Edith R. Saunders 227 



And further in regard to certain abnormal features : — 



11. That the two modifications of the corolla, peloria and heptandry, 

 both recessive to the normal, are inherited quite independently. 



12. That as a rare occurrence the margins of the sepals may be 

 thickened and bear structures having the appearance of rudimentary 

 ovules. 



The expenses incun'ed in the course of this investigation have been 

 defrayed in part by grants from the Royal Society and from the British 

 Association for the Advancement of Science. 



After the foregoing account had been written I received from Professor 

 Coutinho, to whom I here wish to tender my most cordial thanks, in 

 response to my enquiry, a specimen of D. miniana obtained from the same 

 region as Sampaio's original plant (Duriminia, Serro de Castro, Laboreiro, 

 near Alcebaga, about lat. 42° N.), and also further interesting information 

 regarding the occurrence of this form. Professor Coutinho writes that 

 D. iiiiniana is a rare plant, occurring within a very restricted area mixed 

 ivith purpurea ; that it was found again last year by one of his assistants 

 in the original locality, the specimen sent being one of those obtained 

 on this occasion. The appearance of the specimen, taken together with 

 Sampaio's account, and the further details supplied by Professor Coutinho, 

 make clear the close affinity of this form with purpurea. In fact whether 

 it is really distinct from that species seems to me open to question seeing 

 that the smooth stem emphasised as the distinctive character of miniana 

 is a constant character of purpurea nudicaulis. The dimensions given 

 for the stem (2 cm. in diameter at the base, 17 dcm. in height), the deeper 

 colour, smaller size (10—18 mm. long) and narrower form of the corolla 

 which is only slightly ventricose may possibly be fluctuating characters, 

 and in that case may be discovered associated with the pubescens habit, 

 although having hitherto escaped detection. The description of the 

 leaves of miniana would be quite applicable to a plant of purpurea. 

 Although the length of the pedicels in purpurea is generally not more 

 than equal to that of the bracts whereas in this particular specimen of 

 miniana they far exceed the bract-length, this appears to be a decidedly 

 variable character since in his description of miniana Sampaio states the 

 precise contrary, and in purpurea I have occasionally found a pedicel 

 exceed the bract by perhaps 2 mm. or more. Beyond the corolla characters 

 therefore there seems to be no very sharp distinction between purpurea 

 nudicaulis and miniana, and it remains to be seen whether the latter 



