Incumbency, Diversity, and Latitudinal Gradients

Paleobiology, 34(2), 2008, pp. 169–178

MATTERS OF THE RECORD

Incumbency, diversity, and latitudinal gradients

James W. Valentine, David Jablonski, Andrew Z. Krug, and Kaustuv Roy

James W. Valentine. Department of Integrative Biology and Museum of Paleontology, University of California, Berkeley, California 95720. E-mail: [email protected] David Jablonski and Andrew Z. Krug. Department of Geophysical Sciences, University of Chicago, 5734 South Ellis Avenue, Chicago, Illinois 60637. E-mail: [email protected], [email protected] Kaustuv Roy. Section of Ecology, Behavior and Evolution, Division of Biological Sciences, University of California, San Diego, 9500 Gillman Drive, La Jolla, California 92093. E-mail: [email protected]

Accepted: 5 November 2007

Physical environmental factors have been number or proportion of genera that have ex- seen as paramount in determining many panded from the Tropics (the ratio is generally large-scale biodistributional patterns in time ϳ3:1), and even this smaller number is prob- and space. Although this is probably correct ably an overestimate, because the extratropical for many situations, this view has become so post-Paleozoic fossil record is so much better pervasive that it has led to the neglect of the sampled than that of the Tropics (Allison and role of biotic interactions in setting large-scale Briggs 1993; Jablonski 1993; Jackson and John- diversity patterns. (In this paper diversity de- son 2001; Jablonski et al. 2006; Valentine et al. notes taxonomic richness.) New approaches to 2006). The result is a gradient wherein the ma- this perennial debate on the roles of physical jority of taxa in each latitudinal bin is shared and biotic forces in paleoecology and macro- with the Tropics. For genera that originate ex- evolution are needed, and here we explore an tratropically, expansion into the Tropics is vir- argument for the role of incumbency or pri- tually unknown, at least at shelf depths, and ority effects in the dynamics behind the most given the anti-tropical bias in the evidence of dramatic spatial pattern in biodiversity, the first occurrences, most apparent instances latitudinal diversity gradient. may be artifacts (Jablonski et al. 2006; see also A global analysis of the fossil record of liv- Vermeij 2005a). Over the past few centuries, ing marine bivalve genera and subgenera successful species invasions have been less (hereafter simply genera) of the continental frequent in tropical than extratropical regions, shelves provides perhaps the strongest evi- at least for mainland terrestrial communities, dence for the Out of the Tropics (OTT) dy- even though several studies have shown a pos- namic associated with the formation of the itive relationship between the number of ex- present marine latitudinal diversity gradient otic and native species within climatic zones (LDG) (Jablonski et al. 2006). The marine LDG (Rejmańek 1996; Sax 2001; Fine 2002; Sax and appears to be driven primarily by the origin Gaines 2006; Fridley et al. 2007). Fewer data of novel lineages in the Tropics, some of which are available for marine tropical settings, but then expand their ranges into higher, extra- success rates of invasions from the temperate tropical latitudes (see Jablonski 1993, 2005; zones also appear to be low, outside of severe- Clark and Crame 2003; Goldberg et al. 2005; ly disturbed or novel habitats (e.g., Hewitt Jablonski et al. 2006; Martin et al. 2007). Sup- 2002; Paulay et al. 2002; Fridley et al. 2007). port for this pattern comes from the over- Here we explore some features that may reg- whelmingly tropical first fossil occurrences of ulate this evolutionary and biogeographic dy- living bivalve genera and their subsequent ap- namic, and advance a counterintuitive hy- pearances in higher latitudes. Some genera are pothesis in hopes of stimulating research into first found in the extratropical fossil record, the neglected role of biotic factors in shaping but these never match, at any latitude, the the LDG.

ly, a deﬁnitive global data set of the ages of marine bivalve species is not available. How- ever, because only one species need reach higher latitudes to create a record of its genus there, these categories are linked to some ex- tent. High-latitude species might be older than tropical ones, although active recent high-latitude speciation is known or inferred in several terrestrial groups (see Mittelbach et al. 2007 and Wier and Schluter 2007 for refer- FIGURE 1. The median ages of living marine bivalve ences). However, the species/genus ratio is genera (solid line, diamonds) increases with latitude as low for extant high-latitude bivalves globally the number of genera (dashed lines) decreases. Modiﬁed with added data from supporting online material in Ja- (Polar S/G: 1.9; Tropical S/G: 5.1, suggesting blonski et al. 2006. sluggish per-taxon rates of marine speciation in polar regions [Krug et al. 2008]). Related to the OTT dynamic is the increase Possible Causes of Latitudinal in median ages of bivalve genera with latitude Diversity Dynamics as the number of genera decreases (Fig. 1). The tropical fauna retains most of the older genera The genus-level diversity gradient must be as they expand, but its high origination rate explained by some combination of origination, imparts a younger average age to the fauna extinction, and dispersal rates, and recent there (see Foote 2001 on the dynamical inter- models have explored interactions among pretation of taxon ages). This interpretation is these parameters (see Roy and Goldberg supported by Crame’s (2000) observation that 2007). As noted above, origination rates are diversity gradients are steepest for the bivalve highest in the more diverse regions and taxa clades dominated by younger genera. The few are exported to higher latitudes without eras- genera residing in higher latitudes are thus bi- ing latitudinal diversity trends, so either there ased toward those that evolved earliest, many is a gradient in extinction or young genera are demonstrably at low latitudes (Jablonski et al. being prevented from dispersing, or both. As 2006). Even the few groups with peak diver- shown below, paleontological estimates of ge- sity outside the Tropics, such as the anomal- neric extinction rates in the Pliocene and Pleis- odesmatan bivalves, share the same funda- tocene (ϳ0.012–5 Myr ago) are no higher (and mental relationship between taxon age and di- may be lower) in polar regions than in tem- versity, with minimum genus ages roughly co- perate latitudes, so differential extinction inciding with their extratropical diversity alone cannot account for the temperate to po- maxima, and age maxima coinciding with di- lar decline in diversity (we use polar here sim- versity minima (Krug et al. 2006 and unpub- ply to denote regions poleward of 60Њ lati- lished data). In theory, patterns of species-lev- tude). Thus one or more factors not directly el origination might differ from those of line- related to origination and extinction rates per age formation at the genus level. Unfortunate- se must impede the cross-latitudinal expan-

TABLE 1. The qualitative pattern of factors associated with an incumbency model that proposes a ﬁltering of the poleward spread of lineages, chieﬂy arising in the Tropics, to shape the latitudinal diversity gradient. Factors 1 to 4 are based on observations whereas factors 5 and 6 are inferential.

