499 Paleomagnetism and counterclockwise tectonic rotation of the Upper Oligocene Formation, southern ,

Donald R. Prothero, Elizabeth Draus, Thomas C. Cockburn, and Elizabeth A. Nesbitt

Abstract: The age of the Sooke Formation on the southern coast of Vancouver Island, British Columbia, Canada, has long been controversial. Prior paleomagnetic studies have produced a puzzling counterclockwise tectonic rotation on the underlying Eocene volcanic basement rocks, and no conclusive results on the Sooke Formation itself. We took 21 samples in four sites in the fossiliferous portion of the Sooke Formation west of Sooke Bay from the mouth of Muir Creek to the mouth of Sandcut Creek. After both alternating field (AF) and thermal demagnetization, the Sooke Formation produces a single-component , held largely in , which is entirely reversed and rotated counterclockwise by 358 ± 128. This is consistent with earlier results and shows that the rotation is real and not due to tectonic tilting, since the Sooke Formation in this region has almost no dip. This rotational signature is also consistent with counterclockwise rota- tions obtained from the northeast tip of the Olympic Peninsula in the Port Townsend volcanics and the Eocene–Oligocene sediments of the Quimper Peninsula. The reversed magnetozone of the Sooke sections sampled is best correlated with Chron C6Cr (24.1–24.8 Ma) or latest Oligocene in age, based on the most recent work on the Liracassis apta Zone mol- luscan fauna, and also a number of unique marine mammals found in the same reversed magnetozone in Washington and Oregon. Re´sume´ : L’aˆge de la Formation de Sooke sur la coˆte sud de l’ıˆle de Vancouver, Colombie-Britannique, Canada, a long- temps e´te´ controverse´. Des e´tudes pale´omagne´tiques ante´rieures ont donne´ une rotation perplexe dans le sens inverse des aiguilles d’une montre sur les roches volcaniques du socle (E´ oce`ne) et aucun re´sultat concluant sur la Formation de Sooke elle-meˆme. Nous avons pre´leve´ 21 e´chantillons de quatre sites dans la portion fossilife`re de la Formation de Sooke a` l’ouest de la baie de Sooke, de l’embouchure du ruisseau Muir a` l’embouchure du ruisseau Sandcut. Apre`s une de´saiman- tation par champs alternatifs (CA) et par traitements thermiques, la Formation de Sooke a produit une aimantation re´ma- nente, surtout dans la magne´tite, qui est comple`tement renverse´e et tourne´e dans le sens contraire des aiguilles d’une montre de 35 8 ±128. Cela concorde avec des re´sultats ante´rieurs et montre que la rotation est re´elle et non pas cause´e par un basculement tectonique, e´tant donne´ que la Formation de Sooke dans cette re´gion n’a pratiquement pas de pendage. Cette signature de rotation concorde aussi avec les rotations dans le sens inverse des aiguilles d’une montre obtenues de l’extre´mite´ nord-est de la pe´ninsule Olympic dans les roches volcaniques de Port Townsend et les se´diments, datant de l’E´ oce`ne–Oligoce`ne, de la pe´ninsule de Quimper. En se basant sur les plus re´cents travaux sur la zone de faune mollusque a` Liracassis apta et d’un certain nombre de mammife`res marins uniques trouve´s dans la meˆme magne´tozone de polarite´ inverse dans les e´tats de Washington et d’Oregon, la magne´tozone de polarite´ inverse des sections e´chantillonne´es de Sooke est le mieux corre´le´ avec la C6Cr (24,1–24,8 Ma), soit un aˆge Oligoce`ne terminal. [Traduit par la Re´daction]

Introduction Vancouver Island have also been called Sooke Formation (Clapp and Cooke 1917; Jeletzky 1973; Cameron 1980). The Sooke Formation is widely exposed along the First described by Richardson (1876–1877), the rich fossil beaches and creeks of southern Vancouver Island west of deposits of the Sooke Formation were originally collected Sooke Bay (Fig. 1) and includes beach and seacliff outcrops in the 1890s, and described by Merriam (1896, 1897, 1899). from southeastern side of Sooke Bay to San Juan Point Clapp and Cooke (1917) made the first detailed maps of the (southwest of ). Outcrops at Carmanah Point region, and Clark and Arnold (1923) conducted the first and on Nootka Island, further north on the western side of complete description of the fossils. On Nootka Island,

Received 6 September 2007. Accepted 4 March 2008. Published on the NRC Research Press Web site at cjes.nrc.ca on 29 May 2008. Paper handled by Associate Editor F. Cook. D.R. Prothero1 and E. Draus. Department of , Occidental College, Los Angeles, CA 90041, USA. T.C. Cockburn. 7683 Colin Place, , BC V8M 1N6, Canada. E.A. Nesbitt. Burke Museum, Box 353010, University of Washington, Seattle, WA 98195, USA. 1Corresponding author (e-mail: [email protected]).

