Geography Reference
In-Depth Information
seem to decay through time as new ones are superimposed (Grande 1985;
Hunn and Upchurch 2001; Upchurch and Hunn 2002; Upchurch et al. 2002).
Donoghue and Moore (2003) postulated that cladistic biogeographic meth-
ods are susceptible to the confounding effects of pseudo-incongruence and
pseudo-congruence if they do not incorporate information on the absolute
timing of the diversification of the lineages.
Pseudo- incongruence
means
that different area cladograms may show conflict when the taxa evolved at
the same time but diversified in response to different events.
Pseudo-con-
gruence
means that different area cladograms may show the same area re-
lationships although the taxa diversified at different times, presumably un-
der different underlying causes. However, Riddle and Hafner (2006) argued
that time alone might not necessarily cause us to resort to an explanation of
pseudo-congruence.
Cladistic biogeographers avoid using temporal data because of the risk
of incorporating ideas of unobserved processes in the elucidation of biogeo-
graphic patterns. This would imply unverifiable assumptions, with the risk of
falling back on narrative scenarios. However, the need to consider time in
biogeography becomes clearer in cases of biogeographic convergence. The
terms
convergence
and
divergence
were proposed by Hallam (1974) to dis-
tinguish two extreme biogeographic patterns. Widespread taxa and redund-
ancy identify biogeographic convergence, whereas vicariance is the most
common interpretation of divergence patterns. Convergence can be the res-
ult of area coalescence (due to the elimination of geographic barriers). Ana-
lyses of biogeographic convergence are unlikely to show congruence.
Upchurch et al. (2002) noted that biogeographic analyses over an ex-
tensive stratigraphic range may fail to find the correct area relationships.
This point is illustrated by the hypothetical succession of area separations
followed by area coalescence described by Upchurch and Hunn (2002),
where branching relationships are evident only when extensive dispersal
hasnotyetoverwhelmedtheoriginalvicariancepattern.Infact,ifbioticcom-
ponents A and C, formerly separated by vicariance following the pattern (A
(B, C)), merge while the third biotic component B remains isolated, the fi-
nal relation will be uninformative because we will have two biotas and two
areas: A1 (= A + C) and B. Area coalescence causes vicariance patterns
to fade out through time (Grande 1985). Biogeographic convergence leads
to a sort of biogeographic overprinting, and therefore analytical techniques
based on parsimony algorithms may be inappropriate (Young 1995). When
previouslymergedbioticcomponentssubsequentlyundergovicariance,pat-
