ASBS Newsletter – Book Review

Historical Biogeography of the Southeast Asian Genus Spatholobus (Legum.-Papilionoideae) and its Allies.

written by J.W.A. Ridder-Numan Blumea Supplement 10.

(From ASBS Newsletter Number 91, June 1997)

Published by: Rijksherbarium / Hortus Botanicus, Leiden, Netherlands, 1996.

The Rijksherbarium in Leiden is justly famous for its longstanding programme of high quality research on the flora of Malesia (the southeast Asian mainland and the archipelagos stretching from there to New Guinea). Major outcomes are Flora Malesiana and its companion journal Blumea, in which is published the paper reviewed here. This research is impressive in its breadth and depth. It integrates floristics, monographic revisions, phylogenetics, comparative biology and biogeography. It is systematics in the broadest sense. Thus it is not suprising that Leiden has produced leading theoreticians such as van Steenis, Zandee, Roos, Geesink, Kornet and Turner.

Leiden has also produced more than its share of biogeographers. They would have been inspired in part by C.G.G.J. van Steenis, who produced major papers in the 1960s and '70s on the phytogeography of Malesia and the Pacific. His collaborator M.M.J. van Balgooy, who is still active, followed with a series of books on plant patterns in the Pacific. In the last decade a series of workers have produced dissertations integrating phylogeny and biogeography of Malesian plant groups. A nice example is an analysis of the western Pacific flora and its origins by three of these authors (Van Balgooy et al. 1996). The present paper continues this tradition.

Malesia is a special place in the history of biogeography, for this is where Alfred Russel Wallace spent the most productive years of his working life (van Oosterzee 1997). Here he independently discovered the theory of evolution by natural selection. Not only this he also founded the modern science of biogeography and identified one of the most important biotic boundaries in the world. Today this line carries his name, and we now know that it results from a monumental collision between chunks of the long-separated supercontinents Gondwana and Laurasia, reuniting their biotas. However, we are far from understanding fully the complexities of this region's history, as the present paper shows.

This paper by Jeanette Ridder Numan is a beautiful example of the legacy of Wallace and van Steenis, for it is a study integrating evolution, biogeography and geology. It contains the greater part of her Ph.D. thesis, which by Dutch convention was itself published, but in a limited edition (RidderNuman 1996). Only the chapter on pollen morphology has been omitted from the Blumea paper, and has been submitted to the Review of Palynology and Paleobolany. Initially Ridder-Numan's study was supervised by Rob Geesink, and it further develops his work on the primitive legume tribe Millettieae. Sadly and prematurely the supervisor died before the student completed her thesis.

The paper is divided into three parts: a cladistic analysis of Spatholobus and allied genera, a review of the geological history of the Malesian region, and a biogeographic analysis that integrates the previous two parts. It is thorough, meticulous and very detailed throughout. The methods used are by and large sound and up to date, however this is an empirical study. For explanation and discussion of theory and methods, other works are referred to, such as Turner (1995). (Hubert Tumer's thesis is an excellent example of the recent theoretical contributions from Leiden.)

Appropriately, the phylogenetic analysis in part one of this work is based on a monograph of the study group, completed some years earlier by the same author (Ridder-Numan and Wiriadinata 1985). The study group consists of all the species in three genera: Spatholobus (29 spp.), Butea and Meizotropis (2 spp. each). From earlier cladistic analyses, these are known to comprise a monophyletic group, and the closely related genus Kunstleria is used as the outgroup. A thorough knowledge of the group is revealed by the large set of morphological (80), anatomical (10) and pollen (7) characters used. A good rule of thumb is that one needs at least twice as many characters as terminals in a cladistic analysis, and this data set is well above that limit. Thirty five pages, several excellent line drawings and some large tables and graphs are devoted to a detailed discussion of the 97 characters. This section epitomises the thoroughness of the study and is recommended as a model for similar studies. My only quibble is with the discussion of the nine quantitative characters. Several of these are graphed, clearly showing the continuous nature of the variation, yet there is no discussion of how the states are defined. A substantial literature on the problems of 'gap-coding' continuous characters is overlooked (Gift and Stevens 1997 and references therein).

The cladistic analysis itself is a standard parsimony analysis using PAUP. Encouragingly for such a large data set, only three parsimonious trees were found, with minor differences due to uncertain placement of a couple of species. All three ingroup genera are shown to be monophyletic. One of the three trees is chosen as 'best', based partly on a character weighting procedure proposed by Turner (1995), and partly on intuition. The tree is evaluated in detail in terms of the support of branches by the characters. Again I quibble because none of the widely used tests of phylogenetic robustness are employed here (e.g. Bremer support, T-PTP tests or the Felsenstein bootstrap), even though some of them are available in PAUP. However, the dynamic character-weighting cum tree-searching procedure of Goloboff (implemented in his program Pee-Wee) is tried, but the resulting trees are intuitively unsatisfying - they do not even show the genera to be monophyletic.

Part two of the paper is a review of the geological history of the region in which the study group occurs. This comprises the Indian subcontinent and west Malesia (excluding New Guinea). A very large amount of literature is cited, and this review would be a useful reference for anybody interested in the biogeography of this region. It is supported by and excellent series of detailed maps, showing plate boundaries, subduction zones, sea levels and terranes at various time slices from the Palacozoic to the present. It is not light reading, for this region must have one of the most complex geological histories in the world. It is thought to have accreted from many small terranes, most of which rifted north from Gondwana over a very long period, commencing in the Palaeozoic or perhaps even earlier. Of course, major impacts on the region have been caused by the collision of first India and then Australia. Even today it is a highly active zone, as frequent reports of earthquakes and volcanic eruptions testify. Figure 3.17, mapping only the Philippine-Moluccan region, shows eleven currently active subduction zones!

