144 MORGAN— THE CONSTITUTION OF [April 23, 



In the third place, it was shown that in some eggs each of the 

 first two, or first four cells derived from the egg is capable of form- 

 ing a whole embryo. This result creates a strong presumption 

 against the adequacy of Weismann's interpretation of development. 



Meanwhile one of the greatest biological discoveries of the last 

 century — one that had a very direct bearing on the traditional in- 

 terpretations of predetermination — was forgotten. I refer to 

 Mendel's work. Mendel showed that when two related organisms, 

 differing from each other in a single character, are crossed, and 

 their offspring are again bred together, that in the second genera- 

 tion individuals appear that are like their grandparents. He showed 

 that the numerical proportions, in which they appear, could be 

 explained on the assumption of one factor difference between the 

 original forms. This result might be interpreted to mean either 

 that the two original germ cells, taken as a whole, represent such 

 a factor difference ; or it might be interpreted to mean that the 

 original germ cells had one particulate difference. But Mendel went 

 further, and showed that when two related organisms that differ in 

 two, or three, or more different characters are bred to each other, 

 all possible combinations of the original characters appear later. It 

 might seem then that we must abandon the view that each germ cell 

 is to be thought of as a whole, for we see that the parts of each 

 can be separated to become parts of others. In this sense Mendel's 

 results seem to furnish a brilliant confirmation of Weismann's 

 theory, in so far as it relates to preformation in the germ, and in 

 the last edition of his " Vortrage ueber Descendenz Theorie," Weis- 

 mann put in his claim to this verification. 



In fact, Mendel's discovery does furnish a strong argument in 

 favor of that part of Weismann's view that deals with the con- 

 stitution of the germ-plasm, but it by no means confirms that part of 

 Weismann's theory which postulates that embryonic development 

 is a sorting-out process of representative particles. 



Let us turn our attention, then, to Mendel's law and examine in 

 how far it justifies an assumption that there are specific substances 

 in the germ cells. 



Mendel's law postulates that the early germ cells (and it may 

 be added all of the body cells too) contain two of each kind of the 



