The blood of tunicates presents a large variety of cell types, which are subject to considerable variation in different species and some in the same individual. These various cell types may be grouped in five categories, classified on the basis of their structure, the nature of their inclusions, and their probable functions: 1. Lymphocytes, comparable to vertebrate large lymphocytes. Transition stages connecting these with the specialized cell types show them to be the blood-mother cells, and investigations concerning budding and regeneration have shown that they are the blastogenic cells that differentiate into virtually all structures of an individual produced asexually. 2. Macrophages. In some cases these cells are quite large and have the cytoplasm crowded with ingested cells of other types. 3. Vacuolated cells with various sorts of metaplasmic inclusions. Some of these inclusions are histologically similar to the reserve food materials stored in eggs, and chemical evidence reveals the presence of lipoids, proteins, and carbohydrates. The presence of quantities of reserve nutritive material in blood-cells is a common phenomenon in invertebrates generally. One of the primitive functions of blood-cells appears to be to serve as storage places for and bearers of nutritive material. This primitive nutritive function of blood gives some meaning to the fact that in both phylogeny and ontogeny of the vertebrates the early blood-forming organs arise in relation with the food-bearing endoderm and that in later life substances of endodermal origin exercise a profound influence on blood formation. 4. Nephrocytes. Cells with vacuoles containing substances in colloidal suspension or in the form of larger granules or intracellular calculi, that have been identified as purine bases. 5. Cells without vacuoles, but with yellow, orange, red, or blue-green granules present in the cytoplasm. In some species the granules have been identified as carotinoids. There are features of blood histology that have a taxonomic significance. For example, a particular variety of vacuolated cell is characteristic of the Styelidae, and is present in at least some species of the closely related Pyuridae and Botryllidae. One particular type of cell has been found only in the Aseidiidae, and in all species of that family studied. Spindle-shaped orange granules have been found only in the genus Ecteinascidia, and here they are always present in the two species and a subspecies present in Bermuda waters; there are certain specific differences in the orange cells in each of these forms. Other specific differences in blood also occur. The colour and colour markings of ascidians are due primarily to the varieties of blood-cells present, their relative numbers and distribution, and the angle from which the illumination comes. It would seem that the varieties of cellular differentiation found in aseidian tissues might have something to tell us concerning the fundamental differentiation of protoplasm for particular types of function. The generalized, indifferent, pluripotent state of protoplasm appears to be associated with a homogeneous or finely granular condition. This is the condition of the cytoplasm in primitive germ-cells and in lymphoid cells, which serve as blood-stem cells and as blastic cells in budding and regeneration. Vacuolization of protoplasm, on the other hand, seems to be a structural feature associated with great metabolic activity. The vacuolated structure is a conspicuous feature in the egg follicle-cells not only of aseidians but of other animals during the period of the growth of the egg when the follicle-cells appear to be very active contributors to the growing egg, in many specialized blood-cells of aseidians and many other invertebrates, in the food-storing cells of many animals, in the cells of cartilage during the period of deposition of the matrix, &c. Such cells seem to be highly efficient units for chemical reactions. In the case of the various nutrition. bearing blood-cells of aseidians the substances taken into the vacuoles and acted upon by their enzymes are apparently transferred to other cells specialized for efficiency along different lines; in the case of the nephrocytes the absorbed products are stored indefinitely. May we not look upon the vacuolated structure of protoplasm as being a specialization for absorption, intracellular enzyme activity, storage, and transfer of materials just as we have come to look upon fibrillar differentiation as being a specialization for conduction, for contraction, or for tensile strength?