Comment on biogeography of Humphries

John Grehan jrg13 at PSU.EDU
Fri Nov 10 17:15:14 CST 2000

Humphries, C. 2000. Form, space and time; which comes first? Journal of 
Biogeography 27, 11-15.

In this article Humphries makes a number of interesting comments both about 
biogeography and Panbiogeography. For those interested, I offer some comments
concerning Humphries’ views about the current state of biogeography and the
qualities of particular approaches. I realize that my views will be in the 
and accept that most practitioners will probably see the issues differently.

Humphries notes that there are a variety of different biogeographic methods 
and it is hardly surprising to him that Tassy and Deleporte suggest 
biogeography was in a mess,
a subject looking for a method.

        An alternative view might be that biogeography is not in a mess at all. 
Just    because there is no tyranny of consensus does not mean a mess. As far 
as I can        see it is a very, very, very good situation for biogeography to 
have a lot of   diversity.

Humphries identified the relationship between form, space and time as the 
modern ‘three fold parallelism”.

        It is the sequence of this relationship (that Croizat first brought into
        focus) that presents an important difference between panbiogeography and 
other   biogeographic frameworks. As I have commented before on TAXACOM, 
many    biogeographers, including vicariance cladists, don’t really seem to 
like geography
        (space) so that for them form comes before space and time, whereas Croizat 
saw     space and time as central to understanding form  hence the biological 

Humphries notes that Croizat eschewed the somewhat arbitrary areas of 
endemism of de Candolle, Wallace etc.

        I don’t know about “somewhat”  either they are or they are not.

Humphries characterization of the modern ocean basins providing the 
boundaries of former historical units is problematic.

        I am not aware that they have been circumscribed as geographic areas by 
Croizat         or      Craw, but rather they are biogeographic homologies.

He asserts that panbiogeography has yet to shake off the impediment of 
ancestors as part of the explanations it avowed to replace.

        The comment is one of the most intriguing as its meaning is obscure (and 
to at least     one other biogeographer I have contacted).

Humphries asserts that Nelson and Platnick synthesized panbiogeography with 
the systematics of Hennig.

        This remains one of the most problematic claims of vicariance cladists. 
In      rejecting       Croizat’s spatial technique, vicariance cladists seem to 
have removed    any     foundation for “synthesis”.

A key statement is his conclusion that historical biogeography is about 
classification of areas amongst biological and spatial co-ordinates.

        This is problematic since vicariance cladists reject spatial analysis in 
favor of        biological cladograms, and in panbiogeography there are no areas 
of endemism.

Humphries characterizes much of current biogeography to be narrative with 
fossils and centers of origin remaining popular.

        This is something with which I can find some agreement as I have noticed 
the c   continuing plethora of papers of this kind. Perhaps it is not really 
surprising. Most        of      the historical biogeography is generated by 
systematists, and perhaps most  systematists are not really interested in 
patterns of biogeography in general.    Hence the number of papers dealing 
with taxa as unique events of origin and        dispersal  cladistics or no 

Humphries principal criticism of panbiogeography is the claim that it does 
not make a clear statement as to what constitutes relationship between 
different areas on the earth, and rather than provide material evidence 
(whatever is meant by that) for homology it still kings to the mysteries of 
ancestry to hypothesize relationships of areas.

        I’m not sure whether Humphries really believes that there is no “clear” 
(again  whatever that might be) statement, or it’s really a matter that he 
knows what      constitutes relationship in panbiogeography (simply minimal 
distance with respect   to      main massings and baseline features for one or 
more taxa) and disagrees with it.

Cladistic biogeography is offered by Humphries as the alternative since its 
“clear” correspondence between systematic relationships in “similar” areas 
it provides biogeographically informative relationships of the biotas.

        The problem here is that while one may classify the hierarchical order 
of      biological      relationship, the biological cladogram contains no spatial 
information by which to         determine which geographic sector links the areas. 
It is Morrone and his colleagues        who have confronted this problem by 
suggesting that a panbiogeographic analysis is  necessary before biological 
area relationships can be interpreted historically.

Humphries sees the future for biogeography in biological and geological 
cladograms while at the same time noting that the method is fraught with 
problems of combining taxa into general area patterns, delineation and 
recognition of areas, and optimizing multiple simple sequences.

        All these problems are supposed to take us “one step further down the road 
to a    solution than the panbiogeography program which ultimately sees 
made  up of
tracks and nodes of life’” Indeed I support the 
concept of biodiversity as a    complex of tracks and nodes as something 
empirical and unproblematic. Croizat    (1958) already provides the main 
outlines of biogeographic patterns and homology         (that no one whether 
dispersalist or vicariance cladist) has yet refuted)

Unlike Humphries, I see the future for biogeography as more diverse than 
just trying to figure out “a precise idea of what constitutes the relations 
(area homologies) and the things being related (areas) in the analyses. 
With panbiogeography one can be less worried about such conundrums. The 
concept of relations is clearly designated as baseline homologies (spatial 
correlation of tracks and geophysical features) and things being related 
(taxa at any level mapped as tracks). So for now I’ll stay with 

These comments present a purely personal perspective. I do not presume to 
persuade anyone else of the merits. Food for thought if nothing else.

John Grehan

More information about the Taxacom mailing list