PakoBios 21(2):1-11, September 15, 2001 © 2001 University of California Museum of Paleontology Reassessing the Lambdotherium first appearance datum (Wasatchian, early Eocene) in the Bighorn Basin, Wyoming KRISTER T. SxMITH* University of California, Museum of Paleontology, 1101 Valley Life Sciences Building, Berkeley, CA 94720-4780, USA; •Current address: Department of Geology and Geophysics, Yale University, P.O. Box 208109, New Haven, CT 06520 8109, USA; e-mail: krister.smith^yale.edu The Lambdotherium first appearance datum (FAD) is an important biostratigraphic indicator in the early Eocene of the Rocky Mountain Interior, defining the base of the Lostcabinian provincial land-mammal "subage." Previous studies have placed this datum at different levels in the Elk Creek section of the central Bighorn Basin; as currently documented, the correct level is 591 m. However, a survey of the Yale Peabody Museum collections revealed that four specimens of Lambdotherium derive from localities below the previously documented FAD. One of these is from a locality correlated to S01 m in the Flk Creek composite section, 90 m below the previous FAD; another specimen is from the 511 m level, and two others are from the 561 -m level. Magnctostratigraphic correlation with the McCullough Peaks section to the north suggests that the Lambdotherium FAD may be diachronous by as much as 200 kyr. Alternatively, the FAD may actually be lower than currently documented in the McCullough Peaks area; increased sampling intensity may help to resolve this issue. Even if the Lambdotherium FAD is diachronous, that taxon would remain a strong temporal indicator. However, its usefulness for defining chronostratigraphic and therefore geochronologic units (e.g., Lostcabinian) would be compromised. INTRODUCTION Perissodactvls have been utilized as biostratigraphic indi- cators in the early Eocene of North America, especially since the work of Granger (1914). He first provided a complete biostratigraphic zonation of what is now recognized as the Wasatchian NALMA (North American Land Mammal "Age") by establishing four provincial units, viz., the Sand Coulee, based on the presence of the equid Hyracothcrium Owen 1841; the Gray Bull, based on the presence of both Hyracothcrium and the isectolophid Homojjalax Hay 1899; the Lysite, based on the presence of the helaletid Heptodon ("ope 1882; and the Lost Cabin, based on the presence of both Heptodon and the brontotheriid Lambdotherium Cope 1880. All of these index taxa are perissodactvls. Though originally discussed in the context of the Wind River and Bighorn Basins of Wyoming, Granger's scheme was subse- quently extended, with some modification, to other basins (Van Houten 1944, 1945, Krishtalka et al. 1987). The validity of some of the land mammal "subages" within the Wasatchian has been questioned. For example, Jepsen (1930) argued that the Sandcouleean could not be regarded as distinct from the succeeding Graybullian because Homojialax could be found deep within nominally Sandcouleean strata. Despite a rebuttal by Simpson (1937), Jepsen's arguments were largely embraced by later workers (e.g.. Van I louten 1944, 1945). Bown (1979) in particular argued that the stratigraphic distribution of Homo/jalax depends on that taxon's paleoecology; thus, its first appear- ance may be diachronous even across an area as small as the Bighorn Basin (Fig. 1). In subsequent years the Sandcouleean has been revived by Gingcrich (1983, 1989, 1991). By con- trast, the Lostcabinian —the final "subage" of the Wasatchian—has long been accepted by vertebrate paleon- tologists and stratigraphers. Schankler (1980) published the first formal biostrati- graphic zonation of the Wasatchian that was tied to a strati- graphic section, following the suggestions of the then-current International Stmtijjraphic Guide (Hedberg 1976). His scheme was tied to the Schanklcr-Wing F.lk Creek strati- graphic section in the central Bighorn Basin of Wyoming (Fig. 1), and he included meter-levels from this measured section for all of his zonal boundaries. Of particular rel- evance to this note is Schankler's treatment of the upper Heptodon Range Zone, which is equivalent to Zone Wa7 (e.g., Gingerich 1983) and the Lambdotherium Range Zone TS**! lorn Basin 100 mi Wyoming 0 100 km Fig. 1. Map showing position of Bighorn Basin within the state of Wyoming. 