320 Mr. Griffith on the Root-Parasites referred to Rliizantheae, 



might be, perhaps, supposed to be the structure of the Euphorbice, on which 

 they grow. 



Obs. II. — Of the few descriptions I have been able to consult, that of Mr. 

 Harvey, who has noticed the relations of the placentae with the lamellae of 

 the stigma, agrees best with the appearances presented by my specimens. 



M. Endlicher's account, which is much the same as that of Sprengel*, who 

 referred the genus to Monadelphia Triandria, and doubtfully to Cacti\ of 

 Jussieu, who took his from Linnseus's Supplement, adding one of his extraor- 

 dinary sagacious questions regarding its affinity to Cytinece, differs considerably. 

 He considers the male apparatus as consisting of three stamina with multi- 

 locular anthers, a structure, as it appears to me, quite at variance with the 

 appearances in my specimens. It is also, I think, contrary to analogy, no 

 instance being I believe known of a single anther consisting of an indefinite 

 number of regular cells, each of which has its own distinct superficies and 

 dehiscence. In all cases, not arising from the union of two or more stamina, 

 in which the number of loculi is increased, it arises, or appears to do so, from 

 the subdivision of the ordinary quadrilocellar theca ; and in all such the de- 

 hiscence appears to become more or less disturbed, in some being contracted 

 to a common point, as in Rajfflesia ; in others being partially extended, so 

 that each locellus has its proper dehiscence, as in Viscum ; in others being 

 generally extended, so as to cause the separation of the whole or greater part 

 of the superficies of the body of the anther, as in Rhizophora. The indefinite- 

 ness of the anthers is likewise, I think, to be inferred from the occasional 

 existence of what may be termed supplemental anthers, and also from the 

 structure of Cytinus. It must however be kept in mind, that Mr. Brown may 

 be supposed to have entertained a similar view from the nature of his remarks 

 on the affinity of ^/?%#eia with Curcubitacece. It appears to me, nevertheless, 

 that to constitute an affinity in structure between the anthers of Hydno?-a and 

 of some Cucurbitacece, a certain amount of continuity of some of the cells of 

 each lobe of the column of Hydnora would be requisitef. The structure of 



* Genera Plantarum, ii. 526. no. 2585. \V,i,Vi\^ 



t On such a point as this I would express myself with the greatest diffidence, being quite aware of 

 the immense knowledge required to be able to thoroughly understand the remarks of Mr. Robert 

 Brown. 



