of the value per unit by the number of units. In the case of a very small stock this rectangle will be very small. On the other hand, in the case of a very large stock, since the value per unit is small and may even reach zero, the rectangle will also be very small, reaching zero, as we have already seen, in the case of free goods. Somewhere in between these two extremes, of course, there will be a maximum rectangle (Fig. 5, at), a stock the total value of which is greater than that of either a larger or a smaller stock. This fact (brought out also in the third column of the table) that after a certain point an increase of the total stock will result in a decrease of the total value, has been called the "paradox of value." Cases have been known of the partial destruction of a stock of goods by its owners as the result of their calculation that the remainder would actually sell on the market for more than could be secured for the whole original supply.
NcrrE The Weber-Feohner Law. In part the effects of repeated stimuli are probably explained by a law of psychology. It is that geometric increase of the stimuli acting on any of the senses is required to produce an arithmetic increase of sensation. It holds "approximately and within a certain middle re gion of the intensive scale for intensities of noise and tone, of pressure, of various kines thetic complexes (lifted weights, movements of the arms, movements of the eyes), and of smell."
"There is some little evidence that affection on the intensive side obeys Weber's law." (Titchener, "A Text book of Psychology," 1911, pp. 218, 259.) The points on the curve to indicate the total stimuli (measured on the ordinate by scale shown at right) required to produce a given degree of sensation, shown by abscissas measured on line to While the stimuli increase in geometric ratio (1, 2, 4, 8) the sen sations increase in an arithmetic ratio (1, 2, 3, 4.) The relative quantity of sensation per unit of stimulus is represented by the height of a, b, c, d respec tively above the base line. The second R produces sensation equal to the first, the third and fourth R (average) produce as much sensation, the 5th to 8th R (average) produce as much. The curve a-d corresponds with the observed trend of decreasing valua tion in many cases. It is clear, however, that valuation does not al ways (perhaps not usually) rise and fall in curves exactly parallel with sensation; for example, a first and a second unit of R might (if there were no more) be neither pleasurable nor valuable; a third and fourth might raise the total sensation to a degree where it was desirable and valuable; and not until the fifth or some latter unit would an addi tional unit of R add a smaller proportional value. The correspondence between decreasing sensation and decreasing valuation is thus found only at certain middle regions of the scale.