94 Dr. H. Karsten on the Sexual Life of 



racterize porous parenchymatous cells and vessels always corre- 

 spond to each other in the adjoining walls. 



In the course of the further generation of the pollen-cells the 

 soft mother-cell becomes progressively absorbed, until it is re- 

 duced to a most delicate and scarcely perceptible membrane, as 

 we find it to be in the mature pollen -cell at the time of im- 

 pregnation, and when at length it is broken through by the 

 agency of the three vesicles distended by the absorption of 

 fluid. In this way the openings (pores) originate, in the 

 vicinity of which the intine begins to extrude. 



At the time of the separation of the little-cells from the 

 mother-cells of the Ccelebogyne (which in their origin stand 

 related, like the dotted cells of wood, to the peculiar porous cells, 

 and which ought therefore rightly to be called, not porous, but 

 dotted cells), they can scarcely be recognized as such ; for their 

 walls are so thick, that their cavities are, like those of starch- 

 grains, extremely contracted. Since Fritsche's observations on 

 pollen, we know that the " intermediate corpuscles " existing in 

 fully-developed pollen may not rarely be recognized as actual 

 cells. I represented, in my f Flora Columbia?/ vol. i. pi. 44, the 

 pollen of Schachtea, which also displays very clearly this same 

 relative structure. The intermediate corpuscles, distinctly re- 

 cognizable as vesicles, occupy, as do those in Ccelebogyne, the 

 same position as the " opercula " of Fritsche, and are thrust aside 

 by the expanding intine : until this period they constituted the 

 " porous canal " in the wall of the extine. 



The length of this canal necessarily depends in part upon the 

 thickness of the extine of the pollen, and in part on the dimen- 

 sions of the cell which constitutes the intermediate body. In 

 Ccelebogyne the canal is exceedingly short ; in (Enothera and 

 Clarkia* its development is very considerable, and both the cell 

 concerned in its formation is very large, and the outer portion 

 of the distinct canal hollowed out in the highly thickened extine 

 of considerable length. 



How essential the developmental history of organic bodies is to 

 the correct apprehension of structural relations is exemplified by 

 Schacht's most recent work on this subject of pollen-structure ; 

 for, notwithstanding his marvellous skill in the representation 



sake, vesicles. A superfluous aud incorrect designation for these struc- 

 tures is the expression ' vacuoles.' 



* In Clarkia pulchella the wall of the intermediate cell is intimately 

 united with the intine and extine on either side of it, whilst these two 

 membranes are not in union, so that on a transverse section a fissure 

 appears between them, as Schacht has figured (' Physiologie,' pi. 10. fig. 17), 

 and as he states may be demonstrated very clearly in half-developed pollen- 

 granules of this species. 



