1892.] 



MICROSCOPICAL JOURNAL. 



131 



strands are of good size. At the middle lamella the connecting 

 threads of protoplasm are three to seven or more in number, and 

 bulge outward, the whole somewhat resembling a barrel in shape. 

 Unfortunately a drawing of the alder was mislaid. The drawing 

 of the endosperm oi Hetcrospathe elata^ a palm seed (Fig. 3), 

 has been copied from the work of Mr. Gardiner, and very closely 

 resembles that of the alder, above referred to. In A is seen en- 

 dosperm of Hcterospathe clata^ showing threads of the tips of 

 four contiguous strands. The middle lamella is but little devel- 

 oped ; in B, ripe endosperm of PJuviiix dactyllfcra after treat- 

 ment — one cell, with parts of several others. The strands of pro- 

 toplasm extending out from the central mass in the cells of Ani- 

 elaiichicr (Fig. i ) are often forked near the tips. 



The specimen is one of the best examined. The cells are not 

 large, but are comparatively long, and having exceedingly long 

 protoplasmic strands. From one cell (A) many pointed strands 

 of protoplasm seemed to radiate in all directions, their form of 

 cells were numerous, and in some cases strands extended to other 

 cells. The peculiar point illustrated by B, C, D, and E is the 

 forked appearance of protoplasmic strands at the point of junction 

 between the cells. Here in many cases the connection seems to 

 be broken, but in above, as in E, exceedingly fine threads appar- 

 ently unite the two strands. The middle lamella is easily seen 

 in this specimen, and it is here where we always find the break, if 

 any, in the connecting strands. 



In Cratccs^us (Fig. 3) the continuity is very pronoimced. The 

 middle lamellae are distinct. At the middle lamella there is a 

 difference apparently in the strands of protoplasm. Strands do 

 not appear as dense at this point as elsewhere. Strands in all 



