IN reply to Mr. Clement Ley's letter in NATURE, vol. xv. p. 253, I fear I cannot at all agree with him as to the cause of the polar depressions of the barometer. He says: “The ‘polar cyclones’ appear to be themselves aggregates of those local depressions, or cyclones, which have penetrated into the Arctic or Antarctic regions, and have there partially or wholly coalesced.” Now, let us test the question in this way:—Suppose the surface of our planet were all land, so that there was no watery vapour in the atmosphere; there would be no cyclonic storms, for they are due to what Espy truly calls steam power;—would the polar depressions of the barometer be observed as they are in our actual atmosphere? Mr. Clement Ley's reasoning seems to require him to say that they would not; I have no doubt that they would. The causes which produce the west winds of the middle latitudes (Maury's “counter-trades”) would act as in our actual atmosphere, and their centrifugal force, in rotating round the poles, would produce a space of shallow atmosphere at and around each pole, exactly like the depression at the centre of a vortex of water, which would show itself, as at present, by a depression of the barometer.