The origin of a tephra-like bed near Mount Cayley volcano, southwestern British Columbia, Canada1 Gregory R. Brooks and Edward J. Hickin Abstract: A distinct white bed 1-2 cm thick is present within backwater deposits at the upstream end of the Turbid Creek debris fan, southwestern British Columbia, Canada. This white bed and the enclosing backwater deposits accumulated within an impoundment of Squamish River caused by a large ca. 4800 BP debris avalanche originating from Mount Cayley, a Pliocene-Pleistocene volcano. The white bed is composed of silt and clay detritus and resembles tephra. About 5 % of the grains exhibit optical characteristics consistent with volcanic glass. The remainder, however, are mineral and lithic particles (mainly cristobalite and Na-feldspar, with minor quartz, and trace amphibolite, mica group, and chlorite). The white bed is interpreted to be dust derived from the debris avalanche and washed into the impoundment, based upon (i) the high proportion of mineral and lithic grains within the bed, (ii) the contemporaneity of the bed and the debris avalanche, (iii) the lack of Holocene volcanic activity at Mount Cayley, and (iv) the significant age difference between the bed and known tephras in southwestern British Columbia. R&um6 : Une lit de couleur blanche ais6ment reconnaissable, d'une 6paisseur de 1-2 cm, est interstratifik dans les d6p6ts du r6servoir d'eaues de retenue, juste en amont de l96ventai1de debris de Turbid Creek. Ce lit de couleur blanche et les dCp6ts des eaux de retenue qui le confinent, se sont accumults dans un r6servoir sur la riviBre Squamish crte par le barrage Mifi6 lors d'une grosse avalanche de dkbris, survenue aux alentours de 4800 Av.P., qui a d6vale du mont Cayley, un volcan d'gge Pliockne-Pl6istockne. Le matiriel de ce lit blanc est form6 de silt et d'argile et il ressemble h un tephra. Environ 5 % des grains exhibent des propri6t6s optiques analogues i celles d'un verre volcanique. Le reste est compos6 de petits grains de min6raux et de roches (principalement de la cristobalite et du feldspath sodique, avec un peu de quartz et de trks petites quantit6s d'amphibolite, de min6raux du groupe des micas et de la chlorite). Les observations suivantes : (i) la proportion 6lev6e de grains de min6raux et de roches dans le lit; (ii) la contemporan6itt du lit et de l'avalanche de d6bris; (iii) l'absence d'activite volcanique holockne au mont Cayley; et (iv) la diffkrence importante d'gge entre ce lit et les tephra connus dans le sud-ouest de la Colombie-Britannique, suggbrent que ce lit de couleur blanche est un d6p6t 6olien d6riv6 d'une avalanche de d6bris et d6lav6 dans un reservoir d'eau de retenue. [Traduit par la redaction] Introduction Tephra beds can provide information about the age, magnitude, and character of prehistoric volcanic eruptions (Fisher and Schmincke 1984), which is important in assessing hazards associated with future eruptions (e.g., Crandell et al. 1975). Tephras are also useful stratigraphic marker beds for some geological and archaeological studies. A considerable literature exists on the identification and correlation of tephra Received February 15, 1995. Accepted August 14, 1995. G.R. Brooks? Geological Survey of Canada, 601 Booth Street, Ottawa, ON KIA OE8, Canada. E.J.Hickin. Department of Geography and The Institute for Quaternary Research, Simon Fraser University, Burnaby, BC V5A lS6, Canada. * Geological Survey of Canada Contribution 11493. Corresponding author (e-mail: gbrooks@gsc.emr. ca). beds (see recent reviews by Sarna-Wojcicki and Davis 1991; Knox 1993). The misidentification of a bed that exhibits some characteristics resembling a "true" tephra can cause erroneous interpretations of the local stratigraphy and, more importantJy, the eruptive history of the volcano from which the bed is attributed to have originated. This paper reports a distinct white tephra-like bed within backwater deposits that accumulated in an impoundment of Squamish River caused by a large debris avalanche from bunt Cayley volcano. The interpretation of this bed has important implications for the Holocene history of Mount Cayley, which is generally believed to have been dormant since at least the Late Pleistocene. In this paper, we consider eruptive and noneruptive origins for the white