Vicariance Biogeography in Mesoamerica

Abstract

This paper sets out to achieve three objectives. A brief introduction gives the present position of cladistic biogeography. Secondly, the most recent developments in cladistic biogeography for resolving complex and apparently incongruent species cladograms are presented. Thirdly, the methods as applied to the interrelationships of endemic areas in Mesoamerica with particular respect to the data of Rosen (1975, 1978, 1979) on poeciliid fishes is given. INTRODUCTION The analysis of spatial and temporal distribution of life on earth is the principal concern of historical biogeographers. Although many people would agree with this statement, there is still considerable disagreement amongst biogeographers about which methods should be used for the study of distribution and how distributions contribute information about the interrelationships and origins of areas. Distribution patterns can be analyzed in at least three different ways: first, we can compare the number of species or groups of species in different regions and say how many are common to each, thus getting a measure of overall similarity. Second, we can note the species and genera within one region and compare them with the species and genera of another and speculate where and how they might have arrived there. Third, we can make a comparison of the cladistic relations of various groups of organisms occurring in a group of areas of endemism to develop hypotheses on the historical connections of biotas in space and time-the pursuit originally identified by Nelson (1975) and Nelson and Rosen (1981) as vicariance biogeography and called cladistic biogeography by Parenti (1981). Historical biogeography, by any method, is an integration of taxonomy and historical earth science. A comparison of taxonomies in different groups is the basis for determining biogeographic patterns. Historical biogeography is thus an integral part of systematics and a slave to its theoretical foundations. Invariably, changes in historical biogeographic theory follow changes in taxonomic theory. There have been numerous developments in the post-war period and today major divisions exist between the practitioners of cladistics, evolutionary systematics, and phenetics with parallel divisions in historical biogeography between the methods of cladistic biogeography (e.g., Nelson & Platnick, 1981), dispersalism (Darlington, 1965), and area phenetics (e.g., Holloway & Jardine 1968). I would like to concentrate my efforts on the most recent and most thorough developments in systematics and biogeography-cladistics and vicariance. Most of the ideas presented here do not claim my originality, but are the ideas of three zoologists from the American Museum of Natural History-D. E. Rosen, G. Nelson, and N. I. Platnick-and one botanist, L. Croizat, now resident in Vene1 Department of Botany, British Museum (Natural History), London SW7 5BD. ANN. MISSOURI BOT. GARD. 69: 444-463. 1982. 0026-6493/82;0444-0463/$02.05/0 This content downloaded from 157.55.39.215 on Tue, 30 Aug 2016 05:37:43 UTC All use subject to http://about.jstor.org/terms 19821 HUMPHRIES-VICARIANCE BIOGEOGRAPHY 445 zuela, who have all done so much to put historical biogeography into a scientific framework. In the view of Patterson (1981), analytical systematics has been developed by three groups of workers: those workers who have adopted numerical methods for explicit phylogeny construction (e.g., Kluge & Farris, 1969; Farris et al., 1970; Farris, 1970); those taxonomists like myself influenced originally by Hennig's phylogenetic systematics (see Hennig, 1965, 1966); and those biochemists who have independently developed methods of reconstructing phylogenies from molecular data (e.g., Fitch & Margoliash, 1967; Moore et al., 1973; Fitch, 1977). These methods all have various features in common which distinguish them from narrative procedures. They all have taxa or empirical units as terminal branches in the relationship diagrams and the method for producing the diagrams is to interpret the most economical or most parsimonious distribution of homologies (i.e., characters, components, synapomorphies) (Patterson, 1981). As the basic method for producing the diagrams was developed independently in the three different fields, its superiority over narrative methods lies in its simplicity and its accessibility to criticism. By insisting on one property-homology, and only one criterion-parsimony, cladistics is as scientific a method as anyone can use in systematics. Narrative systematics on the other hand allows any explanation and thus immunizes itself against criticism. Patterson (1981) when reviewing the recent history of efforts to clarify the basis of historical biogeography concludes that cladistic biogeography is a development within cladistics. Analytical or vicariance biogeography is the pursuit of a method which encompasses a code comparable to cladistics. By gradually shedding a priori assumptions such as dispersal, extinction, allopatry, and sympatry it has increased our chances of understanding pattern and area interrelationships without circular evolutionary reasoning. In other words, if we are to understand the products of evolution we cannot use the existing principles to interpret patterns. By changing the word homologies to sister groups cladistic biogeography interprets the geographical distribution of sister groups parsimoniously. Different patterns when added together generate general hypotheses of area interrelationships. Thus, by using the two criteria of monophyletic sister groups and parsimony, cladistic biogeography is as scientific a method as anyone can use in historical biogeography. BIOGEOGRAPHIC PATTERNS IN MESOAMERICA Both centers of origin' and 'vicariance' hypotheses are biogeographical statements that presuppose that areas of endemism are interrelated and that these patterns are identifiable. What does it mean though to say that two areas of endemism are related to one another? We could pick on any two areas of the world and pick on any reason to say that they are biogeographically related to one another. For example, consider the southeastern states of North America and the tropical-parts of Central America (Fig. 1). The fact that they have many groups of organisms in common means that they are related. But, do numbers really mean anything? We could equally pick on one endemic species from each area and say that these two are related, if only by saying that they belong to the This content downloaded from 157.55.39.215 on Tue, 30 Aug 2016 05:37:43 UTC All use subject to http://about.jstor.org/terms 446 ANNALS OF THE MISSOURI BOTANICAL GARDEN [VOL. 69