38 



SCIENCE 



[Vol. XVII. No. 415 



discussed by Ferrel, that in every case the latest and most accu- 

 rately determined physical constants are used, and that the theo- 

 retical deductions, while simply offered as such to be tested, are 

 strictly the results of mathematical analyses. If in time these 

 appear inadequate, the measure of praise for the man and his 

 work may be diminished, but only in proportion as it is remem- 

 bered that meteorological data and laws were in a condition more 

 or less chaotic when he took up his labor of developing these into 

 a consistent harmonious science. 



Al^XANDER McADIE. 

 Washington, D.C., Jan. 9. 



Cyclones and Areas of High Pressure. 



I HAD supposed that Professor Davis would give some explana- 

 tion of the argument against the condensation theory of cyclones 

 deduced from the comparisons of the temperatures in cyclones 

 with those in high- pressure areas. He commences with a citation 

 from my book, in which I state that the high pressures in the 

 north-west sides of cyclones in the higher latitudes in winter are 

 caused mostly by their lower temperatures, and consequently 

 greater densities. He thinks the high pressure over the Alps in 

 November, 1889, is a typical case of all such high-pressure areas. 

 While I do not so regard it, yet, for the sake of brevity, I will 

 here concede it, and consider merely this supposed typical case. 

 Over the Alps, during the last five of the fourteen days of the ex- 

 istence of this high pressure, the temperature on the summits of 

 the Alps was found to be several degrees warmer than the normal 

 temperature of the season. There are no observations to show 

 how high this abnormal temperature extended, but I am willing 

 to admit that it may have extended up to a considerable altitude. 

 Professor Davis, because this temperature is found to be above the 

 normal a few degrees, maintains that the descent of the air is not 

 due to its being heavier than the siirrounding air, thus assuming 

 that the surrounding temperatures at a distance at the time are 

 the same as the normal temperature, notwithstanding the well- 

 known great and long-coatinued departures from the normals 

 which frequently occur over large areas of the country. But it 

 is not necessary that this body of heated air in high-pressure areas 

 should have a temperature lower than the surrounding tempera- 

 tures even; for if the great vertical extent of air above 

 it has a temperature only one or two degrees lower than 

 the surrounding temperatures on the same levels, which gives 

 rise to a descending current, the air below, if it even has a 

 little higher temperature than the surroundings, cannot rise 

 up through the descending current, but must be forced down- 

 ward. But suppose it were clearly established that the air in a 

 high-pressure area extending hundreds of miles had a lower tem- 

 perature than the surroundings even, and not merely the normal 

 of the season : how is the greater pressure and the descent of the 

 air to be accounted for ? Professor Davis has never hinted 

 at a probable explanation merely. The deduction, there- 

 fore, from a few surface observations merely in a very limited 

 region, that the air over a large area, and extending to the top of 

 the atmosphere, is warmer than the surrounding air at a great 

 distance in all directions, especially where these few observations 

 are found to give a temperature above the normal merely, and not 

 above the surrounding temperatures at the same levels, should be 

 received with great caution ; for, if there were even a well-established 

 theory to account for the descent of the air under these circum- 

 stances, these observations could scarcely be regarded as having 

 any weight in confirmation of such a theory. 



In what precedes I have gone upon the assumption that a lower 

 temperature is the only cause of the descent of the air in high- 

 pressure areas. While I regard this as adequate to account for 

 it, I have never said or thought that it is the only cause, but 

 simply the principal cause. I think there are other causes, espe- 

 cially in the origin of these high-pressure areas, which, for our 

 present purpose, it is not necessary to discuss here. 



