290 althou.'^h it h^s lon^' been considered to Toe so in the case of the 

 yellow-flowered I!, edule end the r^urple-flov-erecl i-I. aeaullrterrle, 

 which -re other"dse identic?!, "but from a cereful examinetion of 

 the tv'o nlants frrov.'inp side by side in '^ormvpn, where they have 

 becone naturalised, cnS. from infornation received from ^outh 

 Afric? , I am convinced that these tv^-o supposed species are but 

 colour variations of one species. A proof of this colour variation 

 is afforded in the case of ''■■"-, testiculere. — fev/ yecrs ago I sent 

 to Dr. RodieJ? Heath a single seed-pod of ^-. testiculsre and the 

 plants he raised from it are nov: coning to flower, and he writes 

 to inform me that some ere of a very pele v/hitish-yellow, others 

 o-f ? much darker yellov;, and others purr^le. -^11 these varieties 

 bein-cr obtained from the same seed-pod demonstrates clearly that 

 colour alone is not a specific character ^r. ^^^odier J^-^eath has also 

 raised seedlings of -'. simulans, some with white, others with 

 oranre or yellow flovrers. The most curious case of colour variation 

 tliet has yet come under my notice Is thpt of the plant I described 

 as --. fraternum some years ago. A single riant of it, of which I 

 had a portion, was sent to Kew. The first year after it ves intro- 

 duced it flo^'-ered and was figured; the flovrers v;ere of a rosy 

 pink. The same plant has flov/ered several times since, and each 

 time the flowers hax'^e been white; it hps not again produced a pink 

 flower. At the same time was also introduced e sinrle plant from 

 another locality of whrt I consider to be a white-flov/ered variety 

 o-^ ^'. fraternum which has rempined white and shown no tendency to 

 varv, anc! is al"'ays of e. slightly larger size end less glaucous 

 than the tyre. 



The only other peculiprity of which I wish to direct the att- 

 ention of the rlpnt-studen and horticulturist elike, since it is 

 closely connected v.dth the cultivation of the Sphaeroid group, 

 is thrt these curious rlpnts hrve ver^/ sh'ort roots, rarely more 

 than 1 to 1^ inch in length. Yet these plants thrive best in 

 the driest pnd hottest parts of South Africa, some of them in 

 nlaces where the rainfall is only from three to six inches per 

 annum. J'or months eta time the soil is thoroughly dry end parch- 

 ed, Hov do they exist under these conditions'? '^hej certainly do 

 not thri-"e in this country if kept dr?/ too long. This, however, 

 may be due to the feet th^t they want to grow and flower during 

 our autumn, i-hich corresponds to spring in the jiarts of ■^outh 

 Africa —here these Sph^eroids f^row. I'^. 3, Brown 



(To be continued.) 



I--esembrvanthemum end some nev- cenere ser^arated from it. 

 Card. Ghron. HI. 70: 303. 1921. 

 (Continued from page 290.) 



303 1 had one species sent to ne in lvl9, v/hich began to grow 

 '-e'^.l enough, but a'^ter a short time decaj'^ed. This spring the 

 same plant was sent to me v/ith the information that ?.t grew upon 

 f 1p t rocks in about half an inch of soil ^-^ow does it exist under 

 that condition*? A plant grorring in an inch of soil upon one of 

 our rtpvements this year vrauld have had a poor chence of surviving, 

 I think, yet this plant thrives in a similer position in the drier 

 end hotter climate of South Africa. But I found that this species 

 would not submit to be dried un during this sujnmer as its allies 

 did, -Pot again the plant be^gan to die, and I have had considerable 

 trouble to save it. Its beheviour probably indicrtes thet it does 

 not like th^: reversal of the seasons, and thpt, having passed safe- 