Factor Tropical zone Temperate zone Polar zone 1. Diversity High Intermediate Low 2. Median age of genera Low Intermediate High 3. In-situ origination rate High Low Low 4. Extinction rate Low Higher? Low 5. Invasibility Low Higher? Low 6. Niche breadth Narrow Intermediate Broad INCUMBENCY, DIVERSITY, LATITUDE 171 sion of lower-latitude genera through restric- istence of a diverse deep-sea fauna at low tem- tions on their contained species. These poten- peratures indicates that evolution is quite ca- tial constraints fall into three general catego- pable of overcoming physiological problems ries. associated with the low end of the marine 1. Time. Many genera show a latitudinal temperature range. The fact that most of the array of species along shelves, with poleward marine bivalve families absent from the deep spread presumably involving speciation sea are suspension feeders (Knudsen 1979; across latitudes. Perhaps, then, insufficient Gage and Tyler 1991; Valentine et al. 2006) time has elapsed for the accumulation of pole- suggests that the attenuation of trophic re- ward speciations in many lineages. ‘‘Age and sources may be a more important barrier than area’’ arguments, with geographic expansion low temperature (see Rex et al. 2005 for a sim- as a positive function of taxon age, have a long ilar argument for species richness in the abys- pedigree (e.g., Willis 1922; Miller 1997; Jones sal environment). Furthermore, temperature et al. 2005; see Mittelbach et al. 2007 for a re- variability does not correlate with the LDG, view). However, the LDG has been at least a for it is greatest in middle and least at low and semi-permanent feature of the marine bio- high latitudes (Valentine 1973). Air tempera- sphere for much of the geologic past and tures can certainly be very low in high lati- seems to have been present in something like tudes, affecting intertidal forms, and intertid- its current form not just in the Neogene, but al and shallow subtidal zones may be ice cov- since the Mesozoic breakup of Pangaea (Crame ered or scoured, precluding living space sea- 2001, 2002), though evidently with a lower sonally. However, the bulk of the bivalve slope during nonglacial intervals. Given that fauna can live subtidally, in environments even complete polar extinction (e.g., during with more stable though low temperatures. glacial episodes, an improbable event for sub- Even in the terrestrial realm, a quantitative as- tidal organisms) would leave a pool of poten- sessment of factors affecting vascular plant di- tial invaders in adjoining provinces (e.g., versity patterns did not find mean annual Hickerson and Cunningham 2006), expecta- temperature per se to be an important con- tions for a time frame within which to mea- tributor (Kreft and Jetz 2007). Regarding other sure the lag for spreading into high latitudes physical variables, habitat area has also been are uncertain. Today’s Arctic Ocean first be- implicated in terrestrial gradients (e.g., Rosen- came a fully marine feature nearly 17.5 Myr zweig 1995), but along the northeastern Pacif- ago, shortly before the time of peak mid-Mio- ic, shelves are broadest in higher latitudes, cene warming (Jakobsson et al. 2007), when whereas the tropical shelf is quite narrow and temperate latitudes harbored a relatively rich embraces a relatively small area; habitat area bivalve fauna (e.g., Gladenkov and Sinelniko- thus opposes the LDG (Roy et al. 1998). Hab- va 1990; Ward 1992; Ogasawara 1994). Both itat heterogeneity seems important locally, but dispersal and local adaptation can be rapid at least for regions that lack coral reefs, as in relative to the time available. the eastern Pacific, the heterogeneity of phys- 2. Abiotic Aspects of Environmental Harshness ical habitats does not differ obviously between or Invasibility. Many authors have suggested high- and low-latitude shelves. Area and het- that the harsher climate encountered in higher erogeneity differences may help to explain in- latitudes produces physiological barriers that terregional diversity differences within cli- exclude most lineages (see review by Willig et mate zones, however (e.g., Crame and Rosen al. 2003). In this view, only a few lineages can 2002; Rex et al. 2005). adapt to conditions at very high latitudes, but 3. Biotic Aspects of Environmental Harshness those that do so may diversify there. Temper- or Invasibility. Such aspects would include ature has been a prime suspect in regulating food supply, activity of predators or patho- the LDG, and temperature correlates signifi- gens, competition for space or specialized cantly with both bivalve and prosobranch gas- habitat resources, and more generally, ecolog- tropod diversity along north-south shelves ical incumbency, priority, or niche preemption (e.g., Roy et al. 1998, 2000). However, the ex- effects. Levels of primary productivity do not 172 JAMES W. VALENTINE ET AL. strongly correlate with the LDG, either along spread dependency, which is somewhat anal- the shelf or in the open ocean (Longhurst ogous to density dependency (Smith 1935) in 1998); what correlation there is appears to be population biology and to diversity depen- negative. Variability in productivity does seem dency (Valentine 1972) in evolutionary paleo- to correlate well with diversity, owing chiefly ecology. Spreading rates recorded for an as- to a significant latitudinal increase in season- sortment of invaders suggest linear negative ality (Valentine 1973, 1983; Valentine and Ay- feedbacks that cause asymptotic approaches ala 1978), but a gradient in seasonality of pro- to an equilibrium spread rate. As Arim et al. ductivity cannot account