376 RECORDS. 



that they do not consist of homogeneous protoplasm, but of 

 different kinds of protoplasmic materials or stuffs that are at the 

 outset arranged, roughly speaking, in horizontal bands or strata, 

 as indicated in the diagram (Fig. 2, B), where the number of 

 strata is arbitrarily assumed to be four. Now, an examination 

 of the manner in which the egg divides gives strong reason for 

 the conclusion that in such forms sls A7np/iioA?/s the first division 

 bisects these stuffs, so that each of the first two cells receives 

 one half of each stratum (Fig. 2, C, £). In the egg of De/i- 

 talimn, on the other hand, this is demonstrably not the case, for 

 the lower stratum passes over bodily into one of the cells and 

 is quite excluded from the other (Fig. 2, U). The symmetrical 

 division in AinpJiioxiis^ the sea-urchin, or the nemertine, gives 

 the immediate possibility of producing two smaller systems simi- 

 lar to each other and to the whole egg. The symmetrical or 

 qualitative division in Dciitaliiiin, on the other hand, does not 

 give such an immediate possibility, for it produces two different 

 systems neither of which is identical with that of the entire egg. 

 It is highly probable that we find here a proximate explanation 

 of the fact that each of the two cells in AniphioxiLS may produce 

 a perfect dwarf, while in Dcntalnun neither produces such a larva. 

 Facts like these are leading us to the conclusion that the imme- 

 diate determining causes of development are to be sought in 

 specific protoplasmic stuffs, or organ-forming materials, that are 

 distributed to the cells in a definite way during division. These 

 materials, definitely arranged, are sometimes plainly visible in 

 the undivided egg. I have, for instance, been able to show that 

 the egg of DcntaliiLDi contains an area of protoplasm at the lower 

 pole that has a causal connection with the formation of the foot 

 and shell, and probably also of the principal part of the meso- 

 blast structures ; for if this area be cut off from the unsegmented 

 egg the resulting embryo regularly lacks these structures. In 

 like manner, Professor Conklin has recently been able to recog- 

 nize in the protoplasm of the unsegmented egg of a species of 

 ascidian the material of the future tail-muscles of the larva ; and 

 though no necessary connection between this material and the 

 muscles has thus far been experimentally proved, my experi- 



