THE BIOLOGY OF THE CELL SURFACE 



meres lie apart, they may with care by puncture of the 

 vitelline membrane be removed as four independent cells. 

 If brought together again and kept in close contact, by 

 surrounding them with fibres of lens paper, they unite and 

 develop into a single embryo. Also, two blastomeres from 

 one egg when brought together with two from another in 

 some cases united and developed into one embryo; often 

 however union failed to take place. I found that this 

 failure resulted whenever the transferred blastomeres were 

 not in exactly the same moment of development. For 

 example: two lots of eggs from the same female were fer- 

 tilized at an interval of one minute apart from each other. 

 At second cleavage, the four blastomeres were removed 

 from one egg of each fertilized lot. Two blastomeres of the 

 one were next brought together with two blastomeres of 

 the other; no union was established. Observation revealed 

 that the surface-changes in the transposed blastomeres 

 were different. This observation furnished the clue for 

 the failure of union between transposed blastomeres taken 

 from a lot of eggs which had been fertilized at the same time : 

 For the establishment of the interconnection of blastomeres, 

 the ectoplasmic condition must be identical. Therefore, the 

 union of blastomeres can be made certain if by direct obser- 

 vation their surface changes have been found to be in the 

 same stage. Since in any lot of eggs in best condition fer- 

 tilization takes place in every egg at almost the same instant 

 and the development ensues at much the same rate, at 

 second cleavage the difference in the ectoplasmic behavior 

 between the most slowly and the most rapidly developing 

 egg is never as great as that in eggs fertilized at an interval 

 of one minute. The failure of blastomeres from two eggs 

 of the same fertilized lot to unite therefore is to be attrib- 

 uted to changes in ectoplasmic behavior resulting from a 

 difference in rate of development which is of seconds only. 

 To mv knowledge, there exists no observation which so 



338 



