Gran, Norway, 451-452, 521
Grand Canyon, granite of, 98
Puy of Sarcoui, 131
Granite, a mountain rock, 164
chiefly pre-Cambrian, 57, 60
clan, 341 ff
definition of, 341
derived from syntectics, 344 ff,
355 ff
differentiated from andesitic
magma, 365
from dioritic magma, 362-363
from granodioritic magma, 361,
366
from syenitic magma, 367
dominance of, 168-169
family, restricted to continental
plateaus, 42
origin of, 243, 287, 312, 342 ff, 359,
360 ff
problem, 360
Granites, abnormal, 359
alkaline, 368
Granitic differentiates, 360 ff
rocks absent in ocean basins, 162
Granodiorite, 121, 463
aplite from, 368
associated with alkaline rocks, 391
chiefly post-Cambrian, 68, 60, 387
clan, 385
not a basified granite, 362, 866,
389
origin of, 243, 312, 361, 387 ff
relation to mean of granite and
diorite, 385
relation to quarti monionite, 888,
388
replacement by, 99, 106, 111,201
shattering by, 201
specially developed in America,
13,386
542
INDEX
Granodiorite, syiifcenetic with alkaline
rocks, 115,428
Granodioritcs, forming a diistinct family,
3,385
Granophyre, 67, 72, 79, 341-342, 350,
355
origin of, 355-354)
Grant, U. S., 328
Graphite in nopholitc Hyenite, 437
Graton, L. C, 39S, 471
Gravitativo differentiation, we Dif-
ferentiation
Grayback, Colorado, 492
Great Basin, fissure eruptions of, 101,
413
Britain, alkaline rocks in, 517
eruptive sequences in, 472
fissure eruptions in, 191
geosynctinals of, 187
quartz diabases of, 354
syenite clan in, 502
Rift of Africa, 54, 120, 191, 250,
508, 522
Greenland, alkaline rocks in, 510, 527
"batholiths" of, 71,417-418
composite sill in, GO, 447
fissure eruptions of, 118, 119, 310
iron basalt of, 321
quartz diabase of, 317
Green, W. L., 106
Gregory, H. E., 419
J. W., 120, 250
Grenville, Quebec, syenite of, 4(X2, 486
Grewingk Island, 132
Griqualand, intrusive sheet in. 316, 349,
448
Grorudite, 411
Grossprieejen laccolith, 438
Guiana, 502
Gunflint dis^trict, 488
Gunn, W., 79
Gwillini, J. C, 327, 452
IIaar<it Mountains, 518
Ilackman, V., 114, 237, 416
Hague, A., 18
Ilaleakala volcano, I3t), L>)
Ilalemauniau crater, 255 fT imap),
2«W) ff, 2S0, X**- also Kilaura
llaliburto.i-llastinfcs region, 331, 419.
429. 431, 434, 436
Il&lleflinta, 124
Hallock, W., 179
Hamilton, W., 143
Harcourt township, 487, 511
Harden, Australia, 524
Harder, £. C., 454
Harker, A., on aaBimilation, 209
on Atlantic and Pacific brmncbat,
338-339, 412-413
on average igneous rock, 166, 457
on banded gabbro, 324
on Carrock Fell, 356
on differentiation, 221-222, 226,
246, 368, 373
on dikes, 79
on eruptive sequence, 479
on hybrid rocks, 383
on laccoliths, 71-73
on migration of magmas, 416
on Norm classification, 10
on phacoliths, 76-78
on sills, 65-66
on stocks, 100
Harrison, J. B., 236, 355
Hartung, G., 145-146, 150
Harz Mountains, 505
Harzburg, 447
Harzburgite, 446-447
relation to gabbro, 447
Hatch, F. H., 188, 236, 350
Hauran, fissure eruptions of the, 191
Ilaussmann, K., 284
Haute Ari^e, assimilation at the, 215
Hailynite, in alkaline rocks, 437
basalt, 412
Haiiynophyre, 412
Hawaii, 54, 135, 142, 228, 292, 423,
426, 9ee also undar Kauai,
Kilauea, Maui, Mauna Kca,
Mauna Loa, Molokai
alkaline rocks of, 509, 525
Heard Island, 423, 526
Heat developed by compreawon, 156-
157, 276-277
by injection, 210
by reactions in lara, 269 ff
flow of, 198 ff
latent, of rocks, 172, 273-274, 277
lost bv conduction, 255 ff
lost by ra<liation, 257 ff
magmatic, sources of, 155-156,210,
214,268
specific, of rocks, 199, 359
INDEX
543
Heat transfer of, in magmas, 157, 258,
264 ff, 274
Heats of formation and of reaction in
lavas, 271-272
of solution, 276, 278
Hebrides Islands, 415, 517
Hedley district, aplite of, 368-369
syenite clan in, 488, 512
Hedrumite, 393
Hedstrom, H., 354-355
Hegau, 519
HeUprin, A., 130
Heim, Arnold, 66, 402, 447, 454
Helena district, 490
Hellefors diabase, 314
Hengen, neck near, 296
Henry Mountains, Utah, 70, 71, 74,
236
Hemican district, 518
Heronite, 59
Herschel district, necks of, 130
Heterogeneity of volcanic piles, 288,
292, 372, 444
Heumite, 411
Hibsch, J. E., 209, 415, 438, 481
High Plateaus, trachytes of, 493
High wood Mountains, alkaline rocks in,
512
laccoliths of, 75, 223, 238, 463
syenite clan in, 490
Hill, J. B., 79
J. M., 407
Hills, R. C, 75
Hillsboro district, 498
Himalayas, granites in, 94, 96
Hitchcock, C. H., 280, 290
D. W., 291
Hobart, Tasmania, 524
Hobbs, W. H., 123
Hogbom, A. G., on Alno, 215, 419, 451
on anorthosite, 331
on assimilation, 306, 354, 383,
420
on Brefven dike, 356
on leptites, 124
on omoite, 324
on sills, 236, 242
on Sweden, 470
Hohbohl, neck at the, 296
Holbak, 504
Holland, T. H., 357
Holmquist, P. J., 18, 81, 356
Homogeneity, «e« Batholiths
Hopetown, sill near, 66
Hornblende andesites, origin of, 380
gabbro, origin of, 319
Homblendite, 446, 450, 452
Homblendites, origin of, 241, 327, 401
Horne, J., 300
Hornito, 135
Houghton, Australia, 509
Hovey, E. O., 130
Howford Bridge sill, 235, 243, 438
Hozomeen Range, 110, 459
Hrafntinnihraun liparite flow, 120
Hrossaborg volcano, 294
Hualalai, 135, 141, 280
Huber, O. von, 367, 369, 482
Huerfano Park, Colorado, 75
Humboldt Range, 495
Humphrey, W. A., 427
Hunne diabase, 314
Huronian geosynclinal, 186
Hybrid rocks, origin of, 312, 350, 354,
383,436
Hydrogen, thermodynamic properties
of, 278
Hypabyssal rocks mapped, areas of, 44
Hyperite, 313
Hypersthene andesite, 149
nearly restricted to volcanoes of
central type, 380
origin of, 380
basalt, 314
origin of, 319
Hypersthenite, 446
Iceland, dikes of, 81
fissure eruptions of, 118, 119, 120,
191, 226, 250, 289, 413
lava domes of, 136
subordinate volcano in, 294
Ice River, alkaline rocks of, 512
intrusive, 234, 241, 243, 434, 441
Spring volcanic cluster, 129
Idaho, anorthosite in, 323, 334
batholiths of, 98, 307, 388
fissure eruptions in, 381, 458
syenite clan in, 490
Iddings, J. P., on andesites, 880-881
on assimilation, 209, 306
on bysmalith, 84
on differentiation, 221-322, 237
on rock classificatioiit 10, 12
544
JSDEX
Iddings, on Yellowstone Park, 18, 448,
450, 476
Igaliko intrusion, 418, 441-442
Igneous-rock bodies, maximum sise of,
52
volumes of, 45-51
species, average composition of,
13, 19, 37 (special index)
of small total areas, 50
relative quantities of, 43
Igneous rocks, classification of, 9 ff,
311
general distribution and relative
quantities of, 42
primary division of, 39
varieties of, 40
Ijolite, 76, 410, 444
total area of, 50
Ilimausak intrusion, assimilation in,
418
d liferent iat ion in, 240, 439-442
laccolithic, 235, 418, 442
subsidence consequent on, 71
Immiscibility, magmatic, 226
Inchcolm Island, sill of, 234, 243, 435.
451,517
Inclusions, belt of, 200
In<iia, alkaline rocks in, 507, 521
anorthosite in, 323
fissure eruptions of, 118 (map)
quartz diabase of, 317
Injected bo<lies, classified, 61, 63
value in petrogenic theor>', 218,
229, 357-358
Injection, abyssal, 174, 181, 183 ff,
188 flf, 192, 248, 270, 311
causing downwarps, 185
relieving crustal tension, 183
Inoculation of magmas, 431, 436, 450
Insizwa Mountain sheet, assimilation
in, 242
differentiation in. 232. 316. 349
norito of, 318
peridotite clan in, 448, 454
Intermediate rocks, 78, 228, 238, 241,
312, 344, 347, 361, 384. 396,
402
Intrusive bodies, classifies I, 61
need of classification for, 61
rocks, dominant ly a(*id, 50. 53, 57.
