PRIMULA CONFERENCE. 



of Primula at the writer's disposal as well as by his advice on various 

 critical points has greatly assisted the writer in his present task. 

 P. Drummondiana is easily distinguished from P. Winteri by the 

 smaller flowers and by the corolla lobes being bilobed and not at all 

 fringed. The remaining varieties — namely, nana and scapigera — the 

 writer is at present inclined to retain, although it must be pointed out 

 that in the former variety some rather distinct plants are included. The 

 writer is in full accord with Sir George Watt in his remark that 

 P. petiolaris and its varieties are well worthy the attention of horti- 

 culturists ; and now that P. Winteri is so much appreciated he hopes 

 to see more of the allied species imported soon, for it appears quite 

 evident that from living material only can P. petiolaris and its varieties, 

 nana and scapigera, be satisfactorily accounted for. 



With regard to P. Tanneri , Pax and Knuth do not indicate any 

 specimens as having been seen by them, and so have, according to the 

 writer's views, separated it widely from its ally P. Griffithii. 



The section Soldanellgtdeae, Pax, corresponds with the section 

 Soldanelloides, Watt, with the addition of P. pusilla and P. 

 sapphirina, which Watt placed in his section Denticulata. In 

 addition to the two mentioned it includes P. Wattii (fig. 94), P. 

 Reidii (fig. 90), P. uniflora (fig. 91), and P. soldanelloides. 



So far as the Indian species are concerned Pax's section Farinosae 

 corresponds to the section Rosea, Watt, except that Pax refers P. 

 hazarica to the Calliantheae. To this section also belongs a new species 

 collected during the Tibet Frontier Commission expedition which the 

 writer would name P. Walshii in compliment to the collector. In habit 

 the plant is very similar to P. pumilio, Maxim., from Kansu, but it is 

 readily distinguished by the abundance of glandular hairs. 



The section Capitatae, Pax, which includes several well-marked 

 types, contains five species which Watt referred to his section Denticu- 

 namely, P. erosa (fig. 92), P. denticulata (fig. 59), P. glabra (fig. 93), P. 

 bellidifolia, and P. capitala, and two new species described by Pax : P. 

 Schlagintweitiana and P. glomerata. These two species the writer has not 

 seen, but from description plants collected by Duthie in Kumaon would 

 appear to belong.to the former, and they agree fairly well with the figure 

 given except that the pedicels are much shorter ; but, whereas from the 

 figure the pedicels would appear to be up to 6 mm. long, Pax in his 

 description says " pedicelli fere nulli." The plant is closely allied to 

 P. denticulata, from which it may best be distinguished by the absence 

 of basal scales. From description P. glomerata would appear to be well 

 represented in Kew herbarium. A close ally of P. glomerata — or it may 

 be a hybrid — is the plant figured under the erroneous name of P. erosa 

 in the " Botanical Magazine." In 1911 another species belonging to this 

 group was published by W. W. Smith under the name P. atrodentata. 

 Founded on specimens collected in Zemu and Llonakh, Sikkim, it may 

 be regarded as the Eastern representative of P. Schlagintweitiana. The 

 plant was previously collected at Lachen by Sir J. D. Hooker, but the 

 material of that collection was too poor for description. Under the 



