90 ECONOMIC ESSAYS IN HONOR OF JOHN BATES CLARK 
If no obstacles intervened it would increase by the proportion 
A B, which in this case of unit elasticity would bear the same rela- 
tion to A as PP; to P. But since the supply of Y had increased 
and that of X had remained constant, the marginal productivity 
of X would certainly be greater in terms of Y than it would have 
been had their elasticities been equal. The unit return to X 
would therefore rise above Pi, to, let us say, P.. The marginal 
productivity of Y, on the other hand, would have fallen 
because there would be relatively more of it mixed with each 
unit of X than before. Its return per unit would therefore fall 
below P; to, let us say, Ps. But this very decrease in the 
marginal productivity of Y would in turn dampen off the rate 
of growth of the curve and would cause less than B to be 
produced and would lessen the rate of increase in the unit return 
to X and bring it down below Po. 
But how far would this process of readjustment go? It would 
not be sufficient to bring the return to X back to P; or of Y 
to P; since Y would certainly show some increase in its total 
quantity, and any increase in unit return over O P would call 
forth a proportionate increase in the quantity supplied of Y 
while the supply of X would not increase. There would, therefore, 
be a permanent increase in the quantity of Y offered over the 
supply A and hence an increase in the relative marginal pro- 
ductivity of X in terms of Y. The return per unit of X would 
rise above P; while that of Y would fall below P;. X would 
not rise to P, however, because of the dampening off of Y's 
rate of growth, and would settle, let us say, at P;,. The return 
to Y in turn would not be equal to P; but would, instead, be 
something less than this amount but more than P; and would 
be fixed at Ps. The ultimate result will, therefore, be that X 
will secure a greater proportionate return per unit than the 
increase in the total effectiveness of industry, while Y will 
secure a lesser unit increase. 
It is not conclusively demonstrable by graphic methods alone 
whether X as a whole will secure a larger share of the total 
product than before, or whether the greater number of units of 
Y which have been supplied will be more than sufficient to offset 
the lesser increase per unit. From mathematical illustra- 
tions, which have been worked out by my associate, Mr. 
3. W. Wilcox, however, it is apparent that under the assump-