166 ANIMAL LIFE AND HUMAN PEOGRESS 



gametes. Again you will notice it is the two parental com- 

 binations which preponderate. The combinations which go 

 into the cross together tend to come out together. 



Since it was first discovered in the case of the sweet peas, this 

 phenomenon of linkage of factors has been observed in other 

 plants and also in animals. The more thorough the analysis 

 the more frequently does this phenomenon appear. More 

 than two factors may enter into one of these Unkage systems. 

 In the sweet pea, for example, the factor which causes the 

 standard to become erect instead of hooded shows linkage 

 relations with the factors for long pollen and purple flower- 

 colour. In any cross involving all three of these factors the 

 general rule holds good. Those which go in together into the 

 cross tend to eome out together, though the hnkage is never 

 absolute. 



Through the energy of Professor Morgan and his colleagues 

 the study of this phenomenon has recently led to develop- 

 ments of unusual interest. The American workers here had 

 the advantage of using extraordinarily favourable material 

 in the pomace fly (Drosophila), and it is not too much to say 

 that this little creature is to the geneticist what the frog has 

 long been to the physiologist. Besides the rapidity with 

 which it can be bred and the large numbers produced by a 

 single pair, it presents a very considerable number of characters 

 which have been proved to follow the ordinary scheme of 

 Mendelian heredity. We have already seen that the American 

 observers claim to have identified over a hundred factors, 

 affecting most of the external characters. As was to be 

 expected in a form in which the analysis has been pushed so 

 far, linkage phenomena have been encountered in abundance. 



A remarkable feature of the case is that the analysed factors 

 fall into four groups. All the members of a given group show 

 linkage with one another but not with the members of the 

 remaining three groups. Now it is significant that in Droso- 

 phila there are four pairs of chromosomes. This at once 

 suggests that each of the four groups of factors is related to a 

 given pair of chromosomes. Moreover, there are two further 

 facts which strengthen the suggestion of such a relation. In 



