SEEDLINGS 
1 1 
surely to be expected, seeing that Himalayan Rhododen- 
drons are mostly social plants, and it is borne out by 
our experience with the old-fashioned R. ponticum, which, 
when kept apart from other species and hybrids, con- 
tinues true for many generations. Certainly, for the first 
generation, the seedlings of Himalayan Rhododendrons are 
often produced in great numbers, almost absolutely alike. 
Sometimes this uniformity descends to later generations, 
and sometimes even to hybrids. Thus Mr. J. Sheppard has 
raised about 150 hybrids between R. Edgeworthii and R, for- 
mosiim ; most of the seedlings have bloomed, and he can see 
no difference whatever in any of them. I must not, however, 
at present go so far afield, but confine myself to the asser- 
tion that carefully collected native seed will, when grown 
in this country, faithfully reproduce the parent type. 
The circumstances attending the collection of wild 
Rhododendron seeds, and the fact that ripe seed and blos- 
soms are seldom to be seen together, will explain how 
many smaller differences have escaped the observation of 
collectors who have gathered the same species from different 
localities, from different elevations, or perhaps from different 
sides of the same valleys. Such an explanation is surely 
more philosophical than the notion that so many variations 
have arisen — not gradually, but all of a sudden — from 
cultural influences. My readers will apprehend how in- 
teresting botanically it is to find such a variety of types 
packed together in so small a compass, for instance, as 
Sikkim ; but they will, of course, remember the character 
of the country and its climate, where alpine and almost 
tropical conditions occur side by side. The Heaths at the 
Cape of Good Hope give an example of great variation in 
the same family, but Dr. Becarri's remark on the Rhodo- 