Professor Davis says, "Records of temperature made on high 

 mountain-peaks furnish the best means of testing the convectional 

 theory of cyclones, for, even if all other tests were successfully 

 borne, failure under this test would be fatal to the theory." By 



" convectional theory of cyclones" I understand him to mean the 

 condensation tbeory, which requires the air in the ascending cur- 

 rent to be warmer and lighter than that of the surroundings at 

 the same levels. Now, this theory can neither be established nor 

 overthrown by any such tests. Cyclones are usually several hun- 

 dred, sometimes a thousand and more, miles in diameter; and to 

 prove that the air over so large an area up to the top of the at- 

 mosphere, or at least up to high altitudes, has a higher or a lower 

 temperature than its surroundings, would require numerous sta- 

 tions of observation at many different levels, not only over this 

 large area, but also all around this area at great distances. The 

 condensation theory requires that the temperature of the air in a 

 cyclone must be greater, in a general way, than that of the sur- 

 rounding air; but this does not mean that there are no places 

 within the cyclone, especially on the earth's surface, with lower 

 temperatures than those of many places outside. In the theoreti- 

 cal treatment of a cyclone we have necessarily to assume certain 

 regular conditions of uniform temperature at the same distances 

 in all directions; but I have always been careful to explain that 

 such conditions are never found in nature, but generally only 

 rough approximations. In a large cyclone there is a great differ- 

 ence between the north and south sides, due to difference of lati- 

 tude, which is taken into account in the general motions of the 

 atmosphere, and so must be excluded in the treatment of the cy- 

 clones, and the differences of temperature only with reference to 

 corresponding temperatures outside of the cyclone on the same 

 latitudes must be considered. Besides, the temperatures vary all 

 around the cyclone, not only on account of difference of latitude, 

 but likewise from various abnormal causes. It must be expected, 

 therefore, in comparing inside temperatures with the surrounding 

 ones, especially surface tempei'atures, that there would be numer- 

 ous cases in which those within would be found lower than many 

 of those in the surroundings. The theory only requires that there 

 shall be a predominance of higher temperatures in the interior. 

 Besides, the conditions of a cyclone need not extend down to the 

 surface at all, and, in fact, mere surface conditions generally have 

 little or nothing to do with a cyclone. If the necessary conditions 

 exist at altitudes only considerably above the earth's surface, the 

 air is thrown into a great whirl or gyration, which relieves the air 

 below of a part of the pressure upon it, and increases the pressure 

 round about; so that this air tends to rise up, .just as the water 

 does in a suction-pump, and the surrounding air flows in to take 

 its place; and in flowing in it assumes a gyratory motion, not 

 only from the deflecting force of the earth's rotation, but likewise 

 from the action of the air above by means of friction, so that it is 

 brought into the general vertical and gyratory circulation. But 

 suppose that it could be shown that the air in a cyclone is mostly 

 or entii'ely of a lower temperature than the surrounding air at all 

 altitudes, and yet ascends, as it always does: how is this strange 

 phenomenon to be accounted for when there is no force, either 

 real or imaginary, to cause it to ascend ? 



Professor Davis thinks that the snow-faU on the Alps at the 

 time of the cyclone of Oct. 1, 1889, had little effect in lowering 

 the temperature, on account of the wind ; but this is one of the 

 causes which Dr. Hann gave, a few years ago, of the lower sur- 

 face temperatures in cyclones. The air, in being forced up the 

 mountains on the windward side, is expanded and cooled below 

 the temperature of the air generally on the same level. Another 

 reason which he assigned was, that as the lowest pressure above 

 lags behind that below, as was shown by Loomis, and first ex- 

 plained, I think, by Dr. Hann, the cold north westerly winds set in 

 above rather before the lowest pressure-point is passed. The real 

 centre of the cyclone above is not that of lowest pressure. 



I admit that it is not strictly logical to assume that two theo- 

 ries, or two kinds of forces, may not be such as to give the same 

 effects, especially where nothing is known of the nature or man- 

 ner of application of the one kind ; but still this is extremely im- 

 probable. As the general motions of the atmosphere, cyclones, 

 and tornadoes, are all very much alike, consisting of gyra- 

 tions around a centre, — and it is admitted that in the first and 

 last the air rises where it is warmest and lightest and because this 

 is so, and that this is even the case with cyclones in the lower 

 latitudes,— we should hesitate in making an exception in the case 