bed. Background Mount Cayley is a large v o l c a ~ ccomplex in the central Garibaldi Volcanic Belt (Fig. 1) (Souther 1980). The com- Can. J. Earth Sci. 32: 2040-2045 (1995). Printed in Canada I Imprim6 au Canada I I I i Brooks and Hickin Fig. 1. Location map of the Mount Cayley volcanic complex. Also shown are the Quaternaq volcanic centres of the Garibaldi Volcanic Belt and distributions of Mazama and Bridge River tephras (modified from Roddick et al. 1979; Green et al. 1988; Mathewes and Westgate 1980; Westgate and Naeser 1985). Fig. 2. Map showing the location of exposures of the white bed at the edge of the Turbid Creek debris fan (1 ft = 0.3048 m). COLUMBIA QUATERNARY VOLCANIC CENTRES OF THE ' DISTRIBUTION OF MAZAMA TEPHRA APPROXIMATE DISTRIBUTIONOF posite cone consists of andesite, dacite, and minor rhyodacite flows, domes, and pyroclastic deposits that formed in three distinct eruptive stages during the Pliocene and Pleistocene (Souther 1980; Clague and Souther 1982). Rock from the last eruptive stage has been K- Ar dated at 310 & 50 ka (Green et al. 1988). The rugged edifice is deeply dissected with a large amphitheatre-shaped basin carved into its southwestern flank (Fig. 2). The steep slopes of this basin have a long history of large landslides, several of which descended dlrectly into Squamish Valley and impounded Squamish River (Brooks and Hickin 1991; Evans and Brooks 1991, 1992; Brooks 1992). The backwater deposits containing the white bed relate directly to a massive collapse of the volcanic cone at ca. 4800 BP (Evans and Brooks 1991, 1992; Brooks 1992). The debris avalanche descended the slopes of Mount Cayley, crossed Squamish Valley, and spread upstream and downstream along Squamish River. The Turbid Creek debris fan, produced by this landslide, extends about 6 km along Squamish River and covers an area of about 8 km2 (Fig. 2) (Evans and Brooks 1991). The landslide debris impounded Squamish River and formed a reservoir that inundated the valley bottom and extended upstream into the lower portions of Elaho and upper Squamish valleys (Brooks and Hickin 1991). Sediment carried by Elaho and Squamish rivers accumulated within this reservoir (Brooks and Hickin 1991). Subsequent incision of the debris fan by Squamish River breached the reservoir, restoring a riverine environment to the valley bottom and exposing the debris avalanche and backwater deposits. The white bed Distribution We have found the white bed only in backwater deposits of the ca. 4800 BP debris avalanche at the upstream end of the Turbid Creek debris fan (Figs. 2, 3). Exposures containing the white bed are located up to 8 km from the debris avalanche source area. The bed was not found in cores collected from peat bogs in Elaho, upper Squamish, and Ashlu valleys (Fig. 1). Stratigraphic position Brooks and Hickin (1991) described in detail debris avalanche and backwater sediments (their "fan toe" deposit) exposed at the upstream end of the Turbid Creek debris fan. A representative lithostratigraphic section (Fig. 3) is briefly described here. The basal unit (unit 1) consists of diamicton at least 5 m thick that is part of a debris avdanche deposit that impounded Squarnish River. This is overlain by up to 4.5 m of interbedded fluvial and debris flow deposits (unit 2) reworked from the debris avalanche deposits immediately following the landslide. Above this is 20 cm of sand and silt lacustrine rhythmites (unit 3a); in some exposures, this unit overlies directly the debris avalanche diamicton (unit 1). Unit 3a is overlain successively by 6 m of silt and clay lacustrine rhythmites (unit 3b) and 2.2 m of sand and silt lacustrine rhythmites (unit 3 4 , both of which accumulated within the lake impounded by the debris avalanche. Capping the sequence is 2.6 m of fluvial sediments (unit 4) deposited by Squamish River after the lake drained or was infilled. The contact between the lacustrine and overlying fluvial sediments is gradational, and nowhere in the section is there evidence of a major hiatus. Following the deposition of unit 4, Squamish River incised Can. J. Earth Sci. Vol. 32, 1995 Fig. 3. Stratigraphic section of deposits at the upstream end of the Turbid Creek debris fan, showing the position of the white bed. The location of the section is shown in Fig. 2. 