1. A study of the distribution of the granules in rat heart-muscle fibres confirmed the early morphological findings. We prefer to describe these granules under the name of sarcosomes, first introduced by Retzius.
2. In sections, heart sarcosomes are seen as rods in close association with the anisotropic disks of the myofibrils. The rod-shape is believed to be due to compression of the sarcosomes between adjacent myofibrils. Isolated sarcosomes which retain the elasticity of those in the living state would be expected to be spherical.
3. Preparations of sarcosomes isolated with hypertonic sucrose contained many rods. Spheres were obtained with isotonic saline or sucrose. Water, ethylene glycol, or glycerol gave greatly swollen and otherwise altered sarcosomes.
4. In hypotonic media, sarcosomes underwent a series of partially reversible morphological changes termed ‘transformation’. These changes, which are caused by penetration of water into the sarcosomes, are described in detail.
5. Studies of ‘transformation’ gave clear evidence that the intact sarcosome is composed of a central body in the form of a gel surrounded by a membrane. As the tonicity of the medium was decreased, the body progressively disappeared, leaving eventually a large vesicle surrounded by the membrane. In this condition the membrane readily breaks up into particles which contain all the enzymes which constitute the respiratory chain, but is devoid of most dehydrogenases and phosphorylating enzymes.
6. On standing in isotonic media the membrane is rapidly damaged, causing penetration of solutes so that the sarcosomes transform in much the same way as in hypotonic media. This ‘spontaneous transformation’ was greatly delayed by the calcium-chelating agents versene or citrate and by adenosine diphosphate or triphos-phate, and was accelerated by calcium.