BULLETIN OF THE GEOLOGICAL SOCIETY OF AMERICA VOL. 66. PP. 405-426 APRIL 1956

DISTRIBUTION OF IGNEOUS ROCKS IN SPACE AND TIME

By G. W. TYRRELL

ABSTRACT Consanguinity in an assemblage of rocks, associated in time and place, is held to indi- cate that it has been derived by a more or less uniform geological process from a common initial magma. The most fundamental grouping of igneous rocks is thus based on distribu- tional relations, which are in turn dependent on geological process and event. The close association of igneous activity with earth movement is well known. The repetition of petrographic "provinces" or "kindreds" in time and place is clearly a mani- festation of the cyclic character of tectonic events in geological history. The tectono- igneous cycle is elaborated and discussed with special reference to Scotland and North- west Europe, and comparisons are made with eastern North America and Indonesia.

CONTENTS TEXT Page Page Plutonic and volcanic associations 417 Introduction 405 Nature and origin of igneous cycles 420 Consanguinity 406 References cited 424 Kindreds of igneous rocks 407 Petrographical provinces and periods 408 TABLES Igneous action and tectonics 409 Table Page Geological cycle 409 1. Correlation of the .igneous and tectonic Diastrophic subdivision of the crust 409 phases of the Tertiary and Quaternary of Diastrophic (tectonic) cycle 410 Indonesia with those of the Paleozoic of Igneous cycle, with special reference to Scot- Scotland 418 land, Northwest Europe, and eastern North 2. Tectono-igneous cycle 420 America 412 3. Comparison of the igneous phases of the Quartz dolerite dikes and sills of North Britain. 415 Caledonian, Hercynian, and Alpine oro- Tectono-igneous cycle in Indonesia 416 genic cycles 421

INTRODUCTION think were characteristic igneous rocks have been relegated to the category of metamorphism My assignment today is to introduce the (plutonism in Read's sense) and with, I think, subject of your deliberations—The distribution some justification. With some reservations I of igneous rocks in space and time,—-a subject am disposed to accept Read's (1948, p. 3) which, while it concerns high-temperature classification of rocks as neptunic (mainly materials, will probably not generate so much marine sediments), volcanic (magmatic, igneous, heat as the symposium 4 years ago introduced dominantly effusive and basic in composition), by my friend Professor H. H. Read. It is, how- and plutonic (metamorphic, migmatitic and ever, a subject which has attracted much at- granitic, of deep-seated origin). The last- tention in the past 60 years and has given rise named group also includes, beside granite, to a voluminous literature. Nevertheless it many formerly described as plutonic igneous has not yet advanced far beyond the stage of rocks. fact collection, and the important generaliza- Because of the doubts about the origin of tions which I feel are inherent in this field on many plutonic rocks I am now disposed to rest the relations of igneous activity to tectonics generalizations on the distribution of igneous and its correlation with the geological cycle are rocks in space and time mainly on the un- still largely to be made. doubted igneous rocks which have clearly been In recent years many of what we used to derived from liquid or quasi-liquid magmas, 405

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 406 G. W. TYRRELL—IGNEOUS ROCKS IN SPACE AND TIME

together with the intrusive rocks, sometimes of This view is confirmed from the etymological quite coarse textures ("phitonic"), which are point of view by my colleague Professor A. W. associated with them—that is, on what W. Q. Gomme1, to whom I applied for information Kennedy (Kennedy and Anderson, 1938, p. 25) as to what the term magma meant to the ancient has called the volcanic association. But I still Greeks. think that consideration of many of those types It is no answer to say that salt, clay, and now assigned to Read's plutonic group can certain metamorphic rocks must therefore be help us toward the above-desired end. regarded as magmatic because, under some By a curious coincidence some groups of circumstances, they seem to conform to the lavas have a very close chemical resemblance above definition. The writer maintains that to plutonic groups, as rhyolites with granites; rock-flowage materials can readily be dis- dacites, latites, and andesites with members tinguished from magmatic when the associated of the granodiorite - tonalite - quartz - diorite phenomena are taken into consideration. A group; trachytes with alkali syenites; phono- perfect transition must exist, and many Trans- lites with nepheline syenites; and basalts with formists are willing to admit it, between just- gabbros. I cannot help thinking that in at mobilized granitic material and granitic magma least some cases there is a genetic connection which might consolidate as a granitic vein or as between lavas and the corresponding plutonic a flow of rhyolite; but we are not called upon rocks. And this leads me to question the com- to fix an arbitrary boundary between mobilized plete divorce in geology, petrology, and genesis material and magmatic material any more than of the volcanic from the plutonic rocks ad- we are bound to fix a point at which white ends vocated by Kennedy (Kennedy and Anderson, and black begins in the perfect transition of 1938, p. 31) and Read (1948, pp. 3-4). It is shades of gray which intervenes between white true that both authors concede that certain and black. Common sense applied to the ob- intrusive dolerites, gabbros, and other "plu- servation of relevant facts must decide such tonic" rocks may sometimes be included in issues, and there is room for different opinions. volcanic associations. Discussion as to what are and what are not CONSANGUINITY igneous rocks has been hampered by an un- necessarily restricted definition of magma as It is a generally accepted fact that certain "completely fluid rock substance" (Read, 1948, groups of igneous rocks show various degrees of p. 5), and F. F. Grout (1948, p. 46) has rightly family likeness (Judd) or consanguinity (Id- animadverted upon it. If this definition is ac- dings). This community of character is ex- cepted literally, practically the only truly pressed in their chemical, mineral, textural, magmatic rocks are the pure glasses like ob- and geological features. While consanguineous sidian. When examined after sudden quenching suites may range in chemical composition from even the basaltic liquids of the lava lakes acid to ultrabasic, may exhibit many kinds of (e.g., Halemaumau) are seen to have had a textures, and may consist of extrusions or in- few crystals in them, and there are plenty of trusions or both, some mineral and chemical instances of both effusive and intrusive action characters, which may be described as con- in which the magmas concerned were rich in stant, are common to practically all members. crystals. The writer knows of instances of Other characters are serial and show a regular the Lower Carboniferous plagioclase basalts variation throughout the group which may be (Markle type) of the Midland Valley of Scot- illustrated by suitable graphs. Thus, in some land in which the feldspar crystals exceed 50 1 Professor of Greek, University of Glasgow. per cent by volume of the rocks, and are some- Professor Gomme says tie word comes from a root meaning to knead, especially to knead in a mould times so clotted together as to simulate an as, for example, a kind of barley cake which was anorthosite, but which yet form lavas and small not baked but eaten in the soft state. The past tense (kneaded) was also used of a poultice. The intrusions. It would be more accurate to say term magma itself is actually found only in some that a magma is a mush of liquid and crystals Greek medical literature for a thick unguent; and that is capable of effective intrusion or ex- Webster's New International Dictionary, 1st. Ed. (1932 issue, p. 1295) confirms that it is still trusion,—a material something like porridge in used in this sense, but only rarely, by pharmacolo- which crystals correspond to the oat grains. gists.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 CONSANGUINITY 407

