Biogeographic memory

[John Grehan]

>
>One of the reasons why I find Panbiogeography difficult to understand may
>be that it uses a lot of terms in a sense that is distinctly different
>from common usage. Thus,  Panbiogeography builds a "parallel universe'  of
>terms, which is hardly accessible from this one.

I agree. The new research program has necessitated both new terms and new
definitions for old terms. This appears to represent a normal function in
science where new research programs are instituted  - as acknowledged by
Hovenkamp. There was the same kind of hurdle with the introduction of
cladistics, and more recently molecular systematics. These problems seem to
be ephemeral for the most part, but can understandably be frustrating and
even confusing. My involvement in panbiogeography necessitated quite a lot
of time being devoted to understanding meaning and terminology. Some
efforts have been made to clarify what may be understood of
panbiogeographic terms, although these will undoubtedly fail to be
satisfactory to everyone. Problems with panbiogeographic terminology appear
to be a function of one's perspective and experience, and certainly the
more traditional one is in biogeography the more problematic the meaning
and terminology of panbiogeography appears to be.


>Thus, "dispersal" does not mean dispersal, and "tracks" are not tracks,
>"homology" is not homology (not quite, at least), "center of origin" is
>not center of origin (or is it?), "main massing" need not be main nor
>massing (see p. 79, where a single occurrence in the Comoros is treated as
>a "main massing"),  and now "phylogeny" turns out to be not phylogeny (but
>note how in the examples of Ch. 4 the term phylogeny seems to be used in
>its conventional meaning).

These terms are simply not what they might be assumed to be in traditional
Darwinian biogeography.

>I realize that such terminological confusion is inherent in the
>establishment of any new research program, but I would have expected a
>more coherent account of these terms in a book of this kind, summarizing
>40 years of research after the first establishment of its research program.

I guess 'coherence' is a judgement that the individual must make. Some
people think we did quite well, others not.


>Here our ways part...

So there is quite a clearly delineated contrast in biogeographic
methodology that may provide for some interesting debates and analyses in
the coming years.


>>>Would not that be enough to dismiss the claim that "their distiributions
>>>are not distinguishable according to their different means of dispersal"?
>>
>>Not in my view.
>
>Again - how different must two patterns be to be distinguishable? My
>earlier question about this was ignored by John Grehan.

Sorry, not deliberately.My view is that any two patterns 'must' be
distinguishable (potentially) according to a lack of biogeographic
homology. In panbiogeography this homology criterion is not geographic
overlap of distribution range, but corresponding or nested overlap of track
geometry, and shared baseline characters.

>I see a lot of light between Mayr's statement that different dispersal
>abilities result in totally different distribution patterns (did he really
>make such a catergorical statement?

Yes indeed.


>>Since similarity of distribution range is not part of panbiogeographic
>>method
>
>But similarity in baselines is. How is it possible to assign "the same"
>baseline to different tracks if there is no notion of similarity? And that
>similarity is decidedly arbitrary.

As quoted above, I was excluding similarity of distribution RANGE. The
similarity for shared baseline is for the tracks to cross the same
geographic  feature. Since the track is drawn according to the minimal
distance criterion, the selection of a baseline is not arbitrary in the
sense, for example, that a distribution either side of the Atlantic cannot
be given a Pacific baseline.

>Anyway, I never intended to question the obvious discrepancies between
>obvious means of dispersal and sizes of distribution areas -merely the
>choice of the example in fig. 1-1, which I still think does not illustrate
>the point very clearly.

Perhaps you are right. At least this discussion allows one to consider
using better examples in the future or providing a better explanation of
the example used. There's always room for improvement one may hope.


>But I asked for a summary of Croizat (1968), optionally incorporating the
>data that have become available since.

It appears that kind of treatment is yet to come. Then perhaps one may
contrast views expressed by Turner et al with respect to the following:

"New Guinea has a very special history" whereas panbiogeographic aporoaches
suggest a similar kind of history to NZ, Borneo, Philippines, Japan,
California etc.),

"to disperse from e.g. Sri lanka to Tonga it is almost unavoidable to pass
through India, Himalayas, SE Asia, Malay Peninsaula, Sumatra (or Java or
Borneo), Sulawesi, New Guinea..." when in fact there are thousands of
disjunctions along this track.

John Grehan