¹ This Journal, XXV, no. 1, Feb., 1959, 65-70.

² It would, of course, be an astonishing coincidence if all the sum products (β _i Z _i + γ _i t) were exactly equal. What Mr. Lenschow must mean is that they were not significantly different. Since, from Table II, the errors of the coefficients are quite large in relation to the size of these coefficients, that seems entirely plausible.

³ It is not entirely clear whether the price variable y is itself also different for each route: the text on page 66 clearly suggests that this is the case, but the notes to Figure 1 give the contrary impression. This question does not affect the outcome of the argument that follows, and it is obvious how the argument should be amended if the y ⁱ differ between routes.