CHARTING THE LATE CRETACEOUS SEAS: MOSASAUR RICHNESS AND MORPHOLOGICAL DIVERSIFICATION

Joural of Vertebrate Paleotology 29(2):409 416, Jue 2009 # 2009 by the Society of Vertebrate Paleotology ARTICLE CHARTING THE LATE CRETACEOUS SEAS: MOSASAUR RICHNESS AND MORPHOLOGICAL DIVERSIFICATION MARCUS R. ROSS *,1 1 Departmet of Geoscieces, Uiversity of Rhode Islad, Kigsto, 02881 U.S.A. ABSTRACT Abudat, readily idetifiable, ad biostratigraphically resolved specimes make mosasaurs ideal cadidates to test fluxes i geeric richess ad morphological treds amog marie vertebrates durig the late Cretaceous. More tha 1800 globally distributed mosasaur specimes are allocated to fiftee substage-level stratigraphically correlated assemblages. These data are quatitatively aalyzed to illumiate treds withi the group. Followig their first appearace i the Ceomaia, mosasaurs experieced a sigificat radiatio i the Coiacia ad Satoia. Richess levels cotiued icreasig ito the Maastrichtia while mosasaurs icremetally exploited ew predatory iches recogized via accumulatio of ovel detal morphologies. Their extictio at the ed of the Cretaceous occurred at the zeith of mosasaur morphological ad ecological diversity. INTRODUCTION Mosasaurs were geerally large, globally distributed apical marie predators kow exclusively from late Cretaceous marie shelf ad epicotietal sea deposits. Sice the discovery of the first mosasaur i 1766 (Mulder, 2003), thousads of specimes have bee recovered from Europe, North ad South America, Africa, Atarctica, New Zealad, ad Japa. The wealth of material, i cojuctio with a high level of biostratigraphic refiemet, make mosasaurs ideal cadidates for tracig the cotours of richess ad morphologic treds of these marie squamates throughout the Late Cretaceous. This study demostrates that followig their iitial origi ad radiatio, mosasaurs cotiued a growth i geeric richess through at least the early Maastrichtia, followed by overall stable richess levels util their demise at the ed of the Maastrichtia. Throughout this time they diversified i predatory behaviors, as deduced from tooth morphology. Two aalyses were coducted. First, I quatitatively aalyzed fluxes i mosasaur geeric richess durig the middle Coiacia through Maastrichtia to determie if richess variatios are real or samplig artifacts. Secod, I tabulated the distributio of morphological ovelties (represeted by tooth guilds) amog mosasaurs over time to determie the pace of acquisitio ad distributio of ovel adaptive traits. * Curret address: Departmet of Biology/Chemistry, Liberty Uiversity, Lychburg, Virgiia 24502 U.S.A., mross@liberty.edu METHODS: ORGANIZING THE MOSASAUR RECORD Biostratigraphic Framework Existig high-resolutio marie biostratigraphic systems (e.g., ammoites, belemites, calcareous aofossils) offer tremedous precisio i datig vertebrate fossils. However, due to the comparative rarity of vertebrate remais i ay particular microfossil or ivertebrate biozoe, ad the potetial for o-recovery of a idex taxo due to various geographical, ecological, or huma factors, idividual biozoes are ieffective samplig widows. As a result, most aalyses of marie vertebrate diversity cosist either of local/regioal studies with high temporal resolutio (e.g., Sheldo, 1996; Mulder et al., 1998) or studies o far broader scales with cocomitatly lower temporal resolutio (e.g., Russell, 1993; Hirayama, 1997; Kriwet ad Beto, 2004). Stratigraphically correlated assemblages (SCAs) are temporally successive samplig bis defied by groupig several ammoite zoes ad/or calcareous aofossil biozoes, icorporatig sequece stratigraphy, ad, where possible, umerical ages (Fig. 1; see appedix for defiitios). SCAs are desiged to provide a broad framework for compilig late Cretaceous marie shelf ad epicotetal sea deposits. The fiftee SCAs employed here rage from 1.2 to 3.0 myr each, with a average duratio of approximately 2.3 myr (timescale from Gradstei et al., 2004). Although ot uiform i duratio, SCAs provide a practical framework for compilig ad orgaizig marie fossils (both vertebrate ad ivertebrate) from far-ragig localities while retaiig relatively high temporal resolutio. While this study aalyzes the mosasaur record, SCAs ca be used as a biig system for virtually ay set of late Cretaceous marie orgaisms. Database Costructio The database used here cosists of 1805 mosasaur specimes obtaied from museum collectios ad published reports. Specimes are recorded from North ad South America, Europe, Africa, Atarctica, Asia, Australia, ad New Zealad, ad each is idetified to geus ad cofidetly assiged to oe of the fiftee SCAs (Table 1). The occurreces compiled i the database provide biostratigraphic rages for twety six mosasaur geera, ad from this I recostructed the stadig richess of mosasaur geera per SCA durig the Late Cretaceous (Fig. 2). The solid lie represets the miimum stadig geeric richess per SCA, while the dashed lie idicates maximum richess whe taxa with uresolved occurreces i SCAs are icorporated. MOSASAUR RICHNESS THROUGH TIME Early mosasaurs retaiig plesiopedal (lad-capable) limb arragemet appear i the early Ceomaia ad middle Turoia 409

410 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 29, NO. 2, 2009 FIGURE 1. Graphical represetatio of SCAs; see appedix for defiitios. SCA boudary ages give o right; ages with arrows deote curretly accepted date for stage boudary. Late Cretaceous time scale, sea level curve, ammoite ad aofossil zoes costructed usig TS-Creator versio 2.0.

