munitics only one queen is permitted to develop, while the remaining females continue sterile, and become adapted to-working duties. Among the ants numerous queens develop, but each surviving queen usually becomes the mother of a separate community, in which the sterile females are adapted to two or more distinct duties. The problem of the males is a singular one. Among bees and ants they arc never checked at the worker stage, but develop to become a possible burden on the community. Here among the bees a second remarkable instance of intelligent selection is displayed. The males are suffered to live as long as food is abundant, but are mercilessly stung to death as soon as there is danger of lack of food. In ant communities natural selection disposes of the surplus males. Their life-power is reduced to that required for the nuptial flight, and they die as soon as their one necessary duty is performed.
Among the termites, or white ants, we find an interesting extension of this principle. Here restriction applies to both sexes, the workers and soldiers being immature males and females. Some writers, indeed, hold that they are of no sex, but have been checked in development at the larval stage, before sexual differentiation began. And a male as well as a female survives to start the new community, each nest having its so-called king and queen. In polyp colonies we find the same thing in a less developed stage. Each sexual individual is hermaphrodite, and the king and queen powers exist in a single form. In the Siphonophora, or floating hydrozoan colonies, the partly developed forms are adapted to four distinct duties. Some of them become contracting bells, and serve for locomotion; others become stomachal tubes, and digest the food of the colony; others are tentacles, or food-catchers; and others are simply covering or protective pieces; yet in all of them the Medusa type can occasionally be recognized.
It may be well to point out here that a similar division of duties exists in all the higher members of the vegetable kingdom. Each tree is a colony, the product of buds arising in a common stem, and is thus closely analogous to a polyp colony. The analogy goes further—there is a division of duties among the members of the tree colony. Some of these members attain full development and become hermaphrodite sexual individuals. The others are restricted in development, and become adapted to several distinct duties. Thus, two distinct nutritive forms appear, the leaf-bearing individual and the root individual. But greatly restricted protective forms occasionally appear, such as the thorn, whose development is on a level with that of the covering piece in a polyp colony. Other illustrations of this principle of restriction of development and division of duties might be given, but we must go on to consider its significance.
If we consider any of the lower animal forms, it will quickly appear that structural development is checked more or less completely during
