In a recent issue of this journal W. R. Allen criticized Keynes' and my applied instability theorems as a priori.' The main bone of contention in his critique seems to be that a high-saving, lowconsumption economy may not be endogenously so unstable as Keynes, Hansen and others have suggested. I shall in this note concentrate, though not exclusively, on that part of Allen's critique which pertains to my own application. By way of providing perspective may I repeat the Keynesian instability theorems :2 (1) The nearer to unity than to zero the marginal propensity to consume, the greater will be proportional fluctuations in income and employment corresponding to given fluctuations in investment. (2) The lower the average propensity to consume, the greater the preponderance of capital-goods industries in the economy and the greater also the degree of instability associated with those industries. The first of these theorems was advanced by himself,3 while the second theorem is my application of Keynes' analytical distinction between the marginal and the average propensity to consume to a longer time horizon than in his short-period analysis. Keynes' elasticity of income, which he formulated to support his instability theorems in a mathematical footnote,4 gave an opening for Allen's formal attack. As Allen correctly pointed out, failed to specify his consumption function as linear, non-proportional, with the result that the investment elasticity of income regarded as a measure of instability turns out to be an indeterminate function of income for either a poor or a wealthy economy. That formal failing of is, however, not as serious as Allen makes it out to be, for two reasons. First, Allen has himself admitted that Keynes could have salvaged the