Palaeogeography, Palaeoclimatology, Palaeoecology, 63 (1988): 141-157 Elsevier Science Publishers B.V., Amsterdam -- Printed in The Netherlands
141
SAMPLING AND FAUNAL TURNOVER IN EARLY EOCENE MAMMALS C A T H E R I N E B A D G L E Y a n d P H I L I P D. G I N G E R I C H Museum of Paleontology, University of Michigan, Ann Arbor, MI 48109 (U.S.A.) (Received July 21, 1987)
Abstract Badgley, C. and Gingerich, P. D., 1988. Sampling and faunal turnover in Early Eocene mammals. Palaeogeogr., Palaeoclimatol., Palaeoecol., 63: 141-157. Faunal turnovers in the fossil record are episodes of synchronous appearance and disappearance of species from a community, often resulting in net change in species richness. We studied the biostratigraphic record of faunal turnover involving early Wasatchian (early Eocene) mammals from the Clark's Fork Basin, Wyoming, U.S.A. Two faunal turnovers occur in this record - - one at the base of the Wasatchian, comprised mainly of appearances of taxonomically and ecologically distinctive species, and a later one, Biohorizon A of Schankler (1980), comprised mainly of disappearances, especially of carnivorous species. This study focuses on Biohorizon A. In the record of the Clark's Fork Basin, Biohorizon A may be an artifact of sampling. Sample size and species richness are highly correlated (r = 0.95) throughout this record. Moreover, sample size and species richness fluctuate markedly between successive stratigraphic intervals; peaks of appearances coincide with large sample sizes and peaks of disappearances with low sample sizes. The peaks and valleys in fossil productivity over time mask the real timing of appearances and disappearances of species. Changes in fossil productivity in the stratigraphic section may result from changes in exposure area, taphonomic factors, or ecological factors. Evaluation of the effects of sampling is a necessary prerequisite for investigating the chronological and ecological significance of faunal turnovers.
Introduction F a u n a l t u r n o v e r refers to c h a n g e in species c o m p o s i t i o n of a f a u n a t h r o u g h the disappearance of some species and the a p p e a r a n c e of o t h e r species. The d i s a p p e a r a n c e s m a y be local or global e x t i n c t i o n s or m a y be o n l y t e m p o r a r y d i s a p p e a r a n c e s ; the a p p e a r a n c e s m a y be local or global first a p p e a r a n c e s or else reappeara n c e s of t a x a t h a t were f o r m e r l y present. The fossil r e c o r d a b o u n d s in f a u n a l t u r n o v e r s , T h e y m a r k m a n y c h r o n o l o g i c a l b o u n d a r i e s in the g e o l o g i c a l time scale. The m a j o r f a u n a l t u r n o v e r s - - t h o s e of b r o a d t a x o n o m i c scope and wide g e o g r a p h i c c o v e r a g e - - a r e i m p o r t a n t as events in e c o l o g i c a l a n d e v o l u t i o n a r y hist o r y and as guides to r e g i o n a l c o r r e l a t i o n , 0031.0182188/$03.50
F a u n a l t u r n o v e r s h a v e been r e c o g n i z e d over a c o n s i d e r a b l e r a n g e of spatial and t e m p o r a l scales. At a large scale, for example, Kitchell and C a r r (1985) modelled diversification of and interactions among three Phanerozoic marine f a u n a s (at the family level) as compiled by S e p k o s k i (1981). At a small scale, for example, S c h a n k l e r (1980) identified episodes of extinction a n d i m m i g r a t i o n w i t h i n the m a m m a l i a n f a u n a for p a r t of the e a r l y Eocene, w i t h i n p a r t of a d e p o s i t i o n a l basin. At an i n t e r m e d i a t e scale, for example, Webb (1984) described o r i g i n a t i o n and e x t i n c t i o n w i t h i n the mammalian f a u n a of N o r t h A m e r i c a d u r i n g the middle to late M i o c e n e a n d also the i n t e r c h a n g e of m a m m a l s b e t w e e n N o r t h and S o u t h A m e r i c a d u r i n g the late Neogene.
© 1988 Elsevier Science Publishers B.V.
142
While the essence of faunal turnovers is the synchronous a p p e a r a n c e a n d d i s a p p e a r a n c e o f taxa, two other aspects are relevant to understanding their mechanism and significance: net change and rate of change in the number of species present, that is, species richness. Faunal turnovers may be rapid (from 103 to 105 yr) to slow (some 106 yr) in geological time. Some paleontologists have considered different causal mechanisms to pertain to slow versus rapid faunal turnovers (e.g., Olson, 1980; Vrba, 1985), although this issue remains open. In many clear instances of turnover, neither the rate of change of species richness nor the rates of species appearances and disappearances can be specified precisely, owing to the absence of absolute chronological markers. A substantial degree of faunal turnover could result in a net change of zero in species richness, but the usual pattern involves a net increase or decrease. Webb (1984) proposed two types of faunal turnover - - those in which extinctions outnumber originations and those in which originations outnumber extinctions. These types, distinguished on the basis of change in species richness, are associated with different environmental causes, The present paper illustrates the role of sample size in the record of an apparent faunal ....
-T
---~-~ CLARK'S ~ \
o~ L~ FORK .power,
..'%.~
BAs,N
turnover in the early Eocene history of mammals from the Bighorn Basin, Wyoming (U.S.A.). Sample size reflects taphonomic or ecological influences on the fossil record. This aspect of faunal turnover is rarely given explicit consideration in evaluation of the rate or the putative mechanisms of faunal change. In the example presented here, we find a close correlation between the components of faunal turnover and changes in sample size through the stratigraphic section.
Background The Bighorn Basin of northwestern Wyoming, U.S.A. (Fig.l), records a set of major and minor faunal turnovers in the early Cenozoic history of mammals from western North America. Major faunal turnovers correspond to the transitions between the North American land mammal ages (e.g., Puercan, Clarkforkian, Wasatchian). Minor faunal turnovers often correspond to faunal zones within land mammal ages. Faunal zones are typically based on stratigraphic ranges of one to a few taxa, while faunal turnovers are based on surveys of the whole fauna. Here, we discuss the record of a faunal turnover in early Wasatchian (early Eocene)
Montana
w.o,°o
\
I ~-.
'~
,,o. ~
.
t"
b-.
l
~
%
.
.
.
c~.,,o,,
t~
• °°°"
4
~ o
'
. . . . 6
~ -
. ,ore,
80km
,
I
IIA/ /
/
\k'~"-""~__
.
"'"', ,: ,o.o~\\M~x,co "-1
Ao,*-:---~oo.,.~ \ 110"
, I
Fig.1. Location of the Bighorn Basin in northwestern Wyoming, western United States with enlargement of the Bighorn Basin. The Clark's Fork Basin is the northwestern portion of the Bighorn Basin. The study area of Schankler (1980)centers around Elk Creek, that of B o w n centers around Fifteen Mile Creek.
143
o o n~ ~ ~ ~3
Biohorizon
/
A
Biohorizon
A/
A'
B
Biohorizon
/
/1'
I'
C
D -"
Extinctions Immigrations
,,
e
M' ~. I t' Jt
J I ' ' '~ I t .',
' t
'~', :/~,
,, I
o
1;o
-
"
_
2;o
aoo
4;0
500
6;0
70Or.