for the gradient in note, this dynamic suggests the same kind of the antiquity of generic lineages. However, the resource competition that limits population observed OTT dynamic, with depressed di- growth. However, spreading rates associated versities and older faunas in progressively with global or regional warming, for example, higher latitudes, would arise if the poleward might also be constrained by the rate of warm- faunas themselves have features related to sea- ing, or by some intrinsic limitation that might sonality and other latitudinally graded con- lag behind that rate. The geographic range ditions that inhibit immigration. limits reached by spreading species are pre- Although each of these three types of fac- sumably imposed by a mix of factors similar tors might contribute to the present LDG, the to those that constrained their ranges in their third has garnered the least attention, so we previous territories. To be sure, the spread of focus here on evidence and interpretations species across important barriers, as between that bear on the relation of faunal incumbency widely separate habitat islands, requires a in biologically harsher or less receptive envi- founding invasion and favors the better dis- ronments to invasibility. Much has been pub- persers. However, the LDG for species of ma- lished on the tendency for incumbent taxa to jor constituents of western Pacific coral reef depress the invasibility of a region, so that communities, which require long-distance asymmetric biotic interchanges are often at- founding events in many cases, is symmetrical tributable to interregional variations in spe- around a low-latitude peak despite the uneven cialization, diversity and extinction intensity latitudinal distribution of reef-bearing shelves (e.g., Rosenzweig and McCord 1991; Vermeij and island platforms (Roberts et al. 2002; Con- 1991, 2005b; Jablonski and Sepkoski 1996; nolly et al. 2003). Beard 1998; Schluter 2000; Stachowicz and Til- Shifts in latitudinal ranges along shelves are man 2005; Wiens et al. 2006; Sax et al. 2007), best known during climate change, when but incumbency effects have rarely been in- warmer- or cooler-adapted species track iso- voked for the LDG. therms. The most recent fossil records of poleward marine expansions are associated with Spreading Across Latitudes the warming interglacial trends of the Pleis- If latitudinal variations exist in invasibility, tocene (e.g., Valentine 1961 for the Californian the LDG could involve a source-and-filter pro- Province; Tankard 1975 for South Africa; Raffi cess rather than simple diffusion. Movement et al. 1985 for the Mediterranean; Kendrick et of populations along a shelf can be accom- al. 1991 for Western Australia; Kitamura et al. plished by the spreading of populations along 2000 for Japan). When climates then cool, as the more or less continuous shelf habitats, during glacial intervals, some species’ pole- where long-distance transport is probably not ward limits retreat toward the equator, with required; the species are already present in species that have extended their ranges to- the adjoining territory and short-range dis- ward the poles during warming being the persals should usually suffice for invasion, ones most likely to retreat equatorward dur- provided that any former ecological barriers ing cooling (see Roy et al. 1995). The shape of to their dispersal, such as at provincial bound- the LDG thus may change during these cli- aries, have ameliorated. mate swings, but commonly this is reversible; Arim et al. (2006) reviewed data on the evidently some other step or condition is re- spread of invaders and discuss the concept of quired for significant and permanent changes INCUMBENCY, DIVERSITY, LATITUDE 173 to the latitudinal array of taxa across zonal cli- variability in trophic resource availability). mates. However, such short-term climate Thus the average higher-latitude species is changes do provide a mechanism for intro- ecologically more generalized and requires a ducing lineages, at least temporarily, into lat- larger share of available resources than the av- itudes from which they had been absent. erage lower-latitude species. Furthermore, more trophic resources may escape the con- Incumbency and Adaptation sumer portion of the food web when produc- If incumbent lineages preclude or at least tivity is variable, producing lags between rises inhibit invasions, diversity-dependent factors in productivity and increases in consumption. are presumably at work. Diversity-dependent Therefore more resources are required to sup- factors are those arising from resource limi- port the average species, and fewer species tations; when conditions do not favor resource can be supported in trophically variable en- partitioning, diversity must be relatively low, vironments. This is a different view from ap- whereas greater partitioning will accordingly proaches that hold lower species diversity to permit diversity to be higher (Valentine 1972). reflect open niches and greater resource avail- Factors such as temperature that are not ac- ability (for review see Stachowicz and Tilman tually partitioned among individuals, popu- 2005) and from ideas that invoke a latitudinal lations, or taxa—factors that are not used up gradient in environmental tolerances and by organisms—are diversity-independent in therefore geographic ranges, as suggested by this sense. Rapoport’s Rule (Stevens 1989; rejected for Diversity-dependent factors are inferred to marine mollusks by Roy et al. 1994, and ap- be related to environmental heterogeneity, ei- parently not a general phenomenon; see Gas- ther spatial or