8 Dr. Grisebach on the Germination 



the leaf, a new set of germs, some of which descend and form 

 small and somewhat thick roots, of a very simple cellular tissue, 

 while others put forth peduncles and flowers. Such a process 

 is often repeated, for when the plant is in its full vigour and 

 its base has become a thick rhizoma*, as many as twenty long 

 petioles are not unfrequently to be met with, the leaves of 

 which float as well as the flowers, the latter rising from the 

 upper part of each petiole, all at the same point and opposite 

 to petiolar roots. In some other species of this genus, viz. 

 L. cristatum, between the floral buds not unfrequently one large 

 branch shoots forth which does not end in a flower, but grows 

 laterally to a great length and produces another leaf, which 

 again bears floral buds and roots in the same way near the top 

 of the petiole. It may easily be conceived that the develop- 

 ment of buds on those points is highly analogous to the origin 

 of the young plant from the petiolar top of the long-stalked 

 cotyledon. Besides, it seems clear enough that the formation 

 of germs from the petiole is owing to an union of a cauline 

 organ with the petiole, as well as of the plumule with the coty- 

 ledon in the latter case. 



If we compare this germination, if correctly understood, with 

 that of Trapa, we see here also one cotyledon inclosed by the 

 nut and the young plant at a great distance from it ; but here 

 is this remarkable difference, that the inclosed cotyledon is the 

 long-stalked one, while the other is most obliterated, and that 

 the plant grows in the common way from the axis of both co- 

 tyledons, the radicle being the real origin of the root. In Lim- 

 nanthemum the radicle dies soon after the development of coty- 

 ledons, and the true roots rise at another point, viz. from the 

 petiole of the long-stalked cotyledon. This is indeed highly 

 singular, and I know of nothing analogous to this fact ; but I 

 am also as little aware of any analogy to the further vegetation 

 of that genus. Another resemblance to Trapa lies in the for- 

 mation of roots from the higher parts of the stem ; or in Lam- 

 nanthemum from the petioles which perform the function of a 



* This rhizoma must be considered as the real caulis, and I cannot 

 agree with M. A. de St. Hilaire {Voyage au Bresil, vol. ii. p. 413.) who 

 takes for the stem that part of each petiole which is inferior to the inflo- 

 rescence. This view is refuted by the younger plant. 