2 PALEOlilOS, VOL. 21, NUMBER 2, SEFEEMRER 2001 (Stucky 1984); it its boundaries are isochronous, it is the basis for the Lostcabinian. Schankler (1980, p. 108) wrote that: |t|he boundary between the Middle and Upper Hcptodon Range-Zones is placed at the first appear- ance of Lamhdotherium at the 670 m level. It should be noted that two specimens of Lamhdotherium are known from levels below 670 m. Both specimens were found in the collection and their presence was not noted in the field. The date or locality of discovery of both specimens raises the possibility of pocket or slope contamination. In any event, Lamhdotherium first appears as a dominant element of the fauna (7% abun- dance) at the 670 m level. However, Schankler did not cite specimen numbers so that others could easily reevaluate the two specimens or recol- lect the relevant localities. Given that Schankler's pioneering work has served as a biostratigraphic framework for numerous studies in the Big horn Basin (e.g., Bown and Rose 1987, Wing et al. 1991, Wing 1998), the author attempted to locate in the Yale Peabody Museum collections the two specimens mentioned by Schankler as occurring below 670 m in order to reevalu- ate the hypothesized FAD of Lamhdotherium. All specimens of Lamhdotherium popoajjicum (the only recognized spe- cies of the genus) were recorded, many of which are tied to the Sehankler-Wing stratigraphic section (Schankler 1980) and to the revised meter levels for the Elk Creek section presented by Bown et al. (1994). Hcptodon, an animal simi- lar in size to Lamhdotherium, was also documented in order to evaluate possible size-related collecting or taphonomic biases in the local record. Finally, the magnetostratigraphic correlations between the Elk Creek section in the central Bighorn Basin and McCullough Peaks section in the north- ern Bighorn Basin are examined in order to determine whether the FADs of Lamhdotherium, Hcptodon, and/or other biostratigraphic indicators arc isochronous or diachronous (as far as resolution permits) within the basin. Institutional abbreviations are as follows: Johns Hopkins University, JHU; United States Geological Survey, USGS; United States National Museum, USNM; and Yale Peabody Museum, YPM. rf:sults Lamhdotherium and Hcptodon at YPM All specimens of Lamhdotherium (Tables 1, 2) and Hcptodon ('Fable 3) at YPM were recorded, and meter levels determined based on the Bown ct al. (1994) revision of the Flk Creek section. Most specimens of Lamhdotherium de- rive from localities at or above the previous Lamhdotherium FAD of 591 m (see Discussion below). A significant pro- portion (-25%) are not easily correlated with the Elk Creek section. The latter category includes a locality described as "!()()' above loc 17" (YPM 39796). Y17 has been traced "into the Schanklcr-Wing section" (Bown et al. 1994) and is at approximately 561 m. It is unclear whether "100 ft" refers to stratigraphic or topographic feet, but considering the low prevailing dip of the beds, the distinction is incon- sequential. Another such ambiguous specimen (YPM 39795) comes from "near loc 197"; Y197 is at 591 m, so the speci- men must also be from close to this level. One especially problematic specimen is YPM 39797, from the "Pelycodus loc." On July 21, 1961, the field party un- der E.I.. Simons visited a site they dubbed the "Pelycodus loc" after collecting at what later became localities Y26a-d. (Only a typed index remains from this expedition, and lo- cality numbers, as opposed to names, were assigned to lo- calities above Y25 subsequent to the typing of the index.) Following "Pelycodus loc" (which became Y27), the party moved on to "Windy Gap," in "sec. 29, T49N, R98VV." Unfortunately, the index apparently lists the incorrect range for Windy Gap (it should be in R96W, twelve miles to the east), but all authors (e.g., Bown et al. 1994) seem to agree that Windy Gap equals Y28, that this locality lies in a pro- gression from Y26 and Y27 on the maps, and that this locality's range was mistyped in the index. Y27 also appears to have had its range mistyped in the index. It was plotted between Y26 and Y28 on the Dutch Kick Flat Quadrangle map, in the progression of localities encountered on July 21, 1961 (in sec. 29, T49N R96W), but the index gives its range as R97W, six miles to the west. In 1963, another field party under E.L. Simons collected at a locality dubbed "Loc. 27 (of 1961)," which, in chronological order, falls between Y71a and Y40. If Y27 was correctly plotted in 1961, one would expect Y71a and Y40 to lie close to Y27. Indeed they do, both falling within T49N R96W. This second collect- ing effort would seem to confirm that Y27 was correctly plotted in 1961 but misrecorded in the index. In 1977, D. M. Schankler collected at localities named "Pelycodus loc" and "Pseudo-27." "Pelycodus loc" was col- lected after Yl (in sec. 29, T49K R97VV), and "Pseudo-27" after Y18 and prior to "20 strat ft. above Loc. 40" (in sec. 29, T49N R96W). On the Dead Indian Hill Quadrangle map in the YPM collections, there is a "Pelycodus loc" that has been marked with an X in sec. 29, T49N R97W. This plot must correspond with the "Pelycodus loc" collected in 1977. One suspects, then, that the field party used the in- dex locality data, went to the relevant subsections of T49N R97W (where it found fossils), and assumed this new local- ity to be the original "Pelycodus loc." However, since its position conflicted with the Y27 already plotted on the map, Y27 was dubbed "Pseudo-27." In this interpretation, the "Pelycodus loc" collected