STRATIGRAPHIC LEVEL Fig.2. Three episodes of faunal turnover in early Eocene mammals, as identified by the frequency of "extinctions" (disappearances, this paper) and "immigrations" (appearances, this paper) versus stratigraphic level in the central Bighorn Basin. The Clarkforkian-Wasatchian boundary is below the lowest level represented here. Redrawn from Schankler (1980).
deposits of the Clark's Fork Basin, in the northwestern corner of the Bighorn Basin. We refer to this turnover as "Biohorizon A", a name proposed by Schankler (1980). Schankler formulated Biohorizon A on the basis of the biostratigraphy of fossil localities in the Willwood Formation of the central Bighorn Basin, specifically, in an area of about 650 km 2 along Elk, Antelope, Dorsey, and Fifteen Mile Creeks (Fig.l). Schankler (1980)recognized three episodes of faunal turnover in early Wasatchian deposits of the central Bighorn Basin (about 100 km southwest of the Clark's Fork Basin): Biohorizons A, B, and C (Fig. 2). He established these by plotting first and last appearances and a few reappearances of taxa in 600m of Wasatchian section. The three biohorizons mark short intervals with a high frequency of *~extinctions" and *'immigrations", relative to longer intervals with less faunal change. Biohorizon A comprises a large peak of extinctions and a minor peak of immigrations, As defined by Schankler (1980), Biohorizon A marks the boundary between the lower and upper units of the Haplomylus-Ectocion Range-Zone and also divides the lower and middle Graybullian sub-ages of the Wasatchian land mammal age. Gingerich (e.g., 1983a) regards Biohorizon A as the transition from Sandcouleean (early W a s a t c h i a n ) t o Graybullian (middle Wasatchian) sub-ages, In the Clark's Fork Basin, a minor faunal
turnover, involving many of the same taxa, occurs in the same relative position as Schankler's Biohorizon A. We have provisionally considered this turnover to be the same biochronological event and have called it Biohorizon A (e.g., Gingerich, 1982, 1983b, 1985; Gunnell, 1985). By correlation to the magnetic polarity time scale of Berggren et al. (1978), Biohorizon A is estimated to lie between 51.5 and 52.0 myr in the Clark's Fork Basin (Gingerich, 1982). Below, we present the biostratrigraphic data that define Biohorizon A in the Clark's Fork Basin, describe the geological context of Wasatchian fossil assemblages, and evaluate the pattern of fossil productivity through the Wasatchian in relation to the pattern of biostratigraphic change. Materials and methods Data for this study come from two sources, representing broad-scale and fine-scale records of faunal change in the Clark's Fork Basin. The first set of data consists of all mammalian specimens catalogued at the University of Michigan Museum of Paleontology (UMMP) from the Wasatchian portion of the Clark's Fork Basin. Catalogued specimens collected between 1975 and 1985 by Gingerich and coworkers form the basis for determining biostratigraphic ranges of early Wasatchian mammalian species. In this sample, specimens were
144 o r i g i n a l l y c o l l e c t e d f r o m one or m o r e h o r i z o n s w i t h i n a r e s t r i c t e d s t r a t i g r a p h i c i n t e r v a l in a limited g e o g r a p h i c a r e a , d e s i g n a t e d a locality, L o c a l i t i e s w e r e tied to local s t r a t i g r a p h i c s e c t i o n s a n d t h e i r s t r a t i g r a p h i c p o s i t i o n s recorded to t h e n e a r e s t 5 m. T h i s b r o a d - s c a l e r e c o r d s p a n s 700 m of section. T h e s e c o n d set of d a t a c o n s i s t s of m a m m a l i a n fossils c o l l e c t e d f r o m s e v e n fossiliferous h o r i z o n s t h a t s p a n a n i n t e r v a l of 40 m e n c o m p a s s i n g B i o h o r i z o n A. T h e s e fine-scale s a m p l e s w e r e c o l l e c t e d by B a d g l e y a n d c o - w o r k e r s d u r i n g t h e field seasons of 1984 a n d 1985. T h e s t r a t i g r a p h i c p l a c e m e n t of c o l l e c t i n g h o r i z o n s w a s m e a s u r e d to t h e n e a r e s t 0.5 m. T h e two d a t a sets differ in t h e i r r e p r e s e n t a t i o n o f t h e o r i g i n a l fossil a s s e m b l a g e s . I n t h e field, fossil a s s e m b l a g e s of t h e W i l l w o o d format i o n o c c u r as a s c a t t e r of d i s a r t i c u l a t e d a n d often i n c o m p l e t e e l e m e n t s , u s u a l l y w e a t h e r e d f r o m identifiable s t r a t i g r a p h i c h o r i z o n s t h a t p r o d u c e fossils a l o n g m u c h of t h e i r e x p o s u r e ( B o w n a n d K r a u s , 1981b; W i n k l e r , 1983; t h i s paper). T h e b r o a d - s c a l e , c a t a l o g u e d s a m p l e f r o m t h e W a s a t c h i a n is a s e l e c t i v e c u l l i n g of t h e s e s u r f a c e fossil a s s e m b l a g e s . J a w s a r e t h e m o s t n u m e r o u s c a t a l o g u e d items, i s o l a t e d t e e t h of a s s o r t e d t a x a a r e n e x t in a b u n d a n c e , and non-mammalian remains and mammalian postcranials are the least numerous catal o g u e d items. M a t e r i a l c o l l e c t e d b u t n o t c a t a logued f r o m a l o c a l i t y is a s s i g n e d t h e s t a t u s " m i s c e l l a n e o u s " a n d t h e n g i v e n a single catal o g u e n u m b e r for e a c h c o l l e c t i n g visit. T h e "miscellaneous" may include non-mammal m a t e r i a l (e.g., reptile, fish, plant, mollusc),
p o s t c r a n i a l s , i s o l a t e d t e e t h of w e l l - k n o w n t a x a , a n d i n c o m p l e t e j a w s of t a x a a l r e a d y a b u n d a n t l y r e p r e s e n t e d by jaws. In the finescale s t u d y of B i o h o r i z o n A, all fossil m a t e r i a l identifiable as to s k e l e t a l p a r t or t a x o n was collected. D a t a p r e s e n t e d h e r e r e p r e s e n t all g n a t h i c a n d d e n t a l m a t e r i a l of m a m m a l s ; of this m a t e r i a l , t h e m o r e c o m p l e t e j a w s a r e c a t a l o g u e d i n d i v i d u a l l y in t h e U M M P catalogue. T h u s , t h e r e is a s m a l l a m o u n t of o v e r l a p in t h e two sets of d a t a . T h e g e o l o g i c a l c o n t e x t of fossil a s s e m b l a g e s in t h e B i o h o r i z o n A s t u d y a r e a w a s a s s e s s e d f r o m a d e t a i l e d s t r a t i g r a p h i c s t u d y o f fossiliferous as well as non-fossiliferous units. Sediment a r y e n v i r o n m e n t s w e r e e v a l u a t e d on t h e basis of field d o c u m e n t a t i o n of l i t h o l o g i e s f r o m h a n d s p e c i m e n s , v e r t i c a l facies v a r i a t i o n in meas u r e d sections, a n d l a t e r a l facies v a r i a t i o n in l a t e r a l l y c o r r e l a t e d sections. On the basis of personal observations and catalogued records of t h e U M M P , we c o n c l u d e t h a t t h e g e o l o g i c a l c o n t e x t of o t h e r e a r l y W a s a t c h i a n l o c a l i t i e s in t h e C l a r k ' s F o r k B a s i n is b r o a d l y s i m i l a r to t h a t of fossiliferous h o r i z o n s in t h e B i o h o r i z o n A s t u d y area. We constructed a biostratigraphic range c h a r t for e a r l y W a s a t c h i a n m a m m a l s of t h e C l a r k ' s F o r k B a s i n b a s e d on t h e b r o a d - s c a l e c a t a l o g u e d r e c o r d s of t h e U M M P (Fig.3). T h e d o c u m e n t e d i n t e r v a l s p a n s 700 m, f r o m j u s t below the C l a r k f o r k i a n - W a s a t c h i a n boundary to 200 m b e l o w t h e t o p of t h e section. T h e q u a l i t y of t h e b i o s t r a t i g r a p h i c r e c o r d for e a c h species w a s d e t e r m i n e d f r o m t h e s a m p l e sizes for e a c h 20-m s t r a t i g r a p h i c i n t e r v a l a n d f r o m
Fig.3. Biostratigraphic range chart of early Wasatchian mammals from the Clark's Fork Basin. Mammals are listed by order, with Condylarthra divided into herbivorous and carnivorous subunits. Excluded are mammalian orders comprising predominantly small taxa (less than 100 g estimated body weight), because remains of such taxa are generally too rare to be informative of the timing of faunal change. Species names, ranges, and phyletic transitions are taken principally from the UMMP catalogue records as of 1985, with additional information from Gingerich and Gunnell (1979), Gingerich (1983b), Gunnell (1985), and Ivy (1982). The quality of the record for each species is indicated by the width of the line. Heavy lines indicate records of good quality. Lines of medium width indicate records of medium quality. Thin lines indicate "spotty" records. (See text for further explanation.) Gaps of 200 m or greater are represented as blank areas with a question mark. Single specimens with large gaps below or above are marked as an x. For species represented by lines, small horizontal ticks at the base (resp. top) indicate the first (resp. last) records in the Clark's Fork Basin. The absence of a tick at the left (resp. right) indicates that the record continues below (resp. above) the interval represented. Numbers at the left indicate the stratigraphic position (in meters) at which the species is first recorded, if lower than the Clarkforkian-Wasatchian boundary. The stippled band marks Biohorizon A.