Can. J. Earth Sci. 45: 499–507 (2008) doi:10.1139/E08-012 # 2008 NRC Canada 500 Can. J. Earth Sci. Vol. 45, 2008

Fig. 1. Geology and sample sites for the Sooke Formation on the southern tip of Vancouver Island. Geology based on Geological Survey of Canada Map No.1553A by J. E. Muller. Q, ; S, Sooke Fm; M1, Sooke Gabbro; M2, Volcanics; M3, Catface intru- sions. Sample sites this report, 1, 2, 3, 4; Sooke Fm sample sites of Irving and Massey (1990), S1 and S2. Inset map: Location of main map area. LRF, Leech River ; SJF, San Juan Fault.

the Sooke outcrops overlie the Hesquiat Formation, and (1923) correlated the Sooke invertebrate faunas with those Cameron (1980) used foraminiferal biostratigraphy to place of the Blakeley Formation, on Bainbridge Island, across these rocks in the late Zemorrian (latest Oligocene). Puget Sound from Seattle, Washington. However, the corre- Muller et al. (1981) considered the foraminiferal assem- lation used by Clark and Arnold (1923) was based on fossils blage from the Sooke Formation to be facies restricted and from the Carmanah Point area of Vancouver Island that is not stratigraphically useful. Narayan et al. (2005) devel- outside of the Sooke formation between Sooke Bay and San oped a benthic foraminiferal biozonation from wells drilled Juan Point. Durham (1944) noted that the Sooke molluscan in the Basin, offshore southern Vancouver Island. fauna was characteristic of outcrops the Blakeley Formation Although there is no outer neritic to bathyal sediment around Lake Sammamish (east of Seattle). This region is the onshore, this offshore scheme indicate that the Sooke For- inner neritic to bayshore facies of the deeper water type mation belongs in the Rectuvigerina branneri Zone (formally Blakeley Formation on Bainbridge Island and has recently called the Siphogenerina branneri Zone) of Kleinpell et al. been mapped as this formation (Booth et al. in review). In (1980) and Finger (1992), latest Oligocene – earliest Mio- the historic Weaver Commission report (Weaver et al. cene in age. 1944), the Sooke was correlated with the upper Twin Rivers The molluscan fauna of the Sooke Formation is particu- Formation, which was then considered early Miocene. larly diverse, with 28 species of gastropods and 35 species Jeletzky (1973), Muller et al. (1981), and Cockburn et al. of bivalves. In addition, two species of brachipods and bar- (1999) assigned the Sooke Formation in the Nootka Sound nacles; single chiton, sand dollar, and coral species; and a area and between Sooke Bay and San Juan Point to the serpulid worm tube have been described (Clark and Arnold molluscan Liracassis (= Echinophoria) apta Zone of Dur- 1923; Cornwall 1927). Despite its rich faunal assemblage, ham (1944; Moore 1984), which spans the late Oligocene – the age of the Sooke Formation has long been controversial. early Miocene interval between 23 and 28.5 Ma (Prothero Merriam (1896) thought that it was upper Miocene or lower 2001). Pliocene in age. Arnold (1906, 1909) placed it in the upper This wide range of opinions of the age of the Sooke mol- Miocene. Weaver (1912) correlated it with other lower Mio- lusks is largely due to the fact that the assemblage comes cene deposits of Washington State. Arnold and Hannibal from intertidal and shallow-subtidal deposits, with relatively (1913) collected additional fossils and placed it in the few of the mid-neritic mollusks that are used to zone the middle Oligocene. Clapp and Cooke (1917), on the other marine beds of the Pacific Northwest elsewhere. This is par- hand, placed it in the lower Miocene. Clark and Arnold ticularly evident in the abundant rocky intertidal forms, such

# 2008 NRC Canada Prothero et al. 501

Table 1. Taxa from the Sooke Formation indicative of intertidal to estuarine habitats (from Clark and Arnold 1923; Cockburn et al. 1999; Burke Museum unpublished data).