0 m Legend UNlT 4 - Fluvial Lithofacies coding Dmm Gms Sh Sr SP St FI Fm UNlT 3c - Lacustrine - 5m DIAMICTON, massive,matrix-supported GRAVEL, stratified SAND, horizontally stratified SAND, ripple cross-laminations SAND, planar cross-laminations SAND, trough cross-laminations SILTICLAY, laminations SILTICLAY, massive UNlT 3b - Lacustrine Sediments 10 m 'White' bed . UNlT 3a - Lacustrine 1 Dmm Sh StISh StISh 15m UNlT 1 - Debris Avalanche . ri Dmm ,o,-.: the debris fan, effectively terminating aggradation. The entire sequence is attributed to the ca. 4800 BP debris avalanche from Mount Cayley, based on (i) four radiocarbon ages ranging from 4860 to 5310 BP on wood recovered from debris avalanche diamicton of unit 1, and (ii) the lack of major breaks within the sequence (Brooks and Hickin 1991; Evans and Brooks 1991, 1992). Although Squamish River was impounded by subsequent landslides from Mount Cayley, none of the reservoirs was high enough to inundate the 4800 BP backwater deposits at the upstream end of the debris fan. The white bed is 1-2 cm thick and lies at the contact of units 3a and 3b (Figs. 3, 4). It can be traced over a distance of about 1100 m in exposures at the upstream end of the Turbid Creek debris fan. The white bed contrasts markedly with the darker coloured sediments immediately above and below it (Fig. 4). Grain characteristics The white bed consists of silt and clay, and has a mean grain AP. . , .. ' SAND ~ 13 ~ DIAMICTON - Grain Size SILT CLAY size of 15.6 pm and standard deviation of 13.6 pm. The grain morphology includes angular shards and elongated and furrowed particles; some of the latter are slightly curved, with a few containing spherical vesicles. The furrow and vesicle characteristics are morphologic features common to tephra shards (Heiken 1974). About 5% of the grains appear isotropic under cross-polarized light, a characteristic of volcanic glass. X-ray diffraction analysis of a bulk sample of the white bed indicates that the crystalline grains are mainly cristobalite (polymorph of Si02 found in siliceous volcanic rocks) and Na-feldspar; quartz and trace amounts of amphibolite, mica group, and chlorite are also present. We experienced technical difficulties trying to analyze the geochemistry of the isotropic grains using an electron microprobe because of the fine texture of the material. Specifically, problems concerned the grains "floating" and therefore overlapping within the epoxy matrix of the polish thin section, which complicated both the identification and analysis of the grains, and the generally small size of the isotropic grains in relation to the microprobe beam width. Unreliable I I I1 Brooks and Hickin Fig. 4. Photographs of the white bed. (A) Oblique view of an exposure showing the laterally continuous white bed (arrows) at the base of the lacustrine unit 3b. (B) Close-up view of the white bed, which is situated at the contact between lacustrine units 3a and 3b. Table 1. Chemical composition of grains in the white bed having tephra-like morphologic features and glass composition of Mazarna and Bridge River tephras. SiO, (wt.%) A1203 CaO Na20 K2O MgO FeO White bed Mazama tephra Bridge River tephra 51.67f 5.34 (29.09-69.51) 18.58f3.98 (9.28-36.9) 3.54k1.90 (0.65-11.73) 1.17k0.85 (0.0-5.13) 0.95+0.36 (0.14-2.52) 3.56f 2.56 (0.0-15.83) 5.15k3.13 (0.11-18.97) 72.58f 0.31 74.35+0.73 14.40f0.16 13.72k0.27 1.66+0.10 1.40&0.18 5.18k0.20 4.61 k0.15 2.70+0.11 3.16k0.14 0.56k0.08 0.04+0.01 2.11 f0.08 1.62k0.18 Notes: Values are means & 1 SD, with the maximums and minimums given in parentheses for the white bed. The analysis of the white bed was performed by Mineral Resources Division, Geological Survey of Canada on a Cameca SX50 electron microbe, which has four wavelength-dispersive spectrometers. The microprobe was operated at 15 kV with a beam current of 10 nA and a beam width of approx. 