Printer Friendly

Shell microstructure of gastropods from Lake Tanganyika, Africa: adaptation, convergent evolution, and escalation.

Since missionaries and explorers sent back shell collections for the Royal Zoological Society of London in the 1850s, the endemic gastropod fauna of Lake Tanganyika (East Africa) has attracted considerable attention from biologists. Some of the first malacologists to describe the Tanganyikan gastropods placed them in marine gastropod families (Bourguignat 1890; Moore 1898, 1899, 1903). This mistake is understandable, for the convergence between Tanganyikan and marine gastropod shells is striking [ILLUSTRATION FOR FIGURE 1 OMITTED]. Like many marine gastropods, the Tanganyikan gastropods are often heavily calcified and many species display coarse noded ribbing, spines, apertural lip thickening and repair scars [ILLUSTRATION FOR FIGURE 1 OMITTED]. These features deviate markedly from the typically weakly calcified and unornamented shells of most lacustrine gastropods.

A number of workers have attempted to explain the convergence between Tanganyikan and marine gastropod shell morphologies. Moore (1898, 1899, 1903) speculated that a Jurassic seaway once linked Lake Tanganyika to the ocean. Thus, he maintained that the modern Tanganyikan faunas were marine relics thereby explaining their resemblance to marine faunas. However, our current understanding of the African Rift System is that the rift lakes are much younger than Jurassic (Miocene, 5-20 million years ago [M.Y.B.P.]) and were never connected to the ocean but rather were formed by the rifting of the African continent (Baker et al. 1972; Rosendahl et al. 1986; Ebinger 1989). Furthermore, studies of the internal soft-part anatomy of the Tanganyikan gastropods (Leloup 1953) confirm that the Tanganyikan gastropods belong in freshwater families and are not closely related to the marine groups that have similar shells.

Some authors have proposed that the resemblance between the Tanganyikan gastropods and marine species was a result of convergence. Stromer (1901), Germain (1907, 1908) and Dautzenberg and Germain (1914) have argued that although Lake Tanganyika was never connected to the sea, in many aspects, such as the variety of geographical exposures, substrates, depths, wave conditions, weather and its great age, Lake Tanganyika behaves like a sea. Since the Tanganyikan gastropods are evolving in similar environments and responding to similar stimuli as marine species, these authors maintained that it is not surprising, to the contrary, it might be expected that the Tanganyikan shells would resemble marine species.

Fuchs (1936) and Beauchamp (1946) hypothesized that Lake Tanganyika had a unique, calcium-rich water chemistry which accounted for the heavy calcification of Tanganyikan shells. However, limnological studies have shown that no observed features of Lake Tanganyika's water chemistry (including calcium concentrations) are unusual (Beadle 1981; Wombwell 1986; Cohen and Thouin 1987; Van Meel 1987). Lake Tanganyika is slightly alkaline and its water chemistry is similar to some neighboring rift lakes which host gastropods with weakly calcified, unornamented shells.

A number of workers speculated that the anomalous features of the Tanganyikan shells were adaptations to specific environments in Lake Tanganyika. For example, Leloup (1953) proposed that the spines of some species served to stabilize the gastropod on soft substrates. And Yonge (1938) maintained that the thick shells were an adaptation for the rocky, wave-battered Tanganyika shorelines. While these generalizations are true in some cases, many gastropod morphologies defy such environmentally-adaptive explanations. For example, while the spines of Tiphobia horei may provide buoyancy on mud substrates (our observations), the spines of the Paramelania species are oriented such that they could not stabilize the gastropod on soft substrates. Furthermore, both heavily and weakly calcified gastropods are found on a variety of substrates from quiet, vegetated sands and muds to wave-battered, rocky shorelines. Finally, the most heavily calcified shells are invariably found below wave base. The morphology of Tanganyikan gastropod shells thus remains largely unexplained by physical or environmental factors.

West et al. (1991) and West and Cohen (1994) have explored the idea that similar environmental stimuli in Lake Tanganyika and in marine environments have produced similar gastropod morphologies. They proposed that the unusual morphologies of the Tanganyikan gastropods are the product of a coevolutionary relationship with endemic predators in the lake. Similar predator-prey coevolution complexes were recognized in the Phanerozic fossil record and modern marine systems. For example, Vermeij (1977, 1987) noted that architecturally stronger mollusc morphologies (such as close-coiled, conispiral and nonumbilicate gastropods) have replaced structurally weaker forms that were more common during the Paleozoic and early Mesozoic. Vermeij (1977, 1987) attributed this transition in mollusc morphology to the increasing armament of shell crushing predators such as teleosts, stomatopods and decapod crustaceans. In another example, Signor and Brett (1984) described the marked increase in sculpture among gastropods, brachiopods, nautiloids and crinoids, paralleled by the subsequent decline of clades without these features. These morphological transitions, they note, are concomitant with radiations of shell-crushing placoderm and chondrichthyan fish.