Intertidal rocky shore fauna Chiton Oligochiton lioplax Limpets Five species from the family Lottidae Key-hole limpets Megathura vancouverensis, Puncturella sp. Turban shell Homalopoma vancouverensis Periwinkles Littorina sookensis Cap-shells Calyptraea vancouverensis, Trochita sookensis Slipper limpets Crepidula sookensis Rock-boring piddocks Zirfaea sp. Mussels Four species of the genus Mytilus Oysters Crassostrea sookensis Barnacles Solidobalanus sookensis Intertidal sandy sediment fauna Surf clams Three species of the family Mactridae Razor clams Solen clallamensis Cockles Clinocardium sookensis, Glycymeris vancouverensis Sand dollar Scutella newcombei Geoduck Panope generosa Soft-shell clam Cryptomya quadrata Brackish water – estuarine fauna Mud snails Cerithiopsis newcombei Goniobasis sookensis Estuarine clam Corbicula sookensis as the chiton, plus a diversity of limpets, algal-grazing only known from the Oligocene–Miocene Clallam Forma- gastropods, rock-boring clams, mussels, oysters and bar- tion of Washington and the upper Oligocene Nye Mudstone nacles (Table 1). In addition the fauna includes a sand dollar near Newport, Oregon (Tedford et al. 1994). An unpublished and a diversity of bivalve taxa that inhabit intertidal sandy analysis of the pollen from the Sooke (Cox 1962) shows a sediments, as well as two gastropod species indicative of close resemblance to the pollen of the Blakeley Formation, brackish estuarine settings (Table 1). This faunal assemblage and also to the pollen of the middle Miocene Astoria Forma- is most like that from the inland outcrops of the Blakeley tion of Oregon. Formation, east of Lake Washington in the Seattle area. Despite these problems, it is apparent from the more re- The type Blakeley at Restoration Point on Bainbridge Island, cent age estimates mentioned earlier in the text that the west of Seattle is within the Liracassis apta molluscan opinion about the age of the Sooke Formation is converging zone, and at the foraminiferal Zemorrian–Saucesian boun- on the late Oligocene or earliest Miocene based on mol- dary zone, which overlaps in the Pacific Northwest (Pro- lusks, desmostylians, and the ‘‘beach bear.’’ Where exactly thero 2001; Prothero and Nesbitt, in press). The fauna in the late Oligocene or early Miocene could be better from the type section of the Blakeley Formation is deeper resolved by and by comparison with water in origin, and the fauna from the inland sections is recent magnetostratigraphic work on beds of similar age, characteristic of an intertidal embayment. However, the such as the Clallam Formation (Prothero and Burns 2001) gastropod species Liracassis apta, Bruclarkia acuminata, and the Blakeley Formation (Prothero and Nesbitt in press) Mediargo matthewsonii, and Priscofusus hannibali have of Washington, and the Nye Mudstone near Astoria, Oregon been found in both areas (Clark and Arnold 1923; Cock- (Prothero et al. 2001a). burn et al. 1999; Burke Museum unpublished data). Thus, The only previous paleomagnetic study of the Sooke For- the molluscan assemblage from the Sooke Formation is mation was by Irving and Massey (1990). Their focus was stratigraphically consistent with the foraminiferal biozona- primarily on the early Eocene Metchosin ophiolites, which tion indicating a latest Oligocene to earliest Miocene age. unconformably underlie the Sooke Formation near Sooke A few distinctive mammalian fossils have been collected Bay. Their magnetic results from the Metchosin volcanics from the Sooke Formation near Sooke Bay. The most im- (especially the Stock) showed a consistent portant are teeth of the desmostylian Cornwallius sookensis counterclockwise rotation of about 308–408 in rocks of both (Cornwall 1922; Hay 1923; Clark and Arnold 1923). This normal and reversed polarity (so they passed a reversal test). desmostylian has also found in the upper Nye Formation This is consistent with the 208 ±78 rotation reported for the near Astoria, Oregon that is upper Oligocene – lower Mio- same rocks by Symons (1973). Irving and Massey (1990) cene (Prothero et al. 2001a). There is also a lower jaw did sample isolated unfossiliferous outcrops of reddish fragment of a new undescribed genus of ‘‘beach bear’’ Sooke near Sooke Bay, but their results were (McAnally 1996) related to Kolponomos, which is otherwise highly scattered, and they did not believe it represented an

# 2008 NRC Canada 502 Can. J. Earth Sci. Vol. 45, 2008

Fig. 2. Orthogonal demagnetization (‘‘Zijderveld’’) plots and stereonets of representative samples. On Zijderveld plots, solid squares indicate declination (horizontal component); open squares indicate inclination (vertical component). On stereonets, open squares indicated directions in the upper hemisphere. First step is NRM, followed by AF steps of 2.5, 5.0, and 1.0 mT (millitesla), then thermal steps from 200 to 630 8C in 100 8C increments. Each division equals 10–5 emu.

