Tải bản đầy đủ - 0trang
Chapter 61. Geographical isolation in marine species: Evolution and speciation in Ostracoda, I
812 T.M. CRONIN
A PROSPECTUS FOR EVOLUTIONARY
STUDIES IN OSTRACODA
This paper presents the background and initial results of a series of studies specificallydesigned
to investigate speciation and morphologic stasis in marine Ostracoda. My ultimate objectives are
1) to assess the geographical classification of speciation by determining the relative frequency of
types of allopatric, sympatric and parapatric speciation, 2) to estimate the relative frequency and
circumstances of rapid and gradual morphological change, and 3) to assess the role of climate and
tectonics in causing geographical isolation and speciation. The decision to conduct these studies
stems from several trends emerging in evolutionary palaeontology. First, there are few case studies
using fossil sequences that adequately address the question of speciation (Gingerich, 1985, Table 2).
Of greater concern are questions about paleontologists’ ability to identify species in fossil material
(Schopf, 1980, 1982). It may be that Schopf’s concern about fossil species reflects the approaches
adopted by paleontologists, which may not be appropriate for answering current evolutionary
questions, and not the fossil record itself. With few exceptions, paleontological studies of speciation have unnecessarily overemphasized stratigraphic completeness and high sampling resolution
(Gingerich, 1985). Clearly, no stratigraphical sequence is complete, an idea known long before the
comprehensive studies of Schindel (1980, 1982) and Dingus and Sadler (1982). Despite assertions
by theorists that geographical coverage is of paramount importance (Eldredge and Gould, 1972;
Schopf, 1982; Valentine and Jablonski, 1983), the apparent reluctance by paleontologists to delineate biogeographical patterns of speciation using fossils has been a major flaw in methodology.
A stratophenetical approach is important, but in order to critically assess phylogenetic hypotheses,
the geographical scope of a study should also be stated in the context of the paleogeography of the
taxon (see Young, 1984). Climatically-induced shifts in species’ biogeography,which form the
empirical data for many paleoclimatologists, suggests that migration rather than evolutionary origination, accounts for the first stratigraphical appearances of most species in rock sections. It should
therefore be stressed at the outset that comprehensive geographical coverage deserves as much
priority as stratigraphical sampling interval.
Another topic is the question of abiotic. extrinsic factors and their effect on evolution. Widely
divergent schools of evolutionary thinking hold that geographical isolation caused by climatic and
tectonic events plays an important role in speciation. Some models of genetic changes during
speciation postulate extrinsic events to initiate speciation (Carson, 1982). In advocating a cladistic
approach, Eldredge and Cracraft (1980, p. 121) pointed out that isolation by geographical barriers
is a mechanism for the disruption of “within species patterns of parental ancestry” but that isolation is not prima facie evidence for two species. Valentine and Jablonski (1983) discussed modes
of speciation by small founder populations, large vicariant populations. and in clines and offered
hypotheses about which mode might be more common in groups with various reproductive strategies and biogeographies.
Mayr (1982) has maintained that peripatric speciation is the most ccmmon type, but that speciation does not necessarily result from all founder events. However, we still cannot answer the question of which environmental events are more likely to result in speciation, which in extinction, and
which in stasis. Is it true that a smaller population, when isolated, is more likely to “pass through
the bottleneck of deleterious heterozygosity and reorganize itself genetically” (Mayr, 1982, p. 6)?
If so, how can we identify such events in fossils and associate them with documented environmental
changes? If sympatric and parapatric speciation are found to be common, do they support genetic
models of speciation (Templeton, 1980) and relegate geography to a secondary role?
Climatic change represents an obvious candidate to explain evolutionary trends and has frequen-
Geographical Isolation in Marine Ostracoda 873
tly been invoked as such (Cracraft 1982; Stanley, 1984; Valentine, 1984; Vrba, 1985). Indeed, it is
almost always possible for a paleontologist to find a climatic event in the geological record to
“associate” with a faunal or floral diversification or extinction. Yet simple age equivalence of a
faunal and a climatic event does not imply a causal relationship. More rigorous testing of suspected
abiotic influences on evolution is possible and necessary (Cronin, 1985 and below).
