42 - CACTACEZ. 
hardly more than one line in thickness; rim rather narrow, thick, but acutish. Var. Micros- 
perma has seeds only 0.8 or 0.9 line in diameter, more compressed, with quite a narrow rim. 
This species has, by western botanists, generally been considered identical with the eastern 
O. vulgaris. Riddell mentions it as occurring in Kentucky and Illinois, Torrey & Gray in 
their Flora do not give any locality in the Mississippi valley; but Rafinesque had already 
obseryed it in Kentuck , and, in his usual careless manner, had indicated 3 species: Cactus 
hum@fusus, (which growing, ‘‘ from New York to Kentucky and Missouri,’’ probably comprised 
both O. vulgaris and our species,) O. ceespitosa, from Kentucky and Tennesse, and O. Mesa- 
cantha, from Kentucky to Louisiana. As it seems impossible from his incomplete descriptions 
to make out what he meant by three different names, and as we know only one species in those 
States of the Mississippi valley, I take the liberty of discarding those names and of substituting 
the name of the author for the western species. It is not improper to state here that Rafin- 
esque’s vague and partly erroneous descriptions have found their way into Seringe’s Bulletin, 
1831, page 216, into the Linnwa, vol VIII, into Pfeiffer’s Enumeratio Cactearum, page 146, 
and into other works, but with the substitution of Nuttalls for Rafinesque’s name as authority ; 
the ‘‘ rounded joints’’ have, in these works, been taken for ‘‘ globose’’ or “‘ cylindric ’’ joints, 
and our plant has been classed with the Opu /ice glomerate from Chili and Mendoza. 
A large form of O. Rajinesquii was coll cted near Fort Smith, on the western border of 
Arkansas ; further west, where no true O. Rafinesquii has been found, several forms were met 
with, which, though they exhibit some distinctive characteristics, are, perhaps, not sufficiently - 
different to constitute distinct species. The flowers of most of them are unknown as well as 
the leaves, but fruit and seedSwere carefully preserved, which not only furnish important 
characters, but also the means to propagate, cultivate, and further to study them. We append 
them as dab-speciin 
1. RADICE FIBROSA. 
a. QO. CYMOCHILA : diffusa,articulis orbiculatis-pulvillis subremotis griseo-tomentosis stramineo 
s. fulvo-setosis, plerisque atmatis ; aculeis 1-3 robustioribus elongatis teretibus s, subcompressis 
tortisque albidis basi saepe rufescontibus, patulis deflexisve, additis saepe 2-3 gracilioribus 
radiatim deflexis ; flore? stigmatibus 8; bacca obovata umbilico plano s. parum depresso pul- 
villis 20-24 griseo-tomentosis parce setulosis, demum nudatis; seminibus irregularibns angulosis + | 
margine undulato acuto. (Plate XII, fig. 1-3 ¥ XX/l Z. SO F2. 4 
On the Camanche plains east of the ee Estacado, near the 100th dazrod of longitude, and 
from there to Tucumcari hill, 80 miles east of the Pecos. Joints 24-3 inches in diameter, 
orbicular or very slightly ohowabe pulvillif 6-8 lines apart; the very light yellowish-brown 
bristles numerous, and conspicuous pals on the older joints; only the lowest pulvilli of a joint 
unarmed, upper ones with 2-5 spines, 2 or 3 larger ones, often reddish-brown at lower half, 1-2 
inches long, lower, smaller, paler ones 3-9 Ries long. Fruit oval, 1-1} inches long, about 
10 lines in diameter, purplish, pulpy, sweet, and edible, less contracted at base than O. Rajin- 
esquit; seed remarkably irregular and twisted, 24 lines in diameter, with a wavy or twisted 
very sharp rim, whence the name which indicates the undulated border. 
The orbicular joints, the numerous spi:es, the oval not clavate fruit, and curiously twisted 
seed, seem to distinguish this form sufficiently from O. Rafinesquii, but these characters may 
not be sufficiently constant or important to constitute specific difference. The characters of 
Opuntie are not yet sufficently studied to permit usto form satisfactory conclusions about their 
diagnostic importance ; so we find a form collected on the Sandia mountains, near Albuquerque, 
which, in habit and appearance, does not differ from the common form of O. Rajinesquii, but 
which has the seeds of 0. cymochila: 
QO. cymocutta, 8. MonTANA: articulis orbiculatis majoribus inermibus s. margine superiore 
solum aculeatis; pulvillis remotis stramineo-setosis ; acleis singulis binisve validis albidis 
