The Morphology of Ruppia Maritima. 135 



Megaspoeangium 



The usual method of development of the megasporangium cor- 

 responds in every particular to that of Ruppia rostellata, as des- 

 cribed by Murbeck (1902, pp. 10, 11), so that it would be use- 

 less to duplicate his careful description here. Stated briefly, the 

 process is as follows : At a stage in the growth of the pistil represented 

 in PL X, lig. 62, or when it is abovit 0.25 mm. in length, a hypo- 

 dermal cell in the young nucellus, in the region marked x, becomes 

 considerably larger than its neighbors, with a larger nucleus, and 

 more densely staining contents. This then divides by a periclinal 

 wall, forming an outer, primary parietal cell, and an inner, megaspore 

 mother-cell. The primary parietal cell now divides twice succes- 

 sively by anticlinal walls, at right angles to each other, forming a 

 plate of four cells, or, through a third anticlinal division, sometimes 

 six cells. ^ During this time the two integuments successively make 

 their appearance. Meanwhile, the megaspore mother-cell enlarges, 

 and with preparations for the first reduction division the history 

 of the female gametophyte begins. 



In Zannichellia, Campbell (1897, pp. 45, 46) finds two parietal 

 layers formed at first, but these later divide into several layers. 

 A much greater development of parietal tissue has been 'observed 

 by Wiegand (1900, pp. 31, 32) m. Potamogetonfoliosus and by Holferty 

 (1901, p. 341) in Potamogeton natans. In the latter case it is def- 

 initely stated that sometimes eight layers lie between the embryo 

 sac and the epidermis. In other respects the development of the 

 megasporangium in both of these genera is essentially the same 

 as in Ruppia. In Lemna, according to Caldwell (1899, pp. 56, 

 57), there are not more than two parietal layers formed, and the 

 other details of the megasporangial development are practically 

 the same as in Ruppia. 



Coulter and Chamberlain (1903, p. 65) state that the suppression 

 of the parietal tissue among the monocotyledons " is usually asso- 

 ciated also with the greater or less development of this tissue," a 

 point which is illustrated here in the Potamogetonaceae by the 

 condition in Potamogeton. " The strongest argument," to quote 

 these writers further, " that suppression of the parietal tissue of the 

 megasporangium is a strong tendency among Angiosperms, is that 

 this condition is universal among the Sympetalae so far as investig- 

 ated." 



^ Rarely two layers of parietal cells are formed (Fig. 78). 



