RECORDS. 371 



facts which will place clearly before us the form in which these 

 time-honored problems appear to us to-day. 



It is a familiar fact that development begins with the progres- 

 sive segmentation or division of the egg into cells, which, con- 

 tinually increasing in number, finally build up the body of the 

 embryo. Until comparatively recently it was not suspected that 

 the cells thus formed in the earliest stages had any constant and 

 definite relation to the parts of the future body. The fact has 

 now been established, however, that in a large number of forms 

 (though apparently not in all) such a definite relation exists, 

 both the form of division and the prospective values of the cells 

 being constant. In the egg of the ascidian, for instance, the first 

 cleavage-furrow passes pretty accurately through the future 

 median plane of the body, and the two cells thus formed give 

 rise respectively to the right and left sides of the embryo. In a 

 snail's egg the relation is a different one, but is no less definite 

 and constant ; in the four-cell stage, for instance, the material 

 that will produce the shell and foot is located, mainly at least, 

 in one of the four cells. Again, in a worm's egg, after its seg- 

 mentation into sixteen or more cells, we know very exactly how 

 the materials for the head, the segmented trunk-region, the 

 digestive tract, the muscles and the ganglia, are distributed 

 among these cells. In all such cases the embryo seems com- 

 parable to a piece of mosaic-work, each cell apparently having 

 its own inherent particular character, and its own specific role 

 to play. 



These facts place very conspicuously before us a modern form 

 of the problem of preformation which we may conveniently call 

 the problem of ''germinal prelocalization.'" Does this mosaic- 

 like character of the early embryo mean that the cells are 

 inherently different ? Are they in any degree individually pre- 

 destined for their future development ; and if such be the case, 

 can this predestination be traced back to protoplasmic regions 

 in the egg before it has divided into cells ? In other words, 

 does the egg, or does it not, contain prelocalized, predetermined 

 areas that have any necessary or causal relation to the parts of 

 the future embryo ? This is the first guise in which the old 