145 C~ W a
BiohorlzonA
,°?o ,,
17o
,,? . . . . . . . . . . . . . . . . . . . . . . iiii!iiil
PRIMATES Ignocius #roybuNionus Phenocolemurproecox Phenaco/emur sirrmnsi Tin/morays groybu//iensis Niptomomys doreenae Arctodo~tomys wilsoni Arctodontom2s nuptu~
1150 14o5
?
1280 129O
?
x
?
x
q ?
x
.
'
i::iii~i~ii! . . . . . . . . . . . . . . . .
.............
M~o.o.~op. o.gu.,i~... Microsyop$ n sp Tetonoldes t~u,cu/us Talon/us stein/ Talon/us homunculus Conti., t o m , ~ Can#us ralstoni Cant/us mckennoi Coot, o ,
............. :::::::::::::::::: ::+::::F : :; ~:~:i:~:~:~ ::::::::;:::::;:
i . ,,j
.............. 2
::::::.::::
!i;iii;iiii;il
~
~L m,
•
i. ""
: :.h::i:::::
:'-
tr,~o~os
RODENTIA
"~
;;;:;~: ~
Acrltoparornys o t o v u , 1150 Acr/foporomy$ °fwoter/ 1 1 5 0 Po,omys taur~s ,980 [~Tromys cope/" Poromys n. sp Lophioporom¥$ murmu$ Microporomys n sp 1280 " Reithroporomys c~ R debequensls Knightomys n. sp
......
?
i;'!;'.i:lli; ii!;;:;i x
'
?
x
~:::
x
x ?
?
x
~:~i:: :: ::;: :
x
CONDYLARTHRA, HERBIVOROUS Phenocodus intermedius 1 2 0 0 Phenocodus vortmom 1425 Phenocodus rO~$tU$ 1440 Phenocodus brachyPCernus Eclat/on porvus ~ctocion osbornionus 1015 Hyopsodus Ioomisi Hyopsodus iotidens Hoplomylus simpson/ 1090 Hoplomylus spel'rianus Hoplomylus n. sp Aphehscu$ nt'h'dus 1080 Aphel/$cu$ n sp Aphehscus wopthens/s Aphehscus t'nsl'dlbsus
:';!:;:I:!;':I; ;:!;:;;E i ?
I
i;;;;; 34
~
x
;:::: ::~::: ::::::::;:;::::: :::':':1]" i :;i ?::::'.,
I
I
:!L!:!:!~::! "
,~4
~
:,,:,:,:,:,
i
L
PERISSODACTYLA Homogo/Ox n sp Homogolox protapiPinus
I
]::::
Hyrocofherium n .~o Hyrocotheriurn perntx
ARTIODACTYLA O~ocodexls mets~ocus Diocodexls robustus
:::::::::::
TILLODONTIA Esthonyx grongerl Esthonyx spofulorius ~sthonyx bt~ulcotu,
1370
I
: i : ::: P @:|~i-..
PANTODONTA
Coryphodon proterus Coryphodon spp
975
--~
.: : :::: :::::
'1'
CREODONTA Oxyoeno translens 1090 Oxyoen° gulo Ox¥oeno m t e r m e d i ° Pal°on/criB occldentohs
?
.....
;::::i:: i d;i:d
......... I
+
..................... ?
.....
~
~a,eomot,..e/o. . . . . . . A r f i e zele A r f l ° shoshonlensls
i
j. ,(c(~,:;
~r~io op,st~oto~o
H¥oenodonhd, n gen. A, Sp I Hyoenodonhd, n. gen. A , sp. 2 ' Prollmnocyon n $l~ Prolimnocyon otovus Hyaenodonhd, n. gen. ~, n sp. H¥oenodo~/id, n gen. C, sp. / Hyoenodonhd, n. gen. C, sp. 2 Hyoenodonh~, n. gen, C, ~ . 3 Hyeenodonl~d, n. gen C, sp. 4
...... ........ !:i~J::: i ~ " " "'x
?
x
:::::::
......... "'"
CARNIVORA Oldymlchs Didym/ctis D/ldxmlctls Vivertovus viverrovus
proteus leptomylus curt/dens °tutus bowni
975
~ 1455 1280
Vlverrovus n sp U/ntocyon rud/s 1280 Vossocyon prom~oration MfOCIS winkler/ Mioc~s deutschJ M~OClS exiguus Vulpevus c f Oustral/s
CONDYLARTHRA Thryptocodon anfiquus Dissacus praenunfius Dlssocus spp.
~ :[~:
i
I : : :
i
?
;CW,
v~v ~
I "=
~ ::: ~;
x
CARNWOROUB lO80 1150 990
?