Several studies in modern marine systems offer evidence that the relationship between prey morphology and predator efficiency is not merely a correlation but that predators are in fact selective agents for the modification of prey morphology and vice versa. In two separate biogeographical studies, fossil gastropods present prior to the introduction of a shell-crushing predator were high-spired and thin-shelled. Modern conspecific gastropods sympatric with shell-crushing predators possess significantly reduced spires and thicker shells than their ancestors (Vermeij 1982a; Seeley 1986). Also, in laboratory experiments, Appleton and Palmer (1988) showed that predatory crabs and damaged conspecific gastropods alike released water-soluble chemical cues that induced the gastropods to increase their shell armament. But in a coevolution model, prey may also induce changes in their predators. Smith and Palmer (1994) showed that crabs raised on fully shelled prey developed larger and stronger claws than crabs grown on nutritionally equivalent unshelled prey. These patterns in the fossil record, biogeographical observations and laboratory studies offer strong evidence for a causal, coevolving relationship between marine molluscs and their shell-crushing predators.

Based on the results of shell-crushing experiments, predation experiments and comparative morphology, West et al. (1991) proposed a similar predator-prey coevolution complex to explain the unusual Tanganyikan shell morphologies. They reported that when all other shell variables are approximately equal, gastropods with larger shells, thickened apertural lips, or sculptured shells suffered significantly fewer fatal attacks from shell-crushing crabs than gastropods with small, thin or unornamented shells. Finally, West et al. (1991) showed that compared to closely-related cosmopolitan thiarid and viviparid gastropods, shell features associated with predation (such as shell size, apertural lip thickness, sculpture, apertural lip thickening and frequency of repair scars) are extremely derived among the endemic Tanganyikan gastropods. Similarly, features of shell-crushing crabs that are associated with predation (such as claw size and dentition) are extremely derived among Tanganyikan potamonautid crabs compared to closely-related cosmopolitan potamonautid and potomonid crabs.

Although West et al. (1991) described macroscopic shell properties associated with shell strength and predation resistance, in this paper, we discuss variations in microscopic shell properties that allow the Tanganyikan gastropods to achieve great shell strengths that confer immunity to predation. In this paper we describe the crystal architecture of the Tanganyikan gastropod shells. We present data from shell-crushing studies and predation experiments (described in West et al. 1991) which show that the microstructure of the Tanganyikan shells is a function of overall shell strength and resistance to shell-crushing predators. Moreover, our phylogenetic analyses reveal that these shell microstructures are derived and have evolved repeatedly in the endemic radiation of thiarid gastropods in Lake Tanganyika. We discuss the Tanganyikan shell microstructures as examples of convergent evolution and adaptation.


Shell Microstructure

We surveyed the shell microstructure of sixteen gastropod species endemic to Lake Tanganyika and two closely-related cosmopolitan species. After we collected the gastropods, the shells were briefly boiled to remove soft tissues and dried at room temperature. Shells were fractured in a vice and/or mechanically abraded by a grinding wheel parallel to the axis of coiling to expose a longitudinal cross-section. Shells were glued to aluminum stubs, coated with gold-palladium and examined with a Cambridge Stereoscan 120 scanning electron microscope (SEM). Some Tanganyikan gastropods may not deposit the full complement of shell layers in the inner and upper regions of the whorl (pers. obs.). Consequently we only surveyed the presence or absence of different shell layers at the outermost portions of the whorls, representing the leading edge of the aperture as the shell grew. We used the nomenclature of Carter and Clark (1985) and Bandel (1990) to describe the crystal architecture of the shell wall. For each gastropod species, at least 15 shells representing different ontogenetic stages were examined.

Shell-Crushing Experiments

We quantified overall shell strength with mechanical shell-crushing studies (after Vermeij and Currey 1980). Crab and fish predators use a variety of different techniques to break open shells: some break the spire of the shell, some peel the shell from the apertural lip, others break the shell outright. Although shell-crushing studies cannot simulate a predatory attack precisely, they can provide an overall measure of shell strength.

