278 ■ 



Willdenow and Brown still retain. This is not merely 

 a dispute of words. If the bodies in question are not 

 really seeds, the offspring of impregnation, they are gem- 

 moE, buds or offsets. In the former instance, they pro- 

 pagate the plant, with its appropriate indestructible cha- 

 racters and qualities ; in the latter they only increase or 

 extend an individual, with whatever characters or pro- 

 perties any variety from which they are taken may pos- 

 sess. Their impi'egnation however is proved, and their 

 germination well understood. The branching jointed 

 fibres, which they in the first instance send forth, have 

 been taken for cotyledons, but prove to be radicles, ex- 

 actly the same as the whole vegetable body in Mosses, 

 or any portion of it, is so very prone to produce. Even 

 a fragment of the receptacle of the anthers has been found 

 to throw out such fibres, which led to a mistaken idea 

 of their originating in the anthers themselves. I am sur- 

 prised that the impregnation of Musci should, after Hed- 

 wig's correct demonstration of all the parts concerned, 

 and their separate functions, still be contradicted, though 

 it cannot be disproved. I humbly conceive it would be 

 as idle to institute experiments to prove the generation 

 of Mosses, as of Hollyhocks, or any other plants. But 

 let those who doubt the fact make such experiments ; at 

 least before they hazard unfounded suppositions. 



The production of perfect germinating seeds, contained in 

 capsules, and consequently produced by impregnated 

 fertile Jlo'wers, is as clear in Ferns as in Mosses, though 

 nothing is certainly known of their stigmas, any more 

 than of their anthers. We are nevertheless content to 

 plead ignorance on the subject, and to presume, by ana- 

 logy, that such parts may exist, rather than to assume 

 the idea of some other mode of impregnation, hitherto 

 unknown, which would be going contrary to the first 

 principles of philosophy ; or, what is worse, returning to 

 the old gratuitous fancies of spontaneous generation. 



The genera of Dorsal Ferns have been founded on dif- 

 ferent characters by different writers. Ray, Tournefort, 

 Plumier, and other early systematic botanists, resorted, 

 in the first instance, to the shape of the frond, than 

 which nothing is more vague, unnatural, or uncertain as 

 a generic distinction. Linnaeus and his followers have 

 trusted to the shape of the masses of capsides, whether 

 round, oblong, linear, or indeterminate, whence far bet- 

 ter characters are obtained, but not such as prove suffi- 



