Isolation and Evolution of the Amphidromus in Nusa Tenggara
by Richard L. Goldberg & Mike Severns

The eminent malacologist and land snail aficionado William Clench once claimed that if he were transported blindfolded to a tropical Pacific island, he could determine his location based on the color forms of land snails that he would encounter. If, as Clench proclaimed, an island's unique land snails can act as indicators of geographical location, then, ostensibly, a snail sleuth familiar with variations of insular species could also ably track his way through any tropical archipelago of the world.
Variable color and pattern are a hallmark of many tropical land mollusks. Yet this variability has caused countless wrong turns on the road map of taxonomy. The polymorphic nature of these snails confounded early taxonomists who often described many color forms as new, without any consideration of their relationship to adjacent populations.

Studying landshell variation can be a fascinating pursuit for any shell collector. When laid out in collection drawers, a well organized and documented series of landshells can provide a conchologist with the opportunity to observe an evolutionary moment in time of a species. The varied and aesthetically pleasing series of snail shells can also reveal information from which to learn more about the similarities and differences among neighboring species.

A land snail species can develop a shell with myriad color forms within one population (intra-population variability), or may vary from the norm only in isolated populations within its range (inter-population variability). Occasionally a species can exhibit both phenomena. A number of biological concepts may be useful in explaining why a species of snail can vary locally or over its entire range.

The external appearance of a shell -- color and pattern, for instance -- are likely to be influenced by natural selection when determining the genetic make up of a snail. Extreme color polymorphism in a population perhaps indicates that looking different from your neighbor is advantageous, making it more difficult for predators to develop a search image, such as color and contrast, to locate their prey. The unique color form can then spread rapidly in a population. This phenomenon, known as frequency dependent selection, may explain why some intra-populations of snails in the genus Amphidromus from eastern Indonesia often have unpatterned or strikingly different colored shells mixed in among the predominantly multi-colored or strongly patterned shells. Natural selection is typically considered to favor one form over another, thereby shifting the bell curve, yet frequency dependent selection flattens that bell curve and spreads it out.

One explanation for inter-population variability may be a phenomenon called genetic drift. To illustrate this, let's say a population gets split in two, one on island A, and one on island B. Both populations might start off identical, with chances at 50/50 for a particular trait. It is unlikely that the offspring will be 50/50. They may be 47/53, and the subsequent generation may be 43/57. Over many generations, the percentages will fluctuate up and down at random, and they can end up being very different from where they started. Yet, in the real world, a new mutation doesn't start at 50/50. It might be only 1/10,000 or 1/100,000. By chance it can have more surviving offspring than average, and slowly increase in the population. Smaller populations will drift faster than large populations. But, if by chance that first mutant does not have any surviving offspring, that trait disappears. This is a more likely scenario. So, genetic drift tends to eliminate variation within a population and increases differences between populations. Over a period of time, the populations from islands A and B will end up looking very different.

Inter-population variability might also be explained by the founder effect. The snails that colonize an island may represent only a small fraction of the parent population's genetic diversity. If a rare mutant color form is mixed in among the new colony, a colony that was 1 in 1000 in the parent population, it may instantly be 1 in 10 in the new population. Given a brand new set of environmental influences, the population may evolve in whole new directions from its parent population.

Environmental factors such as the geology, climate, flora, fauna, and food are all suspected contributors to the rise of a unique shell appearance. To reduce the competition with other species which overlap in some aspect of their niches, a phenomenon called character displacement occurs. Shell, behavioral, anatomical or biochemical characters eventually deviate from the ancestral form to allow the species to survive with its neighbors. Character displacement might also manifest itself in the color, pattern, and even the form of the shell. Changes brought on by one, or a combination of, these biological concepts can eventually render an isolated population unable to interbreed with the parent population, producing a situation which makes speciation possible.

Isolation is the key ingredient for a species to evolve, and natural barriers provide the necessary separation to allow speciation to take place. Whether the barrier is a series of mountain ridges separated by deep valleys, as in the case of the Hawaiian Achatinella tree snails, or islands separated by wide channels as in the case of some species of Indonesian Amphidromus, isolation allows a species to evolve independently from its ancestral form.

