458 Mr. B. Clarke on Relative Position ; 



of spikes or racemes, one of the flowers only being female. In 

 Mercurialis annua the female inflorescence is peculiar, consisting 

 of two female flowers growing from one side of an axil, the cen- 

 tral bud of which some time after developes as a leaf-bud. Of 

 the two flowers the internal one is a lateral branch of the axil, 

 and the central undeveloped bud is its axis, and the external one 

 springs from its peduncle at or near the base. As thus viewed 

 the carpels are all anterior and posterior instead of right and 

 left, as at first sight they might be supposed to be ; and Vinca 

 speciosa has apparently the same mode of inflorescence and posi- 

 tion of carpels, those of Apocynum (where the inflorescence is 

 regularly branched) being also anterior and posterior. 



Stilagine^. If the female flowers of Antidesma are com- 

 pared with apetalous Flacourtiacese, their near approach will be- 

 come obvious, and on that account Stilaginese are placed near 

 Lacistemacese. 



DiPTEROCARPEiE. It is perhaps a question whether this 

 Order and Chailletiacese are branches of the Urtical and Amental 

 Alliances, as they agree in several parts of their structure with 

 Tihacese and Anacardiacese, and particularly with Chlsenacese in 

 having two collateral pendulous ovules with the raphe next the 

 placenta. There appears also to be in Dipterocarpese a strong 

 tendency to a free central placenta, the dissepiments in Vateria 

 being very fugitive, although the central column is continued up 

 into the style. 



CucuRBiTACE^. The ovary of Sicyos angulata consists appa- 

 rently of two or three carpels united by their margins, and from 

 the form of the ovary and placentation (supposing this to be its 

 structure) the fertile carpel is for the most part anterior, but may 

 perhaps be occasionally lateral if not posterior*. 



MoREiE. In Broussonetia and Morus there are two stigmas 

 terminating two ribs in the ovary; one of these ribs is much 

 thickened, and to it the ovule is attached. The attachment of 

 the ovule is therefore an evidence of the position of the carpel, 

 which is readily ascertained by making a cross section of the 

 ovary. The ventral side of the ovary, besides being thickened, 

 has also a more vascular appearance, as' if the rudiment of a 

 second carpel were present (PI. XV. figs'. 19 & 20). In Morus 



* I find, contrary to my expectation, that the raphe is here not next the 

 placenta hut timied directly away from it, which seems to require that Cu- 

 ciu'bitaceae should be stationed in the Ficoidal Alliance, which would then 

 bring them into closer relation with Onagrai'iae and Campanulacese, to the 

 former of which, according to Auguste de St. Hilaire, they are most nearly 

 alUed. The ovule is attached to the apex of the cell, but a slender and 

 entire cord of fibres continued from the raphe (which is quite distinct) 

 crosses over the exostome and passes down the opposite side of the ovary. 