in 1977 was a new locality, and "Pseudo-27" of 1977, Y27 of 1961 and 1963, and the origi- nal "Pelycodus loc" of 1961 are all equivalent. The 1977 "Pelycodus loc" plots almost precisely on USGS locality D- 1814, which, judging by nearby localities, should fall near 591 m in the revised section. One discrepancy remains in this interpretation, however. In his 1980 paper, Schankler SMJ'/H-RKASSESSING I.AMBDOTHERIUM DATA Table 1. List of all specimens of Lambdotherium popoagicum from the Bighorn Basin at YPM, except those from below 591 m in the Elk ('reek section. "R" and "L" refer to right and left, respectively. Locality I [eight (m )•' Specimen No. (YPM) Element(s) 8 591 17105 L dentary w. P4-M2 8 591 22091 RMsfrag 160 591 30340 LMJ3, RP4, RMxfrag 160 591 41205 ^ 162 591 27724 LMX 196 591 29338 LMX near loc 197 -591" 39795 LMX 199 591 28629 LM, „ RM2 3 frags 199 591 28652 R dentary w. P4-M, 199 591 36811 RMX frag "I'elycodns loc"' -591 39797 LMX, RM3 frag 162c 591 28597 2 RM* frags 3 601 17110 RM\ LMX frag, RMX frag, RMX frag, LdP' 3 601 17112 RMX frag, RM5, LM2 3 601 17113 LM,, RP„ RM,, RM,, L dentary w. 2 Ms ft 3 601 27776 RMX 3 601 30318 LMX, RM\ LM3, RdP4 3 601 39793 RM, 3 601 39794 LM2 frag, LMX frag 33 601 18759 RdP4? 33 601 18762 RM'? 33 601 18763 LMX frag, RP, 163 601 28647 LM" or LdP4' 163 601 29441 LMX (worn) 195 601 27500 RdP 24 611 17109 RP4, 2 RMX frags 24 611 28539 RM, 24 611 33974 R + L dentarics and maxillae 32 611 18744 LP34, M2 „ RMS frag, RM, frag 32 611 18749 LM, 12 17108 LM,, RM,, 12 26265 LdP, 12 26374 RP4, i-P4 12 26383 I. dentary vv. dP4M, 30 18716 LP4, RMX frag 30 18718 2 RMX frags, LMX frag 30 18719 LP54, LMxfrag 31 17738 anterior skull 169 28649 3 RMX frags 243 26326 RM,, 312 26196 LM2 100' above loc 17'' 39796 LM21 frags 3046 71 28581 LM'2, RMX No data 17111 RP4, LM3 Not determined 27683 RM, 2, LP4, LM2,, R+L Mx frags 112-71 28596 R dentary w. P3-M3 112 71 29281 LM, 1572-71 27689 I. dentary w. P2-M, Notes: a-Rcvised meter-levels in Klk Creek section from Bown et al. (1994); b-Y197 is at 591 m; c-See text; d-Y17 is at 561 m 1'ALEOBIOS, VOL 21, NUMBER 2, SEPTEMBER 2001 Tabic 2. List of the four specimens of Lambdothcrium popoajjicum from below 591 m in the Elk Creek section. Two of the speci- mens (YPM 33936 and 56526) were removed from the batches in which they were originally catalogued. "Dare catalogued" refers to the date the individual Lambdothcrium specimen was catalogued, not the original specimen batch. For specimen history, see the text. "R" and "1," refer to right and left, respectively. YPM Spec. No. 56526 17107 33936 39798 Yale Loc. No. 17 I- 23 21 Field No. 61-175 61-271 61-285 62 275 Collection date June 29, 1961 July 4, 1961 July 6, 1961 July 13, 1962 Collector Party L. Radinsky Party J. S. Alperr Description in f ield list "large 1 leptodon "lambdothcrium "miscellaneous "1 lyopsodus I,J and etc." upper" Hyracol herium" Mj-M3" Dale catalogued Feb. 2001 Oct. 1961 Apr. 6, 1977 July 13, 1981 Specimen descri ption partial I.M, LMS RMX fragment partial RM. plots Y27 and "l'clycodus loc" at the same level, -491 m (revised; see below). The discrepancy is resolved, however, if one assumes that Schankler realized, between collecting in 1977 and writing his 1980 paper, that the index had been mistyped; Schankler has confirmed this interpretation (personal communication, 2001). Specimens of Lambdothcrium below the 591 -m level Four specimens of Lambdothcrium are from localities below Schanklcr's Lambdothcrium FAD (Table 2). These are YPM 39798 (locality Y21, 501 m), 33936 (Y23, 511 m), 56526 (Y17, 561 m) and 17107 (Y17, 561 m). [Mcter- levcls are based on Bown et al. (1994). | Significant strati- graphic miscorrelation is unlikely, particularly for locality Y23, as it lies directly on one of Bown et al.'s (1994) mea- sured local sections (Rock Waterhole Creek section). The specimens are readily identifiable as Lambdothcrium (Fig. 2), but the possibility remains that any or all of these four may represent, as Schankler (1980, p. 108) argued, "pocket or slope contamination". Slope contamination for localities Y21 and Y23 is unlikely, given that they occur in areas of low relief (see Bown et al. 1994, PI. 1). Y17, on the other hand, lies close to a region of high relief. Of these "early" lambdothcrium specimens, two had been catalogued prior to Schanklcr's (1980) paper and probably are the two specimens to which he referred (Table 2). The first is YPM 17107 (Fig. 2a). It was collected by L. Radinsky on July 4, 1961, and appears as "Lambdothcrium upper" in the field catalogue of E.L. Simons' 1961 expedition. Clearly, it was recognized in the field, contra Schankler (1980), and it was catalogued in October 1961. The second of Schanklcr's specimens must be YPM 33936 (Fig. 2b). Collected on July 6, 1961, by the field party, this fragment was originally catalogued with "miscellaneous Hyracotherium" teeth and fragments in October 1961, but was subsequently recognized as different