: '~
146 the size of gaps in each species' record, mainly within the interval 1520-1755 m (below Biohorizon A). A "good" record has predominantly large samples with small gaps: these ranged from 3 to 297 catalogued specimens per 20-m interval, with gaps of less than 20m. A "medium" record has small to moderate sample sizes with more and larger gaps: these ranged from 0 to 19 catalogued specimens, with gaps of from 20 up to 130 m, but mostly less than 50 m. A "spotty" record has small sample sizes - fewer than 12 catalogued specimens per 20-m interval - - with gaps of 50-200 m. Taxa with gaps of up to 200 m (not indicated in Fig. 3) were considered to be present but not recorded (yet) by fossils. For taxa with gaps of greater than 200 m in Fig.3, the gaps were recorded for consideration as possible local extinctions and reappearances. For taxa represented by from one to a few specimens for their entire record, all individual specimens were plotted. In addition to the biostratigraphic data, we compiled the number of localities and the number of catalogued specimens for each 20-m interval, Faunal change in Fig.3 is summarized in several ways. We divided the 700-m section into 20-m increments (e.g., 1520-1535, 1540-1555, etc.). We computed species richness for each 20-m interval, a count of all taxa with records somewhere in the interval. We defined appearances and disappearances for each interval in two parallel ways, as follows. Appearances included (1) the number of first records in the Clark's Fork Basin, or (2) the number of first records plus the number of taxa t h a t "reappear" after a gap of 200 m or more. Disappearances included (1) the number of last records, or (2) the number of last records (terminal disappearances) plus the number of taxa with ensuing gaps of 200 m or more (temporary disappearances). Two taxa are known by specimens t h a t occur within only one interval. Such taxa were counted as appearing and disappearing within the same interval. A few taxa are represented by two specimens separated by gaps of 200 m or more; under the second method of tallying appearances and disappear-
ances, each specimen is counted as both an appearance and a disappearance. Speciation by anagenesis involves 33 of the 87 species (38%) in Fig.3 (Gingerich and Gunnell, 1979; Gingerich, 1980, 1983b, 1985, 1986; Gunnell, pers. comm., 1986). Changes in species cornposition resulting from anagenesis were not counted as either appearances or disappearances. Results
The Wasatchian record of the Clark's Fork Basin The biostratigraphy of mammals from early Wasatchian sediments of the Clark's Fork Basin is the basis for identifying Biohorizon A. Figure 3 documents the broad-scale stratigraphic ranges of 87 species known from early Wasatchian localities. The interval recognized as Biohorizon A occurs between 1750 and 1795m. Table I summarizes the principal changes in species composition. Three species appear near the base of the interval. Eight species disappear (terminal disappearances) within this interval; six of these eight are carnivorous mammals. Six species have gaps of greater than 200 m beginning within this interval (temporary disappearances); remains of these species occur again between 1970 and 2100 m. One plausible interpretation of these gaps is temporary extinction and later re-entry of these lineages in the Clark's Fork Basin, an interpretation favored by Schankler (1981) for the phenacodontids in the central Bighorn Basin. The quality of the record for each species involved in the turnover at Biohorizon A is also indicated in Table I. Ten of the 17 species have spotty records; the remainder have records of medium quality. In sum, disappearances outnumber appearances, a pattern that also characterizes Schankler's Biohorizon A (Fig.2). There are additional changes in faunal composition at Biohorizon A (Table II) that do not fit into the categories of Table I. Two taxa are known only at this interval. Two rodent species are represented by one specimen each
147 TABLE I
TABLE II
Principal taxa involved in the faunal turnover at Biohorizon A in the Clark's Fork Basin. Symbols indicate the quality of the biostratigraphic record as explained in the text; signifies a spotty record. The remaining taxa have records of medium quality. Based on Fig.3.
Additional taxa with biostratigraphic records especially pertinent to Biohorizon A, as explained in text. Symbols as in Table I; + signifies a record of good quality. Based on Fig.3. Taxa known only at this interval: 2
Appearances: 3
Apheliscus wapitiensis
Tetonius steini Homogalax protapirinus
Hyaenodontid, n. gen. A, sp. 2
Hyaenodontid, n. gen. C, sp. 3
Taxa represented by one specimen and large gaps above and/or below: 2
Terminal disappearances: 8
Lophioparamys murinus
Phenacodus vortmani Homogalax n. sp.
Microparamys n. sp.
Hyaenodontid, n. gen. B, n. sp. Hyaenodontid, n. gen. C, sp. 1 -
Taxa that change n a m e s through anagenetic speciation: 9
Didymictis leptomylus Viverravus bowni Viverravus n. sp. Uintacyon rudis
Arctodontomys wilsoni - A. nuptus Hyopsodus loomisi - H. latidens + Haplomylus speirianus - H. n. sp. Hyracotherium grangeri - H. n. sp. Esthonyx spatularius E. bisulcatus + Oxyaena gulo - O. intermedia
Temporary disappearances: 6 Arctodontomys lineage Paramys copei Phenacoclus robustus Phenacodus brachypternus Paleonictis occidentalis Wyolestes apheles
w i t h i n this i n t e r v a l and h a v e l a r g e gaps a b o v e a n d / o r below this interval. The r e c o r d s of these four t a x a are even more sparse t h a n the ~spotty" records. In addition, t h e r e are nine i n s t a n c e s of c h a n g e in species n a m e s w i t h i n c o n t i n u o u s l y e v o l v i n g lineages. F o u r of these species h a v e good records, two h a v e poor records, and t h r e e h a v e r e c o r d s of m e d i u m q u a l i t y up to B i o h o r i z o n A a n d s p o t t y r e c o r d s above it. C h a n g e in species n u m b e r w i t h time is depicted in Figs. 4 and 5. Since t h e r e is little i n f o r m a t i o n a b o u t a b s o l u t e time at this level of resolution, we use i n t e r v a l s of s t r a t i g r a p h i c t h i c k n e s s to i n d i c a t e r e l a t i v e time. (The results of this s t u d y do n o t depend on a n y a s s u m p t i o n s a b o u t the a m o u n t of time r e p r e s e n t e d by each interval.) F r o m Fig.3, we h a v e t a b u l a t e d standing diversity for e a c h 20-m i n t e r v a l in two ways. The first, m o r e t r a d i t i o n a l , w a y is to
Arfia shoshoniensis - A. opisthotoma Prolimnocyon n. sp. P. atavus Miacis cleutschi M. exiguus
c o u n t the n u m b e r of species w h o s e biostratig r a p h i c r a n g e s pass t h r o u g h e a c h i n t e r v a l (Fig.4). Thus, gaps do n o t c o u n t as t e m p o r a r y extinctions. In Fig.4, an early, rapid rise in species r i c h n e s s m a r k s the b o u n d a r y b e t w e e n the C l a r k f o r k i a n and W a s a t c h i a n . Species r i c h n e s s c o n t i n u e s to rise t h r o u g h the e a r l y W a s a t c h i a n , t h e n drops at B i o h o r i z o n A and declines t h r o u g h the r e m a i n i n g W a s a t c h i a n section. The final decline in species r i c h n e s s above 2100 m is an artifact, reflecting proximity to the top of the section in the C l a r k ' s F o r k Basin. In the second t a b u l a t i o n of c h a n g e in species r i c h n e s s (Fig.5), gaps g r e a t e r t h a n 200 m w i t h i n a species' r a n g e c o u n t as t e m p o r a r y d i s a p p e a r a n c e s and l a t e r r e a p p e a r a n c e s . The r e s u l t i n g p a t t e r n of c h a n g e is an e x a g g e r a t e d v e r s i o n of Fig.4, with two exceptions. The decline in species r i c h n e s s a b o v e B i o h o r i z o n A is reversed b e t w e e n 1950 a n d 1995 m and b e t w e e n 2040 a n d 2055 m, when, in this view, a n u m b e r of species ~'reappear" in the C l a r k ' s
148
48-
48-
WITHOUT GAPS
44¢/) I.,U 5 I,u
42.... - ....
40-
:
,,
\
re m Z
15:~lW a
1600
1700
1800'
1900
20'00
21'00
22'00m
Biohorizon A STRATIGRAPHIC
LEVEL,
CLARK'S
FORK BASIN
Fig.4. Species richness versus stratigraphic level, compiled for each 20-m interval, based on Fig.3. Gaps within a species range do not count: each species is considered to be present locally from its first to its last records. The highest rates of change occur at the Clarkforkian-Wasatchian boundary and at Biohorizon A. Species number declines above 2100 m as a result of the loss of exposure and of localities at the top of the section.
46
WITH GAPS >200 M
44,
42" 40. w
38
O uJ a.
36 ~
,, O ,,,
342
~ Z
30-
32-
~ -
28-
2422-
1500
Cf Wa
16'00
1700
18'00
19'00
20'00
21'00
22'00 m
Biohorizon A STRATIGRAPHIC
LEVEL, CLARK'S FORK BASIN
Fig.5. Species richness versus stratigraphic level, with gaps of 200 m or more counting as temporary disappearances, based on Fig.3. The decline in species number at Biohorizon A is greater than in Fig.4 because the temporary disappearance of six taxa are now included in the totals.