temporal. The correlation of ton et al. 1998; Ribas and Schoereder 2006; diversity with spatial habitat heterogeneity Ruggiero and Werenkraut 2007). has been demonstrated by experiment or ob- An incumbency hypothesis for the LDG re- servation in ecological time, from the popu- quires that species living in high latitudes lation up to the community level (e.g., Crom- tend to be more generalized with respect to bie 1945; Etter and Grassle 1992; Armbecht et limiting resources than species living in low al. 2004), and analyzed in evolutionary time at latitudes. This idea is supported by studies provincial and global levels (see Valentine documenting narrower diets among tropical 1973; Lomolino et al. 2006). As noted above, gastropods than in high-latitude forms, but however, such variation cannot itself account the data are few (references in Valentine et al. for the LDG. The effects of temporal instability 2002, and see Taylor and Taylor 1977, and also in productivity are likely to involve evolution- Sax et al. 2007 who propose that this notion ary rather than ecological scales and so are not may also apply in terrestrial systems). The ob- amenable to experimentation, but as season- servations that the range of morphological ality in productivity grades latitudinally it is elaboration is greatest among tropical species a candidate for affecting the LDG. The hy- (Vermeij 1978) may also be relevant here. This pothesis is that where trophic resource vari- formulation implies a gradient of limits on lo- ability is high, species persist by maintaining cal and regional carrying capacity. These lim- large populations, and/or by inhabiting a its presumably vary as environments change, wider range of habitats, and/or by consuming and species may well be able to evolve strat- a wider range of food items, particularly items egies promoting denser ecospace packing low in the trophic chain, relative to species in than observed in the modern bivalve fauna. trophically stable environments (Valentine Nevertheless, the persistence of the LDG in- 1973, 1983; the latitudinal shift to nonfeeding dicated by the fossil record suggests that the larvae in higher latitudes documented by, e.g., effects of incumbency that we hypothesize Thorsen 1950; Jablonski and Lutz 1983, and have been a general feature of faunal respons- Laptikhovsky 2006, may also be part of an es to latitudinal variation in diversity-depen- adaptive syndrome for increased seasonality, dent factors over long stretches of geologic as this would buffer larval survival against time. 174 JAMES W. VALENTINE ET AL.

(4.5%) are globally extinct and only five are now locally extinct. Four of the locally extinct genera live today in latitudesat50ЊN or above, from where they could easily be introduced into Arctic waters during climatic ameliora- tion (the other locally extinct genus lives to 45ЊN). In contrast, north temperate bivalves have suffered roughly 12% global extinction plus 8% local extinction, and we estimate the corresponding tropical extinctions to be roughly 8% and 1% respectively. These numbers must be treated as prelimi- FIGURE 2. Preliminary estimates of local and global genus extinction intensity in tropical, north temperate, nary. The north temperate fauna is by far the and north polar regions for the Pliocene and Pleistocene, best sampled; the Arctic fauna is less well based on augmented versions of the spatially explicit known, but its spatial homogeneity indicates databases of living bivalve species and living and extinct bivalve genera described by Jablonski et al. (2006). that fewer samples will capture a large frac- Tropical local extinction estimated by the proportion of tion of the biota; the tropical biota is worst taxa originating in the Tropics in the late Miocene–Pleis- sampled, and the numbers recorded here in- tocene that have lost their tropical presence (Jablonski et al. 2006). Tropical and temperate extinction intensities volve rough estimates (see caption of Fig. 2). differ significantly (p Ͻ 0.001; log-likelihood ratio test); Note, however, that the use of extinction pro- temperate and polar extinction intensities are not sig- portions within a single time frame reduces nificantly different, as expected given lower polar n, but differ in predicted direction; polar intensity is outside the sensitivity to sampling of the reported val- 99% binomial confidence limits of temperate value. ues. For example, for Pliocene-Pleistocene tropical extinction to reach temperate levels would require a twofold error in our estimate Definitive tests for macroevolutionary hy- of tropical extinction (from ϳ30 to ϳ60) or of potheses can be difficult to frame, but consis- tropical diversity (from ϳ450 to ϳ225); the tency arguments can be designed wherein vi- latter is especially unlikely given that Todd et olation of a logical expectation of a hypothesis al. (2002) report ϳ200 Pliocene bivalve genera constitutes disproof. If incumbency is impor- in tropical America alone, omitting the rich tant in regulating generic diversity latitudi- faunas of Florida and the Greater Antilles and, nally, then regional diversity dynamics more importantly, the great Indo-West Pacific (wherein a taxon’s range is not limited by ge- diversity maximum. Thus, the estimates ography) should exhibit reciprocal relation- shown here, with a counterintuitive, non- ships between extinction rates and invasion or monotonic relation of extinction intensity local origination (i.e., local origination pref- with latitude and standing diversity, present erentially occurs following the removal of in- an interesting challenge to the many hypoth- cumbent taxa, assuming