CLASSIFICATION OF THE CRETACEOUS VOLCANIC SEQUENCES OF BRITISH COLUMBIA AND YUKON* By And& de Rosen-Spence and A. J. Sinclair Department of Geological Sciences The University of British Columbia

I

INTRODUCTION alumina arc basalts and non-arc medium-K tholeiites, which plot in the same calcalkaline domain, are well separated on the Cretaceous arc sequences of British Columbia and Yukon are of A1,0, versus Alkali diagram (Kuno, 1960); limitedextent when compared to those of the Triassic-Jurassic period.They are mostlysubalkaline but show such remarkable (3) Manyauthors are uneasy with the term ‘kc tholeiitic” fix differencesin their potash and soda contents that they seemed suites which are potassium-rich but plot in the arc tholeiitic worthy of a detailed investigation. domain, such as that of the West Carpathian arc. Such investigation entails determining and desctibingmagmatic trends, classifying the sequences, and comparing them to Recent suites. Three problems confronting the researcher at the start are: IRON CONTENT the confusing nomenclature expressing alkali and iron contents and Theiron-enrichment trend was originallynamed “tholeiitic” trends; the absenceof a classification of volcanic arcs which would because it was first recognized in tholeiitic suites, whereasthe trend reflect differences in potash content; and the difficultyof determin- lackingenrichment was describedas “calcalkaline” because it ing the original alkali and lime contents of altered sequences. In seemed typical of the calcalkaline suite of the High Cascades and consequence, a discussionof proposed changes in nomenclature, of other similar continentalarc sequences. a new classification of arcs and suites, andof a method for screening As new data on arc volcanism accumulated, it became obvious “unaltered” data will introduce this study of the Cretaceous vol- that arc volcanism was more varied than previously thought. Kuno canic sequences. The revised nomenclature and screening method (1950, 1959, 1966) showed that not only were there three paral.el weredevelopedeadier((1e Rosen-Spence, 1976). butthenew classi- volcanic chains, composed of tholeiitic, high-alumina, and alkaline fication of arcs and suites proposed here is a direct outcomeof the sequences, hut also that so-called “tholeiitic” a.nd “calcalkalint:” present study. trends could be found in anyofthe three suites, even within a single Thevolcanic suites forwhich dataare availahleinclude se- . To avoid confusion, he redefined the Fe-rich and Fe-poor quencesfrom the late Lowerand Middle CretaceousGambier, suites as “pigeonitic” and “hypersthenic” respectively on theba:;is Spences Bridge, Kingsvale, South Forks and KasalkaGroups, and of the groundmass pyroxene, and showed that they are well seFa- the late Upper Cretaceous Mount Nansen, Hutshi and Cmacks rated on theAFM diagram (Kuno, 1954; Aramaki, 1963). NC~W Groups and TipTop Hill Volcanics. This paper establishes that these these terms are rarely used; most authors use the old terminolog;y, sequencesbelong to different arc types, andto different series with resulting confusion.It is therefore proposed that the terms “Fe- withinthese arc types: further, the suites can he comparedto rich” and “Fe-poor” replace the older terms as being more descrip- specific Recent suites. It is hoped that their precise classification tive of the iron-enrichment trend (de Rosen-Spence, 1976). Chl- will be usefulfor correlation purposes and that Recent analogswill calkaline should be reserved for a specific alkali content as definsd enable reconstruction of the subduction environment in Cretaceous on the Alkali versus SiO, diagram, and tholeiitic used onlyfor true time. tholeiites. When dealing with alteredsequences the AFM diagram becomes REVISED NOMENCLATURE inaccurate becauseit is sensitive to any gainof alkali or magnesium. ALKALI CONTENT It is then preferable to use MgO versusFeO, or FeO, versus SiO, diagrams (de Rosen-Spence, 1976). In his pioneering work, Kuno (1959, 1966) divided the Alkali versus SiO, diagram into “tholeiitic”, “high-alumina” and “al- kaline” domains according to the plot of the three main chains CLASSIFICATION OF SUITES AND ARCS whichform a full arc. It is proposedhere to replace the terms Potassium is an essential element in the classification of suitss “tholeiitic” and “high-alumina” by “calcic” and “calcalkaline” and arcs. It isnow well documented that potassium increasesacross used with their strict chemical meaning (see Figure 6-8-6). This an islandarc because it reflectsdepth to the Benioff zone change is needed for three reasons: (Sugimura,1960; Kuno, 1966: Dickinson,1968); however, ar:s (I) Standardization of nomenclature because arc tholeiitic suites may differ from one another in their potash contents (Jakes and are also the most calcic, with Peacock Indicesof 67 to 64, and White, 1972). high-aluminasuites are intermediate between calcic and al- Two diagrams are useful in evaluatingthe potassium content and kaline, with Peacock Indices of 62 to 59; classifyingthediversesuitesandarcs.Theseare:(l)theK,Over~;~s (2) The domainsso redefined can then also be used to describenon- SiO, diagram of Gill (1981). modified to include basalts, rhyolitss arc tholeiitic suites with different alkali contents, such as the and alkaline potassic suites (Spence, 1985). and(2) the K,O vers 1s ridge tholeiitic suite,s of the Galapagos and Thingmuli. High- Na,O (for 70 per cent SO,) diagram developed here.

