Craib — Primulas of Petiolaris-Sonchifolia Section. 259 



with the petiolaris type, but certainly no arguments would con- 

 vince me that Pax was justified in placing P. sonchifolia in the 

 same section as P. japonica, P. Cockburniana, P. Poissomi, etc. 

 In favour of the petiolaris-sonchifolia link now adopted, one has 

 the fact that the development form of P. sonchifolia as also the 

 fruit is that of P. Hookeri and P. vernicosa, and to a slightly less 

 degree of P. Winteri and P. Edgeworthii. 



Reference to fruit character in this section would be incom- 

 plete without an admission that of many, or rather of the 

 majority, the mature fruit is unknown. In P. vernicosa the 

 capsule cracks irregularly round the top and crumbles away. 

 This would appear to be the method of dehiscence also of P. 

 sonchifolia, P. scapigera, P. Whitei, and probably of P. bracteosa. 

 From the immature material of several of the other species, 

 however, the writer concludes that there is evidence, but not 

 wholly convincing evidence, of the presence of longitudinally 

 dehiscing capsules. 



As a preliminary to the examination of Pax's treatment of the 

 individual species it should be noted that of the seven species 

 enumerated, Pax indicates that he saw specimens of only three, 

 and that of the six varieties of P. petiolaris he saw specimens of 

 only three. In his treatment of P. petiolaris Pax follows the 

 Flora of British India, with the single exception that he rightly 

 raises the variety Edgeworthii to specific rank. The fact that so 

 feu- specimens were seen by Pax may account for such unequal 

 treatment of two closely allied plants, viz. var. pulverulent a. 

 Hook, f . and var. Edgeworthii, Hook. f. The former Pax retains 

 as a variety of P. petiolaris and the latter he raises to a species. 



Pax also raises to specific rank P. petiolaris var. odontocalyx, 

 Franchet, but again evidently without having seen any authentic 

 specimens. . 



Three years later G. Forrest added another species from his 

 Yunnan collections, viz. P. taliensis. In the same paper Forrest 

 describes and illustrates another new species-P. gratissnna- 

 which was however recognised later to be Franchet's P. sonchi- 

 folia \nvone working from Pax's monograph would be justified 

 in redesenbing Franchet's species when one considers with what 

 species Pax placed the plant. 



The introduction of P. petiolaris var. pulverulenta, Hook. f. 

 to cultivation led W. Watson to recognise the plant as a distinct 

 species under the name P. Winteri. Whether this plant is really 

 distinct from the true P. nana of Nepaul or not will only be 

 solved by the receipt of additional material from Nepaul. 



In P. Whitei W. W. Smith described the Bhutan represen- 

 tative of the Yunnan and Upper Burma P. sonchifolia. 



At the Primula conference it was deemed prudent to raise to 



