There are only 72 flowering plant species occurring in Bahia, Brazil, which are listed by the Species Survival Commission of IUCN as Endangered, Threatened or Vulnerable.  Our work has revealed that there are far more species that will be shown to meet IUCN's Red List criteria.  To determine these species, the following questions had to be answered:

• What are the species endemic to southern Bahia and adjacent Espírito Santo?
• Exactly where do each of these species occur (that is, what are the geographical coordinates of each collection)?
• Are there patterns of endemism or areas of unusually high occurrence of endemic species?

We have evaluated each of the thousands of species known to occur in southern Bahia and northern Espírito Santo.  To date, we have identified and mapped 258 species with highly limited distributions that will probably qualify for the IUCN Red List, including the 72 already listed.  All the listed species are of limited distribution, occurring only in Bahia and northern Espírito Santo.  All pertinent herbaria were consulted, including the herbaria of The New York Botanical Garden, CEPEC, MBML and CVRD in Espírito Santo, and the Rio de Janeiro Botanical Garden.

For each species, we had to select all available specimens in the CEPEC and NYBG herbaria, ensure they were correctly identified, and determine the exact geographical coordinates for each collecting locality if they were not listed on the specimen label.  This involved locating the collection site on 1:100,000 topographic maps when ever possible.  Then, all the data was transferred into a spreadsheet and, later, into a database format readable by GIS software - we thank Holly Porter-Morgan for invaluable assistance with GIS software and species mapping.  The data for each species was plotted on to a map (with one to several species per map).  One such map shows the limited distribution of three species of the genus Byrsonima (savanna shrubs in the Malpighiaceae).

Once all the data had been mapped for each species, the real anaylsis of the data begins.  Are there specific localities that are rich in endemics?  Are their certain habitas that have more endemic species?  Are there certain plant families that are especially rich?  How does the distribution of these endemics correlate with the distribution of established reserves?  Are there gaps, places where we might predict the occurrence of some species, but which may need to be better collected?

These are the questions we are attempting to answer now.  In an analysis of distributions by plant family, we analyzed all the collection points for fifteen species of the Orange Family (Rutaceae); the data were summarized in a single map to analyze the distribution of the family.  We can see several interesting features.  The first, and most disturbing is the number of species known only from one collection.  Are they poorly collected, are they rare and hard to find, or are they truly limited in distribution to a single forest?  We don't yet know, but the pooling of all the mapped data shows:

• that most collections were made near the coast,
• 2) species that occur near the coast tend to occur only there, or with an outlier in a well-collected area of similar forest (e.g., Conchocarpus diadematus, found in the Serra do Teimoso forest) farther from the coast, and
• 3) species known from only drier forests farther inland (e.g., C. oppostifolius or C. santosii) are not found in more humid forests nearer the coast.

The linked map shows a complete distribution of all 1,452 data points on a single map.  The collections are grouped in ten minute (ca. 20 miles) grid squares.  This provides an indication of where most collections of endemic species have been made.  What is harder to discern, however, is whether this distribution simply represents where most collections have been made, or whether there is actually a higher number of endemic species here.  This will permit us to begin assessing collecting sites and specific habitats for possible conservation.

The Next Steps

We are now in the process of analyzing our data and formatting it for use by conservation organizations. We are also checking distributions of these species in other herbaria, especially the herbarium of the Rio de Janeiro Botanical Garden - this will ensure that the data we have are as accurate and complete as possible.  Our findings for each species and for the patterns of endemism in the region will be published as rapidly as possible.  These will include recommendations for IUCN Red List status categories for each species.  They will also include prioritization of areas in southern Bahia and northern Espírito Santo for conservation based on the diversity of endemic species found in each area.