* This project is a contribution Io the CanaddBritish Columbia Mineral Development Agreement. British Columbia Ministry of Energy, Mines and Petroleum Resources, Geological Fieldwork, 1986, hper 1987.1 419 POTASH CONTENT WITHIN A SINGLE ARC The low-K (LK), medium-K (MK) and high-K (HK) domainson the K,O versus SiO, diagram (Figure 6-8-11 are used mainly to describe the potassium contentof a suite, bearingin mind that actual trendsmay be steeper than the domain boundaries and that an additional sharp increase inK,O may occur in the rhyolites.On this diagram, the three series of any single arc are well separated by sharpboundaries. In some cases,as in theCascades, the cal- calkdine series itself can be subdivided (Wise, 1969, page 1003) into a "uue Cascades" type () with a Peacock Indexof 62, and a "high-alumina" type (Badger Butte, Newberry) with a Peacock Index of 60.5, perhaps on the verge of being alkaline. If, however, arcs with different potash contents are plottedon the same diagram, there is a marked overlap of the calcic, calcalkaline and alkaline series of the different arcs. This is well shown in Figure 6-8-1 by the three positionsof the alkaline and calcic boundariesof the Cascades (I), Honsyu and Hokkaido (11) and West Carpathian and Eolian (111) arcs. I P(rTASHCONTENT0FARCS 45 50 15 60 65 70 75 S#02 AND THEIR CLASSIFICATION Figure 6-8-1. K,O versus SiO, showing the calciclcalcalkaline The new diagram presented here is K,O versus Na,O for 70 per (dotted lines) and calcalkalinelalkaline (heavy lines) boundaries for cent SiO, (Figure 6-8-21,It takes intoaccount theantipathetic arcs with different K,O contents: I = Fiji and Cascades (Shasta and variation of sodium and potassium in suites which have the same Mount Hood); I1 = Honsyu and Hokkaido; 111 =West Carpathian total alkali content but different potash content (up to1.5 per cent and Eolian arcs [domains from Gill (1981) and Spence (198511. difference).Itwasdesignedfor70percentSiO2toofferamoreopen plot, although this hasthe disadvantage thatK,O must be projected on K,O versus SiO, when dealing with less siliceous sequences. This diagram was constructed with data from Kuno(1962) Fiske ef al.(1963),Karoluseral.(l968),Wise(1969),Gill(1970),Ewaner al. (1973), Higgins (1973). Barbieri er al. (1974) and Cantagrel ef al. (1981). On it, arcs can be classifiedinto three types, each containing calcic, calcalkaline and alkaline series: "pe I (more sodic): Izu, Fiji, Cascades, Qpe I1 (moderately sodic): Tonga, Honsyu, Hokkaido, Qpe I11 (less sodic): West Carpathian, Eolian arcs. Among the calcic series, the low-K arc tholeiitic ofsuites Fiji and Izu Islands (TypeI), and of Tonga ('Qpe II) were deposited on thin oceanic CNS~, whereasthe medium-K pre-Mount Hood dacites, Shasta and Hakone (ppe I) and Honsyu-Hokkaido arc tholeiitic suites (Type 11) were deposited on thicker CNS~. The unusual West I- Carpathian arc has the highestK,O content of the calcic series, and is considered to be an intracontinental arc depositedon thick conti- '" nental CNSt, possibly above subducted continental CNS~(Channel and Horvath, 1976).

CLASSIFICATION OF RECENT SUITES Figure 6-8-2. K,O versus Na,O for 70 per centSiO, showing the Table 6-8-1 illustrates how Recent suites are classified according distribution of Recentarcs and individual volcanoes into three to their total alkali, potash and iron contents using Alkali versus types: I= Fiji-Izu-Cascades; II=Tonga-Honsyu~Hokkaido; SiO,, K,O versus SiO,, Na,O versus K,O for 70 per cent SiO, and III= WestCarpathian-Eolian. (Type I: FI,2=Fiji arc; AFM or FeOT versus SiO, diagrams. When rhyolites are absent G = Garibaldi; H = Hakone (Izu Peninsula); I = Iru Islands; MH- from the sequence plotted, N$0 and K,O values for 70 per cent =MountHood; MR=; S= MountShasta; SiO, were projected on Na,O versus and K,O versus SO,. Na,O N = Newberry; NT =Novo de (); PMH= Pre-Mount does not increase above 65 per cent SO, whereas K,O tends to Hooddacites.TypeII:T-Tonga;HKI,2,3=Hokkaidoarc;HSI, increase more rapidly, and care must be taken that the projected 2=Honsyu arc; SS = Sidara sheet (Honsyu). Type III: WC = West values fall in the appropriate total alkali domain on Na,O versus Carpathian arc; El, 2, 3=Eolian arc). K,O.