suites, a constant chemical character is over- kindred is here used for the widest and most saturation with silica; highly silica ted feld- tenuous kind of family likeness as that, for spars and quartz appear in small quantity even example, between the members of the alkalic in some of the basic members. Regular varia- or calcic kindreds. The writer further suggested tions of silica, alkalis, ferrous iron, magnesia, that the progress of research might reveal that and lime may serve as examples of serial char- kindreds included tribes, the rocks of which acters. My late friend and colleague, M. A. would show a closer relationship than those of Peacock, indeed founded a very successful kindreds; and that tribes might be found to be criterion for the discrimination and comparison composed of clans, the members of which would of igneous series on the "lime index"-—a rela- show the highest degree of consanguinity of tion between silica, alkalis, and lime. Some all. Both the alkalic and calcic kindreds, if in- suites or series may be characterized through- deed they are true kindreds, probably contain out by a peculiar mineralogical feature, such several tribes and clans. as the occurrence of anorthoclase in Norwegian, By their nature kindreds, tribes, and clans East African, and Antarctic suites, or the oc- cut across the more formal groupings and classi- currence of orthorhombic pyroxenes in even fications, and certain abundant rock types such the acid members of the so-called charnockite as basalt, andesite, and "granite" (if I may series. Further, whatever views we may hold use this term to denote an igneous rock) may regarding the origin of micropegmatite in cer- recur in several different groupings. A kindred tain dolerite and gabbro sills and dikes, its will be composed of a wide range of igneous presence together with some characteristic rock types, but all or many are marked by geological features in great basic suites of world- common chemical and mineral characters which wide distribution may be cited as yet another may be exhibited in constant or serial relations. criterion of consanguinity. Other factors may serve to establish the in- Consanguinity in an assemblage of igneous dividuality of kindreds. They may be char- rocks, together with spatial and temporal acterized by the different bulk development of propinquity of its members, may be held to their component types. Thus, in one kindred, indicate that it has been derived by a more or andesites greatly predominate in the range ba- less uniform geological process from a common salt—pyroxene andesite—hornblende-mica-an- initial magma, or from a number of closely re- desite—dacite—rhyolite; while another consists lated magmas, however such magma or magmas almost entirely of basalts (e.g., basalt floods). have been generated. Futhermore, there are differences among kindreds in regard to the relative bulk of KINDREDS OF IGNEOUS ROCKS "plutonic," hypabyssal, and volcanic develop- ments within the accessible crust. While this The suggestion the writer made in 1926 feature may, of course, be due to accidents of (Tyrrell, 1926a, p. 136) that, "Groups, series geological environment, erosion, and exposure, or suites of igneous rocks which show con- the writer believes that there are some intrinsic sanguineous chemical and mineral characters, differences in this respect. Finally, kindreds and which appear to be genetically related", may be distinguished by their extension in should be called kindreds has not been taken space and time (petrographic provinces and up to any extent by petrographers, but it has periods) and by their varying relations to value in relation to the subject of the distribu- tectonic processes. These topics will be further tion of igneous rocks. The term kindred in developed in later sections of this paper. In igneous petrology has somewhat the same Principles of petrology (1926a, p. 137-141) the meaning as the term fades in stratigraphy and sedimentary petrology. Several terms have 2 The terms kindred, tribe, and clan are based on been used in much the same sense as, for ex- analogy with those used for peoples and languages. For example, the Aryan kindred comprises (or is ample, series, suite, tribe, clan, branch, and supposed to comprise) tribes (Celtic, Hellenic, stem (stamm), but none appears so suitable as Teutonic, etc.), and the tribes include clans with kindred2 for the major groups of igneous rocks. a closer degree of relationship than obtains within the other two categories. The terms kindred, tribe, The relationship implied by consanguinity may and clan are therefore here used in the proper hier- be of different degrees of closeness; the term archical order.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 408 G. W. TYRRELL—IGNEOUS ROCKS IN SPACE AND TIME is likewise its extension in time" (Tyrrell, 1926a, writer describes and discusses certain kindreds p. 141). regarded as having been tentatively established, but it is unnecessary to go furtehr into that A petrographical province therefore, like ex- subject here. The systematization of kindreds, istence itself, is a space-time concept. The tribes, and clans has barely begun. In all prob- boundaries of petrographical provinces and ability the necessary data are at hand in the periods are necessarily vague and ill-defined vast accumulations of petrographical litera- because the true spatial and temporal associa- ture which have piled up during the last 30 or tions of a kindred are with a particular geo- 40 years, ready for the master mind and pen logical environment and tectonic process which of a petrological Suess. also have ill-defined boundaries in space and time. It is therefore better to approach the con- PETROGRAPHICAL PROVINCES AND PERIODS cept of petrographical provinces by placing the emphasis more on the nature of the kindred In the writer's opinion the most fundamental itself than on its mere geographical or temporal grouping of igneous rocks, as for all rocks, is in extension. For this reason the geographical relation to geological process and event. In names which have been applied alike to petro- effect this means a grouping based on distri- graphic provinces and to the assemblages called butional relations. The idea of petrographical kindreds, such as Atlantic for alkalic kindreds, provinces was dimly foreshadowed by several Pacific for calc-alkalic kindreds, Arctic (Von writers in the middle of the nineteenth century, Wolff, 1914) for the great basalt floods (surely but, as Lowell says, "the credit for an idea be- a most unsuitable term!), and Mediterranean longs to him who utters it clearest and best", (Niggli, 1921) for certain assemblages char- and the credit for this idea therefore belongs to acterized by richness in potash, are to be dep- J. W. Judd. Judd's definition (1886, p. 54) is: recated as useless and misleading. "There are distinct petrographical provinces within The very wide and vague but, the writer which the rocks erupted during any particular believes, valid grouping of igneous rocks into geological period present certain well-marked peculiarities in mineral composition and micro- alkalic and calcic has been much exploited in the scopical structure, serving at once to distinguish past in connection with petrographic provinces. them from the rocks belonging to the same general It is highly probable that both terms cover group, which were simultaneously erupted in other petrographical provinces". several distinct kindreds, and that some kin- dreds pass over the gradational boundary be- A. Marker (1911, p. 3), with the superior liter- tween the two groups, thus being "mixed" ary ability which distinguished his writings, in the sense that they contain both alkaline improved on Judd's definition with the follow- and calcic types. But, despite Whitman Cross's ing: animadversion on this point (1912), this very "A petrographical province is a more or less character is one of those which distinguish cer- clearly denned tract within which the igneous rocks tain kindreds as, for example, those of the belonging to a given period of igneous activity present a certain community of petrographical character Scoto-Irish Tertiary region and the Hawaiian traceable throughout all their diversity, or at least volcanic field. Barker (1911, p. 5) greatly over- obscured only in some of the extreme members of the assemblage." emphasized the importance of the alkalic-calcic contrast in using such phrases as: "A given In both these definitions the phrase italicized province is either of calcic or alkalic facies typi- (by the present writer) insists on a time limita- cal members of the two branches not being tion. In other words a petrographical province found together"; and again: "It is possible to has extension in time as well as in space. Many map out the active parts of the earth's crust objections to the idea have been based on the into great continuous regions of alkalic rocks mistaken notion that the term petrographic on the one hand and of calcic rocks on the province applied to all the igneous rocks within other". This perhaps merited Cross's severe a given region regardless of their ages. strictures (1912, p. 366-367); but in justice to The writer's definition, crystallizing the idea Marker it must be pointed out that he further in terms of kindreds, is that said (p. 4): "A petrographical province is the geographical ex- "This two-fold division of igneous rocks is, of tension of a kindred, and a petrographical period course, in no wise a final or exhaustive treatment of

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 PETROGRAPHICAL PROVINCES AND PERIODS 409

the subject; but as a first step towards a natural or although one must not forget its hypothetical genetic classification it seems to be established be- yond question". nature. W. Q. Kennedy (Kennedy and Anderson, GEOLOGICAL CYCLE 1938, p. 25-29) has divided the igneous rocks The concept of the geological cycle expresses into plutonic and volcanic associations and has the idea that geological events fall into more or related them to erogenic and nonorogenic less regular sequences which may be repeated regions of the earth's crust respectively. This at more or less regular intervals; but in the concept is clearly relevant to the ideas of kin- writer's view the geological cycle is com- dreds and petrographic provinces. He also pounded of five parallel and correlated sub- pointed out that, as the erogenic zones in time cycles of tectonic (or diastrophic), igneous, become welded to the adjacent stable regions metamorphic (or plutonic), sedimentational, of the earth's crust, and then participate in and erosional events, and it is the integration their epeirogenic reactions, it is natural for of all into a connected whole that constitutes different kindreds to be successively established the geological cycle. within the same region. Thus the view that Nature, like history, never repeats herself— petrographic provinces should have time limita- exactly! Or, to put it in the more usual way, tions is confirmed from another angle. The Nature, like history, repeats herself—with writer defers further reference to Kennedy's variations! While geologists recognize that the views until discussion of the tectonic aspect of events of geological history repeat themselves the problem. again and again in a more or less regular se- IGNEOUS ACTION AND TECTONICS quence or rhythm, they do not expect this rhythm to be exactly repeated or invariable. That igneous activity is closely associated In other words the geological them has varia- with earth movement is one of the fundamental tions imposed upon it. A fairly exact analogy observations of geological science, but the is the musical Theme with Variations. Thus nature and laws of the connection are still Brahms, in his "Variations on a Theme by only little understood notwithstanding recent Haydn", takes that composer's noble tune, contributions to the subject by Kennedy and states it simply at first, and follows with a Anderson (1938), A. Holmes (1932), H. H. number of variations in which, however, the Hess (1938), W. H. Bucher (1933), A. Rittmann attentive ear can always discern the underlying (1936), and many others. For further advances theme. The analogy breaks down, of course (as we need to know much more of kindreds and all analogies eventually must), when the com- petrographic provinces and of their correla- poser turns the tune completely upside down tions with earth movements. Closer study of or inside out, as he occasionally does! The these involved questions must still await more geological cycle is a complex conception con- exact discrimination of types and sequences of sisting, as it does, of several interwoven sub- earth movements and of their significance in cycles, but it can be resolved in comparatively the economy and history of the earth's crust. simple terms, especially if the component sub- Even more needed is a deeper understanding cycles are described separately. of the igneous rocks (in a double sense!), of the criteria for their natural grouping, and of DIASTROPHIC SUBDIVISION OF THE CRUST the chemical composition and variations of kindreds, tribes, and clans. The close associa- The term diastrophism covers the sum total tion of igneous activity and earth movement in of earth movements; and, for the purposes of space and time makes it natural to assume a this paper, diastrophism can be roughly divided causal connection between them or, better, into epeirogeny and orogeny. Vertical move- that they are concomitant effects of the same ments in the oceanic sectors have been termed causes operating in and beneath the crust. The thallatogenic by Kober (1921), but the writer work of A. Holmes (1928) on subcrustal cur- regards these as coming under the heading of rents induced by radioactive heat in relation to epeirogenic. No attempt can be made here to tectonirs and igneous activity constitutes a distill the essence of the vast recent literature penetrating exposition of one of these causes, on diastrophism and tectonics; only that which

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 410 G. W. TYRRELL—IGNEOUS ROCKS IN SPACE AND TIME

is considered relevant to igneous action is between comparatively deep subsidences (geo- selected. synclines) and high fold mountains. The incidence of diastrophism varies both He holds that the diastrophic relations of these in time and space. In a given region of the four crustal elements are fundamentally dif- crust orogeny is episodic and may entirely ferent and find expression in their diverse cease for long periods. Epeirogeny, however, magmatic and sedimentational associations. is long-continuing and probably never entirely While kratogens such as the continental ceases. There seems to be a diastrophic rhythm; nuclei remain approximately in constant posi- epeirogenic movement rises to a maximum and tion (pace Wegener!), the sites of the orogens falls to a minimum. Maxirftum erogenic move- which develop around them shift from time to ment and igneous action appear to coincide in time. Old orogens become tectonically "dead" time with the epeirogenic crescendo. Periods of and are so welded onto the adjacent kratogens maximum diastrophism have been termed as to form integral parts of them, acquiring orocratic by W. Ramsay (1924). Orocratic kratogenic characters and reacting in a krato- periods are relatively short and are separated genic manner to subcrustal causes of movement by pediocratic periods in which epeirogenic in succeeding geological cycles. Northwestern movement sinks to a minimum and orogenic Europe affords an excellent example of this movement often does not occur at all. Joly phenomenon for, around the Fenno-Scandian estimated the general ratio of duration between Shield, the loci of orogenic activity in successive orocratic and pediocratic periods at 1:3. cycles since the Precambrian (Caledonian, It has long been recognized that there are Hercynian, Alpine) have shifted gradually strong, rigid, stable portions of the earth's from the northwest to the southeast and south, crust (shields) that have not suffered folding, and each orogen, in its turn, has been welded or any diastrophism except gentle vertical onto the circum-Baltic kratogen. movements, since very early geological time. The above and later remarks are amplified At the same time there exist certain narrow, from the writer's incomplete paper of 1937 elongated, subsiding areas (geosynclines) in presented to the Seventeenth International which, or near which, thick sedimentation Geological Congress at Moscow which was followed by orogeny takes place at intervals. published in English in Volume I of the Report At any given period, according to Kober (1921), of the Seventeenth Session dated 1939, but the earth's crust may be divided into kraiogens which only reached the western geological world (cratons in American literature), strong rigid in 1948. blocks some of which approximate continental dimensions, separated from one another by DIASTROPHIC (TECTONIC) CYCLE orogens, narrow, elongated, weak, labile zones whereon, in the fullness of time, fold mountains A very simple conception of the diastrophic arise. This contrast is fundamental to our sub- cycle, sufficient for much of our present purpose, ject as the igneous, sedimentary, and other is that it begins with the breakdown of the cycles are widely different in orogens and krato- broad high continents ridged with fold moun- gens. tains, which represent the climax of the previ- S. von Bubnoff (1923, p. 4), while adhering ous cycle. The earth's crust begins to sag off to Kober's contrast of kratogens and orogens, the edges of the continents into geosynclines attempts a more detailed tectonic classifica- which are weighted with immense loads of tion of crustal elements, two of which are stable sediments derived from the erosion of the ad- and two variable: jacent highlands. At the same time the con- (1) Permanent continental blocks with a tinents become lower and smaller, the oceans steadily rising tendency broader and shallower, while the geosynclines (2) Permanent oceanic sectors with a steadily continue to deepen until some equilibrium is subsiding tendency attained between the operative and opposing (3) The slightly mobile continental shelves, forces. Toward the end of the geosynclinal varying constantly between shallow seas and period the oceans, filled with sediments, be- lowlands come very broad and shallow, spilling over the (4) The highly mobile orogens alternating lowland continental margins in great trans-

