|
Murilo
Carvalho Dr. Richard Vari, Ph.D.
"It was a great experience to have had the opportunity to meet and work with some of the greatest ichthyologist in one of the World’s best museums." |
|
|
Murilo
Carvalho Dr. Richard Vari, Ph.D.
"It was a great experience to have had the opportunity to meet and work with some of the greatest ichthyologist in one of the World’s best museums." |
|
|
Application of CDA analysis to the catfishes of the genus Hypostomus in the upper Rio Paraná, Brazil (Siluriformes: Loricariidae) The
genus Hypostomus Lacépède, 1803, is the most
diverse group of the armored catfishes. Most species descriptions
of this genus provide limited or incomplete information, making correct
identifications difficult, if not impossible. According to literature,
32 nominal species of Hypostomus occur in the upper Rio Paraná
region, an area of demonstrated ichthyological endemism. As a consequence,
it is likely that at least some, and more likely a number, of the
species of Hypostomus that occur in the headwater streams in
this drainage basin are endemics, some of which may be undescribed.
A geographically focused taxonomic review of the Hypostomus
species in the headwater streams is a prerequisite to an understanding
of the diversity of the genus in the basin in light of the high species
diversity in the region along with the complicated taxonomic history
of the reported species. The species of the genus Hypostomus
that occur in the streams of the upper Rio Paraná in State
of São Paulo, Brazil, were revised, and one potentially undescribed
species was identified. Identification of the species as new necessitated
the application of multivariate statistics to the problem since many
of the characters in these fishes demonstrate a pronounced degree
of overlap. Canonical Discriminant Analysis was utilized in the study
and was found to be a very good methodology for addressing this type
of problem, allowing the species collected in the streams of the region
to be separated and confirm their identifications using data taken
from a number of type series. This research was supported by a grant from the Alice Eve Kennington Endowment. |