RETRIEVAL OF "UNALTERED" DATA FROM ALTERED SEQUENCES CaO LOSS Original magmatic trends in altered sequences generally can be CaO is lost in many types of alteration unless it is trapped as determined by screening the data on certain diagrams in order to calcite.Treatment of dataon MgO versus CaO(Figure 6-8-3) retrieve "unaltered" or "least altered" analyses. separates a large number of altered samples. This diagram has the

420 TABLE 6-8-1. event is subdivided into two episodes, Aptian(?I to Albian, andlate CLASSIFICATION OF RECENT Alhian to Cenomanian, separatedby the uplift of the Coast Plutonic SUBALKALINE ARC SUITES Complex. Remnants of the once extensive submarine AlbianGlm- bier Group are found in roofpendantsof the soutlnern Coast Plutonic Complex. In south-central British Columbia, the subaerial arc ,;e- t TYPE I t TYPEII t TYPEIII .~...... ~~~ ...... ~ ...... ~ ...... ~~~ ...... ~ """ quences of the late Albian Spences Bridge Gmup and AIbian :?) t Fe- t Fe- t Fe- "Kingsvale" Group near Aspen Groveoccureast of the Fraserfal It. + pmr ich t pr ich t pr ich those of Kingsvale Group of Chilko Lake foml a thick sequerce south of the Yalakom fault and are dated as Cenomanian. In Ihe t + + Bella Coola area, submarine and subaerial sequences may belong VHK t t t Eolim.3 respectively to the Gambier and Kingsvale Groups. In central Brit- ish Columbia, the shallow marine Albian volmnics of the Skeena Group and the later subaerial Cenomanian (?) Kasalka Group are MK + t t but small remnants. In the Yukon, the unusual subaerial Albian t + + South Forks Volcanics developedon the North American platforn. The second period,in Maastrichtian time (Grondef a[., 198411,is t t + t one of more subdued subaerial volcanism alonga narrow belt which HK + t + Eolian.2 extends from Yukon to northern British Columbia;it is represented t t Hokkaido.3 t Eolian.1 by the Mount Nansen, Carmacks and Hutshi IGroups. In centxl ...... Newberry ."""~ ...... ~ ...... t MI. Hood t ----Honsyu.Z----t British Columbia, the Tip Top Hill Volcanics mark the southern CAE- MK t St. Helens t ---Hukkaido.Z--~t extension of this belt. Plutonism was also we:& and only snnll ALKALINE +Rainier t Lassen t plutons are found (Armstrong, 1986, in press). + Flji.2 Huzi + Talasea t LK t t t GAMBIER GROUP t t t t The Gambier Group is a submarine arc assemblage of basalts, t t t andesitesand dacite flows and tuffswith associated flysch ;and HK t t t argillites. In theHarrison Lake area, thereis evidence (Arthr, 1986) of episodes of Middle Triassic, Lowerand Upper Jurassica Id t Shasta t t early Lower Cretaceous arc volcanism, all separated by unconlor- CAKCIC MK +Be- t ----Hansvu.l----t West Car- mities. On Gambier Island, the Gambier Group rests unconforrna- tMt. Hwd t pthm + bly on folded, intruded and eroded Triassic (?) greenstones of tle t ~--Hkone -~~t ---Ho!&aida.l--- t Bowen IslandGroupand Late Jurassic diorite (Roddick, 1965).It is + t t intruded by late Albian (?) and Cenomanian plutons (White, 19611). LK t lmt Tonga t The existence of older arc sequences and intrusive rocks, together + Fiji.1 t t with reported Carboniferous zircon ages (Roddick et al., 1979, indicates that the Gambier Group wasdepositedonawell-develop:d XC CNSl. advantage of being valid for all volcanic suites. though lavas with plagioclasecumulates plot as enriched in calcium(de Rosen- BRITMNIA MINE AREA Spence, 1976; Spence, 1985). Datadescribed following are from Margaret McColl (M.Sc. thesis, The University of British Columbia, in preparation). The Na,O GAIN OR LOSS Gambier Groupin the mine area isaltered (Figure6-8.3). basalts are Plotting the screened "unaltered" data from Figure 6-8-3 on an spilitized and dacitic flowsand tuffs are mainly sericitized(Figurzs Na,O versus SO, (Figure 6-8-4) allows thc eliminationof spilitized 6-8-4 and 6-8-5). Andesitic and dacitic dykes and some massvie or sericitized samples whereCaO was retained as calcite,as well as dacite samples plot as "unaltered" on MgO vmsus CaO (Figure sampleswith marked K-Na exchanges.Such exchanges can he 6-8-3) and give consistent magmatic trends. From these, the Britan- recognized by the antipathetic variationof potassium and sodiumon nia mine sequence can be dehcd as a medium.-K (Figure 6-8-2). Na,O versus SiO, and K,O versus SiO, (Figure 6-8-5) plots. calcic (Figure 6-8-6) suite with a Peacock Index of 64; it is an wc tholeiiticsequence. It is Fe-poor,though close to the Fe-rich MgO GAIN (tholeiitic) boundary, and belongs (Figure 6-8-7:l to a Type I arc ,IS defined previously. Heah et a/. (1986) found that basalts of lhe Rocks which have gained MgO also plot as "altered" and MgO nearby Sky Pilot area arealso arc tholeiites showing Fe-rich and17,:- gains can he assessed by plotting MgO versus SiO, (Spence, 1985, poor Wends. The Gambier Grollp is close in cmsmposition to the Figure 136). Care must be taken though, that there has been no and Kuroko suites. The presenceof older arcCNS~ hilt silicification as this would also cause a shift of the data points into deepmarine conditions,together withvolcanogenic deposits(Payrle the magnesium-enriched field. er al., 1980). suggestsan intra-arc extensional environment similar to that of the Green Tuffs-Kuroko trough in Japan (Sillitoe, 19X:!; CRETACEOUS VOLCANISM OF Cathles et al., 1983). BRITISH COLUMBIA AND YUKON Two periods of Cretaceous volcanism have been recognized in the HARRISON LAKE AREA Canadian Cordillera by numcrous workers and re-emphasized by On the west shore of HarrisonLake. the Ciambier Group IS Armstrong (1986, in press): one is in the late Lower and Middle represented by the Fire Lake Group theand Doctor's Point volcanirs Cretaceous, the other in latest Upper Cretaceous time. The first (Ray et a/.,1985). The Fire Lake Group is composed of altered period is also one of vigorous plutonism and uplift, conscquently andesites, whereas the Doctor's Faint volcanics include dacites, and much of the contemporaneous volcanic rock has been eroded. The are relatively well preserved. The latter sequence plots as a low-I< 42 1 MgO

