$94 BALFOUR—NEW SPECIES OF PRIMULA. 
concavus subtus convexus basi expansus vix vaginans. Scapus 
brevis ad 1.5 cm. longus erubescens cum bracteis pedicellis- 
que dense glanduloso-puberulus umbellam 2-3-floram gerens ; 
bracteae circ. 8 mm. longae virides a basi circ. I mm. lata acumi- 
natae pedicellos aequantes vel eis breviores; pedicelli rubri 
1-2 cm. longi fragiles ; anthopodium obconoideum incrassatum 
magnum. Flores inter folia immersi. Calyx viridis circ. 8 mm. 
longus ultra medium fissus tubo poculiformi-tubuloso extus 
glanduloso viscido, lobis a basi lanceolatis acutis subpatentibus 
trinerviis integris. Corollae aureae fl. brevistyli tubus circ. I 
cm. longus cylindricus pallidior supra stamina ampliatus extus 
glanduloso-puberulus intus conspicue transverse rugosus ad os 
pentagonum subconstrictus annulatus annulo ro-lobato lobulis 
duobus strumosis ad basin cujusque petali oppositis, limbi discus 
circ. I mm. latus intensius coloratus glandulosus, lobi obcordati 
circ. I cm. longi 8 mm. lati apice emarginati recurvati subtus 
subalbidi ibique dense glanduloso-pubescentes. Stamina fila- 
mentis conspicuis crassiusculis antheras subaequantibus in flore 
brevistylo os tubi corollini versus antherarum apicibus vix ex- 
sertis inserta. Ovarium viride globosum; stylus brevis calyce 
multo brevior ; stigma capitatum. 
Species P. pseudobracteatae, Petitm. affinis sed efarinosa haud 
scabrida et ad haec umbellis 2-3-floris, calyce longiore ultra 
medium fisso, corolla calycem triente superante recedit. 
N.W. Yunnan. On precipices above a glacier two days west 
of Atuntsu. Alt. 11,000 ft. -F. Kingdon Ward. June 1913. 
Growing in big tufts. 
In 1913 Mr. Kingdon Ward sent to me a small dried specimen 
of this plant which in many ways agreed with the impression I 
had been able to form from Petitmengin’s description of the 
plant named P. pseudobracteata, Petitm., and I included it pro- 
visionally under this name. The living plants which we now 
have do not confirm my earlier impression. Petitmengin speaks 
of his species as having the leaves white mealy below when young 
atleast. There is no sign of this in Ward’s plant, and the glandu- 
lar hairs are not of the kind that produce meal. They have rela- 
tively long stalks, and the globose translucent head is coated with 
sticky glandular secretion. Petitmengin indeed says of P. 
pseudobracteata that it is covered with hairs which are in part 
glandular scabrid, but there is no suggestion of scabridity 
in Ward’s plant. Other characters of difference between P. 
pulvinata and P. pseudobracteata are its calyx, which is larger, 
more deeply cut, and with lobes entire showing no sign of incision, 
its corolla tube not so much longer than the calyx, and then in 
the corolla itself there is the distinctive creamy white and 
giandular under surface of the lobes so conspicuous a feature 
