more dilute solution into the blood. In a large number of experiments which have yielded quite constant results I have been unable to confirm the idea generally held and supported by Hay that subcutaneous and intravenous injections of saline purgatives do not exert a purgative action. I have quite constantly found that with proper conditions these salts do produce increased peristalsis and also an increased secretion of fluid into the intestine when introduced directly into the circulation or under the skin. In most cases also an actual passage of faeces was observed. Several methods were used in testing the action of the salts and it was found that the results could best be studied by observing the loops of intestine directly. In rabbits under the influence of morphine (5 c.c. 1% solution morphine hydrochlorate subcutaneously) the abdominal cavity was laid open and the loops of intestine carefully protected from loss of heat and moisture while under observation. In other cases they were observed under the surface of normal sodium chloride solution, a method devised by van Braam-Houckgeest and used by him in a long series of experiments. In the NaCl solution the intestines normally remain almost quiet and the effect on them of stimulating agents can be readily determined. In addition to this method many experiments were made in which animals were kept in separate cages, some acting as control animals, others for experiment. The amount and character of the faeces were observed during several hours after the administration of the purgative and compared with the control animals. The solutions used were made up as fractions of molecular solutions. Thus to obtain a solution of sodium citrate, for example, approximately isosmotic with the rabbit’s blood, one-sixth molecular weight of the salt in grams (including water of crystallization) was dissolved in 1,000 c.c. distilled water. This gave an m/6 solution, which was taken to be approximately isotonic with the blood. The injections were made usually with a hypodermic needle into the marginal vein of the rabbit’s ear, or into the jugular vein. Subcutaneous injections were made under the loose skin of the back. In most of these experiments small rabbits weighing from 1,200 to 1,500 grams were used.