May 8, 1903.] 



SCIENCE. 



731 



trivalent in their stimulating capacity ap- 

 proximately as l:2 + :3.5, and not as 

 Hardy found in colloidal solutions pro- 

 portional to a power of the valence. There 

 are variations from this rule, some mono- 

 valent anions, i. e., hydroxyl, being nearly 

 as powerful as divalent. The formate, 

 however, is not an exception to the rule in 

 the nerve, whatever it may be in muscle. 

 It is, however, somewhat stronger than the 

 chloride. Such variations are in no way 

 antagonistic to the general conclusion that 

 it is the charge which stimulates, and at- 

 tention was called to them in my original 

 paper. They probably mean that the num- 

 ber of charges is not the sole factor, but 

 possibly, as already suggested, it is rather 

 the motion of the charge around or with 

 the atom or the affinity of the charge for 

 the atom. 



3. The general rule that the inhibiting 

 action of the cathion is proportional to its 

 valence or its electrical charges holds true, 

 but here even more than in the anions 

 there are exceptions, monovalent ions some- 

 times being stronger than one half of a bi- 

 valent ion. These exceptions need further 

 study and do not invalidate the general 

 conclusions, for, so far as I have examined, 

 the only bivalent cathion which is weaker 

 than two monovalent anions is barium. 



4. That barium chloride stimulates by 

 means of the anion is shown also by the fact 

 that its stimulating action may be neutral- 

 ized by any of the agents used to neutral- 

 ize the stimulating action of sodium 

 chloride, i. e., by the addition of small 

 amounts of CaCla, KCl, LiCl, NH^Cl and 

 probably other salts having predominant 

 positive ions. More of these salts are re- 

 quired than are required to neutralize 

 NaCl, which agrees exactly with the theory. 

 These facts were predicted, and experi- 

 ments confirmed the prediction even to the 

 amounts of salts necessary to add. Fur- 



thermore, barium chloride, like sodium 

 chloride or other sodium salts, places the 

 end of the nerve in a condition of catelec- 

 trotonus, so that if the end of the nerve is 

 cut off after immersion in any of these 

 salts, the muscle goes into a tetanus and 

 may remain in a tetanic contraction for 

 many minutes, in some eases even half an 

 hour. This tetanus corresponds closely to 

 the tetanus observed on cutting the nerve 

 between the electrodes with the anode near 

 the muscle during the passage of the cur- 

 rent, and is, I believe, due to the same 

 cause. This similarity of action between 

 sodium and barium salts shows them to act 

 in the same way in the nerve, but the 

 barium salts somewhat more strongly. The 

 fact that barium chloride may be neutral- 

 ized in its poisonous and stimulating action 

 by calcium chloride is difficult to reconcile 

 with the hypothesis that antitoxic action 

 occurs between monovalent and polyvalent 

 positive ions. 



5. If sodium chloride stimulates by the 

 anion, as electrical stimulation clearly in- 

 dicates, it should be possible to neutralize 

 its stimulating action by adding small 

 amounts of any salt of which the positive 

 ion preponderates, but not by any salt of 

 Avhich the negative ion preponderates. This 

 is the case: The stimulating action of 

 sodium chloride may be neutralized by 

 small amounts of the chlorides of lithium, 

 potassium, hydrogen, ammonium, alumi- 

 num, calcium, strontium, zinc, cobalt, man- 

 ganese or magnesium. The amount of salt 

 necessary to neutralize varied in different 

 cases, more lithium being necessary than 

 potassium. The salts were found to range 

 themselves in this action in the same order 

 of efficiency as previous experiment had 

 shown them to act as depressors of ir- 

 ritability. The order was predicted from 

 the theory and confirmed by experiment. 

 The exact figures will be given in the full 



