the wire mesh, than to miss the wire by passing through 

 the openings of the cartridge. 



The ordinary method of oeed spotting, 50 seeds 

 to the spot, 1/2" covering of soil, either firmed or loose, 

 gave as good results as any other method tried, but never- 

 theless suffered greatly through the mouse damage, 

 tfotes dm the Smelling Powers of ITice: 



In the last experiment seed was sown at depths 

 of 1-1/2" and 2-1/2" in some of the tests. Part of the seed 

 in the 1-1/2" sowing was dug up, while the surface soil on 

 one of the 2-1/2" spots was nosed. It is obvious, therefore, 

 that mice can smell seed even through 2-1/2" of soil. Since 

 deeper sowing than 1-1/2" would make the coming up of the 

 plants very uncertain, it is not possible to sow seed deep 

 enough to prevent their detection by rodents. 



In cage tests a box 5" x 6" x 1-1/2" was filled with 

 soil, seed sown at a depth of 1-1/4" in it, and iodoform or 

 napthalene spread over the surface. The following instances 

 show that mice detected the seed despite the strong smell of 

 the substance sprinkled over the surface. (Only the first 

 time the box was put in a given mouse's cage is considered, 

 since on repeating such a test there is a chance memory 

 rather than sense of smell rcould cause the mouse to dig.) 



-6- 



