312 TRANSACTIONS OF THE [Sess. lxxix. 



P. Listeri, King, in which the vars. rotundifolia, Franch., 

 and glabrescens, Franch., are absorbed. 



Pax x adopted the latter alternative and introduced P. 

 Listeri, King, definitely as a member of the Yunnan Flora, 

 keeping it distinct from P. obconica, Hance, and including 

 in it, but not as varieties, Franchet's var. rotundifolia and 

 glabrescens. Franchet's var. hispida he ignores. 



George Forrest 2 took a similar line when describing the 

 primulas collected by him in Yunnan during his first 

 exploration (1904-6) for Bees, Ltd., placing one set of 

 specimens in P. obconica, Hance, and another in P. Listeri, 

 King. 



French botanists after Franchet, who have dealt with 

 primulas — Petitmengin and Bonati — have rather worked 

 in the segregation mood of the first of the foregoing 

 alternatives, going even further than is expressed there by 

 raising one of Franchet's varieties to specific rank, and 

 naming as species others of the forms which Franchet said 

 " could be easily taken for distinct species." 



I imagine the real position is that, at the time of his 

 writing, Franchet, overwhelmed by the multitude of 

 novelties which the work of the French missionaries in the 

 virgin area of the rich West Chinese temperate flora poured 

 so bountifully upon him, could only skim the surface of 

 the inflow, and, having no time for critical microform 

 investigation, contented himself with pointing out some 

 broad features that appeared in this variable type. More- 

 over, I am disposed to suggest that possibly Franchet — as I 

 can assert of the later French botanists — had not adequate 

 material to enable him to form a true concept of P. Listeri, 

 King. In any case I cannot accept Franchet's statements 

 as intended to give critical diagnoses within this variable 

 series of forms, and an examination of some of the type- 

 specimens which he cites confirms me in what I say. In 

 subsequent pages I refer to this in some detail — and the 

 point is of importance, having in view confusions introduced 

 by the naming of one of his varieties as a species. Here I 

 may state in brief : — 



Under var. (a) hispida Franchet cites five types : — 



1 Pax in Engler's Pflanzenr. Primul. (1905), 24. 



2 Forrest m Notes Roy. Bot. Gard., Edin., iv (1908), 216. 