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 DIASTROPHIC (TECTONIC) CYCLE 411

gressions. The continents, now small and low, continents and are broad and relatively shallow. contribute little but fine argillaceous sediments Rivers bring down only small amounts of to the seas, wherein the main sedimentation is muddy sediments to be deposited as shales in organic (limestones, etc.)- Meanwhile tectonic warm shallow seas swarming with life, where forces have been gathering strength beneath the they are augmented by limestones due to the seeming quiet and, toward the end of the cycle, accumulation of remains of the abundant break out, with perhaps one or two preliminary marine organisms. Epeirogeny, orogeny, and spasms, in revolutionary power, buckling the igneous activity are minimal or absent, but geosynclinal sediments into folds which become toward the end of the phase diastrophic unrest lifted above sea level, first as island arcs, and begins again, to culminate finally in the suc- then as great fold-mountain ranges. Hence the ceeding phase. diastrophic cycle may be held to consist of: (3) Orogenic (Emergence) phase, beginning (1) A geosynclinal or evolutionary phase, oc- with regression of the seas from the old con- cupying perhaps three-quarters of the cycle, and tinents, associated with intense compression, (2) An orogenic or revolutionary phase, much folding, and mountain building along the former shorter and comprising both orogenic and geosynclinal zones. This is accompanied and epeirogenic movements followed by vertical movements giving rise to As an aftermath of the orogenic phase, both broad continents surmounted by fold-moun- outside and within the orogenic arcs, the earth's tain ranges, with great volcanic activity at the crust may break into large blocks which suffer surface, and plutonism (metamorphism, graniti- vertical movements along deep-reaching major zation, and magmatic granite) in the depths of faults. This usually begins during the waning the orogenic zones. stages of the orogenic phase or somewhat later (4) Continental phase, during which orogeny and ushers in a new diastrophic cycle. is in abeyance, the continents attain their Many igneous cycles fit in well with this maximum size and remain comparatively stable. simple version of the tectonic cycle. A more There is still some vertical movement and con- elaborate scheme is that by R. A. Sender siderable volcanic activity. Extremes of climate (1922, p. 242-247); based mainly on tectonic result in the formation of great ice sheets and events in Northwestern Europe, it fits the extensive deserts with the deposition of glacial writer's particular field of study very well. and "red-rock" sediments. Sender divides the diastrophic cycle into four The use of Sender's scheme in this connec- phases3: tion does not imply its full acceptance, or that (1) Geosynclinal phase, in which epeirogenic it is applicable all over the earth. Von Bubnoff movements become strong, leading to alterna- (1924, p. 173) has criticized it on the ground tions of continental uplift and oceanic sub- that, far from being a generalization from world sidence. Geosynclines come into existence near geology, it is really based on the sequence of the continental margins which, as they deepen, geological events in Northwestern Europe and are kept filled by sediments eroded from the eastern North America, both highly complicated adjacent highlands. There is a resurgence of and disturbed regions in which the sequence of igneous activity of ophiolitic type within the events may not be the normal one. Further, it geosynclines. Toward the end of this phase does not take into account the fact that the there may be small preliminary orogenic spasms earth's crust consists of dissimilar units, with bringing island arcs to the surface over the different histories, and most probably with geosynclinal areas, and accompanied by vol- differing and nonsynchronous geological cycles. canoes yielding andesitic and rhyolitic lavas. Moreover, geological cycles do not always coin- (2) Transgression (Fluctuation) phase. The cide with the standard geological eras. The continents are now worn down to minimum great Russian Platform, for example, has re- size and height, and climate has become more mained almost undisturbed since Precambrian or less equable over the whole earth. The oceans time, and must have experienced a quite dif- have transgressed widely over the edges of the ferent cycle of events from those of geosyn- 3 Sonder begins his cycle with the Continental clinal and orogenic regions such as western Phase. The writer prefers to begin it with the Geo- synclinal Phase for reasons which will appear later. Europe and eastern North America. Despite

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 412 G. W. TYRRELL—IGNEOUS ROCKS IN SPACE AND TIME

these criticisms Sender's cycle does seem to fit Sender's diastrophic cycle is peculiarly suit- events in at least the Paleozoic of northwestern able, for reasons given above, as a frame for Europe, and seems therefore best adapted as discussion of the Paleozoic igneous succession the frame within which the sequence of igneous in Scotland, supplemented by reference to the events can be discussed. corresponding Scandinavian and North Ameri- can igneous fields. The use of eastern North IGNEOUS CYCLE, WITH SPECIAL REFERENCE TO America in this connection, although it lies SCOTLAND, NORTHWEST EUROPE, AND 2000 miles or more west of the Scottish and EASTERN NORTH AMERICA Scandinavian scene, is justified by the probable The Paleozoic succession of igneous rocks in consideration that the North Atlantic did not Scotland affords an excellent example of an exist in Paleozoic time (Bailey, 1928, p. 76); or igneous cycle parallel to the simplified tectonic better, perhaps, that North America was much cycle (Tyrrell, 1926, p. 284). It may be repre- closer to Northwest Europe until at least the sented as follows: end of the Paleozoic. The above fields present evidence of one complete diastrophic and Early: Early Ordovician igneous cycle (in Sender's sense) together with Ophiolitic series; basalts, spi- the closing phase of an earlier one, from the lites, keratophyres, etc. Late Precambrian to the Permian. Geosyncli- Albite dolerite, gabbro, ser- nal Phase pentine, etc. (4A) Continental phase. Late Precambrian Late: Late Ordovician-Middle Silu- (Torridonian) and Early Cambrian rian (IB) Geosynclinal phase. Late Cambrian and Little or no igneous activity Early Ordovician (Scotland) Early: Late Silurian? Igneous rocks belonging to this phase are not Foliated granites and grano- represented in Scotland, but the equivalent diorites, augen gneiss, oligo- Scandinavian formations (Dala, Jotnian, Spa- clase gneiss, lit-par-lit gneiss, ragmite, etc.) contain flows of amygdaloidal etc. in concordant sheet basalt, with great sills of dolerite (mainly intrusions and injection com- plexes (?). quartz dolerite but some olivine dolerite) and Metamorphosed ultrabasic and a dike swarm consisting of the same rocks. basic igneous rocks; foliated The equivalent North American formation lamprophyric sills and dikes. (Keweenawan) contains the enormous Lake Orogenic I Late: Late Silurian-Lower Old Red Superior basalt flood of estimated volume 24,000 Phase I Sandstone cubic miles (C. R. Van Hise and C. K. Leith, Intrusions of serpentine, gab- 1911), which is probably associated with ex- bro-, and norite. Granite, tensive dike swarms of dolerite and two of the granodiorite, tonalite, and largest gabbro lopoliths (Duluth and Sudbury) quartz diorite in massive in the world. discordant batholiths, sheet- and ring-intrusions; associated Submarine volcanic activity along the geo- with numerous small bodies synclinal zones gave rise to an ophiolitic (or of kentallenite, appinite, horn- spilitic) kindred. The volcanic rocks may be blendite, scyelite, etc. normal basalts, but there are many spilitic and Andesitic and rhyolitic lavas. some keratophyric types relatively rich in soda Carboniferous-Permian as compared with potash, and with albite or Basaltic lava fields with oligoclase as the dominant feldspar. Pillow olivine basalt, trachybasalt, Post-Oro- trachyandesite, trachyte, and structure is often spectacularly displayed as an genie Phase phonolite. indication of subaqueous eruption. The ac- Associated intrusions, mostly companying hypabyssal and (old-style) plu- sills, of picrite-teschenite, ther- tonic rocks are sills and dikes of albite dolerite, 4 alite, essexite , monchiquite. and masses of gabbro, pyroxenite, and peri- Quartz dolerite sills and dikes. dotite (serpentine)—an ophiolitic assemblage 4 Luscladite-Lacroix (Steinmann, 1926, p. 637). These rocks are

