VOL. I. ] Californian Lobeliacee. 375 
The appended notes of material accessible to me are furnished as a 
contribution to the better knowledge of these plants, and in the hope 
that directing attention to their unstable characters, sufficient data 
may be collected to determine whether even D. elegans and D. 
pulchella can be held sufficiently distinct. 
All the specimens here mentioned, excepting the first, are in the 
herbarium of the California Academy of Sciences, and even these 
few examples invalidate all the recently proposed species, including 
Dr. Gray’s D. bicornuta The color character cannot well be in- 
vestigated in herbarium specimens, but it is known to be infinitely 
variable in cultivation. 
Baillon is probably right in discrediting the generic value of the 
tufts or sete of the anthers in Lobeliaceze; they certainly differ very 
much in Downingia. The saccate protuberances of the lower lip,* 
so conspicuous in D. dzcornuéa, are found greatly reduced in other 
forms, and the position assumed,by the narrow, more or less deeply 
divided posterior lobes of the corolla is obviously unworthy of at- 
tention as a factor in classification. 
The most robust form of Downingia known to me is No. 289 
of the collection of T. S. Brandegee, from The Dalles, Oregon, 
about a mile below the town, in depressions, which are filled with 
water in the winter. They are a foot or more in height, with stout, 
branching stems, and probably represent the apparently little known 
D. corymbosa. The capsule is stout, ordinarily less than an inch in 
length, with broad bracts about one-third as long, and is hispid on 
the angles, though less so than in many of the Californian forms. 
According to Bentham & Hooker, the cotyledons in Downingia are 
conferruminate. In this form, as well as all the others of which we 
have mature fruit, they are either united to the tip and rounded at 
both ends or cleft for a variable but always short distance—both 
forms in varying proportions in the same capsule. In fact, the em- 
bryo scarcely differs from that of small lobeliaceous plants in gen- 
eral. 
In the typical specimen of D. concolor the tube is cleft no deeper 
at the back than at the sides, upper lobes not deflexed nor appressed 
‘to the sides of the tube; lower lip saccate, the slender posterior seg- 
ments of corolla variously disposed—rolled up, bent forward or 
backward, or even crossed. 
“In all these notes the apparent position of the flower is meant. 
