EVOLUTION, VARIATION AND HEREDITY 467 



and in many crosses we get a condition intermediate between the 

 parent forms. 



What is the explanation of this phenomenon ? In the first place, 

 it is obvious that the germ cell of the parent may be considered 

 to carry something which we term a factor, let us say D or R, that 

 determines the appearance of the character in the next generation. 

 If there were a complete blending of these characters in the F. i 

 generation, then we should expect all the F. 2 and subsequent 

 generations to remain the same, but they do not. Suppose we 

 imagine two sorts of pollen grain to be produced in the Fig. i 

 generation, one with the factor D and the other with the factor R, 

 and in the same way two sorts of ovules. We can easily see all the 

 chance combinations possible and the result. 



D may fertilise D giving DD 

 D R DR 

 R D RD 

 R R RR 



This is the proportion that we actually do get, and so it seems 

 the reasonable explanation. That is to say, in the germ cells of the 

 F. i generation the unit characters are not blended, but segregated 

 in different cells, a phenomenon that we term segregation. It can 

 be checked by the consideration of other cases, e.g. suppose the 

 F. i generation, i.e. DR, be back-crossed with the parent R, then 

 we get offspring in the proportion 2DR : 2RR, exactly what we 

 should expect. To take a more complicated example, suppose we 

 cross a green smooth pea with a yellow wrinkled seed. The F. i 

 generation is smooth, as this is dominant to wrinkled, and it is 

 yellow, as this is dominant to green. Thus we can represent this 

 diagrammatically : 



Sg X wY 



^ 

 F. i. SwYg 



Now if we consider the possible ways in which these factors can 

 be segregated in the gametes of the F, i generation we shall find 

 it is four, namely, SY, Sg, wY, and wg. The possible combinations 

 of these gametes can be seen if we place them along the sides of a 

 square and then divide the square up into 16 smaller squares. In 

 each square of the vertical columns place the symbols above the 

 column, and in each square of the horizontal rows place the symbols 

 at the beginning of each row. The squares then indicate all the 

 possible fertilisations from gametes that are so constituted. It will 

 be seen that in 9 squares the letters S and Y occur ; in 3 squares the 



