chemical compounds, not of cells, of unicellular animals; and as the molecules of a crystal arrange themselves in an aggregate which has a particular shape—the shape of the crystal—so, he thinks, do the physiological units of which an animal is compounded, arrange themselves in an aggregate which has a particular shape—the shape of the animal. Differences in animal shapes are by him accounted for by supposing that there is a slight difference in their physiological units, which in the aggregate produces considerable differences in shape. Thus, a dog differs from a sheep because of a difference in physiological units. A son resembles his father because their physiological units are much alike; he differs somewhat from his father because of a slight, a very slight, difference in physiological units. Physiological units according to him are modifiable by all forces that impinge on the aggregate—the animal. The germ, like the rest of the animal, is compounded of physiological units, and, like the rest of the animal, its units are modifiable by the play of forces to which the aggregate is subjected. Therefore, when the aggregate is modified by forces impinging on it, the germ is likewise modified, wherefore, when, by the assimilation of food-molecules, which in it are built up into physiological units similar to those composing it, growth occurs, it develops or crystallizes into an organism, which reproduces the modifications of the parent organism. In this way is the transmissibility of acquired traits accounted for, and, in Mr. Spencer's opinion, to doubt that transmissibility is to doubt the persistence of force.
Mr. Spencer's arguments are marshalled with great skill, and the theory which he supports by them is highly philosophical; but there are fatal objections to it. It is impossible to admit that that form of force which causes, under fit conditions, the development of
