4 PITTONIA. 
tion; but Dr. Torrey's name being appended as authority for 
the names of the two species indicates that he, at that time, 
was entertaining the thought of founding a genus upon these 
plants ; an opinion which it is evident that he shortly after- 
wards relinquished ; for only a few years later, in preparing 
his elaborate report upon the Botany of the Wilkes’ Exploring 
Expedition he referred the same plants to Echinocystis. As 
late as 1875, two years or more after Dr. Torrey's death, Mr. 
Sereno Watson published in the Proceedings of the American 
Academy, what was indexed as a Revision of Megarrhiza ; 
but this paper, so far from being the revision of a previously 
established genus, is the very first appearance of it as charac- 
terized. 
Our Pacific American plants differ from the Atlantic type 
of Echinocystis only in their more turgid seeds, hypogzeous 
cotyledons and perennial tuberous roots. In the seeds there 
is displayed every gradation between the obovate and orbicu- 
lar, and from nearly globose to much compressed. Eminent 
botanists who have given special attention to the Cucurbi- 
tacec agree in the opinion that the species in question form 
no more than a section of Echinocystis. In the natural orders 
most nearly allied, Cactacez and Loasacez for example, much 
more striking differences in the character of seeds are allowed 
in a genus. To take the case of Mentzelia, the diversity of 
seeds, all the way from nearly shapeless to thin, orbicular, flat 
and winged on the one hand, and to almost exactly cubical on 
the other, is manifold greater than what we shall have in 
Echinocystis as allowed by Bentham and Hooker and Cog- 
niaux. 
It is to be hoped that, in the small matter of the priority of 
subgenerie or sectional names, no strifes may ever arise; but 
I ean not conjecture what need there was seen to be of naming 
this $ Mana, as M. Cogniaux has done in his admirable mono- 
graph, when Bentham and Hooker had, years before, applied 
MEGARRHIZA to that use. 
The species, now in fair number, should be named as fol- 
lows. . 