a reasonably con- eses invoking predominantly direct abiotic stant climatic regime). One testable prediction controls on the dynamics of the LDG. Even is that the low-diversity, high-latitude faunas given that our extinction figures are prelimi- should exhibit relatively static compositions— nary, there is no evidence that extinction alone a high degree of incumbency—and so should can account for or promote the diversity dif- have had about the same generic compositions ference between temperate and polar zones. before the strong Pleistocene glaciations as These new results, though preliminary, are they do today. Data are not yet available for a a substantial refinement of the extinction pat- complete analysis, but a preliminary test can terns reported by Jablonski et al. (2006). The be made in the Northern Hemisphere Neo- earlier paper simply compared extinction gene, where bivalve faunas have been de- rates between tropical and extratropical scribed from a few localities north of 60Њ (Fig. zones, and Arctic diversity is such a small 2). Of 66 genera reported north of 60ЊN in Pli- fraction of the extratropical total (Ͻ15%) that ocene and Pleistocene assemblages, only three it had little effect on the first-order signal. Par- INCUMBENCY, DIVERSITY, LATITUDE 175 titioning the northern extratropics into tem- cal invaders. Combined molecular and pale- perate and polar zones allows us to show that ontological analyses are needed to assess bio- extinction intensity does not simply increase geographic sources of the marine Neotropical monotonically with latitude and supports a evolutionary rebound. potential role of biotic factors in the low in- Adaptations to the high-latitude environ- vasion rates evident for both the Tropics and ment, whether biotic, physical, or both com- the poles relative to the temperate zones. bined, seem to result in the suppression of Although temperature may not directly de- per-taxon origination relative to lower lati- termine standing diversity, its fluctuations tudes. This would explain the progressive fail- may set the pattern of extinction intensities, ure of generic lineages to penetrate into in- both being highest in temperate latitudes to- creasingly higher latitudes where ecological day and during Pleistocene climate swings generalists are most successful and are incum- (see Jansson 2003). Temperature adaptation is bent. Indeed, such preemptive interactions clearly involved in establishing the geograph- may be a significant factor in molding the ic ranges of many taxa, and as noted above, LDG to the gradient of those diversity-depen- changes in range end-points accompanying dent environmental conditions that correlate isothermal shifts are well documented from with latitude. The role played by extinction in midlatitudes, especially during glacial-inter- the dynamics of the LDG would thus be to re- glacial cycles. Our data are insufficient at pre- duce the incumbent fauna sufficiently below sent to study these relations in detail, but the its carrying capacity to provide ‘‘openings’’ patterns of midlatitude thermal variability for the invasion of new species (Walker and (which coincide with fluctuations in many ad- Valentine 1984), some of which would add ditional factors) and the patterns of extinction genera to the region. If a polar extinction spike clearly deserve further scrutiny. could be found, our hypothesis might be test- ed by whether there was a compensatory re- Summary and Conclusions placement of lost genera from the temperate Through much of the Neogene at least, spe- biota. At any rate, the lack of any clear differ- ciation has evidently been easier within low- ence between temperate and polar extinction latitude, high-diversity ecosystems than with- rates accords well with the dynamic outlined in high-latitude, low-diversity ecosystems, above. There is no evidence in the data re- creating an abundance of young lineages in ported here that the extinction rate directly af- low latitudes. This dynamic implies that re- fects carrying capacity or is the primary con- sources are more easily obtained by new spe- trol on the LDG, although it may push diver- cies in high-diversity than in low-diversity sity below regional carrying capacity and thus ecosystems, a counterintuitive observation, drive temporal or spatial divergence from an particularly in light of apparent resistance of idealized trend. tropical habitats to invasion from high lati- The hypothesis we develop here thus goes tudes over both the short and the long term. counter to classic suggestions that equator- Such a latitudinal trend in origination could ward range limits are set by biotic interactions be promoted by ecological feedbacks, such as and poleward limits are set by physical factors proposed by advocates of niche construction (e.g., Dobzhansky 1950; MacArthur 1972). In- (e.g., Odling-Smee et al. 2003; Erwin 2007), stead, we suggest that physical environmental but the question of why there should be a gra- factors determine regional differences in the dient in feedbacks remains. The rapid diver- niche breadth of taxa throughout the LDG, sity rebound in the New World Tropics follow- with resulting variations in biotic resistance to ing a late Neogene extinction pulse (see Todd invasion and local origination. The LDG, and et al. 2002; O’Dea et al. 2007) might have been by implication other large-scale spatial varia- partially fueled from higher latitudes, but the tions in diversity, is thus primarily an evolu- huge pool of tropical lineages available as di- tionary rather than an ecological phenome- versification sources would have swamped the non, albeit, as in most evolutionary cases, much smaller number of potential extratropi- driven by adaptations to ecological condi- 176 JAMES W. VALENTINE ET AL.

