Species crosses in Rats. Ill 



males and several females. This was often done to continue strains of 

 particular colours, which were in danger of dyinj^: out. ft is obvious, 

 that in such colonies the exact breeding of the young produced is 

 unknown. Results of such matings have for this reason not been 

 given in this papei-. In Bussum, Holland, our colony of rats of this 

 series generally filled thirty to thirtyfive cages and comprised between 

 150 to tiOO animals. 



In Huitenzorg, .lava, the Department of Agriculture constructed a 

 rathouse according to our specifications, with a series of ratproof rooms 

 in which the rats could be observed without knowing it, and a numbei- 

 of masonry tanks. The majority of our rats were kept in this building 

 for about a year, and this was the only time in which the cost of our 

 experiments was not entirely defrayed by us. In Java we started the 

 fieldrat series, the work on which was mainly done in Berkeley, Cal., 

 where a fairsized colony of around a hundred rats was kept till the 

 end of the experiments. 



Crosses between other species. 



All through the eight years in which we w^ere breeding the rats, 

 we kept trying to produce hybrids between species less closely related 

 than the houserat and the treerat. The tame laboratory rats, which 

 I'eadily interbreed with Miis nurvegicv!<, and must have been derived 

 partly from this species, have been tried in crosses with our houserat- 

 treerat animals and later with fieldrats. We observed, that when male 

 houserats were removed from a cage containing females of their species, 

 with which they had been mating, and put with adult laboratory females, 

 they often copulated with these. Although we observed these cross- 

 matings several times, we never obtained any young. In Berkeley we 

 observed the mating of a fieldrat male and a laboratory female on at 

 least three occasions, with equally negative results. 



In Java, crossmatings between houserats and treerats resulted in 

 one litter of four whitebellied agouti young, who escaped before they 

 reached breeding age. As the genotype of the Javanese treerat must 

 be different from the North African one used by Bonhote (see results 

 of crossing it to yellow), we expected more novelties from this cross. 



Several mixed pairs of one houserat and one fieldrat were kept 

 in the tanks of the Buitenzorg rathouse for long periods, but no young 

 were ever produced. This is not very surprising if we remember, that 

 houserats only very rarely bred in captivity during our stay there, 



