In the course ofexperiments in which acetylcholine (ACh) released during nerve stimulation was collected from the isolated rat diaphragm (Krnjevid & Mitchell, 1960), it became a matter of some importance to know how ACh is distributed within the diaphragm and at what rate it can be expected to diffuse out. The following experiments were therefore undertaken, and, as no information was available about the rate of diffusion of ACh in physiological solutions, estimations were also made of the diffusion coefficient of ACh in gels of Ringer-Locke solution and agar.
METHODS
Diffu8ion of ACh chloride in agar gelsThe method used was derived from one described by Eggleton, Eggleton & Hill (1928) in which measurements were made of the amounts of substances diffusing out of an agar gel. In our first experiments the solution of ACh chloride and agar was poured into a boilingtube (cross-section about 4 cm') and allowed to set, and the rate of diffusion of ACh into physiological saline determined exactly as in the experiments of Eggleton et al. (1928). The apparent coefficients of diffusion obtained in this way were unexpectedly high, and, for control, measurements of the diffusion of NaCl into water were made under the same conditions. These also showed very high apparent rates of diffusion (as much as 50 % higher than the values given in the International Critical Tables). Inspection of the dirfusion tubes showed that this anomalous diffusion could be ascribed to a substantially greater (+ 10-20 %) area of diffusion at the curved surface of the gel than would be expected from the area of cross-section of the tube. The method was therefore modified as follows.'Bacto-agar' (Difco Laboratories, Detroit, Michigan) was dissolved in hot Ringer-Locke solution so as to make a 0-5 % (w/v) solution. It was allowed to cool to about 400 C, at which temperature it remained liquid. ACh chloride was then added from a standard solution to give a final concentration in the agar solution of 5-0 x 10-6M, the standard solution (5 x 10-3M) being prepared from a known weight of acetylcholine chloride (Roche Products) dissolved in a solution of 0-lM-NaH2PO4. The Ringer-Locke solution had the following composition (mM): Na+ 150, K+ 5.0, CaO+ 2-0, Mg2+ 1-0, Cl-148, H2POj-1-0, HCO,-12-0 and glucose 11, the ionic strength being 0-164. Diffu8ion chamber. This consisted of two Perspex tubes. One, with an internal diameter of about 3 cm and length of about 5 cm, was closed at one end; the other (about 4 cm long) was open at both ends and fitted accurately over the open end of the first. Some 25 ml. of the warm solution of ACh and agar was poured carefullv into the bottom tube so as to give