Published in

Wiley, Journal of Biogeography, 3(34), p. 437-446, 2007

DOI: 10.1111/j.1365-2699.2006.01604.x

Links

Tools

Export citation

Search in Google Scholar

Biogeographical relationships among tropical forests in north‐eastern Brazil

This paper was not found in any repository, but could be made available legally by the author.
This paper was not found in any repository, but could be made available legally by the author.

Full text: Unavailable

Green circle
Preprint: archiving allowed
Orange circle
Postprint: archiving restricted
Red circle
Published version: archiving forbidden
Data provided by SHERPA/RoMEO

Abstract

Aim  To use parsimony analysis of endemicity and cladistic analysis of distributions and endemism to evaluate two hypotheses addressing biogeographical relationships among Amazonia, the Caatinga forest enclaves, Pernambuco Centre and the southern Atlantic Forest.Location  North-eastern Brazil, South America.Methods  To find the most parsimonious areagram we analysed a matrix composed of the presence (1) or absence (0) of 745 taxa (i.e. 293 genera and 452 species of woody plants) within 16 localities belonging to the four large regions addressed in this study.Results  One most parsimonious areagram was found and it shows a basal separation between the southern Atlantic Forest and all other regions. This break is followed by a separation between all Caatinga forest enclaves (except Baturité) from a cluster composed of Baturité, the Pernambuco Centre and Amazonia. In this cluster, the most basal separation isolates Baturité from the cluster formed by localities from Amazonia and the Pernambuco Centre. The biogeographical relationships among sites could not be explained by either a random distribution of species among sites or by the geographical distance between sites.Main conclusions  We found strong cladistic signal within the raw distribution and phylogenetic data used in our analysis, indicating structured species assemblages in the surveyed localities. They have resulted from the fragmentation of an ancestral biota that was once widely distributed in the region. Our results also support the hypothesis that Atlantic Forest is not a biogeographically natural area, because the Pernambuco Centre is more closely related to Amazonia than to the southern Atlantic Forest. Finally, our data do not support the notion that Caatinga forest enclaves comprise a single biogeographical region, because one Caatinga forest enclave (Baturité) is much more closely related to the cluster formed by Amazonia and the Pernambuco Centre than to other sites. These relationships suggest the occurrence of forest connections between Amazonia and the Atlantic Forests across Caatinga during several periods of the Tertiary and Quaternary. However, palaeoecological data currently available for the Caatinga region are still scarce and do not have either the spatial or temporal resolution required to reconstruct the history of connections among the forests in north-eastern Brazil.