Muscle when most highly developed has an extraordinarily definite structure. It is minutely subdivided into units which appear, looked at separately, simple in design. We are tempted to believe that the explanation of the way in which each of these units works is not far to seek. It is disappointing to be obliged to admit that, notwithstanding all the thought which has been devoted to the problem, we are as far as ever from a definitive solution. We understand the principles on which steam-engines, combustion-engines, electric motors are planned. We compare muscle with each of these mechanical contrivances in turn, expecting to discover the principle of its construction. Many ingenious hypotheses have been formulated; but the fact that some of these are mutually destructive shows clearly enough that as yet no approach to certainty has been made. Probably the fundamental error lies in attempting to compare muscle with a mechanical contrivance. The apparent simplicity and regularity of structure of “striped muscle” misleads us. We ought to have commenced our investigations at the other end of the scale of mobile tissue—to have begun with semifluid and apparently homogeneous animal matter, working upwards to the tissue which, being limited to the one function of movement, and movement in one direction only, has, as it were, crystallized along the lines of force. All protoplasm is mobile. Its particles move one on another. Hence follows either circulation of the living matter within the cell or change in shape of the cell. The two phenomena are identical in nature. Circulation is best studied in a large-celled, transparent part of a plant. A filamentous water-weed is suitable for the purpose. If this be examined with a microscope while still alive, its cells are seen to contain a watery juice enclosed in spaces of denser cell-substance. Bridges of cell-substance span the spaces. The particles of which these bridges consist are in a state of constant streaming motion, which has, it is needless to say, no effect upon the shape of the cell (_cf._ p. 9). The unicellular animal amœba, leucocytes, and certain spores of plants, are devoid of cell-wall (_cf._ p. 28). Their soft protoplasm is not limited by a rigid case. When it streams, the form of the cell is changed. True, we must not think of the body-substance of an amœba as homogeneous. It exhibits an internal structure. Yet its architecture is not, so far as we can see, sufficiently fixed to restrict the directions in which it can stream. Any change of shape is possible. We cannot find in Nature an isolated clump of living protoplasm; nor do we suppose that, if we found it, it would prove to be homogeneous. It appears to be necessary that protoplasm and metaplasm—the terms have no chemical significance; “primary” and “secondary,” or “chief” and “subsidiary” would be equally distinctive—should be intermixed. Streaming is apparently due to alterations in the surface relations of the two substances.