Fig. 6. 



Fig. 7. 



The first furrow divides the egg into two very unequal parts — a large 

 macromere and a small mieromere. The latter is about one-fourth the size 

 of the first. Each segment lias a single nucleus. The next furrow — at 

 riglit angles to last — divides each of the-first two into two ecjual halves. In 

 some cases the larger cell divided five minutes before the smaller, but in 

 other cases the reverse process took place. The furrows dividing the 

 egg into these four parts nearly coincide for the micromeres and macro- 

 meres. The third furrow divides both micromeres and macromeres into 

 four each, and is at right angles to the first two planes of division. The next 

 furrow is seen to divide the four macromeres into eight, is at right angles to 

 the last two furrows, or is parallel to the first plane of segmentation. At the 

 same time each micromere divides into two, but no definite plane of division 

 is apparent. There are now eight macromeres and eight micromeres. Each 

 of the eight macromeres divides into two in planes at right angles to the last 

 furrow or parallel to the second and third planes of division. This is followed 

 later by a division in the macromere in a plane at right angles to the last or 

 parallel to the first and fourth ; but only those cells above the horizontal far- 

 row (fourth) were seen to divide. The micromeres were not seen to keep 

 pace with these last divisions, so that the upper (micromere) pole of the egg 

 is covered with a mass of cells of about the same size. Sections of eggs in 

 this stage show pyramidal figures somewhat similar to Fig. 6, but the upper 

 pyramids are smaller, and some do not run to the centre of the egg. Each 

 pyramid has a nucleus in its outer part, and each nucleus is accompanied 

 by a mass of protoplasm which sends out processes into the surrounding 

 yolk of the cell. Soon after this the formation of yolk pyramids ceases, 

 I and the nuclei (and their protoplasm) lie at the periphery of the egg. At 

 I the upper pole the nuclei are much more numerous, but smaller than at 

 ^ the lower, and the protoplasm forms a thick covering to the egg. Here, 

 1" also, the blastoderm develops rapidly. At the lower pole there are scat- 

 I tared nuclei at the surface of the yolk. 



The early separation of the egg into two unequal parts is apparently 

 closely connected with the more rapid development of the embryo in the 

 region of the smaller segment. About the time when the pyramids become 

 lost (and perhaps at that time) each of the peripheral nuclei divides radi- 

 ally into an outer and an inner nucleus — each of course with its cell proto- 

 plasm. It takes place first over the upper pole, and not till very late over 

 the lower area. This is undoubtedly the same thing as the delamination 

 in the smaller eggs. The difterences are these — that in Pallene the pyra- 

 mids do not divide themselves each into two cells, but only the nuclei and 

 protoplasm ; and this takes place later at the lower than at the upper pole. 

 The delaminated n\iclei remain just under the outer cells — ectoderm — and 

 only exceptionally do one or two wander into the yolk. These nuclei 

 form the endoderm of the mid-gut after having devoured the yolk. The 

 protoplasmic layer at the upper pole becomes wider and the nuclei more 

 numerous, each nucleus being the centre of a distinct cell. At (me place 

 may be seen from surface veins an opaque area (much like the early stage 

 of the primitive cumulus in spiders), and sections show that here an invagi- 

 nation of ectoderm is forming — the stomodajum. Around its periphery 

 there is a collection of cells which are, no doubt, the beginnings of the 

 mesoderm. The stomoda-um increases in depth, and at this time the 



