76 CROSSING. 



F2 generation however, we obtained waltzers. These are of 

 course in no way related to the waltzers in the house-rat series. 

 Both physiologically and genetically these waltzing rats behave 

 differently. They spin around with the same speed as house-rat 

 waltzers or waltzing mice, but they are better able to take care 

 of themselves. They can climb fairly well, whereas the house-rat 

 waltzers cannot. They can walk around their cage, and eat, 

 without having to interrupt their meal for seemingly involun- 

 tary spinning. The inheritance of this character in this series of 

 rats is not yet worked out. Curiously enough, we find, that 

 matings of two waltzing rats can produce all normal off-spring, 

 so that it seems as if we had at least two genetically different 

 kinds of waltzing field-rats. The analysis in this case is made 

 very difficult, and may prove to be impossible, by the fact 

 that waltzing females have never in our experience been known 

 to raise their young. A foster-mother must be provided for the 

 few young which are not killed by the mother at birth. A 

 complete litter of living young from a waltzing female can be 

 obtained only very rarely. 



No evidence for the production of novelties has ever been 

 found by us in pure families of wild rats, not even in the famil- 

 ies which produced the novelties when inter-crossed. In sev- 

 eral instances, however, rats of aberrant colour have been 

 caught wild. In one case we obtained several new colours in 

 one shipment, among rats all caught in one locality. From the 

 genetic behaviour of this group it is probable that a cross with 

 some other species, with the house-rat or the tree- rat or Mus 

 concolor caused the production of the new colours. 



In these few examples from our work with rats, we saw the 

 origin of waltzing individuals in two separate series. In both 

 cases waltzing rats originated in the F2 of a cross between spe- 

 cies which were very closely related, in the last case pheno- 

 typically indistinguishable. This origin of waltzing in rats gives 

 an explanation of the way in which the same variation may 

 have originated in mice. Most authors who have studied the 

 anatomical, physiological and psychological peculiarities of 