and re-catalogued on April 6, 1977. The specimens of Hyracotherium with the same field number are YPM 16997 and 16998. Two additional specimens of Lambdothcrium have been discovered subsequent to Schanklcr's (1980) study. Based on its field number, YPM 39798 (Fig. 2c) appears lo have been collected by J.S. Alpert on June 13, 1962. However, the field index only lists "Hyopsodus LJ M,-M." (Tabic 2). This Lambdothcrium specimen, a partial right M., was cata- logued on July 13, 1981. The Hyopsodus specimen to which the field index refers could not be found. The second addi tional Lambdothcrium specimen is YPM 56526. It was col lected on June 29, 1961 by the Simons field party and noted in the field catalogue as "large I lepiodon and etc." The specimen batch was catalogued in October 1961 as YPM 17092, and the specimen box indeed contained a Hcptodon left Mx and several unidentified fragments of teeth; but two of these fragments constitute a partial left M, of Lambdothcrium popoajjicum (Fig. 2d; now catalogued as YPM 56526), and two others, a right M. of Hyracotherium index (now catalogued as YPM 56525). The stratigraphic distribution and abundance of Lambdothcrium and Hcptodon specimens, which are of com- parable size, are shown in Figure 3. There does not appear to be a size bias— Hcptodon is abundantly represented in some intervals (e.g., 531-551 m) even when Lambdothcrium is not—although a comprehensive evaluation of all fossils and localities in this interval would be necessary to quantify this assessment. In terms of number of specimens, Hcptodon is more abundant in the collections than Lambdotherium. No specimen of Hcptodon was discovered from a locality lower than its previously documented FAD (Fig. 3; Schankler 1980, Bown et al. 1994). In summary, a total of four Lambdothcrium specimens in the collections come from localities below 591 m in the Elk SMITH REASSESS INC; LAMBDOTHERIUM DATA B Fig. 2. Specimens of Lambdothcriiim poponjjiciim from below the previously documented 591-m FAD, A. YPM 17107, left Mx, lo- cality Y17. B. YPM 33936, right Mx fragment, locality Y23. C. YPM 39798, partial right M,, locality Y21. D. YPM 56526, partial left M„ locality Y17. Scale bars are 5 mm. Creek section. Two were mentioned by Schanklcr (1980). Of these, one was recognized in the Held as Lamb doth crium (contra Schanklcr 1980), and the other, a tooth fragment, had originally been batch-catalogued with Hyracothcrinm and was subsequently removed. A third specimen of Lambdothcriiim has little documentation and could be a spurious occurrence; this specimen is the lowest of the four discussed here. The fourth specimen was batch-catalogued (in two pieces) with a Hcptodon upper molar and other frag- ments but later recognized and re-catalogued. The speci- mens from Y17 were not collected in the year (1977) that the party officially strayed up to "100 ft" above the actual locality. DISCUSSION The Lostcabinian was first characterized biostra- tigraphically by the presence of Lambdothcriiim (Granger 1914), and the term has been widely used for designating a temporal unit in the early Eocene. It is equivalent to the Upper Hcptodon Range Zone of Schanklcr (1980) and Wa7 of Gingerich (1983). Although rocks of the northern Big horn Basin and contiguous Clark's Fork Basin have been 6 1'ALEOBIOS, VOL. 21, NUMBER 2, SEPTEMBER 2001 Tabic 3. List of all catalogued specimens of Heptodon .ocality Height (m)-" Specimen 227N 457 34419 44 463 22127 34 469 17080 34 469 18811 45 470 22116 45 470 22129 45 470 22132 45 470 22139 45 470 24094 6) 474 17081 61 474 17098 61 474 37116 318 478 36088 249 480 24045 249 480 24093 40 481 22131 40 481 22133 40 481 41177 42 481 22123 42 481 22124 42 481 36083 28 491 36051 21 501 17065 21 501 17068 21 501 17070 21 501 22134 -7 501 22140 77 501 41175 71a 501 22130 23 511 17063 23 511 17066 23 511 17067 23 511 17069 23 511 17071 23 511 17072 314 511 29242 314 511 36079 36 521 18835 36 521 18845 36 521 18852 18b 521 27687 18b 521 36807 175 531 27837 175 531 27838 175 531 27839 175 531 27842 175 531 27851 175 531 36077 176 531 28913 176 531 35197 176 531 36078 176 531 36089 178 531 28024 185 531 27750 185 531 28356 185 531 28727 from the Bighorn Basin at Yl'.Yl. No. Elemcnt(s) LM\ LMS frag R maxilla w. P!-M\ L maxilla w. M1 ¦' RM3 (worn) LM12, LM, frag R dentary w. P„ M, R dentary w. M, ,, I. maxilla w. P5'4 I. dentary w. M, ,, RMX LM13?, I. dentary vv. P2-M_, R dentary w. M, , LMS LMS LMS frag LMX, LP\ RMX frag, LdP4? frag LM23 RM2, RM, frag R dentary w. P, M,, LM" frag R dentary w. P.M., L dentary w. I\ M, RM, I- dentary vv. P^-M, R dentary w. M, , RP4 LMX frag I.P,, I.MS, 2 LM. frags, 2 RMX, RM. LP4, RP4, RP4, RMS frag, LM, . frags R dentary w. P2-M, 2 LMX frags R maxilla vv. M2, I. maxilla frag RM, frag LMX, RMS maxillae: RdP24, M1, LdP84, M1, M2 frag LM3 LM1-', RP4 MISSING RM„ RM! LM2" I. dentary w. dP4-M, RMX frag LM2 LMS LM3 I. dentary w. P,-M,, R dentary w. P;-M, LMX (worn) L dentary w. Pi4 R maxilla w. M,; R dentary w. M. R dentary vv. P.