ROSS MOSASAUR RICHNESS AND DIVERSIFICATION 411 TABLE 1. Total umber of mosasaur specimes recorded per SCA. Geus A B C D E F G H I J K L M N O Cariodes 5 11 Clidastes 1 2 5 624 17 Dallasaurus 1 2 Ecteosaurus 1 1 1 Globides 5 7 3 1 7 27 2 Goroyosaurus 3??? Haasiasaurus 1 1 Haiosaurus 1 6 1 2 Halisaurus 2 4 113 1 5 6 Igdamaosaurus 4??? Kourisodo 1 1 1 1 2 Lakumasaurus 1 1 Liodo 1??? 11 Moaosaurus 1?? Mosasaurus 1 3 7 10 4 42 29 135 Platecarpus 1 1 21 23 60 103 Plesiotylosaurus 1 1 Plioplatecarpus 1 3 3 8 6 14 2 47 Plotosaurus 2 20 2 Progathodo 1,4 5 1 2 8 5 6 Russellosaurus 1 1 Selmasaurus 1 1 Taiwhasaurus 1 1 Tethysaurus 1 1 Tylosaurus 1 26 37 269 26 1 1 Yaguarasaurus 1 4 Total Specimes 1 1 8 49 71 1087 155 11 21 15 77 89 220 Superscripts idicate feedig guild assigmet followig Massare (1987, 1997; see text for discussio). 1, Cut; 2, Pierce II/Geeral; 3, Smash; 4, Cruch; 5, Crush;?, taxo with uresolved SCA assigmet. (Bardet et al., 2003; Bell ad Polcy, 2005; Polcy ad Bell 2005). Hydropedal (fully marie-adapted flipper) mosasaurs first appear i the middle Turoia (Ligham-Soliar 1994). Based o the stadig richess curve (Fig. 2), hydropedal mosasaurs diversified steadily to the Campaia, experiecig peak richess levels durig the early ad late Campaia (SCAs H ad L; Fig. 2). Mosasaur richess appears geerally stable throughout the Campaia, ad falls slightly durig the Maastrichtia. This may idicate that mosasaur richess durig the latest Maastrichtia (8 geera) was lower tha the early or late Campaia (10 geera), or alteratively, it may be a artifact of ueve samplig itesity (e.g., Fastovsky et al., 2005). Aalytical Techiques To determie if samplig artifacts are ifluecig the perceptio of mosasaur richess, I follow previous studies addressig vertebrate richess fluxes over time (e.g., Sheeha et al., 1991; Pearso et al., 2002; Fastovsky et al., 2004) i applyig rarefactio to quatify richess through time. Rarefactio (Saders, 1968; Hulbert, 1971; Simberloff, 1972) estimates the predicted umber of taxa, E(Ŝ), from a dataset of a umber of idividuals (N), whe oly a portio of that dataset is sampled ( N), eablig compariso of datasets of uequal sizes. Tipper (1979) suggests that for /N>0.1, the rarefactio equatio give by Hulbert (1971) is best selected: 2 3 N N i Eð^S Þ¼ Xs 61 7 4 N 5 i¼1 Variace is calculated idepedetly usig the equatio of Heck et al. (1975): varð^s Þ¼ðNÞ 1 6 4 2 X s i¼1 þ2 Xs 1 X s i¼1 j¼iþ1 2 3 N N i ðn N i Þ61 7 4 N 5 3 2 3 ðn N i Þ ðn i N j Þ 4ðN N i N i Þ 57 ðnþ 5 Stadard deviatio for the rarefied geeric richess, s(ŝ), was calculated as the square root of the variace, var(ŝ). Mosasaur richess per SCA was calculated by rarefyig selected samples to a particular value for stadardized compariso. I this case, datasets with larger specime couts were rarefied to a dataset with a smaller specime cout, ad their geeric richesses compared. The aalysis preseted here, of surface ad quarry-collected Late Cretaceous macrofossils from a sigle family of pelagic marie squamates satisfies the criteria of Tipper (1979) for the applicatio of rarefactio to fossil databases. Seve SCAs: F, G, H, I, M, N, ad O, were selected for compariso o the basis of adequate sample size ( 30 per SCA; Table 1). The data from these SCAs were rarefied to = 49, the smallest umber of specimes amog the SCAs (SCA-F). For 30, 95% cofidece itervals are calculated as 1.96 s from that sample s rarefied geeric richess. Results Results of the rarefactio aalysis are preseted i Table 2 ad Figure 3. All rarefied SCA richess values ad cofidece itervals are compared to those of SCA-F. All SCAs ad their 95% cofidece itervals lie above the richess of SCA-F (three geera) whe rarefied to = 49. Utilizig miimum umbers of

412 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 29, NO. 2, 2009 idividuals (MNI) for statistical aalysis yielded similar results (see Ross, 2006). Discussio The low richess value of SCA-F i compariso to all other rarefied SCA values (Fig. 3) idicates that despite differeces i sample size, the middle- ad late Coiacia mosasaur faua was sigificatly less rich tha those from ay subsequet SCA. Thus the apparet trajectory (Fig. 2) of the hydropedal mosasaur radiatio durig the Coiacia ad Satoia is ideed real. Give the first appearace of Platecarpus (=Agolasaurus) i the late Turoia (Ligham-Soliar 1994) ad high mosasaur richess i the early Campaia (Table 1), this iitial radiatio of hydropedal mosasaurs occurred over approximately 7 myr (usig the timescale of Gradstei et al., 2004). However, the above aalysis could ot determie amog the competig hypotheses for mosasaur richess durig the Campaia ad Maastrichtia (above). To address this, I rarefied the Campaia ad Maastrichtia SCAs to = 77, the lowest sample size amog viable Campaia ad Maastrichtia SCAs. These data are preseted i Table 2 ad Figure 4. Agai, usig MNI for statistical aalysis yielded similar results (Ross, 2006). Whe rarefied, both of the cofidece itervals for the Campaia SCAs available for aalysis (H ad I) fall below the richess of SCA-M. Thus the taxoomically rich mosasaur commuities i the early Campaia are actually statistically sigificatly less rich tha those i the early Maastrichtia. Durig the Maastrichtia, the rarefied richess level of latest Maastrichtia SCA-O caot be distiguished from SCA-M, while middle Maastrichtia SCA-N is statistically sigificatly lower tha SCA-M. This may idicate that richess levels i the Maastrichtia were fluctuatig, though the overall tred appears stable. Likely affectig this picture are taxa kow to exist i the Maastrichtia which caot be cofidetly assiged to oe or more SCAs (Table 1). FIGURE 2. Stadig richess of mosasaur geera per SCA. Solid lie ad squares idicate miimum richess per SCA. Dashed lie ad circles idicate the upper limit of mosasaur geeric richess i istaces where the precise biostratigraphies of certai geera are ot fully resolved. Late Cretaceous time scale costructed usig TS-Creator versio 2.0. RICHNESS LEVELS AND MORPHOLOGICAL DIVERSITY Give the above, it is possible to trace the morphological treds withi the cotext of the first appearace of plesiopedal mosasaurs, followed by the appearace ad cotiued radiatio of hydropedal mosasaurs durig the late Cretaceous. Throughout their history, mosasaurs followed two morphological trajectories: icreasig adaptatio to their marie eviromet ad icreasig detal morphological complexity. Here I will examie the latter. Massare (1987; 1997, p. 406-407) defied six feedig guilds for Mesozoic marie reptiles based upo tooth morphology: Cut Poited, robust coe with two or more sharp logitudial cariae or ridges, which ca be serrated. Pierce II/Geeral Poited, somewhat sleder, curved tooth of moderate legth. Two cariae or fie logitudial ridges are ofte preset. Pierce I Very sleder, sharply poited, log, delicate tooth which ca be smooth or have fie logitudial ridges. Smash Small, straight or slightly curved tooth with a rouded but acute apex. TABLE 2. Rarefactio of mosasaur richess per SCA, rarefied to = 49 ad = 77. SCA-F SCA-G SCA-H SCA-I SCA-M SCA-N SCA-O # Idividuals, N 49 71 1087 155 77 89 220 # Geera (S) 3 6 9 7 8 7 8 E(Ŝ) for = 49 3 5.37 4.75 5.35 7.01 6.32 6.24 s(ŝ) for = 49 (-) 0.66 0.82 0.92 0.81 0.66 0.94 E(Ŝ) for = 77 x x 5.20 6.11 8 6.84 7.00 s(ŝ) for = 77 x x 0.90 0.76 (-) 0.36 0.79

ROSS MOSASAUR RICHNESS AND DIVERSIFICATION 413 FIGURE 3. Plot of mosasaur richess amog SCAs F, G, H, I, M, N, ad O. Data are rarefied to = 49 (solid rectagles) ad 95% cofidece itervals are plotted (vertical bars). Crush Small, robust, straight or slightly curved tooth with a blut apex. Cruch Very blut, bulbous or peg-like. She the (1987, 1997) assiged mosasaur taxa to at most three guilds (cut, crush, ad possibly pierce II/geeral). Re-examiatio of mosasaur detal morphology expads the variety of guilds exploited to iclude cut, pierce II/geeral, smash, crush, ad cruch, ad illumiates treds i the acquisitio of ovel detal morphologies (Fig. 5). Mosasaur geera were assiged to feedig guilds (Table 1), ad I tabulated the umber of geera per feedig guild per SCA (Table 3). The relative abudaces of guilds from selected SCAs are preseted i Figure 6. Discussio Novel detal morphologies, iferred to reflect differeces i feedig adaptatios, appear sporadically amog mosasaurs durig the Late Cretaceous (Table 3). The earliest, plesiopedal mosasaurs from the Ceomaia ad Turoia (SCAs A through D) were small ear-shore predators with cut guild FIGURE 4. Plot of mosasaur richess amog SCAs H, I, M, N, ad O. Data are rarefied to = 77 (solid rectagles) ad 95% cofidece itervals are plotted (vertical bars). FIGURE 5. Tooth morphology guilds (from Massare, 1987, 1997) assiged to a simplified phylogey of the Mosasauridae. Phylogey based o Bell ad Polcy (2005), Ligham- Soliar (1988), ad Schulp et al. (2004). The geus Progathodo icludes species assiged to both the cut ad cruch guilds. tooth morphologies (Fig. 6, Table 3). Early hydropedal mosasaurs from the Turoia ad Coiacia (SCAs D through F), icludig Clidastes, Platecarpus, ad Tylosaurus, were geeralist ambush predators also withi the cut guild. They possess the domiat detal morphology amog all mosasaurs, ad differeces i prey amog these taxa are thought to have bee related to body size: the > 10 m Tylosaurus is kow to have cosumed fish, sharks, ad swimmig birds, as well the 4 m Clidastes (Marti ad Bjork, 1987). For early 14 myr, all kow mosasaurs retaied the cut guild tooth morphology (Fig. 6). The first appearace of Halisaurus i the middle Satoia itroduces the first ovel tooth morphology amog mosasaurs ( pierce II/geeral ; Fig. 6, SCAs A G). Based o detal morphologies, throughout the bulk of the Campaia (SCAs H K), geeralist cut guild members comprised approximately 80% of the mosasaur geera at ay give time, while pierce II/geeral ad cruch guilds are sparsely represeted (Table 3, Fig. 6). The first declie (73%; Fig. 6) i cut guild mosasaurs durig latest Campaia SCA-L coicided with the first appearace of the cruch guild mosasaurs. These chagig percetages presage the more sigificat chages i mosasaur predatory behavior i the Maastrichtia. Durig the Maastrichtia, both the percetage ad raw umber of cut guild geera plummets, mosasaur feedig behavior becomes more evely distributed amog guilds (Table 3, Fig. 6). Far from their predatory domiace durig the Campaia, cut guild members are reduced to 37% of geera by the close of the Maastrichtia (SCA-O). Durig this same time, mosasaurs which preyed upo shelled orgaisms ( cruch ad crush guilds) become as taxoomically rich as geeralists. The chage i mosasaur predatory behavior durig the Maastrichtia was complex, beig drive by at least three factors. First, as oted above, durig at least part of the Maastrichtia, mosasaur richess levels were higher tha durig the early Campaia. However, ad most iterestig, Maastrichtia richess levels also appear geerally stable (see above), so the largest shifts i tooth guild acquisitio ad distributio are occurrig withi the cotext of stable richess. Secod, cut-guild russelosauries (Bell ad Polcy, 2005; Fig. 5) declie precipitously. Third, expasio of the mosasauries (Bell ad Polcy, 2005; Fig. 5) fuels rapid tooth guild diversificatio (Fig. 6). Here I will focus o the last two poits.

