394 OVULES OF LOKANTHADS. [BOOK i. 



stigma and the style, as in other plants, are exposed to the 

 direct influence of the pollen. 



The occurrence of a naked ovule has been asserted by 

 Schleiden to take place in Viscum ; but Mr. Griffith has 

 pointed out this error in the Linncean Transactions (vol. xix.), 

 and refers the appearance to what has been termed a solid 

 ovary, upon the evidence of M. Decaisne, on Viscum album, 

 and that derivable from some casual observations made by 

 himself in 1838 on a Himalayan species of the same genus. 

 But he has shown this to have been a mistake ; and he in 

 his last paper declares that there is, perhaps, nothing more 

 constant than the existence of a cavity in the pistil, nor is its 

 absence compatible with the rule, that a pistil is formed from 

 one or more involute carpellary leaves. " It is easy," says 

 Mr. Griffith, " to conceive a pistil without any very manifest 

 cavity ; for the space, which must exist from the disposition 

 of its component parts, may be filled by an extension of the 

 placenta, or the margins of the laminae of those component 

 parts, and indeed by several other modes of extension of its 

 inner surface. But the solidity which I prematurely an- 

 nounced as existing in Loranthus, was of a very different 

 nature, and could not be reconciled to that idea of a pistil, 

 which I have been led to adopt. The anomalies of the mere 

 pistil of Loranthus I at present consider to be confined to 

 the obscurity of the cavity, particularly as connected with 

 obscurity of the placenta. I have, however, seen in Loranthus 

 bicolor appearances which lead me to suspect that much still 

 remains to be observed, not only as regards the conical 

 eminence from the fundus of the cavity, but as regards the 

 true limits of the ovarium. I have not been able to find any 

 such ovarial cavity in two species of Viscum I have lately 

 examined; but I almost feel convinced, that an obscure 

 ovarial cavity similar to that of Loranthus does exist ; for, in 

 addition to the strong doubts that must arise from any appa- 

 rent infraction of a general law, M. Schleiden has stated that 

 in Viscum album there is a nucleus, and consequently an 

 ovarial cavity. The late appearance of the ovule does not, 

 I think, present so remarkable an anomaly as the solidity of 

 the ovarium, unless it can be shown that the development of 