An excellent illustration of how isolation influences independent evolution is seen in the species Amphidromus inconstans Fulton, 1898, which inhabits a limited group of islands in Nusa Tenggara, eastern Indonesia. Nusa Tenggara has served as a natural laboratory of evolution for Amphidromus and other biota, because of the region's long isolation from the surrounding Indonesian islands, and from Australia. Many endemic species are found here, including the exotic Komodo dragons.

Amphidromus inconstans has also caused a great deal of taxonomic confusion due to the complexity of the inter-relationship among neighboring species. A. inconstans, as it is traditionally understood and revised by Laidlaw & Solem in 1961, exhibits a broadly differing array of colors, patterns and, at times, shell form, throughout its range. A perfunctory examination of these color forms is deceiving, and gives a false impression of one widely ranging species. Recent field collecting throughout the range of A. inconstans has provided a much clearer picture of this intriguing group of snails. Initial observations of the comprehensive series of shells from this field work, and comparisons with neighboring species has shown that A. inconstans has a much narrower range than originally thought.

Taxonomically, Amphidromus inconstans is placed in the subgenus Syndromus, a group of sinistral species noted for their brightly colored and variably patterned shells (see Am.Conch 21(1): 5, "Variation in the Amphidromus"). Older collections with limited habitat documentation made it difficult to comprehend the full extent of the variability of A. inconstans, and its associations with species from adjacent islands. Also, early collectors often sorted out various color forms for aesthetic reasons, giving a false impression of inter- and intra-population variability.

For instance, many species of Syndromus contain a solid yellow intra-population color form. These yellow shells were often separated by field collectors because of their desirability. Since the shells had limited habitat data, by the time they reached museum collections, they were assumed to have come from different localities. It is now known that, in most cases, intra-population variability is commonplace in this group. Even with this in mind, A. inconstans poses a dilemma for taxonomists since there is strong evidence to indicate that we are dealing with more than one species.

Determining whether the color forms traditionally thought of as A. inconstans actually represent more than one species requires an understanding of Laidlaw & Solem's view of its distribution. As currently understood, the string of islands north of Timor between eastern Flores and Romang are home to the Amphidromus inconstans complex, a seemingly whimsical group that displays an enormous range of color and pattern. Each island in the geographical range of A. inconstans, starting in the west on Pantar and moving east to Romang, has populations that take on a unique appearance. Yet a traceable pattern to the whimsical nature of these extreme variations exists, and clearly offers visible evidence that populations of A. inconstans in the eastern and western portions of the range have evolved from different ancestral species.

The palette of colors and patterns exhibited by these snails becomes a paradigm for the evolutionary changes that are influenced by isolation. In this case, sea water provides a formidable barrier for isolating these insular snails. The bright, but mostly unicolored forms on the islands in the western part of its range are in stark contrast to the extremely intricate patterns of the eastern island forms.

Basically, the western islands of Pantar and Pura are home to the solid yellow color form described as A. gracilis von Martens, 1899 (Fig.1: 1 - 2). On the small island of Pura, the species begins to exhibit faint banding around the periphery of the body whorl (Fig.1: 4 - center).

Fig.1

Moving east to Alor, three distinct A. inconstans forms are found. The solid yellow A. i. gracilis form previously seen on Pantar and Pura lives side-by-side with an intricately patterned form considered typical A. inconstans (Fig.1: 3). This nominate form typically has four spiral brown bands on the body whorl, the upper two broken into irregular brown squares. It has a base color of yellow and white. This pattern varies considerably and, in a series, shows a remarkable gradation into the unicolored gracilis form (Fig.1: 4 - left & right).

Also on Alor, a thin shelled form, Amphidromus i. oscitans von Martens, 1899 (Fig.2: 5), with its wavy radial streaks and more convex outline, lives at higher elevations. A solid yellow form of A. i. oscitans, in the past also referred to as A. i. gracilis, can also be found (Fig.2: 5 - right). The appearance of this distinct form on Alor, isolated from the lower elevation A. inconstans and A. i. gracilis forms, is one piece of evidence that A. i. oscitans may in fact be a distinct species. Early accounts report finding it only on Pura and Alor. Recently a few specimens were also found at higher elevations on Atauro (Fig.2: 6), the next island in the chain.