464
Inyo county, 496
Ireland, batholith in eMteni, 96 (oMp)
plateau basalUof, 118
Iron baaalty 314
origin of, 321
ores (magmatic), origiii of, 241, 247,
312, 325-327, 350-351, 4M ff
Irving, J. D., 74
R. D., 329, 347
Ischia, 503
Is^re, 503
Ishan peninsula, 507
lahawooa quadrangle, 475
Ifllands, alkaline clana in, 509
syenite dan in, 524
Isogeotherms, rise of, 211
Italy, alkaline rocks of, 420, 484, 503, 518
Jackson, New Hampdure, 500
quadrangle, 366, 452
Jacupirangite, 397, 444
Jaggar, T. A., 75, 130, 132, 353, 876
Japan, cone chains of, 139 (map)
eruptive sequence in, 473
volcanoes of, 139, 147
Jarilla district, 498
Java, alkaline rocks oT, 421, 426^ 525
eruptive sequence in, 483
volcanoes of, 138 (map), 151, 158
Jensen, H. I., 58, 237, 435, 430, 484-485
Jevons, H. S., 237, 242, 438
John, C. von, 212, 216
John Day Basin, eruptive sequeoee in,
477
Johnson, D. W., 127
Johnston-l4ivis, H. J., 140, 174, 209,
305,423
Joly, J., 175,201,211
Jones Camp, 498
Juan Femandes Islands, 228, 423, 509,
525
Judd, J. W., 80, 137, 139, 371^873, 448
Judith Mountains, Montana, 68 (map),
74, 75, 491, 512
Jukes, J. B., 82
Julianehaab intrusions, alkaline char-
acter of, 527
assimilation in, 418
different iation of, 240, 417, 489-442
eruptive sequence in, 444, 488
laccolithic, 1 14, 235, 418, 442
stoping in, 194
syenite clan in, 510
INDEX
545
Jumimte, 412
Juvenile carbon dioxide, 419
gases, 246, 249, 270-271, 274, 278,
285, 287, 370
Kadi-Kale, Smyrna, 507
Kaersutite syenite, 402
Kaimekite, 394
Kaiserstuhl, 288, 518
Kakortokite, 411, 439 ff
Kalahandi, 507
Kamerim, 508, 522
Kamloops, 459
Karelia, quartz diabase of, 317
Karroo, intrusive sheets in the, 2(V5, 349-
350
quarts dolerites of the, 317
Karsuarsuk, Greenland, 402, 454
Kasaan peninsula, 489
Katzenbuckel, 505, 519
Kauai Island, 136
Kawich Range, 494
Kawsoh Range, 495
Keekeek Lake, 486
Keewatin rocks, basaltic, 56
Kekequabic.Lake, 499
Kelly Hill laccoUth, 71
Kelvin, Lord, 172, 177, 198, 256
Kemp, J. F., 454
Kentallenite, 394
Kenyite, 394
Keratophjrre, 394
origin of, 426
Kerguelen Island, 423, 510, 526
Kerne, 321
Kerr, H. L., 367, 451
Kesmek-Kopru, 521
Kettle River, 488, 512
Kewagama Lake, 486
Kiama-Jamberoo, 523
Kichatna valley, 489
Kikuchi, J., 148, 285
Kilauea, 56
activity of, explained, 254 ff
a subordinate volcano, 293-294
(map)
crystals in lava of, 376
dormancy of, 279
laccolith at, 76
lava fountains, 266
ring, 135
little explosiveness of, 285, 287
Kilauea sink, 151 (map), 153
small size of vent, 127, 280
superheat at, 414
surging lava at, 279
temperatures at, 212, 414
terraced crater of, 141
tumuli of, 134
two-phase connection at, 25^268
vent located, 255 (map), 266, 280
Kilauea volcano, satellitic origin of,
293-294
Kilbum crater, 285
Kiloran Bay, 502
Kilpatrick Hills, 517
Kilsyth-Groy laccolith, 232, 242, 354
Kimberlite, 446, 451
Kimituria Mine, 507
King, L. v., 180
Kinne diabase, 314
Kirai-guir River, 508
Kiruna, differentiation at, 367, 408,
453 (map)
eruptive sequence in, 397-398, 470
iron ores of, 453, 455
syenites of, 506
Kjerulf, T., 209, 306
Kluane River, 489
Knight, C. W., 427, 436
Knopf, A., 107
Kola peninsula, alkaline rocks of, 235,
505,520
differentiation in, 443-444
laccolith of, 76, 114
nephelite syenite body, 53
Kolderup, C. F., 58, 237, 322, 331-332,
335, 337, 367, 483
Konga diabase, 314, 316, 357
Konigsberger, J., 199, 367, 414
Korea, syenite clan in, 506
Kosciusko, 523
Kosvite, 446
Kraats, K. von, 416
KrabHte, 342
Kragerdite, 313
Krakatoa, 149 (map), 310, 483
Kranz, W., 423
Kruger alkaline body, 428, 443, 612
KrystallizatioDS differenUation, 221
Kula, 521
Kulaite, 411
Kuolajftnri, 520
Kuppe, 130
54fi
ISUEX
KuuMiujfi. ]j'jbt« of. Xi^j. 'j'Jft
Kylit^-. 4;J^
Kyxib^v^ii. H . %j
Kv-^btvmjt-t. 313
L&4i'.-kier .See brecciiLfe, 401. 505
Lh}frwioT. BiionLosiU^ of J41. 323. 3>4
Lii^»rbdorjte [^jrpb>Tit«. 374
rwk. 322
Lk'tvjlitliic onimi of aiiortboiute. 322
LhToLthf . 09. 29»3
ttiiunho6Jie ixL 33o
cozDp&red ftreally vith kI11<. ^m.
327.335
fjrzijxjrite. 72-73
ciUipfj^itJOD of. 75. 236
coinjj'juijd. 74
differentiate 240 ff
diviiie*!. 74
lllu^trht:rl^ peTroeeries^i*. 2'i''», •'^4;^,
4o2. 402
iDUrform&tioriiLl. 72-74
intruBioii of. 70
inward dip« at. 70
iD'^tiple. 72
Larrt»ix. A . on alkahne rc»rk>. 4.'»1
on aASimilatJon. 2<»9. 21'). 3»to-
3J>5
on differentiation. 'ii)ti
on fusion of xenolitli*. 212. 21''»
on Madaeasoar. 1>
on Mont TeUV. i:ii>-l.>l. 4vi
on Trel'izonil l-re^r'.a. 427
LadenburjE. R. 2<«
I^ke Champlain. 515
District, intrusion in KnEli-sh. :>.'»^
Labarce. 4S9
.Superior Regit >n. ys. 3oO
eruptive >eq'.ienix* in, 4r»l*
fissure eniptions in. l'.»l
fceosyu('linab> nf. IvJ
Laki fissure. lA), 250
Lane. A. C . 249. :<70
Lanfce)>er|cen. ^heet near. tW'i
Laplare. I*., nebubir byj>otbt*si«<. l.V>.
304
I^ riata quailransle. 3*N. 4'.»1
I-apparent, A. de. 21*)
Lappland. hnvobtb in. 11 1. 4lM
Laramie Mountains. *)s. 491
Larder Lake, 3SS. 4S7
Larwn. £ S.. 214
La SKiuk raUrr. 503
Lfefi Parroquiaa. ipbeDQlhh at, 86
La0pe>Te& H.. 130
Lac Vegv. 514
Latian diBtjict, 518
Latite. 201. 391, 398. 406
included in •jrenite das, 363
Laukkujarri, 506
Laurcotian batboUtlit, 406
Laurrik.53
LaurrikHe. 393, 397
Lauaiu. 504
batholith. 108. 200
Lava, bkick. 133
caeicades. 133
flov-B. snail ase of, 182, 279, 2iB
types of. 290
f ountaina, 256. 266, 268
lake, emptying of, 280
lakes. 130. 266. 280-281
manep. beierofliefieit j of, cz|4aiacd.
2&S. 292. 372, 444
outflov. causes of, 290
pjlow. 133
ropy. 133
«rarp». 133
tunneU. 133
%*L<co6ity of. 203, 288, 376
La%*as. oriidn of extrusiTe. 288
I^m « (*artle. diatreme at^ 252
Lan-ji<n. A. C. on anorthosite, 333
on ammilation. 242, 305-306, 336.
346
on Berkeley Hills, 478
on ^abbro. 331
on plumasite, 435
on stoping. 194
Leadville. Tertiary *"^f*— of, 404,
41S, 492
LeUimbo Range. 350
Le Cunte. J.. 177
I.«dmonte. 439
Lee. W. T., 1 19, 285
Leffendre. A. M., 304
Leitb, C. K., on Bad River beeolith,
3*29, 347
on Duluth lacooUth, 75, 114, 236.
325. 328
on Vox River basin, 123
on Lake Superior d]alriet« 469
Lemoine, P., 425
IXDEX
547
Lepsius, R.. 4S. lOS, 144. 200, 424, 444.
