1921-22.] BOTANICAL SOCIETY OF EDINBURGH 125 



above sea-level. The ranges are consequently separated 

 in some cases by a sub-arid tropical region, in others by 

 a moist valley with dense tropical vegetation. Such a 

 configuration induces isolated botanical regions with 

 resultant specific divergences in many genera. As Pro- 

 fessor Sir Isaac Bay ley Balfour pointed out to me, this 

 may not affect Lilium and allied bulbous genera to the 

 same extent as other genera which are more responsive 

 to edaphic conditions. 



The first group I propose to discuss is that which centres 

 round Lilium taliense, Franchet. 



Lilium taliense, Franchet, and its Allies. 



Among the Asiatic Martagons this species and its near 

 allies are well defined by the nectariferous furrow being 

 without papillae, and consequently smooth and naked. 

 As Franchet pointed out (Journ. de Bot., vi, p. 320, 

 1892), the Asiatic species having this peculiarity are L. 

 speciosum,, ochraceuni, taliense, and polyj^Jiyllum. L. 

 speciosum (one of the Archelirions) has no affinity with 

 taliense, and I need not discuss it here. L. jjolyphyllum 

 is apparently confined to the Western Himalaj^a, and in 

 its bulb and other characters is readily distinguished from 

 the Chinese members of this alliance. In China, therefore, 

 we are concerned in the first place with Franchet's two 

 species, L. ochraceum and L. taliense. But since Franchet's 

 time there have been additions. Leveille has described 

 five lilies of this series from Yunnan and Kweichow — 

 L. Bodinieri, L. Feddei, L. majoense, L. Pyi, and L. Tenii. 

 Of these the Edinburgh herbarium now possesses the 

 types with the exception of that of L. Pyi, which is 

 unfortunately absent from Le\eille's collection. Of it I 

 can therefore say nothing but that Leveille places it near 

 L. concolor of the Isolirions, while his description suggests 

 a Martagon. L. nepalense, D. Don, must also be considered 

 in this connection as it has been held, and with some 

 reason, that L. ochraceum is only a form of that Himalayan 

 plant. The admission of L. nepalense for discussion in the 

 series implies also an examination of L. primulinum, 



TRANS. BOT. SOC. EDIN. VOL. XXVIU. 11 



