Mr. J. Miers on the Affinities of the Olacaccae. 173 



Opilia, although often with only a single suspended ovule, some- 

 times exhibits two or three ovules, as I have distinctly seen in 

 O. amentaeea. This fact was evidently more than suspected by 

 Mr. Bentham, who says (loc. cit. p. 674) that it appeared to him 

 there were two ovules in Opilia, three or four in Cansjera, a cir- 

 cumstance rendered probable by the evidently compound nature 

 of the stigma in both genera, but which on account of the ex- 

 cessive minuteness of the parts he could not ascertain from dried 

 specimens : after fecundation he never found traces of more than 

 one ovule. The order however will admit of being divided into 

 tribes, by some of the characters already indicated, but in a sub- 

 sequent memoir I will offer my views on this subject. 



As I shall have shortly to treat of Leretia, and other correla- 

 tive genera, I shall be able to detail at greater length the nume- 

 rous observations that have induced me to propose the separation 

 of Mr. Bentham's tribe Icacineae from the Olacacece ; it will at 

 present be sufficient to state, that they constantly differ in having 

 the stamens alternate with, not opposite to the petals ; they always 

 want the hypogynous disk that forms so frequent and so remark- 

 able a feature in that family, although they sometimes exhibit a 

 similar epigynous gland upon a superior ovarium; they differ 

 also most essentially in the structure of their somewhat gibbous 

 ovarium, which normally will be seen to be 5- celled, but which 

 with a single exception is by abortion always completely uni- 

 locular, and without the smallest indication of any free central 

 placenta, the ovules being generally two in number, attached 

 somewhat laterally, from near the summit of the cell. The 

 fruit differs most essentially in structure from that of the Ola- 

 cacea, being a drupe, enclosing a single nut, with a solitary 

 albuminous seed, that is covered with the usual testa and inner 

 integumental envelopes, and distinguished by a well-marked 

 chalaza and raphe, which, as in Euonymus, is averse or dorsal 

 in respect to the axis of placentation. This is very manifest in 

 Pennantia, a genus clearly belonging to this family. 



In a former page [ante p. 169), while speaking of Villaresia and 

 Bursinopetalum, genera belonging to Aquifoliacece, I pointed out 

 the existence of the identity of structure of the ovarium in those 

 genera with that of the Icacinece, and I stated many other cir- 

 cumstances, tending to prove how closely this tribe is related to 



than the linear cotyledons, of which there were three, equal in size, in the 

 specimen I examined : from the extremity of the cotyledons a thread ex- 

 tended to the umbilicus in the axis of the albumen, which was probably the 

 remains of the embryonary sac. These characters cannot in any single 

 respect be made to correspond with the Olacacece, and Cansjera must again 

 be assigned to its former place, as an anomalous genus of the TJiymeleaceee, 

 until a more fitting position can be given to it. 