473
Le Roy, O. E.. 402
Les cheires, 133
Lestivarite, 411
Lethan HiU, 451
Leucite, 437
basalt, 415, 432
porph>T>% 238
basatiite, 378, 412, 422
Hills, necks of, 130. 513
syenite, 410-411
tephrite, 423. 434
Leucitite, 412. 414, 423, 427
Leucitophyre. 411, 423-424
Leucoph>Te, 314
Level of no strain, 176 ff, 18^5
L<5vy, A. Michel, on a&similation. 209,
305-306
on earth's acid shell. 164
on eruptions at Mont Dore, 480
on fusion experiments, 213
on rock classification, 12
on two priman- magmas, 52. 308
Lewis, C, 451
G. X., 278
J. v., 207, 236, 237, 316, 4.50
Lherzolite, 446
relation to gabbro, 448
Liebncrite prtph>Ty, 411
Liesegang, R. £., 443
Limagne, 424, 444, 481
Limburgite, included in alkaline clan,
410, 412
melting temperatures of, 213-214,
376
origin of, 377, 422-423, 444
Lime-alkali series of rocks, 393
Lime-silicat«s in alkaline magmas, 436
Limestone, magmatic absorption of,
414 ff, 420 flf, 430 ff
Limestones, composite analysis of,
400-401
Lindgren, W., 136, 364, 385, 471, 479
Lindoite, 411
Linebarger, C. E., 227
Lipari Islands, see Eolian
Liparite, 120, 342, 422
origin of, 284, 370
Liquation, 170, 221, 225 flf, 227, 229,
238, 441, 443, 454
Liquefaction of inclusions, 212
Liquid fractioiis, 221, 229
phase preceding crysUUittlioii,
225 ff
Liquidity, magmatic, 213, 9e€ Viscoaty
Litchfieldite. 411
Lit par lit injection during the |»e»
Cambrian, 60
Little Belt Mountains quadrangle, 53,
405,513
Falls quadrangle, 499
Livingston quadrangle, 405, 420. 476
Lix-radois, le, 424, 432
Loch Borolan, alkaline rocks of, 420,
502,517
differentiation at, 241, 361, 439,
444,451
laccolith at, 235
syenite clan at, 502
syntexis at, 243, 420
Lockyer, J. N., meteoritic hypotheBis,
155
Loco Mountain stock, 363 (map), 365
Loewinson-Lessing, F., on assimilation,
209,305
general theory, 307 ff
on differentiation, 221-222, 227
on general magmas, 52
on rock averages, 169, 382
on rock classification, 12
Ix)foten Islands, anorthosite of, 323,
334, 337
syenite clan in, 510
Lof tahammar granite, 356
Logan, W. E, 186
Loh oelo, 525
Long Lake quadrangle, anorthosite of,
240 (map)
eruptive sequence in, 469
syenite of, 499
Loon Lake, 500
Los Islands, 508, 510, 526
Loughlin, G. F., 236, 352-353
Lourous volcano, 426
Low, A. P., 331
Lowe, H. J., 237, 435
Lugar sill, analcitic rooks in, 243, 517
differentiation in, 234, 239, 438,
451
magmatic viscosity in, 241
Lujavrite (lujaurite), 411, 439 ff
Lundbohm, H., 367, 897, 468, 470
Lunn, A. C, 157-158
548
I\DEX
Luray, 498
Lurcombc, sill near, 235, 435, 517
Lutterworth townphip, 487
Luxullianitc, 341
Lydgate, J. M., 255
Lyell C, 145
Lynx Mountain, 489
Lyon Mountain, 499, 515
Maars, 144
MacDonald, D. F., 345
Maclcar, South Africa, ncrks of, 130
MadagaRcar, alkaline rocks of, 423,
425 (map), 510, 526
Madeira Itflund. 423. 510. 526
River, Brazil, 502
Madoc district, hatholithic Hhattering
in, 200
Madras, 507. 521
Madupite, 412
Madiu-a L<iland, 421
Maenaite, 411
Magdalena, New Mexico. 498
Magma basalt. 412
complementary', 241. 373, 444. str
Diaschistic
primar>', 1(>4 fT. 315
Hccondary, 207. 216. 219. 242. see
Assimilation
MagnioH. genetic classification of. 312
general. 165 fT, 308
Mngmatic gases, 249, hcc Gas
heat, sources of, 155, 210
(»res, 226, 349, 446 fF, 454-455
Magnet l>erg, 505
Magnet Cove, alkaline rocks of, 420,
498, 514
Magnetic abnormalities due to intrusion
of n)ck, 284
Maine, alkaline n)cks of, 500. 515
Mainz b:isin, 518
Malignite, 1 15. 411. 428. 443
Malpais of Mexiro. 133
Mamelons, 131
Manchuria, ne|)hclite basalt in. 522
Mandelsh>h. C ount. 2%
Mangerite, 332
origin of. 337
Murico, 522
Mariu|K»lite, 411
Maroochy-Cooran. 523
Marquesas, 509
Marquette district, 50, 400
Marr, J. E., 100
Marshall, P., 424
Martinique, eruptive sequence in* 483
MartiUB, S., 401
Martonne, E. de, 100
Maryland, diabase dikes io, 80
Mar>'8vale, 494
Mar>'8ville, Montana, sills near, 346
stock, 100, 104, 105 (map), 106, 461
Masafuera Island, 228
Masaya volcanoes, 147 (map)
Mass, igneous, 62
M assachusetts, alkaline rocksof , 801, 515
granodiorites of, 388
Matanuska valley, 480
Matatiele, Cape Provinoe, gi«nt dOw
in, 80
necks in, 130
Matavanu volcano, 268, 287, 838, 414
Matopo granite, 356
Matto Grosso, 501
Maufe, H. B., 85, 122, 106, 480
Maui Island, 136, 281
alkaline rock in, 525
neck in, 265
Maima Kea. 228. 280, 426
Loa, a lava dome, 135
dormancy of, 270
lateral eruptions of, 250
map of, 203
nested sinks of, 152 (map)
small sixe of vent, 280-n2
superheat in, 212
terraced craters of, 141
volumes of flows, 200
Mawson. D., 437
Mayenne, 503
Mayon volcano, 275
Maxapil Valley, 501
McCulloch cone, 132
Mcdford dike, 353
Me<lvo8 Mountains, 520
Meister, A., 427
Melanite pyroxenite, 430, 444, 451
Melaphyre, 314
Mclilitc basalt, 410, 412.422-tfI, 432.
435
lacks a plutonic equiYalent, 432
origin of, 436
basanite, 424
in alkaline rocks, 436-4S7
r inclusions in basal tic Ibvb, 218
of rocks, 214, 273
s of Tocka, 210, 213 B
F. P., 169, 356
G., 124, 134
1,506
&nge, 499
«s, 66
i Lake, 486
«re of eartb, 160
ihic aureoles, 103, 106-108,
163-364, 416, 419, 431
>hi8m, dynamic, of igneous
ocks, 454
lation to magmatic tempera-
ures, 211, 370
ir static, 176
me, 132
9, composition of, 160, 166 ff
toirastiip, 487, 511
Ikaline rocks in, 516
dioritic rocks of, 388
X clan in, 601
noes of, 138
;, 138, 444, 480
Midciltility, limited, 170, 225-226
410
*11
[iass), Ural Mountains, 434,
05,520
tsitea, oriipn of, 380
08,313
[in of, 317
B, composite analysis of, 400-
01
quarts gabbro of, 317
;eB in, 499
>ro, 313
natite, 241, 317-38, 345, 348-
50, 353-357, 359-^360, 406
district, British Columbia,
44, 476, 488
of magmas, 404, 418, 460
174
341
isin, 494
07,434-435
, intrusive rocks of, 68, 75,
»)-231, 241,336,346.361,406
: diabase of, 317
e clan in, 499
total area of, 50
Mitlagong, 509, 523
Mittelgebirge, Bohemian, 504, 520
Mixed rocks, tee Hybrid
Moberg, J. C, 356
Modder Fonlein volcano, 153
Mode clasaificatioD, 9, 10, 12, 13, 14
Moira Sound, 489
Moisie Ri jer, anortbosite of, 53, 323,
334
Mokuaweoweo, 141, 152, 212, 266, 268,
280, 287
Molt-ngrnafT, G. A. F., 236, 350-351, 427
Molokai Island, 136
Molten surface of the earth, 159
Moluccas, 525
Monarch and Tomichi districts, Colo-
rado, 86 (map), 493
Monchique intrusion, 416 (map), 503,
518
Monehiquite, 411, 452
Mondhaldeite, 411
Monmouthite, 411
Monmouthshire, 452, 517
Monmouth township, 487, 512
Monomineralic magma J 447
Mont Dore, 424,4^4, 480, 503, 517
Pelfe, 130 ff, 483
Montana, alkaline rocks in, 420, 512
batholiths in, 307, 388
sapphires, 434-435
sills in, 346
i^tooks in, 110, 195, 363-365, 402
syenite clan in, 490
Montarville, 511
Monte Adamello, ethmolith at, SS
Perm, 527
Summa, 145
Vulture, 518
Montea^Ie township, 487, 612
Montenegro, 504
Monteregian Hills, 54, 396-397 (map)^;
451
Montezuma flange, 495
Monnoni, alkaline rocks of, 504, 519
chonolith at, 85
differentiation at, 369, 462
eruptive sequence at, 367, 482
Monzonite and the t«lluric magma. 41
in relalion to pyroxenite, 450
550
r.WDEX
Monzonitp, inrliided in syenite clan,
oriKin of, 337. 3(>S, 4a-). 408. 414
p«»ri)hyry, 72, 74
rcpla/^emonl hy, 101
:«ynKen<:tic with fEr:ino«iiorite, 428
with hiKhly alkaline rork*». 307, 414
with latite, 201
Mf>on, volranir nurfarf* of. 150
M«K>re. J. E. S.. 250
Moreno dLstrict, 497
Morin diMtrict, anortho^itc of, Xi^i, 323,
337
differentiation in, 240, 327
laccolith c»f. 233. 241, 32S. 3:i0
(map;. 331
Mon>zewi(!z, J., 227, 303
M or van. 503
Mother Lode district. 380, 452, 49ti
Moiilton. F. R., 155
Mount A.-'ciitney. see Af^ntney
liaker, me Front i?*pi#^#* and paRrs
101. 370, 31H)
Belrj«'il. 511
Bronif. 4sr». 511
Diahlii. pyrr>\fnitc' of. .-»3
Flindor.**. 524
(lirmir. 522
iliialtlai. 13.'i. 111. JsO
Ihlh*r>* laf^otilith, 71
I Inline;* hysmalith. si
Johnson. 22h rnap . 4V». 511
I/ofty KanKos. umnitcs of, *»S
Maretlon. KnintHlioritcof. 301-302,
r>(H», .V.M
Orfonl. 4s»)
Pni-iMTt. 235. 237. 212. 43^. 523
Ki^aud. 4st»
Koraini:i. lai'colith :it. !{55
Royal. 511
St. Hilain*. 511
ShffTt.r.l. 403 map . ls»>. 511
Stuart f{uailrani!h'. dikes of. 121
iinap . 45s
eruptive sim|Ui*ihv in. 177
peri" lot Iff of. .Vi
Ta\lor. nerk- of. 130. 251
Tripyraniid. 3ii7. 171. .'iiMl
\'t*'«uviii>. tin \f-iuviu^
Vaniaxka. 51 1
Mountain-huddnu:. i-oiiijitions for. Iss,
193
Moyie sills, Msimilfttion in* 343, 33A,
345
differentiation of, 228, 231, 237.