A third reason behind these studies is the unrealized potential of Ostracoda for understanding
evolutionary processes. Ostracods have ideal characteristics for evolutionary research - sexual
reproduction, abundantly fossilized carapaces with easily identified homologous features for
morphometric study, and growth through moulting yielding a clear ontogeny (Cronin, 1985; Reyment, 1985). Many problems associated with assumptions of the “paleontologic approach” (see
Eldredge and Novacek, 1985, p. 69,70) are minimized by using ostracods. In addition to distinguishing ontogenetic and sexual dimorphism from other morphological variation, the ostracod carapace affords an opportunity to study the exoskeleton of the whole organism, not just a single part,
such as the vertebrate tooth. Mention should also be given to important new studies, among them
papers on ostracod histology and morphology (Okada, 1982), the relation between ecology, reproduction and morphology (Kamiya, this volume), and various aspects of polymorphism (Reyment, 1985; Abe, 1983; Ikeya and Ueda, this volume), that together boost our confidence that
biological species can be recognized from carapace morphology. In an important paper, Kamiya
(this volume) has demonstrated how distinct carapace shapes in two related extant species of Loxoconcha directly reflect different copulatory positions in different microhabitats. One species
is phytal in habitat, and requires a rounded posterior shape and circular-oval lateral shape to
copulate on the plants; the other species is flattened posteriorly and rectangular laterally to copulate
on the sand bottom. Therefore, not only does carapace shape distinguish two species but it is functionally related to the primary means of distinguishing two species-the criterion of reproductive
While some authors have emphasized species diversity and overall faunal patterns to test evolutionary laws (Hoffman and Kitchell, 1984), my approach focuses on species and monophyletic
groups for which there is strong evidence for a common ancestor. I agree with Eldredge and Novacek (1985) that cladistic and paleontological methodologies of phylogentic analysis differ from
one another more in style than in substance, and that both are confronted with similar subjective
decisions. In my ostracod studies, relationships among species are known with varying degrees of
confidence that will be stated in each case. While subjectivity and judgement will always play a role,
testing genealogical hypotheses is treated as an iterative process, in which securing new modern
and fossil material confirms or refutes suspected phylogenies.
Three points about the approach should be stressed: [l] As stated explicitly by more and more
evolutionists (Stenseth and Smith, 1984; Vrba, 1984), the fossil record provides a unique, albeit
imperfect, opportunity to test hypotheses about the influence of environment on organisms. Each
study in this series of papers represents an effort to understand the evolutionary history of a
species or monophyletic group of species, selected on the basis of their ecology, biogeography
and exposure to known climatic or tectonic changes. One school of vicariance biogeography has
criticized the role of fossils in biogeography and systematics, stating “recent distributions . . .
provide the only unequivocal data of biogeography” (Patterson, 1981, p. 464). To test vicariance
versus dispersal explanations of why a taxon lives in an area, some propose seeking concordant
biogeographical trends in separate lineages subjected to the same vicariant events (Rosen, 1978).
However, different taxa will not necessarily respond in similar ways to the same extrinsic event
(see Endler, 1982; Young, 1984). Although methodologically sound in trying to falsify hypothesized
genealogies, this vicariant-cladistic approach tends to ignore valuable geologic evidence for
The fossil record itself can be used both to test phylogenetic hypotheses (Young, 1984), and,
more importantly, hypotheses about the causal relationships between organisms and their environments. In one example, I documented a period of speciation in the ostracod Puriunu, which I hypothesized was caused by gradual climatic and oceanographical change along the eastern United
States about 4 to 3 million years ago (mya) (Cronin, 1985). This idea was tested in three ways: 1)
by examining Puriuna’s response to cyclic climatic change, considered by palaeoclimatologists to be
qualitatively distinct from gradual climatic change (no speciation events were found, Cronin, 1985);
2) by examining the entire endemic ostracod fauna off the eastern United States for evolutionary
first appearances during the last 5 million years to see if other genera diversified when Puriana
did (about 50 % out of 127 ostracod species originated during this time, Cronin, in press a); 3) studies
have been started on ostracods from the western north Pacific near Japan where a distinct ostracod
fauna was subjected to similar climatic changes in temperate and subtropical climatic zones to those
in the western North Atlantic. In summary, for some extant marine groups having a fossil record,
paleobiogeography is known as well as modern distributions so that both hypotheses about genealogical relationships and hypotheses about the causal relationship between speciation (vicariant
and dispersal) and extrinsic events can be tested and falsified.
