Incidental reference has been made to the great lymph-spaces—peritoneal, pleural, and pericardial. The brain and spinal cord are separated from their outer membranes by a lymph-space. There are also spaces within the brain—the ventricles—and a central canal in the spinal cord. The aqueous and vitreous humours of the eye are also lymph-spaces, although the latter contains some remnants of tissue. The joint cavities are lymph-spaces. So also are the bursæ which surround tendons or separate them from bones. It is not, however, justifiable to include all these cavities in a single category, either from the point of view of their purpose, their mode of formation, or the nature of their contents. The peritoneal, pleural, and pericardial spaces are parts of the great primitive body-cavity, or cœlom. The two first are potential rather than actual. Normally they contain just sufficient fluid to moisten the apposed surfaces of the endothelium which lines their walls and covers the organs which they contain. There is no fluid in them which can be collected and labelled “peritoneal” or “pleural” fluid. The purpose of the spaces is to allow of movement without friction—in the one case of the intestines, in the other of the lungs. It is possible to take a spoonful or so of fluid out of the space which surrounds the heart. It has the usual composition of lymph. It contains proteins, but is not spontaneously coagulable. Leucocytes are absent, a fact which probably accounts for its not clotting. The fluid inside the cerebro-spinal system is extremely dilute. Its principal salt—its principal constituent, indeed—is sodic chloride. It contains hardly a trace of proteins, and these in a modified condition—proteoses. It also contains pyro-catechin, a benzoic alcohol. This substance has long been recognized as a constituent of cerebro-spinal fluid, owing to the fact that, like sugar, it reduces copper salts when heated with them in an alkaline solution. It appears to be one of the products of proteid decomposition. Although exuded as lymph from the bloodvessels of the chorioid plexuses, the composition of cerebro-spinal fluid has been profoundly changed by the activity—it might almost be called the digestive activity—of the epithelium which lines the cerebro-spinal canal. There is a theory that the ancestors of all vertebrate animals were organized on a very different plan from that of their distant descendants. Our cerebro-spinal canal was their stomach and intestine. It would appear that the lining epithelium of these organs, although disused for millions of years, cannot resist the temptation to digest the lymph which they contain! The fluid in joints contains mucin (the essential constituent of mucus), or a substance resembling mucin. In this case the joint-membrane has added something to lymph without removing or destroying any of its other constituents.