414 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 29, NO. 2, 2009 TABLE 3. Number of mosasaur geera per tooth guild per SCA. Tooth Guilds A B C D E F G H I J K L M N O Cut 1 1 4 1 3 5 8 8 6(7) 7(9) 7(8) 7 4(5) 4 Pierce I Pierce II/Geeral 1 1 1 1 1 1 1 2 2 Smash (1) (1) (1) Cruch 1 1(2) 1(2) 1(2) Crush 1 1 1 1 1 1 2 2 Numbers i paretheses represet highest possible umber of geera per guild whe stratigraphically uresolved taxa are cosidered. Feedig guilds from Massare (1987, 1997). With the exceptio of the Maastrichtia smash guild geus Goroyosaurus, all russelosauries fall withi the cut guild (Fig. 5; Table 1). After the last appearaces of Tylosaurus ad Haiosaurus i the early Maastrichtia (SCA-M), russellosauries were reduced to o more tha two geera. Mosasauries, however, displayed cosiderable morphological differetiatio early i the clade, with the first appearace of the cruch guild Globides i the early Campaia (Russell, 1975). Eve amog the plesiomorphic cut guild mosasauries, such as Clidastes, Kourisodo, ad Progathodo, there are markedly differet tooth morphologies (see Russell, 1967; Nicholls ad Meckert, 2002). These basal morphologies served as platforms for the acquisitio of derived pierce II/geeral, cruch, ad crush guilds durig the latest Campaia ad Maastrichtia. Moreover, the umerical domiace of mosasaurie cut guild geeralists such as Mosasaurus compared to the russellosaurie Plioplacearpus (Table 1) idicate that itra-clade geeralist competitio was domiated by mosasauries. By the latest Maastrichtia (SCA-O), mosasaurs had developed tooth morphologies represetig five of Massare s six feedig guilds, ad also displayed cosiderable itra-guild diversity, particularly amog mosasauries. Mosasaurs apparetly coverged o similar tooth morphologies see i extict groups. The highly derived mosasaurie Plotosaurus, a pierce II/geeral guild member ad probable pursuit predator, coverged o features see i early ichthyosaurs (Camp, 1942; Lidgre et al., 2007). Likewise, the tooth morphology of cruch guild members Progathodo ad Igdamaosaurus coverged o the ammoite-cosumig ichthyosaurs (Ligham-Soliar, 1988; Christiase ad Bode, 2002; Massare, 1987, 1997). Other mosasaurs competed directly with other taxa already occupyig certai feedig guilds. Plesiomorphic cut guild members ad the smash guild Goroyosaurus may have competed for the prey of other large predatory reptiles such as plesiosaurs ad crocodilias, respectively (see Ligham-Soliar, 2002, cocerig the latter). The success of the cut guild mosasaurs may be due to advatages i ambush-style predatory behavior durig much of the Late Cretaceous (Massare, 1997). Regardless of whether they opportuistically exploited ovel feedig strategies or competed directly agaist established groups, by the close of the Maastrichtia mosasaurs established themselves as the top predators i every ecological area i which they competed. CONCLUSIONS Throughout their history, mosasaurs developed far greater diversity of predatory behavior amog fewer taxa ad over a shorter time tha ay other sigle group of Mesozoic marie reptiles, perhaps eve amog all marie vertebrates. By the time of their extictio, mosasaurs had udergoe a sigificat radiatio early i their history, ad maitaied high geeric richess levels throughout the Maastrichtia. That their ultimate extictio evet coicides with the zeith of their morphological ad ecological diversity, icludig large ad small geeralist predators, pursuit-style piscivores, mollusc-crushers, cephalopodcruchers, ad crocodiliform estuary dwellers, oly serves to cast their demise ito greater relief. FIGURE 6. Relative abudace plots of mosasaur feedig guilds for all SCAs i the late Cretaceous. Maximum umbers of geera per tooth guild (Table 1) are plotted. Tooth guilds draw after Massare (1987). Miimum couts produce slight differeces but display the same treds. ACKNOWLEDGMENTS May thaks to G. Bishop, D. Brikma, J. Bolt, M. Brett- Surma, T. Culver, J. Ebersol, M. Everhart, W. Gallagher, N. Gilmore, P. Holroyd, J. Igle, R. Johso, D. Kig, S. MacLeod, L. Marti, C. Mehlig, D. Miao, D. Parris, A. Risburg, A. Schulp, ad G. Taguchi, who provided great assistace i researchig collectios, ad for helpful discussios o local stratigraphy ad collectios history. I thak G. Bell for discussios o mosasaurs, ad H. Cappetta for aid i refiig mosasaur records from Morocco. I am idebted to D. E. Fastovsky ad three aoymous reviewers for commets ad critiques of earlier drafts of this mauscript. The Departmet of Geoscieces at the Uiversity of Rhode Islad provided much of the support ad facilities, ad the Geological Society of America provided partial support used i the completio of this project.