No& 10’* t

8-

.“

45 50 155 61 65 70 75 SlO*

Figure 6-8-3. MgO versus CaO. Gambier Group, Britannia mine Figure6-8-4. N%Oversus SO,. GambierGroup, Britanniamine (circles = lavas; triangles= tuffs; squares= dykes in the mine; open (notice the spilitized basalts and Na-depleted tuffs) (domains from symbols = altered; filled symbols = “unaltered’). Analyses from de Rosen-Spence, 1976). M.McColl (in preparation), domains from de Rosen-Spence (1976).

Figure 6-8-5. K,O versus SiO,. Gambier Group, Britannia mine Figure6-8-6. Alkali versus SiO,. Gambier Group, Britannia (notice the well-defined trendof the “unaltered” dykes and the high mine (notice the well-defined trendof the “unaltered” dykes) (do- K,O content of the tuffs). mains from Kuno, 1966).

422 ”.

(Figure 6-8-8). calcic (Figure 6-X-9) suite of Typ I (Figure 6-8-7) posed of differentiated flows andtuffs (Wood and Armstrong, similar to the Britannia sequence, though apparently less potassic. It1982). Theseare described as potassic with a calcalkaline(Fe-poor) is generally Fe-poor except for a few Fe-rich basalts (G.E. Ray, trend and are characterizedby a high initial strontium ratio indica:. personal communication, 1986). ing a strong crustal influence. They have been datedas Albian and are intruded by a Cenomanian quartz monzonite (Wood and Arm MOUhT RALEIGH PENDANT strong, 1982). The Mount Euridyce dacite is a low to medium (?)-K dacite of The South Forks Volcanics plot as a high-K (Figure 6-8-121, Type I (Wwdsworth, 1979). calcic (Figure6-8-8) suite with a PeacockIndex of66. They are lfery calcic and poor in sodium (Figure 6-8-9) in spite of their high CALLAGHAN PENDANT potassium content and distinctly belong an to arc ofType 111 (Fipre The andesites aroundthe Northair mine are heavily altered. They 6-8-7). The most comparable Recent suite is that of the intraccrti- are subalkaline and aluminous, and possibly- but not certainly - nental West Carpathianarc I:Karolus et a/., 1968).The quartz akin to those of the Britannia mine area (Miller, 1979). monzonite, alsoofType 111, iscalcalkaline, indncating an increase in alkali with time. SPENCES BRIDGE AND KINGSVALE GROUPS KASALKA GROUP The SpencesBridge and Kingsvale Groupswere first defined east Maclntyre (1976) showed that the andesites and rhyolitesof the of the Fraser fault, where they form a 2M)-kilomet~-long belt of Kasalka Group were preserved in a cauldron subsidence complex. differentiated subaerial volcanics resting unconformably over the They were deposited subaerially,in early Late Cretaceous time, over Triassic and Jurassic volcanicsof Stikinia. Both have recently been a folded and eroded sequenceof the shallow marine SkeenaGroup re-examined and analysed (Thorkelson, 1986) and redated by fos- of Albian age. The sequence was intruded by the Mount Bolurn sils lo be of Late Albian age (Thorkelson and Rouse,in preparation). granophyres and the varied plutons of the Bulkley suite. The liuter In the Kingsvale area, the two groups were found to be in strat- was dated as latest Upper Cretaceous (Carter, 1982). conterrpo- igraphic continuity and similar in major element chemistry. As a raneous with the next volcanic period. result the KingsvaleGroup has now lost its statusas a group. The Kasalka Group data plot along the medium-K to high-K Kingsvale andesites were renamed Spius Formation and integrated boundary(Figure 6-8-13), calcalkaline (Figure 6-8-14) and Fe- into the Spences Bridge Group. Near Aspen Grove, to the east, a poor, and belonging to a Type 11, near Type I arc (Figure 6-8-71. small inlier of Lower Cretaceous volcanics was attributed to the similar to the Kingsvale Group of Aspen