of Limnanthemum lacunosum. 9 



stem. May not this analogy help to prove how erroneously 

 some authors have considered those of Trapa to be stipules ? 



If we investigate more particularly the structure of Limnan- 

 themum, a genus which adorns the waters of far the greater 

 part of both hemispheres, we shall not without some interest 

 inquire into the peculiarities of its organs. For my present 

 purpose it seems sufficient to touch chiefly upon the structure 

 of the seed in this genus, which certainly does not well corre- 

 spond with the theory of its germination just proposed, or at 

 least there are no characters in the seed by which the follow- 

 ing phenomena may be foreseen. I am highly indebted to 

 my friend Dr. Schleiden* for an examination of these parts ; as 

 he has been for years, and with the fullest success, occupied in 

 investigating the development of ovula ; and the acknowledged 

 precision and admirable acuteness of his microscopical re- 

 searches afford the following observations a greater weight 

 than my own inquiries, if unassisted, could do. 



The question whether the Menyanthidece really belong to the 

 Ge?itianece maybe solved more easily if we compare the essential 

 parts of their flowers before we are biassed by a view of their 

 vegetative organs. The Gentianece have an ovarium com- 

 posed of two carpels, which bear an indefinite number of ovules 

 at their sutures. The genera with a placenta centralis make 

 no exception to this rule, the latter being combined of four 

 placentas, whose vessels rise from the introflexed margins of 

 the carpella in an earlier state. The seeds have only a single 

 testa, are antitropous, and contain a small cylindrical embryo 

 which lies in the middle of a fleshy albumen ; the latter being 

 formed in the interior of the sacculus embryonis, while the 'nu- 

 cleus is obliterated. The cotyledons in that state are oblong 

 and somewhat thickish (fig. 14 — 16), and between them may 

 be observed a thin layer of albumen, so that Mr. Brown called 

 the cotyledons of Menyanthes and Gentiana lutea (i semi-dis- 

 cretae." This structure of the seed is quite identical in the Me- 

 nyanthidece and in the other Gentianece ; there is not the least 

 trace of disproportion in the cotyledons of Limnanthemum 

 while still in their seed, though the radicle be very slightly 



* Dr. Schleiden's interesting paper on the development of the organiza- 

 tion in Pheenoganious plants will be found in Phil. Mag. for February 1838. 



