242 Dr. L. Kuy on the Structure and 



that a series of divisions takes place in the central cell, leading 

 to the formation of the special mother cells. The first septa 

 are usually directed exactly in accordance with the longitudinal 

 axis of the antheridium, and placed at right angles to each 

 other in three directions ; afterwards radial walls alternate 

 several times with tangential ones. The cells of the last 

 grade, the number of which is not constant, become rounded 

 off from each other. Their very delicate membrane is followed, 

 immediately Avithin, by a hyaline plasma-zone ■ the central 

 part of the contents is distinctly granular. 



The opening of the antheridium is here also evidently 

 effected by the turgescence of the two ring cells. After the 

 opercular cell is irregularly ruptured, and the special mother 

 cells are evacuated, these extend themselves inwards, at the 

 same time becoming slightly shortened. By this means are 

 formed radially perpendicular folds, which, Avhen looked at 

 laterally, often present a delusive resemblance to true septa in 

 appearance*. Here also, as in Aneimia iiirta, we may easily 

 convince ourselves, by examination from above, that they do 

 not attain the outer membrane. 



Exceptionally we sometimes observe antheridia with only 

 07ie ring cell. This has then, so far as the mature state en- 

 ables us to judge, exactly the same origin as the upper ring 

 cell in normal antheridia : it is the sister cell of the opercular 

 cell. 



Eather more frequently antheridia Avith three ring cells are 

 observed. The middle one, in this case, is probably formed 

 by a funnel-shaped septum in the same Avay as the lower one. 

 This was certainly the case in two abnormal antheridia, in 

 which the second ring cell had attached itself laterally and 

 obliquely to the lower one (fig. 16). 



Gihotium Schidei directly approaches Asjjlenium alatum, but 

 sliOAVs some remarkable peculiarities. The lowest of the two 

 ring cells, which are here present in the great majority of the 

 antheridia, usually rests upon a basal cell which is only deve- 

 loped on one side, and is then loAver upon one side than on 

 the other, Avhilst the upper ring cell is more regularly deve- 

 loped (fig. 19). The opercular cell does not remain undivided, 

 but is divided into two daughter cells of unequal size by a 

 wall, which is perpendicular to the outer Avall, but strongly 

 convex toAvards the middle point of the cell. The larger cell 

 is crescentiform, the smaller one elliptical, pointed at both 



* In two cases I believe I positively ascertained tlie presence of a single 

 true longitudinal Avail in one of the ring cells. I regard them as supple- 

 mentary structures. As to the mode of their production, I can, unfortu- 

 nately, say nothing further. 