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 IGNEOUS CYCLE 413

frequently involved with the later orogeny and and during the folding (synorogenic). In altered to the so-called "green rocks" or green- Sutherland there is a group of ultrabasic and stones (chlorite-, hornblende-, glaucophane-, basic intrusions and augen gneisses, associated talc schists, "prasinites", etc.). The constant with Moinian and Lewisian rocks, which have submarine eruptions and the accompanying been highly metamorphosed along with these earthquakes attested by fine slumping phe- formations. In addition there are two great nomena (Henderson, 1935) sent clouds of basic injection complexes (injection gneisses, permea- debris mingled with the more normal sediments tion gneisses, granite gneisses, granites, and the of the sea floor up through the water. The peculiar intermediate and basic hybrid rocks gradual settlement of this material led to the known as the Ach'uaine type), which Read formation of the characteristic graded gray- (1931, p. 11-12) assigns to a pre-Torridonian wackes and shales which are now recognized date. Also, in the Grampian Highlands, asso- by many writers as the typical sedimentary ciated with rocks similar to the above-men- rocks of the geosynclinal phase along with tioned acid gneisses, is a group of greenstones cherts (often radiolarian) and chemical lime- (amphibolite, epidiorite, hornblende schist) and stones (Tyrrell, 1933; Krynine, 1941; Pettijohn, large bodies of ultrabasic rocks (now serpentine 1943; Krumbein and Sloss, 1951). Dewey and and associated types), which were intruded prior Flett (1911, p. 246) and Steinmann (1905; to the metamorphism of the Dalradian (Read, 1926, p. 662) had already associated the spilitic 1948, p. 39). Their exact date therefore de- (ophiolitic) suite with slow oceanic subsidence; pends on which is considered to be the age of the and Von Bubnoff (1923, p. 65) had pointed out Dalradian. It is now all but certain that Cam- the association of these rocks with geosynclinal brian rocks are involved with the Dalradian depression. (J. G. C. Anderson, 1947, p. 482-484), and Ophiolitic zones of the same Early Paleozoic therefore the most probable date of the "Older ages extend southwestward from the Southern Igneous Rocks" and the Dalradian meta- Uplands of Scotland into the center of Ireland morphism is Post-Cambrian—i.e., they belong and northeastward, along the entire length of to the earliest part of the Caledonian orogeny the Scandinavian peninsula (Goldschmidt, (Late Silurian). D. L. Reynolds (1942, p. 43) 1916, p. 7-21), right up into Spitsbergen, a total also regards the "Older Granites" as migmatites distance of some 3000 miles. The same rocks belonging to a syntectonic episode of the are no doubt found on the same scale in eastern Caledonian orogeny, although she might demur North America (e.g., in Newfoundland); but to calling them igneous. The writer therefore the writer has neither time, space, nor compe- regards all these rocks, with the possible excep- tence to discuss the American equivalents as tion of the Sutherland complexes, as belonging thoroughly as they deserve. This task must be to the earliest phase of Caledonian orogeny. left to American geologists who are more con- They probably represent the products of plu- versant with the terrain. tonic processes (in Read's sense) which took (2B) Transgression (Fluctuation) Phase. place at the roots of the geosynclinal prisms. Late Ordovician and the greater part of the The assemblage is unevenly distributed Silurian in Scotland throughout the Scottish Highlands, and rocks In this phase there was only slight epeirogenic of the same types and presumably of the same movement, and, in Scotland, igneous activity age extend up the length of the Scandinavian was wholly in abeyance. peninsula from south to north, especially in the (3B) Orogenic (Emergence) Phase. The western (Norwegian) part. From a very cursory Caledonian Orogeny. Late Silurian and Early dip into the literature the writer believes that Devonian (Lower Old Red Sandstone) the igneous rocks of both parts of the Orogenic In Scotland there seem to have been at least Phase are not so widely distributed in eastern two peaks of igneous activity. The earlier was North America as they are in Northwest the injection of foliated granites and grano- Europe, but he is open to correction on this diorites, oligoclase gneiss, etc., in comparatively point by American geologists more familiar with small masses as concordant intrusions before the ground.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 414 G. W. TYRRELL—IGNEOUS ROCKS IN SPACE AND TIME

The second period of igneous activity oc- and the Southern Uplands of Scotland south of curred after the folding had ceased and pre- the Midland graben. It is not yet known sumably when deep-seated plutonic processes whether these masses are to be associated with had reached their end stage of complete mobi- the first or second of the post-orogenic groups, lization, with the production of eruptible but certain large intrusions of felsite along or magma capable of truly igneous injection into near the Southern Boundary Fault seem to be the upper parts of the erogenic zones. There intrusive into the Lower Old Red Sandstone. seem to be two distinct series. One, which has The above post-orogenic suite was preceded apparently contributed many boulders of and accompanied by andesitic and rhyolitic similar rocks to the conglomerates of the eruptions from small volcanoes, forming the Downtonian and Lower Old Red Sandstone, is Lome, Sidlaw, Ochil, Pentland, Carrick, and most probably of Late Silurian age. It consists Cheviot Hills. The principal lava types erupted of a group of ultrabasic and basic sheet in- were highly feldspathic "basalts" with sporadic trusions (peridotite, olivine gabbro, troctolite, olivine6, pyroxene-, hornblende-, and biotite olivine norite, hypersthene gabbro, etc.) oc- andesites, dacites, and rhyolites. At Taynuilt curring in Aberdeenshire and Banff shire; and a (Argyllshire) in the Lome field there are rela- group consisting of massive intrusions of biotite tively potash-rich lavas called orthoclase- granite, granodiorite, tonalite, and diorite with olivine basalt which may correspond to the a hypabyssal entourage of porphyrites, lam- kentallenite type among the "plutonic" rocks prophyres, aplites, and pegmatities, which are (Bailey and Maufe, 1916, p. 185). These lavas distributed throughout the Central Highlands were erupted onto a land surface as there are and include the great Cairngorm mass and the sedimentary intercalations and cavity fillings Aberdeen granites. They are all massive, un- of Old Red Sandstone type, marked with the foliated, and structureless (except for lineations tracks and trails of crawling animals. Many of and what one may call the Cloos Gefuge). In the lavas have reddened tops and rusty patches their relations to the country rocks they are which may mark the sites of temporary pools mostly crosscutting and discordant, though (Tyrrell, 1914, p. 66-68). with some suggestions of flattish tops and even A great unconformity exists between the bases (Read, 1948, p. 48-51). Lower and Upper divisions of the Old Red The second series is in part later than the Sandstone during which period the new con- Lower Old Red Sandstone and is best de- tinent reached stability. In one or two parts of veloped in the Southwest Highlands. It con- the Midland Valley of Scotland a final spasm of sists of plutonic masses of granite, granodiorite, Caledonian orogeny caused gentle open folding tonalite, and quartz diorite (Ben Nevis, Glen in the Lower Old Red Sandstone. (Cf. connec- Etive, Ben Cruachan, etc.), associated with tion of andesitic volcanoes with this type of dike- and sill-swarms of porphyries and lam- folding.) prophyres. An abundant suite of small bosses of (4B) Continental Phase. Latest Old Red monzonite, kentallenite, and cortlandtite, some Sandstone, Carboniferous and early Permian of which are relatively rich in potash, are as- The tectonic history of the continental phase sociated with the above acidic types. Some of begins with the breakdown of the Old Red these basic rocks also occur in intrusive com- Sandstone continent by block faulting, with the plexes along with granite and granodiorite. It formation of a great graben or rift valley in the is in this series that the remarkable cauldron Scottish midlands along northeast to south- ring complexes of Ben Nevis, Glen Coe, and west faults—i.e., in the main Caledonian direc- Glen Etive are developed (Read, 1948, p. 53- tion. The Midland Graben is bounded to the 57). north by the Highland Boundary Fault which, A group of large and small plutonic masses as it affects the Lower but not the Upper di- of granite, granodiorite, and quartz diorite associated with a small number of basic types B The writer prefers to call these rocks olivine andesites. In chemical composition they are very and accompanied by dike swarms of lam- different from the true olivine basalts of the Car- prophyre and porphyry, occurs in Galloway boniferous.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 IGNEOUS CYCLE 415

vision of the Old Red Sandstone, must have are known at the present level of erosion, except been formed in Middle Old Red Sandstone fragments of peridotite, pyroxenite, anortho- times. To the south the graben is bounded by an site, gabbro, etc. in some vent agglomerates'. echelon of faults which may be collectively re- In Scandinavia the probable equivalent of ferred to as the Southern Uplands Boundary this Scottish period-province is the famous Fault. Both these fault systems run in the alkalic province of Oslo, Norway, now known general northeast to southwest Caledonian to be Permian. In eastern North America it is direction. The rift depression is filled with Old the White Mountain Magma Series of New Red Sandstone and Carboniferous rocks; and England (together with the Monteregian field its floor is broken by differential fault move- of the adjacent part of Canada?), of which the ments into a mosaic of smaller blocks. This age is now believed to be Mississippian (Wil- faulting continued right through the Carbonif- liams and Billings, 1938, p. 1025). Possibly the erous and after it. As time went on the faults New Hampshire Magma Series, syntectonic veered more and more to the east and west with the Acadian Revolution, is the tectono- direction, thus becoming parallel to the major igneous equivalent in New England of the direction of the Hercynian folds which were Scottish granodiorite-andesite kindred which corning into existence some 400 miles to the is roughly syntectonic with the Caledonian south in Southern England. Revolution (Billings and Keevil, 1946, p. 810- Igneous activity in the Midland Valley of 811). The writer invites the attention of Ameri- Scotland is associated with the initiation and can geologists, especially those whose field early history of the Midland Graben. It began of work is in the New England States, to the in late Upper Old Red Sandstone time with the close similarity among the Oslo, Scottish, and eruption of mainly olivine basalts from hun- New England petrographic provinces in point dreds of small volcanoes of central type, giving of age, petrographic composition, and igneous rise to comparatively large basalt fields (Tyrrell, structures. Such a comparison may suggest 1931a, p. 186-188). The major period of erup- another line of evidence which would lead us to tion was early Carboniferous, but the activity suspect that North America was much nearer continued with ever-diminishing strength Northwest Europe in the later Paleozoic than through the epochs of the Carboniferous Lime- it is now. stone Series and the Millstone Grit, and flared up finally in the Carbo-Permian. All this QUARTZ DOLERITE DIKES AND SILLS igneous activity seems to be definitely earlier OF NORTH BRITAIN than the east-west faulting. The lavas belong To complete the consideration of the Paleo- to an olivine basalt differentiation series lead- zoic igneous cycle in Scotland let us turn now ing to trachyte and phonolite (trachybasaltic to discuss a late igneous event quite different kindred—A. Holmes) at least in the greater from those dealt with above—namely, the in- part of the petrographic period; but the later trusion of a great suite of quartz dolerite dikes eruptions (Late Carboniferous-Early Permian) and sills (F. Walker, 1935; J. E. Richey, 1939; are coarse olivine-rich basalts (Millstone Grit) A. G. MacGregor, 1948). They occur as thick and analcite- and nepheline-bearing olivine (60-100 foot) extended dikes which run in a basalts (Early Permian). Neither group is ac- general east-west direction, some of them for companied by trachytes or phonolites. Their scores of miles. They are associated with a hypabyssal representatives are differentiated number of thick sills which, although only one sills of teschenite-picrite-peridotite, bekinkinite, direct connection has been proved, have cer- theralite, and monchiquite, showing fine ex- amples of gravity differentiation, and forming 6 In one of these vents, that of the Black Rock at Fairlie (Ayrshire), the writer recently found a what is probably the largest petrographic region block of carbonatite rich in soda hornblende, of analcite-bearing igneous rocks known. Dikes closely resembling types from Fen (Norway) and associated with this phase of igneous activity Alno (Sweden). In others the writer has found biotites like those described by H. S. Washington are scarce except around some large centers of from Italy. Xenocrysts of anorthoclase, titanaugite, trachytic eruption. No "plu tonic" equivalents soda hornblende, and biotite are common.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 416 G. W. TYRRELL—IGNEOUS ROCKS IN SPACE AND TIME