We included seven endemic gastropod species from Lake Tanganyika in the shell-crushing studies. We prepared the shells as described above, then measured three shell variables: maximum shell length, maximum shell width, and maximum apertural lip thickness with calipers to the nearest 0.01 mm, prior to crushing. In addition, shell mass was measured to the nearest 0.001 g on an analytical balance. All shells were rewetted before crushing so our results would be comparable with earlier studies (Vermeij and Covich 1978; Vermeij and Currey 1980).

Shells were oriented aperture down on an Instron Universal Testing Machine at the U.S. Air Force Academy's Mechanical Engineering Laboratory and crushed at a rate of 1.27 mm/min. Between 13 and 25 individual shells were crushed for each species, depending on available specimens. We plotted mean maximum load strength values (in kN) against mean shell mass values (in g) for each gastropod species on a logarithmic scale. We used a one-way ANOVA to describe the relationship between shell strength and shell microstructure ([Alpha] = 0.05).

Predation Experiments

We used predation experiments to investigate the adaptive significance of particular shell forms. Representatives from four genera of Tanganyikan gastropods, Lavigeria, Neothauma, Paramelania, Spekia, and one species of shell-crushing crab, Platytelphusa armata, were used in the studies. Before the predation experiments began, we recorded maximum length of the shell, maximum width of the shell, maximum apertural lip thickness, and measures of shell sculpture for the gastropod specimens. For the crabs, we recorded breadth of carapace and three measures of claw robustness. Identification letters were painted on all gastropods and crabs used in the experiments.

For each experiment, one crab was placed in a 40-gal aquarium filled with lake water, sand, algae-encrusted rocks, and 10 gastropods from each of seven different species. In preliminary experiments, we found that predation rates were high at the onset of the experiment but quickly declined after 36 h. Consequently we terminated each experiment after 48 h. The status of each gastropod was recorded into one of the following categories at the end of each experiment: no visible attempt at predation, unsuccessful predation (shell chipped or cracked but gastropod survived), or successful predation (gastropod fatality). For cases of successful predation, the method of predation (peeling the shell from the apertural lip, breaking the shell at the spire, or crushing the shell at the apertural lip) was also recorded. We ran this experiment with a fresh complement of gastropods for each of 41 different crabs. Live gastropods lacking visible predation scars were reused in subsequent experiments.

We tabulated the frequency of successful versus unsuccessful attacks according to the number of crossed-lamellar layers in the gastropod shell wall, regardless of other shell variables. The relative frequencies of unsuccessful and successful attacks were plotted as a function of shell microstructure. We used a two-sample log-likelihood ratio test for goodness of fit (G-test) to evaluate the significance of the relationship between shell microstructure and susceptibility to predation ([Alpha] = 0.05). The G-statistic was subjected to the Williams adjustment (Williams 1976) before it was compared to the chi-square probability distribution.

Character Evolution Studies

To compare the resolution and congruence of paleontological and neontological data in phylogeny reconstruction, West (1991) studied conchological and allozyme characters of the endemic thiarid gastropods of Lake Tanganyika. West studied morphology and microstructure of the shells and conducted allozyme electrophoresis (across 30 presumptive genetic loci) on the tissues of 20 individuals from 17 endemic Tanganyikan gastropod species and two closely related cosmopolitan outgroups. Phylogenetic analyses were conducted separately on the conchological and electrophoretic data sets as well as on the combined conchological and electrophoretic data sets using the PAUP program (Swofford 1990). Bootstrapping was employed to quantify support for the various nodes on the trees. The shell microstructure of the gastropods, determined from the SEM studies described above, was mapped onto the resultant cladograms.


Shell Microstructure

The shells of the endemic Tanganyikan thiarid gastropods are composed primarily of layers of crossed-lamellar crystal architecture [ILLUSTRATION FOR FIGURE 2 OMITTED]. Each layer consists of needle-like aragonite crystals (third-order lamellae) which are packed into laths (second-order lamellae). The laths are stacked to form sheets (first-order lamellae) such that the aragonite needles of adjacent sheets are never parallel (they are typically offset 70-90 [degrees]). The number of crossed-lamellar layers in the shell wall varies from one to four among different Tanganyikan gastropod species (Table 1). The number of crossed-lamellar layers was constant throughout the ontogenies of all shells surveyed and uniform for all individuals within a species. In species with two, three, or four crossed-lamellar layers, the orientation of the first-order lamellae is offset by approximately 90 [degrees] between the different layers [ILLUSTRATION FOR FIGURE 3 OMITTED].