149 15
Appearances
o_,
,>,~
[~j Disappearances
o10
i~
~
I ~
+
//. a,,t£i
,I.
e,h./, i
1500 Cf Wa
_. _ •
16'00
l
/!,,. A
/
W
17'00
18'00 t I Biohorizon A
"-,
19'00
STRATIGRAPHIC
20'00
21'00
22'00 m
LEVEL
Fig.6. A p p e a r a n c e s a n d d i s a p p e a r a n c e s v e r s u s s t r a t i g r a p h i c level, w i t h g a p s of 200 m or m o r e c o u n t i n g as t e m p o r a r y d i s a p p e a r a n c e s w i t h l a t e r r e a p p e a r a n c e s . B a s e d on Fig.3.
Fork Basin. According to either Fig.4 or Fig.5, the base of the Wasatchian and Biohorizon A represent periods of greater t h a n usual rate of change in species number, The number of appearances and disappearances of species for each 20-m interval (Fig.6) best illustrates episodes of faunal turnover, With t e m p o r a r y d i s a p p e a r a n c e s a n d t h e a s s o c i ated reappearances counting as legitimate biostratigraphic events, the major faunal turnovers are clearly identified as an early episode dominated by appearances (the beginning of the Wasatchian Land Mammal Age) and a later episode dominated by disappearances (Biohorizon A). Before documenting the relationship between faunal turnover and sample size in this record, we present some results of the fine-scale study of Biohorizon A.
Biohorizon A study area In the Clark's Fork Basin, the stratigraphic interval of Biohorizon A outcrops principally in an area of about 4 km 2 within badland hills. The area is subdivided geographically into several localities; each locality spans part or all of Biohorizon A, which is about 45 m in total thickness. Seven distinct fossiliferous levels can be identified and traced laterally through much of the area. These levels are the
smallest operational units for collection of fossil material on the surface. Figure 7 indicates the stratigraphic spacing and some facies characteristics of the fossiliferous levels at one locality. Each level consists of one or more stratigraphic units; levels range in thickness from 0.3 to 3.0 m, with some lateral variability in thickness within each level. All levels occur in fine-grained floodplain deposits with pervasive pedogenic features. Intensive surface collecting of all accessible portions of the fossiliferous levels in the Biohorizon A study area yielded approximately 1500 specimens identifiable to genus or species. Species richness (oftaxa listed in Fig.3) ranged from 8 to 29 among the collecting levels, with an overall decline from older to younger levels (Fig.8). These data depict, in an enlarged view, the same trend that characterizes the catalogued data (Figs.3-5).
Sedimentary environment of vertebrate fossil assemblages The fine-grained, pedogenic facies of the fossiliferous levels in the Biohorizon A study area is the source of virtually all vertebrate remains in the Willwood Formation, both within the Clark's Fork Basin (e.g. Winkler, 1983) and elsewhere within the Bighorn Basin
150 Collecting Levels
meters
.~
45
",40
e
o.
Gravel
Cross-bedding
a-
Laminated
Z
bedding
'30 col . . . . . . . . .
~
.+
a
b
25 ~ Slickensldes
~ 20 Mottling
/~
~-
I~'J
~.
~ ),
5
~ 1111iJ
~
"~
Bivalves
~P l a-n~t
T
, • •~ y "
•
.
1 / ~ I ~
~ ~
200-
-20
u~
/
O
70
a
I~ b / c
c
(~+ d
LOW
e
i
o HIGH
STRATIGRAPHIC
LEVEL
Fig.8. Species richness (solid line) and sample size (number of jaws and teeth recorded as to species, dashed line) for the fossiliferous collecting levels of the Biohorizon A study area. The horizontal axis indicates superpositional order, not measured stratigraphic spacing.
Gastropods
.,
I
ao i I
~
=
•
~ 1 ~ x ~ a I
~
. . . .
y
d.le,
[ ' ~ R o o t s or b u r r o w s
o'~::~'
.~ ~ ~ o
C
ill o. U)
cleats
/"
"
fragments
Co.tacts: ~ Sh°,, 5
....
Gradational
.......... o,ff°se LOCAUTY 310 j/jzjy
o
szc Fig.7. A stratigraphic section through Locality 3 1 0 , spanning the interval of Biohorizon A, with fossiliferous horizons identified by letters. (All horizons but f occur at Locality 310.) The stratigraphic position of f in localities nearby is indicated; at Locality 310, level f and adjacent floodplain sediments are cut out by a thick erosional sandstone,
(e.g. Bown, 1979; B o w n and K r a u s , 1981a, b). This facies r e p r e s e n t s the floodplain environm e n t t h a t was a l t e r n a t e l y i n u n d a t e d and a e r a t e d e a c h year. Thus, differences in f a u n a l c o m p o s i t i o n t h r o u g h time c a n n o t be a t t r i b u t e d to different t a p h o n o m i c processes a s s o c i a t e d with b r o a d l y different s e d i m e n t a r y environments. But, b o t h in the B i o h o r i z o n A s t u d y a r e a and e l s e w h e r e in the Willwood Formation, t h e r e is c o n s i d e r a b l e v a r i a b i l i t y w i t h i n the pedogenic facies in t h e r e l a t i v e p r o p o r t i o n s
of silt, clay, and sand; in color, t e x t u r e , and stratification; and in the sizes and kinds of nodules and m o t t l i n g features. M o s t of these facies c h a r a c t e r i s t i c s reflect v a r i a t i o n in postd e p o s i t i o n a l h i s t o r y of the sediments, r e l a t e d to differences in m i c r o e n v i r o n m e n t s and correlated soil-forming processes. Possibly, some of the c h a n g e in f a u n a l composition is r e l a t e d to c h a n g e s in paleosol c h a r a c t e r i s t i c s - - a hypothesis t h a t has y e t to be fully explored in the Clark's F o r k Basin.
Speciesrichness and sample size T h e biological significance of c h a n g e s in species richness can be e v a l u a t e d o n l y a f t e r sample sizes are t a k e n into a c c o u n t . F i g u r e 8 i l l u s t r a t e s h o w sample size ( n u m b e r of specimens) and species richness c o v a r y for the B i o h o r i z o n A s t u d y area, and Fig.9 i l l u s t r a t e s this c o v a r i a t i o n for the e a r l y W a s a t c h i a n as a whole. (Figs.8 and 9 are based on p r e s e n c e or a b s e n c e in e a c h i n t e r v a l o f t a x a listed in Fig.3.) In t h e B i o h o r i z o n A s t u d y area, sample sizes exhibit a declining t r e n d from older to y o u n g e r levels (Fig.8), a l t h o u g h the d e c r e a s e in sample size is n o t as c o n s i s t e n t as the decline in species number. In o r d e r to i n v e s t i g a t e the r e l a t i o n s h i p b e t w e e n sample size and species n u m b e r in the
151 co 40- 9 LU ~
o
J
"
• Species •.4--- Specimens
-
0 LU
-1200 .
co 30 Q LU O O < 2C ~--
-1000 " "800
.6
-600
rr '"
1E
: :
•
:: ~:
. ~
=
:~ Z
:
o
~ .:".
" .. "
.
b .:'
ii
-400
_
? 6
"
¢r U.I
~
".d
" :.
i
/
'"q"
Cf
1600 Wa
<
O
0 1500
C3 LU (D -J
_
O OU-
(D zLU n LU a.
17'00
18'00
-200
'~
~E ,.n
"'~"o
o.
19'00
.e.-e---
20'00
21'00
z
2200 meters
I I Biohorizon
A STRATIGRAPHIC LEVEL
Fig.9. N u m b e r o f s p e c i e s ( s o l i d l i n e ) a n d s a m p l e size ( d o t t e d l i n e ) v e r s u s s t r a t i g r a p h i c level. B o t h w e r e c o m p i l e d for e a c h i n t e r v a l f r o m t h e U M M P c a t a l o g u e as o f 1985.
