A step higher in the animal scale, or a further advance in the development of the schematic embryo (depicted in Diagrams 3 to 6), and we find that these special digestive cells are losing their sturdier qualities and being placed in a position protected by cells which have specialized in another direction. This is shown in Diagram 4, where the hollow ball of cells which resulted from the repeated division of one cell is represented in section. One side of the ball is pushed in, and now the beast consists of two layers of cells, an outer protecting and an inner digesting (Hydra and sea-anemone). Soon, however, it is found more convenient to have a tube for digesting food, for then different substances can be digested and absorbed in different parts; and the refuse, of which the animal can make no use, need not be brought back to the mouth to be got rid of. This, however, requires a number of other changes in the structure of the animal, which are roughly shown in Diagrams 5 and 6. It is not to our purpose here to discuss the development of animals or an animal; but the figures are worth glancing at, as they show not only how certain of the cells are set apart for digesting food, but also that a large body consists really only of a mass of protoplasm, composing kindred cells of common origin. [Illustration: DIAGRAM 6.—SHOWING DEVELOPMENT OF AN EMBRYO.] Now, for obvious reasons, the longer, within certain limits, this tube is the better. All sorts of different food-stuffs have to be acted upon in it, and some offer considerable resistance to digestion; and the further they have to travel in the tube, the more chance there is of their being successfully treated. Besides, different parts have different functions, and the longer the tube—again within necessary limits—the greater scope is there for division of labour, and consequent economy. The comparative length of the alimentary canal is not the same in all animals by any means. Carnivorous animals, like the cat, whose food is soft and easily digested, have a fairly short one. Herbivora, like the sheep, whose food is difficult to digest and mixed with much husk, which is wholly indigestible, have a comparatively very long one. Man, who is omnivorous, but eats less and more judiciously chosen food than either of the above classes, has one of medium length. But in all cases among the higher animals there is an attempt made to obviate the necessity of increasing the length of the animal by coiling the tube within the body. The annexed diagram (7) illustrates this principle. It shows a schematic animal whose digestive canal is much longer than itself. [Illustration: DIAGRAM 7.—SHOWING HOW THE DIGESTIVE CANAL IS LENGTHENED.] [Illustration: DIAGRAM 8.—CROSS-SECTION OF THE DIGESTIVE TUBE.]