Oligocene field. Their samples were too friable to be treated Table 2. Paleomagnetic data. by thermal demagnetization, and they only used alternating Site nD I k field (AF) demagnetization on rocks with a distinct reddish 95 tint. Such a procedure would not have removed any young 1 4 149.4 –58.7 3.1 63.2 chemical remanent overprinting due to hydroxides 2 5 132.1 –48.8 28.0 14.7 (which must be thermally demagnetized), so this inconclu- 3 6 146.3 –53.1 9.3 23.2 sive result is not surprising. Irving and Massey (1990) also 4 6 151.2 –44.1 6.4 32.8 sampled sixteen boulders in a Sooke conglomerate and got Note: n, number of samples; D, declination (8); I, inclination a random scatter of points, showing that the Sooke Forma- (8); k, precision parameter; 95, ellipse of 95% confidence around tion has not been remagnetized since deposition (a positive mean. conglomerate test). Irving and Massey (1990) did not report any results from the classic fossiliferous outcrops of the vectors decayed to the origin. The fact that nearly every Sooke Formation to the west of Sooke Bay. sample showed a significant drop in intensity during AF demagnetization indicates that most of the remanence is held by a low-coercivity mineral, such as magnetite; this is Methods consistent with the fact that nearly all the remanence was During the summer of 2002, we took oriented block sam- lost when the Curie temperature of magnetite (580 8C) was ples from beach and stream exposures of the Sooke Forma- exceeded. Thus, it appears that the Sooke samples all show tion west of Sooke Bay. The local section was only 10 m a primary or characteristic remanence that is reversed and thick in this region, so four separate sites with as many as rotated counterclockwise. Although it is not possible to con- six samples apiece were spaced along the beach from the duct a true reversal test on these rocks without an antipodal classic fossil locality at the mouth of Muir Creek west to normal direction, the fact that the direction is clearly French Beach Park and ending at the mouth of Sandcut reversed with minimal overprinting can only be explained if Creek (Fig. 1). Most of the rocks were well indurated and it is a primary or characteristic remanence. In addition, these did not crumble like those reported by Irving and Massey results are consistent with those of Irving and Massey (1990), but dilute sodium silicate was used to harden sam- (1990) on the Metchosin volcanics, which did pass a rever- ples that required it. In the laboratory, each block was then sal test. Those same authors showed with a conglomerate subsampled into standard cores using a drill press. They test that there has been no post-depositional remagnetization were then measured on a 2G cryogenic using of the sediments. an automatic sample changer at the California Institute of The mean for all the Sooke Formation samples (n = 21) Technology, Pasadena, California. After measurement of was D (declination) = 144.18, I (inclination) = –50.98, k NRM (natural remanent magnetization), each sample was (precision parameter) = 7.9, 95 (ellipse of 95% confidence AF demagnetized at 2.5, 5.0, and 10 mT (millitesla) to around mean) = 12.4. This result is rotated about 358 ±128 remove any remanence held by multidomain grains and also counterclockwise with respect to the Oligocene North Amer- to determine the coercivity behavior of each specimen. After ican cratonic pole of Diehl et al. (1983) (Fig. 3). As can be AF demagnetization, every sample was then thermally seen from the stereonet, our Sooke Formation results are demagnetized at 100 8C increments from 200 to 630 8C. highly consistent with the rotations reported on the under- This helps get rid of high-coercivity chemical overprints lying Metchosin volcanics by Irving and Massey (1990), due to iron hydroxides, such as goethite, and allows us to and also with those reported by Symons (1973). Thus, it determine how much remanence was left after the Curie appears that the counterclockwise rotation is real and not temperature of magnetite (580 8C) was exceeded. caused by local tilting, as suggested by Irving and Massey Results were plotted on orthogonal demagnetization (1990), since the Sooke Formation exposures have almost (‘‘Zijderveld’’) plots, and average directions of each sample no dip. were determined by the least-squares method of Kirschvink (1980). Mean directions for each sample were then analyzed Discussion using Fisher (1953) statistics, and classified according to the scheme of Opdyke et al. (1977). Although a counterclockwise tectonic rotation seems sur- prising at first, it is consistent with the results reported by Results Symons (1973) and Irving and Massey (1990), so it is not an artifact of a particular study or a single laboratory. In ad- Results are given in Table 2. Representative orthogonal dition, this counterclockwise rotational trend is opposite the demagnetization plots and stereonets of the same samples clockwise rotation reported throughout the coastal region of are shown in Fig. 2. As can be seen from all the samples, Washington and Oregon (e.g., Wells and Heller 1988; Wells the NRM direction of the samples was southwest and up (a 1990). To our knowledge, the only counterclockwise paleo- reversed direction rotated counterclockwise), and most of magnetic directions previously reported were those of Beck the samples showed only this single component as their and Engebretson (1982), who found a mean reversed D of