tions. We suggest that marine latitudinal gra- Crame, J. A., and B. R. Rosen. 2002. Cenozoic palaeogeography and the rise of modern biodiversity patterns. Geological So- dients in diversity-dependent ecological fac- ciety of London Special Publication 177:227–246. tors produce an OTT dynamic, with a low-lat- Crombie, A. D. 1945. On competition between different species itude source and higher-latitude filters, where of gramnivorous insects. Proceedings of the Royal Society of London B 132:362–395. the filtering processes are mediated by the in- Dobzhansky, Th. 1950. Evolution in the tropics. American Sci- cumbency of local populations that belong on entist 38:209–221. average to progressively older and ecological- Erwin, D. H. 2007. Disparity: morphologic pattern and devel- opmental context. Palaeontology 50:57–73. ly more generalized lineages at progressively Etter, R. J., and J. F. Grassle. 1992. Patterns of species diversity higher latitudes. Moreover, the role of biotic in the deep sea as a function of sediment particle size diver- interactions as proximate factors in structur- sity. Nature 360:576–578. ing spatial and temporal diversity patterns Fine, P. V. A. 2002. The invasibility of tropical forests by exotic plants. Journal of Tropical Ecology 18:687–705. will be severely underestimated if, as pro- Foote, M. 2001. Evolutionary rates and the age distributions of posed here, the nature of those interactions living and extinct taxa. Pp. 245–194 in J. B. C. Jackson, S. Lid- varies with physical environmental parame- gard, and F. K. McKinney, eds. Evolutionary patterns. Uni- versity of Chicago Press, Chicago. ters. Diversity-dependent factors do not only Fridley, J. D., J. J. Stachowicz, S. Naeem, D. F. Sax, E. W. Sea- come into play at the high end of the diversity bloom, M. D. Smith, T. J. Stohlgren, D. Tilman, and B. Von Holle. scale. Depending on regional history and en- 2007. The invasion paradox: reconciling pattern and process in species invasions. Ecology 88:3–17. vironmental context, we suspect that biotas at Gage, J. D., and J. A. Tyler. 1991. Deep-sea biology. Cambridge any diversity level can be at, or distant from, University Press, Cambridge. a state in which diversity dependence is a sig- Gaston, K. J., T. M. Blackburn, and J. L. Spicer. 1998. Rapoport’s rule: time for an epitaph? Trends in Ecology and Evolution nificant force. 13:70–74. Gladenkov, Y. B., and V. N. Sinelnikova. 1990. Miocene mollusks Acknowledgments and climate in Kamchatka. Trudy Geologicheskogo Instituta We thank James Crampton, Susan Kidwell, Akademiya Nauk SSSR 453:1–173. [In Russian.] Goldberg, E. E., K. Roy, R. Lande, and D. Jablonski. 2005. Di- and Wolfgang Kiessling for valuable reviews. versity, endemism and age distribution in macroevolutionary This work was supported by NASA (Exobiol- sources and sinks. American Naturalist 165:623–633. ogy/Evolutionary Biology) and a Faculty Re- Hewitt, C. L. 2002. Distribution and biodiversity of Australian tropical marine bioinvasions. Pacific Science 56:213–222. search Grant from the University of Califor- Hickerson, M. J., and C. W. Cunningham. 2006. Nearshore fish nia, Berkeley. (Pholis gunnellus) persists across the North Atlantic through multiple glacial episodes. Molecular Ecology 15:4095–4107. Literature Cited Jablonski, D. 1993. The tropics as a source of evolutionary nov- elty: the post-Palaeozoic fossil record of marine invertebrates. Allison, P. A., and D. E. G. Briggs. 1993. Paleolatitudinal sam- Nature 364:142–144. pling bias, Phanerozoic species diversity, and the end-Perm- ———. 2005. Evolutionary innovations in the fossil record: the ian extinction. Geology 21:65–68. intersection of ecology, development and macroevolution. Arim, M., S. R. Abades, P. E. Neill, M. Lima, and P. A. Marquet. Journal of Experimental Zoology B 304:504–519. 2006. Spread dynamics of invasive species. Proceedings of the Jablonski, D., and R. A. Lutz. 1983. Larval ecology of marine National Academy of Sciences USA 103:374–378. benthic invertebrates: paleobiological implications. Biological Armbrecht, I., I. Pefecto, and J. Vandermeer. 2004. Enigmatic Reviews 58:21–89. biodiversity correlations: ant diversity responds to diverse re- Jablonski, D., and J. J. Sepkoski Jr. 1996. Paleobiology, commu- sources. Science 304:284–286. nity ecology, and scales of ecological pattern. Ecology 77: Beard, K. C. 1998. East of Eden: Asia as an important center of taxonomic origination in mammalian evolution. Bulletin of 1367–1378. the Carnegie Museum of Natural History 34:5–39. Jablonski, D., K. Roy, and J. W. Valentine. 2006. Out of the trop- Clark, A., and J. A. Crame. 2003. The importance of historical ics: evolutionary dynamics of the latitudinal diversity gradi- processes in global patterns of diversity. Pp. 130–151 in T. M. ent. Science 314:102–106. Blackburn and K. J. Gaston, eds. Macroecology: concepts and Jackson, J. B. C., and K. G. Johnson. 2001. Measuring past bio- consequences. Blackwell Science, Oxford. diversity. Science 293:2401–2404. Connolly, S. R., D. R. Bellwood, and T. P. Hughes. 2003. Indo- Jakobsson, M., J. Backman, B. Rudels, J. Mycander, M. Frank, L. Pacific biodiversity of coral reefs: deviations from a mid-do- Mayer, W. Jokat, F. Sangiorgi, M. O’Regan, H. Brinkhuis, J. main model. Ecology 84:2178–2190. King, and K. Moran. 2007. The early Miocene onset of a ven- Crame, J. A. 2000. Evolution of taxonomic diversity gradients in tilated circulation regime in the Arctic Ocean. Nature 447: the marine realm: evidence from the composition of Recent 986–990. bivalve faunas. Paleobiology 26:188–214. Jansson, R. W. 2003. Global patterns in endemism explained by ———. 2001. Taxonomic diversity gradients through geological past climatic change. Proceedings of the Royal Society of Lon- time. Diversity and Distributions 7:175–189. don B 270:583–590. ———. 2002. Evolution of taxonomic diversity gradients in the Jones K. E., W. Sechrest, and J. L. Gittleman. 2005. Age and area marine realm: a comparison of Late Jurassic and Recent bi- revisited: identifying global patterns and implications for valve faunas. Paleobiology 28:184–207. conservation. Pp. 141–165 in A. Purvis, J. L. Gittleman, and T. INCUMBENCY, DIVERSITY, LATITUDE 177