tainly been fed from the dikes. They are prob- alkalic magmas, erupted from local and com- ably slightly earlier than the picrite-teschenite paratively shallow reservoirs, were gradually sills and are thus Permo-Carboniferous. Their exhausted, and the province became mag- area of distribution is much more extensive matically "dead". The Scottish region then than that of the preceding basic alkalic suite, became part of a great kratogen bounded to which is largely confined to the Midland the south by the rising Hercynian folds. The Graben. They cover an area from Aberdeen- tensions thus created caused fissures which ex- shire on the north to south of the English tended deep enough to tap the intermediate Border, where the Great Whin Sill of the North layer of olivine-free basalt (Kennedy and of England appears to be the southernmost and Anderson, 1938, p. 35-39; MacGregor, 1948 largest member of the suite. p. 148). The chemistry and mineralogy, the mode of The above, of course, raises difficult ques- differentiation or, rather, the almost complete tions of the contemporaneity of tectono-igneous lack of differentiation, the prominence of thick cycles proceeding in different, although ad- dikes, and the much wider area of distribution jacent, parts of the earth's crust, and of varia- of this suite are in complete contrast to the tions in such cycles. At the same time (Late preceding alkalic suite. The magmatic com- Silurian to Lower Devonian) that an erogenic position of these rocks and their geological oc- phase was in progress in the Caledonian orogen, currence, so alien to those of the Carbo- a geosynclinal phase accompanied by an ophio- Permian, support the view that they are derived litic series (Devonian) was proceeding in the from new and different magmatic sources, and Hercynian orogen. The question is also raised that they may mark a completely new beginning as to whether we ought to treat orogens and in the sequence of igneous events. It can also kratogens separately in regard to tectono- be surmised that the quartz dolerite magma was igneous cycles. The writer thinks we should. derived from more deeply seated sources than These questions are far-reaching in their im- the limited and localized reservoirs of the Carbo- plications, and it is doubtful whether the data Permian igneous rocks which, as we have seen, are yet at hand for their solution. Their further diminished greatly in bulk and became highly exploration must be left for future work. differentiated and specialized toward the end of their eruption period. The alkalic magmas thus TECTONO-IGNEOUS CYCLE IN INDONESIA seem to have been almost completely exhausted before the arrival of the quartz dolerites. For an example of a more recent tectono- In the writer's opinion the east to west ex- igneous cycle we may turn to Indonesia which tension of the quartz dolerite dikes and the has been closely investigated by R. W. van 7 limitation of their age to the earliest Permian Bemmelen (1949; 1950) . He starts from a cor- give a clue to their tectonic affiliation. The relation between petrographic provinces and writer believes they represent a phase of igneous the phases of structural evolution (diastrophic activity in the northern kratogen corresponding cycle) from the Cretaceous to the present time to the Hercynian orogeny which had been pro- within the great Tertiary orogen of that region. ceeding in the geosynclines far to the south. Starting from the south an "Atlantic suite" A. G. MacGregor (1948, p. 148) relates them to of igneous rocks occurs in or near the southern the Borcovician stress system which operated foreland of the orogen (Christmas I. etc.), fol- from late Carboniferous to earliest Permian lowed by an "ophiolitic suite" in the adjacent time. The situation is rather involved and diffi- foredeep to the north which is of Late Cretace- cult to disentangle. There is here perhaps what ous to Paleogene age. Then come three suc- may be called a sudden transfer of allegiance cessive erogenic phases in the course of the from one tectono-igneous system to another. Tertiary and Quaternary, each accompanied The alkalic rocks of the Carbo-Permian graben by the metasomatic emplacement of "acid were clearly associated with the waning Cale- 7 donian orogeny, and with block faulting of the The writer has only very recently come across van Bemmelen's book on the Geology of Indonesia Old Red Sandstone continent as compression (1949), and has not yet had time fully to understand relaxed after the cessation of folding. The and assimilate his views on igneous origins.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 TECTONO-IGNEOUS CYCLE IN INDONESIA 417

plutonites" (quartz diorite, granodiorite, gran- post-folding phases, respectively Late Silurian ite) in the cores of the geanticlines. Each or- and Lower Old Red Sandstone. There is even a ogenic phase was succeeded by a cycle of phase of potash-rich magmas in both the In- "basalto-andesitic" volcanicity in the Oli- donesian and Scottish fields at approximately go-Miocene, Mio-Pliocene, and Quaternary, the same position in the sequence of tectono- respectively, the lavas being erupted from the igneous events. The Indonesian cycle concludes crests of the geanticlines. In at least the second with a Post-Orogenic Phase of fissure eruptions erogenic phase the "rising front of granitiza- emitting tholeiitic basalt lavas which find their tion" invades the lavas belonging to the first parallel in the Scottish quartz dolerite dikes cycle of erogenic volcanism ("Old Andesite and sills; but there seems to be no indication Formation"). The first and second of these of an Indonesian trachybasaltic kindred be- cycles were terminated by enormous eruptions longing to the Post-Orogenic Phase. The gen- of "acid volcanites" (mainly tuffs)8 respec- eral parallelism between the Indonesian and tively generated by and in connection with Scottish tectono-igneous cycles thus appears to the upward advance of the acid plutonites of be very close. the second and third erogenic phases. Toward the end of the third orogenic phase in Java some PLUTONIC AND VOLCANIC ASSOCIATIONS aberrant differentiates of "Mediterranean affinities" were produced besides the normal In an important paper W. Q. Kennedy "Pacific" andesitic lavas (e.g., the Muriah (Kennedy and Anderson, 1938) put forward volcano with potassic lavas). The post-orogenic the fruitful conception of plutonic and volcanic igneous rocks of the cycle consist of large associations. The plutonic association includes spreads of "tholeiitic plateau basalts" which all subjacent igneous masses (concordant and cover the northern hinterland of the Indonesian discordant batholiths, stocks, sheet complexes, orogenic arcs. etc.) which are intruded at the roots of an oro- Van Bemmelen's summary of the above genic belt during or immediately after an active sequence is given in the first three columns of orogeny, together with their associated minor Table 1, to which the writer has added two intrusions, but with practically no effusive columns showing what he could gather of the equivalents. The volcanic association, on the petrographic characters and ages of the igneous other hand, consists mainly of superficial ef- rocks of the Indonesian cycle, and the corre- fusive rocks, with the intrusive masses geneti- sponding stages in the Scottish region. The cally associated with them and derived from comparison makes clear a parallelism between the same magmatic sources, within a limited the two sequences. From Van Bemmelen's cycle of volcanic activity. The intrusions of the rather compressed and involved style of Eng- plutonic association are overwhelmingly acid, lish writing it is sometimes difficult to extract consisting almost entirely of granodioritic and the relevant facts of age and petrographic com- granitic types. Typical basic and ultrabasic position, and there are a few apparent incon- rocks are characteristically rare or absent. The sistencies in his account. His Indonesian cycle volcanic association, on the other hand, is com- starts with a Pre-orogenic phase which the posed mainly of basalts and their less abundant writer is inclined to regard as the final con- differentiates which belong to the two basaltic tinental phase of the preceding cycle. There is lines of descent postulated by Kennedy. Acid no clear indication of a Transgression phase. lavas and intrusions are thus rare in compari- Within his Orogenic phase there appear to be son with the basic. True plutonic associations three tectono-igneous subcycles, and there are are connected with orogenic belts; but, al- also indications of three in the Scottish field— though Kennedy recognizes volcanic associa- namely, the pre- and synorogenic phase at the tions within the orogenic zones he is unwilling beginning of the Caledonian orogeny, and two to regard them as genetically connected with the 8 In a summary account of van Bemmelen's plutonic rocks, the great bulk of volcanic ef- 1950 paper (Science Progress, July 1951, p. 506) the fusions and intrusions occur within, and are writer inadvertently wrote "plutonites" instead of "volcanites". genetically connected with, the stable, non-

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 TABLE 1.—CORRELATION OF THE IGNEOUS AND TECTONIC PHASES OF THE TERTIARY AND QUATERNARY OF INDONESIA WITH THOSE OF THE PALEOZOIC OF SCOTLAND oo Zonal stages of Petrographic provinces Orogenic zones orogenic evolution Age and petrographic characters Scottish Paleozoic equivalents I Atlantic Suite Foreland (1) Pre-Orogenic Trachybasaltic kindred of Christmas I. vol- cano, etc. (Early Tertiary) II. Ophiolitic Suite Foredeep (2) Geosynclinal Ophiolitic kindred (Late Cretaceous to Early Ophiolitic kindred of Geosynclinal Phase Tertiary. The age varies in different parts of (Early Ordovician) the region) Transgression Phase. No igneous activity (Late Ordovician-Silurian) III. Pacific Suite Geanticline (3) Orogenic (sensu First Orogeny Early Orogenic Phase stricto) Plutonites: granites etc. (Eocene-Oligocene) Plutonites: foliated granites, oligoclase Volcanites: basalto-andesites and acid tuffs gneiss etc., migma-rocks; foliated basic and (Oligo-Miocene) ultrabasic rocks (Late Silurian) iw Second Orogeny Second Orogenic Phase o Plutonites: granites, etc., and basic rocks Plutonites: granite, granodiorite, quartz- ("basic front") (Miocene) diorite, gabbro, norite, and peridotite (Late Volcanites: basalto-andesites and acid tuffs Silurian) (Mio-Pliocene) C/8> Third Orogeny Third (?) Orogenic Phase h-1 Plutonites: granites, etc. (Quaternary) Plutonites: granite, granodiorite, quartz- 2 Volcanites: basalto-andesites and acid tuffs diorite, etc. with numerous small masses of C/3 (Plio-Pleistocene) monzonite, kentallenite, etc. Volcanites: lavas ranging from basaltic O andesite to rhyolite (Lower Old Red Sand- H stone) IV. Mediterranean Backdeep (4) Late Orogenic Potash-rich lavas of the Muriah and Ringgit (Kentallenite etc. and orthoclase-olivine Suite volcanoes (Plio-Pleistocene) basalts within the above Third (?) Orogenic Phase) w V. Tholeiitic Pla- Hinterland (5) Post-Orogenic Post-orogenic Phase teau Basalts Volcanites: Trachybasaltic kindred (Car- boniferous-Early Permian) Fissure eruptions giving rise to over-satu- rated basalts (tholeiitic basalts) (Quaternary) Quartz dolerite dikes and sills (Earliest Permian)

