A REVISION OF PHILIPPINE CONNARACE/E. 
By E. D. Merrill. 
( From the Botanical Section of the Biological Laboratory, Bureau of Science, 
Manila, P. I.) 
This family is a small one so far as the Philippine flora is concerned, 
but like most groups of plants found in the Archipelago, has long been in 
need of revision. Our herbarium contained numerous unclassified speci- 
mens, and there was considerable doubt as to the proper specific name to 
use in the case of several identified species. Blanco’s imperfect descrip- 
tions have always been the cause of more or less doubt as to the identity 
of his species, and F.-Villar’s erroneous identifications of these have 
added to the confusion. In the case of extant herbarium material, no 
less than three specific names have been published by as many different 
authors, all based on a single number of Cuming’s Philippine collection, 
No. 851 , while Cuming 1172 has had two specific names applied to it. 
Blanco described the first Philippine representatives of the family, five 
species, all of which he placed under the genus Cnestis. Two of them are 
properly referable to this genus, although reducible to a single species, 
but the other three are referable to Connarus and Rourea. 
F.-Villar, in the Novissima Appendix to the third edition of Blanco’s 
Flora de Filipinas, enumerates twelve species in four genera, but only 
five of these actually occur in the Philippines, so far as the study of 
material now available shows. Most of Blanco’s species were erroneously 
reduced to species that do not extend to the Philippines. 
In the present paper five genera and seventeen species are recognized 
as occurring in the Philippines, but the list of both genera and species 
will undoubtedly be considerably increased as botanical exploration of the 
Archipelago progresses. With the exception of two, or possibly three 
species, all of those enumerated below are endemic in the Philippines. 
There are apparently also two or three additional species of Connarus, 
probably undescribed, but so far represented in our herbarium by im- 
perfect material, so that it is not deemed advisable to consider them at the 
present time. 
117 