Grove. The Mount Bolum Kingsvale Group by Preto (1979). Analyses from both sequences granophyres are at the alkaline limit whereas the Bulkley intrusions will be examined following. In the Chilko Lake area south of the are calcalkaline and less sodic (Figure 6-8-7). Yalakom fault, other subaerial volcanicswere also attributed to the Kingsvale Group hut are reponed to be unlike those east of the MOUNT NANSEN GROUP AND Fraser fault and have been dated as Cenomanian (Kleinspehn, TIP TOP HILL VOLCANICS 1985). No analyses are available. Datapresented are fromCiond PI al. (1984) for the Maunt SPENCES BRIDGE GROUP IN NansenGroupinYulronandfr~~mChurch(197(1)foroneanalysisof KINGSVALE AREA the Tip Top Hill Volcanics in #central British Columbia. This sub- aerial volcanism is similar to that of the Kasalka Group (Figure; Thorkelson (1986) showedthat the Lower Spences Bridge Group 6-8-13,6-8-14and6-8-7).TheMontanaMount~insequence,wtich and the Spius Formation, though similar in their major element belongs to the Hutshi Group (Roots, 19821, is also similar but ve?’ compositions, differintheirtitanium, phosphorusand traceelement altered. contents. The Spius andesites seem to have a plume component indicating the beginning of rifting of the Spences Bridge arc. CARMACKS GROUP Plots of the Spences Bridge Group reveal that it is a medium-K (Figure 6-8-81. calcic (Figure 6-8-9) suite with a Peacock Index of Datapresented are from Grand et a/. (1984). The Cmack:i 65 to 66; it is Fe-poor and belongs to a ppe II arc (Figure 6-8-7) Group overlies the Mount Nansen Group and has the same rarlio- similar to the hypersthenic arc tholeiitic suite of Honsyu. metric age. It is composed of shoshonite flows and of breccia:; including calcalkaline andesite clasts (Figures 6-8-13 and 6-8-1 4). “KINGSVALE’ GROUP OF ASPEN GROVE This shoshonitic volcanism is important as it indicates disturbance through collision of the MaastrichUian subduction zone. Reto(1979)recognizedtwounits. lOand11,intheAspenGrove area and correlated them with the Kingsvale Group. Unit IO, con- taining andesites and rhyolites, was broadly dated as Albian and is CONCLUSION intruded by a Cenomaniao granite. Unit 11 is basaltic, is not dated, and isnot in contact with UnitIO. Both units belong to a medium-K The methods presented here for screening altered samples and (Figure 6-8-10). calcalkaline (Figure 6-8-1 I) suite with a Peacock classifying volcanic suites have allowed more accurate definitlon Index of 62. They are Fe-poor and belong to an arc transitional and comparisonof Cretaceous volcanic sequences with morerec,:nt between spes I and II iFigure 6-8-7). Unit I1however, is slightly suites. It is hoped that this may eventually lead to a more accurate richer iniron, titanium and phosphorus, suggesting latea or behind- reconstruction of the old subduction zones and arc margins. the-arc setting. The Kingsvale sequence from Aspen Groveis there- The main results of interest are: fore different in total alkali content fromthe Spences Bridge Group, (I) Identification of the Gambier Groupas an aic tholeiitic suite of including the SpiusFormation. suggesting that it belongsto a Type I, similarto other arc sequences hosting copper-zinc different volcanic event. massive sulphides, such as the Miocene Kuroko and Archean Noranda sequences, SOUTH FORKS VOLCANICS (2) Identification of the Spences Bridge Group and South Forks The South Forks volcanic rocks are a subaerial intracontinental Volcanics - previouslydescribed as “c;dcalkaline” - as arc sequence deposited on the North American platform and com- calcic suites (Peacock Index of 66) with high potash and CDI- 423 I LK