Shell-Crushing Experiments

Figure 4, a plot of mean species mass versus mean species load strength, and Table 1 summarize the results of the shell-crushing studies. Most of the variance in strength between gastropod shells can be explained by the shell microstructure (one-way ANOVA F(3,229) = 321, P [much less than] 0.001). The Tanganyikan gastropod shells that were able to withstand the greatest load strengths invariably had four crossed-lamellar layers. Shells with three crossed-lamellar layers were stronger than shells with two crossed-lamellar layers. The shells [TABULAR DATA FOR TABLE 1 OMITTED] that fractured most readily had a single crossed-lamellar layer. In shell-crushing experiments, many Tanganyikan shells fractured at load strengths comparable to tropical marine shells, denoted by the line log[load strength] = -0.3 + 0.67log[mass] (from Vermeij and Currey 1980). This line represents the load-strength boundary between tropical and temperate marine thaidids.

Predation Experiments

In aquarium experiments, the number of crossed-lamellar layers in the shell had a significant effect on the gastropod's vulnerability to shell-crushing crab predators ([ILLUSTRATION FOR FIGURE 5 OMITTED]; N = 390, Gadj = 105.7, 2 dr, P [much less than] 0.001). Shells with four crossed-lamellar layers were the least susceptible to shell-crushing crabs. Shells with three crossed-lamellar layers suffered significantly fewer fatal attacks than shells with only two crossed-lamellar layers.

Character Evolution Studies

Phylogenetic analyses of conchological and electrophoretic data from the endemic gastropods of Lake Tanganyika produced highly unresolved cladograms (West 1991). Nonetheless, we found several phylogenetic patterns to be uniform across all analyses. In the conchological data only, electrophoretic data only, and combined conchological and electrophoretic data analyses, the cosmopolitan thiarid outgroups, Melanoides and Cleopatra, invariably nested among the endemic Tanganyikan gastropods, suggesting that the endemic Tanganyikan thiarids are at least diphyletic [ILLUSTRATION FOR FIGURE 6 OMITTED]. Though the branching order of clades could not be resolved, all analyses of conchological, electrophoretic and the combined data sets uniformly identified six monophyletic clades of endemic Tanganyikan thiarid gastropods [ILLUSTRATION FOR FIGURE 6 OMITTED] with bootstrap values of 85 or higher.

The number of crossed-lamellar layers, determined from shell microstructure studies, was mapped onto the monophyletic clades established through phylogenetic analysis [ILLUSTRATION FOR FIGURE 6 OMITTED]. Outgroups Melanoides and Cleopatra have one and two crossed-lamellar layers, respectively, and represent the ancestral condition of the Tanganyikan thiarid gastropods. Among the endemic Tanganyikan thiarids, four taxa have three crossed-lamellar layers in their shell wall and three taxa have four crossed-lamellar layers in their shell wall, representing increases from the ancestral condition. The distribution of shell microstructures among the six monophyletic groups of Tanganyikan thiarids indicate at least two, perhaps three, separate origins of three crossed-lamellar layers and two separate origins of four crossed-lamellar layers.


Our shell microstructure studies show that many of the gastropods of Lake Tanganyika build their shells in much the same way plywood is made. Plywood is thin sheets of wood laminated together such that the woodgrains of the different sheets are juxtaposed by about 90 [degrees]. This construction gives plywood an overall tensile strength that is much greater than the sum of the individual sheets. In Tanganyikan gastropod shells, the crossed-lamellar layers are juxtaposed such that the orientation of the sheets, or first-order lamellae, are offset by about 90 [degrees] [ILLUSTRATION FOR FIGURES 3, 7 OMITTED]. The implications of this architecture are clear: If a crack is propagating through a crossed-lamellar layer, each time it intersects a layer in which the first-order lamellae are offset, it will be retarded. Depending on the energy of the propagating crack, enough changes in the orientation of crossed-lamellar layers could halt the crack completely. By retarding crack propagation, changes in the orientation of crossed-lamellar layers effectively strengthens the material without necessarily thickening it.

Currey and Kohn (1976) first described these anisotropic properties of crossed-lamellar structure in documenting the microstructure of Conus shells. The seven Conus species they studied uniformly possessed three crossed-lamellar layers in their shell walls. Gastropods in Lake Tanganyika, however, show one, two, three or four crossed-lamellar layers in their shells. That is, they may have from zero to three changes in the orientation of first-order lamellae between crossed-lamellar layers. Our shell-crushing studies show that the number of crossed-lamellar layers present in the shell is correlated with shell strength. The more crossed-lamellar layers present, the greater the load the shell can withstand. Similarly, predation experiments show that the number of crossed-lamellar layers present in the shell is positively correlated with the gastropods' resistance to shell-crushing crab predators. The more crossed-lamellar layers (and thus changes in orientation of first-order lamellae) present, the lower the frequency of successful predation by shell-crushing crabs.

