80 Mr. J. Miers on the Tribe ColletiejE, 



one half of which Hes within the sinus, the other half in the 

 nipple of the albumen; the cotyledons are anterior and posterior 

 in regard to the axis of the fruit. 



The presence of the distinct external crustaceous shell, always 

 more or less pervious at the basal extremity (or one of its sides), 

 and the position of the raphe, as above described, attached to an 

 intermediate integument, making nearly a complete circle round 

 the seed, — are features that occur not only in all the Culletiece, 

 but in all the genera of the Rhamnacece that 1 have examined, 

 though subject to several modifications, which I will mention. 

 In Rhamnus catharticus, however, as well as in R. Alaternus and 

 a few others, the raphe, though still peripherical, is differently 

 situated, for it is everted yet further from the position mentioned 

 in CoUetia ; it is seen upon the dorsal face of the seed, — a well- 

 known occurrence, that has been ingeniously accounted for by 

 Mr. Bennett* ; but before we adopt his conclusion, it is desira- 

 ble to investigate the circumstances attendant on the pheno- 

 menon. Mr. Bennett says that he has universally found, through- 

 out all Rhamnacece, that the ovules, in their early state, uniformly 

 exhibit a ventral raphe ; and hence he infers that any variation 

 subsequently observed in its relative position, whether it be 

 shifted to one side or round to the dorsal face, must be due to 

 a simple twisting of the funicle, by which the ovule is turned 

 upon its axis, either to the extent of a quarter or of half a 

 revolution. Were this explanation required only in the case of 

 the raphe becoming dorsal, we could scarcely refuse our assent 

 to it, because in that case it ajjpears to solve the difficulty with 

 apparent satisfaction, notwithstanding that Mr. Bennett con- 

 fesses his inability to assign any cause for so singular a rotation 

 of the ovule. 



The first point to be considered is the soundness of the foun- 

 dation of the argument, that at the period of its fertilization the 

 ovule has its raphe directed to the ventral side, and that it sub- 

 sequently turns awaj)^ from it by a lateral torsion of its funicle. 

 Mr. Bennett states {loc. cit. p. 131) that he has invariably found 

 this to be the case in the Rhamnacece', and this statement has 

 been accepted as an indisputable fact by nearly all botanists. 

 Some few, however, have ventured to cast a doubt on the sub- 

 ject : Mr. Ciarkcf says he has always found the raphe dorsal or 

 lateral in that family from an early period ; while Bi'of. Agardh 

 affirms the reality of this latter circumstance, and protests in 

 positive terms against the prevalent belief on this subject J ; in 

 his work above referred to, he gives a figure as an example of 



* mant. Juv. nir. 131. 



t Linn. Proceed, ii. 148-150; Ann. Nat. Hist. scr. L'. .\i. p. 85. 



X Theor. Syst. Plant, p. 17^, and jdatc 15. fig. 5. 



