Page:Popular Science Monthly Volume 77.djvu/90

84 abruptly ended at term 3 or 4, as a result of fear, through discovery or disturbance of the nest, and a new series is promptly begun at 3. This is the simplest type of disturbance which we can record (Fig. 16).

The old nest may be torn down by the little builders, and its materials used again, but this does not commonly happen. Since fear is rapidly depressed, with the rise of the brooding instinct, beginning at term 5, interruptions are less liable to occur after this point is reached, but wherever the thread is dropped, it is usually picked up again at stage 3.

Of far greater interest is the fact that a new cycle may be begun at the very close of the breeding season, when it seldom goes far, and is bound to fail for lack of time. Probably no stronger witness to the instinctive basis of the behavior of birds could be found than this recrudescence of the reproductive activities at a time when most must answer the fatal summons of the migratory impulse. It is typically illustrated by the great herring gulls, which toward the close of their usual cycle in mid-July begin to build new nests, and will even lay eggs in them, though all are eventually abandoned. It would not be surprising to find that many young were also left to their fate, but my observations have never extended late enough to determine this definitely. At the Great Duck Islands, Maine, where these facts were gathered, the birds arrive early in March, and depart about September 1, according to the warden and lighthouse keeper, Captain Stanley, who has found that the first eggs are laid about the middle of May, while the first young begin to appear the second week in June.

In a census of one hundred nests of this gull taken on the island July 17, 1902, at the close of the breeding season, some interesting facts were brought out, which may be summarized as follows: