242 The Botanical Gazette. [April, 
North American Cactacee. 
Our North American species of Cactacee are now brought together! 
as a result of the studies of Dr. John M. Coulter. 
The first part (no. 2, published in June, 1894) contained the genus 
Cactus (Mamiliaria) and its small outliers, Anhalonium and Lopho- 
thora; the second part (no. 7, published in April, 1896) completes the 
work by presenting Echinocactus, Cereus, and Opuntia. The revision 
is called “preliminary” because confused and inadequate material, a 
badly tangled synonymy, and paucity of types could result in nothing 
else. It is also preliminary in the sense that it professes to do little 
more than to bring together the widely scattered material, sift it s0 
far as possible, and thus lay the foundation for more elaborate study. 
It is probable that no group among the higher plants presents greater 
difficulties in the way of classification, both on account of the meaget 
and fragmentary material, and also on account of the almost entirely 
unknown possibilities of variation. Having been very largely culti- 
vated in Europe as well as America many modified forms have been 
produced, many- garden species and varieties have been described, 
and these have vastly complicated the work of revision. It is certall 
that very many forms described, both in the revision before us and 
elsewhere, will be found undeserving of specific and varietal rank, but 
this can be discovered with certainty only by long continued and pa 
tient scientific cultivation. In few families are there such poorly oF: 
fined generic lines, even our well-known North American genera pie 
hibiting most puzzling intergrading forms. It may be safely said that 
in Cactacez there are few good species, as that term goes, 40 in 
genera. The monographer had a rare opportunity in his access ® 
Dr. Engelmann’s types and notes, many of the latter being first pub 
lished in the present revision. If the North American species north 
of the Mexican boundary were the only ones concerned, some Teas? 
able degree of certainty might be reached, but the well-nigh unknow? 
ground. Taking up the three genera of the Confribution just i 
we find Echinocactus containing fifty-two species and varieties, ap 