45 50 55 60 55 70 15 s,02

Figure 6-8-7.K,O versus Na,O for 70% SO,, showingthe Figure 6-8-8. Alkali versus SiO, for the Gambier and Spences distribution of the Cretaceous arcs into three types: I = Gambier Bridge Groups and South Forks Volcanics. Legend as in Figure Group (GDP=DocIor's Point; GB=Britannia mine) and Noranda 6-8-12, (N) and Kuroko (K) dacites for comparison; 11 = Spences Bridge (SB),"Kingsvale" of AspenGrove (KI), Kasalka and Mount Nansen (KA) Groups and Mount Bolum (BO) and Bulkley (BU) intrusions; Ill = South Forks Volcanics(SF) and Quartz Monzonite (QM).

Figure 6-8-9.Na,O versus SiO, for the Gambier and Spences Figure 6-8-10. Alkali versus SiO, for the "Kingsvale" Group of Bridge Groups and South Forks Volcanics. Legend as in Figure Aspen Grove (UnitIO = circles; Unit11 =triangles). Analyses from 6-8-12. Preto (1979).

424 45 50 55 60 65 70 15 S102

Figure 6-8-1 I. K,O versus SiO, for the "Kingsvale" Group of Figure 6-8-12. K,O versus SiO, for the Gambier Groupin Doc-. Aspen Grove (Unit IO=circles: Unit 11 =triangles). tor's Point areas (dots), and Britannia mine area (VVV), Spen:e!, Bridge Group (lines) and South Forks Volcanicr, (dashes).Analyse fromRay (personal communication), McColl (in preparation), Thorkelson (1986) and Wwd and Armstrong (1982).

All

12

I"

lo t / .lily B

/ ,

~, q5 50 55 6" 65 70 is S,Ol 45 50 55 6" 65 70 75

Figure6-8-13. Alkali versus SiO, for the Kasalka(circles), Figure6-8-14.K,OversusSiO~fortheKasalka,MountNansc~, Mount Nansen (squares), Tip Top Hill (diamond) and Carmacks Tip Top Hill and Carmacks Groups. Legend as in Figure 6-8-13, (triangles)Groups. Analyses from Church (197O), Maclntyre (1976) and Grond et ai. (1984).

