311 ^lut to hpve pccer>ted them fs uninpepchable , so thet his determina- 

 tions are often very imtrustY/orthy. Another cpuse of the diffi- 

 culty in correctly defining the species is th^t in this country 

 we hpve no means of* knoxving to v^h&t extent the same species rea- 

 lly vrries under naturrl conditions in -South Af ricr , nor have we 

 any knov;ledp;e of the localities in '• hich the species grov;. 



Thpt there fire several distinct species among them I feel 

 sure, but they ^re not the si^ecies that ere usually'" found in grr- 

 dens, v'hich are mostly forms or hybrids of one species (G. longum, 

 I think) grovm under different specific names, -^t the same time 1 

 be! ie""'e thf't some of the slants thpt have been described as distin- 

 ct sriecies pr'^' merely forms of other species, for there can be no 

 auestion th? t som.e of these r>lants very so greatly that it is very 

 di-^ficnlt to find absolute characters for distinguishing them. 

 As en illustration of this, I may m.ention that one of the most 

 prominent distinctive characters is vliether the leaves are all 

 arranged in tvo ranks spreading right and left, or v/hether the suc- 

 cessive neirs cross one another. This is a v;ell marked and e? sily 

 recoptnised distinction. ^'^, S, Bpown 



(To be continued.) 



1-esembrvanthemium and some nev/ genera senarated from it. 

 Oerd. Chron. HI. 70*. 327. 1921.' 

 (Continued from page 311.) 



'^27 "^ew T'-oulf^ hesitate to saj'' that the ti-ro plents illustrated in 

 ^iff. 146 v^ere two perfectly distinct species, the only thing they 



?ig. 146 — ^lottiphyllum pus tula turn "^ 

 1V«ro rl^nts raised from seeds taken from the same seed pod and grov'n 

 in the same 6-inch pot under similar conditions, -t-bout one-fourth 

 natural size. 



have in common being the prllid sv.'elling (or so-called pustule) 

 at the base of each leaf on the urper side, which is scarcely evi- 

 dent in the much reduced -photograph. Yet these tv/o plents v;ere 

 ■pc^c^ofi >)y myself from seeds taken by my ovm hand out of one seed- 

 "^o*: "^.is seed-pod "--as upon a living plant^ collected at Alice, in 

 Sou't'h Africa, and sent to me last ^'•ear by ^'^r. T. N. Leslie. 'Ih.e 

 tvo ^eedslings figured are of the same age and have been growing 

 in the sam^- TOt, out of doors, and exposed to the full sunlight 

 of this summer from Lfey to October 24th. --ost of the seedlings 

 •^re like the small T^l?nt, ?nd three like the long-leaved one. 

 '^eir distinct aT^pearance is one o-" the greatest surprises I have 

 hr ^ , ^n" clearly demonstrates that the proper defining of the 

 srecies belon^in?- to this genus cannot be done in this country. 

 In co'^nection with the two-renked arrangement of the ler^ves com- 

 mon to many of the species of this genus, it am.y be o'' interest to 

 sa-^ th^t in the seedling stage the first few pairs of leaves that 

 ore form.ed are always developed in a crucirte manner, the alter- 

 n^^tinp- pairs being placed at right angles to each other; the change 

 to p. two-ranked arrangement takes place when the seedlings are a 

 ■•"ew months old. This would seem to indicate that this genus has 



