APART from other considerations, I think that Mr. Craig's contention (NATURE, January 7, p. 281) as to the isothermal layer is disproved by the results of theodolite observations on ballons-sondes. When a balloon is observed, its altitude above the horizon is in general seen to decrease, showing an increase of wind velocity with height. If Mr. Craig's supposition were true, this decrease in apparent altitude would become still more marked when the balloon had reached such a height that it no longer ascended; but the contrary is the case. In nearly all the ascents in which I have observed balloons for a considerable time the angular altitude, after decreasing, commences to increase again; on Mr. Craig's supposition this would mean that the balloon, after it reaches the floating condition, enters a current of air that brings it nearer to the observer, and for this to occur frequently is extremely unlikely.