Biogeography of Humboltia

[John Grehan]

Fred Rickson asked for suggestions on the biogeographic analysis of
Humboltia. Here are some, although they are of a general nature since I
am not a specialist in the biogeography of the region or taxa in question.

>
>Humboldtia is a genus of small legume trees with 4 species distributed
>along the Western Ghats of south India (moist areas).
 DNA analysis shows the Sri Lankan tree sister to the most
>eastern of the Indian species, also an ant form.  On the basis of
>morphology the Sri Lankan species is also the most advanced of the five.
>No other species of this genus exist.
>So, it seems to me as if the south Indian area is ancestral and Sri Lanka
>is derived.

What is the biogeographic evidence for this? Is it not also possible that the
ancestor ranged over both areas and that the Sri Lankan species and its sister
species just represent the most recent differentiation rather than a
migration from
one to the other (at least that's usually what is impled by refering to one area
as ancestral). Further, can the Sri Lankan species be considered more
derived than
its sister species?

  Now, do I draw a track from the Indian sister species to Sri Lanka
>and claim vicariance across the former, but now submerged, land bridge, or
>do I just do the botanical thing and suggest a wanton seed sometime after
>the Eocene became established, and subsequently evolved into the Sri Lankan
>species?  I really don't have any idea of an answer to this distribution
>question.

Drawing a track between the taxa would just map the spaital structure of
their differentiation.
If one is interested in the possible history of the distribution it would
be neceesary
to consider a broader range of geographic information. What other taxa in
the region share
the same kind of geographic links? Do they share any common boundaries
(nodes) or
disjunctions? What is the distribution of taxa related to Humboltia? With a
general picture of biogeographic patterns in the region one might be in a
position to at least observe the level of generality of the genus. Along
with the spatial context is the pattern of biological relationships in
these taxa involed with the two areas. Further considerations might then
involve correlated geological and topographic features, which together with
ecological information might lead to predictions about their origin and
history.

  I may
>have some help because a proposed comparison of the invertebrates living in
>association with each species (within swollen, hollow, self-opening stems)
>might shed some light on whether whole plants and there pals marched across
>the land bridge

At present their seems to be a lack of biogeographic analsysis by which to
assert any marching across a landbridge.

John Grehan