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 PLUTONIC AND VOLCANIC ASSOCIATIONS 419

erogenic or kratogenic regions of the earth's dominantly andesites, eruptions of rhyolitic crust. tuff reach a thickness of 1000 feet and cover While these ideas in their present form have thousands of square miles, and are thus clearly influenced and advanced petrogenetic theory, a most important feature of Andean volcanicity. they have never been subjected to the de- He suggests, in consequence, that the prevalent tailed criticism which, in my opinion, would idea of an overwhelming excess of andesitic add considerably to their usefulness. Kennedy, lava in the Andean region may require modifi- for example, practically rules out volcanic rocks cation. The same relations probably hold in from his plutonic association, and one may the Katmai region of Alaska and in the North wonder how the ophiolitic (spilitic) suite and the Island of New Zealand. These facts are clearly pyroxene andesite to rhyolite suite fit into at variance with the characters Kennedy re- his scheme. He does not mention the ophiolitic gards as typical of the volcanic association suite at all and only refers to the basalt- While in some passages Kennedy appears andesite-rhyolite suite in connection with his to exclude the possibility of his volcanic asso- two postulated lines of basaltic descent (1938, ciation occurring in connection with plutonic p. 34, 37). Both suites are of worldwide distri- masses, and cites certain facts as "suggestive bution in orogenic zones of all ages, and occur of the lack of intimate connection between the in great bulk (Tyrrell, 1937, p. 91). The ophio- plutonic and volcanic associations", he goes litic suite is recognized by almost all petrolo- on to say (Kennedy and Anderson, 1938, p. gists as being connected with the geosynclinal 29): stage of the diastrophic cycle, and its eruption is the igneous event which corresponds to the "This conclusion appears to find support also in the observed relations of batholiths to their opening stage of that series of tectonic events roofs, and Bucher [1933] in particular has empha- which culminates in the orogenic phase. Scheu- sised the fact that the large acid plutonic masses mann (1924, p. 8) indeed calls it prolotectonic. of the orogenic belts are capable of penetrating the crust to points very near the surface without The typical basalts, spilites, gabbros, and peri- bursting their roofs in catastrophic fashion. Under dotites (serpentines) are never found along such circumstances, moreover, they do not appear to give rise to surface eruptions but actually cut with the rocks of the basaltic lines of descent the effusive rocks of the same igneous episode [italics which are characteristic of nonorogenic regions. present writer's]. This universal absence of lavas In the writer's opinion their provenance, dis- belonging to a period contemporaneous with the rise of batholithic intrusions to their highest levels tribution, and association with orogenic events in the crust is of considerable importance". places them indubitably with the plu tonic association. But, on Kennedy's own showing, it is pos- In regard to the basalt—pyroxene andesite— sible for effusive rocks of the same ignevus hornblende-mica andesite—dacite—rhyolite se- episode to reach the surface before the batho- ries, which appears to be characteristic of liths arrive at their highest levels in the crust. orogenic zones and times, basalt is just as Moreover, as in the Scottish Caledonian and subordinate in this series as it is predominant the Indonesian Tertiary orogenic phases, the in the volcanic association. Pyroxene andesites lavas which are cut by \hefinal intrusive efforts are dominant in this suite, and basic pyroxene of the batholiths are invariably those of the andesites often rich in hypersthene are the andesite—dacite—rhyolite series,—a series par- most abundant lavas erupted from the present- allel in chemical composition to the rocks of day volcanoes of the entire circum-Pacific the quartz diorite—tonalite—granodiorite— orogenic zone. There is reason to believe that granite series, i.e. the typical rocks of the plu- acid lavas and tuffs are nearly as abundant tonic association. as the andesites in some regions as, for ex- Parallelism in chemical composition between ample, the "cataclysmal eruptions" of acid the plutonic and volcanic suites is shown by tuffs, thrice repeated, in the orogenic phases comparison of average analyses of the various of the Tertiary Indonesian cycle. Fenner types. For example, the average composition (1948, p. 890-892) states that, while the normal of hypersthene andesite which the writer has lavas of the Arequipa (Peru) region are pre- computed from 114 analyses of rocks called

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 420 G. W. TYRRELL—IGNEOUS ROCKS IN SPACE AND TIME

hypersthene andesite, and pyroxene andesite NATURE AND ORIGIN OP IGNEOUS CYCLES containing hypersthene, culled from the litera- ture on Recent and Tertiary volcanoes and At this stage the writer does not attempt to lava fields of the circum-Pacific region (includ- discuss fully the views of the many workers in ing the East and West Indies), and certain this field of study, but merely to present in European fields (Hungary, Sardinia, Aegean brief his own views, which have been partly Sea) (Tyrrell, 1931b, p. 195), is extraordinarily close to the average of 125 diorites and quartz TABLE 2.—TECTONO-IGNEOUS CYCLE diorites computed by Daly and to the average Diastrophism Kindreds Locus of the 55 quartz diorites which are included in the above (Daly, 1933, p. 15-16). Also, com- I. Geosyncli- 1. Ophiolitic parison of the average analyses of 40 gran- nal Phase Kindred odiorites and 90 dacites (Daly, 1933, p. 15) and of 546 granites and 126 rhyolites (Daly, II. Orogenic 2. Granodio- In Orogen Phase (with rite-Ande- 1933, p. 9) strongly reinforces the parallelism. two or three site Kin- The lava series ranging from pyroxene ande- subphases) dred site to rhyolite, therefore, has the same chem- ical composition, the same geological and geo- III. Post-Oro- 3. Trachyba- graphical distribution, the same tectonic genic Phase saltic Kin- environment, and appears at the same stage (with two dred • In Kratogen of the tectono-igneous cycle as the plu tonic subphases) 4. Quartz dol- series ranging from quartz diorite to granite. erite Kin- If the latter belongs to the plutonic association dred there seems to be no escape from the conclusion that the andesite—dacite—rhyolite series adumbrated in earlier pages of this paper, and likewise does. Indeed if the volcanic association compare them with a few selected from the consists predominantly of basaltic lavas with abundant literature. their differentiates, together with subordinate For the highly complicated region of North- coarse-grained intrusions of "plutonic" type, western Europe, consisting, as it does, of three there is no reason why the predominatingly ancient orogens welded onto the Scandinavian- coarse-grained, more or less "acid", subjacent Baltic Shield, the complete tectono-igneous intrusions of the plutonic association should cycle worked out for the Scottish Paleozoic, not give rise to subordinate series of volcanic which may be regarded as typical for the above rocks corresponding to them in chemical range. region, is shown in Table 2, reduced to its The writer suggests that the whole trouble simplest terms and using the simple diastrophic with plutonic and volcanic associations is that cycle already given. they are wrongly named. As Kennedy has well The cycle begins with an ophiolitic kindred shown there are certainly two strongly con- associated with the geosynclinal or prototec- trasted series of igneous rocks; but the two tonic stage of orogeny, and confined to the groups are contrasted not so much in their orogen. At the climax of orogeny, after a long plutonic or volcanic characters, or in their period of comparative quiescence in diastrophic chemical compositions, as in their tectonic and igneous activity, vast irruptions of the associations. One group is associated with granodiorite-andesite kindred take place, and orogenetic processes during a period of active may reach two or three peaks of intensity orogeny in an erogenic zone; the other is preceding, during, and immediately subsequent associated with epeirogenetic movements in to the folding. In the earliest subphase processes stable nonorogenic regions (kratogens). Ken- of palingenesis, anatexis, and granitization nedy's concept would gain greatly in clarity transform the roots of the geosynclinal prism and usefulness if the plutonic and volcanic into plutonic rocks (Read's sense). In the later associations were renamed orogenic and krato- subphases the completion of mobilization genic, respectively. makes available magmas capable of injection

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 NATURE AND ORIGIN OF IGNEOUS CYCLES 421

into the upper brittle parts of the crust, where throughout the eras within the western half they solidify to form igneous rocks of the gran- of Europe, and will possibly be found to hold odiorite-andesite kindred. Finally, toward the for eastern North America. The third column

TABLE 3.—COMPAEISON or THE IGNEOUS PHASES OF THE CALEDONIAN, HEECYNIAN, AND ALPINE OROGENIC CYCLES Alpine (Central Europe and Diastrophic Cy cle -> Caledonian (Scotland) Hercynian (Western Europe, Etc.) Mediterranean)*

III. Post-oro- 4. Quartz dolerite kindred; 4. Newark basalts, Pali- genic Phase dikes and sills (Earliest sade sill, etc. of eastern Permian) North America (Trias- sic)

3. Trachybasaltic kin- 3. ? Exeter lavas, some pot- 3. Trachybasaltic kindred dred; lavas and associ- ash-rich types (Middle of Extra-Alpine re- ated intrusions (Lower to Upper Permian) gions (Late Tertiary to Carboniferous to Early Recent) Permian)

II. Orogenic 2b. Granodiorite-andesite 2b. "Porphyrite" and "Me- 2b. Andesite fields, Hun- Phase kindred; plutons and laphyre" lavas of west- gary and Balkans (Ter- lavas (Late Silurian to ern Germany (Permian) tiary) Lower Devonian) Granite and granodi- Granodiorites, to- orite plutons of South- nalites, etc. Adamello west England and Ger- & Tonale Alps, etc. many (Late Carbonifer- (Late Cretaceous) ous to Permian)