Because the shell-crushing and predation experiments were conducted prior to our studies of the Tanganyikan shell microstructure, they were not designed (nor can the data be analyzed) such that the shells are uniform with respect to all characters (e.g. shell mass, shell size, shell thickness, shell sculpture) except the number of crossed-lamellar layers. We have explored, for example, the possibility that shell strength may be a function of shell size or thickness or sculpture and that the number of crossed-lamellar layers in the shell might be correlated with one of these other variables. But we have found no simple relationship between crossed-lamellar layers and shell size, thickness or sculpture. For example, although the shells with the most crossed-lamellar layers included some of the largest species in the study, the shells with two and three crossed-lamellar layers were the smallest species in the study, and shells with a single crossed-lamellar layer were intermediate in size. The relationship between shell thickness and crossed-lamellar layers is similar. The thinnest shells in the study had one or two crossed-lamellar layers, but one species with four crossed-lamellar layers was thinner-shelled than all three-layered and some two-layered species, and there is considerable overlap in shell thickness between species with two and three or three and four crossed-lamellar layers. Finally, there is not a clear relationship between shell sculpture and crossed-lamellar layers; shells with four crossed-lamellar layers included both the most and the least sculptured species. Thus, the relationship between shell strength and crossed-lamellar layers is not a simple function of shell size, thickness or sculpture.

Numerous studies have described the merits of various shell features for deterring predators. Noded ribs, fluted or spiny shell sculpture, reduced spires, and/or narrow, folded, thickened or denticulated apertures have been shown to decrease the efficacy of shell-crushing predators (Ebling et al. 1964; Kitching et al. 1966; Vermeij 1973, 1982b; Palmer 1979; Bertness and Cunningham 1981; West et al. 1991). Our results show that morphological defenses by molluscs against shell-crushing predators need not be macroscopic. Even closely related species with grossly similar shells (such as the Lavigeria or Paramelania species) may show subtle differences in crystal architecture that profoundly affect shell strength and the gastropod's vulnerability to predation.

Currey and Kohn (1976) showed the crossed-lamellar structure of Conus shells to be highly anisotropic depending on the orientation of crack propagation with respect to first-order lamellae. Our study affirms that there are significant variations in strength among shells composed of crossed-lamellar structures, depending on the number of crossed-lamellar layers and thus the number of changes in the orientation of first-order lamellae. Previous studies have noted that nacreous and prismatic microstructural configurations are stronger than crossed-lamellar architecture (Currey and Taylor, 1974; Currey 1976, 1988). But Currey (1988) pointed out that although nacre is stronger, crossed-lamellar structures may be deposited up to 2.4 times faster than nacre. He predicted that some gastropods sacrifice shell strength for rapid growth so that they may reach enemy-free space sooner where, owing to their size, they are virtually immune to predators. It is interesting to note that even in Tanganyikan gastropods that have a homogeneous or spherulitic microstructural component to their shell, repair scars were exclusively crossed-lamellar structure. And in Tanganyikan shells that are almost exclusively crossed-lamellar structure, repair scars typically showed fewer than the standard number of crossed-lamellar layers for the species. These observations support Currey's (1988) predictions. Gastropods are most vulnerable after shell damage. In such cases they may opt for rapid repair over shell strength by sealing shell damage with crossed-lamellar material, often with less than the full complement of crossed-lamellar layers for the species.

Like the other unusual rift lake faunas, the thiarid gastropods of Lake Tanganyika evolved in situ in the Tanganyika basin. West (1991) is attempting to resolve the radiation of endemic thiarids in Lake Tanganyika. They selected Melanoides tuberculata and Cleopatra ferruginea, which are known from fossil deposits predating the opening of the rift lakes and are the only cosmopolitan thiarids presently residing in Lake Tanganyika, as outgroups for their analysis. Their Tanganyikan thiarid phylogeny, based on conchological and electrophoretic data, is not well resolved. But despite the present lack of resolution, this phylogeny elucidates some interesting patterns about the evolution of shell microstructure [ILLUSTRATION FOR FIGURE 6 OMITTED]. The outgroups Melanoides and Cleopatra possess one and two crossed-lamellar layers, respectively. Assuming a maximum of two crossed-lamellar layers to be the ancestral condition for the endemic Tanganyikan thiarids, note that many of the modern Tanganyikan thiarids possess three or four crossed-lamellar layers. The Lavigeria, Paramelania, and Chytra/Limnotrochus clades show an increase from two to three or three to four crossed-lamellar layers. Though the branching order of the major clades remains unresolved, there were at least two separate origins of the four crossed-lamellar layer structure and at least two, and perhaps three separate origins of the three crossed-lamellar layer structure.

