172 BALFOUR—NEW SPECIES OF PRIMULA. 
“Yunnan. Hills N.W. of Tengyueh. Alt. 6000 ft. Lat. 
25° to’ N,” G. Forrest. No. 11,904. April 1913. In Herb. 
n. 
“Upper Burma. Feng-shui-ling Pass. 8000 ft. On open 
or shady banks by streams, or in marshes, on fallen tree trunks, 
etc. 5th June 1914. Flowers sulphur-yellow, pendent. At- 
tains a height of 14-2} ft. Flowers May and June.” F. K. 
Ward. No. 1635. In Herb. Edin. 
This grand species is now in cultivation through seeds sent 
by George Forrest to Mr. J. C. Williams of Caerhays Castle. 
Seedlings have been raised in abundance. It closely resembles 
the East Himalayan P. Smithiana, Craib, which is altogether a 
smaller plant ; but the two plants will, I think, when we know 
more of them, prove to be geographical microforms of one 
aggregate. 
Type specimens sent by Forrest show that the plant is variable 
both in foliage and in the bracts. The most of the plants under 
number 7561 have the smaller rosulate leaves described above, 
and also the small bracts much shorter than the flower pedicels. 
The later specimens under Nos. 9802 and 11,904 have most 
of them the longer leaves and the longer bracts far exceeding 
the pedicels, and becoming more or less foliaceous. Could one 
have correlated without question the leaf and bract form there 
might have been grounds for making a distinction between the 
two forms, but I do not find there is a constant association of 
short rosulate leaves with short bracts and long leaves with 
large foliaceous bracts, or the opposite. - The former holds 
more frequently than does the latter. In the flowers I have 
discovered no constant difference between the two forms, al- 
though there are observable minor distinctions of size and relative 
length of parts. Upon the evidence before me I must unite all 
the plants under one specific name without making distinction 
of a variety, leaving the question of possible segregation for de- 
cision after more specimens have been obtained and the plants 
have been in cultivation. 
That the plant will be an acquisition to our gardens one feels 
assured from the specimens Forrest has sent home. One speci- 
men is remarkable because it shows what I have not seen in any 
other Candelabra Primula—a compound umbel of flowers ending 
the scape and umbels and compound umbels upon the lateral 
branches of the lower whorls of the inflorescence. A plant which 
‘sports ” in this fashion in nature, and with the bracteal varia- — 
tions I have described above, may be expected to show as great 
or greater variation under cultivation leading it along desired 
lines. It flowered in cultivation with Mr. Williams of Caerhays 
Castle in 1915, and also at Edinburgh. 
