102 
8. C. robustum, Williams. 
OC. morrisonense, Hayata. 
10. C. arvense, Linn. 
ll, C. schizopetalum, Maxim. 
= 
Critical. 
Having had the opportunity of examining at Kew most of the 
original specimens of Cerastium from Japan as well as the valuable 
specimens referred to in the list given above, I have been able to 
make a careful study of the various Japanese specimens in connec- 
tion with those of my own collection. Before entering 1 
detailed account and criticism of each plant, I should like to preface 
my remarks with a few words as to the characters of systematic 
importance in the genus. Species of this genus cannot be distin- 
guished by only a few characters. The habit, hairiness of stem, 
size and shape of leaves, characters of bracts, length and direction 
of pedicels, proportion of length between sepals and petals, size and 
shape of petals and direction of capsules, must all be consulted ; still 
they are to some extent not constant. Great care must be paid in 
examining the direction of the pedicels and capsules ; the pressing 
of the plants often causes them to assume a quite unnatural direc- 
tion. 1t is also my conviction that the size and marking of mature 
seeds afford a very good distinctive character, though the various 
kinds of marking are rather difficult to describe satisfactorily in 
words. 
specimens were not always at my disposal. 
According to Dr. Hayata the specimens of C. pilosum collected 
perennial, though weak annual specimens are often met with. 
So far as I am aware the Japanese form of this plant, and especially 
usually has leaves oval in form like those figured in Iwasaki’s 
Dzufu* and Tin 
Id. op. xlviii., fol. 14 verso. — + Id. op. viii. fol. 71. 
d. tab. xxx., g. 1. : § Id. op. i., 2, tab. 42 
| Id. op. v.-vi., fig. 4972. 
es 
