PHANEROGAMIA : FLOWERS. 349 



until the correct expression of the real common character of a group of 

 forms is discovered. This expression can, and will, only be given by 

 the history of development ; and if we wish to understand ourselves and 

 nature, we must now unconditionally give up that clumsy method of 

 preconception. And thus we must establish and maintain in the special 

 case, that nothing which has not an anther-cell and pollen is a stamen, but 

 another form of the foliar organs of the flower, which we are by no 

 means entitled to refer to that particular form. If we take the Comme- 

 linacece as an example, it is part of their general character to develop 

 five tri-merous circles of foliar organs in the flower; the particular 

 group is characterised by the development of the two outer into 

 calyx and corolla, of the innermost into a germen ; but it lies in the 

 character of the family that the two intermediate are sometimes all, 

 sometimes partly, developed into stamens, and that in the latter case 

 the remaining foliar organs assume peculiar forms, which however are 

 not stamens. Now, these six organs are all called stamens, and it is 

 added that the anthers (consequently the sole exclusive character of the 

 stamens) are wanting ; in this the character of the family is main- 

 tained for all: but does the similarity in different plants lie in our 

 imperfect mode of description, or is it not rather in the plant itself? 

 If the latter were not the case, all our systems would be but a childish 

 game with our words. Such a mode of describing a family is therefore 

 entirely superfluous, so soon as the character of the family has been 

 correctly unfolded. In this example the reference is to the analogous 

 position in different genera, and the position in one and the same circle, 

 from which it is presupposed that all its foliar organs must be de- 

 veloped in a similar manner. But the last is just as much as the first, 

 an empty prejudice ; here there is, indeed, some loop-hole to creep out 

 at, since the accessory stamens which are formed are by no means so 

 strictly characterised organs as to make the term stamina castrata at 

 once evidently inapplicable ; but in Canna exigua (see Plate III., figs. 

 ]2 20.) we have the most striking instance of the entire perversity of 

 this mode of conception, where, in the inner circle of leaves, one is 

 abortive, one becomes the stamen, and one the style. If this circle of 

 leaves were described either as a staminal circle or a carpellary circle, a 

 monstrous Phanerogamous plant would be produced, in which was typi- 

 cally suppressed an organ without possessing which it cannot be a Pha- 

 nerogamic plant at all. 



I next turn to the analogy of the stamen with the sporophyll of the 

 higher Cryptogamia. An unprejudiced examination renders it manifest 

 that the latter is a true foliar organ, in which determinate cells become 

 parent-cells, which, after the formation of four spores, are dissolved, so that 

 the spores, in their peculiar form of simple cells, invested by a peculiar 

 secreted layer, lie free in certain cavities of the leaf previously filled by 

 the parent-cells, and, by the regular rending of the walls of these cavities 

 through desiccation, become scattered. We find this structure per- 

 fectly identical in the phanerogamous anther. I have, in earlier pages, 

 as well as in the paragraphs, remarked upon the analogies, which may be 

 carried out even into individual cases, between the sporophyll and the 

 stamens, more particularly in the Cycadacece and Coniferte. We are 

 unfortunately destitute of any account of the development of the stamens 

 of the Cycadacece; but, familiar with the development of other forms, we 

 may tolerably safely come to a conclusion in this case. In Cycas, on a woody 

 axis with abbreviated internodes, we find a number of foliar organs, 