 I do not share Schopf’s (1982) pessimism about identifying biological species from fossilized
hard parts. With the exception of sibling species, distinguishing among species of ostracods is
possible as demonstrated by a growing literature (e.g., Cronin, 1985, and below; Reyment,
1982, 1985; Abe, this volume; Ikeya and Tsukagoshi, this volume) provided one has adequate
sample size and geographical coverage.  I place emphasis on intraspecific variability, particularly
by comparisons among completely or partially isolated populations. Unless morphological
variability within a species is understood over its geographical range, phylogenetic relationships
among related species cannot be determined.
A wide range of environmental changes will be considered in these studies including: (1) cyclic
and long-term, gradual climatic changes, (2) sea level changes, (3) vicariant geographic isolation
of large populations, (4) founder events isolating small populations, ( 5 ) relict populations and
others. The first set of studies will focus on shallow water marine taxa, and they will hopefully
expand to deep water marine and fresh water forms in the future.
This first study, appropriately published in the Proceedings of the Ninth International Ostracod
Symposium, whose theme was evolutionary biology, examines two major categories of geographical
isolation. First, isolation during the Cenozoic of small populations .of shallow water species on
remote Pacific atolls and islands. In theory, such isolation could lead to peripatric speciation
(sensu Mayr, 1982), which is equivalent to type Ib speciation of Bush (1975). Conversely, classic
dumbbell isolation of large populations by the development the Isthmus of Panama about 3 mya
might theoretically lead to allopatric speciation, type la of Bush (1975). This important distinction between types of allopatry was emphasised by Mayr (1982) because of his hypothesized
inverse relationship between population size and rate of divergence.
Geographical Isolation in Marine Ostracoda 875
Benthonic marine ostracods have no planktotrophic larval stage and therefore dispersal of
shallow water species across deep water is almost certainly passive. This situation is similar to that
of a terrestrial organism passively dispersed to small islands, but differs from that in some marine
gastropods having larval dispersal and gene flow among dispersed populations (Scheltema, 1971).
Study of the relationship between Atlantic/Caribbean and Pacific marine ostracods reveals interesting examples of taxa that disperse over long distances and show very little intraspecific variability
in carapace morphology among widely separated populations. However, taxonomic studies of
faunas from different regions have frequently not recognised the strong affinities of isolated populations. This may reflect an unfamiliarity with the literature from other regions or unstated assumptions by taxonomists about the endemic nature of shallow water ostracods. The evidence
presented below shows the conspecific nature of many small, widely dispersed populations of a
Herrnanites transoceanica Teeter, 1975
Teeter (1975) described the extant species Hermanites transoceanica from the Belize carbonate
platform near eastern Central America, and noted its wide distribution in the Pacific off New
Guinea and Hawaii. Holden (1976) identified specimens of the same species from the Pleistocene
of Midway Island as Jugosocythereis luctea (Brady, 1866), and Hartmann (1981) described the new
extant species Quadracythere insulardeaensis from eastern Australia based on specimens that are
also conspecific with Teeter’s species. Bonaduce et al., (1980) described Quudrucythere auricolatu
1-LOCALITIESOF Hermanites transoceanica TEETER,1975t
1. Eastern Australia (Hartmann. 1981)
2. Rangiroa Atoll (Tuamoto Archipelago) (Hartmann, 1984)
3. Kwajalein Atoll (Marshall Islands), this paper
4. Okinawa Island (Ryukyu Islands), this paper