20-
m
macroscopic record, it was necessary to tally the number of recorded species for each 20-m interval. Figure 3 does not contain the requisite information, since taxa are represented there by biostratigraphic ranges but are not necessarily recorded in every 20-m interval. (Only taxa with good records are found in every 20-m interval of their biostratigraphic ranges between 1520 and 1755 m; above this level, even the good records show some gaps of 20 m or more.) Figure 9 depicts, interval by interval, the numbers of specimens and of species recorded for the early Wasatchian. Clearly, peaks in species richness coincide with peaks in sample size, and the drop in species richness at Biohorizon A coincides with a large drop in sample size. The correlation between root sample size and species number (Fig.10) is high for the combined data sets (r = 0.95). (The square root is the appropriate transform for linearizing the scatterplot when the underlying model corresponds to Poisson sampling (Snedecor and Cochran, 1980).) The data from the Biohorizon A study area by themselves exhibit a correlation coefficient of r = 0.94 (n = 9); the U M M P catalogued data alone have a correlation coefficient of r = 0.95 (n = 29), indicating a similarly tight relationship between sample size and species
number, despite differences in the methods of collection and documentation. Discussion The high correlation between sample size and species number in Fig.10 indicates that we have no basis for rejecting the following null hypothesis: observed changes in species richness are explained satisfactorily by changes in sample size alone. This relationship is further illustrated in the correspondence between biostratigraphic events (appearances, reappearances, and terminal and temporary disappearances) and sample size (Fig.ll). All five peaks in appearances between 1500 and 1995 m occur in intervals with peaks in specimen number; two minor peaks in specimen number (1800-1815 m, 1840-1855 m) are not associated with appearances. The only reappearances (following temporary disappearances) in Fig.ll occur between 1960 and 1975 m, likewise a local peak in sample size. All of the Wasatchian lineages with "good" records appear in the first peak. Taxa that appear later have records of "medium" or "spotty" quality. Among the species that appear at Biohorizon A, the three principal appearances (Table I) occur within
152 40'
r =0.95
o
30-
ILl
•
0
.,
•
g
O. ~ u.
o
o
0
•
20-
••
n"
I,LI
~e
"~ Z
•
O0
0
O
10"
• UMMP catalogue o Biohorizon A study area
o •
0
0
2'5
160
2~5 NUMBER
460
6~5
960
'
OF SPECIMENS
( s q u a r e root s c a l e )
Fig.10. Scatterplot of sample size versus number of species, based on data in Figs.8 and 9. Sample size is represented by the s q u a r e r o o t of s p e c i m e n n u m b e r in o r d e r to m a k e t h e d i s t r i b u t i o n o f p o i n t s m o r e n e a r l y linear. T h e r e p o r t e d c o r r e l a t i o n is
between species richness and this transform of sample size.
I-
9..
z,, >m to
£
'o Appearances 15
~
-1200
-100o
I.U
'"
O.
0
5=
Specimens
~ :
10
:' ..o
~
i
."
"
~ ~ i
;
A
!!
9""
/i',
.
s , i,, ~~
~.
::3 Z
].d
,
_ t
%"
1500 Cf Wa
I
i!l
ii. ,=,, ~
,1\
_ _11~_\
xe-e-q;
16'00
.r..f,
h',k
/ , .'1 \ ! ~,, "~ ~ '1\
%\/1
\/
"b'
I
~
17'00
\:u". :1 -~"~'..'1 ~
-800
!
/:.~
~ .:
5 '="
.
Disappearances
.
18'00
~
....,.
/,\
I ,
~!
19,00
2()00
-
e
We"
<
ii /1".',\ D'.2"A // :~..',r'~
I<.I
,oo
200
..!lw "~ l
,~i'o,. \1 ~ "o. "~ l
'x:b~4mm~,e
21,00
..~
500 o,
I
il/ 1
~"9.. • ~/ "o.,.,,,
o
to
,,,
=
~i =
Z
2200 meters
e I Biohorizon A STRATIGRAPHIC LEVEL
Fig.ll. The number of appearances and disappearances (from Fig.6), with gaps of 200 m or greater counting as temporary disappearances and later reappearances, plotted with sample size (from Fig.9). Data points representing appearances or disappearances are plotted at the interval boundaries. Points representing number of catalogued specimens are plotted in the middle of each interval.
intervals of very high sample sizes. Four additional species are rare or unknown outside this interval (Table If). Thus, the timing of appearances closely tracks peaks in fossil productivity, while the number and taxonomic identities of appearances depend on the prior history of appearances,
Disappearances generally coincide with drops in sample size. The most pronounced example occurs at Biohorizon A. Fourteen speciesdisappear (six? temporarily) when sample size drops from 1027 specimens to 20 specimens in successive 20-m intervals.Nine of the 14 species have poor records even when
153 sample sizes are relatively high (below 1780 m). The six reappearing taxa occur again in intervals with moderate peaks of sample size. The correlation between disappearances and drops in sample size at and above Biohorizon A indicates that some temporary disappearances are to be expected on the basis of changes in sample size alone. (Whether the disappearance of a particular species is statistically significant can be determined under the assumption of Poisson sampling, as Schankler (1981) demonstrates for phenacodontids.) These reappearances are an example of the "Lazarus effect" (Jablonski, 1986) in a fluvial environment, The anagenetic speciations that occur at Biohorizon A in the Clark's Fork Basin (Table II) do not in themselves offer persuasive evidence for a significant faunal change, since the separation between ancestral and descendant species was sometimes determined on the basis of a large gap in the record or by noting the fact of the faunal turnover itself. These speciations are, in any case, arbitrary. All nine lineages t h a t change species name at Biohorizon A exhibit increase in size, as indicated by increased cheek-tooth area. But only in the case of Hyopsodus is size increase a distinct reversal of the trend (size decrease) j u s t b e l o w Biohorizon A (Fig.2, Gingerich, 1985). If new taxonomic interpretations were to view any of these changes as cladistic speciation events, the number of appearances and disappearances recorded at Biohorizon A would be altered, but not the relationship between sample size and species number that we find in these data. What is the real pattern of appearances and disappearances of species? As the record stands, new taxa of diverse taxonomic and ecological affinities appear early in the Wasatchian. Disappearances occur in pulses and involve species known from the Clarkforkian, along with some of the Wasatchian immigrants (Fig.3). In general, appearances coincide with increases in sample size, and disappearances coincide with decreases in sample size. Common taxa appear early. When sample sizes fall off and later do not rise above moderate values, a number of rare taxa disappear (Fig.ll). Some
disappearances occur even as sample sizes are increasing (e.g., between 1700 and 1715m). Clearly, substantial changes in faunal composition were occurring through the early Wasatchian record. But since almost 90°//o (r 2 = 0.88) of the apparent pattern can be attributed to variation in sample size (Fig.10), the real pattern of faunal change is little constrained by data. The apparent pattern is consistent with a number of hypotheses of real faunal change. For example, all appearances (29 species) recorded between 1520 and 1815 m in Fig.3 could have occurred at the beginning of the Wasatchian, to be detected in pulses as sample sizes rose and fell. According to this hypothesis, had the sample size between 1520 and 1535 m been increased to 1000, many of the 29 taxa would have been found, and some new ones as well (Koch, 1987). In the real data, this hypothesis suggests t h a t the appearances are unveiled in pulses because progressively larger sample sizes are required for the rarer taxa. Another hypothesis is t h a t actual species diversity did not change between 1780 and 1795 m, but rather that the apparent decrease is due solely to the decrease in sample size. Had our samples been enlarged to some 1000 specimens between 1800 and 1815 m, many of the "disappearing" taxa would have been present. Both these expectations derive from the observed relationship between sample size and species number (Fig.10). This relationship between sample size and species richness is widely known in ecology and paleontology (e.g., May, 1975; Raup, 1975; Tipper, 1979; Koch, 1987). For a specified geographic area and time - - that is, for a particular c o m m u n i t y - larger sample sizes of individuals record ever increasing species richness. The relationship between sample size and species number is typically non-linear; the ratio of species number to sample size is generally greater in small samples than in large ones. If the total number of species present is fixed by assumption, there results the familiar asymptote in species-versus-individuals and rarefaction curves. The ecological
154
basis for this pattern is the shape of speciesabundance distributions in nature. Communities have a few common species and many rare ones (Elton, 1927; May, 1975; Koch, 1987; and many others). But these distributions vary in detail, making it difficult to apply a universal standard of correction for sample-size effects in samples from modern and ancient communities.
