264 



ACOTYLEDONOUS PLANTS. 



the earliest stage examined, very cucullate, the point deflexed 

 anticously and meeting the axis. 



It contains at this time two nuclei, one rather more ad- 

 vanced than the other : protruding from the cup or annulus, 

 rudimentary paraphyses, and confervoid filaments occur, but 

 these have no relation with the ovula, beyond that of being 

 contained in the in voider um. 



These filaments, appear to begin to get access to the ovula 

 at the time when the nucleus is just enclosed by the tegu- 

 ment, but the access is by degrees, for at this period, two were 

 observed only just entering. 



The grains disappear very slowly, a few may be found even 

 when the capsules are really mature, or the gongylus quite 

 marked out. 



One very remarkable circumstance was noticed regarding 

 them : in one instance, a bundle of them was seen attached 

 to the axis, and one or two single ones in which some of 

 the joints were a good deal enlarged, of a yellow green 

 tint, with two evident nuclei approached either end (or one, 

 when the enlarged joint was terminal) near its base, some- 

 times with a rudiment of a third : those unchanged in size, 

 still often presented important changes ; thus those near the 

 terminal one had become divided into two, by aline across the 

 middle ; the joints hence resulting being broader and more 

 beadlike, in others again, at the ends of each was a small nu- 

 cellus, corresponding with the upper and base one of its next 

 neighbour ; and near this was the shadowing out of a large 

 cell ! ! 



All this is quite confervoid, and opens a rich source of spe- 

 culation. At the same it may throw difficulties in the way of 

 proving the grains to be pollen, at least with some. 



Appearances are in favour of the condensing points being at 

 some period ternary, or quaternary, always they are an- 

 gular, often appearances are seen of division ; but this is so 

 obscure, as to probably arise from suggestions of analogy. 



Appearances are in favour of the nucleary grume developing 



