508 PROCEEDINGS OF THE AMERICAN ACADEMY. 



ratio — -, taken from a table of conductivities to the nearest integer for 



A, = .002 n, is as follows: CaCla = 110/115, NasHPOi = 100/126, 

 NaH2P04 = 84/126, KNO3 = 122/125, NaNOg =^ 101/104, KCl = 

 127/130, NaCI = 107/110, NasSO^ = 105/110, K^SO^ = 126,133. 

 This ratio has the approximate value of 1. P'or uuivalent salts, the 

 number of ions, A', into which a molecule of salts dissociates = 2 ; for 

 bivalent salts = 3, for the above phosphates =^ 4. Hence i = I + 

 (k — 1) a = 2, 3, 4. In other words, the osmotic concentrations of 

 those salts at .002 n. is equivalent to 2, 3, 4 times that of an undis- 

 sociated salt. Hence if it be desired to take any one of the physiological 

 salts to represent the osmotic value of a mixture of all of them, one 

 whose value for t = 3 is probably the best choice. Partly for this 

 reason and partly for the sake of good physiological adjustment I se- 

 lected calcic chloride. Its degree of dissociation is near the average of 

 that of all the salts in question, provided none of tliem is present in 

 excessively large proportion. A concentration of calcic chloride having 

 the same conductivity as that of the culture medium of the Stentors of 

 tills experiment may hence be regarded as approximately isotonic with 

 the medium. Conductivity determinations were made with the test-cell 

 previously described connected in parallel with the measuring tube W", 

 upon which the dial was used. The conductivity of the culture liquid 

 from which the Stentors were to be taken was then measured, that is, the 

 position of the electrodes in W", as indicated by the pointer on the dial, 

 was noted. It was not necessary to know the absolute value. Succes- 

 sive equal quantities of calcic chloride solutions of different concentrations 

 were then introduced into the test-cell and the position of the electrode 

 noted as before. After two calcic chloride solutions had been measured 

 in this way the concentration for the third trial was found by inter- 

 polation from the Urst two. It developed incidentally that under the 

 conditions of the apparatus then prevailing tlie sensitiveness of the 

 instrument was = .0001 n. calcic chloride solution. It was found that 

 a .0022 n., that is a .0011 tn. calcic chloride solution had the same con- 

 ductivity as the culture medium. Of course the same result could be 

 obtained with fewer measurements and more calculation, but since con- 

 ductivity measurements are made with certainty and rapidity, this method 

 seemed the better. 



Several hundred Stentors were introduced into 200 cc. of .0011 m. 

 calcic chloride solution. The liquid was contained in a narrow gradu- 

 ate and the animals settled upon the bottom. A glass siphon of narrow 



