and with much longer cilia; Lindley suggests its being a 
variety of (. barbatum. A comparison of these excellent 
drawings in the Register suggests that lanciferum, bar- 
batum, and cornutum may be local forms of one variable 
species, and Garnettianum another, but very closely allied 
species, possibly to be connected by intermediates not 
hitherto known on cultivation. As to their localities, 
Garnettianum, barbatum, var. proboscoidewm, and lanct- 
ferum are natives of Brazil, the latter from the province. 
of Goyaz, and the variety from Sertao; whereas C. bar- 
batum itself and C.cornutwm are native in Demerara, the 
ng of the Massarony River, near the Falls of Wapo- 
pekai. 
More interesting in a scientific point of view than the 
variations in the perianth of Cataseta, is the now well- 
known sexual dimorphism of almost all the species of the 
genus. This dimorphism was misinterpreted by Mr. 
Darwin. The subject has been lately investigated by Mr. 
Rolfe, Assistant in the Herbarium of the Royal Gardens, 
and the results have been read before the Linnean Society, 
and alluded to in the “ Gardener’s Chronicle,”’ 1889, i. 407. 
Mr. Rolfe divides the genus into four sections, of which 
he has been so good as to give me the hitherto unpublished 
characters. They are :— 
1, Eucatasetum. Flowers unisexual. Column of ¢ 
with a pair of deflexed filaments. Lip of both sexes 
posticous. 
2. Myanthus. Flowers unisexual. Column of das in 
Hucatasetum. Lip of 3 anticous, of ? posticous. 
8. Heirrhose. Flowers unisexual. Column of male 
without filaments (? flower unknown). 
4. Pseudo-catasetum. Flowers hermaphrodite. 
C. Garneitianum belongs to the second section, The 
figure of it here given is from a plant presented to the 
Royal Gardens by P. F. Garnett, Esq., of South Bank, — 
Grassendale, Liverpool, who received it from the Amazons 
River in North Brazil in 1888. It flowered in a tropical @ 
house at Kew in the month of November.—J. D. H,. 
Fig. 1, Lip and column; 2, pollinia :—doth enlarged. 