Fig.2

Atauro (also sp. Atuaro) is a small island in the wide, deep channel between Alor and Wetar, just north of Timor. It possesses one of the truly beautiful and unique Amphidromus. This population exhibits a highly developed color pattern not seen anywhere else in the range of A. inconstans. Its white base color, with a dark blue/black to pale sky blue pattern overlaid with bands or flushes of yellow, provides a bridge to the more intensely colored eastern island forms (Fig.3: 7). Unlike the A. inconstans color forms from other islands, which tend to exhibit intra-population variability, the Atauro population is amazingly consistent in form throughout the island. This form has never been recorded in literature, and is presented here for the first time.

Fig.3

Continuing northeast to Wetar and the surrounding islets, one finds A. i. wetaranus Haas, 1912, a form which represents the greatest divergence from the previous forms (Figure 4: 8 - 9). Until recently, only a few specimens of A. i. wetaranus were available for study. Haas originally described the shell as a full species. Later, Laidlaw and Solem (1961) reclassified it as a subspecies of A. inconstans based on its similarity to the Romang Island form named A. i. subporcellanus discussed below. There is a strong argument for removing A. i. wetaranus from the shadow of A. inconstans because of the consistently different shell, generally smaller size, more slender profile, and a unique and intense color pattern. It also exhibits red bands not seen in any of the A. inconstans color forms from the islands west of Wetar. The Wetar snails, like the low elevation Atauro snails, more closely resemble some forms of A. contrarius (Mueller, 1774) endemic to Timor and its western satellite islands (Figure 5: 12). Much like the forms found on Alor and Romang, A. i. wetaranus also exhibits a great deal of intra-population variability.

Fig.4

Fig.5

Due east of Wetar is the termination point in the traditional range of the A. inconstans complex, Romang Island (Roma in older literature) and its satellite islets. Romang is home to an extraordinarily varied population of snails (Figure 6: 10 - 11). It is currently known as A. inconstans rollei Laidlaw & Solem, 1961; Rolle described the population as A. i. gracilis Rolle, 1903, but that name was preoccupied by the unicolored westerly forms from Pantar, Pura and Alor. In that same paper, Rolle also described three distinct color forms of his extremely variable A. gracilis (=rollei). Shells with narrow radial bands broken into spots by spiral bands of ground color were given the form name A. subporcellanus; A. viridistriata are those shells with a green band above the red columellar color patch; and A. subsimplex are those unicolored shells with only a red columellar color patch. These varieties of A. inconstans rollei intergrade so completely that the names have no real taxonomic value except as a means for collectors to single out a color form.

Fig.6

The geology of these islands plays a significant role in understanding why the western and eastern forms of A. inconstans probably represent distinct and separate species (see map). The islands west of Wetar, namely Alor, Pura, and Pantar, sometimes referred to as the Alor Archipelago, are geologically part of a string of volcanic islands that extend west through Flores, Indonesia and belong to the Asian continent. These islands are covered with lush green rain forests. The islands are very close together, allowing some genetic communication among snail populations. East of Alor, the islands of Atauro, Wetar, Reong, Nyata, and Romang are all part of the Australian Plate (Timor and Kissar are also grouped here, but are not inhabited by A. inconstans). These uplifted limestone islands are covered with dry-land forest and, in some cases, eucalyptus.

It is significant to note that the forms on these limestone islands of the Australian plate, long considered to be related to A. inconstans, almost always have a black dot protoconch, a characteristic that is exhibited by A. contrarius on Timor (Figure 6: 15). But to the west, on the volcanic islands of the Alor Archipelago, the forms of A. inconstans lack the black dot protoconch; instead, they have either light or white embryonic whorls. The only exception to this on the western islands is the high elevation A. i. oscitans of Pura and Alor, which does exhibit the black dot protoconch of the eastern populations.

Fig.6

This apparent evidence suggests that migration has occurred west to east on the volcanic islands, and southwest to northeast on the limestone islands beginning with western Timor and including the islands north and east of Timor. The genus Amphidromus may have originated in Asia and spread south and east into Indonesia. As the genus reached the small islands of eastern Nusa Tenggara, populations may have become more isolated than they were previously in their range and the distances between habitable islands may have increased. Those that made the leap between islands may have exploded into new forms and species with spectacular colors. It was those colors that immediately attracted naturalists to study these brightly colored snails, brought back to Europe aboard the ships of spice traders after they hopped among islands while sailing through eastern Indonesia.

Low sea levels, in some cases low enough to provide a land bridge from Singapore to Bali, provided Amphidromus populations the opportunity to move eastward. To the east in Nusa Tenggara, migration was most probably by storms, birds such as eagles, or even fruit bats, which may have carried young snails between their feeding grounds and their nesting islands. The result is random dispersal over long distances. Populations became established on remote islands, leading to evolutionary isolation and independence from the ancestral form. With little or no genetic communication with the parent population, these isolated, island-bound snail populations could diverge to form new species.

A. i. wetaranus and A. i. rollei, which inhabit the eastern limestone islands, show a remarkable likeness to A. contrarius from Timor and its adjacent islands to the west. The general shell shape, black dot protoconch, and similarities in color/pattern of A. wetaranus mimic that of A. contrarius populations from the Niki Niki area of Timor, due south and west of Wetar Island. The shells of A. wetaranus tend to be slightly less inflated than A. contrarius, but other features are very close. These apparent similarities lead us to believe that A. contrarius, and not A. inconstans as has been believed, was introduced to Atauro, Wetar and from there to Romang where isolation and evolution have taken hold.

Other Timor populations of A. contrarius (Figure 5: 13) also show a close affinity in shape, size, and dark protoconch to the low level population of Amphidromus from Atauro, in close proximity to Timor and of similar geological makeup. Yet the Atauro snails somewhat mimic the shell patterns of typical A. inconstans from Alor. This anomaly begs the question of hybridization.

A. i. oscitans from Pura, Alor, and Atauro presents another anomaly. The black dot protoconch and general shell characteristics exhibited by this species link it to A. contrarius, A. wetaranus, and A. rollei, yet isolation has allowed the snail to develop enough distinct shell characters to warrant specific status. Its habitat at higher elevations, removed from limestone, may be the reason for its having a thin shell, a trait often seen in species of snails living at high elevations.

This convincing evidence suggests that A. inconstans, long thought to be one highly variable species, has, in fact, succumbed to the changes influenced by isolation. The complex has seemingly evolved into four or five distinct species stemming from two ancestral forms; A. contrarius from Timor being the parent population for the Amphidromus inhabiting the limestone islands; and A. poecilochrous (Figure 7: 14) from Flores & Sumbawa Islands possibly being the parent population for the Amphidromus on the volcanic islands. Yet additional field work is necessary to confirm or deny our speculation that there is a connection between A. poecilochrous and A. inconstans.

Fig.7

If there is a pattern to this hypothesis, then the first species, Amphidromus inconstans, would include the color forms that inhabit the volcanic islands of Pantar, Pura, Alor (Figure 1: 2 - 4); a second species, A. oscitans found only at higher elevations on Pura, Alor and Atauro would also be distinct (Figure 2: 5 - 6); the limestone islands of Wetar and Romang would be home to the third and fourth species, A. wetaranus (Figure 4: 8 - 9) and A. rollei (Figure 6: 10) respectively; and finally the low elevation Amphidromus from Atauro may, in fact, represent another new species (Figure 3: 7). A paper currently being drafted will delineate the Atauro snail.

Equipped with this empirical data, not only would William Clench have been assured of which island he had landed and collected on, but also this data would have clarified for him what direction he was heading if he had journeyed in the path of the spice traders through the islands in eastern Nusa Tenggara. This fascinating complex of Amphidromus is just one of many in the genus that offers a challenging puzzle for students of landshells, professional and amateur alike.

About the Authors | Glossary | References | Acknowledgements