74 



termined axes or poles, pure accident will determine the relation of 

 those poles in the attached eggs to the direction of gravity. 



Segmentation was first seen about twenty-four hours after fertili- 

 zation. After twenty-four hours more the upper half of all the eggs 

 was divided in to from fifteen to twenty segmentation spheres, well 

 rounded and distinct in the middle of the upper surface, and becoming 

 flatter and less sharply defined on the sides. The under half of the 

 eggs showed no trace whatever of segmentation. 



On stripping ofi" a sheet of one hundred or more eggs, one can 

 compare the two surfaces and note the contrast between them. The 

 under surface of each egg may be identified by the presence of a distinct 

 scar on the chorion showing where it was attached to the glass. Now 

 when such a sheet of eggs, about forty-eight hours old, was turned 

 over, the unsegmented hemisphere, now uppermost, begins to segment 

 rapidly, and in from ten to fifteen minutes it is broken into spheres 

 of approximately the same size and number as those on the original 

 upper surface of the egg. It should be observed that the original 

 under surface on being turned upwards, does not segment in the regular 

 sequence, two, four, etc., but breaks up at once into as many spheres 

 as are found on the original upper side for that age. 



The original upper side, after reversing the egg, becomes flattened 

 against the chorion by the weight of the overlying yolk and conforms 

 perfectly to it so that the furrows become fainter and fainter. However 

 they do not disappear altogether but are still visible after the opposite 

 side has begun to segment. This shows conclusively that the egg has 

 not rotated inside the chorion, and indeed one can watch the process 

 closely enough to feel satisfied that such a possibility is entirely ex- 

 cluded. 



These fixed eggs then, externally, appear to be meroblastic, while 

 the eggs taken from the sand are holoblastic. However if the artificially 

 fixed eggs are removed from the glass and rolled about, they soon show 

 equal segmentation of all surfaces and appear holoblastic. The eggs 

 naturally deposited in the sand are no doubt fixed and if examined 

 »in situ« would probably appear to be meroblastic and with the seg- 

 mented side uppermost. It is no doubt the rolling about incident 

 upon freeing the eggs from the sand that causes them to segment equally 

 over the whole surface, as described by most authors. 



It is well known that under normal conditions the segmentation 

 nuclei are pretty uniformly distributed throughout the yolk, and it is 

 therefore highly probable that in the »fixed«, apparently meroblastic 

 eggs nuclei are already present in the unsegmented under side. It 

 would appear therefore that the weight of overlying material has pre- 



