BRITISH WARBLERS
the reeds, or rather a territory including reeds, are not available. When the opportunities of building in the willows fringing the reeds, or in the reeds themselves, are equal, the latter situation seems to be chosen in preference. I have examined a number of nests for the purpose of discovering whether, in their construction, new reeds were made use of more frequently than old, and whether both new and old were utilised at the same time; for it occurred to me that, if the latter were the case, the growth of the new reeds would cause the nest to become tilted on one side, and bring about a calamity by precipitating the young into the water. All three methods, I find, are followed, although of the nests examined only one was built upon both old and new reeds; this, however, must not be regarded as a correct proportion, the number examined being much too small to enable such a proportion to be correctly estimated. As a rule the nest is attached somewhat loosely to the stems, and it might thus be able to slip and automatically adjust itself to the growth; but there are budding leaves on the new stems, and an uneven surface and joints on the old, which might at any moment prevent its slipping and result in a gradual tilting of the nest to one side. Therefore I am inclined to think that the danger is a real one, but one that is kept in check by selection, and it would in no way surprise me to find that the number of individuals that constructed their nests after this fashion was but a small proportion of the whole. Whether old reeds are used more frequently than new is difficult to determine, since in many places the natural conditions have been altered, the reeds being cut in the winter months and the birds being consequently forced to wait until the new ones have made sufficient growth before commencing to breed. Neither is it possible to say, or even to guess, what the original nest-building instinct of the species was, whether, that is to say, the species as a whole was limited to the use of new or of old reeds, or whether some individuals used the one and some the other. The latter alternative is not impossible, and there may be two races, so
42