5. Enewetak Atoll (Marshall Islands), this paper
6. Rongelap Atoll (Marshall Islands), this paper
7. Ponape, this paper ,
8. Bikini Atoll (Marshall Islands), this paper
9. Guam (Mariana Islands), this paper
10. Guadalcanal (Solomon Islands), this paper
11. Shortland Island (Solomon Islands), (Titterton, 1985)
12. New Guinea, (Teeter, 1975)
13. Belize (Central America), (Teeter, 1975)
14. Bahamas, this paper
15. Virgin Islands (St. Thomas), this paper
16. Vera Cruz (Mexico), (Krutak, 1982)
17. Midway Island (Hawaiian Islands), (Holden, 1976) Pleistocene
18. Taiwan (Hu, 1981) Pliocene-early Pleistocene
19. Enewetak Atoll, Marshall Islands, this paper, Miocene-Pleistocene
20. #Pinaki and Niau, (Tuamoto Islands), this paper
21. #Pago Pago, (Samoa Islands), this paper
22. #Gulf of Aqaba, Red Sea, (Bonaduce et a!., 1980)
t : Numbers refer to map in Text-fig. 1. ## : Not shown in Text-@. 1. All samples are modern except 17,
18 and 19. Locality data-can be obtained from the author.
876 T.M. CRONIN
TEXT-FIG.1-Localities yielding Hermunites trunscoceunicuTeeter, 1975. Fossil samples are indicated by open circles, modem samples by solid circles (See Table 1 for list of localities). Since this paper was first written, this
species has also been identified in samples from Pinaki and Niau, Tuamoto Islands, South Pacific, and from
Pago Pago, American Samoa, from Truk, Majuro and Arno in Micronesia, and in the paper by Bonaduce et
ul., (1980) on ostracodes from the Gulf of Aqaba, Red Sea.
from the Gulf of Aqaba in the Red Sea for a species that is conspecific with H. transoceanicu.
Similarly, Hu (198 1) described the species Rudimella microreticulata from the Maanshan Mudstone
of southern Taiwan, and although citing Teeter (1975), Hu did not identify these Asian forms as
conspecific with H. trunsoceanica, although they are clearly so.
To study H. trunsoceunica in detail, I assembled fossil and modern collections from shallow
carbonate environments in the western equatorial Pacific and Caribbean and, where possible,
examined material from published studies and theses. Selected localities yielding H. trunsoceunica
are shown in Text-fig. 1 and are listed in Table 1. Text-fig. 2 plots carapace length versus height
for measured specimens and plots length and height of holotypes using measurements given by
Teeter (1975) and for Australian material from Hartmann (1981). As suggested by Teeter (1975),
H. transoceanica includes widely dispersed populations from a single, morphologically well-defined
species. The length/height plot shows the range of carapace sizes for fossil specimens (MiocenePleistocene) from the Marshall Islands is about the same as that from recent populations from the
same region. Measurements taken from Holden (1976) for fossils from Midway, and from Hartmann (1984) for modern material from Polynesia show specimens slightly larger, but not outside the
expected range for a species. Sexual dimorphism is weak in this species and the few specimens believed to be males are indicated. Carapace ornamentation shown in Plate 1 reveals a striking degree
of morphological stability over time (at least 6 million years) and geographical area. The fossil record
of this species shows that it lived in the Marshall Islands from the late Miocene until modern times,
on Taiwan during the late Pliocene or early Pleistocene, and on Midway during the Pleistocene.
Geographical Isolation in Marine Ostracoda 811
MARSHALL ISLANDS fossil
BELIZE. CENTRAL AMERICA
SOLOMON ISLANDS, AUSTRALIA
HAWAIIAN ISLANDS fosd
LENGTH I MICRONS)
2-Plot of carapace length versus height of Hermanites transoceanica Teeter, 1975. Specimenspresumed
to be male are indicated with the male symbol.Measurements of the holotype of H. transoceanica from Belize
were taken from Teeter (1975). Hartmann (1981) gives ranges of 500-530 microns and 280-290 microns as
measurements for length and height for Quadracyfhere insufardeaensis from Australia, and the midpoint of
these values was plotted.