-M, RMX R dentary w. M, , R dentary w. Mx RM, LMX frag R dentary w. M, , R maxilla w. dP24 R dentary w. dP -M, LMX SMITH-REASSESSING LAMBDOTHHRWM DATA 7 Table 3. (continued) List of all catalogued specimens of Heptodon from the Bighorn Basin at YPM. Locality Height (m)* Specimen No. Element(s) 185 531 28950 RM3 185 531 36084 RMX frag 185 531 36086 LMN 185 531 36087 LMX, RM1 185 531 41179 LMS frag 191 531 29621 I. dentary w. Mv LM' 16 541 17056 RMX 16 541 17060 I. dentary w. P 16 541 17061 RM* 16 541 17062 1. dentary vv. M, 2, RM, 167 541 36808 RMS 180 541 28740 RM, frag, LM', LM2, RM'2, 2 RM! 181 541 28062 lu\\ 181 541 28859 LM, 181 541 36081 LM' 181 541 36582 LMX 181 541 36809 LMX, LP4, RMX frag 184 541 28912 LM2 184 541 29047 RM, frag 184 541 36080 LP4, LM3 192 546 27603 I. maxilla w. M' 192 546 35192 LM. 192 546 35239 I. dentary w. M,, frags 192 546 36074 RMS 193 546 27436 I.MX frag 193 546 27521 I. dentary w. P, 193 546 27636 R dentary vv. P,-M, 193 546 28534 L dentary w. I'4-M, 193 546 29465 RM, 193 546 35259 L dentary w. P.-M, 315 546 30312 L dentary w. dP4-M2, R dentary vv. dl'.-M, 315 546 41180 R dentary w. M 10 551 17051 R maxilla w. M2 3 188 551 27588 R dentary w. M, , 188 551 32882 L dentary vv. P.-M,, R dentary vv. P4-M, 17 561 17092 RMX, many frags 2 571 30267 RM' 160 591 28651 L dentary vv. Ms, R dentary vv. M,. frag 33 601 41214 RMX 32 611 18753 LMX frag 7 641 17054 RM3 4 17050 skull and jaw frags, phalanx 5 17053 LM" frag 11 17093 LM13, RM\ RM3, Rl\? IS 17058 RMS frag 18" 17064 LMV (worn) 18" 17074 RP\ RMX, RMS, LMS frag 18" 17077 LM,, frags, RM, , 38 27662 LM\ RMX 38 36090 LMX frag 246 41 181 RM, frag 255 24074 LM,, RM,, RM,, LP, 310 33017 LM3 331b 32637 RM3 Notes: a Revised meter-levels in Elk Creek section from Bown et al. (1994); b—Sublocality information (i.e., 18a, 18b) is unknown, so meter level is unavailable. 30 25 0) c CD E o CD I 20 15 10 PALEOBIOS, VOL. 21, NUMBER 2, SEPTEMBER 2001 XL 1_I ¦ Lambdotherium ? Heptodon n L H 451 461 471 481 491 501 511 521 531 541 551 561 571 581 591 601 611 621 631 641 Meter-level category Fig. 3. Number of.specimens of Lambdotherium and Heptodon in ten-meter stratigraphic categories (e.g., 451-460 m) for the Elk ("reck section in the central Bighorn Basin, based on all catalogued specimens at Yl'M. If original abundance is assumed to be tem- porally constant for each animal, it is evident that no size bias can account for the lack of appearance of Lambdotherium in the 521- 560-m portion of the section. studied using paleomagnctism (Clyde et al. 1994) and bio- stratigraphy (e.g., Gingerich 1989, 1991, Clyde 1997), the central and southern parts of the Bighorn Basin are impor- tant in that they preserve higher strata (including Lostcabinian) not found in more northerly regions. For ex- ample, the McCullough Peaks area includes only the base of the Lostcabinian (Clyde et al. 1994), and magnetostratigraphic study of the Clark's Fork Basin (But- ler et al. 1981) extends only to the basal Wasatchian. Other regions yielding Lostcabinian faunas, such as the Wind River Basin, have not been studied using magnetostratigraphic or radiometric methods. Thus, the accurate depiction of this biostratigraphic zone in the central Bighorn Basin is impor- tant. Significantly, it also provides crucial tests of the pre- sumptive isochroneity of certain mammalian FADs. Previous treatment of the Lambdotherium FAD Three different meter-levels in the central Bighorn Basin have been published as the Lambdotherium FAD. The first is that ofSehankler (1980, p. 108), who argued that the datum should be placed at 670 m despite possible occur- rences lower in his section, because that is where Lambdotherium first appears as a "dominant element of the fauna (7% abundance)". Bown et al. (1994) revised the meter levels for the upper portion of the F.Ik ("reek section, and using their formula (p. 46: subtract 79 m from original sec tion) one arrives at 591 m for the Lambdotherium FAD. This figure is used by Wing et al. (2000). The second Lambdotherium FAD used is found in Tauxe et al. (1994, Fig. 4), who illustrated it as -600 m in the revised section. These authors give no reference, but K.D. Rose (personal communication, 2000) states that no Lambdotherium from USGS/JHU collections (stored at USNM) occurs below -600 m; it seems then that Tauxe et al.'s (1994) FAD was informed by these collections. 