2a. Granodiorites and gran- 2a. Granodiorite and gran- 2a. Synorogenic granodio- ites; Gneisses, migma- ite syntectonic intru- rite gneisses, etc. (Mid- rocks, injection com- sions, gneisses, and dle Cretaceous to Pa- plexes (Late Silurian?) granulites (Late Paleo- leocene — long believed zoic) to be Hercynian)

I. Geosynclinal 1. Ophiolitic kindred 1. Ophiolitic kindred (Mid- 1. Ophiolitic kindred Phase (Lower Ordovician) dle Devonian to Lower (Mesozoic —variou s Culm) ages) * Most of the data for this column derived from Burri and Niggli (1945)

end of the cycle, when the orogen has been of Table 3 covers igneous events of the Alpine united with the adjacent kratogen and then cycle throughout the Mediterranean and Alpine participates in the ensuing kratogenic tectonic region as far as the Balkans and Turkey. Table reactions (block faulting, etc.), there are 1 shows that the above theoretical tectono- abundant local eruptions of the trachybasaltic igneous cycle applies to the Indonesian Ter- kindred, followed by an extensive series of tiary rather closely; and we may surmise, dikes and sills of the quartz dolerite kindred. therefore, that it fits tectono-igneous events The first and second tectono-igneous phases throughout the whole length of the Alpine- are products of erogenic, the third and fourth Himalayan system. In the third column of of kratogenic activity. Table 3 the writer has omitted the enormous This cycle of tectono-igneous events is North Atlantic region of flood basalts and the probably too neat to be the whole truth, but accompanying local centers. This is because Table 3 shows that it holds fairly generally he believes it to be a thalattogenic event

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 422 G. W. TYRRELL—IGNEOUS ROCKS IN SPACE AND TIME

connected with the formation of the North phase is separated by a long period of quiescence Atlantic Ocean, entirely alien to, although from the granitization processes which set in approximately contemporaneous with, the during the opening stages of active folding in tectono-igneous events of the Alpine diastrophic the orogen. In many regions for which granitiza- cycle. So also with the Deccan Traps which tion has been advocated it is these very rocks are considered to be similarly connected with of the ophiolitic kindred, and especially the the inbreaks which formed the northern part graywackes with which they are invariably of the Indian Ocean. associated, which have been affected by that The ultimate origin of the kindreds that the process. writer has associated with the tectono-igneous Another point which seems to be of consider- cycle, and many others, is still largely a matter able importance in this connection is the un- of conjecture and speculation. Umbgrove expected fact that the order of size of ophiolitic (1939) has provided a valuable review of some accumulations is little smaller than that of the of the ideas that have been put forward. Many flood basalts. Extrapolating from the volume authors have pointed out that basic and ultra- of the ophiolitic suite of the Caledonian geo- basic rocks develop in geosynclines at the syncline in southern Scotland (Tyrrell, 1937, beginning of the tectono-igneous cycle, and p. 31), which is estimated at 1000 cubic miles this is perhaps the best-established generaliza- for :very 100 miles along the strike, it is pos- tion in the subject (Holmes, 1932; Bucher, sible that 30,000 cubic miles of basaltic and 1933; Rittmann, 1936). The writer subscribes spilitic lavas were erupted along the Cale- to Hess's views (1938) on the origin of geo- donian orogen in Northwestern Europe; and synclinal ultrabasic rocks, and would extend as there was almost certainly an equal quantity them to explain the ophiolitic suite of which of basic and ultrabasic intrusive rocks asso- serpentine in particular is such an important ciated with the lavas the total amount of ir- constituent. In the development of a geosyn- rupted material may have been of the order cline the lower parts of the crystalline crust of 60,000 cubic miles. If this estimate is ac- are buckled downward with the result that ceptable the enormous bulk of the ophiolitic the basaltic layer and even the ultrabasic suite in most old fold-mountain ranges may be material at greater depths enter regions of regarded as more favorable to the idea of its higher temperature with the formation of sheet- rapid derivation from elongated reservoirs of like magmatic pools at various levels which fused material of basic and ultrabasic composi- may ultimately reach shallower positions in tion along the geosynclinal zones rather than the crust along planes of tectonic weakness as a "basic front". (Kennedy and Anderson, 1938, Fig. 1, p. 39; In regard to the Orogenic Phase the writer Hess, 1938, Fig. 1, p. 334). Sir E. B. Bailey has shown that what he has called the gran- and W. J. McCallien have even announced odiorite-andesite kindred is the characteristic the discovery of serpentine lavas in Anatolia igneous product. The writer believes with (1950). This may explain the irruptions of Eskola (1932) that, in orogens, processes of basic and ultrabasic rocks along the geosyn- palingenesis and anatexis with assimilation clines ; but there is still no adequate explanation and differential melting, to which we may of the frequent excess of soda resulting in the now legitimately add the Transformist mecha- characteristic spilites and keratophyres of the nisms of granitization, are responsible for the ophiolitic phase which remains an unsolved ultimate appearance of magmas of the gran- petrological mystery. odiorite-andesite kindred in the upper levels I cannot entertain Van Bemmelen's tenta- of the crust. Von Bubnoff's fine diagram of tive hypothesis (1950, p. 218) that "the ophio- the Schwarzwald and Odenwald tectono-igneous litic suite is a regional low-level basic front" levels, reproduced with slight modification representing a "geochemical culmination of in the writer's Principles of petrology (1926a, cafemic constituents in front of acidification p. 335), perfectly illustrates the point of view. and migmatization of the crustal base", for In kratogens, on the other hand, the writer the sufficient reason that, in all the tectono- believes that granite (and usually a more igneous cycles yet investigated, the ophiolitic alkalic variety of granite than that of the oro-

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 NATURE AND ORIGIN OF IGNEOUS CYCLES 423

gens) is due to processes of differentiation Apparently it is only the later and less (Tyrrell, in Eskola, 1932, p. 481); but, of course, powerful paroxysms of tectonic activity which processes of differentiation may also take place are associated with volcanic eruptions of the in the later stages of the erogenic phase when granodiorite-andesite kindred accompanied or true magmas have been generated and are followed by the emplacement of large granitoid held for a time in subcrustal reservoirs. All plutons. This fact consorts well with R. T. the above-mentioned processes co-operate in Chamberlin's distinction between thick-shelled generating "granitoids"3. This is, of course, and thin-shelled fold mountains (1919). In the eclecticism; but the writer would rather err in thin-shelled type such as the Pennsylvania that direction than the opposite one, in which Appalachians, the Canadian Rockies, and the the inventors of a new and promising theory, Alps, a relatively thin superficial portion of in the first flush of enthusiasm, overextend the crust is characterized by intense folding, its legitimate field of application to the exclu- thrusting, and dynamic metamorphism, with a sion of all others in the attempt to make one maximal amount of shortening and little or theory alone responsible for the whole range no volcanism. On the other hand thick-shelled of tectono-igneous phenomena. mountains such as the Colorado Rockies and There are at least two, and often more, peaks much of the Andes are marked by compara- of intensity during the orogenic phase of tec- tively gentle open folding, moderate shortening tono-igneous activity. In the Caledonian of affecting a relatively thick zone of the crust, Scotland and the Tertiary cycle in Indonesia and strong epeirogenic uplift with normal there seem to be three; and in the Argentinian faulting which may likewise reach and tap Andes Backlund (1926) distinguishes no fewer considerable depths. It is with the latter type than five. The first sub-phase, which is pre- or that the granodiorite-andesite kindred, in syntectonic, is marked by the formation of both its volcanic and plutonic manifestations, gneisses, migma rocks, and injection complexes is connected. The same contrast may be found mainly of "acid" composition, although more in one and the same orogenic zone as, for ex- basic types may participate to varying extents. ample, in the Colorado Rockies (thick-shelled) This is the period in which palingenetic and and the Rockies of Montana and Alberta anatectic processes, inmelting and differential (thin-shelled). The intense compression suffered melting, assimilation, and granitization in its by thin-shelled mountains appears to be in- narrower sense reach their highest intensity hibitory of surface volcanism. in the deep-seated theaters of operation at the When the orogenic belt has been welded onto bases of the geosynclinal prisms with which the adjacent kratogen the very different con- marginal parts of the adjacent forelands are ditions of the Post-Orogenic (Continental) also involved. As time goes on true magma is Phase supervene. The diastrophism is of epeiro- generated from the mixed materials as the genic character; great faults and zones of weak- mobilization admitted by all but the most ness penetrate deeply into the continental extreme Transformists proceeds. Magma de- basement, and may tap pools of magma which rived from these processes reaches higher have been formed within both the intermediate levels in the crust during later and less intense layer (layer of olivine-free or tholeiitic basalt, spasms of tectonic activity and gives rise to Kennedy and Anderson, 1938, p. 38) and the granitoid plutons with 'approximately con- olivine basalt layer of the earth. Kennedy and temporaneous outbursts of andesitic and rhyo- Anderson (1938) have investigated the thermal litic volcanoes. How these bodies are emplaced and tectonic conditions under which such or erupted is admittedly difficult to explain, pools may be formed, and Figure 1 of their but as these vexed questions, however impor- paper provides a succinct statement of their tant, are largely irrelevant to the topic of views. On the other hand it is arguable that distribution in space and time, the writer does the olivine-free or tholeiitic basalt layer does not propose to enter upon them here. not exist as such, but that magma pools of that composition are formed as the result of 9 Granitoid—an omnibus term for the series of granite-granodiorite (adamellite, etc.)-tonalite- interaction between activated olivine basalt quartz diorite, etc. and the more acid underside of the continental