425 relatively low soda contents, and their comparison to the Hon- Dickenson,W.R. (1968):Laye Cenozoic Shonshonitic Lavas in syu arc tholeiites and West Carpathian arc respectively; Northwestern Vitu, Fiji, Nature, Volume 219, page 148. (3) Differentiation of Spences Bridge Group and Kingsvale vol- Ewarr, A,, Bryan,W.B. and Gill, J.B. (1973): Mineralogy and canic rocks fromthe Aspen Grove area, which precludes their Geochemistry of the YoungerVolcanic Island onTonga, correlation; Southwest Pacific, Journal of Petrology, Volume 14, pages (4) Identification of important differences between the Gambier 429-465. and Spences Bridge Groups indicating thatthey belong to two Fiske, R.S., Hopson, C.A. and Waters, A.C. (1963): Geology of different episodes or to two different arc segments. Similar Mount Rainier National Park, Washington,UnitedStaresGeo- differencesare found between the Miocene Green Tuffs- logical Survey, Professional Paper No. 444, pages 1-93. Kuroko rhyolites and the present Honsyu arc in the same area, Gill, J.B. (1970): Geochemistryof Viti Levu andits Evolution as an or between the Recent Izu and Honsyuarcs. This difference in Island Arc, Contributions to Mineralogy and Petrology, Vol- composition is thought to result from differencesin the condi- ume 27, pages 179.203. tions of subduction in time and/or location; __ (1981): Orogenic Andesites and Plate Tectonics, (5) Recognition that there is a need for data from the Cretaceous Springer-Verlag, New York, 390 pages. sequences of the Chilko Lake and Bella Coola areas. Green, N.L. (1977): Multistage Andesite Genesis in the Garibaldi LakeArea, Southwestern Brit.ish Columbia,Unpublished Ph.D. Thesis, The Univerrity British Columbia. ACKNOWLEDGMENTS of Grond, H.C., Churchill, S.J., Armstrong, R.L., Harakal, J.E. and This paper represents a small part of a study of whole rock Nixon, G.T.(1984): Late Cretaceous Age of the Hutshi, chemical data for volcanic sequences in British Columbia under- Mount Nansen and Carmacks Groups, Southwestern Yukon taken at The University of British Columbia, and funded by the Territory and Northwestern British Columbia,Canadian Jour- Science Secretariat of British Columbia and the British Columbia nal ofEarth Sciences, Volume 21, pages 554-558. Ministry of Energy, Mines and Petroleum Resources through the Heah, T.S.T., Armstrong, R.L. and Woodsworth, G.J. (1986): The CanaddBritish Columbia MineralDevelopment Agreement. We Gambier Group in the Sky Pilot Area, Southwestem Coast wish to thank M. McColl, G. Ray and D. Thorkelson for their Mountain, British Columbia, Geological Survey of Canada, unpublished analyses and R.L. Armstrong for discussions on Dr. Paper WIB, pages 685.691. Cretaceous volcanism. Higgins, M.W. (1973): Petrology of Newberry Volcano, Central Oregon, Geological Sociery oJAmerica, Bulletin, Volume 84. REFERENCES pages 455-488. Aramaki, S. (1963): Geologyof Asama Volcano, Tokyo University, Jakes, P. and White,A.I.R. (1972):Major and Trace Element Facultyof ScienceJournal, Section 14, Number 2, pages Abundances in Volcanic Rocks of Orogenic Areas,Geological 229-443. Society of America, Bulletin, Volume 83, pages 29-40. Arthur, A.S. (1986): Stratigraphy along the West Side of Harrison Karolus, K., Fogal, J. and Kdnecny. V. (1968): Neovolcanics of the Lake, Southwestem British Columbia, Geological Survey of West Carpathians, 23rd International Geological Congress, Canflda. Paper 86-lB, pages 715-720. Czechoslovakia, FieldExcursion 18AC, Guidebook, page 26. Armstrong, R.L. (1986, in press): Mesozoic and Early Cenozoic Kleinspehn, K.L. (1985): cretaceousSedimentationandTectonics, Magmatic Evolution of the Canadian Cordillera, Geological vaughton-MethowBasin, SouthwesternBriitsh Columbia, Society @America, Memoir. CanadianJournal of EarthSciences, Volume 22, pages 154-174. Barbieri, E, Innocenti, E, Ferara, G., Keller, 1. and Villari, L. (1974): Evolution of Eolian Arc Volcanism,Earth and Plane- Kuno, H.(1950): Petrology of Hakone Volcano (Japan),Geological tary Science Lerrerr, Volume 21, pages 269-276. Society of America. Bulletin, Volume 61, pages 957-1014. Cantagrel, J.M., Robin, C. and Vincent, P. (1981): Les Grandes -(1954): Geology and Petrologyof Omuro-Yama Volcano Etapes d’Evolution d’un Volcan Andesitique Composite: Ex- Group, North Izu, Tokyo University, Faculty of Science Jour- emple due Nevado de Toluca (Mexique), Bulletin, Vol- nal, Section 29, Part 2, pages 241-265. canology, Volume 442. -(1959): Origin of Cenozoic Petrographic Provinces of Carter, N.C. (1982): Porphyry Copper and Molybdenum Deposits, Japan and Surrounding Areas, Bulletin, Volranology, Volume West-central British Columbia, B.C. Ministry of Energy. 