60 INTEODUCTION TO CRYPTOGAMIC BOTANY. 



of the radicle away from, and not towards the apex of 

 the nucleus, are all anomalies which have no parallel in 

 its closely-aUied genera. "Nothing," says Mr. Griffith, "can 

 prove more satisfactorily than the present instance, the abso- 

 lute necessity of tracing anomalous forms back to the earhest 

 period of their development. In this case, the process is at- 

 tended with the desired effect, viz. of reducing anomalies to 

 the ordinary type of formation. It is very evident that the 

 form of the embryo, immediately before its conical apex projects 

 through that of the nucleus, closely resembles the usual form 

 of these organs in other Aroidese, since we have a superior 

 radicle,* a cotyledon, and a tendency to the formation of a 

 lateral slit, as indicated by the depressed areola." 



48. The best characters are often derived from the germi- 

 nation and structure of seeds, but in nearly related plants, 

 these matters are often very different. In Poeonia and 

 Nyraphcea, we have a germination greatly resembling that 

 of Endogens ; in Delphinium, fissum, and some other species, 

 there are two cotyledons joined by their petioles, so that 

 the plumule is obliged to force its way out at the base ; 

 a similar structure occurs in Bunium luteuTn, Prangos 

 ferulacea,-\' and Dodecatheon Meadia. In some Um- 

 bellifers the anomaly is, however, much greater, the two 

 cotyledons are soldered into one, and no plumule is produced, 

 but the energies of the plant are devoted to the formation of 

 a tuber, which sends up a bud in the succeeding year ; this is 

 the case in Bunium Bulhocastanum, and in various other 

 genera ; I believe it is the case, also, in Thapsia villosa ; in 

 Corydalis cava, and many other species, however, a single 

 cotyledon only is produced, and, as in the last case, it pro- 

 duces a bud the following season from the tuber. In 



* The radicle is, in fact, pushed aside by the enormous plumule. If 

 analogies are wanted, compare the phmaule just when the cotyledon 

 has fallen with Riccia nutans. 



t In Prangos feridacea, the first leaf of the plumule makes its appear- 

 ance at the point of confluence of the two cotyledons, the others being 

 free. I have availed myself here of an excellent paper, by Bernhai'di, 

 in Linnaia, vol. vii. p. 661. See, also, Irmisoh, in Botanische Zeitung, 

 1856. 



