117 



the length of the style-arms. He draw attention to the existence 

 of gyno-dioecism In several species (virescens^ giganteum sensu 

 lato, hterrense and Decaisnei), and described four new ones 

 Webhii, e^asperatum, Bdtirgaenum and gentianoides), three of 

 which had heen proposed hj Webb in manuscript. Bolle* had 

 previously described two of Webb's manuscript species iliierrense 

 and callithyrsum) , and E, leucophaeum, Webb, mss., is recog- 

 nised as distinct in the present paper. 



Webb seems to have projected a revision of the Canarian 

 Echiums^ and it is to be regretted that this never appeared, as 

 his knowledge of the flora of the Canaries was unrivalled. It is 

 evident, from the names which he left in manuscript, that he 

 would have adopted in certain cases smaller specific units than 

 those formerly recognised by him in the Phytographia Canarien- 

 sis, and that in no case would he have united species so diverse 

 in habit and with such well-markedi technical characters as 



giganteum and aculeatum. Christt regarded these two as dis- 

 tinct species, and described leucophaeum as a variety of 



aculeatum^ whereas Bornmuller+ treated leucophaeum and 

 aculeatum as varieties of giganteuTn^ and de Coincy 

 who knew them only in a dried state, was unable to 

 distinguish leucophaeum from giganteum and regarded 

 aculeatum as a mere form of the latter. BornmuUer considered 

 that the differences in breadth of leaf, spininess, and indumentum 

 were dependent on local conditions, aculeatum occurring, accord- 

 ing to him, in the hottest and driest places, leucophaeum in 

 somewhat moister localities, and giganteum usually in shady 

 bushy places. Neither giganteum nor aculeatum, however, alters 

 appreciably under cultivation, so that the characters mentioned 

 are relatively fixed; and as they are associated with others 

 derived from habit, the shape of the calyx-segments, the degree 

 of zygomorphy of the corolla, the height of insertion of the 

 stamens, the length of the style-arms, and the nature of the 

 nutlets, the three species can hardly be regarded as otherwise 

 than distinct. 



The Canarian species of Echium and certain other genera have 

 an extremely restricted distribution, and some which look very 

 much alike m a dried state may be distinguished at a glance in 

 .the field, so that they require for their elucidation a knowledge 

 of local topography and observation in the field, in addition to 

 critical work in the herbarium. 



During the course of an expedition to the Canaries in the 

 summer of 1913 (see K. B., 1913, pp. 287-299), the writers were 

 fortunate in seeing both E. giganteum and E. aculeatum grow- 

 ing near Puerto Orot^ava, Tenerife, the former in a wild state, 

 the latter in the garden of the Grand Hotel Taoro. On visitin 

 the island of La Palma, two other Echiums were seen and col- 

 lected which obviously represented distinct species. These are 



* Ind. Sem, Rort. BeroL 1867, app. 1, pp. b'-7. 

 t Engl. Jaltrb. vol. ix. p. 128. 

 X Bne:l. Jahrb. vol. xxsiii. p- 466. 

 Bull. Herb. Boiss, ser. 2, vol. iii. p. 27^. 



