Donn E. Rosen

VICARIANT PATTERNS AND HISTORICAL EXPLANATION IN BIOGEOGRAPHY

Rosen, D. E. (Department of Ichthyology, American Museum of Natural History, New York, New York 10024) 1978. Vicariant patterns and historical explanation in biogeography. Syst. Zool. 27:159-188.-Geographic coincidence of animal and plant distributions to form recognizable patterns suggests that the separate components of the patterns are historically connected with each other and with geographic history. To seek evidence of these historical connections, cladograms of geographic areas, representing sequences of disruptive geologic, climatic, or geographic events, may be compared with biological cladograms, representing sequences of allopatric speciation events in relation to those geographic areas. Such comparisons, when they meet the minimum requirements of being among dichotomized threetaxon cladograms, can resolve similar or dissimilar historical factors; two-taxon statements do not distinguish between groups with different histories. Congruence of biological and geological area-cladograms at a high confidence level (such as congruence of a five-taxon cladogram or four three-taxon cladograms with a geological cladogram, where the confidence level can be shown in cladistic theory to be 99%o) means that specified events of paleogeography can be adopted as an explanation of the biological patterns. In such a cause and effect relationship, where the earth and its life are assumed to have evolved together, paleogeography is taken by logical necessity to be the independent variable and biological history, the dependent variable. Drawing a mathematical simile, the biological cladogram y (dependent variable), is a function of the geological cladogram x (independent variable), as in a simple regression of effect y on cause x where we are given no free choice as to which is the independent variable. Such a view implies that any specified sequence in earth history must coincide with some discoverable biological patterns; it does not imply a necessary converse that each biological pattern must coincide with some discoverable paleogeographic pattern, because some biological distributions might have resulted from stochastic processes (chance dispersal). Determining that all discoverable biological patterns conflict with a given corroborated or observed sequence of geologic, climatic, or geographic change (i.e., that y is not a function of x), in theory, therefore should falsify vicariance biogeography. Because dispersal biogeography presupposes stochastic processes, and any failure to meet the expectation of a postulated dispersal is explained by an additional dispersal, dispersal biogeography is immune to falsification. Without resort to paleontology or earth history, whether a given historical relationship implied by congruence of biological area-cladograms is the result of dispersal or vicariance can also be thought of in terms which minimize the number of necessary assumptions: did the sedentary organisms disperse with the vagile ones or did the vagile organisms vicariate with the sedentary ones? Cladistic congruence of a group of sedentary organisms with a group of vagile ones rejects dispersal for both. Hence, distributions of sedentary organisms have the potential to falsify dispersal theories as applied to vagile organisms, but distributions of vagile organisms cannot falsify vicariance theories as applied to sedentary ones. The problems that arise in various kinds of historical explanation are exemplified by several specific distributions of fishes and other organisms in North and Middle America and in the larger context of Pangaean history, and are discussed in relation to current species concepts. [Vicariance; species concepts; biocladistics; biohistory; geocladistics; geohistory; Neotropics; Gondwanaland.]

A Cavernicolous Form of the Poeciliid Fish Poecilia sphenops from Tabasco, Mexico

LADIGES, W. 1958. Der malaiische Gurami Spaerichthys osphromenoides Canestrini. D. Aquar. u. Terr. Z. 11:3-7. MAGOUN, H. W. 1958. The waking brain. Thomas, Springfield, Ill. 138 p. MORRIS, D. 1958. The reproductive behaviour of the ten-spined stickleback (Pygosteus pungitius L.). Behaviour Suppl. 6:1-154. MORRIS, R. W. 1956. Clasping mechanism of the cottid fish Oligocottus snyderi Greeley. Pacif. Sci. 10:314-7. NIEUWENHUIZEN, A. V. D. 1958. Badis badis (Hamilton). D. Aquar. u. Terr. Z. 11:35962.

Evolution of Secondary Sexual Characters and Sexual Behavior Patterns in a Family of Viviparous Fishes (Cyprinodontiformes: Poeciliidae)

Aronson and Clark (1952) and Clark, Aronson and Gordon (1954) summarized the extensive behavioral work on these fishes, and they and Clark and Kamrin (1951) and Hubbs and Reynolds (1957) specifically studied the behavioral significance of fin and genitalic structure in Poeciliidae. Rosen and Gordon (1953) reviewed the mechanics of poeciliid genitalic and fin action in relation to copulatory motor patterns and provided an evolutionary interpretation of their findings. The present paper takes its origin from two separate reports, one behavioral and the other systematic3. It concerns the integration of structure and behavior in the evolution of the highly specialized, viviparous poeciliid fishes from a generalized, probably oviparous ancestor. The results of this study show, not unexpectedly, that the combination of comparative studies of behavior and functional anatomy may furnish insights on questions of evolution that do not emerge from either alone. The demonstrable effec-