A 
1913] Brainerd,— Viola arenaria 111 
Dr. Gray had before him only fresh and well-dried specimens from 
Howell and Suksdorf. As a matter of fact, good specimens of V. 
Howellii, markedly punctate with minute dots, are to be seen in the 
herbaria both at St. Louis and at New York; in the latter herbarium 
is also an older specimen,? marked in pencil by Dr. Gray as “large 
V. canina." 
The acaulescent appearance that V. adunca sometimes assumes at 
vernal flowering has been a cause of further confusion. The American 
caulescent species that bear violet-colored flowers all belong to the 
group “RosuLANTEs,' marked by bearing in spring at the crown of the 
rootstock a cluster of leaves. From the axils of these the stems later 
appear, and often petaliferous flowers on long scape-like peduncles, 
especially if the growth of the stems is retarded. These undeveloped 
forms are easily mistaken for stemless species. Even Dr. Gray, thus 
misled, has in the Synoptical Flora placed V. Langsdorfii ( a stemmed 
violet of the Northern Pacific coast, allied to V. adunca and the Eura- 
sian V. mirabilis) in the section of species "strictly acaulescent." 
Dr. Greene has published as a species (V. filipes) a seemingly 
acaulescent form of V. adunca. (Pitt. iv. 289.) On the other hand, 
Howell and others have distributed V. nephrophylla, a truly stemless 
species as V. Howellii, supposing it the form with undeveloped stems.— 
In Garcke's German Flora, at the end of his treatment of the stemless 
violets, there is a sign-board, warning every one not to locate there 
forms of V. mirabilis that are stemless at first flowering? We may 
need to have a like caution inserted in our American manuals. 
MIDDLEBURY, VERMONT. 
EXPLANATION OF PLATE 104. 
1. Viola adunca J. E. Smith, (E. Brainerd, Bristol, Vt., May 26, 1912), X $. 
‘A. stipule X 2; B. immature capsule X 3. 
2. Viola rupestris Schmidt var. arenaria (DC.) Beck.; after Schlechtendal 
revised by Hallier, fig. 1273. X $. C. stipule X 2; D. capsule X 2; 
both after Reichenbach, Pl. crit. Ixxii. 
1 Pt. Reyes, Marin Co., Cal, July, 1903, A. D. E. Elmer; labeled ''V. cuneata.” 
? Swamps at Noyo, Mendocino Co., Cal. Bolander, coll. 1867. 
3 Man hüte sich, die zuerst blühenden, stengellosen Pflanzen von V. mirabilis 
hieher zu rechnen.” 