Part three (historical biogeography) is the main guts of the paper. Here an attempt is made to explain the origin, spread, fragmentation and diversification of the study group by reconciling its phylogeny and distribution with the geological history. This section has two parts: (1) a component analysis of several taxa including the present study group, with the aim of deriving a general area-cladogram that shows the history of vicariance events in the study area; and (ii) a reconciliation of the Spatholobus group with both the general area cladogram and the map to elucidate the unique history of this group.

Like the first two parts of the study, this analysis is thorough and rigorous. The critical first step of defining areas for analysis is discussed in detail, with citation of some but not all of the relevant literature. The standard problems, such as empty areas and areas containing widespread species, are canvassed. It is stated that delimitation of areas is based on distributional discontinuities but the method seems subjective or at least unclear (this is no different from most studies). However, it is noted that many (but not all) the areas correspond with geological units described in the previous section.

Several methods are available for deriving general area cladograms, all are flawed and vigorous debate continues about them. Ridder-Numan chooses Brooks parsimony analysis (BPA), which is in widespread use, and component compatibility analysis (CCA),whose use is largely restricted to the Leiden school. With both these she uses two different rules ('assumptions' 0 and 1) for dealing with the confounding effects of widespread distributions and missing data. Phylogenies input for analysis are not only the Spatholobus group but additionally some from distantly related taxa, with the aim of finding a more general area-cladogram - one that reflects a history of externally imposed vicariance, rather than one-off dispersal events that only affected the history of Spathlobus. The results of the four different analyses have some area-clades in common and some differences. None is completely resolved. Again using largely subjective reasons, she prefers one of these (that from BPA with assumption '0').

I would not dare to report here in all its complexity the general area-cladogram preferred by Ridder-Numann. It shows an early history of vicariance events in what is now mainland Asia, and later ones among the islands. She attempts to reconcile this with the geological history, which is a brave act, given the complexity of the latter. More interesting perhaps is her final section, a reconciliation of the Spatholobus cladogram with both the general area-cladogram and the geological history. Again these are too complex to report here. Of course, this section is highly speculative, but was probably fun to do. Briefly, she hypothesizes the origin of the group (and of all three genera) in the early Tertiary of mainland Asia, perhaps before the impact of India. From there Spatholobus migrated to the islands of Malesia in a series of episodes mediated by rifting, lower sea levels and corridors of suitable vegetation, and differentiated gradually into the extant species. The lineage even appears to have migrated back to mainland Asia (e.g. Indochina), and differentiated further there.

It would be difficult to test such a detailed scenario. One way would be to use a molecular clock to date the nodes on the cladogram, and to test these against the dates of the matched vicariance events. Ridder-Numan is careful to avoid hypothesizing anachronistic events. For example she notes (p. 128) that the area patterns shown by her taxa do not reflect the preangiosperm (Triassic) collision of Sibumasu with Indochina / East Malaya, whereas geological entities within Borneo, which assembled during the Tertiary, are evident in the distributions of her species. (Sibumasu is an ancient fragment of north Gondwana that today comprises parts of east Burma, west Thailand, west Malaya and west Sumatra.)

A cynic might dismiss the final scenario thus. The geological history of the southeast Asian region, comprising numerous plate splits and collisions, climatic changes and sea-level fluctuations, is incredibly complex. Therefore it would be possible to construct a plausible scenario of vicariance and dispersal events to fit any taxon's phylogeny and distributional pattern. However, I am not a cynic. We now know far more about the history of Malesia than Wallace did, and as more taxa are studied with the rigour of Ridder-Numan and her colleagues, the many pieces of this gigantic mobile jigsaw will gradually be put into place.

References

Gift, N., and Stevens, P.F. (1997). Vagaries in the delimitation of character states in quantitative variation - an experimental study. Systematic Biology 46, 112-125.

Ridder-Numan, J.W.A. (1996). Historical biogeography of the southeast Asian genus Spatholobus (Leguminosae-Papilionoideae)and its allies. Rijksherbarium/Hortus Botanicus.180. (Leiden University: Leiden.)

Ridder-Numan, J.W.A., and Wiriadinata, H. (1985). A revision of the genus Spatholobus (Leguminosae-Papilionoideae). Reinwardtia 10, 139-205.

Turner, H. (1995). Cladistic and biogeographic Analyses of Arytera Blume and Mischarytera Gen. Nov. (Sapindaceae). (Leiden University.)

Van Balgooy, M.M.J., Hovenkamp, P.H., and Van Welzen. P.C. (1996). Phytogeography of the Pacific - floristic and historical distribution patterns in plants. In 'The Origin and Evolution of Pacific Island Biotas, New Guinea to Eastern Polynesia: Patterns and Processes'. (Eds A. Keast and S. E. Miller.) pp. 191-213. (SPB Academic Publishing: Amsterdam.)Van Oosterzee, P. (1 99 7). 'Where Worlds Collide: The Wallace Line.' (Reed Books: Kew,Victoria.)

Reviewer: Mike Crisp
Botany and Zoology
Australian National University
[Received 10 June 1997]