268 



THE AMERICAN NATURALIST [Vol. XL 



Fig. 2 (PL 1) also seems to show three prothalHal cells, one with 

 no protoplasm, and two with very little. In this case the walls of 

 the cells had persisted after the protoplasm had almost or entirely 

 disappeared, and so far as appearance was concerned seemed to be 

 like the inner layer of the pollen grain wall and continuous with it. 

 Usually, however, these cells do not show so persistent and evident 

 a membrane. Fig. 1 has the number regarded as the normal for 

 Picea. In this case also the membrane persisted after the proto- 

 plasm was entirely gone in the first one formed, showing that the 

 membranes were not plasmic, but in all probability of cellulose. 

 No microchemical tests were made on them. Fig. 1 shows a 

 distinct separation of the protoplasm of the central cell from that 

 of the large tube cell. This separation was the result of plas- 

 molysis, probably due to the killing fluitl. The boundary of this 

 cell, however, appeared to be only a plasmic membrane. At least 

 it was exceedingly thin and closely applied to the surface of the 

 protoplasm. 



Fig. 3 (PI. 1) shows the typical arrangement of cells, with two 

 prothallial cells, but between the main [)art of the pollen grain 

 and the wings there are two s{)ace.s cut oM' l)y cellulose walls, con- 



whether these should be considcrcil ;is cells or not. Tlicv sccnied 

 to bear no definite relation to the main part of the j)oll<'n grain. 

 In this grain the menibranr siirronnding tiic (-(Mitral cell, and con- 

 tinued around the stalk (rll, i)laiiil.v proj(H-ted beyond the limits 



the pollen grain l)y })lasin()lysis. This membrane was very much 

 thimier than the inner cellulose layer of the wall of the pollcii grain, 

 but if it was only a plasmic membrane it was at hvist nuicli firmer 

 than the protr)|)lasin which it bounded, and from which it was 

 partlv separated by plasmolysis. 



There i. .li,.un in Ki- 7 d>l ha variation > .m-A hat .lilVerent 

 from th..M- alreaK di^cuw-.l. (),,K nne . ii.int.-rat in,^- pro- 

 thahial cell is appare.it, l.nt proiccin,; int , th.- inl.c cell of the 

 pollen grain i. a row of three ceil, of a;.f .ro^in,atel^ e.,nal size 

 and structure except that the inneiin ist one ot" the row is dome- 

 shaped while the others are (lattene I a,-ain>l the adjacent cells. 

 These three cells have equally large, prominent, ami well preserved 



