PHILIP H. MITCHELL AND J. WALTER WILSON 47 



marked loss of irritability in the muscle substance. Our experiments 

 though not including any quantitative measurements of irritability 

 have shown to rough observation a progressive loss of excitability 

 accompanying potassium depletion. This is to be expected from the 

 well known contrast between the physiological effects of a potassium- 

 free physiological saline and those of Ringer solution. 



We have further undertaken to test the possibility that muscular 

 contraction is favorable to the process by which potassium is absorbed. 

 We have found that only a contracting muscle, in sharp contrast to 

 a resting one, can take up rubidium and cesium, substances whose 

 chemical properties are more like those of potassium than sodium, 

 so as to retain them in the same sense that potassium is retained. 



EXPERIMENTAL. 



Large bull frogs weighing from 125 to 400 gm. were used. Some 

 of them were collected in Louisiana in February, shipped in March 

 and used for the experiments in April, May and June. Others were 

 taken in Rhode Island or Massachusetts in summer and used shortly 

 after collection. Perfusions were made through a glass canula in the 

 dorsal aorta of the pithed animals. The muscles used for analysis 

 were lightly drained on paper after removal from the animal and 

 immediately weighed. After digesting the muscle in a mixture of 

 nitric and sulphuric acids potassium determination was made by the 

 method described by Clausen (7) for blood analysis. Muscles removed 

 from frogs without experimentation, merely killing and bleeding the 

 animal, gave results as shown in Table I. 



The averages of these results, whether computed for the same 

 muscles of different frogs or for the various muscles of the same frog, 

 are fairly uniform and are in agreement with averages obtained by 

 Fahr (8) and others, placing the normal content of potassium in frog 

 muscle at 0.34 per cent. The considerable variation in the results 

 is partly due to errors in sampling. The sartorius is too small in 

 many frogs for a satisfactory analysis. The vastus cannot be removed 

 intact in its sheath so as to yield clean cut prisms of muscle but is 

 apt to suffer loss of some of its fluids during sampling. For this reason 

 most of the experiments reported below were done with the gastroc- 

 nemius muscles which can be more satisfactorily sampled. Another 



