324 



Dr. J. H. Long on the 



at 50 cubic centims. per hour, and when suddenly changed to 

 80 cubic centims. gave results as in the Table. This anoma- 

 lous case will be explained when I come to speak of the influ- 

 ence of concentration. 



From the Tables given above it is seen that the rate of dif- 

 fusion is constant for 30 hours at least. To ascertain how 

 much longer this constancy might continue was the object of 

 the next experiments. Two tubes of the same size were filled, 

 one with n and the other with 2nNaCl solution. After 

 waiting until the rate of diffusion had become uniform, the 

 titrations were commenced with the following results. The 

 velocity of the flowing water was 40 cubic centims. per hour. 



Duration of experiment. 



4 P.M 



.- 8 A.M. 



= 



16 1 



IOU 



8 



-12 



= 



4 



?? 



12 



- 6 



= 



6 



n 



6 



-12 



= 



18 < 



;? 



12 



- 4 



= 



4 



?? 



4 



- 8 



= 



16 



v 



8 



-12 



= 



4 



)> 



12 



- 6 



= 



6 



» 



6 



- 8 



= 



14 



?? 



8 



- 4 



= 



8 



?? 



4 



-12 



= 



20 



)) 



12 



- 6 



= 



6 



V 



6 



-10 



= 



16 



)} 





Total 





138 



?? 



Cub. cent. ^AgN0 3 



solution 



required for 



r~ 



\ 



n sol. 



2n sol. 



12-8 



26-0 



3-7 



7-5 



6-1 



11-2 



17-7 



34-6 



3-5 



7-8 



15-8 



31-2 



3-8 



7-8 



5*9 



11-6 



14-3 



26-0 



8-2 



16-4 



20-5 



40-5 



6-5 



11-8 



16-0 



30-0 



134-8 c.c. 262-4 c.c. 



As the results for the last days are no smaller than those 

 for the first, it must be concluded that the duration of constant 

 diffusion is rather long, for NaCl at any rate. 



Indeed this seems to be true of the chlorides in general. In 

 the case of the weaker of the above solutions the glass con- 

 tained at the outset 14*5 grams salt, and the whole amount dif- 

 fused was only 0*2 gram in the 138 hours. It is probable that 

 the same rate would have been conserved for a much longer 

 period. 



The above Table also shows something regarding the effect 

 of concentration. As stated, one solution was twice as strong 

 as the other ; and the titrations show that, of the stronger, 

 nearly twice as much has diffused as of the weaker solution. 

 This corresponds with the observation of Weber on ZnS0 4 



