Page:Popular Science Monthly Volume 58.djvu/259

Rh standpoint, but from a desire to learn the principal source from which our houses are supplied with this eternal nuisance, with a view to being able to suggest remedial measures. Experimental work in this direction was continued for some years. In the course of this work he early decided that an overwhelming majority of the house-flies found in domiciles breed in horse manure. This substance is its favored larval food, and experimental work showed that by the semi-weekly treatment of the horse manure in one large stable, the house-fly supply of the neighborhood was very greatly reduced. In confined breeding cages he had been unable to breed house-flies in any other substance than horse dung, and consequently when the camp typhoid question and the agency of flies became a matter of such general comment in 1898, he saw the desirability of a careful study of the insects which frequent or breed in human excrement, in order to give exact data from which reliable statements

could be made and upon which reliable conclusions could be based. This work was begun and carried on through the summer of 1899 and to some extent in the summer of 1900, with results which will be briefly summarized in the following paragraphs. The exact details, somewhat too technical, altogether too long and certainly too unpleasant for publication in a journal of this character, will be published in the Proceedings of the Washington Academy of Sciences.

In all seventy-seven distinct species of flies, belonging to twenty-one different families, were found by actual observation, either by rearing or by captures, to be coprophagous; thirty-six species were found to breed in human fæces under more or less normal conditions, while forty-one were captured upon such material. All have been studied with more or less care, and their other habits ascertained. The most