Fossil productivity The striking correlation between sample size and species number raises the important issue of the factors controlling fossil productivity in a stratigraphic section. We suggest three possibilities: area of exposure, taphonomic changes, and ecological changes, (1) Fossil productivity may be proportional to the area of exposure for any stratigraphic interval. This factor is certainly critical within the Biohorizon A study area: the pattern of badland topography results in a decrease of available exposure up-section. One way to evaluate this hypothesis would be to measure (e.g., by digitization of a base map) the exposure area of selected stratigraphic intervals, then correlate exposure area with the number of catalogued specimens. If the correlation were high, then there would be no basis for inferring a real faunal turnover within the interval currently recognized as Biohorizon A. (2) Fossil productivity may be controlled by basin-wide changes in sedimentary or climatic processes that affect the preservation of mammalian remains: for instance, change in the rate of sediment accumulation, change in seasonality, or change in the duration of annual floods. In this case, similar changes in fossil productivity should have occurred in the central Bighorn Basin, since it was part of the same climatic and depositional system during the Eocene. Also, such changes should have left some record within the sediments themselves, such as long-term changes through the Wasatchian in the maturity of soils. If this factor were important, then the fact of a faunal turnover would be questionable, although
there would be a significant environmental change with far-reaching sedimentological taphonomic effects. It is possible, however, that such environmental changes could cause a faunal turnover as well. (3) Changes in fossil productivity might be caused by ecological changes in the fauna itself that have taphonomic ramifications, such as changes in the intensity of scavenging. In the case of Biohorizon A, this hypothesis would be most plausible if taxa that appear when fossil productivity is falling can be interpreted reasonably as having caused the drop in productivity (e.g., the appearance of bone-crushing carnivorous species). If this ecological hypothesis were operative, then similar faunal changes elsewhere in the Bighorn Basin and in other basins should be correlated with similar changes in fossil productivity, and no correlated changes in the sediments themselves would be expected. We cannot conclusively reject any of these factors at present, although there is information that bears on each of them. With regard to the factor of exposure area, the exposures in the Clark's Fork Basin do become restricted and often steeper up-section, resulting in less accessible outcrop area (but possibly more rapid renewal of fossils by erosion) for higher levels. This general observation does not, however, explain the modest increases in sample size above 1800 m. With regard to the taphonomic factor, the Biohorizon A study area is capped by thick (over 30 m), multistoried sandstones (Fig.7), which extend laterally through the Clark's Fork Basin. These sandstones probably account for the initial drop in fossil productivity around 1800 m, since vertebrate remains are rarely found in this facies in the Willwood Formation. Also, these sandstones probably signify a period of widespread erosion, as is associated with the '~boundary sandstone" at the base of the Wasatchian around 1500-1515 m (Kraus, 1980). But above the interval of sandstone there is a return to predominantly fine-grained silts and silty clays, the appropriate facies for vertebrate
155
remains. Thus, from the standpoint of sedimentary environments, the sample size should be able to rise (as it does between 1840 and 1815 m) to values typical of Biohorizon A and below, The pattern of fossil productivity in the central Bighorn Basin bears some similarity to that of the Clark's Fork Basin. Schankler (pers. comm., 1986) reports t h a t at localities around Dorsey and Elk Creeks (Fig.l), sample sizes fall across Biohorizon A and remain moderately low until the level of Biohorizon B, when sample sizes increase markedly. The decrease at Biohorizon A is not of the same magnitude as in the Clark's Fork Basin. Bown (pers. comm., 1986) reports that there is a decrease in sample size above Biohorizon A based on localities around Fifteen Mile Creek. With regard to the ecological factor, the faunal turnover at Biohorizon A involves the appearance of one carnivorous species and the loss of six others. This pattern neither suggests a mechanism by which faunal change could cause the change in fossil productivity, nor quite rejects the possibility either. Conclusion This study of the early Wasatchian record in the Clark's Fork Basin indicates t h a t changes in sample size through a stratigraphic record may influence the perception of faunal turnover. Although clearly a major change in faunal composition occurred early in the Wasatchian, the peaks and valleys in sample size conceal the timing and rate of this change, Appearances are correlated with peaks in sample size, and disappearances with valleys, We cannot rule out the possibility that the observed pulses of appearances and disappearances, i.e., the faunal turnovers, are an artifact of widely varying sample sizes through the section. Sample size is influenced by a number of ecological and taphonomic factors, including original population sizes and sediment accumulation rates, The role of sampling in this biostratigraphic record was demonstrated only when sample size was documented along with biostrati-
graphic change. Even though sample-size effects are present in the biostratigraphic range chart (Fig.3), in the graphs of change in species richness through time (Figs.4 and 5), and in the record of appearances and disappearances through time (Fig.6), the effects are not immediately evident, since none of these figures document sample size per se. But when changes over time in sample size are plotted together with species richness (Figs.8 and 9) or appearances and disappearances (Fig.ll), the nature of our inferential problem becomes clear. Sample size and species richness should be highly correlated in modern and ancient communities. The confounding situation occurs when large fluctuations in sample size occur over time. In such situations, the record of appearances, disappearances, and species richness is not a reliable indicator of actual faunal change. Species number and composition vary widely with sample size because so many species are rare. In the record of Biohorizon A in the Clark's Fork Basin, many of the disappearing taxa are carnivorous species. These would have been rare in the original community because of their position in the trophic hierarchy. The herbivorous taxa that disappear have medium to spotty records - - low to moderate numbers from stratigraphic levels with moderate to large sample sizes. Whether for taphonomic or ecological reasons, these species are never abundant in the fossil samples. Common species change principally by phyletic evolution. Thus, it is difficult to make a strong case for a faunal turnover in the interval within which Biohorizon A has been located in the Clark's Fork Basin. Yet we refrain from concluding that Biohorizon A does not exist in the Clark's Fork Basin or anywhere else. The primary historical changes in the data presented here are a rise and fall in fossil productivity. These changes may themselves have chronological and environmental significance as widespread events within the basin of sedimentation. Biohorizon A could be identified more clearly by the collection of additional data, notably above
156 1780 m, so as to e n l a r g e s a m p l e sizes to a b o u t the v a l u e s of s a m p l e s b e t w e e n 1700 a n d 1775 m. T h e n , t h e r e c o r d w o u l d n o t be free of samplesize effects, b u t t h o s e effects w o u l d a t l e a s t be m o r e c o m p a r a b l e f r o m i n t e r v a l to i n t e r v a l , Episodes of a p p e a r a n c e a n d d i s a p p e a r a n c e of species are, in principle, i m p o r t a n t ecological a n d e v o l u t i o n a r y events, w i t h h i s t o r i c a l signific a n c e for c o r r e l a t i o n a n d for p a l e o e n v i r o n m e n tal a n a l y s i s as well. I f o t h e r t a p h o n o m i c factors a r e equal (e.g., if t h e r e is no s u b s t a n t i a l c h a n g e in the facies of fossil a s s e m b l a g e s b e i n g compared), t h e n a p p e a r a n c e s a n d / o r d i s a p p e a r a n c e s of a b u n d a n t t a x a a r e less susceptible to t h e vicissitudes of s a m p l i n g processes t h a n a r e a p p e a r a n c e s or d i s a p p e a r a n c e s of r a r e taxa. U l t i m a t e l y , the identities of the a p p e a r i n g a n d d i s a p p e a r i n g t a x a a n d t h e i r possible ecological a n d c o - e v o l u t i o n a r y r e l a t i o n s h i p s hold t h e k e y to the m e c h a n i s m a n d significance of f a u n a l
turnovers.