20, pages 37-76. Mines and Petroleum Resources,Bulletin 64. (1960): High Alumina Basalt,Journal oJPetrology, Vol- Cathles, L.M., Guber, A.L., Lenagh, T.C.and Dudas. F.O. (1983): ume I, pages 121-145. Kuroko-type Massive Sulfide Deposits of Japan: Products of an Aborted Island-arc Rift,Economic Geology, Monograph 5, (I 962): Catalogueof Active Volcanoesin the World,Part pages 96-114. X,Japan, Taiwan and Marianas, International Volcanological Channel, JET. and Horvath, F. (1976): The AfricaMAdriatic Pro- Association, Naples, 332 pages. montory as a Paleogeographical Premise for Alpine Orogeny -(1966): Lateral Variation of Basalt Magma Types across and Plate Movements in the Carpatho-Balkan Region. Tec- Continental Margins and Island Arcs, Bulletin, Volcanology, tonophysics, Volume 35, pages 71-101 Volume 29, pages 195-222. Church, B.N. (1970): GeologyoftheOwen Lake, Parrot Lakes and Lipman,P.W.andMullineaux,D.R.(1981):The1980Eruptionsof Goosly Lake Area, B.C. Ministv $Energy. Mines and Pe- Mount St. Helens, Washington, UnitedStares GeologicalSur- troleum Resources. Geology, Exploration and Mining in Brit- W.Y. Professional Paper 1250 ish Columbia, 1970, pages 119.125. Maclntyre, D.G. (1976): Evolution of Upper Cretaceous Volcanic deRosen-Spence, A.F. (1976): Stratigraphy, Development and Centres and Associated Porphyry Copper Occurrences,Tahtsa Petrogenesis of the Central Noranda Volcanic Pile, Noranda, Lake, British Columbia, Unpublished Ph.D. Thesis, Univer- Quebec, Unpublished Ph.D. Thesis, Universit? of Toronto. sir?, of Western Ontario. 426 McColl,M. (in preparation): M.Sc.Thesis, The University of Smith, A.L. andCarmichael, I.S.E. (1968): Quaternary lmas British Columbia. from the Cascades, Western United States, Contributio,w to Miller, J.H.L. (1979): Geologyof the Central Part of the Callaghan Mineralogy and Petrology, Volume 19, pages 212-238. Pendant, Southwestern British Columbia, Unpublished Spence, A.F. (1985): Shoshonites and Associated Rocksof Central M.A.Sc., The Universiry of British Columbia. British Columbia, B.C. Ministry of Energy, Mines and PC- Payne, J.G., Bratt, J.A. and Stone, B.G. (1980): Deformed Meso- rroleumResources, GeologicalFieldwork, 1984, Rlpfr zoic Volcanogenic Copper-zinc Massive Sulfide Deposits in 1985-1, pages 426-442. the Britannia District, British Columbia, Economic Geology, Sugimura, A. (1960): Zonal Arrangementof someGeophysical and Volume 75, pages 700-721 Petrological Features in Jnpan and its Environs,Tokyo Unwer- sity, Faculty of ScienceJournal, Volume 12, 2, pages Preto, V.A. (1979): Geology of the Nicola Group between Merritt Part 134.154. and Princeton, British Columbia, B.C. Ministry of Energy, Mines and Petroleum Resources, Bulletin 69. Thorkelson, D. (1986): Volcanic Stratigraphy and Petrology orthe Mid-Cretaceous Spences Bridge Group near Kingsvalc, Ray, G.E.,Coombes, S., MacQuarrie,D.R., Niels, R.J.E., Southwestern British Columbia, Unpublrshed M.Sc. Thfsi!:, Shearer, J.T. and Cardinal, D.G. (1985): Precious Metal Min- The University ofBritish Columbia. eralization in Southwestern British Columbia,Geological So- White, W.H. (1968): Granitic Rocks of Southwestern British Cw ciety ofAmerica, Field Trip 9, May 1985, pages 9-1 lo 9-1 1 lumbia, TheUniversity of BritishColumbia, Department Roddick, J.A. (1965): Vancouver North, Coquitlam and Pitt Lake Geology Report 6, pager, 13-17. Map Areas, British Columbia, Geological Survey ofCanada, Wise, W.S. (1969): Geology and Petrology of the Mount Hood Memoir 335, 276 pages. Area: A Study of High Cascade Volcanism, Geological Set- Roddick,J.A., Muller, J.E. and Okulitch, A.V. (1979): Fraser ciery ofAmerica. Bulletin. Volume 80, pages 969.1006. River, Geological Survey ofCanada, Map 1386A. Wood,D.H.andArmstrong,R.L.(1982):Gec~logy,Chemisvyand Roots,C.F. (1982): Geology of theMontana Mountain Area, Geochronometry of the Cretaceous South Forks Volcartic!;, Yukon,Unpublished M.Sc. Thesis, Carleton Univerriry, YukonTerritory, in CurrentResearch, Part A, Geolo~icd Ottawa. Survey ofCannda, Paper 82-IA. pages 309-316. Sillitoe, R.H. (1982):Extensional Habitats of RhyoliteHasted Woodsworth, G.J. (1979): Mctamorphism, Deformation and Plu- MassiveSulfide Deposits, Geology. Volume IO, pages tonism in the Mount Raleigh Pendant, Coast Mountains, Brit- 403-407. ish Columbia, Geologicnl Survey of Canada, Bulletin :?Y:i.

427