3 pm. Metals and simple compounds were used as standards. The raw counts were converted to elemental concentrations using the Cameca PAP program. Data for the Mazama and Bridge River tephras are from Westgate and Gorton (1981). results were obtained by this method. Analysis of grains contained upon a polished disk was done using tephra-like morphologic features as a guide to grain identification. This produced technically reliable data; however, 26 of the 46 grains analyzed were crystalline minerals (ferromagnesium group, quartz, feldspar, and lithic fragments). While it is not equivocal that the other 20 grains represent volcanic glass, they collectively have a range of compositions that vary considerably from the volcanic glass of known tephras found in southwestern British Columbia (Table 1). Interpretations Eruptive origin The presence of isotropic grains within the white bed suggests that it could be tephra produced by a volcanic eruption. The occurrence of the white bed within deposits relating to the ca. 4800 BP impoundment of Squamish River has been suggested to indicate that an eruption of Mount Cayley accompanied, and even caused, the massive debris avalanche that blocked the river (Evans 1990). Although a subsurface heat source exists within the volcano, as indicated by the presence of hot springs in the Turbid Creek basin (Souther 1980), we have found no evidence of recent eruptive products anywhere in the area. More significantly, the white bed seems too fine to be the product of a local eruption. No Holocene tephra deposits are known to occur in the Turbid Creek debris fan area, although two are present regionally in southwestern British Columbia. Mazama tephra was produced by an eruption at Crater Lake (Mount Mazama), Oregon, ca. 6800 BP (Bacon 19831, and occurs throughout the Pacific Northwest. Mazama tephra has been identified in Cheakamus Valley south-southeastof Mount Cayley (Mathews 1972), and in Ashlu Valley to the south-southwest (Brooks 1992), but it is not clear if the tephra covered the area now buried by the Turbid Creek debris fan (Fig. 1). Bridge River tephra is a product of the ca. 2350 BP eruption at Plinth Peak, Mount Meager volcanic complex about 70 km northwest of Mount Cayley (Fig. 1) (Nasmith et al. 1967; Mathewes and Westgate 1980). It probably does not occur as far south as Mount Cayley (Fig. 1). The white bed reported here is younger than Mazama tephra and older than Bridge River tephra. Although it conceivably could consist of grains reworked from a deposit of Mazama tephra upstream of the impoundment, there are no known occurrences of Mazama tephra in Elaho and upper Squamish valleys (Fig. 1). Even if Mazama tephra is present in these areas, it likely is of limited extent and thus could not produce such a concentrated reworked layer at the toe of the Turbid Creek debris fan, especially since the tephra would have to have been eroded and redeposited some 2000 radiocarbon years after the eruption. In addition, the composition of the white bed with its high proportion of crystalline grains differs significantly from Mazama and Bridge River tephras. Although the white bed is not associated with either the Mount Mazama or Mount Meager eruptions, it is conceiv- Can. J . Earth Sci. Vol. 32, 1995 ably a product of a distant but unknown volcanic eruption. Noneruptive origin The fact that the white bed was deposited shortly after the ca. 4800 BP debris avalanche suggests that the two are related. The source rocks of the debris avalanche are the product of the Mount Cayley and Vulcan's Thumb eruptive stages (Evans and Brooks 1991) and consist of lava flows, dykes, breccia, and pyroclastic deposits (Souther 1980; Clague and Souther 1982). The isotropic grains within the white bed thus could have been inherited from the source rocks. There are two reasonable noneruptive explanations for the formation of the white bed. The bed may consist of silt and clay grains that were transported from the debris avalanche diamicton by subsurface runoff, and redeposited as a discrete bed. However, the fact that the white bed is a single, welldefined layer (Fig. 4B) rather than an irregular, diffuse zone does not support this hypothesis. Alternatively, the white bed could be dust generated by the debris avalanche. Large dust clouds commonly rise from and envelop rapidly moving streams of fragmented rock and may be carried considerable distances by powerful air blasts (Eisbacher and Clague 1984). Brief reference to dust clouds is contained in the descriptions of many landslides (e.g . , Plafker and Ericksen 1978; Hewitt 1988). Some of this literature also mentions dust deposits (e.g., McConnell and Brock 1904; Mathews and McTaggart 1978), but these generally are not