# 2008 NRC Canada Prothero et al. 503

# 2008 NRC Canada 504 Can. J. Earth Sci. Vol. 45, 2008

Fig. 3. Stereonet of means of magnetic sites. Solid (oval) dots and pics, it seems clear that the central core of the Olympics solid circle indicate mean for normal sites (lower hemisphere pro- has been pushed eastward, causing the north flank to rotate jection). Open dots and dashed line indicate mean of reversed sam- counterclockwise and the south flank to rotate clockwise. A ples (upper hemisphere projection). Results of reversed Sooke similar model has been suggested by Johnston and Acton Formation samples, and their directions inverted by 1808 are shown (2003), although they only suggest about 208 of clockwise in comparison with the results from the sheeted dikes and East rotation in the southern Vancouver Island region. Contrary Sooke Stock in the Metchosin volcanics (Irving and Massey, 1990). to Johnston and Acton (2003), our results show that the This shows that the directions are antipodal and the primary rema- counterclockwise rotations on the north Olympics are not nence has been obtained and overprinting removed. Fm., Formation. Eocene in age, but early Miocene or younger. McCaffrey et al. (2007) used GPS (global positioning sys- tem) studies of the entire Pacific Northwest coastal region to analyze the block rotations and crustal deformations of the area. Consistent with previous paleomagnetic studies, they found clockwise vertical-axis rotations in the coastal regions of Oregon and Washington south of the Olympics. They found counterclockwise block rotations in northern Van- couver Island, but not on the southern tip of the island, as we have found. If their interpretations are correct, the Neo- gene rotations that we have documented for southern Van- couver Island (and also for the northern coast of the Olympic Peninsula) have ceased to operate today, except in the northern part of Vancouver Island. We know of no paleomagnetic studies from Cenozoic sedimentary rocks in northern Vancouver Island, but it would be interesting to see if they also show counterclockwise rotation as predicted by these models. Although the single reversed magnetozone of the Sooke Formation may not seem diagnostic at first, biosratigraphic age constraints (Fig. 4) allow us to pin down the age of the Sooke Formation. Previous results from the molluscan fauna (discussed earlier in the text) most closely match the Sooke Formation with the Blakeley Formation Liracassis apta molluscan zone assemblages of the uppermost Pysht Forma- tion of the northern Olympic Peninsula, or the Blakeley Formation of the eastern Olympics, and older than the Clallam Formation, which overlies the Pysht Formation 158.58 and I of –71.58 ( 95 = 16.1) for the lower Eocene (Fig. 4). The uppermost Pysht Formation (Prothero et al. Port Townsend on the northeast corner of the Olym- 2001b) was correlated with reversed magnetic Chron C6Cr pic Peninsula and due south of Sooke Bay across the Straits (24.1–24.8 Ma) or latest Oligocene in age. This is slightly of Juan de Fuca. This counterclockwise reversed direction is older than the Blakeley Formation, which correlates with statistically indistinguishable from our result, and from the the latest Oligocene – early Miocene Chron C6Cn (23.5– results of Irving and Massey (1990). In addition, Prothero et 24.1 Ma) (Prothero and Nesbitt in press), and with the al. (in press) have obtained the same counterclockwise rota- Clallam Formation (source of the ‘‘beach bear’’ Kolponomos tions on the Eocene and Oligocene sedimentary rocks clallamensis), which correlates with the same interval (Pro- (Quimper Sandstone, Marrowstone Shale, ‘‘Lyre’’ Forma- thero and Burns 2001). Corroborating this correlation are tion) of the Quimper Peninsula, overlying and just north of the magnetostratigraphic results for the upper Nye Mudstone the Port Townsend basalts, so the rotation is real and found near Astoria, Oregon, which are correlated with Chron throughout the northeast Olympic Mountains, as well as in C6Cr. These rocks are the only ones besides the Sooke the Sooke Bay area. The eastern side of Olympic Peninsula Formation to produce the desmostylian Cornwallius sooken- and western Puget Sound, however, shows a clockwise rota- sis, and also produce the ‘‘beach bear’’ Kolponomos newpor- tion (Beck and Engebretson 1982, on the Bremerton basalts; tensis (Tedford et al. 1994). Thus, the marine mammals Prothero and Nesbitt in press, on the Blakeley Formation), confirm the latest Oligocene age of the Sooke Formation as do the Pysht and Clallam Formations of the northern and tie it to rocks correlated with the same magnetozone in coast of the Olympic Peninsula (Prothero and Burns 2001; the northwest Olympics, on Puget Sound, and near Astoria, Prothero et al. 2001b). Oregon. A full analysis of the tectonic implications of these results will be presented in a forthcoming paper by Mark Brandon Conclusions and the authors. However, some things seem immediately apparent. Based on the overall trend of clockwise rotations Previous studies have given widely varying age estimates to the south and east of the Olympic Peninsula and mostly of the Sooke Formation, and puzzling paleomagnetic direc- counterclockwise rotations on the north side of the Olym- tions. Our study shows that the Sooke Formation is entirely

