414 ON THE KEMALK rEOWKR AND FUMT OF 



233] form merely different orders of the same natural class, in 

 giving an account of a new and remarkable genus of the 

 latter family.^ 



At present I shall only remark, that the sole remaining 

 character employed to unite these two families and supposed 

 to distinguish them from all others, namely, the simple or 



stages is entirely cellular, aud that this contimics to be the case not only until 

 that mutual adaptation of parasite and stock wliich enables the former to 

 complete its development has taken place, but until the first indications of its 

 future structure have become perceptible. It may also be remarked, that even 

 after the formation of vessels in the parasite is obvious, the direct union be- 

 tween Edfflesia and the Vifis continues to be chiefly if not entirely cellular, the 

 connection consisting in a slight mutual penetration or indentation of the two 

 substances, whose cells are easily distinguishable. 



I may here advert to one of the most difficult points in the economy of 

 Ra^esiacecCy namely, by what means their minute embryos, which are at the 

 same time of an extremely loose texture, are enabled to penetrate through the 

 bark of the plants on wdiich they vegetate, so as to account for such appear- 

 ances as tiiose exhibited in the nascent RaffJesia Arnoldi represented in Tab. 

 25 (XXYI), A, in which I have been unable to trace any perceptible commu- 

 nication with the surface, and where the parasite seems rather to grow out of 

 than into the stock. 



Connected with this point a question may also arise, whether the earliest 

 effort of the seed after its deposition in the proper nidus, by whatever means 

 tiiis is effected, may not consist in the formation of a cellular tissue extending 

 laterally under the bark of the stock and capable of producing the fully 

 developed parasite. 



This question might not occur in regard to Raffleda and Bnigmansia, in both 

 of which the individual plants are in general sufficiently distant on the root of 

 the Vitis to make it probable that each developed parasite is produced from 

 a distinct seed. But in Rilostyles^ and even Cyiinus^ where they are closely 

 approximated, their possible origin from one common basis or thallus is more 

 readily suggested, especially on considering that in the former genus, which is 

 dioecious, each group of parasites is generally, perhaps always, exclusively of 

 one sex; and that these groups, often of great density, not unfrequently 

 surround completely the branch of the stock. But although this view did 

 occur to n.e as not very improbable, and as tending to remove some of the 

 apparent difficulties, I have never been able to trace any substance decidedly 

 distinct from the proper tissue of the stock ; there are, however, some a))pear- 

 ances favouring the hypothesis in both genera, especially in Pilodyh's, but 

 which require careful examination in the living plants. 



^ This geims, which was first found by Francis Masson, is the MydropetaJon 

 of Mr. Harvey (in South Afr. Gen. p. 41$), who has described two species, 

 from both of which Masson's plant is perhaps distinct. 



I may here advert to a note at p. 225 of my former memoir (in Linn. Soc. 

 Trans, vol. xiii), {Ante, p. 390] in which 1 thought it not improbable that a 

 parasite briefly noticed by Isert (in Keise nach Guinea, p. 288) might be re- 

 lated to Rafflesia. I have now, however, reason to believe that Isert's plant 

 is the Thonniiujia sanguinea of Vahl (in Act. Soc. Hist. Kat. Hafu. t. vi, p. 

 124, t. 6, and iSchumacher, Guincische Plant, p. 431), a genus nearly related 

 to, if really distinct from Balaiiophorn. 



