Feb. 12 ,1917 
Studies of the Genus Phytophthora 
245 
table gives the class in microns, the number of conidia out of a total of 
200, both for length and width, falling into each class, and the results 
of combining the data of sets A and B. In figures 1 to 6 curves are 
plotted for the length and width of conidia in microns, and the number 
of conidia falling into each class out of a total of 400 measurements. 
These curves were obtained by combining the data given in Table II 
and making the classes differ by 4 n instead of 2 fx. They show at a 
glance the variation that may reasonably be expected in each species, 
33.3- 
3/. 
/S.3 
II* 
u \ 
v 
N 
■V 
T 
V 
X, 
\ 
■ L 
1 
/ 
/ 
r 
• 
V 
N 
V 
S ( 
\ 
4 
Ji 
\Bt-i 
. / 
K, 
s 
*•* 
* «*■ 
•* 
> 
- f 
i 
• 
r 
t 
j 
i 
X 
> 
-V 
✓ 
X 
X 
r 
/ 
/ 
0 * 
E? X 
17 0 
E? * 
V 3 
V € 
a? ? 
7 6 
fc? 3 
0 A. 
*7 // 
t7 /i 
V A 
/vUMse# <30 aw/0//* ovroF* mr/u orioQ. 
Fig. 2.—Graph showing the variation in length of conidia of 
Phytophthora phaseoli, P. cactorum (.Phyllocactus sp.), P. 
faberi, and P. erythroseptica. 
as well as the class containing the greatest number of individuals for 
each species. Further examination of these curves shows that some of 
them, as, for example, the length of P. jatrophae , exhibits a mode of 
47 n and of 55.5 fx. This is probably not to be interpreted as meaning 
that the normal-sized variation curve is a multimodal one, as too few 
observations or too numerous classes may cause a similar condition. 
A similar condition is indicated by the length of P. nicotianae . Theo¬ 
retically all these curves are smooth, and probably would be if an in¬ 
finite number of measurements were made. 
