In this controversy, however, there was another string to the Lamarckian bow. Worker-ants, bees, etc. are imperfectly developed females. They have four or five egg-tubes where the queen has two hundred, but they cannot be fertilized by the drones. It occasionally happens, however, that these neuter insects do lay a few eggs. These unfertilized eggs always develop into drones. One of these drones might, it was suggested, now and then fertilize a genuine queen, and thus hand on the traits of the worker from which it sprang. But apart from the fact that an occasional occurrence of this sort would hardly suffice to maintain the worker-characteristics unimpaired throughout the ages, there is the decisive answer, as Weismann points out, that we know at least one species of ant in which the evolution of a neuter caste is absolutely complete, for the workers of _Tetramorium caespitum_ possess no egg-tubes at all. Yet the transmission of characteristics from queens and drones who never exercise them to workers who cannot pass them on, goes forward in this species of any ant just as in any other. Nature, therefore, while doing in the case of these insect communities exactly what she appears to be doing elsewhere by the accumulation of acquired characteristics, must, in reality, have been working on entirely different lines. If we can discover what those lines were, they will cover the apparently Lamarckian cases as well, but the Lamarckian principle certainly will not cover these. In the next chapter we shall review the alternative explanation offered by Darwinism, the explanation of Weismann; and we shall see whether Spencer was not as successful in demolishing it as Weismann was in showing that, if evolution exists at all, some other basis must be found for it than that on which it was so largely rested by Herbert Spencer. CHAPTER V THE MECHANICAL THEORY OF EVOLUTION: THE DARWIN-WEISMANN EXPLANATION “Chance guides all things: mind and forethought must call it God alone!”—Menander. “In the end,” writes M. Edmond Perrier, “every imaginable theory of evolution must lead up to one or other of two absolute doctrines, essentially antagonistic to each other. Either the inheritance of acquired characteristics must be admitted in its full scope (_dans toute sa généralité_), or else we must believe in the predestination of protoplasm, developing by virtue of its own internal forces. But in the latter case we pass from the domain of pure science to enter that of metaphysics.”[66] We have now to consider the most conspicuous attempt made in recent times to escape from this tragic dilemma.