ORGANS OF EEPEODrCTIOX. 345 



envelopes, which so resemble each other in their general appear- 

 ance and colour, that it is next to impossible to say where the 

 sepals end and the petals begin. In many other cases also, there 

 is no other way of distinguishing between the parts of the calyx 

 and those of the corolla than by their different positions, — the 

 calyx being the outer series, the corolla the inner. The analogy 

 between petals and leaves is still further shown by the fact, that 

 the former are occasionally green, as in certain species of Cohaa, 

 in a variety oi Eanicnculus, and in one of Campanula rapuncu- 

 loides ; and also from their being occasionally converted, either 

 entirely or partially, into leaves. We may therefore consider 

 that petals like sepals and bracts are homologous with leaves. 



The stamen is, of all organs, the one which has the least re- 

 semblance to the leaf. In describing the structure of the stamen 

 we have shown, however, that the different parts of the leaf 

 may be clearly recognised in those of the stamen. "We find 

 moreover, that in many plants the petals become gradually 

 transformed into stamens. This is remarkably the case in the 

 VvTiite Water-Lily {fig. -iSS) : thus in the flowers of this plant the 

 inner series of petals gradually become narrower, and the upper 

 extremity of each petal exhibits at first two little swellings, which, 

 in those placed still more internal, become true anthers containing 

 pollen. From the fact that the stamens can thus be shown to 

 be merely modified petals, while the latter have been abeady 

 proved to be modified leaves, it must necessarily follow that the 

 stamens are so also. If we now refer to what takes place in 

 many cultivated flowers, we have conclusive evidence at once 

 afforded to us of the leaf-like nature of stamens. Thus in what 

 are called double flowers, the niimber of petals is principally 

 increased by the conversion of stamens into petals ; hence the 

 number of the latter iucreases as the former decreases. Thus, 

 if a double Kose be examined, all sorts of transitions may be 

 observed between true petals and stamens. In other cases, the 

 stamens have been actually transformed into true leaves. The 

 stamen is therefore, consequently, also to be considered as a 

 modification of the leaf. As far as the stamens therefore, we have 

 no difiieulty in tracing both in the normal and abuonnal condi- 

 tions of the parts of the flower, a regular and gradual transition 

 from the ordinary leaves, thus forming conclusive evidence of 

 their being developed upon a common t}^e with them. 



If we now pass to the carpel, we find that transition states 

 between the stamen and carpel are unknown in the normal con- 

 dition of flowers, the difference in the functions performed by 

 them respectively being so opposite, that it necessarily leads to 

 corresponding difierences in structiire. We must, therefore, 

 look to monstrosities or deviations from ordinary structure for 

 examples of such conditions. Even these are by no means 

 common. Such mav, however, be occasionally foimd in the 



