Sapindacer.] THE HIMALAYAN MOUNTAINS, 137 
first found by him at Goalpara, specimens of, apparently, the same species were pro- 
cured by Dr. Wallich from Silhet and from Ava. 
The species of this family in Dr. Roxburgh’s Flora Indica are thirty in number, in 
which are included a few from the countries to the eastward of India. More than 
twice as many are enumerated in the unadjusted Sapindacee of the East-Indian 
Herbarium; but among these are necessarily included many of the described species, 
while others are probably only varieties, as are some, perhaps, of Dr. Roxburgh’s species ; 
but, of these, without having an opportunity of seeing their habits in the living state in 
addition to their appearance in a dried one, I am unwilling to*suppress any, from the care 
and acuteness which he usually displayed in the elucidation of species. It is seldom 
that any botanist enjoys the opportunities of Dr. Roxburgh, who, after studying the 
botany of the Peninsula, was transferred to the Calcutta Garden and the ‘tropical riches 
of the eastern frontier of Bengal. | 
The only genera which are found over all parts of India are, Cardiospermum, Sapindus, 
and Melicocca. Of the first, the species C. Halicacabum, which extends also into 
Arabia, is found in every part of India. Of Sapindus, though so tropical a genus, 
we find a species, S. acuminatus, in the most northern parts of India, having been 
found by Dr. Wallich in Silhet, at Bechiaco on the entrance into Nepal, and by 
myself, attaining a great size on the banks of rivers in the valleys within the Himalaya 
in 303° N. latitude. Melicocca trijuga (Schleichera, Roxb.) is found in similar situ- 
ations, being common at the foot of the hills in the Deyra Doon, and also in the 
most southern parts, as well as in the Peninsula. The species of Schmidelia (Ornitrophe, 
Roxb.) do not extend northwards, though common in Silhet, and southwards in the 
Malayan as well as in the*Indian Peninsula. Adlophyllus ternatus (Roxb. Fl. Ind. 2. 
p- 263), of which there is a specimen in the East-Indian Herbarium under Dr. Rox- 
burgh’s first name, A. ornitrophioides, appears from the characters which he gives of the 
genus, to be distinct from Schmidelia. The species of Dodonea seem to be entirely 
confined to the, Peninsula. — 
The Sapindacee having been divided into sections on account of differences in botanical 
structure, a like difference is observed in the properties of the plants belonging to the 
respective sections. Thus, of the Dodonea, Magonia pubescens and glabrata, according 
to M. A. St. Hilaire, are, like some of the Paulline@ which are acrid and narcotic, used 
as Paullinia pinnata for stupifying fish. Seriana triternata is employed for the same pur- 
pose. Paullinia australis, and some other species of the genus, are said to be poisonous ; 
and P. cupania Humb. is used for making an intoxicating liquor. P. subrotunda is an 
exception, in furnishing, like so many of the following section, an edible aril. Of 
these, the Sapindee, the species of the genus Sapindus, are remarkable for so many of 
them being used as a substitute for soap. Thus, in America, the S. Saponaria, and in | 
Java S. Rarak, and in India S. acuminatus, laurifolius, emarginatus and detergens (the 
last, according to Dr. Roxburgh, nearly allied to S. Saponaria,) are all used for the 
purposes of soap; owing, it is now ascertained, to the presence of the vegetable prin- 
T ciple 