The presence of this species in the Carribbean suggests it probably lived in this region before
about 3.0 mya when the formation of the Isthmus of Panama separated Atlantic and Pacific
populations. Specimens from Belize, the Virgin Islands and the Bahamas (Plate 1) cannot be distinguished from Pacific forms. Its presence in the Red Sea, however, leaves the remote possibility of
westward migration across the Atlantic during the last few million years.
Populations of H. trunsoceunicu were compared with two sympatric species, H . purvilobu (Hu,
1981), believecfto be a separate lineage, and H . mooneyi n. sp., the only known descendant species
evolving from H. trunsoceunica. Text-figure 3 plots length versus height for these three speciesshowing the similar dimensions of H. trunsoceunicu and H. mooneyi. However, Plate1 shows H. mooneyi
differing significantly from H. trunsoceunica in the development of surface ridges and reticulation.
The results for H. trunsoceunicu show that at least 10 times during the Neogene, and probably
dozens more, populations became established off remote islands or in isolated atolls, but maintained
morphological stability for several hundred thousand to several million years. Since the Miocene,
only one species has evolved from ancestral H . trunsoceunicu. Although short-lived species may
be as yet undiscovered, the evidence from well-studied fossil faunas from Midway (Holden, 1976),
Enewetak (Cronin, unpublished), Taiwan (Hu, 1984 and references), and Okinawa (Nohara and
Tabuki, 1985) shows no evidence for any. If new species had evolved during the Pliocene or
Pleistocene, one would expect to find living representatives in the many well-known tropical
Geographical Isolation in Marine Ostracoda 819
of carapace length and height for three species of Hermanites. The plots of H. transoceanica
are the same as in TEXT-FIG.
2. See text for discussion.
It is appropriate to consider mechanisms to account for the dispersal of H. trunsoceunicu.
This species could not have actively migrated across deep water barriers as it does not have a
planktotrophic larval stage. Dispersal was almost certainly passive and there are several possible
mechanisms which are summarized by Teeter (1 973). The encysted (double-walled) eggs of fresh
water ostracods can withstand dessication and therefore, have probably been dispersed by
migratory birds, either on their feet, trapped in feathers or in the intestinal tract (Sandberg, 1964).
1-8. Hermanites transoceanica Teeter, 1975. 1. Lateral view, left valve, female carapace (USNM
401834, Holocene, Guam). x 99. 2. Lateral view, left valve, female (USNM 401835, Holocene, St. Thomas,
Virgin Islands). x 99. 3. Lateral view, right valve, female (USNM 401836, Holocene, Rongelap Atoll, Marshall
Islands). x99. 4. Lateral view, left valve, female carapace (USNM 401837, Holocene, Acklins Island,
Bahamas). x99. 5. Lateral view, left valve, female carapace (USNM 401838, Holocene, Enewetak Atoll,
Marshall Islands). x99. 6. Lateral view, left valve, female (USNM 401839, Holocene, Bikini Atoll, Marshall
Islands). x99. 7. Lateral view, left valve, female (USNM 401840, late Pliocene, Enewetak Atoll, Marshall
Islands). x99. 8. Lateral view, left valve, female carapace (USNM 401841, middle to late Miocene,
Enewetak Atoll, Marshall Islands). X 99.
Fig. 9. Hermanites parviloba (Hu,1981). Lateral view, left valve, female (USNM 401842, early Pleistocene,
Enewetak Atoll, Marshall Islands). x 99.
Figs. 10-12-Hermunites mooneyi Cronin n. sp. 10. Lateral view, left valve,female (Holotype, USNM 401843,
Pliocene, Enewetak Atoll, Marshall Islands). x 99. 11. Lateral view, right valve, female carapace (Paratype,
USNM 401844, Pliocene, Enewetak Atoll, Marshall Islands). x 99. 12. Lateral view, left valve, ?male
(Paratype, USNM 401845, late Pleistocene, Enewetak Atoll, Marshall Islands). X 99.
