NEW OR NOTEWORTHY SPECIES. 95 
ably accepted as typical of Andersonii the grumose-rooted 
plant of the Sierra Nevada, which Gray had confounded 
with it, and named as a part of it. 
SOPHIA Sonnet. Sisymbrium incisum var. Sonnet, Robin- 
son, in Gray, Syn. Fl. i. 140. From the typical SopHra 
INCISA (Sisymbrium incisum, Engelm.)this plant ofthe eastern 
borders of California appears constantly to differ in more 
important particulars than those of its leaf character. Its 
short and very few-seeded pods are remarkably and quite 
eharacteristically acute; and the valves are both extraordi- 
narily firm in texture and almost reticulately venulose. It 
grows in company with S. HARTWEGIANA (Sisymbrium Hart- 
wegianum, Fourn.) with no apparent intergrading between 
the two. 
The adoption of this type of plants as a distinct genus by 
recent authors must be eminently satisfactory to all students 
of the Cruciferze who study them in the field as well as in 
the herbarium, thereby becoming convinced as it were by 
nature herself, that differences of habit are of prime impor- 
tanee in distinguishing genera. But the name Descurainea 
taken up in the Engler & Prantl series of monographs was 
not well selected. It is long antedated by Descuria, while 
SoPHIA is much older than either. 
Nropeckia aquatica. Cochlearia aquatica, Eaton, Man. 
9 ed. 181 (1829). Nasturtium natans var. Americanum, Gray, 
Ann. Lye. N. Y. iii. 223 (1836). Nasturtium lacustre, Gray, 
Gen. Ill. i. 132 (1848). Roripa icem Mem. 
Torr. Club. v. 169 (1894) Not many generaNin the in- 
tricate family of the Crucifere are more satisfactory than 
Ror ipa, if it be duly restricted according to what is indicated 
in nature. I have,in the Bay Region Manual, shown its ~ 
essential character, and have excluded, as seemed indispen- : 
sably necessary, the water-cress and horse-radish types, point- j|. 
‘ng out their characters. And the above eastern type is even 