described in detail. However, there are accounts of significant accumulations of landslide dust. Heim (1882) mentions a 2-3 cm thick layer of black slate dust that was deposited upon the rubble of the 1881 landslide at Elm, Switzerland, and Crandell and Fahnestock (1965) refer to "nearly an inch" of silt and fine sand that settled on trees following a series of large 1963 rockfalls from the eastern flank of Mount Rainier volcano. In both cases, the sediment was later washed away by rain. The ca. 4800 BP debris avalanche transported volcanic material from the slopes of Mount Cayley to the bottom of Squamish Valley, pulverizing large amounts of rock into sand and silt, which now constitute the matrix of the debris avalanche deposit (Evans and Brooks 1991; Brooks 1992). The mechanical disintegration of rock during movement of the debris avalanche undoubtedly generated a large dust cloud that would have settled on the valley floor. It is unlikely, however, that the white bed is dust that settled out directly from this cloud because the bed occurs within the backwater deposits rather than lying directly on the debris avalanche diamicton (Fig. 3). It is likely that the white bed consists of sediment that was washed into the reservoir shortly after it began to fill, and thus is a secondary dust deposit. As mentioned above, the dust deposits described by both Heim (1882) and Crandell and Fahnestock (1965) eventually were washed away by rainfall. The dust hypothesis seems the most likely explanation for the origin of the white bed, but this is not unequivocal because of uncertainty concerning the interpretation of the geochemical evidence. Although the high proportion of crystalline grains within the white bed probably supports a noneruptive origin, this could result from the incorporation of lithic fragments into a tephra cloud, or perhaps from the mixing of tephra and other silt and clay grains carried by Squamish River into the lake impounded by the debris avalanche. Similarly, if the 20 grains with tephra-like morphologic features are volcanic glass (Table I), their collective heterogeneous composition does not clearly support a noneruptive hypothesis because lithic volcanic glass fragments also could have been incorporated into a tephra cloud. Conclusions A distinct, 1-2 cm thick, tephra-like bed containing a high proportion of crystalline grains occurs within backwater deposits that accumulated in response to the impoundment of Squamish River by a large debris avalanche from Mount Cayley ca. 4800 BP. This bed is interpreted to be landslidegenerated dust that settled in the valley and was later washed into the lake impounded behind the debris avalanche deposit. It does not constitute evidence of a volcanic eruption associated with the ca. 4800 BP debris avalanche from Mount Cayley. We recommend caution when a new tephra bed is encountered close to a dormant or extinct volcano that has experienced large landslides; the bed is not necessarily the product of a volcanic eruption. Acknowledgments The possible landslide-dust origin of the white bed was first suggested to us by Peter Jordan of the British Columbia Ministry of Forests. Drafts of this paper were read by Steve Evans, John Westgate, Thomas Pierson, Jon Riedel, John Clague, and an anonymous reviewer; their comments greatly improved the paper and are appreciated. The research was supported by the Natural Sciences and Engineering Research Council of Canada and the Geological Survey of Canada (GSC). We thank Mark Gawehns (Simon Fraser University) for assistance in the field; David Walker (GSC) for scanning electron microscope images; Michael Beaulne (Canadian Centre for Mineral and Energy Technology) for microprobe sample preparation; John Stirling (GSC) for the microprobe analysis; Gordon Pringle (GSC) for discussion of the microprobe analysis; and Bob Delabio (GSC) for the X-ray diffraction analysis. References Bacon, C.R. 1983. Eruptive history of Mount Mazama and Crater Lake caldera, Cascade Range, U.S.A. Journal of Volcanology and Gwthermal Research, 18: 57- 115. Brooks, G.R. 1992. Aspects of post-glacial sediment supply and its control upon the morphology of Squamish River, southwestern British Columbia. Ph.D. thesis, Simon Fraser University, Burnaby , B.C. Brooks, G.R., and Hickin, E.J. 1991. Debris