Teeter (1973) considers avian transport unlikely for marine species because (1) eggs of marine
species are single-walled and are not known to withstand dessication; (2) many species live in the
subtidal zone and have a very small probability of contact with birds; (3) those tropical and
subtropical ostracod species that are geographically widespread are latitudinally restricted over
broad oceanic areas that are not related to north/south continental pathways of migratory birds.
Teeter (1973) similarly dismisses wind and fish as agents for widespread dispersal because of the
vulnerability of the single-walled egg to destruction.
The possibility that transoceanic shipping is the cause of widespread distributions of some
species has been discussed (Teeter, 1973), but an increasing number of species now have fossil
records back to the Miocene and Pliocene showing they were widely dispersed millions of years ago.
Teeter considers drifting on aquatic plants the most probable means of dispersal and notes the
occurrence of living ostracods on the marine alga Turbinaria, which lives near tropical coasts, and
on Sargassum near the Florida Keys. Sohn (1954) also reported benthonic marine ostracods living
on Sargassum that had drifted near Cape Cod, Massachusetts. Surface water currents therefore
could easily carry living ostracods on algae to new habitats on atolls and tropical islands. Another
passive mechanism is transport on drifting pumice rafts (Newton and Bottjer, 1985), but there are
no data on ostracods living on floating pumice. At present it is unclear how frequent these dispersal
events were and whether they were frequent enough to postulate continued, albeit occasional,
genetic interchange among seemingly isolated populations. In my opinioil, the probabilities are low
that significent genetic interchange occurred between Pacific atoll populations over the millions of
years of the species history.
In summary, if the taxonomic literature from the last decade were taken at face value, populations of H . transoceanica would be recorded as five species in four separate genera, Radimella,
Jugosocythereis, Hermanites, and Quadracythere. Actually, intraspecific stasis, both in the sense of
species integrity over geographic range (Van Valen, 1982) and in terms of temporal stability (Eldredege and Gould, 1972) have characterised this single species despite many opportunities thought
to be conducive to lineage splitting.
BY THE ISTHMUS
During the middle Pliocene 4 to 3 mya, the Isthmus of Panama formed as a barrier preventing
interchange between Pacific and Atlantic/Caribbean marine faunas (Keigwin, 1978; Quinn and
Cronin, 1984). This event represents a classic dumbbell allopatric split of large populations of
tropical species. Indeed, Vrba (1985) has cited it as a potential natural experiment for testing the
punctuated equilibrium model and Valentine and Jablonski (1983). pointed out that identifying
gradual patterns of change resulting from vicariant events such as the Isthmus would strengthen
evidence for punctuational patterns found elsewhere.
Ostracods from the Caribbean and western Atlantic are better known than are eastern Pacific
faunas, but strong affinites between populations from the two sides have long been known. In a preliminary report, Swain et al. (1964) identified Orionina vaughani (an “Atlantic” species) from the
Gulf of California, but later Swain (1967) renamed it 0. pseudovaughani for Pacific populations.
In a remarkable coincidence, Pokorny (1970) and Hazel (1977) independently identified a unique
species of Caudites having asymmetrical ornamentation of left and right valves and both called it
asymmetricus (Hazel’s name was nomen nudem). Hazel’s material was from the Pleistocene of
South Carolina, Pokorny’s was Holocene from the Galapagos. Subsequently, Hazel (1983) formally
described the Atlantic form as C. paraasymmetricus citing slight differences in the location of the
caudal process and position of the carinae. In ostracod faunas from the Clipperton Islands (Allison
Geographical Isolation in Marine Ostracoda 881
and Holden, 1971) and the Galapagos (Bate et al., 1981) similarities with the Caribbean have also
been noted for species of Paracytheridea, Neocaudites, Cytherelloidea, and Caudites. Carreiio (1985)
described Miocene-Pliocene ostracods from Maria Madre Island, Mexico and also noted strong
affinites in species of Loxocorniculum, Cativella, Puriana, and Triebelina.