It is probably not a coincidence that upgrading of crossed-lamellar layers has occurred repeatedly in different clades. This convergence alludes to its selective advantage. Tanganyikan gastropods are routinely subjected to predation by crabs and fish (West et al. 1991; West and Cohen 1994). West et al. (1991) published shell repair scar frequencies for Tanganyikan thiarid gastropods, noting that 15% to 55% of the individuals in a population bore scars commemorating an unsuccessful attack. With such predation pressure, there is clearly a strong selective advantage for increased crossed-lamellar layers that strengthen the shell.

Increases in the complement of crossed-lamellar layers may be an adaptation by Tanganyikan gastropods to thwart shell-crushing predators. There is an increasing literature disputing the definition of adaptation and the criteria for identifying a particular trait as an adaptation (Gould and Vrba 1982; Sober 1984; Coddington 1988; Baum and Larson 1991; Harvey and Pagel 1991; Reeve and Sherman 1993; Frumhoff and Reeve 1994). It is beyond the scope of this paper to summarize this literature or add to the dispute. Coopting parts of several of these definitions and criteria for adaptation, we maintain that three and four crossed-lamellar layers in the shell wall of Tanganyikan gastropods is an adaptation because it is a derived trait that carries a strong selective advantage and confers a higher fitness to Tanganyikan gastropods in the face of shell-crushing predators.

Moreover, gastropods in Lake Tanganyika with only one or two crossed-lamellar layers in their shell walls appear to be ecologically restricted to safe places (such as the undersides of rocks or at the waterline where they are periodically emerged) where predators are rarer or weaker. These biogeographical observations and the multiple independent origins of three and four crossed-lamellar layers among the Tanganyikan thiarid gastropods attest to the selective and fitness advantage of this adaptation.

In Lake Tanganyika, three and four crossed-lamellar layers have evolved repeatedly over a relatively short evolutionary time period (the Tanganyika Basin has a maximum age of late Miocene, approximately 8-12 M.Y.B.P.[Ebinger 1989; Cohen et al. 1993] and recent work has shown that the lake may have dried or nearly dried several times since the mid-Pleistocene [Lezzar et al., in press]). Shell microstructures preserve well in the fossil record (Batten 1972; Runnegar 1984, 1985). Crossed-lamellar layer composition may be used to assess the relative strengths of fossil shells. We encourage workers interested in the coevolution of molluscs and their predators to investigate not only the history of macroscopic mollusc armament but the changes in microstructure as well.


This work was funded by National Science Foundation Grants BSR 8415289, and BSR 8696074, and an H. M. Keck Foundation grant to A.C.P. Ndabaneze, G. Ntakimazi, L. Ntahuga and the Universite du Burundi provided logistical support in Africa. The Baudouin Hussein family, D. Gevirtzman, M. Johnston, R. Jones, E. Michel and M. Soreghan provided assistance in the field. C. Meadows and the U.S. Air Force Academy offered access to their facilities for the crushing experiments. R. Mantonya and R. Schmidtling assisted with photography and drafting, respectively. We thank J. Houbrick, P. Kat, C. Marshall, E. Michel, S. Rice, B. Runnegar and G. Vermeij for helpful discussions and/or reviews.


APPLETON, R. D., AND A. R. PALMER. 1988. Water-borne stimuli released by predatory crabs and damaged prey induce more predator-resistant shells in a marine gastropod. Proc. Nat. Acad. Sci. 85: 4387-4391.

BAKER, B. H., L. A. J. WILLIAMS, T. A. MILLER, AND P. A. MOHR. 1972. Geology of the Eastern Rift system of Africa. Spec. Pap. Geolog. Soc. Am. 136:1-67.

BANDEL, K. 1990. Shell structure of the gastropoda excluding archaeogastropoda. Pp. 117-134 in J. G. Carter, ed. Skeletal biomineralization: Patterns, processes, and evolutionary trends. Vol. I. Van Nostrand Reinhold, New York.

BATTEN, R. L. 1972. The ultrastructure of five common Pennsylvanian Pleurotamarian gastropod species of Eastern United States. Am. Mus. Novit. 2501.

BAUM, D. A., AND A. LARSON 1991. Adaptation reviewed: A phylogenetic methodology for studying character macroevolution. Syst. Zool. 40:1-18.

BEADLE, L. C. 1981. The inland waters of tropical Africa. Longman, London.

BEAUCHAMP, R. S. A. 1946. Lake Tanganyika. Nature 157:183-184.