240-243, 344
Kabbro of, 319-320
injection of, 219, 3411
intermediate rocks in, 384
Miigearite, 67, 411
Mugodjaren Mouotains, 507
Muir, M. M.. 271
MQUer, J.. 414
Murgoci, G. M., 437
Mveni Creek, 490
Naknek Lake, 489
Namqualand, alkaline rocka in, 523
necks of, 130
Nandewar Mountains, 50B, 523
Naril)erRet, 402, 455
Natal, sills of, 240
anorthnsite in, 234, 323
aMumilation in, 350
differentiation in, 325
hybrid rock in, 232, 350
pyroxenite in. 325
Naujaitc, 411, 439 ff
Navite, 374
N'erkH. volranir, 83, 126, 228
aJ«oriated with sills, 298 ff
eompo^iite, 126, 128 ff
cylindrical form of, 282
dimensions of, 130
lava, 126, 127
limall site of, 127, 129 ff
tuff, 126, 128
Nelson district, 488
Nephelite, synthesis of, 437
basalt, 410. 412, 423-423, 433, 444
bafianite, 424
dolerite, 412
minctte, 411
syenite, 76. 115, 397. 410. 444
associated with linwatone, 335.
410, 428 ff
wit h subalkaline rocka, 851, 427
calcite in, 434
origin of, 411. 419, 438 ff
NephelinitG, 412, 432
Neix»n.'*et Valley, 501
Nmist, W., 27i
Ne:«te<l caldoras, 147, 150
craten, 143-144, 146
INDEX
551
Nested sinks, 152-153
Neuffen, boring at, 296
Nevada, alkaline rocks of, 514
City quadrangle, 366, 496
granodioritic rocks of, 388
syenite clan in, 494
Nevadite, 342
New Brunswick, anorthosite of, 323
New England, batholiths in, 98, 307
Newfoundland, anorthosite of, 323, 334
New Hampshire, alkaline rocks of, 419,
515
syenite clan in, 500
New Jersey, alkaline rocks in, 515
anorthosite. in, 323
sills in, 66, 240, 316, 449
New Lake in Halemaumau crater, 266,
280
New Mexico, eruptive sequence in, 471
necks of, 130, 251
subjacent bodies of, 98, 388
syenite clan in, 497
New Mountain (Usu-San), 298-300
New Pomerania, 509
New South Wales, alkaline rocks of,
508-509, 523
batholiths of, 96, 98, 115, 389
eruptive sequence in, 444, 483-484
geosynclinals of, 188
melilite basalt in, 435
New York, alkaline rocks in, 515
anorthosite in, 114, 241, 334
batholith in, 98
syenite clan in, 499
New Zealand, alkaline rocks of, 423-
424, 509, 524
granites of, 94, 96, 98
Nicaragua, volcanoes of, 147, 250
Nickel Plate Mountain, aplite of,
368-369
Nicola volcanics, 459
Nicolson, J. T., 179
Nicoya peninsula, 516
Niger-Benu6 district, 522
Nigger Hill laccolith, 74, 414
Nightingale Island, 526
Nijni-Tagilsk, 507
Nindiri volcano, 147 (map)
Nipissing district, sills of, 339
syenite in, 487
Noble, L. F., 237, 406
Nodules, ultra-femic, 448
Nogal district, 497
Non-consolute magmas, 222, 226, 328
Nordenskjold, O., 124
Nordmarken, nordmarkite of, 53
Nordmarkite, 113, 393, 396-397, 403
origin of, 439
Norite, 313, 348-351
origin of, 318, 337
Norm classification, 9, 10, 12
Norrbotten, 452
North America, igneous provinces of,
43
Atlantic fissure eruptions, 1 18
Carolina, anorthosite of, 323
Creek quadrangle, 383, 499
Peak, Colorado, 491
Star district, 494
Norway, alkaline rocks of, 506, 521
anorthosites of, 241, 322, 334, 336-
337, 396
batholiths of, 98
eruptive sequence in, 396
Norwood, J. C, 346
Noss Sound, necks in, 300
Nova Scotia, syenite clan in, 486
Oahu Island, 525
Oberwiesenthal stock, 444
Ocean, origin of, 246
salinity of, 159
basins, general absence of acid
rocks, 317
Oceanic sodium, origin of, 159, 163
Odenwald, 505
Okanagan composite batholith, 115-
116
differentiation of, 366
eruptive sequence in, 444, 477
rock association in, 428
vesicular dikes cutting, 265
Oki Islands, 510
Old Hampshire county, asBimilation
in, 389
Oldham, R. D., 119, 160» 304
Oligoclase granite, 353
Oligoclasite, 313, 437
Olivine basalt, 314
gabbros, 313
norite, 313
segregations, 451
Olivine-free basalts and gabbros, 310
Omori, F., 298
552
ISDEX
00
Ontario, alkaline roclw of, 419, 428. 451,
511
anorthoflites of, 323, 325
hatholiths in, %, 98, 388
quartz diaha.«e of, 317
sill»« in. 23S, 241. 323, 325, 339. 361,
432
hvcnitc clan in, 4 87
Oorlou's* Poort, volcanic pipe at, 21*4
Ophir diMrict. Utah, 494
Ophite, 314
Oquirrh Range, 493
Orbicular rocki<, 226
Order, eruptive, 469, iee Se<iueD<'e8
of cr>'8tallization, 375
Oregon. firi.sure eruptions of. 119. 459
granorlioritic roclui in. 3HH. 390-391
Orcndite. 412
Ores, magmatic. 226, 349, 446 If, 454-^55
Ornoite. 324. 374
Orogeny in relation to abyssal injection,
188
Oronsay. 415
Orthoclatje gabbniH. 313
origin of, 317
Orthophyre. 391
Ortleritc, 374
(H*ann. A.. 16. 18. 168
Osmotic transfer, Ii53
Ostwald, W., 225
Ottawa county. Q'.iel>ec, 4S6
OttfjfiHdiabaw. 314
Ouachitite, 411
< hiray quadrangle, 476, 492
Overt hruMing in relati<»n to batholithic
intrusion, 91. 94. 190, '206. 460
Ovifak iron, 321
Pacheco. K. H.. 135
Pacific branch 'suited of igneous n>cks,
42. 54. .33S. 412-413
geoHynclinal. main. l>>7
v<»lcanoes of the. 310
I'ah«>eh«M' lava. 133, 260. 291
P:ihn>c Itange. 495
Pahute Range. 4<.>4
l*ainin>va. .VKi
Paisunite. 113. 342. 411
Palache. C . 47S
Palagonite. 314
P.ilandokan plateau,
Palatinite. 314
Paleophyrite, 374
Palisades sheet, New Jeney, chilkd
phase of, 237
differentiation in, 231, 316, 450
intennediate rocks in, 384
map of, 449
stoping in, 207
Pansmint Range, 405
Paotelleria Island, 423, 527
Paotellerite, 342
Panxerdecke, 301
Paradox Lake, 4W
Paraguay, alkaline rocks in, 516
Pardee, J. T., 236, 345
Patagonia, alkaline rocks of, 407, 502,
516
batholith of, 92 (map)
Peach, B. N., 300
Peekflkill, 500
Pegmatites, 342
origin of , 368 ff, 452
Pelham, Massachusetts, dikM in, 79
Pembrokeshire, syenite dan in, 502
Penck, W., 429, 482
Penobscot Bay quadrangle, eraptive
sequence in, 474
flyenite clan in, 500
8yngenetic granite and diorile in.