# 2008 NRC Canada Prothero et al. 505

Fig. 4. Correlation of the Sooke Formation, based on the dates and age constraints discussed in the text. Magnetic of the Yaquina Bay, Oregon, section, after Prothero et al. (2001a), and of the northwest Olympics section after Prothero et al. (2001b) and Prothero and Burns (2001). Magnetic stratigraphy of the Blakeley section after Prothero and Nesbitt (in press). scale after Berggren et al. (1995) and Prothero (2001). Fm., Formation; L. Eoc., Late Eocene; Mioc., Miocene; Sauc., Saucesian; Pillar, Pillarian.

# 2008 NRC Canada 506 Can. J. Earth Sci. Vol. 45, 2008 reversed in polarity and rotated 358 ±128 counterclockwise Vancouver Island, with a description of a new coral by T. Way- with respect to the Oligocene cratonic pole. This is consis- land Vaughn. University of California Publications in Geologi- tent with previous results of Symons (1973) and Irving and cal Sciences, 14: 123–234. Massey (1990), and also with results from the northeast Cockburn, T., Dunlop, S., and Landry, M. 1999. Field trip guide to Olympic Peninsula (Beck and Engebretson 1982; Prothero the Tertiary marine faunas of the Sooke Formation. 3rd British et al. in review). These results seem best explained as a con- Columbia Paleontological Symposium, Victoria, B.C., pp. 1–28. sequence of the eastward movement of the central Olympic Cornwall, I.E. 1922. Some Notes on the Sooke Formation, Vancou- block, which rotated the rocks on its northern flank counter- ver Island, B.C. Canadian Field Naturalist, 36: 121–123. clockwise and those on the southern and eastern flank clock- Cornwall, I.E. 1927. Fossil Cirripedia from the Upper Oligocene wise. SookeFormation of Vancouver Island. University of California Publications in Geological Sciences, 16: 399–408. The single reversed magnetozone of the Sooke Formation Cox, R.L. 1962. Age and correlation of the Sooke Formation with a can be correlated with any of the three small reversal peri- section on its palynology. Unpublished M.Sc. thesis, The Uni- ods that span the Oligocene–Miocene boundary: Chron versity of British Columbia, Vancouver, B.C., 64 p. C6Cr (24.1–24.8 Ma), Chron C6Cn3r, or Chron C6Cn2r Diehl, J.F., Beck, M.E., Jr., Beske-Diehl, S., Jacobson, D., and based on the molluscan Liracassis apta fauna and the pres- Hearn, B.C., Jr. 1983. Paleomagnetism of the Late Cretaceous- ence of the desmostylian Cornwallius sookensis and a early Tertiary north-central Montana alkalic province. Journal ‘‘beach bear’’ similar to Kolponomos newportensis. This of Geophysical Research, 88: 10 593 – 10 609. doi:10.1029/ combination of mollusks and mammals is elsewhere known JB088iB12p10593. only from latest Oligocene rocks correlative with Chron Durham, J.W. 1944. Megafaunal zones of the Oligocene of north- C6Cr (24.1–24.8 Ma), such as the Nye Mudstone near western Washington. University of California Publications in Astoria, Oregon, and the uppermost Pysht Formation and Geological Sciences, 27: 101–211. lowermost Clallam Formation in the northwest Olympics. Finger, K. 1992. Biostratigraphic atlas of Miocene foraminifera from Monterye and Modelo formations, central and southern California. Cushman Foundation for Foraminiferal Research, Acknowledgments Special Publications 28, pp. 1–271. The authors thank Josh Ludtke for help with sampling, Fisher, R.A. 1953. Dispersion on a sphere. Proceedings of the and Teresa LeVelle and Erik Prothero for moral support. Royal Society of London, 217: 295–305. We thank Gavin Hanke for drafting Fig. 1. We thank Hay, O.P. 1923. Characteristics of Sundry Fossil