BERTNESS, M. D., AND C. CUNNINGHAM 1981. Crab shell-crushing predation and gastropod architectural defense. J. Exp. Mar. Biol. Ecol. 50:213-230.

BOURGUIGNAT, J. R. 1890. Histoire malacologique du Lac Tanganika. Libraire de l'Academie de Medicine, Paris.

CARTER, J. G., AND G. R. CLARK. 1985. Classification and phylogenetic significance of molluscan shell microstructure. Pp. 50-71 in D. J. Bottjer, C. S. Hickman, and P. D. Ward, eds. Mollusks, notes for a short course. University of Tennessee, Knoxville.

CODDINGTON, J. A. 1988. Cladistic tests of adaptational hypotheses. Cladistics 4:3-22.

COHEN, A. S., AND C. THOUIN. 1987. Nearshore carbonate deposits in Lake Tanganyika. Geology 15:414-418.

COHEN, A. S., C. A. SCHOLZ, AND M. J. SOREGHAN. 1993. New age estimates for Lake Tanganyika: Implications for the tectonic evolution of the East African Rift. Geology 21:511-514.

CURREY, J. D. 1976. Further studies on the mechanical properties of mollusc shell material. J. Zool. Lond 180:445-453.

-----. 1988. Shell form and strength. Pp. 183-210 in E. R. Trueman and M. R. Clarke, eds. The mollusca. Vol. 11: Form and function. Academic Press, New York.

CURREY, J. D., AND A. J. KOHN. 1976. Fracture in the crossed-lamellar structure of Conus shells. J. Mat. Sci. 11:1615-1623.

CURREY, J. D., AND J. D. TAYLOR. 1974. The mechanical behavior of some molluscan hard tissues. J. Zool. Lond. 173:395-406.

DAUTZENBERG, P., AND L. GERMAIN. 1914. Recoltes malacologiques du Dr. J. Bequaert dans le Congo belge. Rev. Zool. Bot. Afr. 4:1-74.

EBINGER, C. J. 1989. Tectonic development of the western branch of the East African rift system. Geol. Soc. Am. Bull. 101:885-903.

EBLING, F. J., J. A. KITCHING, L. MUNTZ, AND C. M. TAYLOR. 1964. The ecology of Lough Ine. XIII: Experimental observations of the destruction of Mytilus edulis and Nucella lapillus by crabs. J. Anim. Ecol. 33:73-83.

FRUMHOFF, P. C., AND H. K. REEVE. 1994. Using phylogenies to test hypotheses of adaptation: A critique of some current proposals. Evolution 48:172-180.

FUCHS, V. E. 1936. Extinct Pleistocene mollusca from Lake Edward, Uganda, and their bearing on the Tanganyika problem. J. Linn. Soc. Lond. 40:93-105.

GERMAIN, L. 1907. Essai sur la malacographie de l'Afrique equatoriale. Arch. Zool. Exp. Gen. 6:103-106.

-----. 1908. Mollusques du Lac Tanganika et de ses environs. - Resultats scientifiques des voyage en Afrique d'Edouard Foa. Bull. Mus. Hist. Nat. Paris 14:612-628.

GOULD, S. J., AND E. VRBA. 1982. Exaptation - A missing term in the science of form. Paleobiology 8:4-15.

HARVEY, P. H., AND M. D. PAGEL. 1991. The comparative method in evolutionary biology. Oxford University Press, Oxford.

KITCHING, J. A., L. MUNTZ, AND F. J. EBLING. 1966. The ecology of Lough Ine. XV: The ecological significance of shell and body forms in Nucella. J. Anim. Ecol. 35:113-126.

LELOUP, E. 1953. Gasteropodes. Exploration hydrobiologique du Lac Tanganika - Resultats scientifiques, 3. Institut Royal des Sciences Naturelles de Belgique, Bruxelles.

LEZZAR, K., J. TIERCELIN, T. BANDORA, P. VAN RENSBERGEN, M. DE BATIST, A. COHEN, W. VERSTEEG, E. MAES, W. MIFUNDU, AND J. KLERKX. In press New seismic stratigraphic data and the Late Tertiary history of the North Lake Tanganyika Basin, East Africa Rift System. Basin Res.

MOORE, J. E. S. 1898. On the zoological evidence for the connection of Lake Tanganyika with the sea. Proc. Zool. Soc. Lond. 62:451-458.

-----. 1899. On the hypothesis that Lake Tanganyika represents an old Jurassic sea. Quart. J. Microscop. Sci. Lond. 41:303-321.

-----. 1903. The Tanganyika problem. Hurst and Blackett, London.

