CROSSING. 75 



acters must have originated as the result of crossing. In our 

 work with the house- rat group we saw the origin of several new 

 recessive characters, and we were in a position to study the 

 behaviour of these new characters in crosses,immediately upon 

 their origin. In no case did we observe a mono-factorial differ- 

 ence between the parents and the first representatives of the 

 new varieties. But we often found quite normal uni-factorial 

 ratios in crossing the new colours with pure-bred un-related 

 rats. For instance, we found that the waltzing character, 

 although it originated by a 'double absence" behaved as a 

 simple recessive, if we tested our waltzers of composite origin 

 by mating them to pure house- rats from Holland. And the same 

 held true for yellow colour. Although it obviously originated 

 by the simultaneous absence of two genes, one of which only 

 was present in each of two wild species, the yellow colour be- 

 haved as a simple recessive in a cross with wild Javanese 

 house-rats. 



From the hybrids between our first waltzing male and a 

 wild-caught house-rat from Holland, we obtained three wal- 

 tzers among eleven young in two litters, obviously a mono- fac- 

 torial ratio, showing that the difference between the waltzer 

 and the wild rat was a difference in presence and absence of one 

 gene. 



And when we mated hybrids between a yellow male and a 

 Javanese house-rat female back to yellows, we obtained a 

 mono-factorial ratio, five agouti to five yellows, in three litters. 



In the last few years we have also been cross-breeding strains 

 of field- rats, belonging to the Mus Rattus group. The common 

 field-rat found throughout the Malayan Archipelago, from Sing- 

 apore south through Sumatra and Java to Bali is remarkably 

 homogeneous. The rats from south Sumatra, which we are 

 still breeding, have a tail which may be very slightly longer 

 than that of the Javanese rats, but otherwise we have not 

 beeen able to find any constant difference, even in an examin- 

 ation of hundreds of freshly-killed rats. The hybrids between 

 these two rats are in every way similar to both parents. In the 