Vertebrates. Pan- R. Enkin, F. Cook, and an anonymous reviewer for helpful American Geologist, 39: 101–120. Irving, E., and Massey, N.W.D. 1990. Palaeomagnetism of ocean comments on the paper. Prothero thanks Joe Kirschvink for layers 2 and 3: evidence from the Metchosin Complex, Vancou- access to the Caltech paleomagnetics laboratory. Prothero ver Island. Physics of the Earth and Planetary Interiors, 64: 247– was supported by NSF grant EAR00-00174 and by a grant 260. doi:10.1016/0031-9201(90)90041-U. from the Donors of the Petroleum Research Fund, adminis- Jeletzky, J.A. 1973. Age and depositional environment of Tertiary tered by the American Chemical Society, during this re- rocks of Nootka Island, British Columbia: mollusks against for- search. aminifera. Canadian Journal of Earth Sciences, 10: 331–365. Johnston, S.T., and Acton, S. 2003. The Eocene Southern Vancou- ver orocline—a response to seamount accretion and the cause of References fold-and-thrust belt and extensional basin formation. Tectono- Arnold, R. 1906. Tertiary and Quaternary pectens of California. physics, 365: 165–183. doi:10.1016/S0040-1951(03)00021-0. United States Geological Survey, Professional Paper 47, p. 10. Kirschvink, J.L. 1980. The least-squares line and plane and analy- Arnold, R. 1909. Environment of Tertiary faunas of the Pacific sis of paleomagnetic data. Geophysical Journal of the Royal Coast. The Journal of Geology, 17: 509–533. Astronomical Society, 62: 699–718. Arnold, R., and Hannibal, H. 1913. The marine Tertiary strati- Kleinpell, R.M., Horneday, G.R., Warren, A.D., and Donnelly, graphy of the north Pacific Coast of America. Proceedings of A.T. 1980. The Miocene stratigraphy of California revisited. the American Philosophical Society, 52: 559–605. American Association of Petroleum Geologists, Tulsa, Okla., Beck, M.E., Jr., and Engebretson, D.C. 1982. Paleomagnetism of 181 p. small exposures in the west Puget Sound area, Washing- McAnally, L.M. 1996. Paleogene mammals on land and at sea. In ton, and speculations on the accretionary origin of the Olympic Life in stone: a natural of British Columbia’s fossils. Mountains. Journal of Geophysical Research, 87: 3755–3760. Edited by R. Ludvigsen. The University of British Columbia doi:10.1029/JB087iB05p03755. Press, Vancouver, B.C., pp. 202–211. Berggren, W.A., Kent, D.V., Swisher, C.C., III, and , Aubry, M.-P. McCaffrey, R., Qamar, A.I., King, R.W., Wells, R., Khazaradze, 1995. A revised Cenozoic and chronostrati- G., Williams, C.A., et al. 2007. Fault locking, block rotation, graphy. SEPM (Society for Sedimentary Geology), Special Pub- and crustal deformation in the Pacific Northwest. Geophysical lication 54, pp. 129–212. Journal International, 169: 1315–1340. doi:10.1111/j.1365- Cameron, B.E.B. 1980. Biostratigraphy and depositional environ- 246X.2007.03371.x. ment of the Escalante and Hesquiat Formations (Early Tertiary) Merriam, J.C. 1896. Note on two Tertiary faunas from the rocks of of the Nookta Sound area, Vancouver Island, British Columbia. the southern coast of Vancouver Island. University of California Geological Survey of Canada, Paper 78-9, 25 p. Geological Bulletin, 2: 101–108. Clapp, C.H., and Cooke, H. 1917. Sooke and Duncan map areas, Merriam, J.C. 1897. New species of Tertiary Mollusca from Vancouver Island. Geological Survey of Canada, Memoir 96, Vancouver Island. The Nautilus, 11: 64–65. pp. 1–445. Merriam, J.C. 1899. The fauna of the Sooke beds of Vancouver Clark, B.L., and Arnold, R. 1923. Fauna of the Sooke Formation, Island. California Academy of Sciences, 1: 175–180.

