Ae 
BALFOUR—-NEW SPECIES OF PRIMULA. 155 
not that species and which are perhaps not P. cardiophylla, 
Balf. fil. et W. W. Sm. I cite here :— 
Sikkim. 11,000 ft. Tanner. In Herb. Calc. 
Sandakphu. 12,000 ft. Clarke. No. 13,707 C. (in Herb. 
Calc.) ; 13,707 J. (in Herb. Kew). 5th June 1884. 
Sandakphu. King’s Collector. June 1887. 12,000 ft. 
Calc. 
Sandakphu. 11,800ft. Gammie. No.14. roth June 18or. 
In Herb. Calc. 
Sandakphu. Rhomoo. No. 285. 30th May 1908. In Herb. 
di 
Edin. 
Sandakphu. Rhomoo. No. 782. 21st July 1910. In Herb. 
Edin. 
I may point out that the increase in size of leaf by the time 
fruit is mature is remarkable, the petiole and lamina sharing 
equally in the increase. And this leads me to speak of a specimen 
in the Kew Herbarium which bears the label “ Primula rotundt- 
folia, Wall. var.—Laka, 11,000 ft. Dhurmsala, Clarke, No. 24,559. 
17th Oct. 1874.” In pencil on the sheet is written “ P. tricostata, 
Watt.” There are specimens of two distinct species on the 
sheet. One specimen is a scrap of a fruiting scape, and 
it might belong to P. cardiophylla. The rest of the specimens, 
also fragmentary—four leaves, a rhizome, and portions of 
two scapes bearing fruit—supposing them to be of one species 
—belong to a plant of quite a different Section. The calyx 
is that of one of the Geranioides, and its ribbing has 
given origin apparently to the MS. name. The leaves and the 
rhizome bud would suit Geranioides, but there is no Indian species 
of the Section with similar leaves and fruit. The leaves recall 
those of P. mollis, Hook., but the scapes do not. It isa plant to 
be looked for. Why it calls for mention here is that in “ Flora 
of British India”’ the area of distribution of P. rotundifolia, 
Wall. is given as “‘ Temperate Himalaya; from Kashmir, alt. 
11,000 ft. to Sikkim, alt. 12,000-13,000 ft.’”’ The statement 
is, I suspect, based upon this Kew sheet. But neither P. rotundi- 
folia, Wall., nor any one of its immediate allies, is found in the 
West Himalaya. P. rotundifolia, Wall. is to our present know- 
ledge a plant of Nepal. VP. cardiophylla, here segregated, is 
a plant of S.W. Sikkim, and this is the species to which most of 
the P. votundifolva of living collections belongs. 
Primula chrysochlora, Balf. fil. et Ward. (Candelabra.) 
Glabra efarinosa. Folia membranacea rosulata tenuia vir- 
idia ad 8 cm. longa ad 3 cm. lata oblonga vel oblongo-obovata 
obtusa margine irregulariter denticulata basi in petiolum 
brevissimum vaginantem attenuata subtus pallidiora glanduloso- 