245
Penokee District, 499
Peri<iotite, 402, 427, 446 ff, 440
clan, origin of, 446 ff
small volume rcptfented, 53» 452
Peridot ites, association with andesiu,
450
origin of, 241, 312, 324. 326, 377.
446ff
relation to gabbro, 447
Peridot itic sheU, earth's, 106
Periodicity in central veatai 248, 286.
275
Perovskite in alkaline rocks, 436
Perret, F. A., 212, 376
Perr>' Basin, granite of, 98
Perthitophyre, 313
Perthshire, neck in, 126
Petnuwh, K., 213
Petrogenic theory, Tahie ol, xarit, 456,
465
Petrographical provineea, M
Pfaff, F. W., 178
' eolitlH,76
INDEX
553
Phenocrysts in Kilauean lava lake, 376
Philippi, E., 415
Phlegrean district, 518
Phoenix district, 488
Phonolite, 72, 410-411
associated with basalt, 422-426
origin of, 414 ff, 426, 444
Phonolites, small volumes of, 45-46, 48,
414, 432-434
Phreatic explosions, 282 ff, 285
fluids, 249-250, 282
Phyllites, composite analysis of, 400-
401
Piatigorsk, 505
Picacho Range, 493
Picrite, 427, 436, 446
in relation to alkaline rocks, 451
origin of, 239, 377, 451
Picrites, lack of vesicularity in, 452
Picton granite. 111
Pier, M., 272
Pigeon Lake, 487
Point intrusive, 206, 230, 242, 325
(map), 336, 346-347 (map)
Pilandsberg, 351, 420
Pillow basalts, 133, 338 ff, 340
Pilot Knob, 514
Pinon Range, 495
Pinos Altos, 497
Pirsson, L. V., on assimilation, 209
on Bearpaw Mountains, 408
on Belknap Mountains, 367, 474
on differentiation, 306
on Highwood Mountains, 18
on Judith Mountains, 68, 70,74-75
on Little Rocky Mountains, 73
on Red Hill, N. H., 419, 474
on rock classification, 10
on Shonkin Sag laccolith, 228-224,
237
on Tripyramid Mountain, 474
on Yogo Peak, 402
Pitchfltone, 342, 371
Pits, lava, 141
Plagioclasite, 448-449
Plagioliparite, 386
Planetesimal hypothesis, 155 ff
Planets, composition of, 158
densities of, 158
temperatures of, 158
Plasticity of rock, 178
Plattenlava, 133
Plauen'scher Grund, 504
Plug, destruction of volcanic, 276
formation of volcanic, 275
Plug-domes, volcanic, 130 ff, 133
Plumasite, 435
Plutonic rocks mapped, areas of, 44
classification of, 14
largest bodies of, 53
Plutonics chemically contrasted with
corresponding volcanics, 19 ff,
39, 229, 315, 370, 384, 386,
409,443
Pneumatolysis, 339, 402, 431, 452
Point Sal, 514
Polymorphic changes, 176-177
Ponape Island, 525
Pooh Bah Lake, malignite of, 53
Porphyrite, 374
Port Coldwell intrusives, 367, 419, 451.
487, 512
Port Cygnet, 509
Port Henry, New York, 499
Port Orford quadrangle, intrusives of,
233, 243, 391, 460
Portugal, alkaline rocks of, 416-417,
503,518
Possession Island, 423, 509, 524
Potchefstroom, 508, 522
' Potentialized energy in earth magma,
272, 278, 302
Pouzac, 517
Pre-Cambrian, a time of intense igneous
action, 59, 322, 330, 335
complex, average composition of,
162, 382
Predazzo, alkaline rocks of, 519
eruptive sequence at, 396, 444, 482
pyroxenite at, 451
syenite clan at, 504
syngenesis at, 367, 429, 439
Preobrajensky, P., 434, 437
Pressure in relation to melting temper-
atures, 210
Preston, Connecticut, laccolith of, 231,
352-353 (map), 448
Pretoria district, sills of, 232
Preuss district, 494
Principal volcanoes, 300, 302-303, 460
Prior, G. T., 237, 242, 319, 326, 339, 350
Propylite, 374
Prospect Mountain intnisiye, 236, 237,
242, 438, 623
554
ISDEX
ProtcrohaAC, 314
PrutcK*la8tic structure, 3ii5
Proworsite, 412
pHCiKlolcurito jHjrphyry, 414
Pulttskite, 113, 22S,'393, 306-397, 4(W,
410, 440-441
Puna district, pit craters of. 142. 2*»:»
Purcell Mountains, rhyolite flow in, 370
»\\U of, aK.*(iinilation in. 242. 33<>.
345, 400
(lifTorcntiation of. 220, 231,
237, 240-243, 344. 4(M), 450
gabhroof, 310 320
injection of, 210, 340
Pt oping in. 2(M>
thicknesH of. 0)0
Pure-iliffercntiation theor>*, 3()0, 410
Puy-cie-Dome, ,50.3
Pyramid Lake, 405
Peak quadrangle. 3«)0
Pyrenees, hatholiths of, 05, 07 (niap».
112
Pyr<M-la.stic andesites. 4r»4
Pyroxene andesitc. 450, 4<hi
porphyrite, 450
Pyroxenic magma, lti5
Pyroxenite, 351. 420, 440. 452
Pyroxenites. *' alkaline." 450
derive*! from setliiiH'ntary ?iyn-
tectirs. 450 flf
origin of, 241, 325^ 327. 350, 130.
451
Quantitative rlassifiratiim. x*v Norm
study of igneous rocks, need for, 42
Quartz hiisalt, 314. 422
origin of. 310. 35r»
diahasi'. 311, 3ir) ff
generally a>snria(eil with mrks
of l»n>altir romi>iisition. 317,
357
scconilary origin of, 317. 321.
3.*)4- 355
diorite, in relation to diorite. 105
in relation to granodiorite.
3.S5-3S0
origin of, ,3ti5 'M\\\, 3s7
syngeneti«' with nioii/.onite. 12s
dolerites. origin (»f. 31(>
gahhro. 313, 352 353
origin of. 31i». 3'>0 357
kcratophyrc^, 342
(juarti kcratophyrefl, origin of, 347,
370
latitc |)orphyry, 86
monzonite, MX, 309, 385-386, 454.
403
origin of, 334, 407
porphyry, 86, 364
syngcnetic with alkmline rocks.
429
noritc, 313
origin of, 317, 337
pantellcrite, 342, 371
ix)ri>hyrite, 386
porpli>Ty, 342
syenite, 342, 408, 429
originof, 361.402, 441
trachyte, 342
(Quebec, anorthociiteH of, 241, 333, 331,
334
(|uartK diabatic of, 317
Mtr>rkH in, 396-397, 402
syenite clan in, 486
Qiu*ensland. alkaline rocka of, 509, 523
hatholithn of, 98, 389
eniptive sequence in, 485
Quensel, P. I>., 92, 228, 407
Hack, (;., 420
Radiation, heat, 157
Radioactivity aM sourre of omgBiatie
heat. 172, 175, 211, 288, 272,
3(i0
Raglan township, 4H7, 512
Ragunda. 5(Mi
I{aiatea Island, 525
Rainy Lake, anorthosite of, 323. 327,
3:^3
svenites of, 4S7
Rajputana, 521
Rampart region. Alanka, 489
Ramsay. W., 114.237
Ran deck niaar, 296
Raniganj district, anorthosite in, 323
Ransome. F. L.. 18, 237, 354, 389, 403,
407. 472, 479
Rapakivi granite, 341
Rath Jordan. 517
Rca.le. M.. 177-178
Rerk. II.. -JIM
Red Hill. New Hampahire, 419^ 444.
