Reverend William Herbert, to whom we are indebted for 
the specimens from which our drawing was made. We 
learn from Mr. Herbert that the stem is woody, and sends 
out frequently horizontal or drooping runners, which cross 
along to the furthest extremities of the hothouse, rising 
up here and there to the top of the building, and again 
hanging down from the wires upon the rafters. 
A tender stove plant, flowering during most of the 
summer months. Our drawing was made in August last. 
Propagated, we presume, by cuttings. 
The remarkably dilated upper lip of the calyx, the 
minute lower lip, and the monadelphous stamens, distin- 
guish Canavalia satisfactorily from the other genera re- 
cently separated from Dolichos. The species now figured. 
differs from the essential character ascribed to the genus 
by M. Decandolle, in the lower lip consisting of one 
minute tooth-like process, instead of three. We are not 
acquainted with the fruit; but we learn from Mr. Herbert 
that the seed was large, and very similar in size and colour 
to that of Wisteria frutescens. ai : 