'The third published FAD is that of Wing et al. (1991), who cited Schankler (1980) and Bown et al. ("1991") as placing the Lambdotherium FAD at 580 m (in the revised section). Bown et al. ("1991") was the in-press version of a monograph that ultimately appeared three years later (Bown et al. 1994) and did not provide new information on the Lambdotherium FAD. Schankler .SA//7H-REASSESSING LAMBDOTHERIUM DATA 9 (1980) is the only source for the FAD, and he gave it as 670 m, or 591 m (revised). Magnetostratigraphic correlation and implications Magnetostratigraphic studies of the Bighorn Basin in the past decade allow the correlation of biostratigraphic events willi the Geomagnetic Polarity Time Scale (GITS). In Clyde et al.'s (1994) study of the McCullough Peaks section (northern Bighorn Basin), Bunophorus, represented by B. ctsajjicus, first appeared near the top of C24r (Fig. 4), Heptodon during C24n.3n, and Lambdothcrium in the brief reversed subchron C24n.lr. In the central Bighorn Basin section, both Bunophorus and Heptodon first appeared dur- ing C24n.3n (Fig. 4). Both a -600-m Lambdothcrium FAD (Tauxe et al. 1994) and the previous 591 -m FAD (e.g., Wing et al. 2000) would fall in C24n.lr (Flynn and Tauxe 1998). However, a FAD at 501 or 511 m in the revised Flk Creek section would pull the FAD down to the top of C24n.3n (Fig. 4). If we accept either of these stratigraphically lower records as correct, Lambdothcrium may have appeared up to - 200 kyr later in the McCullough Peaks region than further south, rendering its FAD diachronous. The Bunophorus FAD also appears to be diachronous across this same region (Flynn and Tauxe 1998), but that taxon ap- pears later in the central Bighorn Basin than in the north. Other Wasatchian FADs all appear to fall within the long reversed interval C24r. The limits of magnetostratigraphic resolution do not permit testing the isochroneity of these FADs, such as those of Hymcotberium pernix and Estbonyx bisukatus. In selecting taxa as biostratigraphic indicators, it is hoped that [heir appearances are essentially isochronous across the geographic area of interest. Local and regional diachroneity limits the utility of a given taxon for correlating strata in non-contiguous stratigraphic sections. In the case of the Lambdothcrium FAD, there are several, non-mutually ex- clusive possibilities for resolving the diachroneity dilemma. First, all four occurrences below 591 m in the Flk Creek section could be regarded as contaminants. This solution is favored by Schankler (1980) and is not controverted by the fact that no specimens of Lambdothcrium have been col- lected from localities below -600 m by USGS/JHU crews (K.D. Rose, personal communication, 2000). If the prob- ability of a mix-up in the collections or in the field is pro- portional to the number of specimens, then it would be surprising that no Heptodon specimens were found in the collection from localities below the 451-m FAD for that taxon, given its greater relative abundance (see also Bown et al. 1994). Contamination does not appear to be a likely explanation for all four specimens. A second possibility is that Lambdothcrium actually oc- curs lower in the McCullough Peaks section than current collections indicate. Sampling density in the McCullough Peaks section falls from a maximum of approximately 1500 specimens in Wa5 (Bunophorus Interval Zone) to just over 500 specimens in the upper portion of Wa6 (approximately 1600 52.51 53.01 53.5 H 54.0 54.5 55.0 55.5 Vvvi GPTS 8001 600 \ 400 ^ 200 0 m 1400- 1200-1 10001 S Elk Creek 800 600 400 200 0 m McCullough Peaks Fig. 4. Magnetostratigraphic correlation of the Klk ("reek and McCullough Peaks sections to the GITS, hung on the base of C24n.3n. Correlations follow Tauxe et al. (1994) and Clyde et al. (1994). The ranges of taxa discussed in the text are indicated in each section, with the following abbreviations: /{- Bunophorus, H= Heptodon, and /. = Lambdothcrium. Last ap- pearance data in the McCullough Peaks section are not intended to reflect ture last occurrences (see Clyde 1997). Data in the GPTS are taken from Wing (1998); Elk Creek stratigraphic dis- tributions are based on the Yl'M data, except tor the Bunophorus FAD from Bown et al. (1994); all McCullough Peaks data are from Clyde et al. (1994). The asterisks indicate the approximate position of the carbon isotope excursion. the Middle Heptodon Range Zone; Clyde 1997). If the ap- parent stratigraphic range of Lambdothcrium in the Elk Creek section is correct, then it is within this less-sampled interval that we would expect to first find Lambdothcrium in the McCullough Peaks section. Statistical analysis of known occurrences and intensified sampling in the McCullough Peaks region could test this hypothesis. Third, cither Tauxe et al. (1994) or Clyde et al. (1994) may have miscorrelated their sections to the GPTS. If an entire section were miscorrelated, this possibility would be contravened by the opposite sense of diachroneity of the Bunophorus FAD. Even if only the upper part of a section had been miscorrelated, it would remain true that in the Elk Creek section, the Lambdothcrium FAD occurs in a normal zone, while in the McCullough Peaks section, the FAD occurs in a reversed zone. Flynn and Tauxe (1998) rejected the idea of miscorrelation in their discussion, though it remains conceivable that future studies may alter this as- sessment. Finally, the Lambdothcrium FAD may in fact be diachronous. As noted above, stratigraphic distribution may 10 