474, 500, 515
River, New Mexioo, 498
Red rock, 242, 329, 336, 346-347,
350-351,354,370,391.407
ReentraDts, stoping, 196
Re-fusion of earth's crust, 300
Regatta Point, 524
Regelmann, K., 295
Reioisch, R.,41S, 451
R«nts, volCBJoic, 153
Replacement, magmatic, 109, 111 fF,
195, 216, 383, 388
Residual magma, expuleion of, 246,
445
ReHurgent gasea, inBuence of, 243, 340,
346, 361, 370, 381, 401, 405,
437
origin of, 246, 249, 2S5
Reunion Island, alkaline racks of, 526
differentiation in, 228, 378, 450
driblet cones in, 135, 292
rock association In, 423, 451
ateam-explosion in, 287
syenite clan in, 510
volcanic sink in, 150
Rhine region, rock associations in, 423,
444
Rhodesia, dikes of, 356
Rhomb-porphyry. 67, 39^
Rhfin, 505, 519
Rhyolite, 83, 120, 342, 380, 463
origin of, 360, 357, 370-372
Rice, W. N., 367, 474
Richards, T. W., 225
Rico district, 418, 491
Riebeckite. formation of, 437
Riesengebiige granite, 341
Rieekeseel, 283 S (map)
Rigaud Mountain, 402
Rigidity, and temperature, 181
increased by pressure, 172, 179
Ringguit volcano, 426
Rings, lava, 135
Rio Magdalena, 501
Payne, 502
River Range, 495
Riverside, Arizona, 497
Riuonite, 412
Robert»-Austen, W. C, 198
Robinson diorite. 364
Mine, 495
Roccamonfina volcano, 423 (map)
Roche, E., 304
River, 488
^EX 556
Rock, average igneouB, 168 ff, 308
bodies aaaociatwl wifli central
vents, 126 ff
Rocky District, *M
Mountain geoByndinal, 187, 457
MountaioB <rf Canada, absence of
granites in, 94, IW, 460
fissure eruption in, 191
Rogers, A. W., en assmilatian, 383
on dikes, 80-81
on kimberlite, 451
on sills, 67, 114,265,360
Romberg, J., 429
Roof-found»ing, batholithio, 121, 203,
206
Roof-pendanta, 100, 103, 104, 105
Roofu of subjacent bodies, 100, 103,
106 ff, 205
"Rooi Hoogte sheet, 67
"Ropy iava, 133. 291
Roue, G., 341
Hoseburg district, Oregon, 75, 113, 390-
391 (map), 400
Rri^nbusch, H,, on alkaline and sub-
alkaline suites, 426
on andesit«a, 380
on assimilation, 209
on classification of rocks, 9, 12, 13,
16, 18, 39, 40, 51, 59, 313, 341,
374-375, 385, 393, 410
on diabase, 319
on differentiation, 306
on general telluric magma, 413
handbooks, 317, 356, 399. 4fi0
00 peridotitea, 447-448
on picrite, 451
on pyroxenite, 450
Roaita Hills, syenite clan in, 492
volcanic sequence in, 372, 475
Rosiwal, A.. 12, 426
Ross Archipelago, 509, 524
RoasUnd district. 488, 512
Rothschonberg, 305
Rotomahana caldera, 146 (m
Rougcmont, 511
Rueker. A. W., 19S
RUdemann, K., lOS
Hudaki, M. M. P., 177
Rum, Island of, 234. 323-334
Runn of Cutch, ace Cutrh
o5G
INDEX
Russia, alkaline rocks of, 505, 520
anorthosites of, 323
St. Hcloriii. 423, 520
St. Joe River, 490
St. Kilcia U]&ml 70
St. IVbain, anorthopite of, 323, 334
Sabatinian <li.strirt, 518
Sagiicnay district, anorthosite of, 53.
323, 327-328, 335
eruptive sequence in, 469
Saleyer, 525
Salisbury, R. D.. 15<). 159
Salite diabase, 314
Salmon River. British Columbia, 48^
Salomon, W., 87
Salta Province. 510
Salvages Islands, 526
Samoa, 423, 525
Samos Island. 510
San .lose <listrirt. 501, 516
San Juan, Argentine, 516
San Luis Obispo, 514
San Luis quadrangle, 421, 477
San Miguel Islan<l. 146
Samlstones, c(»mix)site analysis of, 100-
401
Sandwich, New Hampshire, 500
Sangre de Oisto Range, 492
Sanidinite. 394
Santorin, 1 is (map)
Sanukite. 412
Sao Pjiuhi, 501, 510
Sao Thomi^. 423, .>02
Sardinia, alkaline rocks in, 527
Sarna diaba«JC. 314
Sary-Houlak gorge, rA)7
Sat I'll it e.»*, maginatir, 2.*)5, 291, 3.V,»,
3l)l-3iV2. 3S3
Satellitic injertions, vulcanism origi-
nating in. 291 flf. 300
Sauer. .V., 2S4. 444
Sauk Center, 499
Savaii. alkaline rocks in, 525
volcano in. 2t>.S. 292. 33S
Savoy, orthophyre in, .503
."^axony. alkaline rocks of, 519
granites of, KHi. 108
Scandinavia, overt hrusts of, 1*H)
Scapolite in alkaline rocks. 43(» -137
ScIkuIc. II.. 225
Schalstein, 314
SchisUwity, peripheral, 96, 102
Schliers, ultra-femic, 448
Schneeberg bathoiith, 106, 106
SchoGeld, 8. J., 236, 320, 345
Schollcndome, 133
Schwantke, A., 321
Schweig. M., 221, 222
Scotland, alkaline rocks in, 517
great dikes in, 81
laccoliths of, 354
plateau basalts of, 1 18, 191
volcanic necks of, 126-129, 296-
298
Scyelite, 446
Sedcrholm, J. J.. 50, 209, 306. 470
Sedimentary control over ssmtectia,
3K7, 395, 399, 403, 405
shell, earth's, 162, 171. 304
Sediments, average analyses of, 399-
401
Seebach, K. von. 147, 250
Seeburg. neck near, 296
Scetova Mountains, 495
Segregations, basic, 226. 247, 452. 454
Sckiya, S., 148, 285
Selective solution, 370
Selkirk Mountains, gran it« stock in, 107,
111
Selvagem Grande (Salvages Uands;.
526
Semoroe volcano, 290
."H^pulrhre Mountain, 288, 380
Se(|uences. eruptive, 55, 311, 409
in alkaline provinces, 443 444
in batholithic areas, 244, 302
in composite stocks, 390-397
in the Cordillera, 390. 401. 470, 475
involving syenite, 396, 402
Seri>entine, 448-449
Serra <ie Monchique, 503, 518
.*^:ward peninsula, granite stock in, 107
S<'y<'helle8, 510
Shales, composite analysis of, 400-401
Shand, S. J., 237, 361, 439, 444, 451
Sliannon Tier, 524
Shan-Tung, batholiths of, 08
syenite clan in, 506
Shap granite, 100 (map)
Shatter-blocks, sinking of, 203 ff
Shattering aided by impnaooad fluids,
■200
marginal, 197-201
Sheet, 64
interformational, 6(1, 348
Sheets, importance of, iii petrogcneHis,
229, 343, 391
ShefFord Mountain, 53, 403
Sheibner, C. P., 4X7
Shell o[ fracture, depth of, 179
Shells, earth, tee Acid, Compression.
Sedimentary, Tension
Shepherd, E. S., 212, 376
Sherman quadrangle, anorthoiiite of,
323
Shetland, necks in, 300
Shinumo, Arieona, sill at, 233, 243, 396,
406
Shonkin Sag laccolith, contact chilling
in, 237-238
dilTerentiation in, 223 ff, 234,
238, 243, 438
magmatic viscosity in, 76, 240
stock, 53
ShoDkinite, 410, 450
origin of, 238, 405, 40S, 429, 444
total area of, 50
Shoshone Range, 495
Shoshonite, 412
Shuswap I^ke, 48S
Siberia, alkaline rocks in, 427, 522
quartz diabase of, 317, 321
Sicily, alkaline rocks of, 527
Siebengebirge, 130, 481, 505, 519
Siegl, K., 257, 302
Sierra Caliuro, 496
Eacudillo, 496
Luera, 497
Nevada, Califomiu, batholith of.
53, 98, 102, 105, 366, 388, 390
eruptive sequence in, 478
homblendite of, 452
Tertiary volcanii^ of, 406
Silesia, 504
Silicic acid, compared to carlxinic acid
in strength, 414
Sills, 64, 230
areas of, 67
chemical composition of, 68, 230,
236
compasite, 66-67, 447
differentiated, 65, 230. 240 ff
illustrating petrogeni^ais, 230, 343,
452, 462
intrusion aided by gasntension, 346
iEX 557
Sills, multiple, 65
of British Columbia, 66-67
of New Jersey, 66
of South Africa, 66
Silver Cliff, 492
King, 497
Peak Range, Nevada, 494
Silverton quadrangle, 476, 492
Silvestri, O., 250
Similkameen batholith, 104 (map), 366,
428
Simotomai, H,, 133
Sinking of (Tj-stals, 222, 227, 238, 376
of xenoliths, 202 £F
Sinks, volcanic, 150, 152
Sinni valley, laccolith of, 232, 448
Sin-ash Creek, 488
Skagit Range. 488
.Skaptar Jokull eruption, 117, 2S9
Skeata, E. W,, 391, 484
Skidoo district, 496
Skve, Island of, anorthoaite of, 233,
241, 323-324
handed gabliro, 324
dike in, 80
eruptive sequence in, 372, 479
laccoliths in, 72, 76
ailU in, 66-67, 233
syenite clan in, 502
Klatowralaky, N., 225
Smaland, 606
Smith, G, O., 121, 187, 356, 474
Smyrna, .521
Snake River, Gisiire eruptions of, 191,
380
Snoqiiahiiic bathulith, 105 (map), 121
Society Islands, 626
Socotra, .'■>08
Soda granitea, 427, 437
origin of, 337, 437, 439
granopbyre, 342
lipsrites, 342
origin of, 370
rhyobie, 342
syenites, origin of, 337
lrai?faytB, 228
Sodalilv, concentration of, 441
foyaitft, 439 B
syenite, 393, 411
Soembawn (Sumbawa) toIcboo, 426
Solomon Islands, 509
Solfatoric action, 276
558
INDEX
Solution aided by chemical contract,
217, 218
SolvsberKite, 391, 411
Sonora quadrangle, 496
Sooke gahbro, 454
Sorct principle, 239, 245
South Africa, dikes of, 80
fissure eruptions of, 120
quartz dolerites of, 317, 350
sills of, 60, 114, 241, 336, 350, 383
South America, alkaline rocks of, 516
syenite clan in, 501
South Australia, alkaline rocks of, 509,
524
granodioritic rocks of, 389
South Dakota, alkaline rocks in, 513
Southern Klondike Hills, 494
Spain, alkaline rocks of, 518
syenite clan in, 503
Spanish Peaks, California, 496
district, Colorado, dikes of, 81, 82
(map)
sill in. 64
syenite clan in, 491
Specific gravities of rocks and magmas,
38, 202, 261 ff
Specific gravity of lava afTecte<l by
vesiculation, 261 ff
Sphen(»lith. SS
Sphcn>itliil state, 338
Spiegel Iliver, 523
S|)ilite, 314. 330
Spilitir suite. 338 flf
Spines, volcanic, 130 ff
Spotted Fawn Creek, tinguaite of, A'M\
Spring. W.. 179
Springs, hot. 123. 431
Square Butte. Montana, 5:^, 76. 221.