avalanche impoundments of Squamish River, Mount Cayley area, southwestern British Columbia. Canadian Journal of Earth Sciences, 28: 1375- 1385. Clague, J.J., and Souther, J.G. 1982. The Dusty Creek landslide on Mount Cayley, British Columbia. Canadian Journal of Earth Sciences, 19: 524-539. Crandell, D.R., and Fahnestock, R.K. 1965. Rockfalls and avalanches from Little Tahoma Peak on Mount Rainier Washington. United States Geological Survey, Bulletin 1221A, pp. A1-A30. Brooks and Hickin I I I Crandell, D.R., Mullineaux, D.R., and Rubin, M. 1975. Mount St. Helens: recent and future behavior. Science (Washington, D.C.), 187: 438-441. Eisbacher, G.H., and Clague, J.J. 1984. Destructive mass movements in high mountains: hazard and management. Geological Survey of Canada, Paper 84-14. Evans, S.G. 1990. Massive debris avalanches from volcanoes in the Garibaldi Volcanic Belt, British Columbia. Geological Association of Canada - Mineralogical Association of Canada, Annual Meeting, Program with Abstracts, 15: A38. Evans, S.G., and Brooks, G.R. 1991. Prehistoric debris avalanches from Mount Cayley volcano, British Columbia. Canadian Journal of Earth Sciences, 28: 1365- 1374. Evans, S.G., and Brooks, G.R. 1992. Prehistoric debris avalanches from Mount Cayley volcano, British Columbia: Reply. Canadian Journal of Earth Sciences, 29: 1343- 1347. Fisher, R.V., and Schmincke, H.-U. 1984. Pyroclastic rocks. Springer-Verlag, New York. Green, N.L., Armstrong, R.L., Harakal, J.E., Souther, J.C., and Read, P.B. 1988. Eruptive history and K- Ar geochronology of the Late Cenozoic Garibaldi Volcanic Belt, southern British Columbia. Geological Society of America Bulletin, 100: 563 -579. Heiken, G. 1974. An atlas of volcanic ash. Smithsonian Contributions to the Earth Sciences, No. 12. Heim, A. 1882. Der bergstruz von Elm. Zeitschrift der Deutschen geologischen Gesellschatt, 34: 74- 115. Hewitt, K. 1988. Catastrophic landslide deposits in the Karakoram Himalaya. Science (Washington, D.C.), 242: 64 -67. Knox, R.W.O'B. 1993. Tephra layers as precise chronostratigraphical marks. In High resolution stratigraphy. Edited by E.A. Hailwood and R.B. Kidd. Geological Society of America, Special Publication 70, pp. 169- 186. Mathewes, R.W., and Westgate, J.A. 1980. Bridge River tephra: revised distribution and significance for detecting old carbon errors in radiocarbon dates of limnic sediments in southern British Columbia. Canadian Journal of Earth Sciences, 17: 1454- 1461. Mathews, W.H. 1972. Geology of Vancouver area of British Columbia- field excursion A05 -C05. International Geological Congress 24th Session. Mathews, W .H., and McTaggart, K.C. 1978. Hope rockslides, British Columbia, Canada. In Rockslides and avalanches. Edited by B. Voight. Elsevier Scientific Publishing Company, New York, pp. 259-275. McConnell, R.G., and Brock, R.W. 1904. The great landslide at Frank, Alberta. Department of the Interior, 1902- 1903 Annual Report, Part 8. Nasmith, H., Mathews, W.H., and Rouse, R.E. 1967. Bridge River ash and some other Recent ash beds in British Columbia. Canadian Journal of Earth Sciences, 4: 163- 170. Plaflcer, G., and Ericksen, G.E. 1978. Nevados Huascaran avalanches, Peru. In Rockslides and avalanches. Edited by B. Voight. Elsevier Scientific Publishing Company, New York, pp. 277-314. Roddick, J.A., Muller, J.E., and Okulitch, A.V. 1979. Fraser River, British Columbia - Washington. Geological Survey of Canada, Map 1386A, scale 1 : 1000000. Sarna-Wojcicki, A.M., and Davis, J.O. 1991. Quaternary tephrochronology. In Quaternary non-glacial geology: conterminous U.S. Edited by R.B. Morrison, Geological Society of America, The Geology of America, Vol. K-2, pp. 93-116. Souther, J.G. 1980. Geothermal reconnaissance in the central Garibaldi Belt, British Columbia. In Current research, part A. Geological Survey of Canada, Paper 80-lA, pp. 1 - 11. Westgate, J.A., and Gorton, M.P. 1981. Correlation techniques in tephra studies. In Tephra studies. Edited by S. Self and S.J. Sparks. D. Reidel, Dordrecht, The Netherlands, pp. 73 -94. Westgate, J.A., and Naeser, N.D. 1985. Dating methods of Pleistocene deposits and their problems: V. Tephrochronology and fission-track dating. In Dating methods of Pleistocene deposits and their problems. Geoscience Canada, Reprint Series 2, pp. 31-38.