No. 607] 



EATS AXD EVOLUTION 



401 



In Solo and Djocjacarta in Java, the great tobacco- 

 growing companies erect enormously big sheds con- 

 structed of a very complicated scaffolding of heavy bam- 

 boo, with a thick thatched roof made of palm leaves. 

 Such sheds are erected in the midst of the fields, mostly 

 far from native villages. The native laborers leave food 

 about the structure, so that it very soon becomes inhabited 

 by rats. Now the rat population of these drying sheds is 

 always composed of house-rats ; field-rats are too timid to 

 live permanently in places where human beings move 

 about so much. But as the sheds are built in isolated 

 places, they do not get their house-rat population as such 

 from neighboring houses. We are convinced that into the 

 composition of such rat populations, field-rats, and 

 hybrids between field-rats and house-rats enter to some 

 extent. This is the explanation of the fact, that very 

 often the rat population of such a shed is found to be 

 composed of an aberrant type of rats. If the population 

 of such a shed becomes very numerous and a native vil- 

 lage of some sort springs up in the immediate neighbor- 

 hood, the aberrant new type may have a swamping influ- 

 ence upon a minority of t}T>ical house-rats brought along 

 by the natives, so that the type may become locally com- 

 mon, and temporarily supersede the ordinary house-rat. 



We remember Major Ouwens showing us great num- 

 bers of white-bellied house-rats, received from a tobacco- 

 growing firm in one of the big centers, Klaten. 



When there exists in a certain locality an abundant pop- 

 ulation of rats of a certain species, immigration of a few 

 rats belonging to the same group but to a different species 

 will have no effect. And of course it will make no dif- 

 ference whether the multitude belongs to the common 

 species and tlie few immigrants to an aberrant new type, 

 or rovcrsely. as in the case of the rats in the tobacco- 

 sheds. 



Ships may .xvasioimlly ])riim- rats to .Java, from Eng- 

 lish India, or tVorn Aii-trMli;i nv Sinu;ii>nrr. hut the rat 

 fauna of Java will not bo fui-iduHl Ly a nvw species, as 



