160 
BOTANY: J. ROSENBAUM 
P. infestans, 
P. cactorum, 
P. phaseoli, 
P. nicotianae, 
P. syringae, 
P. fagi, given as a synonym of 
P. cactorum in Saccardo.^ 
P. arecae, 
P. parasitica, 
P. faberi, 
P. jatrophae, 
The characters to be relied on in taxonomic work must be constant 
under all conditions and not correlated with certain adaptations to 
habitat. In the work here set forth separation and relationship of spe- 
cies are made on the aggregate of characters, it being borne in mind that 
the proportionate value to be attached to each character must neces- 
sarily vary. 
The most valuable characters that can be employed are doubtless 
those which are exhibited by the reproductive organs, and which fur- 
nish means of separating the genus into different groups. Other 
characters which have been found can be conveniently used in sepa- 
rating the various species, arranged as nearly as possible in the order 
of their importance, are (1) the size, and morphology of the sexual 
reproductive bodies, conidia, chlamydospores and other structures; 
(2) shape of the terniinal papillum in the conidium; (3) the ratio of 
the length to the width of conidia; (4) variation in the mycelium, 
and (5) the macroscopic growth, time of appearance of spores and 
kind of spore forms produced on various media. 
As artificial media were most generally employed, it seemed impor- 
tant to determine whether these media influenced the morphology and 
size of the various structures to any appreciable extent. Preliminary 
trials were made with P. infestans on potato foliage and artificial media, 
but no appreciable differences between the spore production on the 
potato foliage and those grown on artificial media could be detected. 
As far as possible, all the forms were subjected to identical conditions. 
Measuring spore forms to delineate species in the various fungous 
groups is a well established practice. In forms less variable than those 
of Phytophthora a small number of such measurements might be suffi- 
cient to determine the average size. The conidia of Phytophthora, 
however, are so variable that a small number of measurements is apt 
to be misleading. For this reason four-hundred measurements were 
made of each form, including the length and width of conidia, the ratio 
of the length to the width and diameter of oospores and chlamydospores. 
On account of the great variability in the individual measurenients, 
biometrical methods were used to obtain constants more readily com- 
