Palaeobiogeography: using fossils to study global change, plate tectonics, and evolution (Topics in Geobiology; vs. 16) by . New York : Kluwer Academic/Plenum Publishers , 2000 . 208 pp. B&W illustrations. Hardback. £50.00
‘A science of dispersal is only as wide or as narrow as the preoccupations of those who entertain it. Those who look to it for a pleasant account of sundry theories and curious cases of distribution may rest satisfied with much less, of course, than those who intend to reach the bottom in everything of nature they touch’
Croizat (1958; p. 1)
The premise of this volume is that by interpreting the fossil record we may infer changes in geographical patterns and continental drift. This goal is an interesting and exciting undertaking, though the book disappoints from the stage of data interpretation due to Lieberman’s blinkered methodology.
Plate tectonics has had a profound effect on the Earth. It shapes and moves continents, creates mountain chains and forms large oceans. From an evolutionary perspective it creates and destroys geographical barriers and isolates populations of species. Theoretically we should be able to discover that some taxa were isolated by a geographical boundary (allopatry) or not (parapatry, sympatry, etc.). The known isolation mechanisms that may cause allopatry are vicariance and dispersal. One way in which we can uncover patterns due to allopatry is to replace the names of the taxa with the areas in which they occur, thereby creating taxon-area cladograms – a first step into the relatively complex science of cladistic biogeography.
Lieberman’s book gives a somewhat biased account of the history of palaeobiogeography, a subject that developed during the 19th and 20th centuries, with respect to his preferred biogeographical and evolutionary models. The book also examines how we can use combinations of several taxon-area cladograms to find geographical patterns. However, the focus of Lieberman’s work is to leave the reader with a clear working knowledge of how to generate certain geographical patterns amongst equally parsimonious taxon-area cladograms and how to individually interpret each node (see Ebach & Humphries, 2002).
The importance of finding several taxon-area cladograms and combining them to form area cladograms is not only to uncover information about the biogeographical history of a group. It is to find a common relationship between areas, to discover geographical congruence. The congruence we find in cladistics, as in the relationships among taxa, is the same that we find among areas. The methods Lieberman uses to find such congruence are problematic. These are geodispersal (Lieberman & Eldredge, 1996) and Brooks Parsimony Analysis (BPA) (Wiley, 1987) and the combination of both is the main thrust of his work.
Geodispersal is seen to result from the collapse of geographical boundaries. The disappearance of boundaries brings faunas previously isolated into contact, thus forming congruent range expansions across a variety of groups (p. 22). However, the collapse of barriers and the movement of faunas across such fallen barriers are no more than a mixture of dispersal and vicariance.
Lieberman emphasizes geodisperal as a separate concept, contrasting it to a predominantly vicariant cladistic biogeogeography. This is a misinterpretation. All forms of geographical isolation are assumed within cladistic biogeography, whether vicariance, dispersal or a complex mixture of both, all are probable causes for geographical congruence, or indeed incongruence. Lieberman claims that geodispersal is the only cause of geographical congruence. Not only is this subjective, but it is also unnecessary. The concepts of dispersal and vicariance form the basic premises of geodispersal. To present these basic premises as rivals or to choose one over the other does not aid in finding geographical congruence.
The next part of Lieberman’s book interprets taxon-area cladograms. Cladograms are based on homology and the congruence of its units, character-states. Phylogeny provides an insight to the interrelatedness of taxa, and its nodes are the shared common characters of taxa and clades. To assume that nodes are ‘ancestors’ would compromise the process of cladistics (see Nelson & Platnick, 1984). The same is true of areagrams. Although they share the same topology and branching structure of a cladogram, they are not functional cladograms, that is, the nodes do not represent synapomorphies. We cannot assume that areas are related by homology, any more than we would we assume that nodes are ‘ancestors’. Lieberman assumes that nodes in area cladograms indicate real isolation/speciation events. He claims to be able to clearly identify geodispersal, vicariance or extinction at each node. His logic proceeds to claims of finding migration patterns and actual speciation events, the equivalent of finding ancestor-descendant lineages within the basal nodes of cladograms. The method by which his area cladograms are generated is called modified BPA (mBPA).
The simplicity of BPA and mBPA is appealing. The ease with which general area cladograms can be generated is perhaps the biggest draw card in the book. The biogeographical occurrence of each taxon within a cladogram is coded as characters within a matrix. If the area is present, it becomes an apomorphy and/or a potential synapomorphy, and if the area is absent it is treated as a plesiomorphy. Hence BPA treats the presence of areas as homologies, and the areas as taxa. The difference between BPA and any normal cladistic procedure is minimal, even the same parsimony programs are used to implement BPA. The method seems logical at first and we are reassured by the ease with which we can find biogeographical signals. Perhaps we should ask how do we find a biogeographical signal when we treat areas exactly as though they were taxa? We cannot. Areas do not behave in the same way. By overlooking the possibility of other unknown isolation mechanisms that may be unique to each organism we go on creating a version of evolution and history that may be seriously flawed. We must accept that the biogeographical signal is small. We can only hope to find geographical patterns expressed as the relative interrelationships of areas, and cannot use each node to represent specific events.
Choice of BPA and mBPA methodology demonstrates the conviction that biogeographical history of a living or extinct group can be worked out solely by using cladograms. Surely this attitude is overly simplistic, given the incredible variety of possibilities over vast tracts of time.
Lieberman has valid cause to look for biogeographical congruence. He valiantly tried to give us reasons to explain biogeographical patterns, however, he has chosen the easy path by glossing over the incompleteness of the fossil record, while claiming to have a vision of evolution in action. Lieberman believes that there is a signal, but, he does not seem to acknowledge that it is difficult to find nor that his results are entirely generational. Geographical congruence is still a worthy goal, and we may be able to find it using other more subtle, objective methods. For me, the path to understanding is far removed from Lieberman’s ‘pleasant account of sundry theories and curious cases of distribution’. Uncovering biogeographical patterns in time will prove far more problematic and complex, but it is a journey worth undertaking for ‘those who intend to reach the bottom in everything of nature they touch’.