Brooks, eds. Phylogeny and conservation. Cambridge Uni- scale biogeographic patterns in marine mollusks: a confluence versity Press, Cambridge. of history and productivity? Ecology 86:2288–2297. Kendrick, G. W., K. H. Wyrwoll, and B. J. Szabo. 1991. Pliocene- Ribas, C. R., and J. H. Schoereder. 2006. Is the Rapoport effect Pleistocene events and history along the western margin of widespread? Null models revisited. Global Ecology and Bio- Australia. Quaternary Science Reviews 10:419–439. geography 15:614–624. Kitamura, A., H. Omote, and M. Oda. 2000. Molluscan response Roberts, C. M., C. J. McClean, J. E. N. Veron, J. P. Hawkins, G. to early Pleistocene rapid warming in the Sea of Japan. Ge- R. Allen, D. E. McAllister, C. G. Mittermeier, F. W. Schueler, ology 28:723–726. M. Spalding, F. Wells, C. Vynne, and T. B. Werner. 2002. Ma- Knudsen, J. 1979. Deep-sea bivalves. Pp. 195–224 in S. van der rine biodiversity hotspots and conservation priorities for Spoel, A. C. van Bruggen, and J. Lever, eds. Pathways in mal- tropical reefs. Science 295:1280–1284. acology. Bohn, Scheltema and Holkema, Utrecht. Rosenzweig, M. L. 1995. Species diversity in time and space. Kreft, H., and W. Jetz. 2007. Global patterns and determinants Cambridge University Press, Cambridge. of vascular plant diversity. Proceedings of the National Acad- Rosenzweig, M. L., and R. D. McCord. 1991. Incumbent replace- emy of Sciences USA 104:5925–5930. ment: evidence for long-term evolutionary progress. Paleo- Krug, A., D. Jablonski, K. Roy, and J. W. Valentine. 2006. Con- biology 17:202–213. trasting bivalve latitudinal diversity gradients shaped by Ce- Roy, K., and E. E. Goldberg. 2007. Origination, extinction and nozoic origination patterns. Geological Society of America dispersal: integrative models for understanding present-day Abstracts with Programs 38(7):169. diversity gradients. American Naturalist 170(Suppl.):S71– Krug, A. Z., D. Jablonski, and J. W. Valentine. 2008. Species-ge- S85. nus ratios reflect a global history of diversification and range Roy, K., D. Jablonski, and J. W. Valentine. 1994. Eastern Pacific expansion in marine bivalves. Proceedings of the Royal So- molluscan provinces and latitudinal diversity gradient: no ev- ciety of London B (published online 13 February 2008). idence for ‘‘Rapoport’s rule.’’ Proceedings of the National Laptikhovsky, V. 2006. Latitudinal and bathymetric trends in Academy of Sciences USA 91:8871–8874. egg size variation: a new look at Thorson’s and Rass’s rules. ———. 1995. Thermally anomalous assemblages revisited: pat- Marine Ecology 27:7–14. terns in the extraprovincial latitudinal range shifts of Pleis- Lomolino, M. V., B. R. Riddle, and J. H. Brown. 2006. Biogeog- tocene marine mollusks. Geology 23:1071–1074. raphy, 3d ed. Sinauer, Sunderland, Mass. ———. 2000. Higher taxa in biodiversity studies: patterns from Longhurst, A. 1998. Ecological geography of the sea. Academic eastern Pacific marine mollusks. Philosophical Transactions Press, San Diego. of the Royal Society of London B 351:1605–1613. MacArthur, R. H. 1972. Geographical ecology. Harper and Row, Roy, K., D. Jablonski, J. W. Valentine, and G. Rosenberg. 1998. New York. Marine latitudinal diversity gradients: tests of causal hypoth- Martin, P. R., F. Bonier, and J. J. Tewksbury. 2007. Revisiting Ja- eses. Proceedings of the National Academy of Sciences USA blonski (1993): cladogenesis and range expansion explain lat- 95:3699–3702. Ruggiero, A., and V. Werenkraut. 2007. One-dimensional anal- itudinal variation in taxonomic richness. Journal of Evolution- yses of Rapoport’s Rule reviewed through meta-analysis. ary Biology 20:930–936. Global Ecology and Biogeography 16:401–414. Miller, A. I. 1997. A new look at age and area: the geographic Sax, D. F. 2001. Latitudinal gradients and geographic ranges of and environmental expansion of genera during the Ordovi- exotic species: implications for biogeography. Journal of Bio- cian Radiation. Paleobiology 23:410–419. geography 28:139–150. Mittelbach, G. G., D. W. Schemske, H. V. Cornell, A. P. Allen, J. Sax, D. F., and S. D. Gaines. 2006. The biogeography of natural- M. Brown, M. B. Bush, S. P. Harrison, A. H. Hurlbert, N. ized species and the species-area relationship: reciprocal in- Knowlton, H. A. Lessios, C. M. McCain, A. R. McCune, L. A. sights to biogeography and invasion biology. Pp. 449–480 in McDade, M. A. McPeek, T. J. Near, T. D. Price, R. E. Ricklefs, M. W. Cadotte, S. M. McMahon, and T. Fukami, eds. Concep- K. Roy, D. F. Sax, D. Schluter, J. M. Sobel, and M. Turelli. 