ROSS MOSASAUR RICHNESS AND DIVERSIFICATION 415 LITERATURE CITED Bardet, N., X. P. Suberbiola, ad N.-E. Jalil. 2003. A ew mosasauroid (Squamata) from the Late Cretaceous (Turoia) of Morocco, Comptes Redus Palevol 2:607 616. Bell, G. L., Jr., ad M. Polcy. 2005. Dallasaurus tureri, a ew primitive mosasauroid from the Middle Turoia of Texas ad commets o the phylogey of Mosasauridae (Squamata). Netherlads Joural of Geoscieces 84:177 194. Camp, C. L. 1942. Califoria Mosasaurs. Uiversity of Califoria Press, Berkeley, Califoria, 67 pp. Christiase, P., ad N. Bode. 2002. A ew species of gigatic mosasaur from the late Cretaceous of Israel. Joural of Vertebrate Paleotology 22:629 644. Cobba, W. A. 1993. Diversity ad distributio of Late Cretaceous Ammoites, Wester Iterior, Uited States; pp. 435 451 i W. G. E. Caldwell ad E. G. Kauffma (eds.), Evolutio of the Wester Iterior Basi. Caadia Associatio of GeologistsSpecial Paper 39. Cobba, W. A., E. A. Merewether, T. D. Fouch, ad J. D. Obradovich. 1994. Some Cretaceous shorelies i the Wester Iterior of the Uited States; pp. 393 425 i M. V. Caputo, J. A. Peterso, ad K. J. Fraczyk (eds.), Mesozoic Systems of the Rocky Moutai Regio. USA. Fastovsky, D. E., Y. Huag, J. Hsu, J. Marti-McNaughto, P. M. Sheeha, ad D. B. Weishampel. 2004. Shape of Mesozoic diosaur richess. Geology 32:877 880. Fastovsky, D. E., Y. Huag, J. Hsu, J. Marti-McNaughto, P. M. Sheeha, ad D. B. Weishampel. 2005. Shape of Mesozoic diosaur richess, Commet ad Reply. Geology (olie) DOI 10.1130/0091-7613(2005) 31<e75:R>2.0.CO;2. Gradstei, F. M., J. G. Ogg, ad A. G. Smith. 2004. A Geologic Time Scale 2004. Cambridge Uiversity Press, Cambridge, Uited Kigdom, 589 pp. Haq, B., J. Hardebol, ad P. Vail. 1987. Chroology of fluctuatig sea levels sice the Triassic. Sciece 235:1156 1167. Heck, K. L. J., G. Va Belle, ad D. Simberloff. 1975. Explicit calculatio of rarefactio diversity measuremet ad the determiatio of sufficiet sample size. Ecology 56:1459 1461. Hirayama, R. 1997. Distributio ad diversity of Cretaceous cheloioids; pp. 225 241 i J. M. Calloway ad E. L. Nicholls (eds.), Aciet Marie Reptiles. Sa Diego: Academic Press. Hulbert, S. H. 1971. The o-cocept of species diversity: a critique ad alterative parameters. Ecology 59:577 586. Keedy, W. J., N. H. Ladma, W. K. Christese, W. A. Cobba, ad J. M. Hacock. 1998. Marie coectios i North America durig the late Maastrichtia: Palaeogeographic ad palaeobiogeographic sigificace of the Jeletzkytes ebrascesis Zoe cephalopod faua from the Elk Butte Member of the Pierre Shale, SE South Dakota ad NE Nebraska. Cretaceous Research 19:745 775. Kriwet, J., ad M. J. Beto. 2004. Neoselachia (Chodrichthyes, Elasmobrachii) diversity across the Cretaceous-Tertiary Boudary. Palaeogeography, Palaeoclimatology, Palaeoecology. 214:181 194. Ladma, N. H., R. O. Johso, ad L. E. Edwards. 2004a. Cephalopods from the Cretaceous-Tertiary boudary iterval i Atlatic Coastal Plai, with a descriptio of the highest ammoite zoes i North America. Part 1. Marylad ad North Carolia: America Museum Novitates 3454, 64 pp. Ladma, N. H., R. O. Johso, ad L. E. Edwards. 2004b. Cephalopods from the Cretaceous/Tertiary boudary iterval o the Atlatic Coastal Plai, with a descriptio of the highest ammoite zoes i North America. Part 2. Northeaster Momouth Couty, New Jersey. Bulleti of the America Museum of Natural History 287:107 pp. Lidgre, J., J. W. M. Jagt, ad M. W. Caldwell. 2007. A fishy mosasaur: the axial skeleto of Plotosaurus (Reptilia, Squamata) reassessed. Lethaia 40:153 160. Ligham-Soliar, T. 1988. The mosasaur Goroyosaurus from the upper Cretaceous of Sokoto State, Nigeria. Paleotology 31:747 762. Ligham-Soliar, T. 1994. The mosasaur "Agolasaurus" bocagei (Reptilia: Mosasauridae) from the Turoia of Agola re-iterpreted as the earliest member of the geus Platecarpus. Paläot. Z. 68:267 282. Ligham-Soliar, T. 2002. First occurrece of premaxillary caiiform teeth i the Varaoidea: Presece i the extict mosasaur Goroyosaurus (Squamata: Mosasauridae) ad its fuctioal ad paleoecological implicatios. Lethaia 35:187 190. Marti, J. E. M., ad P. R. Bjork. 