PA LEO BIOS, VOL. 21, NUMBER 2, SEPTEMBER 2001 be strongly dependent on the sampled paleocnvironment, and Bown (1979, p. 138) submitted that Lambdotberium may have "differential local range-zones." Although no physi- cal barriers to dispersal are known to have been present within the early Eocene Bighorn Basin (Bown 1979), different paleoenvironments may have restricted the geographic range of Lambdotberium if it was in some respect stenotopic (as suggested by Bown 1979 for Homogalax). Bown and Beard (1990) found no statistically significant support tor the dif- ferential preservation of perissodactyls in particular pedofacies. However, their analysis was conducted at the ordinal level and did not address the distribution of indi- vidual taxa. A diachronous FAD would serve to emphasize the purely biostratigraphic nature of the Upper Hcptodon Range Zone (=Zone Wa7, Lambdotberium Range Zone). Biozones are not required to have isochronous boundaries (North Ameri- can Commission on Stratigraphic Nomenclature 1983, Sal- vador 1994), whereas chronozones are. If Lambdotberium does have a diachronous FAD, the use of this FAD as the basis for chronostratigraphic and geochronologic demarca- tion (i.e., the Lostcabinian "subage") is undermined. The utility of the Middle Hcptodon Range Zone as either a bio- siraiigraphic or chronostratigraphic unit (as denned b) Schankler 1980) would also be questionable, since the base of both Middle and Upper Hcptodon Range Zones would be nearly coincident in the Elk Creek section. It must be emphasized that, even if the Lambdotberium FAD is diachronous, it would remain a good indicator (sensu Murphy 1977, Woodburne 1977) for the time represented by C24n.lr and later. Even if all of the occurrences from below 591 m are real, it is clear that Lambdotberium was not abundant prior to the 591 -m level (as Schankler noted). Hcptodon, which is similar in size to Lambdotberium, is abun- dantly represented by specimens below 591 m, while the latter taxon has only scattered representation (Fig. 3). In summary, the Lambdotberium FAD used to define the base of the Lostcabinian North American provincial "subage" deserves more careful consideration. All references to particular meter levels for central Bighorn Basin sections trace their origin to Schankler's (1980) work. The two Lambdotberium specimens he cited as occurring below the accepted FAD have been identified (YI'M 17107 and 33936). This paper provides locality numbers (Y17, Y21, and Y23) and specimen numbers (YPM 17107, 39798, 33936, and 56526) that may be used to further examine the possibility of a lower Lambdotberium FAD. If one accepts the data presented here, this important FAD may be diachronous by 200 kyr within the same depositional basin. Alternatively, less intense sampling in the upper part of the McCullough Peaks section may have resulted in an artificially high FAD. It is a testament to the growth of stratigraphic science that one can argue about differences of a "mere" two hundred thousand years. However, as stratigraphic precision increases, certain assumptions that were acceptable when absolute correlation was possible only over million-year rimescales may not remain tenable. ACKNOWLEDCEM ENTS The author would like to thank P.A. Holroyd for first suggesting this project and for her comments on drafts of this paper; M.A. Turner and I.. Murray (YPM) for access to specimens, field notes, and maps, and for their time dis- cussing the locality 27/Telycodus loc" problem; D. M. Schankler for discussion of the "Pelycodus locality" and in- formation about late 1970s collecting practices in the Big- horn Basin; M. Fox and W. Sacco for assistance with the photography; and P.D. Gingerich, K.D. Rose, and S.L. Wing for their criticisms, which greatly improved this paper. LITERATURE CITED Bown, T.M. 1979. Geology and mammalian paleontology of the Sand ("reck facies, lower Willwood Formation (lower Eocene), Washakie County, Wyoming. Geological Survey of Wyoming Memoir 2A-151. Bown, T.M., and K.C. Beard. 1990. Systematic lateral variation in the distribution of fossil mammals in alluvial paleosols, lower Eocene Willwood Formation, Wyoming, pp. 135-151 iiiT.M. Bown and K.D. Rose (eds.). Dawn of the Age of Mammals in the Northern Part of the Rocky Mountain Interior, North America. Geological Society of America Special Paper 243. Bown, T.M., and K.D. Rose. 1987. Patterns of dental evolution in early Eocene anaptomorphine primates (Omomyidae) from the Bighorn Basin, Wyoming. Palcontological Society Memoir 23:1-162. Bown, T.M., K.D. Rose, E.L. Simons, and S.L. Wing. 1994. Dis tribution and stratigraphic correlation of upper Paleocene and lower Eocene fossil mammal and plant localities of the Fori Union, Willwood, and Tatman Formations, southern Bighorn Basin, Wyoming. U.S. Geological Survey Professional Paper 1540:1-103. Butler, R.F., P.D. Gingerich, and E.H. Lindsay. 