BoT.— Vol. I.] CAMPB ELL—SPA RG A NIUM. 309 



cells. The absence of the large vesicular suspensor cell 

 found in Naias, Alisma and other Monocotyledons is largely 

 a physiological phenomenon, doubtless connected with the 

 nutrition of the developing embryo, and is intimately asso- 

 ciated with the degree of development of the endosperm. 

 Hegelmaier (1874) found that in S. rafnosum the terminal 

 segment gave origin to all the structures of the embryo ex- 

 cept a portion of the root-apex. As he worked entirely 

 with embryos which were simply rendered transparent, but 

 not actually sectioned, he was unable to make out success- 

 fully all the details of the arrangement of the tissues in the 

 older embryos. Nevertheless, his results, on the whole, 

 were remarkable accurate, and show a close correspond- 

 ence to my own preparations of S. simplex which were 

 obtained from actual sections. He found that the cotyle- 

 don and stem-apex are both products of the terminal seg- 

 ment, and that the plerome of the root is also derived from 

 this segment, and terminates at the wall separating seg- 

 ments I and II. 



My own studies of the embryo were necessarily confined 

 almost exclusively to S. simplex, as the material collected of 

 the other species did not contain the proper stages of the 

 young seed. The youngest embryos obtained are shown in 

 figs. 37-40. While these agree in general with Hegel- 

 maier' s figures of similar stages in S. ramosum, there are cer- 

 tain differences which may be noted. The basal, or primary 

 transverse wall {x^ which corresponds probably to his wall 

 (/), is regularly lower down than his figures, indicating 

 that the terminal cell of the young embryo is from the first 

 larger than in ^S". ramosum. It is possible that his segment 

 II may be included in the portion above the basal wall, but 

 this is not probable, as evidences of median vertical walls 

 extending from the apex of the embryo to the basal wall 

 can generally be made out, and from a study of the older 

 stages it is evident that the plerome of the root extends 

 quite to the basal wall, as Hegelmaier describes for S. 

 ramosum. The hypobasal portion of the embryo narrows 

 rapidly, and this portion shows ordinarily but two transverse 



(2) July 17, 1899. 