2007. tual ecology and invasions biology: reciprocal approaches to Evolution and the latitudinal diversity gradient: speciation, nature. Kluwer, Dordrecht, The Netherlands. extinction and biogeography. Ecology Letters 10:315–331. Sax, D. F., J. J. Stachowicz, J. H. Brown, J. F.Bruno, M. N. Dawson, O’Dea, A., J. B. C. Jackson, H. Fortunato, J. T. Smith, L. D’Croz, S. D. Gaines, R. K. Grosberg, A. Hastings, R. D. Holt, M. M. K. G. Johnson, and J. A. Todd. 2007. Environmental change Mayfield, M. I. O’Connor, and W. R. Rice. 2007. Ecological and preceded Caribbean extinction by 2 million years. Proceed- evolutionary insights from species invasions. Trends in Ecol- ings of the National Academy of Sciences USA 104:5501–5506. ogy and Evolution 22:465–471. Odling-Smee, F. J., K. N. Laland, and M. W. Feldman. 2003. Schluter, D. 2000. The ecology of adaptive radiation. Oxford Niche construction: the neglected process in evolution. University Press, Oxford. Princeton University Press, Princeton, N.J. Smith, H. S. 1935. The role of biotic factors in the determination Ogasawara, K. 1994. Neogene paleogeography and marine cli- of population densities. Journal of Economic Entomology 28: mate of the Japanese Islands based on shallow-marine mol- 873–898. luscs. Palaeogeography, Palaeoclimatology, Palaeoecology Stachowicz, J. J., and D. Tilman. 2005. Species invasions and the 108:335–351. relationships between species diversity, community satura- Paulay, G., L. Kirkendale, G. Lambert, and C. Meyer. 2002. An- tion, and ecosystem functioning. Pp. 41–64 in D. F. Sax, J. S. thropogenic biotic interchange in a coral reef ecosystem: a Stachowicz, and S. D. Gaines, eds. Species invasion: insights case study from Guam. Pacific Science 56:403–422. into ecology, evolution and biogeography. Sinauer, Sunder- Raffi, S., S. M. Stanley, and R. Marasti. 1985. Biogeographic pat- land, Mass. terns and Plio-Pleistocene extinction of Bivalvia in the Med- Stevens, G. C. 1989. The latitudinal gradient in geographic iterranean and southern North Sea. Paleobiology 11:368–388. range: how so many species co-exist in the tropics. American Rejmańek, M. 1996. Species richness and resistance to invasions. Naturalist 133:240–256. Pp. 153–172 in G. Orians, R. Dirzo, and J. H. Cushman, eds. Tankard, A. J. 1975. Thermally anomalous Late Pleistocene mol- Diversity and processes in tropical forest ecosystems. Spring- luscs from the south-western Cape Province, South Africa. er, Berlin. Annals of the South African Museum 69:17–45. Rex, M. A., J. A. Crame, C. T. Stuart, and A. Clarke. 2005. Large- Taylor, J. D., and C. N. Taylor. 1977. Latitudinal distribution of 178 JAMES W. VALENTINE ET AL.

predatory gastropods on the Eastern Atlantic shelf. Journal of Vermeij, G. J. 1978. Biogeography and adaptation. Harvard Uni- Biogeography 4:73–81. versity Press, Cambridge. Thorsen, G. 1950. Reproductive and larval ecology of marine ———. 1991. When biotas meet: understanding biotic inter- bottom invertebrates. Biological Reviews 25:1–45. change. Science 253:1099–1104. Todd, J. A., J. B. C. Jackson, K. G. Johnson, H. M. Fortunato, A. ———. 2005a. From phenomenology to first principles: toward Heitz, F. Alvarez, A. Marcos, and P. Jung. 2002. The ecology a theory of diversity. Proceedings of the California Academy of extinction: molluscan feeding and faunal turnover in the of Sciences, series 4, 56(Suppl. I, No. 2):12–23. Caribbean Neogene. Proceedings of the Royal Society of Lon- ———. 2005b. Invasion as expectation: a historical fact of life. Pp. 315–339 in D. F. Sax, J. S. Stachowicz, and S. D. Gaines, eds. don B 269:571–578. Species invasion: insights into ecology, evolution and bioge- Valentine, J. W. 1961. Paleoecologic molluscan geography of the ography. Sinauer, Sunderland, Mass. Californian Pleistocene. University of California Publications Walker, T. D., and J. W. Valentine, 1984. Equilibrium models of in Geological Science 34:309–442. evolutionary species diversity and the number of empty nich- ———. 1972. Conceptual models of ecosystem evolution. Pp. es. American Naturalist 124:887–899. 192–215 in T. J. M. Schopf, ed. Models in paleobiology. Free- Ward, L. W. 1992. Molluscan biostratigraphy of the Miocene man, Cooper, San Francisco. Middle Atlantic coastal plain of North America. Virginia Mu- ———. 1973. Evolutionary paleoecology of the marine bio- seum of Natural History Memoir No. 2. sphere. Prentice-Hall, Englewood Cliffs, N.J. Webb, T. J., and K. J. Gaston. 2000. Geographic range size and ———. 1983. Seasonality: effects in marine benthic communi- evolutionary age in birds. Proceedings of the Royal Society of ties. Pp. 121–156 in M. J. S. Tevesz and P. L. McCall, eds. Biotic London B 267:1843–1850. interactions in recent and fossil benthic communities. Ple- Wiens, J. J., M. C. Brandley, and T. W. Reeder. 2006. Why does a num, New York. trait evolve multiple times within a clade? Repeated evolution Valentine, J. W., and F. J. Ayala. 1978. Adaptive strategies in the of snakelike body form in squamate reptiles. Evolution 60:13– 141. sea. Pp. 323–345 in B. Battaglia and J. A. Beardmore, eds. Ma- Wier, J. T, and D. Schluter. 2007. The latitudinal gradient in re- rine organisms: genetics, ecology, and evolution. Plenum, cent speciation and extinction rates of birds and mammals. New York. Science 315:1574–1576. Valentine, J. W., K. Roy, and D. Jablonski, 2002. Carnivore/non- Willig, M. R., D. M. Kaufman, and R. D. Stevens. 2003. Latitu- carnivore ratios in northeastern Pacific marine gastropods. dinal gradients of biodiversity: pattern, process, scale, and Marine Ecology Progress Series 228:153–163. synthesis. Annual Review of Ecology, Evolution, and System- Valentine, J. W., D. Jablonski, S. M. Kidwell, and K. Roy. 2006. atics 34:273–309. Assessing the fidelity of the fossil record by using marine bi- Willis, J. C. 1922. Age and area: a study in geographical distri- valves. Proceedings of the National Academy of Sciences USA bution and origin of species. Cambridge University Press, 103:6599–6604. Cambridge.