12 



suggested, and if when two salts are present the ratio of 

 absorption of both will be decreased as would seem to be 

 the case from I'orowitz* hypothesis, then it would seem that 

 a proportion of 8C molecules of calcium to 1 of aluminum 

 should decrease the toxic effect of A1(N03)3 at least to a 

 limited extent. 



The whole matter might be explained by assuming 

 that the protoplasm is made impermeable to the toxic salt 

 by the presence of the other salt in the medium ^as suggested 

 by Loeb in his later theory and by Cst^rhout from his vrork 



on Laminaria. Yet if this were the case it seems probable 



constant 

 that a/quantity of the salt of the lighter metal would be 



required irrespective of the concentration of the toxic 

 substance. This, howevey,is obviously not the case in the 

 experiments described above. In the combination of Pb(NG3)2 

 with Ca(NC3)2 the amount of the calcium salt required to 

 bring about a given response in the presence of the lead 

 salt varies with the concentration of the latter. In this 

 case the concentrations of the two salts are always in con- 

 stant molecular ratio to each other. A similar state of 



obtains 

 affairs/when the nitrate of magnesium is used instead of 



that of calcium, although the molecular ratio is here 

 somewhat different. Again in the combinations of CuCNCsJg 

 with Ca(KC3)2 here employed the amount of th- calcium 

 needed for a given response varies with the concentration 

 of the copper salt. This combination differs from that 

 with lead nitrate however for here the molecular ratio of 



