170 THE JOURNAL OF BOTANY 



This is not the only difficulty connected with Sonder's treat- 

 ment of Ctenomcria. In 1843 E. Meyer (Drege Zwei Pfl. Docu- 

 mente 226) had issued as Tragia capensis, or as Tragia capensis ft, 

 certain plants collected by Drege under the belief that they are, 

 or belong to a variety of, T. capensis Thunb. This is true of a 

 specimen collected by Drege at Galgebosch, in Uitenhage, which 

 is identical with Drege 8239, but it is not true of any of the other 

 specimens collected by Drege, and issued as T. capensis by E. 

 Meyer." 



Baillon's treatment of the specimens known to him marks a 

 distinct advance on that of Hochstetter and Sonder. The two 

 plants, Drege 8239 and Krauss [186] , which Hochstetter had 

 treated as distinct species, Baillon treated as one species, under 

 the name C. cordata Harv. But while taking this step Baillon 

 still maintained C. Kraussiana Sond. vix Hochst. ; not having 

 seen either of the specimens cited by Sonder, and apparently 

 confused by the double citation of Tragia capensis, Baillon was in 

 doubt as to whether Sonder's plant could be a Ctenomeria at all. 

 This doubt was not justified ; G. Kraussiana, as described by 

 Sonder, is not exactly C. Kraussiana Hochst. ; it is, however, 

 exactly C. cordata Hochst. non Harv., and is, moreover, exactly 

 Tragia capensis Thunb. ; what Baillon, had he been aware of 

 these facts, must have done was not to suggest the exclusion 

 of C. Kraussiana from the genus, but to include it in his widened 

 C. cordata. The doubt expressed by Baillon in 1862 explains the 

 suggestion of Hooker in 1868, that in the genus Ctenomeria there 

 are 1-2 species. 



It is interesting to note that, while Baillon was certainly right 

 in treating the pubescent Ctenomeria and the nearly glabrous 

 one as conspecific, there would appear, after all, to be another 

 recognisable South African form. This is the species described 



* These specimens, from Pondoland and Natal, named T. capensis or 

 T. capensis /3 by Meyer, represent two states, a more pubescent and a nearly 

 glabrous, of a climbing Tragia. Taken conjointly these two conditions consti- 

 tute T. capensis E. Mey. ex Sond. (Linnaea xxiii. 110) and ex Baill. (Adanso- 

 nia iii. 162) ; Sonder and Baillon were right in considering that T. capensis /J 

 does not, as E. Meyer supposed, differ varietally from the more pubescent 

 form, mistaken by E. Meyer for the true 2'. capensis Thunb., which is really a 

 Ctenomeria. Muller in 1886 (DC. Prodr. xv. 2, 938) after eliminating the true 

 T. capensis as Leptorhachis capensis, in intention followed E. Meyer, and 

 altered the name T. capensis E. Mey. ex Sond. & Baill. non Thunb. to 

 T. Meyeriana, a hirsute, and /3 glabrata. But while T. Meyeriana, /3 glabrata 

 Mull.-Arg. is exactly T. capensis p E. Mey., T. Meyeriana a hirsute Mull.-Arg. 

 is a mixture of a pubescent form of T. Meyeriana /3 glabrata, and of a very 

 distinct erect and never twining species which Drege did not collect and Meyer 

 never saw. Of these two strikingly distinct species, one erect, the other twining, 

 the one which has been described by Muller is the erect species ; the name 

 T. Meyeriana, therefore, must be retained for the species with which Meyer had 

 nothing to do. For the intelligent segregation of these two species, as 

 T. Bolusii and T. durbanensis respectively, we are indebted to Dr. 0. Kuntze 

 (Rev. Gen. PI. iii. 2, 293). His name T. Bolusii for the erect plant must 

 unfortunately be sunk in T. Meyeriana Mull.-Arg., but his name for T. dur- 

 banensis stands because T. capensis E. Mey. non Thunb., which is its precise 

 homonym, though accepted by Sonder and by Baillon, was not accompanied by 

 a description. 