1987. Gastric residues associated with a mosasaur from the Late Cretaceous (Campaia) Pierre Shale i South Dakota. Dakoterra 3:68 72. Massare, J. A. 1987. Tooth morphology ad prey preferece of Mesozoic marie reptiles. Joural of Vertebrate Paleotology 7:121 137. Massare, J. A. 1997. Part VI: Fauas, behavior ad evolutio itroductio; pp. 401 421 i J. M. Callaway ad E. L. Nicholls (eds.), Aciet Marie Reptiles. Academic Press, Sa Diego, Califoria. Miller, K. G., P. J. Sugarma, J. V. Browig, M. A. Komiz, J. C. Heradez, R. K. Olsso, ad J. D. Wright. 2003. Late Cretaceous chroology of large, rapid sea-level chages: Glacioeustasy durig the greehouse world. Geology 31:585 588. Mulder, E. W. A. 2003. O latest Cretaceous tetrapods from the Maastrichtia type area. Publicaties va het Natuurhistorisch Geootshap i Limburg, Reeks XLIV, afleverig, Natuurpublicaties Limburg, Maastricht, 188 pp. Mulder, E. W. A., J. W. M. Jagt, M. M. M. Kuypers, H. H. G. Peeters, ad P. Rompe. 1998. Prelimiary observatios o the stratigraphic distributio of late Cretaceous marie ad terrestrial reptiles from the Maastricht area (SE Netherlads, NE Belgium). Oryctos 1:55 64. Nicholls, E., ad D. Meckert. 2002. Marie reptiles from the Naaimo Group (Upper Cretaceous) of Vacouver Islad. Caadia Joural of Earth Scieces 39:1591 1603. Obradovich, J. D. 1993. A Cretaceous time scale; pp. 379 396 i W. G. E. Caldwell ad E. G. Kauffma (eds.), Evolutio of the Wester Iterior basi. Geological Associatio of CaadaSpecial Paper 39. Pearso, D. A., T. Schaefer, K. R. Johso, D. J. Nichols, ad J. P. Huter. 2002. Vertebrate biostratigraphy of the Hell Creek Formatio i southwester North Dakota ad orthwest South Dakota; pp. 145 167 i J. H. Hartma, K. R. Johso, ad D. J. Nichols (eds.), The Hell Creek Formatio of the orther Great Plais: a itegrated cotietal record of the ed of the Cretaceous. Geological Society of America Special Paper 361, Boulder, Colorado. Perch-Nielse, K. 1985. Mesozoic calcareous aofossils; pp. 329 426 i H. M. Bolli, J. B. Sauders, ad K. Perch-Nielse (eds.), Plakto Stratigraphy. Cambridge Uiversity Press, Cambridge, Eglad. Polcy, M., ad G. L. Bell. 2005. Russellosaurus cohei. ge.,. sp., a 92 millio-year old mosasaur from Texas (USA), ad the defiitio of the parafamily Russellosauria. Netherlads Joural of Geoscieces 84:321 333. Ross, M. 2006. Richess treds of mosasaurs (Diapsida, Squamata) durig the late Cretaceous. Ph.D. dissertatio Uiversity of Rhode Islad, 197 pp. Russell, D. A. 1967. Systematics ad morphology of America mosasaurs. Peabody Museum of Natural History Bulleti 23. Yale Uiversity, New HaveCoecticut 237 pp. Russell, D. A. 1993. Vertebrates i the Cretaceous Wester Iterior Sea; pp. 665 680 i W. G. E. Caldwell ad E. G. Kauffma (eds.), Evolutio of the Wester Iterior Basi. Geological Associatio of CaadaSpecial Paper 39. Saders, H. L. 1968. Marie bethic diversity: A comparative study. Biometrics 25:517 535. Schulp, A. S., J. W. M. Jagt, ad F. Foke. 2004. New material of the mosasaur Cariodes belgicus from the Upper Cretaceous of the Netherlads. Joural of Vertebrate Paleotology 24:744 747. Sheeha, P. M., D. E. Fastovsky, R. G. Hoffma, C. B. Berghaus, ad D. L. Gabriel. 1991. Sudde extictio of the diosaurs: latest Cretaceous, upper Great Plais, USA. Sciece 254:835 839. Sheldo, M. A. 1996. Stratigraphic distributio of mosasaurs i the Niobrara Formatio of Kasas. Paludicola 1:21 31. Simberloff, D. S. 1972. Properties of the rarefactio diversity measuremet. America Naturalist 106:414 418. Sissigh, W. 1977. Biostratigraphy of Cretaceous calcareous aoplakto. Geologie e Mijbouw 56(1):37 65. Tipper, J. C. 1979. Rarefactio ad rarefictio the use ad abuse of a method i paleoecology. Paleobiology 5:423 434. TS-Creator visualizatio of ehaced Geologic Time Scale 2004 database (Versio 2.0; 2006). J. Ogg (database coordiator) ad A. Lugowski (software developer). http://www.stratigraphy.org ad http://www.chroos.org. Submitted November 15, 2006; accepted August 14, 2008.