In the majority of cases, while acknowledging morphological affinities between Pacific and
Atlantic populations, most authors either described new species based on relatively minor differences or postulated possible post-Isthmus faunal interchange to explain the similarities. I call
this the “pacifica-atlantica” tendency and suspect it reflects a strong bias by many that geographical
isolation should lead to allopatric speciation. There seems to be a reluctance to call populations
from opposite sides conspecific. Yet on morphological grounds alone, splitter and lumper alike
would not hesitate to call two populations conspecific if they were not separated by a land barrier
and thousands of miles.
Resolving questions of species identification and rates of morphological divergences are extremely important. Consequently, to test whether many Pacific and Atlantic/Caribbean populations
are really conspecific, I collected or obtained from colleagues Neogene material from numerous
formations in southern California and Baja, Mexico to compare pre- and post-Isthmus populations of genera found on both sides of the Isthmus. In this way, the monophyletic relationships of
descendant populations for each lineage can been demonstrated by establishing that a single contiguous (presumed “panmictic”) population existed before the Isthmus formed. New Pacific material includes the Imperial Formation (Quinn and Cronin, 1984), a unit long known for containing
mollusc and other macroinvertebrates with “Caribbean” affinities, the Santa Barbara and Pic0
Formations (Cronin et al., 1983), and in Baja, the Loreto, Tirabuzon (Carreiio, 1981), Salada,
Ysidro, Boleo, and San Ignacio Formations. Material from 10 Caribbean formations and at least
20 more formations from the southeastern U.S. was examined. The following paragraphs briefly
summarize the results for key species whose populations were split by the Isthmus, but which in
most cases shown negligible morphological change during the last 3 mya.
Cativella navis Coryell and Fields, 1937: This species is known from late Miocene to Holocene
deposits in numerous localities in the Caribbean, Central America, Atlantic Coastal Plain and
Baja, Mexico (Bold, 1974). In addition to many published occurrences, Cativella navis was found
in the Tirabuzon and Loreto Formations and modern sediments off La Paz (Baja, Mexico), the
Gatun Formation (Panama), the subsurface Pliocene of south Florida, the “Ecphora” zone of the
Jackson Bluff Formation (Florida), the Bowden Formation (Jamaica) and the Raysor Formation
(South Carolina). It is characterised by its small size three longitudinal ridges, the ventral one
sometimes having perforations. The medial ridge sometimes consists of a row of tubercles and can
be slightly curved or straight (Pl. 2, fig. 8). This species is easily distinguished from C. pulleyi
Teeter, 1975 and C. semitranslucens Couch, 1949 by its size, shape and ornamentation.
From examination of intra- and inter-population variability in Atlantic, Caribbean and Pacific
forms, I consider the following forms conspecific with C. navis: C . dispar Hartmann, 1959; C . semitranslucens and C. unitaria of Carreiio (1985); C. sp. A of Valentine (1976); C . dispar and Costa?
variabilocostata seminuda of Swain (1967, P1. 3, figs. 2a,b, 14). Plate 2 illustrates 5 specimens that
show the similarity of widely distributed populations. Cativella navis therefore represents a cohesive
group of populations showing very slight morphological variation among them. Before the Ithsmus
of Panama formed, and during its formation, early-middle Pliocene populations lived in Baja,
Mexico (Loreto, Tirabuzon Formations) and the Tres Marias Islands on the Pacific side, in Central
America (Gatun Formation), and in the Atlantic/Caribbean/Gulf of Mexico (Bowden, Conception,
Aquequexquite Formations, Pinecrest Member of the Tamiami Formation, and “Ecphora” zone of
the Jackson Bluff Formation). Today, modern populations are still known from the Pacific in the
Gulf of California, off the west coast of Baja, and off El Salvador, and in the Caribbean (see