# 2008 NRC Canada Prothero et al. 507

Moore, E.J. 1984. Middle Tertiary molluscan zones of the Pacific Formation, Clallam County, Washington. Pacific Section, Northwest. Journal of , 58: 718–737. SEPM, Special Publication 91, pp. 224–233. Muller, J.E., Cameron, B.E.B., and Northcote, K.E. 1981. Geology Richardson, J. 1876–1877. Report on the coal fields of , and mineral deposits of the Nootka Sound map area (92E), Comox, Cowitchen, Burrard Inlet, and Sooke, British Columbia. Vancouver Island, British Columbia. Geological Survey of Geological Survey of Canada, Reports of Explorations and Sur- Canada, Paper 80-16, pp. 1–53. veys 160–192. Narayan, Y.R., Barnes, C.R., and Johns, M.J. 2005. Taxonomy and Symons, D.T.A. 1973. Paleomagnetic zones in the Oligocene East biostratigraphy of Cenozoic foraminifers from Shell Canada Sooke Gabbro, Vancouver Island, British Columbia. Journal wells, Tofino Basin offshore Vancouver Island, British Colum- of Geophysical Research, 78: 5100–5109. doi:10.1029/JB078i bia. Micropaleontology, 51: 101–167. 023p05100. Opdyke, N.D., Lindsay, E.H., Johnson, N.M., and Downs, T. 1977. Tedford, R.H., Barnes, L.G., and Ray, C.E. 1994. The early Mio- The paleomagnetism and magnetic polarity stratigraphy of the cene littoral ursoid carnivoran Kolponomos: systematics and mammal-bearing section of Anza-Borrego State Park, California. mode of life. Proceedings of the San Diego Society of Natural Quaternary Research, 7: 316–329. doi:10.1016/0033-5894(77) History, 29: 11–32. 90024-2. Weaver, C.E. 1912. A preliminary report on the Tertiary paleon- Prothero, D.R., 2001. Chronostratigraphic calibration of the Pacific tology of western Washington. Washington Geological Survey Coast Cenozoic: a summary. Pacific Section, SEPM, Special Bulletin, 15: 1–18. Publication 91, pp. 377–394. Weaver, C.E., and 22 others. 1944. Correlation of the marine Ceno- Prothero, D.R., and Burns, C. 2001. Magnetic stratigraphy and tec- zoic formations of western North America. Geological Society tonic rotation of the upper Oligocene – lower Miocene (type of America Bulletin, 55: 569–598. Pillarian stage) Clallam Formation, Clallam County, Wash- Wells, R.E. 1990. Paleomagnetic rotations and Cenozoic tectonics ington. Pacific Section, SEPM, Special Publication 91, pp. 234– of the Cascades arc, Washington, Oregon, and California. Jour- 241. nal of Geophysical Research, 95: 19 409 – 19 417. doi:10.1029/ Prothero, D.R., Bitboul, C.Z., Moore, G.W., and Niem, A.R. JB095iB12p19409. 2001a. Magnetic stratigraphy and tectonic rotation of the Oligo- Wells, R.E., and Heller, P.L. 1988. The relative contribution of cene Alsea, Yaquina, and Nye formations, Lincoln County, accretion, shear, and extension to Cenozoic tectonic rotation in Oregon. Pacific Section, SEPM, Special Publication 91, the Pacific Northwest. Geological Society of America Bulletin, pp. 184–194. 100: 325–338. doi:10.1130/0016-7606(1988)100<0325:TRCO Prothero, D.R., Streig, A., and Burns, C. 2001b. Magnetic strati- AS>2.3.CO;2. graphy and tectonic rotation of the upper Oligocene Pysht

# 2008 NRC Canada