416 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 29, NO. 2, 2009 APPENDIX: SCA DEFINITIONS Each SCA is defied o the basis of ammoites ad calcareous aofossil biozoes, eustatic sea level cycles, ad umerical ages. Uless otherwise idicated, the ammoite zoes follow Cobba s (1993) ad the Cobba et al. s (1994) zoatio from deposits of the Wester Iterior of North America. Calcareous aoplakto zoes are icluded accordig to the formulatio of Sissigh (1977) ad Perch-Nielse (1985). The third-order eustatic cycles of Haq et al. (1987) are employed to desigate global sea level sequeces, ad aid i refiig the timeframe for lithologic uits of ucertai age. Numerical ages for the upper ad lower boudaries of each SCA are discussed to the resolutio available, accordig to the curret time scale for the late Cretaceous (Gradstei et al., 2004). The SCAs ad their respective biostratigraphic ad sequece-stratigraphic criteria are illustrated i Figure 1. SCA-A spas the early Ceomaia, from ammoite zoes Neogastroplites haasi to N. macleari. It icludes calcareous aoplakto subzoe CC9a ad lower subzoe CC9b, ad third-order cycle UZA 2.3. The base of the Ceomaia (ad the Late Cretaceous) is curretly placed at 99.6 0.9 Ma (Gradstei et al., 2004). Obradovich (1993) reported the age of the Clay Spur betoite (above the N. macleari zoe) i the Mowry Shale as 97.2 0.7 Ma. The duratio of SCA-A is approximately 2.3 myr. SCA-B spas the late early Ceomaia through the middle Ceomaia, from ammoite zoes Colioceras tarratese to Plesiacathocera wyomigese. It icludes the third-order cycled UZA2.4 ad the upper portio of aoplakto subzoe CC9b. The Clay Spur betoite at top of the Mowry Shale i Wyomig provides a age of 97.2 0.7 Ma (Obradovich, 1993) for the base of SCA (B). A betoite from the Duvegaoceras podi zoe (which overlies the P. wyomigese zoe) of the Frotier Fm. i Wyomig is dated at 94.6 0.6 Ma (Obradovich, 1993). This provides a upper limit for the age of SCA (B), ad a maximum duratio of approximately 2.4 myr for this SCA. SCA-C spas the late Ceomaia through early Turoia. It icludes sedimetary uits deposited durig the highest eustatic sea levels of the Cretaceous (UZA2.5; Haq et al., 1987). It icludes the late Ceomaia ammoite zoe of Duvegaoceras podi to the early Turoia Mammites odosoides, ad aofossil zoes CC10 ad lower CC11. A betoite from the Duvegaoceras podi zoe i the Frotier Fm. of Wyomig is dated at 94.6 0.6 Ma (Obradovich, 1993), ad provides a date for the base of SCA-C. A proouced drop i sea level markig the UZA 2.5/2.6 boudary defies the top of SCA-C. Based upo 40 Ar/ 39 Ar dates (Obradovich, 1993), this boudary must be youger tha 93.3 Ma (Neocardioceras juddii zoe; latest Ceomaia) ad older tha 90.5 Ma (Prioocyclus hyatti zoe; middle Turoia). Based upo Sr-isotope curves from the Rarita Fm. i New Jersey, the age of top of the Bass River III sequece is 92.8-92.1 1.0 Ma (Miller et al., 2003). SCA-C is likely o more tha 2.5 myr i duratio. SCA-D spas the middle to late Turoia, ad icludes ammoite zoes Colligoiceras woollgari through Prioocyclus germari as well as the upper portio of aofossil zoe CC11, all of CC12, ad perhaps the base of CC13. It icludes third-order eustatic cycles UZA2.6, UZA2.7, ad the base of UZA3.1. The base of SCA (D) is approximately equivalet to the top of the Bass River III sequece i New Jersey, from 92.8-92.1 1.0 Ma (Miller et al., 2003). The curret age for the Turoia/ Coiacia boudary is 89.3 1.0 Ma (Gradstei et al., 2004), makig the duratio of SCA-D betwee 1.8 ad 2.5 myr. SCA-E spas the early- to early-middle Coiacia, ad icludes the ammoite zoes of Forresteria peruaa ad F. alluaudi. It icludes aofossil zoe CC13 ad much of the third-order cycle UZA3.1. The base of the Coiacia is curretly dated at 89.3 1.0 Ma (Gradstei et al., 2004), ad a betoite from the Marias River Shale, Motaa, i the Scaphites prevetricosus zoe (correlative to the F. alluaudi zoe) is dated at 88.3 0.6 Ma (Obradovich, 1993). The top of UZA3.1 is dated at 87.5 1.0 Ma based, o Sr-isotope curves (Miller et al., 2003). The duratio of SCA (E) is approximately 1.8 myr. SCA-F spas the early-middle to late Coiacia, ad icludes the ammoite zoes of Schaphites vetricosus ad S. depressus, aofossil zoe CC14 ad the third-order cycle UZA3.2. The base of UZA3.2 is placed at 87.5 1.0 Ma (Miller et al., 2003), ad the Coiacia/Satoia boudary is curretly assiged a date of 85.8 0.7 Ma (Gradstei et al., 2004). The duratio of SCA (F) is approximately 1.7 myr. SCA-G spas the Satoia stage. The duratio of the Satoia, 2.3 myr; from 85.8 0.7 Ma to 83.5 0.7 Ma (Gradstei et al., 2004). It icludes the ammoite zoes Clioscaphites saxitoiaus through Desmoscaphites bassieri ad aofossil zoes CC15 through CC17. It icludes sea-level