THE EXCHANGE OF IONS 
In the group having a total salt concentration of 360 N X io~^ 
the introductory period during which the solutions gained ions was 
strongly marked in all solutions except that containing 2/4 Ca(N03)2 
+ 2/4 Mg(N03)2. In that solution net absorption became evident 
more quickly than in the other members of the group and the lead 
was held throughout the experiment. The net change in concentra- 
tion of the solution containing 4/4 Ca(N03)2 amounted to over 150 
N X io~^ a change exceeded only in the most favorable mixture, that 
containing 2/4 Ca(N03)2 + 2/4 Mg(N03)2. In this mixture the 
absorption corresponded to a change in concentration of about 200 
N X io~^ by the 17th day. The absorption from mixtures containing 
smaller proportions of Ca(N03)2 was less than that from 4/4 Ca(N03)2 
solution. The Mg(N03)2 solution showed the same features as 
heretofore. 
In the group having a total salt concentration of 480 N X io~^ 
it will be noted that in general the courses of all curves representing 
calcium-containing solutions agreed more closely than in the more 
dilute solutions, even that containing but i/io Ca(N03)2 to 9/10 
Mg(N03)2 acting relatively more favorably than in the less concen- 
trated solutions. Apparently the mixtures were approaching a situa- 
tion in which the Ca demand was satisfied even in those containing 
the smaller proportions of this ion. In the 4/4 Ca(N03)2 solution 
the maximum net absorption was attained as heretofore near the end 
of the experimental period. By the i8th day the concentration of 
the solution had diminished about 185 N X io~^. The most favorable 
nixture was that containing 1/4 Ca(N03)2 + 3/4 Mg(N03)2 from which 
the net absorption on the 15th day corresponded to a drop in con- 
centration of about 200 N X io~^ It is clear that a toxic concentra- 
tion of 4/4 Mg(N03)2 had been reached since at the end of six days 
very marked leaching began, and at no time was there a net absorption. 
Looking at this experiment as a whole a number of features 
attract attention. The contrast between Ca(N03)2 and Mg(N03)2 
in regard to their influence on root absorption is very marked. The 
calcium salt favored this process in all concentrations here tested. 
Magnesium nitrate in no case supports more than a slight net absorp- 
tion in the most favorable concentrations, and the poisonous action 
of the Mg ion promptly appeared at a concentration between 360 
and 480 N X IQ-^ All mixtures with Ca(N03)2 in all proportions 
were more freely absorbed than Mg(N03)2 alone. The solutions most 