1981. Magnetic polarity stratigraphy and biostratigraphy of Paleocene and lower Eocene continental deposits, Clark's Fork Basin, Wyoming. Journal of Geology 89:299-316. Clyde, W.C. 1997. Stratigraphy and mammalian paleontology of the McCullough Peaks, northern Bighorn Basin, Wyoming: Implications for biochronology, basin development, and com munity reorganization across the Paleocene-Eocene boundary. Ph.D. diss.. University of Michigan, Ann Arbor, MI. Clyde, W.C, J. Stamatakos, and P.D. Gingerich. 1994. Chronol ogy of the Wasatchian Land-Mammal Age (early Eocene): Magnetostratigraphic results from the McCullough Peaks sec lion, northern Bighorn Basin, Wyoming. Journal of Geology 102:367-377. Flynn, J.J., and L. Tauxe. 1998. Magnetostratigraphy of upper Paleocene-lower Eocene marine and terrestrial sequences, pp. 67-90 in M.-P. Aubry, S. Lucas, and W.A. Berggren (eds.). Late Paleocene lvarly Eocene Climatic and Biotic Events in the Marine and Terrestrial Records. Columbia University Press, .W/7H-RKASSESSING LAMHDOTHE1UUM DATA 11 Now York. Gingerich, P.D. 1983. Paleocene-Eocenc fauna] zones and a pre- liminarv analysis of Laramide structural deformation in the Clark's Fork Basin, Wyoming. Wyoming Geological Association Guidebook 34:185 195. Gingerich, P.D. 1989. New earliest Wasatchian mammalian fauna from the Eocene of northwestern Wyoming: Composition and diversity in a rarely sampled high floodplain assemblage. Uni- versity of Michigan Papers on Paleontology28:1-97. Gingerich, P.D. 1991. Systematics and evolution of early Eocene Perissodactyla (Mammalia) in the Clarks Fork Basin, Wyoming. Contributions from the Museum of Paleontology, University of Michigan 28:181-213. Granger, VV. 1914. On the names of the lower Eocene faimal ho- rizons of Wyoming and New Mexico. Bulletin of the American Museum of Natural History 33:201-207. Hedberg, H.D. (ed.). 1976. International Stratigraphic Guide; A Guide to Stratigraphic Classification, Terminology, and Pro- cedure. John Wiley and Sons, New York. 200 pp. Jepsen, G.I.. 1930. New vertebrate fossils from the lower Eocene of the Bighorn Basin, Wyoming. American Philosophical Soci- ety, Proceedings 69:117-131. Krishtalka, I.., R.M. West, C.C. Black, M.R. Dawson, J.J. Flynn, VV.D. Turnbull, R.K. Stucky, M.C. McKenna, T.M. Bown, D.J. Golz, and J.A. Lillegraven. 1987. Eocene (Wasatchian through Puchcsnean) biochronology of North America, pp. 77-117 in M.O. Woodburne (ed.). Cenozoic Mammals of North America: Gcochronology and Biostratigraphy. Univer- sity of California Press, Berkeley, CA. Murphy, M.A. 1977. On time-stratigraphic units. Journal of Pale- ontology 51:213-219. North American Commission on Stratigraphic Nomenclature. 1983. North American stratigraphic code. American Associa- tion of Petroleum Geologists Bulletin 67:841-875. Salvador, A. (ed.). 1994. International Stratigraphic Guide; A Guide to Stratigraphic Classification, Terminology, and Pro- cedure. Geological Society of America, Boulder, CO. 214 pp. Schankler, D.M. 1980. Faunal zonaiion of the Willwood forma- tion in the central Bighorn Basin, Wyoming. University of Michi- gan Papers on Paleontology 24:99-114. Simpson, G.G. 1937. Notes on the ("lark fork, upper Paleocene, fauna. American Museum Novitates 954:1-24. Stucky, R.K. 1984. Revision of the Wind River faunas, early Eocene of central Wyoming. Part 5. Geology and biostratigraphy of the upper parr of the Wind River Formation, northeastern Wind River Basin. Annals of Carnegie Museum 53:231 294. Tauxe, 1.., J. Gee, Y. Gallet, T. Pick, and T. Bown. 1994. Magnctostratigraphy of the Willwood Formation, Bighorn Basin, Wyoming: New constraints on the location of the Pale- ocene/Eocene boundary. Earth and Planetary Science Letters 125:159 172. Van Houten, F.B. 1944. Stratigraphy of the Willwood and Tatman formations in northwestern Wyoming. Geological Society of America Bulletin 55:165-210. Van Houten, F.B. 1945. Review of latest Paleocene and early Eocene mammalian faunas. Journal of Paleontology 19:421 461. Wing, S.I-. 1998. Late Palcocenc-carly Eocene floral and climatic change in the Bighorn Basin, Wyoming, pp. 380 400 in M.-P. Aubry, S. Lucas, and W.A. Berggren (eds.). I .ate Palcoccnc- Early Eocene Climatic and Biotic Events in the Marine and Terrestrial Records. Columbia University Press, New York. Wing, S.L., H. Bao, and P.L. Koch. 2000. An early Eocene cool period? Evidence for continental cooling during the warmest part of the Cenozoic. pp. 197-237 in B.T. Huber, K.G. MacLeod, and S.L. Wing (eds.). Warm Climates in Earth His- tory. Cambridge University Press, Cambridge, UK. Wing, S.L., T.M. Bown, and J.D. Obradovich. 1991. Early Eocene biotic and climatic change in interior western North America. Geology 19:1189-1192. Woodburne, M.O. 1977. Definition and characterization in mam- malian chronostratigraphy. Journal of Paleontology 51:220-234. Note added in proof: The USNM collections include twelve specimens ofLambdotherittm, three of which are tied to the Elk Creek section (A. Chew, personal communication, 2001).