Such changes in pressure may not be of any great importance, but they can easily be avoided by substituting a horizontal coil of polythene tubing, of not more than 5 mm diameter, for the vessel outside the body. Movement of fluid into or out from the viscus then makes no alteration in the pressure. With this diameter of tubing the interface between air and fluid remains practically vertical and the column of fluid does not break up. A coil large enough to allow changes in volume of the order of 60 ml., is more than adequate to record movements of most hollow organs and is reasonably compact. The friction in the tubing seems to be of little moment and the system as a whole records delicately even small movements with little force behind them. During the course of investigation into pre-eclamptic toxaemias of pregnancy it was desired to ascertain whether the urinary excretion of chorionic gonadotrophin (CG) was increased during the last trimester above the normal range d PROCEEDINGS OF THE PHYSIOLOGICAL of 4000-11,000 i.u./24 hr as given by Loraine & Matthew (1950), and if so, to form an estimate of the amount of CG excreted.The assay method to be used was the spermiation test of male toads Wohlzogen, 1953). Toads respond to the injection of CG with the expulsion of spermatozoa in their urine, this response being quantal in character. The dose-response curve is rectilinear if the probit of the percentage reactors is plotted against the log. dose. The slope of the regression line did not vary significantly over a period of a year, and a significant shift of the line (i.e. alterations in the sensitivity of the toads towards CG) was only encountered during time intervals of more than 2 months (Wohlzogen, 1954).It has been shown that the problem of testing any material for its CG content to be above or below a given level is most efficiently solved by a method of sequential sampling (Bukovics & Wohlzogen, 1953, 1954. For this purpose the material to be tested is assumed to contain an arbitrary concentration of CG which, in the present case, is taken as the upper limit of the normal range of urinary excretion. The hypothesis that the urine contains the actual CG concentration assumed or a lower one, is then tested against the single alternative hypothesis that its concentration is higher. Depending on the assumed concentration, the amount to be injected into each toad is adjusted to contain the median effective dose (ED 50), so that the animals tested may be considered to come from a population with 50 % or less positives as against a population containing 50 % or more positives. In the present instance a grouping technique was adopted; an initial group of six, followed by subsequent groups of three toads, were injected and an answer was obtained with between six and fifteen animals. The accuracy is 80-125 % with a probability of committing errors of the first or second kind of P = 0-033. This sampling plan is summarized in Table 1. The test is terminated, as soon as the number Fig. 1 segment of ileum from a large rab...