NESTING-HOUSES 37 



mixture would answer equally well. A small col- 

 ony resulted from this experiment, which would 

 probably have proved permanent but for the fact 

 that additional houses were put up in West Ches- 

 ter, and after two years the colony deserted the 

 old neighborhood. Though only temporary suc- 

 cess was achieved, the experiment is encouraging 

 and points the way to ultimate success. For the 

 above facts I am indebted to Mr. Charles J. Pen- 

 nock of Kennett Square, Pennsylvania. 



" As the young can be raised by hand without 

 serious difficulty, it may be easier to start the 

 new colony with nestlings alone, feeding them on 

 meal-worms, grasshoppers, and the like. Six or 

 eight pairs can well be spared from a strong col- 

 ony without unduly weakening it. This method 

 promises well, and if the experiment can be tried 

 from year to year, even on a small scale, a grad- 

 ual increase in the number of martin colonies is 

 likely to resiJt and new centres of distributions 

 to be formed." 



Martin-houses. — The martins prefer to nest 

 in colonies, so that houses should be provided 

 which contain a number of compartments. Each 

 dimension of a single room should be about six 

 or seven inches. The entrance hole may be either 

 square or circular, and should be about two and 



