62 NEILGHERRY PLANTS. 
in the intervals between two crops, and hence, that the Tobacco plant must yield some 
secretion fitted to stimulate into active operation the latent vitality of the seed. This 
deduction is so self-evident that it seems unnecessary to insist further on it, beyond merely 
remarking, en passant, that if the stimulating properties of the secretions of one plant 
are necessary towards exciting vegetation in the seed of another, there seems no yery 
obvious reason why they should not prove injurious to successive crops of itself, or one 
of the same genus or even order. At all events, it is a well known fact that natural forests 
when, from whatever cause, destroyed, are almost invariably replaced by plants belonging 
to widely different families. 
CHRISTISONIA. 
Calyx tubular, quinquifid, equal or sublabiate. Corolla hypogynous, tube funnel-shaped, limb 5-lobed, 
bilabiate. Stamens didynamous, inserted on the tube of the corolla, all fertile, incluse, or rarely exserted ; 
anthers 2-celled, one polleniferous, dehiscing at the apex by an oblique pore, the other sterile, prolonged into 
an acute spur. Disk none. Ovary ovate, oblong, l-celled: placentiferous margins deeply inflexed, revolute 
within the cell; ovules numerous. Style filiform, simple; stigma bilabiate, or orbicular. Capsule enclosed 
in the calyx, sub-globose, 1-celled, 2-valved, dehiscing loculicidally, and bearing the placenta on the middle 
of the valves. Seed numerous, oblong, obtuse, supported on a short thick funiculus; outer seed-coat loose, 
men, orthotropous. Cotyledons short, obtuse: radicle thick, blunt. Herbaceous plants growing parasitically 
on the roots of other plants. Stems short, simple or ramous, scaly below, floriferous towards the apex; 
flowers large, rose-coloured, or yellow, or deep purplish blue; pedicels racemose, The following essential 
character of the tribe and genus is copied from my Illustrations of Indian Botany. 
Hropancnrs. Ovary imperfectly 2-celled, that is, the inflexed carpels only partially meet in the 
axis; while the placentiferous margins, remaining free and spreading to the right and left, form two broad 
lamellar placente. 
Cupistisonta. Calyx tubular, 5-toothed. Corolla infundibuliform, sub-bilabiate. Anthers 2-celled, one 
sterile, subulate. Placente free, revolute. 
We are indebted to the researches of the late Mr. Gardner of Ceylon, for the separation of this genus 
from Phelipea with which the few species, previously known to Botanists, had been combined. 
When he defined the genus and published it in the 8th volume of the Calcutta Journal of Natural 
History, he was of opinion that it belonged to the natural order Cyrtandracee, with which it certainly 
accords in many respects, and accordingly indicated that as its proper place in the natural system. Subse- 
quent consideration and a more intimate acquaintance with Oro ew led him to coincide with me in 
thinking it better located in the latter, with which it accords in both habit and structure. It in truth goes 
far to combine three large natural orders, Gesnerace@, Orobanchacee, and Scrophulariacee, to either of 
which, with the exception of habit, it might be referred. 
The species, so far as yet known, are few, about 10, but I feel certain that many others will yet be 
found. One species, as already mentioned, is abundant on the north-western slopes of the Neilgherries. 
Ihave not met with it elsewhere. It comes very near Gardner’s C. bicolor, if indeed it be not that very 
species, from which, on recomparison of the drawing with the character, I find it mainly differs in the 
Position or rather direction of the sterile cells of the anther, in mine horizontal, in his erect; a distinction 
scarcely of specific value. The stigma also seems to differ, but I fear more in words than substance. 
He views the stigma as 2-lobed, but with one of the lobes frequently aborting, which I do not think 
the case; but taking that view, he describes the stigma of his C. bicolor as “2-lobed, the upper lobe abor- 
tive, the lower flattened, somewhat triangular, emarginate,” a character which does not accord with that 
shown in the plate, which is peltate, umbilicate, somewhat 2-lobed; the lobes slightly emarginate. Not- 
U 