234. 230 ft, 243, 438
Squeezing-out of residual magma. 2ir»,
445
Stability of earth's crust, 172, 20')
Standard bubble. 2m fT
Stansbury Range, 403
Starabba. S., 436
Stark, New Hampshire, 500
Steam-explosion an adventitious phase
of vulcauism. 287
Sterher. E.. 355
Steinheim basin. 2S4 ff
Stenhouse, A. O., 435
Stinkingwater Peak, 399
Stirlingshire, necka of, 128 ff
Stocks, 90
classified, 115
composite, 115
composition of, 114
cross-cutting contada of, W ff, lil6 ff
multiple, 115
relation to batholiths, 253, 462
satellitic, 361-362
Stokes formula for a unking iphcrr.
203,261
Stone, R. W., 121
Stonewall Mountain, 494
Stoping, arrested, 200
breccia, 462
in sills and laccoliths, 206
lateral, 141
magmatic, 190, IM ff, 107, 21M,
207,214,216, 30&-306, 404,461
underhand, 207, 219, 438
Stormbergen, Cape Province, 153, 265
Strain, level of no, 176 ff, 183
Stratification according to densty, 161,
167. 170, 227 ff, 304
Streaming in lava lakes, 255, 250 ff
Stresses due to differential haatiag, 199
Stromboli volcano, 268, 378-379 (aapi,
422
Stnitt. R. J., 175
Stulwl. A.. 301
Sturgeon I^ke, 487
"Subalkaline** a preferred dedgnalion.
414
Subalkaline clans, relatiya abundanee of ,
43. 49, 52
series of rocks, 13, 303, 413
dominance of, 49, 50
Subjacent bodies, 62, 89
chemical composition of, 113
replacement by, 109 ff, 196
SulK>rdinate volcanoes, 292 ff, 397, 903 ff,
303,460
Substratum, basaltic, 164 ff, 166^ 171.
183-185, 192, 300b 303-304.
311,315,361,458
ph^-sical condition of, 17L
magmatic, wrong crilerioB for, 300
Sudbury laccolithic sheet, assimilalioo
in, 242, 406
granodiorite of, 389, 391
gravitative differentiatioo at, 226,
230, 240, 347 ff (inap). 455
Sudbury laccolithic eheet, interfonna-
tional, 69
intermediate rockB of, 3S4
large size of, 330
norite of, 318
ore of, 454
stopingin, 206
Suess, E., 89, 161, 209, *21'J. 282-283.
296. 412
F. E., 481
Suldenite, 374.
Sulphides, magmatic, 226, 349, 455
Sumatra, quarts diabase of, 317
SunUDera, H. S., 391
Sun, composition of, IS8
temperature of, 159
SundaDce quadrangle, 4H, 451, 491, 513
Sunken calderas, 149-150
Sunlight intruBives, 398-aoii
Supercooling of magmas, 210
Superheat, magmatic, 210-212, 217,
242, 249, 273-274, 459
SilBsmilch, C. A,, 188, 237, 438, 483~t84
Sutherland, South Africa, 523
Svir diabase, 316
Swabia, maars of, 144
necks of, 130, 295 (mnp)
Swabian Alp, 619
Sweden, alkaline rocks in, 506, 520
anorthosite in, 331
batholiths in, 96, 98, V2i
diorites in, 383
eruptive sequence in, 3flfi, 470
Sweet Gruss Hills, 490
Swentna River, 489
Switzerland, syenite in, 5()3
Syenite. 68, 425
due to acidification, 215, 419
in eruptive sequences. 367
oridin of, 238, 243, 287, 312, 336,
346, 354, 393 ff, 395 ff, 405,
427-429, 437, 440-441
porphyry, 72, 74, 77, 414
relation to essejtite. 3!"(-397
to other alkaline clau'^, 414
svmall size of bodies, iOH
syngenetie with gabliro, 240, 334,
3:16-337, 397
Syenite clan, 393 ff
compared volumetrically with
clans bearing fejdspathoida,
400
iBX 559
Syenite clan, in relation to sediments,
403, 405, 408, 4S3
Syenites compared chemically with
trachytes, 409
Sycnitic- batholiths rare, 408
Symmonds. R. 8., 229
Syntectic affected by basic volcanics,
300,404
Synteotic-differentiation theory, 287
Syntectic-liquation theory, 307
Syntectics, 216, 221, 242, 318. 321,
382
sedimeiitary, 243, 312, 387, 391,
395, 414
Syntexis, 215, 243, 309, 356
Synthesis of minerals. 437
Syria, alkaline rocks of, 521
Tachylite, 79, 314
Tahiti, 54, 423, 451, 509, 525
Tammann, G., 200, 225
Tanqua Valley, sheet near. 66
Tarawera rif t, 146 (map)
Taixyall district, 493
Tarumai, plug-dome nf, 131, 133 (nap)
Tasmania, alkaline rocks of, 509, 524
Tavite, 411
Taylor, T. G., 237, 438
Toall, J, J. H., 12, 221-222, 237, 306,
324.420
Teanaway basalt, 119, 356, 459
Telluridc. Htoek near, 53. 101 (map).
110,383.390.491
Temperatures, at volranic vr-nts, 311-
212, 267
earth 'b internal, 157
in sills, 218
magmatic. 210 S
of erj-stftllittlion, 375-376
of high fluidity, 376
Tentcger vulimno, 151 (luap). I^^
Tcnmile District, Colorado, 72
Tenxion, increase of, in earth's rr<ist,
178
shell of, 176 IT, 18:i. ISS, 189, 192,
450
Tephriie. 410, 412. 423. 426, 4J2
Termier, P.. 480
Tesehen, silU of, 242, 451, 62U
Tescbenite, 239, 419. 421. 451
origin of, 242, 435-43G
TelTahedral theory of the eurlh, 102
560
ISDEX
Texas, alkaline rocks of, 422, 514
granites of, 98
syenite clan in, 498
Theory, value of petrogenic, xjrtt, 45C,
465
. Theralite, 76, 239, 397, 410, 429, 450
origin of, 405, 414 ff
total area of, 50
Thermometer, geological, 214
Tholeiite, 314
Thomas, A. P. W., 146
H. H., 472
Range, 493
Thomsen, J., 271, 278
Thomson, J. A., 317
Thorrodsen, T., 79, 81, 118, 119, 120,
135, 289
Thousand Islands region, batholithic
replacement in, 111
syenites of, 367, 499
Three Forks quadrangle, 490
Thugutt, S. J., 434
Thunder Bay, sills of, 233. 241, 323, 325
Tidal stress, cause of abysnal fissures,
182, 185, 192
reviving of volcanic activity, 279
Tierra Blanca, 498
Tillo, A. von, 119
Time-scale, eruptive types in relation
to, 55, 57
Timiskaming disttrict, sills of, 339
Timor, 525
Tinguttite, 68, 411, 427,
Tintic, Utah, 398-399, 493
Titanic oxi<ie in alkaline rocks, 43(>-4:i7
Titanifcrous iron ores, 454
Titanitc in alkaline nx-kn. 436-437
Tolman, C. F., 179
Tonalite. 374. 385-386
aplitc, 374
Tonalitic* zone of the Alps, 190
Tongues, intrusive, 83
Tonopah. lavait at, 2S8
T6n.sl)ergitc, 394
Tordriliito, 341
Tornc Triusk, 5(H3
Tornolmhm, A. E., 321
Tnu'liyjiiulcsitc. 374, 394
Trachydolcritc, 2S9. 411, 424, 426, 452
Trachyte, 394. 419, 423-425, 427
compared chemically with syenite,
401
origin of. 395 ff
Trachytic magma, 165
Trail bathoUth, 201, 366, 463
Transfer, gaseotis, am Qmmom
Transgreasive injactionB, 63
junctions in batholitha, 306
Transition rocka, 238, 241 , 840, 344, 347,
361, 306, 398, 402, 421, 436
general absence of, 216
Transvaal, laccoUth of the, 360, 437, 455
Trass, 401
Treadwell Mine, 480
Trebisond, breccia off, 427', 521
Tres Hermanaa, 408
Treuen granite stock, 106 (map)
Trimineralic magma, 448
Trinidad Island, 526
Tripyramid Mountain, 367, 306, 474,
500
Tritriva crater, 140
Troad, 521
Troctolite, 313
Tnickee quadrangle, 105, 366
Trusentbal, Thuringia, 77 (map)
Tsang province, 506
Tschandy Bei, 521
Tschermak, G., 268
Tulameen district, 488
Tumulus, volcanic, 133-184, 202
Turkestan, nephelite ^y«aite ol, 484,
437, 521
Turner, H. W., 478
Turquoise Mountains, 407
Tuscan district, 518
Twin Butte, Cok>fmdo, 402
Peak, Utah, 403
Two Buttes, Colorado, 402
Tyrol, subjacent bodies of, 06
Tyrrell, G. W., on aasimilation, 242, 351
on contact chilling, 230
on differentiation, 438, 451
on Kilsyth-Croy diatriet, 236, 351
on Lugar, 230, 451
on quarts diabaee, 317, 354
on sills, 236-237, 451
I'ani, Nicaragua, 516
rhlig, v., 190, 243, 436, 481
I'inta Range, 493
Ul t ra-fomic rocks, explanation ol tmnXj^
444. 452
riu Kawas, 510
Tmptekite, 303, 306
I'mqueme Mountains, 310
United States Geological Survey, 38,
43, 52, 461
Upolu Island, 525
Urach region, SwabiA, 295 £F (mup)
Ural Mountains, peridotite clan in. 44S
Bubjacenl bodies in, 95
Uranium, energy potentialised in, 272
Urtini Highlands, 389, 507
Urtite, 76, 411,414
Ussing, N. v., 71, 114, 1B4, 237, 305,
417-418, 439 IT, 483
Uau-San volcano, 298-300 (map)
Utah, granodioritic rocks of, 388
syenite clan in, 493
Ute Creek, 497
Uvalde county, alkaline rock.' of, 514
Mountains, 422
Vadose waters, 249, 283, 285, 28S
Vancouver Island, anorthoaite in, 323
geosynclinal in, 187
granodiorite in, 3SS-389, 3'J9
hornblendite in, 454
Van Hise, C. R., 75, 114, 33(1, 323.