cycle UZA3.3, ad early all of UZA3.4 SCA-H is early Campaia i age, ad icludes the ammoite zoes of Scaphites leei III through Baculites sp. (weak flak ribs). It icludes calcareous aofossil zoe CC18, ad third-order eustatic cycle UZA3.5. The base of the Campaia is dated at 83.5 0.7 Ma (Gradstei et al., 2004), ad the Baculites obtusus ammoite zoe of at the base of SCA-I is dated at 80.5 0.5 Ma (Obradovich 1993). The duratio of SCA-H is approximately 3.0 myr. SCA-I is early middle Campaia i age, ad icludes ammoite zoes Baculites obtusus through B. sp. (smooth), calcareous aofossil zoe CC19, the lower part of CC20, ad the third-order eustatic cycles UZA4.1 ad lower UZA4.2. At the base of the B. obtusus zoe, the Ardmore betoite from the Claggett Shale, Wyomig, is dated at 80.5 0.5 Ma (Obradovich, 1993), ad Miller et al. (2003) provide a age estimate from the Sr-isotope curve of 81.0 0.5 for the base of UZA4.1. The upper boudary of SCA-I is less well resolved. Miller et al. (2003) provide a Sr-isotope date of 77.8 0.5 Ma for the top of UZA4.2, which is i lower SCA (J); the duratio of SCA (I) must therefore be less tha 2.7 Ma. Give a duratio of 12.9 m.y. (Gradstei et al., 2004) ad 23 ammoite zoes, Campaia ammoite zoes represet a average of 0.56 m.y. Sice oly oe ammoite zoe (Baculites perplexus [late]) is i both SCA (J) ad UZA4.2, I subtract 0.6 Ma from the 77.8 Ma age of upper UZA4.2, ad propose a 2.1 myr estimate for the duratio of SCA-I. SCA-J spas the late middle Campaia, ad icludes the ammoite zoes Baculites perplexus through B. scotti. It also icludes the upper part of aofossil zoe CC20, most of CC21, ad the upper portio of third-order eustatic cycle UZA4.2 ad UZA4.3. The base of UZA4.3 is placed at 78.5 0.5 Ma (Miller et al., 2003), ad Obradovich (1993) reports a age of 75.9 0.5 Ma from the B. scotti zoe i the Lewis Shale of New Mexico. The duratio of SCA-J is approximately 2.6 myr. SCA-K is early late Campaia i age, ad spas the Didimoceras ebrascese through Baculites compressus ammoite zoes. It icludes calcareous aofossil zoe CC22 ad subzoe CC23a, ad the lower half of third-order eustatic cycle UZA4.4. The zoe of Baculites scotti, uderlyig the D. ebrascese zoe, is dated at 75.9 0.7 Ma, while the base of the B. compressus zoe is dated at 73.4 0.4 based o a betoite from the Bearpaw Shale i Motaa (Obradovich, 1993). The duratio of SCA (K) is approximately 2.5 myr. SCA-L is latest Campaia i age, ad spas the ammoite zoes Baculites cueatus through B. reesidei. It icludes the calcareous aofossil subzoes CC22b, ad CC23a-b, ad the upper portio of the third-order eustatic cycle UZA4.4. The B. compressus zoe, uderlyig B. cueatus has bee dated at 73.4 0.4 (Obradovich, 1993). The date for the Campaia/Maastrichtia boudary is 70.6 06 Ma (Gradstei et al., 2004). The duratio of SCA (L) is approximately 2.8 myr. SCA-M spas the early Maastrichtia, ad icludes the ammoite zoes Baculites baculus through B. cliobatus (Cobba et al., 1994). It also icludes calcareous aofossil subzoe CC23c ad zoe CC24, as well as the lower portio of third-order cycle UZA4.5. The base of the Maastrichtia is assiged a date of 70.6 0.6 Ma (Gradstei et al., 2004), ad a betoite from above the B. cliobatus zoe (top of lower Maastrichtia) i the lower Fox Hills Formatio i Red Bird, Wyomig is dated at 69.4 0.4 Ma (Obradovich, 1993). The duratio of SCA-M is approximately 1.2 myr. SCA-N is early late Maastrichtia i age, ad spas the ammoite zoes of Hoploscaphites birkeludi through lower Discoscaphites coradi (Cobba et al., 1994, Ladma et al., 2004a,b). I the Wester Iterior, the highest ragig ammoite zoe, Jeletzkytes ebrascesis, is at least partially correlative to the D. coradi zoe of the Atlatic ad Gulf coasts, but it is at preset ukow precisely how high J. ebrascesis rages (Keedy et al., 1998). SCA (N) icludes calcareous aofossil zoe CC25a,b ad the upper portio of thirdorder eustatic cycle UZA4.5. The base of the H. icolleti zoe is dated at 69.4 0.4 Ma (Obradovich, 1993). Miller et al. (2003) provide a Sr-isotope date of 67 0.5 Ma for the top their Navesik I sequece (equivalet to the top of UZA4.5). Thus the duratio of SCA-N is approximately 2.4 myr. SCA-O spas the latest Maastrichtia, icludes the upper portio of the ammoite zoe of Discoscaphites coradi through Discoscaphites iris (Ladma et al., 2004a,b), calcareous aofossil subzoe CC25c ad zoe CC26. Third-order eustatic cycle TA1.1 ad the base of TA1.2 are icluded i SCA-O. The base of TA1.1 is dated at 67.0 0.5 Ma based o Sr-isotope curves (Miller et al., 2003), ad the upper boudary of SCA-O is the K/T boudary, defied by the Chixulub impact-derived Ir-aomaly ad dated at 65.5 0.3 Ma (Gradstei et al., 2004). The duratio of SCA-O is approximately 1.5 myr.