TWO NEW SPECIES OF POPULUS. 409 
form of leaf and others the other form ; and I have sinee repeat- 
edly discovered gradations between them. It will be seen that 
the cordate base of the second form nullifies the only character 
given in Gamble's ‘Manual’ for discriminating between this 
species and P. ciliata, Royle, P. Gamblei grows at an elevation 
of 1000-4000 feet. The photographs and examples exhibited 
show the two forms of leaves and the female inflorescence. 
The most interesting characters of P. glauca are the 
frequency of 2-sexual charaeters and the large development of 
the perianth. The Poplars are usually described as diccious 
without any qualification, while Pax, in Engler u. Prantl, * Die 
natürl. Pflanzenfamilien, remarks that the separation of the 
sexes is complete. Sargent, ‘ Silva Americana,’ ix. p- 151, states 
that “individual trees bearing staminate and pistillate amenta, 
and also amenta with staminate and pistillate flowers mixed 
together, occasionally oceur in the United States.” But I know 
of no record of 2-sexual flowers. 
I first observed a tree of Populus glauca in June, in ripe fruit. 
I visited this tree again on 20 April, the following year, in full 
flower; the flowers were all 2-sexual and 1 in. in diameter; the 
so-called dise is expanded into an unequally 5-7-fid membrane 
with segments 3-¢ in. long. There were 6-12 stamens inserted at 
the base of this perianth on its very short tube; these are 
practically hypogynous. At the same date (20 April) I examined 
trees of Populus glauca, distant about 10 miles from the first- 
discovered tree, scattered for two miles along a ridge. I examined 
13 trees. Of these, 4 were entirely female; the others were 
entirely 2-sexual, or with both 2-sexual and female flowers. 
There was not a single male tree. This scarcity of male trees 
has also been observed in the case of Populus ciliata, Royl 
Very many of the 2-sexual flowers on these trees had only 1, 2, 3, 
or 4 stamens. The flowers appeared protogynous. 
lam greatly indebted to Dr. Stapf aud Sir D. Brandis for 
having kindly undertaken for me the arduous task of ascertain- 
ing whether the name of “glauca” proposed by me for this 
species was preoccupied or not, and for the above reference to 
the ‘Silva Americana.’ 
The localities mentioned are all in Sikkim and Darjeeling 
District, or adjacent thereto. 
2H 2 
