OF THE OVULE IN ORCHIS MORIO. V 



down the side of the summit of the embryo-sacs, which always contained the vesicles 

 within. In some embryo-sacs (figs. 20-26) one of the vesicles had begun to develope 

 further, dividing into two cells by a horizontal septum, the upper dividing again and grow- 

 ing out in a conical form through the endostome, to produce the confervoid filament which 

 was described by Mr. Brown, and which Schleiden has certainly mistaken for a develop- 

 ment of the pollen-tube. 



June 3rd. Traced the pollen-tubes to the embryo-sac, and saw them lying on the out- 

 side, and again satisfied myself that the vesicle within the embryo-sac (the germinal ve- 

 sicle) is the first cell of the embryonic body. It generally exhibits a slight collection of 

 protoplasm at its base, and soon after the pollen-tube reaches the surface of the embryo- 

 sac divides into two cells, the upper dividing again and growing out into an articulated 

 filament, the cells of which are formed by the production of septa in the same way as in 

 Confervas, hairs of Phanerogamic/,, &c, the mucilaginous layer (or primordial utricle of 

 Mohl) being rendered very evident by the application of iodine (fig. 29). The lower part 

 of the embryonic body enlarges while the filament is growing out, and soon perfectly fills 

 the embryo-sac. It appears to me that the process of cell-formation in this lower part, 

 by which the embryo is produced, varies in different cases ; generally the lowest cell 

 enlarges very much and becomes filled with dark mucilaginous matter, and then this is 

 soon divided into a number of cells by the formation of septa. Nuclei were visible in all 

 the cells very soon after their origin, but I could not form an opinion as to their relation 

 to the cell-formation, or determine how or at what period they were really produced. In 

 the earliest condition they resembled clear vesicles, not granular bodies such as Schleiden 

 describes. 



In some cases two confervoid filaments are produced, two of the germinal vesicles 

 undergoing development. I met with this several times, but omitted to draw them, in 

 the hope of subsequently finding a more favourable specimen, which I was not fortunate 

 enough to do. 



The obvious conclusions from the foregoing observations appear to be, that the embryo 

 is really produced by the ovule itself; that a germinal vesicle exists within the embryo- 

 sac before the pollen exerts its influence ; that the pollen-tube penetrates the coats of the 

 ovule to reach the embryo-sac ; and that the passage of the pollinic fluid through the 

 intervening membranes impregnates the germinal vesicle and determines its development 

 into an embryo. 



Since the investigations were made with every precaution, and their results are in per- 

 fect accordance with those of Amici, Mohl, Muller and others, I think that I am justified 

 in believing them to be a sufficient refutation of Schleiden's views, so far as the plant in 

 question is concerned ; but as to their positive value, as to the evidence they afford of the 

 actual nature of the process of impregnation, I still regard them as insufficient. I am 

 not convinced that the whole of the pistillary cords are composed of filaments directly 

 produced by the pollen-granules. It is not yet shown whether there is any relation 

 between the application of the pollen on the stigma and the development of the germinal 

 vesicles ; it is only clear that these last exist before the pollen-tubes enter the ovules. 



VOL. XXI. c 



