NEILGHERRY PLANTS. 99 
XIX.—SAPINDACEA. 
This is a large order, exceedingly tropical in its habits, and interesting in an econo- 
mical point of view from several of its species affording edible fruit, and others the well 
known detergent known under the name of soap-nut. The Lichi and Ramboatan are exam- 
ples of the former, the fruit of several species of Sapindus supply the latter. It is not, how- 
ever, my intention to dwell on this order, as, with one or two exceptions, all the species be- 
longing to it are found on the plains or lower slopes; and being, therefore, purely tropical 
plants, scarccly come within the scope of these remarks on alpine vegetation. One species 
of the family, Dodonoea viscosa, certainly does ascend to the higher levels ; but being much 
more abundant on the plains, and having no other peculiarity beyond its power of adaptation 
to different climates to reecommend it to our consideration, need not be further noticed 
here. It may however be remarked, in passing, that it differs from most of the other plants of 
the family, in having simple, not compound leaves, the predominant form in this order, and in 
having no corolla. The want ofa corolla is probably of less note than the other, as the flowers 
generally are unsymmetrical, parts being wanting or irregularly formed. The ovary is pretty 
constantly 3-celled, which, combined with the irregularity of the flowers, forms the _prin- 
ciple distinguishing character of this family. This, however, is not constant, as in the genus 
Schmidelia two cells is the usual number. The following character and figure of that genus 
will serve to illustrate these peculiarities of the order. 
SCHMIDELIA.—Jnss. 
Sepals 4, unequal. Petals 4, the fifth or superior one deficient, and its seat vacant, either naked on 
the inside or usually furnished with a scale above the unguis. Disk incomplete, with 4 glands opposite the 
petals. Stamens 8, inserted ou the receptacle, and connate round the ovary at its base. Ovary usually 2-, 
sometimes 3-lobed: style from between the lobes of the ovary, 2. 3-cleft, the segments recurved, longi- 
tudinally stigmatose on the inside. Fruit indehiscent, 1-2, or rarely 3-lobed: lobes somewhat globose, fleshy 
or dry, l-celled. Seeds with or without an arillus,—Trees or shrubs, usually.with trifoliate, rarely with 
simple, exstipulate leaves. Flowers white, small, in axillary racemes.—W. and A. Prod. P. 10 
The species of this genus are generally somewhat rambling shrubs, and when growing in thickets’ 
elimbing among the trees and bushes often to a considerable height. When growing in open ground, they 
are low bushy shrubs. They resemble each other so closely that their discrimination is generally most 
difficult. The one here represented affords a striking example of this fact; a passing good figure of it was 
published by Rheede in the Hortus Malabaricus abont 150 years ago, and yet, from that time until now, his 
plant has never been recognized, while the figure itself has been quoted by almost every writer on Indian 
Plants for one or other of the recognized, species. In our Prodromus it is quoted as a synonym for S. 
Cobbe with the addition “ not good” which is most true, but for the plant here figured it is good. I have 
therefore dedicated the species to the memory of the original discoverer, feeling quite certain, at the same 
Ume, that it is his plant, partly from its likeness to his figure, and partly from haying found it in Malabar as 
Well as on the hills, 
