Tiny fragments from the cytoplasm of human skin fibroblasts with about 2% of the original cell volume ("microplasts") were prepared by treatment with cytochalasin B, vigorous pipetting, and trypsinization of the attached fragments. They remained alive for 8 hr or longer. Some of the micro lasts were able to produce and move filopodia, ruffle, or boththers blebbed continuously. Slow flattening was observed in the larger microplasts. In all cases tested, microplasts avoided contact with other cells or microplasts. The observations suggest that the cytoplasmic matrix and the membranes of animal cells are so constructed as to express locally and autonomously any one of the elementary amoeboid movements listed above. More importantly, whatever types of motile surface projections a microplast expressed, it continued to produce and move them in a stereotypical way as if there were long-lived structural or material determinants for each type. The microplasts were unable to locomote autonomously. Therefore, it is conceivable that directional movement of whole cells may require a supervising mechanism that confers a certain coordination and strategy on its component cytoplasmic bits. Otherwise they would continue to move in stereotypical and autonomous ways without ever displacing themselves, as suggested by the behavior of the microplasts. Inasmuch as sarcomeres are the structural units of muscle cell contraction, are there also structural units of amoeboid motion of nonmuscle cells? (In this context, "amoeboid motion" is understood to mean animal cell movements involving complex body deformations, as opposed to longitudinal contraction of muscle cells or the use of cilia and flagella.) It seems logical to begin the search for such units in the smallest cytoplasmic fragments-one may call them "microplasts"-that can still display autonomously the entire spectrum of amoeboid motion.Important steps toward the investigation of autonomous cytoplasmic movements have already been taken. Allen et al.(1) showed that the cytoplasm of the giant amoeba (Chaos chaos) can be forced into glass capillaries, where it continues to stream. Kojima (2) demonstrated that enucleated halves of sea urchin eggs can be induced to cleave. Several years ago, Goldman et al. (3) showed that enucleated animal cells (cytoplasts) generated and moved surface projections and also locomoted like whole cells. Even smaller motile cytoplasts of leukocytes were produced by Keller and Bessis (4). In response to temperatures of 460C, the cells segregated into nucleoplasts and cytoplasts, which continued to migrate, to phagocytose, and to accumulate around chemotactic targets.Cytoplasts can be fragmented further. Goldstein et al. (5) used fine glass needles to cut sizable pieces away from human cells and obtained motile fragments. Much smaller motile fragments were described by Shaw and Bray (6) and later by Wessels et al. (7), who found that the similarly amputated distal fragments of axons were able to rebuild a ruffling growth cone.The smallest motile ...