A LIST OF BRITISH UOSES 15 



be diagnosed. The less biserrate forms are near B. insignis 

 Des^gl., the glabrous-styled ones approach R. leiostyla Eip., and 

 those with more globose fruit are near B. sphcBroidea Eip. V.-c. 3, 

 5, 14, 17, 22, 34, 43, 55, 58, 62, 78. 



E. RUBELLiFLORA Eip. ex Desegl. Ess. Monogr. p. 109. This 

 runs very near B. dumalis, of which it is regarded by some as a 

 variety differing only in its bright rose flowers, but it is also [s, 

 taller, stronger bush, with much more glaucous and more strongly 

 biserrate leaflets. As it grows on White Down in Surrey, it looks 

 a very well-marked species, but specimens so named from other 

 stations present less distinctive features. Dingier refers two of 

 my gatherings to B. dumalis, which in the absence of knowledge 

 of the colour of the flowers is quite possibly correct. V.-c. 3, 

 17, 58. 



E. LEIOSTYLA Eip. OX Crep. Bull. Soc. Bot. Belg. p. 238. Glabrous 

 styles, oval leaflets, not much narrowed at either end, nor, on the 

 other hand, broadly oval, and ovoid fruit, often with spreading or 

 ascending sepals, distinguish this species, but it merges insensibly 

 into forms with more subglobose fruit, narrower or broader 

 leaflets, slightly hairy styles, &c. V.-c. 3, 11 ?, 14, 17, 49 (or 52), 

 58, 62. 



[B. Chahoissm Gren. Fl. Jur. p. 241. This name is not used 

 by Sudre or Dingier, and on account of its indetiniteness I propose 

 to exclude it. The examples mentioned in E. p. 44 may be pro- 

 visionally placed to B. leiostyla Eip.] 



\_B. cladoleia Eip. ex Desegl. Cat. Eais. p. 163. A Leicester- 

 shire specimen, collected by Mr. C. P. Headly at Great Stretton 

 (see Watson Exch. Club Eept. 1906), was referred here by Sudre, 

 but its styles seem to be too hispid, and it will fall readily under 

 B. dumalis.} 



[B. canina var. dumalis f. Schlimperti Hofm. in Sitz. Abhandl. 

 Nat. Ges. Isis, p. 13. This is a name given by Dingier to six of 

 my gatherings, which have been referred to as many different 

 species by Sudre. The specimens undoubtedly vary very greatly, 

 and it is not easy to see what they have in common, nor does the 

 description, though a long one, throw much light on the peculi- 

 arities of the variety, unless they are its spreading sepals and 

 glabrous styles. It is best excluded until it is better understood.] 



E. SPH^ROIDEA Eip. ex Desegl. Cat. Eais. p. 169. This corre- 

 sponds in this subsection to B. sphcerica and B. glohularis in the 

 two last. V.-c. 3, 5, 17, 58. 



E. ERiosTYLA Eip. & D6segl. in Cat. Eais. p. 165. The only 

 example I have seen like this species, besides those referred to in 

 E. p. 42, is one from Hunts {Ley). Sudre confirms the name, but 

 Dingier thought it rather B. canina var. grcgaria H. Braun, 

 tliough it does not agree with other specimens so named. It is 

 best under Sudre's name, which is already in our list, though not 

 very well confirmed. Ley's specimen is very near B. spharoidea 

 Eip., though with more woolly styles. V.-c. 3 ?, 31, 49 or 52. 



