3'22 Mr. Griffith on the Root-Parasites referred to Rliizantheie, 



ternary grouping of the plates of the stigma, the evident, thongh in a much 

 smaller degree, ternary division of the inner surface of the ovarium, and the 

 general structure of the flower and its affinities, all militate in a greater or less 

 degree against this supposition ; and if Mr. Harvey's and my own observations 

 be found to be correct, we are compelled, I think, to admit that the composi- 

 tion of the pistillura is definite. In this case the hypothetical explanation 

 becomes exceedingly complex, and perhaps paradoxical, when I consider the 

 simple state in which the vegetable leaf is generally presented to us in the 

 pistillum. 



Hydnora appears to me much more perfect (complex) in its organization 

 than Cytinus. It is also remarkable for being hermaphrodite, and for pre- 

 senting, excepting those parts of the anthers on the outer face of the columna 

 stam'mea, perhaps the greatest known facilities for impregnation*. 



Cytinus, Juss'ieu, Gen. Plant, p. 73. Endl. Gen. PI. p. 75, 723, 



Uypolepis, Spr. Gen. PI. 11. no. (38.) (char, pessimo). Harv. Gen. S. African 

 Plants, 300. Brongniart, Ann. Sc. Nat. i. p. 40. t. 4. 



Char. Gen. Flores monoici vel dioici. Mas. Perianthium duplici serie 4-6-partitum, aesti- 

 vatione imbricatum. Stamina monadelpha 7-8 (vel 14-16); antherae lineares, rectae, 

 adnatae, coronula lobata dentata terminatae. Rudimentum Pistilli nullum. Fcem. 

 Perianthium maris. Ovarium inferum, 1-loculare; placentae plures parietales; ovula 



* On my arrival at the Botanic Gardens several months after the above, with the exception of one or 

 two of the notes, was written, 1 had an opportunity of consulting M. Meyer's description of Hydnora 

 africana and H. triceps in the ' Nova Acta Physico-Medica,' &c., vol. xvi. p. 773. M. Meyer describes 

 the stamina as indefinite, the anthers as dithecous, the stigma as trilobed, as having a striate appear- 

 ance, and as opposite to the lobes of the columna staminea and to the segments of the perianth, and the 

 placentae as being pendulous from the stigma-bearing disc. He also seems to be of opinion that three 

 ovaria enter into the composition of its pistillum. He alludes to Mr. Brown's remarks on the affinity 

 in the structure of the anthers of Hydnora and those of Cucurhitacece, and appears to think that Mr. 

 Brown may have been misled by Thunberg's description, since he finds the structure of the anthers 

 of Hydnora extremely different from the flexuose ones of CucurbitacecB. He suggests the probability 

 of the processes on the inner faces of the lacinise of the perianth (pulvini), the petals of Thunberg, 

 being the rudiments of an inner series : this, it appears to me, would either increase the degree of 

 opposition, or require a paradoxical hypothesis regarding the composition of the outer lacinise. Lastly, 

 M. Meyer denies that it has any affinity with Fungi, and places it among AsarinecB, believing it to have 

 albuminous, embryonate seeds. 



