In order to document a tropical forest flora, each species in the flora must be documented through collection.  Because the flowers or fruits of many species (especially trees, lianas, and epiphytes) are difficult to see and reach, collecting in tropical forests is a difficult and time consuming activity.  Many tree species are widely dispersed (e.g., one individual per hectare) and their flowering period is brief.  This necessitates repeated visits to document the flora.

To select collecting sites, we use all available information, including maps and satellite images, to analyze elevation, soil type, and vegetation type. Geographical coordinates of collecting sites are determined as precisely as possible using GPS devices (if readings underneath the forest canopy are not possible, readings are taken from a clearing near the forested collection site).

At each site, we search for species in flower or fruit.  These are collected unless we are sure that they have been previously collected (this minimizes the repeated collection of common or weedy species).  Fertile plants are sought by walking through each site and searching on the ground for fallen flowers or fruits or using binoculars to find those remaining in the trees.  In cases where trees, lianas, or epiphytes cannot be collected from the ground (aluminum pole pruners can be used to collect material up to ca. 10 m above the ground), trees are climbed.  Specimens are pressed in the field, numbered, catalogued, and dried at preparation facilities.  Collection and specimen drying is done to maximize the preservation of DNA material for possible analysis.  Flowers are often preserved in FAA or ETOH when appropriate.  Photographs are taken to document plant features not easily preserved through standard collecting and drying.

Specimens usually are collected in sets of four to eight duplicates.  Half of the duplicates of any collection and any unicates remain in Brazil (at CEPEC) as dictated by Brazilian law and are distributed to specialists and interested herbaria within Brazil.  The remaining specimens are sent to NY for distribution to the international botanical community.  At NY, the best sheet is  mounted for the NY collection and the second is sent to the appropriate specialist (the current NY specialist list includes 185 botanists identifying material in 180 families of plants) for identification.  For groups in which no specialist exists or for which identifications are not returned in a timely manner, material is identified by project staff.  Once a collection is identified, duplicates are distributed to herbaria whose acquisition policies emphasize the flora of tropical Brazil or the taxonomic group of the specimen.