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 424 G. W. TYRRELL—IGNEOUS ROCKS IN SPACE AND TIME

crust. Whichever view is adopted it is certain REFERENCES CITED that both types of basaltic magma become Anderson, J. G. C., 1947, The geology of the High- available in great quantity at the Post-Oro- land Border: Stonehaven to Arran: Royal genic (Continental) Phase of the tectono- Soc. Edinburgh Trans., v. 61, pt. 2, p. 479-515. Backhand, H. G., 1926, Magmatic activity and igneous cycle. mountain folding in the Andes of South Men- According to the conception of the tectono- doza: Geol. Mag., v. 63, p. 410-422. igneous cycle that is developed in this paper Bailey, E. B., 1928, The Paleozoic mountain sys- tems of Europe and America: British Assoc. the Continental Phase is characterized by the Adv. Sci. Sect. C, Glasgow, p. 57-76. appearance of a trachybasaltic kindred and a Bailey, E. B., and Maufe, H. B., 1916, The geology of Ben Nevis and Glen Coe: Geol. Survey quartz doleritic or tholeiitic kindred. Whereas Scotland Mem., 247 p. in the Caledonian tectono-igneous cycle the Bailey, E. B., and McCallien, W. J., 1950, The trachybasaltic kindred largely precedes the Ankara Melange and the Anatolian thrust: Nature, v. 166, p. 938-940. quartz dolerite sills and dikes, this order in Bemmelen, R. W. van, 1949, The geology of Indo- time does not necessarily hold for other cycles, nesia, vol. 1A: The Hague, Govt. Printing nor does it hold for the basalt floods and asso- Office, 732 p. 1950, On the origin of igneous rocks in Indo- ciated intrusions connected with the major nesia: Geol. en Mijnb. 12, no. 7, p. 207-220. inbreaks by which parts of the oceans come into Billings, M. P., and Keevil, N. B., 1946, Petrog- raphy and radioactivity of four Paleozoic existence. There is, in fact, plenty of evidence magma series in New Hampshire: Geol. Soc. for the simultaneous availability of the two America Bull., v. 57, p. 797-828. kindreds. Both the North Atlantic basalt Bubnoff, S. von, 1923, Der Gliederung der Erd- rinde: Fortschr. Geol. Palaeont., 3, 84 p. region and the Deccan traps of India contain 1924, Die Gebirgsbau Europas: Geol. Rund- rocks assignable to the olivine basalt line of schau, Bd. 15, p. 147-174. descent (trachybasaltic kindred) and the tholei- Bucher, W. H., 1933, The deformation of the earth's crust; Princeton Univ. Press, 518 p. itic basalt line of descent (quartz dolerite or Burri, C., and Niggli, P., 1945, Die jungen Eruptiv- tholeiitic kindred). In the North Atlantic gesteine des mediterranen Orogens. I. Haupt- teil. Einfuhrung-Die Ophiolithe-Allgemeines region the general order of magmatic production iiber das Verhalten basischer Magmen: Vul- is trachybasaltic followed by tholeiitic; but, kaninst. Immanuel Friedlander, Zurich, Pub. no. 3, 654 pages. while there are many exceptions to this rule, Chamberlin, R. T., 1919, Building of the Colorado there are also later magmatic resurgences Rockies: Jour. Geology, v. 27, p. 148-164, which do follow the above general order (Tyr- 225-251. Cross, Whitman, 1912, Petrological abstracts and rell, 1949, p. 419). The petrography of the reviews: Jour. Geology, v. 20, p. 326-372. Deccan traps is not yet fully known, but, from Daly, R. A., 1933, Igneous rocks and the depths of the earth: N. Y., McGraw-Hill Book Co., 2d the literature, they seem to belong mainly to ed., 598 p. the tholeiitic affiliation. Nevertheless, dikes Dewey, H., and Flett, J. S., 1911, On some British pillow-lavas and the rocks associated with and ring intrusions of alkalic types cutting them: Geol. Mag., v. 48, p. 202-209, 241-248. and doming the Deccan traps are known in Eskola, P., 1932, On the origin of granitic magmas: Min. u. Petr. Mitt, Bd. 42, p. 455-481. the Kathiawar Peninsula (Mathur et al., Fenner, C. N., 1948, Incandescent tuff flows in 1926). This seems, therefore, to represent a southern Peru: Geol. Soc. America Bull., reversal of the more usual order. v. 59, p. 879-893. Goldschmidt, V. M., 1916, Ubersicht der Eruptiv- The tectono-igneous cycle which the writer gesteine im Kaledonischen Gebirge zwischen has endeavored to expound is based on the Stavanger und Trondhjem: Vidensk. Skrift. I. Math.-Nat. Kl., Kristiania, no. 2, 140 p. complicated, highly disturbed, and perhaps Grout, F. F., 1948, in Gilluly, James (Chairman), rather abnormal western part of the Eurasian Origin of granite: Geol. Soc. America Mem. 28, p. 45-54. continent. In conclusion the writer expresses Harker, A., 1911, Some aspects of modern petrol- the hope that American petrologists will ogy: British Assoc. Adv. Sci., Pres. Address, 12 p. now take up the task of setting forth the tec- Henderson, S. M. K., 1935, Ordovician submarine tono-igneous cycle in North America which, disturbances in the Girvan district: Royal Soc. in many respects, is the type continent of the Edinburgh Trans., v. 58, p. 487-509. Hess, H. H., 1938, A primary peridotite magma: world. Am. Jour. Sci., v. 35, p. 321-344.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021 REFERENCES CITED 425

Holmes, A., 1928, Radioactivity and earth move- phismen im Lichte der Kontraktionslehre: ments: Geol. Soc. Glasgow, v. 18, p. 559-606. Geol. Rundschau, Bd. 13, p. 217-272. 1932, Origin of igneous rocks: Geol. Mag., v. Steinmann, G., 1905, Die geologische Bedeutung der 69, p. 543-558. Tiefseebildungen unde der ophiolitischen Erup- Judd, J. W., 1886, On the gabbros, dolerites and tiva. Geologische Beobachtungen in den Alpen basalts of Tertiary age in Scotland: Geol. II: Ber. naturf. Ges. Freiburg. Bd. 16, p. 44- Soc. London Quart. Jour., v. 42, p. 49-89. 65. Kennedy, W. Q., and Anderson, E. M., 1938, Crus- 1926, Die ophiolitischen Zonen in den medi- tal layers and the origin of magmas: Bull. terranen Kettengebirgen: 14th Internat. Geol. Vole. Naples, Ser. 2, T. 3, p. 23-82. Cong., Madrid, fasc. 2, p. 637-667. Kober, L., 1921, Der Bau der Erde: Berlin, Gebrii- Tyrrell, G. W., 1914, A petrographical sketch of the der Borntraeger, 2d ed. 1928, 324 p. Carrick Hills, Ayrshire: Geol. Soc. Glasgow Krumbein, W. C., and Sloss, L. L., 1951, Stra- Trans., v. 15, pt. 1, p. 64-83. tigraphy and sedimentation: * San Francisco, 1926a, The principles of petrology: London, W. H. Freeman and Co. Methuen, 349 p. Krynine, P. D., 1941, Paleogeographic and tec- 1926b, Igneous action and earth movement tonic significance of graywackes (Abstract): in Scotland and Scandinavia: Geol. Mag., Geol. Soc. America Bull., v. 52, p. 1916. v. 63, p. 284-286. MacGregor, A. G., 1948, Problems of Carbonifer- 1931a, Volcanoes: London, Home Univ. Lib., ous-Permian volcanicity in Scotland: Geol. 256 p. Soc., London Quart. Jour., v. 104, p. 133-153. 193 lb, The South Sandwich Islands; Report Mathur, K. K., Dubey, V. S., and Sharma, N. L., on rock specimens: Discovery Repts., v. 3, p. 1926, Magmatic differentiation in the Girnar 191-198. Hills: Jour. Geology, v. 34, p. 289-307. 1933, Greenstones and greywackes: Compte Niggli, P., 1921, Lehrbuch der Mineralogie: Berlin, Rendu Reunion Internat. pour 1'etude du Borntraeger, 694 p. Precambrian et des vieilles chaines de Mon- Pettijohn, F. J., 1943, Archaean sedimentation: tagnes, Finland, 1931, Helsinki, May 1933, Geol. Soc. America Bull., v. 54, p. 925-972. p. 24-26. Ramsay, W., 1924, The probable solution of the '• 1937, Flood basalts and fissure eruption: climate problem in geology: Geol. Mag., Bull. Volcanologica, Naples, ser. 2, tome 1, p. v. 61, p. 152-163. 88-111. Read, H. H., 1931, The geology of central Suther- 1939, Relations of igneous activity to tec- land: Geol. Survey Scotland Mem., 238 p. tonics: 17th Internat. Geol. Cong., Moscow, 1948, in Gilluly, James (Chairman) Granites 1937, vol. 1, p. 213-217. and granites: Geol. Soc. America Mem. 28, 1949, The Tertiary igneous geology of Scot- p. 1-19. land in relation to Iceland and Greenland: Read, H. H., and MacGregor, A. G., 1948, The Medd. Dansk. Geol. For. 11, p. 413-440. Grampian Hills: Great Britain Geol. Survey Umbgrove, J. H. F., 1939, The relation between and Mus., 2d ed., Edinburgh, 83 p. magmatic cycles and erogenic epochs: Geol. Reynolds, D. L., 1942, The albite-schists of An- Mag., vol. 76, p. 444-450. trim and their petrogenetic relationship to Van Hise, C. R., and Leith, C. K., 1911, The geol- Caledonian orogenesis: Royal Irish Acad. ogy of the Lake Superior region: U. S. Geol. Proc., v. 48, p. 43-66. Survey Mon. 52, 641 p. Richey, J. E., 1939, The dykes of Scotland: Edin- Walker, F., 1935, The late Palaeozoic quartz- burgh Geol. Soc. Trans., v. 13, pt. 4, p. 393-435. dolerites and tholeiites of Scotland: Min. Rittmann, A., 1936, Vulkane und ihre Tatigkeit: Mag., vol. 24, p. 131-159. Stuttgart, Ferdinand Enke, 186 p. Williams, C. R., and Billings, M. P., 1938, Petrol- Scheumann, K. H., 1924, Pravariskische Glieder ogy and structure of the Franconia quadrangle: Geol. Soc. America Bull., v. 49, p. 1011-1044. der Sachsisch-Fichtelgebirgischen Kristallinen Wolff, F. von, 1914, Der Vulkanismus: Stuttgart, Schiefer I. Die magmatisch-orogenetische F. Enke, Bd. 1, 711 p. Stellung der Frankenberger Gneisgesteine: Abh. d. Math.-Nat. Kl. Sachs. Akad. Wiss. UNIVERSITY OF GLASGOW, GLASGOW, SCOTLAND Bd. 39, 61 p. MANUSCRIPT RECEIVED BY THE SECRETARY OF THE Sonder, R. A., 1922, Die erdgeschichlichen Diastro- SOCIETY, OCTOBER 1, 1952

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/66/4/405/3426860/i0016-7606-66-4-405.pdf by guest on 25 September 2021