Abstract
From theoretical considerations in conjunction with data on heat production during prolonged fasting it has been shown 1 that the optimal relative rate of starvation at which a human individual would conserve his body reserves to the utmost ought to be closely in the neighborhood of 0.0064 Kg./Kg./Day. It is now proposed to examine this result in the light of several observations on the point.
By far the most accurate of the data at hand are those reported by Benedict, 2 but in Figure 1 we have also included the curves of Succi for several fasts, as well as that for Beauté. The ordinates are in log10 (Weight) = .4343g, the abscissae in days. There is, here, a striking similarity not only in the almost identical shape of these curves, but also in their slope, especially after the 4th or 5th day of fasting. This slope appears so decidedly linear that purely graphical analysis could hardly detect any significant curvature. We shall call this the “apparently linear” phase.
Taking the data on Levanzin by Benedict for each day from the 8th to the 30th inclusively and analyzing this apparently linear phase by least squares, we find that the slope is 0.00645 Kg./Kg./Day, a numerical result completely confirming that given above and already reached by an independent method previously described. 1
Straight lines have also been drawn through the other examples and indicate the expected slope, 0.0064. These likewise pass excellently through their respective points in every case, notwithstanding the diverse conditions and times of observation.
But, even though the foregoing value for an “optimal” rate of starvation is actually witnessed in practice—we must avoid the conclusion that starvation would necessarily continue indefinitely in this way. Indeed, as it will appear in a subsequent paper, the “linear” phase in such experiments discloses merely a temporary, though an admirably close approach to what are fundamentally “ideal” conditions of starvation for the individual subject and for the kinetic system his organism represents.
Get full access to this article
View all access options for this article.