328-329, 347, 469
Vapor transfer, 454
Variability of alkaline types, 410, 41.'].
431-432, 444
of dioritic types, 384
of syenitic types, 399
of vokanic rocks, 287, 292, 372, 444
Variolite, 314
Veidivatnahraun lava flow, 120
Veins, contemporaneous, 83
feldspar, 437
Vdain, C, 142, 150, 228, 378
Velay, the alltaline rocks of, 517
cone cluHt«ra of, 137 (ma|i)
crust fracture, 138 (map)
eruptive sequence in, 444, -If
phonolites ol, 433 (map)
rock association in, 424
Hyenite clan in, 503
Venetian district, 518
VenlH, alignment of volcanic,
404
302,
cenlral, moohanism of, 24S IT, 273,
292 ff
heat problem of, 273, 302
indepcndenceof volcanic, expliiirteU,
292 fT, 300
5BX 561
Vents, long lives of central, 248, 254,
262, 268, 274
opening ftnd localization of volcanic,
250, 275, 281
origin of volcanic, 251 ff
revival of activity at, 275
small size of volcanic, 127. 120 IT,
256. 26fi, 271, 27S, 280, 282
Verbcek, R. D. M., 138, 144, 149, 151,
153, 421, 426, 483
Verite, 412
Vermilion Creek, Idaho, 490
Vermont, alkaline rocks of, 515
syenite clan in, 5(X>
Vesbian district, 518
Veeiculation in lava, 260 ff, 2G4 IT. 2ii7,
273, 452
Vesuvius, 285, 432
dormancy of, 275
lava Countuins at, 268
nested craters of, 143
periodicity of, 280
present crater of, PI. II
two-phase connection at, 2fi8
\'i<'toria, Australia, alkaline rocks ol,
509, 524
eruptive sequence In, 444, 4S4
granodioritea oT, 389, 391
^'illoriat-Gutin Mountains, 504
Vintlite. 374
Viola, C, 237, 448
^'i^ginia, syenite in, 498
Dale, 496
Range, 495
\'i'H'ONity, magmatic, affected by gas,
288
in b&tholiths, 200, 204, 210
in laccoliths, 70, 342
in sills. 206, 242
of alkaline ma«maH, 241
of anorthoeitic magma, 322, 335
of lavaa, 203, 271, 289, 376
of subsUatum, 172, 210
Viti iBland. 525
Vitrophyre, 342
Vitrophyrite, 386
Vizagatapatam, 507
VogcUberg, 505, 519
Vogesitic rock, origin of a, 300-3ti7, 462
Vogt, J. H. L,. 199. 209, 210, 222. 235,
277, 360, 454
Volatile matter, fluxing by, 212
562
rXDEX
Volcanic activity, continuance of, 250,
251, 254, 262. 270, 286, 302
recurrence of, 275, 279
bodies and forms classified, 125 fT,
140 fr
islands, alkaline rocks of, 417
Volcanics contrasted with correspond-
ing plutonics, 19 flf, 39, 229
Volcanoes, principal, 300, 302-303
subordinate, 292 ff , 297. 300 ff, 3a3
Volhynia, anorthosite of, 323, 334
Volume, change of, before cr>'Bt&llixa-
tion, 226
Volume-change in fusion, 202
Volumes of igneous species, 45, 47, 50,
52, 413
Vordoreifel, 519
Voss-Sogn district, anorthosite of, 323,
334
Vulcanello, 422
Vulcanism, accompanying geos^oiclinal
down warping, 186, 193, 459
following mountain-building, 191
originating in satellitic injections,
291 ff
Vulcano Island, 422 (map), 482
Vulkanembrj'onen, 295
Vulsinian district, 518
Vulsinite, 39 1
Wahl, W., 317
Wah-wcah Range. 495
Walcott. C. D., 1S6
Walker. T. L., IS
Walls of subjacent Ixxlics, 1(X), KKi
Walsenburg quadrangle, 497
Waltershausen. W. S. von. 143. 290
Ward district, 495
Waring, G. A.. 119
Warm Spring laccolith. 70, 71
Warren. H. X.. 272
county. New York, stock in, 401
Warrumbungle Mountains. 435, 508. 523
Wartenberg, melilite basalt of t^e. 436
Washington, H. S.. 10. 17, is, HVS. 2(H»,
457, 474, 5 IS
Statw. basalts of. 119
batholithsin. 9S, 3SS. M)}
ti>sure eruptions of. 121, 191,
35t), 45s
geosynclinal in, 187
syenite clan in, 490
Water, in amphibole, 320, 381, 401
in mica, 381, 401
in syntectics. 437
sources of volcuiie, 249, 209, 282 IT.
287
Weber, H. F., 198
Websterite, 446
Wedges, abyssal, aasimiUtioii in, 207.
216, 244, 384
illustrated, 183, 185
in Cordilleran regioii, 464
origin of, 181 ff, 189 ff, 305
relation to batholitha, 240
shattering by, 198
stoping in, 305
temperatures of, 211, 249
Weed, W. H., on a|^t6,300
on Barker Mountain, 73
on Bearpaw MountaiiiB, 408
on corundum, 434-435
on Klkhom district, 470
on Judith MountaiiiB, 08, 70, 74-
75
on Montana stoclu, 304-300
on transition rocks, 399
on underhand stoping, 207
Wehrlite, 446-448, 452
relation to gabbro, 447
Wehrlitic type at Kilauea, 76
Weidman, S., 428, 469
Weimam, P. P. von, 225
Weiselbergite, 374
Wescr-Werra-Fulda district, 519
Westerwald, alkaline rocks in, 504,519
eruptive sequence in, 481
West Indies, Tertiary eruptives of. 413
Kootenay district, batholitha of, 96
Wettem Lake, sills near, 231, 354
Wheaton River district, 389, 480
Wliin sill. 66, 81 (map)
White, I. C, 268
Horse River, 489
Ircm Lake, 499
Pine district, 405
Whiteoaks district, 496
Wiechert, E.. 160
Williams canyon, fissure eruption at,
119, 120
Willis. H.. 188
Wiiichell, H. v., 325
N. H., 209, 236, 242, 325, 330, 347
Winge. K., 78, 356
Winnemucca Peak, 405
Wisconsin, alkaline rocks of, 42S, '>H
eruptive sequence in, 396, -IG!)
syenite clan in, 499
Wisharde Peak sill, 231,
Witwatererand geosynclinal, 18S
Wodehouse district, necks of, ViO
Wolff, J. E., 76, 194
Wollsstonite in alkaline rocks, 436—13!
Wright. C. W., 383
F. E., 13,214,383
W. B., 415
Wyja-Teich, Urals, 507
Wyoming, alkaline clans in, SVi
anorthosiI« in, 323, 334
syenite clan in, 491
Wyotningite, 412
Wyssotzky, N,, 448
Xenoliths, 197,200
corrosion of, 353
prove high magmatic
sinking of, 201, 203
why relatively small, 204
a basalt, 119, 459
3BJt £68
Yamaaka Mountain, 4'M, nil
'^'amaskite, 307, 444
Yankalilla, 600
Yellowstone Park, andeaites of, 380
batholith in, 08
de-roofing at, 122, 2U5
eruptive sequence in, 470
geysers of, 122-123 (map)
lavas of, 380-381 (map)
quadrangle, 405
rhyolite of, 49, 123. 203, 381, 463
Yenisei district, 427, 507, 522
Yentna River. 489
lentnite, 374
■^'|^go canyon, dike at, 207
Peak stock, 402
Yoahiwara, S., 139
Young, G. A., 444
% Mountain, 494
Yukon, granodiorite stocks in. 388-389
e clan in, 480
Zalas. 505
Zurafshan. nephelite syenite i
521
Zirkel, F., 12, 40, 221, 317, 321, 375
Zululand, sills of, 234, 319, 323, 325
r
"N