examining the food in thousands, old and young; old ones from all sorts of places at all times of year, young ones from wherever I could get any all through the breeding season. The result, confirmed by occasional examinations up to the present time, was that I found that sparrows destroy even fewer insects than I had supposed. The food in the old ones was almost all corn during the whole year; green peas (of course bitten up small) were also found in them in summer; and in May and June, when corn is scarce, a few wild seeds, chiefly of grass. No insect has been found by me in a sparrow between September and March. I have not often found one at any season (particularly between June and March) in a sparrow old enough to feed itself, and have very seldom found any number of insects in one—even when corn could scarcely be got. The food of young sparrows was found to vary greatly; of those taken at the same time from one farmyard, some broods contained insects, some corn, green or ripe, or green peas, and a few green seeds, chiefly of grass, and in many would be a mixture of some or all these. The kind of food by no means always depends on the age of the birds; the first food after being hatched given to a young sparrow is commonly a small green caterpillar, but large callow ones a few days old are often full of ripe wheat, and some ready to fly contain insects chiefly. My observations showed plainly that to get results of any value an investigation of sparrows' food must be made on a large scale; otherwise very erroneous deductions may be made. For instance, once in June I found in forty-seven nestling sparrows of all ages from one farmyard scarcely anything except old wheat and green peas; there were only about six insects in the whole number. From such an instance it might
