278 Mr. H. McE. Knower on the 



resulting from segmentation become about equally distributed 

 to all parts of the Qgg. At an early stage most of the cells 

 have reached the surface of the yolk, only very few remaining 

 behind as vitellophags. After attaining the surface the cells 

 of the blastoderm (it may, perhaps, be spoken of by this- 

 term, though protoplasmic continuity between its cells cannot 

 be shown) continue to divide at all points, though the nuclei 

 in the posterior end divide more rapidly. For a number of 

 stages this becomes more noticeable, but it is also evident that 

 the actively dividing nuclei are not confined to the restricted 

 area on the ventral surface to be occupied by the disk. The 

 nuclei of the dorsal surface near the posterior pole are as 

 numerous and as near together at this time as those on the 

 ventral surface. From this stage to one exhibiting a sharply 

 outlined germ-disk about to be covered by an amnion there is 

 apparently a concentration of the cells on the surface toward 

 the ventral side of the tgg to a point just beneath the micro- 

 pyles. This, as has been said, is the place where the primary 

 rudiment of the embryo is finally situated. Hence the em- 

 bryonic disk is seen to be due not simply to an active multi- 

 plication of the cells of a restricted area of the blastoderm, 

 but likewise to a concentration of the blastoderm-cells. This, 

 it will be remembered, is what McMurrich has recently shown 

 to be true in the development of Isopods (5). A similar 

 concentration has been observed in the establishment of the 

 first rudiment of other insect embryos; but in the termite it 

 is especially marked, owing to the comparatively small size 

 of the germ-disk (see Patten for the Phryganids (6) and 

 Wheeler for Doryphora (8)). Sections of the disk during 

 this concentration show that cells are crowded beneath the 

 surface from a very early stage in its formation. This takes 

 place at all points in the area of the disk, and the surface 

 nuclei also divide tangenlially here and there to separate cells 

 which adhere to the lower surface. This is the beginning of 

 the formation of the " under-layer." 



Surface views of older embryos show two changes in the 

 disk. Near the centre a dark spot appears, and in the same 

 stage the posterior margin becomes marked out as a semi- 

 circle of especially closely crowded nuclei. Sections of such 

 disks show that the " under-layer " cells have become more 

 numerous, and have collected into a plug projecting into the 

 yolk and making the dark spot seen on the surface. The 

 posterior semicircle of crowded nuclei represents the first 

 rudiment ot the amnio-serosal fold. It is, at this period, 

 merely a more thickened margin of the disk. The area 

 between the central plug and the amnion thickening (if it 



