2 Contributions from the Gray Herbarium 
The need of a general revision of the Amsinckias of North 
America will be apparent when it is stated that no complete treat- 
ment has appeared since that by Dr. Gray in the Synoptical Flora 
ii. pt. 1, 197 (1886) where he recognized only six species and two 
varieties, practically all that had been proposed at that time. 
Since then, however, over a dozen species have been described, 
mostly from California, and although some of these are apparently 
not valid this study has disclosed five more plants which seem to: 
require characterization. As a result twenty-one species are 
recognized, three times as many as were recorded in 1886. The 
justification of this augmentation will be questioned, especially 
since species in this group have been proverbially difficult of dis-. 
crimination. However, it is believed that this state of affairs has. 
been induced, at least partly, by the lack of a comprehensive treat- 
ment in which the chief characters are brought into contrast. 
Nevertheless it must be conceded that characters which admit of 
clean-cut statement are all too few. The ready response of all the: 
species to every change in ecological conditions makes it difficult 
to determine the reliable diagnostic characters. However I have 
become convinced that shape of leaves, size of corolla and nature 
of pubescence are features that are of the utmost importance, in. 
spite of their supposed susceptibility to environment, for the dis-. 
crimination of species. In view of the fact that fruit-characters. 
are unusually important in the delineation of the components of 
many boraginaceous genera it might be expected that they would 
serve as a means of discriminating species in this genus but with few 
exceptions the variations in the character of the nutlets have not. 
proved stable nor definite enough to warrant their use in classifi-. 
cation. Altogether Amsinckia is the most perplexing group I have 
studied and I can only hope that this effort to define its natural, 
components may lead to careful field-work by some one who may 
then be in a position to prove or correct my interpretation. It is. 
conceivable, with the better understanding gained from field-. 
observation of specific limitations within the group, that it will be 
found to consist of many more species than the twenty-one here 
recognized. I feel, however, that in a genus in which the species 
lend themselves so readily to many diverse environments and are: 
subject, consequently, to the stimulus for great variation, specific 
lines must not be drawn very fine lest the classification become 
artificial and thus fail to serve a useful purpose. 