PALMER, R. A. 1979. Fish predation and the evolution of gastropod shell sculpture: Experimental and geographic evidence Evolution 33:697-713

REEVE, H. K., AND P. W. SHERMAN 1993. Adaptation and the goals of evolutionary research. Quart Rev. Biol. 68:1-32.

ROSENDAHL, B., D. REYNOLDS, P. LORBER, C. BURGESS, J. McGILL, D. SCOTT, J. LAMBIASE, AND S. DERKSON. 1986. Structural expressions of rifting: Lessons from Lake Tanganyika, Africa Pp. 29-43 in L. Frostick and I. Reid, eds. Sedimentation in the African Rifts. Geol. Soc. Lond. Spec. Publ. 25.

RUNNEGAR, B. 1984. Crystallography of the foliated calcite shell layers of bivalve molluscs. Alcheringa 8:273-290.

-----. 1985. Shell microstructures of Cambrian molluscs replicated by phosphate. Alcheringa 9:245-257.

SEELEY, R. H. 1986. Intense natural selection caused a rapid morphological transition in a living marine snail. Proc. Nat. Acad. Sci. 83:6897-6901.

SIGNOR, P. W., AND C. E. BRETT. 1984. The mid-Paleozoic precursor to the Mesozoic marine revolution. Paleobiology 10:229-245.

SOBER, E. 1984. The Nature of Selection. The M.I.T. Press, Cambridge, MA.

SMITH, L. D., AND A. R. PALMER. 1994. Effects of manipulated diet of size and performance of Brachyuran crab claws. Science 264: 710-712.

STROMER, E. 1901. Ist der Tanganyika ein relikten-see? Dr. A. Peterman's Mittelungen Justus Perthes Geogr. Anst. 47.

SWOFFORD, D. L. 1990. PAUP: Phylogenetic analysis using parsimony, version 3.0. Computer program distributed by the Illinois Natural History Survey, Champaign.

VAN MEEL, L. I. J. 1987. Contributions a la limnologie de quatre lacs du Zaire oriental: Tanganika, Kivu, Mobutu Sese Seko (ex. Albert), Idi Amin Dada (ex. Edouard). Fasicule A: Le Lac Tanganika. Institut R. Sci. Nat. Belg. Doc. Trav. 41.

VERMEIJ, G. J. 1973. Molluscs in mangrove swamps: physiognomy, diversity and regional differences. Syst. Zool. 22:609-624

-----. 1977. The Mesozoic marine revolution: Evidence from snails, predators and grazers. Paleobiology 3:245-258.

-----. 1982a. Phenotypic evolution in a poorly dispersing snail after arrival of a predator. Nature 299:349-350.

-----. 1982b. Gastropod shell form, breakage, and repair in relation to predation by the crab Calappa. Malacologia 23:1-12.

-----. 1987. Evolution and escalation. Princeton University Press, Princeton, NJ.

VERMEIJ, G. J., AND A. P. COVICH. 1978. Coevolution of freshwater gastropods and their predators. Am. Nat. 112:833-843.

VERMEIJ, G. J., AND J. D. CURREY. 1980. Geographical variation in the strength of thaidid snail shells. Biol. Bull. 158:383-389.

WEST, K. 1991. Resolution and congruence of biological and paleontological data sets in phylogenetic reconstruction. M.Sc. thesis, University of Arizona, Tucson, AZ.

WEST, K. AND A. COHEN 1994. Predator-prey coevolution as a model for the unusual morphologies of the crabs and gastropods of Lake Tanganyika. Arch. Hydrobiol. Beih. Ergebn. Limnol. 44:267-283.

WEST, K., A. COHEN, AND M. BARON. 1991. Morphology and behavior of crabs and gastropods from Lake Tanganyika, Africa: Implications for lacustrine predator-prey coevolution. Evolution 45:589-607.

WILLIAMS, D. A. 1976. Improved likelihood ratio tests for complete contingency tables Biometrika 63:33-37.

WOMBWELL, J. V. 1986. The chemical limnology of Lake Tanganyika. M.Sc. thesis, Colorado College, Colorado Springs, CO.

YONGE, C. M. 1938. The prosobranchs of Lake Tanganyika. Nature 142:464-466.
COPYRIGHT 1996 Society for the Study of Evolution
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1996 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:West, Kelly; Cohen, Andrew
Date:Apr 1, 1996
Previous Article:Character discriminatory power, character-set congruence, and the classification of individuals from hybrid zones: an example using stone crabs...
Next Article:Independent evolution of terrestriality in Atlantic truncatellid gastropods.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |