The Helicidae are most abundant in the north and west, and are represented by several very striking sub-genera, some of which possess remarkably toothed apertures, and perhaps betray an ancestry common to some of the West Indian genera. Of these, Labyrinthus has 12 species in Venezuela and Colombia, 5 in Ecuador, and 3 in Peru and Bolivia; Isomeria 12 in Venezuela and Colombia, 20 in Ecuador, and 2 in Peru and Bolivia; Salaropsis is represented in these countries by 6, 3, and 7 species, and Systrophia by 4, 5, and 8 species respectively.

Clausilia—in the group Nenia—appears in some numbers along the Andes chain, the only other representative in the New World being the solitary species occurring at Porto Rico. There have been described, from Venezuela and Colombia 10 species, from Ecuador 5, and from Peru and Bolivia 12.

Another marked feature of the region is the occurrence of the Orthalicidae, in the two genera Orthalicus and Porphyrobaphe. The latter of these magnificent forms is peculiar, while the former reaches Mexico, the West Indies, and Brazil. Ecuador, which contains 23 species, seems the metropolis of the group.

Bulimus and Bulimulus, the former genus being peculiar to S. America and the adjacent islands, are largely represented, the former in the three groups Borus, Dryptus, and Orphnus. These attain their maximum in Peru, with 25 species, but Venezuela and Colombia have as many as 17. Bulimulus has been subdivided into a number of groups, e.g. Drymaeus, Mesembrinus, Thaumastus, Mormus, Scutalus, with many others,—the exact scientific limits of which are not easily discernible. It must suffice here to state that Peru seems to be the headquarters of the group with about 190 species (which probably may well be reduced), Ecuador having about 70, and Venezuela and Colombia between 80 and 90.

Two very remarkable forms belonging to the Pupidae, Anostoma (Fig. 154, p. 248) and Tomigerus, occur in Venezuela, the metropolis. Rhodea, another very peculiar shell (Fig. 234), whose exact family position is uncertain, is peculiar to New Grenada. The land operculates are few in number, and in Bolivia almost disappear. They belong principally to Neocyclotus (of which 11 species occur in Venezuela and Colombia) and Helicina (10 species in the same district), besides the stragglers already mentioned from West Indian sources, and a few Cyclophorus. Bourcieria is a form of Helicina peculiar to Ecuador. Ampullaria, with Ceratodes, a peculiar planorbiform sub-genus, and Hemisinus, form the bulk of the fresh-water operculates.

The Galapagos.—Thirty-four species of land Mollusca, all peculiar, are known from these islands; 25 of these are forms of Bulimulus. There are no Helicidae, one each of Hyalinia, Leptinaria, and Helicina, and two Pupa. The Bulimulus are mostly of the group Nesiotis, and in their brown colour bear some outward resemblance to the dark Achatinella of the Sandwich Is., living as they do mostly under scoriae on the ground, and not on trees. In type, however, they appear to be derived from Chili and Peru, rather than from the parts of S. America immediately contiguous. Another section (Pleuropyrgus 2 sp.) closely resembles a marine Chemnitzia. The islands are all volcanic, and are probably not the result of subsidence; thus the existing species are not to be regarded as the relics of a more widespread fauna, but as a new set of inhabitants.

(4) The Brazilian Sub-region.—This immense district is very little known, except in the south, and it is consequently impossible to give any satisfactory account of its Mollusca. It is possible that eventually it will be found that it falls into provinces which correspond more or less to (a) the Amazon basin; (b) the mountainous district in the east, drained by the Tocantins and the San Francisco; (c) the Parana basin in the south central district; and (d) the Argentine or Pampas district in the extreme south. But at present the data are insufficient to establish any such subdivisions, whose existence, if proved, would have an important bearing on the problem of the coalescence of S. America into its present form.[379]

The Agnatha are represented by Streptaxis alone (17 sp.). Helix is rare, but includes the peculiar Polygyratia (Fig. 150 A, p. 246), while Labyrinthus (2 sp.), Solaropsis (5 sp.), and Systrophia are common with the Colombian Sub-region, and Oxychona (4 sp.) with the Central American. Bulimus has in all 36 species, the sub-genera Pachyotus (Fig. 233) and Strophochilus being peculiar. Bulimulus, though not so abundant as in Peru and Ecuador, has about 60 species, of which Navicula (Fig. 235) is the most remarkable group. Megaspira is peculiar. Orthalicus has only 4 species, while Tomigerus (4 sp.) and Anostoma (3 sp.) are common with Venezuela. Land operculates are scarce, and appear to include only Neocyclotus, Cyclophorus, and Helicina.

In Argentina, which may probably rank as a separate province, the tropical forms greatly decrease, Streptaxis being reduced to 2 species, and Bulimus and Bulimulus together to 40, while Orthalicus, the great Helices, and the land operculates disappear altogether. Odontostomus (Fig. 236), a genus of the Pupidae, is abundant in the northern part of the province. Two or three species of Chilina occur.

(5) The Chilian Sub-region.—The greater part of Chili, from its arid and rainless climate, is unfavourable to the existence of land Mollusca. Bulimus (Borus) still has 3 or 4 species, and Bulimulus (Plectostylus 11, Scutalus 9, Peronaeus 7) is fairly abundant, but the profusion of the tropics is wanting. There are no carnivorous genera, and only two land operculates. A remarkable form of Helix (Macrocyclis, Fig. 237) is quite peculiar, but the majority of the species belong to two rather obscure groups, Stepsanoda and Amphidoxa. Chilina, a singularly solid form of Limnaea (of which 8 sp., with a sub-genus Pseudochilina, occur in Chili), is peculiar to Chili, S. Brazil, and Patagonia. From the two islands of Juan Fernandez and Masafuera, are known several Helix, of Chilian affinity, several curious Succinea, a Homalonyx, Leptinaria, and Nothus, and three species of Tornatellina, with the almost universal Limax gagates.

The question of the existence at some remote period of a Neantarctic continent, which formed a communication between the three great southern peninsulas of the world, is one on which the Mollusca may offer evidence. Von Ihering holds that an essential difference can be observed between certain of the Unionidae which inhabit S. America, Africa, and Australia with New Zealand, and those which inhabit Europe, Asia, and N. America, but the point can hardly be regarded as definitely established at present. Something perhaps may be made of the distribution of Bulimus and Bulimulus. It seems difficult to explain the occurrence of sub-fossil Bulimus on St. Helena except on some such lines as have been recently adduced to account for the presence of struthious birds in the Mascarenes, and possibly the form Livinhacea may be a trace of the same element in S. Africa. Again, the Liparus of S. and W. Australia, with the Caryodes of Tasmania, and the Leucotaenia and Clavator of Madagascar (which all may be related to Bulimus), together with the Placostylus of New Caledonia and the adjacent islands, reaching even to New Zealand, and perhaps even the Amphidromus of Malaysia (which are more akin to Bulimulus), may be thought to exhibit, in some remote degree, traces of a common ancestry.

The land operculates give no help, and, of the carnivorous genera, Rhytida is a marked link between Africa and Australia, while Streptaxis is equally so between S. America and Africa. As regards fresh-water Gasteropoda, Ampullaria is common to S. America and Africa, while Isidora is common to Africa, Australia, and New Zealand, but is altogether absent from S. America. Gundlachia occurs in Florida, Trinidad, and Tasmania, but has not been detected in Africa. It must be concluded, therefore, that the present state of the evidence which the Mollusca can afford, while exhibiting certain curious points of relationship between the three regions in question, is insufficient to warrant any decided conclusion.