The sportsman out for quail or woodcock uses dogs to drive out the birds for him. Starlings and cowbirds about Chicago use the same principle in hunting grasshoppers. Instead of dogs they use cows, though of course the cows are intent on something else and presumably unconscious of the fact that they're helping the birds. As the cow grazes slowly across a meadow, it scares up grasshoppers close in front of it. The cowbirds and starlings take advantage of this. Instead of covering the meadow on foot, constantly alert for a sitting grasshopper, or to chase one they flush, the birds keep with a grazing cow. They take up a position by the head, or a foot, and catch the insects the cow disturbs. The cow is so much larger than the bird that it is likely to flush more insects. The grasshoppers on the wing are much easier to see than when at rest in the concealing grass, and some fly directly toward the bird. Too, the grasshoppers fleeing a cow are less likely to be alert to other dangers. CONFIRMED BY OBSERVATION The advantages of this to the bird are obvious. But we've just assumed they were, and we had no data on the relative efficiency of the two methods of hunting. A few years ago, however, while in El Salvador, I was able to get quantitative data proving that using a cow as a beater was advantageous, as we suspected, and showing how much more effective it was, something we did not know. The bird concerned was not the starling, which does not occur there, or a cowbird, which occurs but consorts little with cows, but was the grove-billed ani, a black cuckoo about twelve inches long of the tropics of Central and South America. Like our starling and our cowbird, it kept with cows, catching the grasshoppers and other insects that flew up. Both anis and cows were common in the grassy fields about our headquarters in San Salvador. We decided, my son Stanley and I, to watch anis with cows for a few hours, and then without cows for a few hours; thus getting the average rate for each type of feeding. We quickly found it wasn't as easy as that. Something always happened; even on the levelest and most open fields the birds were constantly disappearing behind a tuft of grass, or in a hollow, or, if nothing else, behind the cow's head or feet. Then, too, the ani we elected to watch wouldn't pay attention to the job in hand. It would wander off, or go to sleep. And sometimes, when we were about to discontinue watching a somnolent bird, it would snap up an insect. Perhaps it had been watching all the time. Finally we found we had to record observations of many short periods, of from three to fourteen minutes each, and add them together.