of Limnanthemum lacunosum. 11 



points may be decisive for bringing the Menyanthidea and its 

 congeners into a distinct tribe, but for retaining them in the 

 same family. 



The regularity of the embryo of Limnanthemum lacunosum 

 may be esteemed a strong argument against the correctness of 

 my theory of its germination ; and though it is not decisive I 

 was much disposed to believe my views erroneous, and to con- 

 sider the body inclosed by the seed, either as consisting of 

 both cotyledons (in which case the germination would be ana- 

 logous, e. g. to that of Tropceolium), or as having lost already 

 one cotyledon by mutilation ; but as both cases would have 

 been made out by examination, and the analysis, so far as it 

 depends on the materials, was against them, I could not but 

 recur to my former opinion. Therefore I leave the matter, as 

 it stands now, to those who may examine living specimens ; 

 and until their decision the point may be deemed doubtful. 

 1 have still to add a few words on the reasons by which I was 

 induced to propose a new nomenclature for these plants. 



After the discovery of Villarsia Cristagalli and some Indian 

 Limnanthema, it seemed impossible to retain the old genera 

 Menyanthes and Villarsia, as distinguished by the appendices 

 of their corolla. I followed an observation of Mr. Brown, that 

 the floating species have a nondehiscent fruit; when I proposed, 

 according to their habit, to adopt three genera, one irregularly 

 dehiscent, with the seeds affixed to the middle of the valves 

 [Menyanthes), another with a capsule at the same time loca- 

 licidal and septicidal, and the seeds hanging to the margins of 

 the valves ( Villarsia) ; and a third, comprehending all floating 

 species with an indehiscent fruit, and the seeds affixed as in 

 Villarsia [Limnanthemum, Gmel.). The species of the latter 

 genus have hitherto not been correctly distinguished, though 

 some of them have lately been well defined by Sir W. J. Hooker, 

 and Dr. Wight. I observe that there are many variations as 

 to the largeness and surface of the leaves according to their age 

 which are not to be applied as specific distinctions. Gene- 

 rally the leaves are, in the first instance, smooth, then appear 

 glands on the inferior surface, these are obliterated and more 

 or less roughness remains ; sometimes the glands are to be seen 

 in the bottom of small circular processes, from which appear- 



