166 



oue at least of the specimens collected by Soulie should belong to 



that species. 



from 



thej- ^ere collected, that all three are 31. Prattii, which species 



WP L'TlriTi- ■n-\r\Tar\-rT/^t% +« T. ^ -rr^ 1 11 J^ _ _1 1. O !•' m 



near 



chien-lu (SouUS n._635). .¥. racemosa, Fedde, is tL 



composite ^species including M. horridula, var. racemosa:^ in 



Prattii. It would follow, however, tliat the speci- 



M 



pedicels. 



M 



impos 



stated to have had purple flowers with white stamens, was figured 

 as M. racemosa in 1909.* The plant in question could hardly 

 have been mistaken for M. racemosa had its flowers been sub- 

 tended by bracts; the fact that it is said to have had white 

 stamens— no doubt " white anthers " was meant— shows, how- 

 ever, that it was M. Prattii, described above, not the plant with 

 yellow anthers and dark blue petals to which the name M. 

 racemosa is_ usually applied in our gardens. 



This particular Gardeners' Chronicle reference renders it neces- 

 sary to call attention to a point which should not be overlooked. 

 It IS one which, for the moment, is not very important, but which 

 may prove to be of consequence at some future date. We know 

 that the plant figured by Maximowicz as M. racemosa in 1889 is 

 only aform of M. horridula, described by Hooker and Thomson 

 in mob. T\'e know that the plant collected in Eastern Tibet bv 

 rrzewaiski, and the plant collected by Potanin in Northern 

 bzechuan, both of which were distributed by Maximowicz as M. 

 racemosa, also belong to M. horridula, Hook. f. & Thorns. 

 We know, besides, that M. horridula is essentially a Tibetan 

 species, and that its occurrence on the high alps of the Eastern 



from Tibet. This bein? the case, 



^ is remarkable that the locality in which the specimens upon 

 which J/, racemosa was based should be situated in the Province 

 of Kansu, well to the east of Kuku-nor. Until the contrary has 

 been proved we must accept the figure of M. racemosa, given in 

 wJ "■? ^"^^f ^^'^ ^J «ne of the most careful students of 

 tlT\'' ^ ' ' ''''''' moreover, the author of the species con- 

 pX.t /? conveying an accurate conception of the Mcconopsis 

 collected by Przewalski at Chobsen and described as M. racemosa 

 in ioio. 



^■,o^w!n^ nl'^f ^^^^^^* «een a specimen of the original M, race- 

 IhCllT ^^P^^''''' m Kansu, in anj of the herbarm he has been 

 this llahT'''\''''- '\^^''^ ^^^ indication that duplicates of 

 It is noflfl^'f ^'''°A\^''^ "^^^ ^^'^ issued to other^ herbaria. 

 ^Zor/rL 'f/?^ ^ '*'*'^ *H^ *^^ %^^^ ^^ict appears in the 

 Chobsen 3hr- ^^' Vr^V^'^ii from a specimen of the original 

 cXcTm fnvwh^^^^^^^^ Moreover, no one^ince Przewalski ha s 

 wi h da k hi^- fl '"^ '^''' ?r'^°^'^^ °^ *^« Tibetan Meconorsi^ 



On_the^«ier^^^ tnown that a Mcconopsis with grey 



« 



Gardeners' Chronicle. 1909, vol. xlvi, p. 91, fig. 40. 