Acknowledgments We
w i s h to t h a n k the field c r e w s that w o r k e d
in t h e C l a r k ' s F o r k B a s i n d u r i n g t h e s u m m e r s of 1975-1985 a n d helped to collect the d a t a presented here. W e t h a n k especially B a r a B a d g l e y a n d J o n P o l l a c k for t h e i r collections in the
Biohorizon A study area and Gregg Gunnell b o t h for collecting fossils and identifying a s u b s t a n t i a l p o r t i o n of the collection f r o m the Biohorizon A s t u d y area. F r e d B o o k s t e i n provided i m p o r t a n t counsel in statistics a n d imp r o v e d t h e m a n u s c r i p t . We t h a n k K a y Behrensmeyer, David Schankler, and Gerald S m i t h for c o n s t r u c t i v e discussions of b o t h the g e n e r a l subject a n d the m a n u s c r i p t . K a r e n Klitz p r e p a r e d the illustrations. This w o r k w a s s u p p o r t e d in p a r t by N S F g r a n t E A R 83-05931 to CB a n d by N S F g r a n t BSR 86-07841 to PDG.
References Berggren, W. A., McKenna, M. C., Hardenbol, J. and Obradovich, J. D., 1978. Revised Paleogene polarity time scale. J. Geol., 86: 67-81. Bown, T. M., 1979. Geology and mammalian paleontology of the Sand Creek Facies, lower Willwood Formation
(Lower Eocene), Washakie County, Wyoming. Mere. Geol. Surv. Wyo., 2, 151 pp. Bown, T. M. and Kraus, M. J., 1981a. Lower Eocene alluvial paleosols (Willwood Formation, northwest Wyoruing, U.S.A.) and their significance for paleoecology, paleoclimatology, and basin analysis. Palaeogeogr., Palaeoclimatol., Palaeoecol., 34: 1-30. Bown, T. M. and Kraus, M. J., 1981b. Vertebrate fossilbearing paleosol units (Willwood Formation, Lower Eocene, northwest Wyoming, U.S.A.): Implications for taphonomy, biostratigraphy, and assemblage analysis. Palaeogeogr., Palaeoclimatol., Palaeoecol., 34: 31-56. Elton, C., 1927. Animal Ecology. Methuen, London, 207 pp. Gingerich, P. D., 1980. Evolutionary patterns in early Cenozoic mammals. Annu. Rev. Earth Planet. Sci., 8: 407-424. Gingerich, P. D., 1982. Time resolution in mammalian evolution: sampling, lineages, and faunal turnover. In: Proc. Third North Am. Paleontol. Conv., Montreal, 1982, 1: 205-210. Gingerich, P. D., 1983a. Paleocene-Eocene faunal zones and a preliminary analysis of Laramide structural deformation in the Clark's Fork Basin, Wyoming. In: Wyo. Geol. Assoc. 34th Annu. Field Conf., 1983, Guidebook, pp. 185-195. Gingerich, P. D., 1983b. Systematics of early Eocene Miacidae (Mammalia, Carnivora) in the Clark's Fork Basin, Wyoming. Univ. Mich. Contrib. Paleontol., 26: 197-225. Gingerich, P. D., 1985. Species in the fossil record: concepts, trends, and transition. Paleobiology, 11: 27-41.
Gingerich, P. D., 1986. Early Eocene C a n t i u s torresi - oldest primate of modern aspect from North America. Nature, 320: 319-321. Gingerich, P. D. and Gunnell, G. F., 1979. Systematics and evolution of the genus E s t h o n y x (Mammalia, Tillodontic) in the early Eocene of North America. Univ. Mich. Contrib. Paleontol., 25(7): 125-153. Gunnell, G. F., 1985. Systematics of early Eocene Microsyopinae (Mammalia, Primates) in the Clark's Fork Basin, Wyoming. Univ. Mich. Contrib. Paleontol., 27(2): 51-71. Ivy, L. D., 1982. Systematics and biostratigraphy of the earliest North American Rodentia (Mammalia), latest Paleocene and early Eocene of the Clark's Fork Basin, Wyoming. Thesis, Univ. Michigan (unpublished). Jablonski, D., 1986. Causes and consequences of mass extinctions: a comparative approach. In: D. K. Elliott (Editor), Dynamics of Extinction. Mus. North. Ariz. Press, Flagstaff, Ariz., pp. 183-229. Kitchell, J. A. and Carr, T. R., 1985. Nonequilibrium model of diversification: faunal turnover dynamics. In: J. W. Valentine (Editor), Phanerozoic Diversity Patterns: Profiles in Macroevolution. Princeton Univ. Press, Princeton, N.J., pp. 277-309. Koch, C. F., 1987. Prediction of sample size effects on the measured temporal and geographic distribution patterns of species. Paleobiology, 13: 100-107. Kraus, M. J., 1980. Genesis of a fluvial sheet sandstone, Willwood Formation, northwest Wyoming. In: P. D.
157 Gingerich (Editor), Early Cenozoic Paleontology and Stratigraphy of the Bighorn Basin, Wyoming. Univ. Mich. Pap. Paleontol., 24: 87-94. May, R., 1975. Patterns of species abundance and diversity. In: M. L. Cody and J. M. Diamond (Editors), Ecology and Evolution of Communities. Belknap Press (Harvard Univ.), Cambridge, Mass, pp. 81-120. Olson, E. C., 1980. Taphonomy: its history and role in community evolution. In: A. K. Behrensmeyer and A . P . Hill (Editors), Fossils in the Making. Univ. Chicago Press, Chicago, Ill., pp. 5 19. Raup, D. M., 1975. Taxonomic diversity estimation using rarefaction. Paleobiology, 1: 333-342. Schankler, D. M., 1980. Faunal zonation of the Willwood Formation in the Central Bighorn Basin, Wyoming. In: P. D. Gingerich (Editor), Early Cenozoic Paleontology and Stratigraphy of the Bighorn Basin, Wyoming. Univ. Mich. Pap. Paleontol., 24: 99-110. Schankler, D. M., 1981. Local extinction and ecological reentry of early Eocene mammals. Nature, 293: 135-138.
Sepkoski, J. J., Jr., 1981. A factor analytic description of the Phanerozoic marine fossil record. Paleobiology, 7: 36-53. Snedecor, G. W. and Cochran, W. G., 1980. Statistical Methods. Iowa State Univ. Press, Ames, Iowa, 507 pp. (7th ed.). Tipper, J. C., 1979. Rarefaction and rarefiction: the use and abuse of a method in paleoecology. Paleobiology, 5: 423 434. Vrba, E. S., 1985. Environment and evolution: alternative causes of the temporal distribution of evolutionary events. S. Afr. J. Sci., 81: 229-236. Webb, S. D., 1984. On two kinds of rapid faunal turnover. In: W. A. Berggren and J. A. Van Couvering (Editors), Catastrophes and Earth History. Princeton Univ. Press, Princeton, N.J., pp. 417-436. Winkler, D. A., 1983. Paleoecology of an early Eocene mammalian fauna from paleosols in the Clark's Fork Basin, northwestern Wyoming (U.S.A.). Palaeogeogr., Palaeoclimatol., Palaeoecol., 43:261 298.