WHAT IS ASTRAGALUS HYPOGLOTTIS ? 225 



But if this be indeed the type of A. hypoglottis, why has 

 Linnaeus marked it '' ejnglottis"? The reply is obvious if, as 

 seems reasonably probable, this plant was in his herbarium before 

 he wrote the Mantissa altera. Until the pubhcation of A. hypo- 

 glottis in that work, and its clear separation from A. epiglottis, 

 this specimen of A. purpureus was A. epiglottis pro parte, and he 

 could not have assigned it to any other species that he had up 

 till then recognized. In fact, in Mant. p. 274, he says quite 

 clearly that A. hypoglottis is the A. epiglottis of Syst. Nat. (exclusis 

 synonymis) ; that is to say, he there declares it to be this very 

 plant. That he did not, after publishing A. hypoglottis, alter the 

 name of this specimen from epiglottis to hypoglottis does not 

 throw doubt on its being really hypoglottis ; because in exactly the 

 same way he has not altered the name of specimen no. 3 from 

 epiglottis to pentaglottis, though it is indisputably the latter 

 species. It must, however, be admitted that Smith's opinion that 

 this specimen represents A. hypoglottis gives us no assistance, 

 because, as already explained, he was blind to the difference 

 between A. danicus and A. purpureus. 



So far, therefore, A. purpureus holds the field, and A. danicus 

 has no claim to represent A. hypoglottis. But the later views of 

 Lange in the Bull. Ac. Roy. Dan. for 1873 [loc. cit.) demand con- 

 sideration. In his paper in that publication, '■' the proposition 

 that hypoglottis = danicus is again refuted — for the third time by 

 Lange — at great length. His objections fall under four heads : — 



(1) Linnceus's diagnosis does not correspond icith danicus. 

 Indeed, it does not altogether fit, but the objections on this 

 ground are not quite so strong as Lange seems to think. As they 

 apply equally to yurpureus, I reserve discussion of them till we 

 come to that species. 



(2) Linnceus describes hypoglottis as annual, whereas danicus 

 is very obviously perennial, and " unless there are peculiar cir- 

 cumstances one cannot suppose that Linnaeus could have deceived 

 himself as to this character." There are peculiar circumstances, 

 as will be pointed out under A. purpureus. 



(3) The reference by Linnceus of Bay's synonym\ and that of 

 Plukenet for the British A. danicus to arenarius and not to hypo- 

 glottis. This is a very strong argument, and is really conclusive 

 by itself against the hypoglottis = danicus theory. It is reinforced 

 by the fact that the synonyms quoted for hypoglottis, i.e. A. epi- 

 glottis Syst. Nat. and A. villosus pirocumbens, &c., of Ray, Appx. 

 p. 455, give no support to that theory. It is strange that Lange 

 was not able to find these last words of Ray in that author's 

 writings ; so he says on p. 49, but the words are there. 



* References are to the French version, but the plates only accompany the 

 Danish original. 



t The phrase " Glaux montana purpurea nostras" is due to Ray, Hist, 

 p. 939 (1686), but its synonym, "Astragalus incanus parvus purpureus nostras," 

 is Plukenet's, in Aim. p. 59 (1696). Ray, however, in the third edition of 

 Syn. Stirp. Brit. p. 376 (1702), adopts Plukenet's name, adding his own as a 

 synonym. 



Journal of Botany. — Vol. 50. [July, 1912.] r 



