May ii, 1882] 



NATURE 



33 



while the cyclone was yet at the Windward Islands, about 

 1200 miles distant. 



A sudden and abnormal rise of the barometer thus 

 constitutes as important a warning to a navigator as a 

 similar fall, or a band of cirrus-cloud, only to be able to 

 make effective use of this danger-signal he ought accu- 

 rately to know the normal height of the barometer where 

 he is, and at the time of year. The author shows how 

 this may be done in Part I., where he has constructed 

 charts, based on a large series of observations in the 

 northern hemisphere, showing not only the curves of 

 mean annual pressure, but also those which represent the 

 coefficient of annual inequality. From a simple equation 

 involving these two elements, the normal pressure at any 

 time and place can be approximately reckoned, and hence 

 the amount of abnormality determined. 



The author next applies the cyclone theory in explana- 

 tion of the various iniqualities of barometric pressure, 

 which are observed on the same latitude in different 

 longitudes. These inequalities he considers to be mainly 

 dependent on the deviations of the mean temperature 

 (annual or monthly) from the mean of all longitudes, 

 which he gives in a tabular form for every fifth degree 

 of latitude and every tenth degree of longitude in 

 the northern hemisphere, by means of interpolation 

 from the observations discussed in Part I. From these 

 tables it appears that in addition to, and superim- 

 posed upon, the general system of two polar cyclones 

 due to the normal differences of temperature between 

 the Equator and the Poles, we have throughout the 

 year, and more especially in the winter, the conditions 

 for the existence of a large fixed warm-centred cyclone in 

 the North Atlantic, with its centre near Iceland. The 

 barometric pressure should consequently be lower here 

 than the mean of the latitude taken round the globe. 

 That this is the case is well-known, and also that the pre- 

 valence of south-west winds in these islands is due to our 

 generally lying on the south-east edge of this nearly per- 

 petual cyclone. A similar cyclone similarly produced lies 

 in the North Pacific. 



Two corresponding regions of abnormally low tem- 

 perature lie one on the east side of Asia, and the 

 other on the east side of America, which, according to 

 the author's theory, should give rise to cyclones with cold 

 centres. As a matter of fact, however, these conditions 

 are found to be completely reversed ; the pressure being 

 above the average, especially in winter, when the tem- 

 perature-gradients are steeper, and therefore, according 

 to the author's views, the cyclonic conditions should be 

 more developed ; while the motion of the air at the surface 

 is anticyclonic, and outwards from the region of greatest 

 relative cold. 



The least satisfactory part of the author's work is that 

 which relates to these cyclones with cold centres. Their 

 non-existence in the progressive form is admitted, and 

 where they should occur according to theory in a station- 

 ary form, they are notably absent, except in the two 

 circumpolar cyclones. It is possible, however, that they 

 may be identified, though in a modified form, and lacking 

 the central barometric depression at the earth's surface, 

 with what are termed "winter anticyclones," which usually 

 coincide with areas of great cold, and which, while they 

 exhibit at the earth's surface an anticyclonic outflow 

 of air, are fed above a certain level by a cyclonic 

 inflow. 



Finally, as regards rainfall, which is an almost unfailing 

 accompaniment of cyclones, the author, while admitting its 

 assistance in helping to maintain a cyclone when once 

 started, by the forces which operate whenever vapour is 

 condensed, is strongly opposed to its being a primary 

 source of energy, and cites in favour of this notion the 

 following conclusion, arrived at by Prof. Loomis, after a 

 careful study of the U.S. Signal Service charts. " Rain- 

 fall is not essential to the formation of areas of low 



barometer, and is not the principal cause of their forma- 

 tion or of their progressive motion." 



The last chapter of the author's work which relates 

 to tornadoes, waterspouts, and hailstorms, has already 

 been referred to in a special article in Nature, 

 and it only remains for us to observe in connection 

 therewith, that while tornadoes differ specifically in 

 many respects from cyclones, the condition of the at- 

 mosphere in the latter is eminently favourable to their 

 production. To this circumstance, according to Ferrel, 

 may be attributed the occurrence of sudden blasts of tor- 

 nado violence in the middle of cyclones, accompanied by 

 a rapid oscillation of the wind-vane. It is these sudden 

 gusts which do the main damage in such cases, since, as 

 might be expected, the velocity of the wind increases per 

 sallum where the gyrations of the tornado and the 

 cyclone coincide in direction. They are found to occur 

 more on the cold or clearing-up side of a cyclone, which 

 Ferrel explains to be due to the cold upper strata over- 

 lapping the warmer central part of the storm, and thus 

 promoting a condition of vertical instability of equi- 

 librium in which tornadoes are generated with facility. 

 Viewing the work as a whole, Mr. Ferrel may be 

 congratulated on having presented to the world a memoir 

 of such luminous research as well as practical utility. 

 When we compare it with the numerous other crude 

 treatises and hypotheses evolved during the past half- 

 century on the same subject, which have not only brought 

 the science of meteorology into ridicule, but encumbered 

 our libraries, we feel a deep sense of relief at finding the 

 question dealt with by a mathematician of more than 

 ordinary ability, and one who does not shrink from 

 tackling the real difficulties of the subject. He has for 

 some time been known by his writings on hydrodynamical 

 questions of great importance, especially those applying to 

 the general motions of the atmosphere. The present work 

 will go far towards placing him in the very front rank of 

 physical and theoretical meteorologists. The deductive 

 method has been fairly applied throughout to the equa- 

 tions of motion, and its success will do much towards 

 counteracting the too prevalent tendency at the present 

 time to induct from every solitary phenomenon, or experi- 

 ment, to some otherwise baseless hypothesis. If the author 

 has not accounted for all the peculiarities of cyclones, he 

 has at least shown that the views entertained by the leading 

 meteorologists regarding their formation, characteristics, 

 and general movements accord with their mechanical 

 theory, and that the sources of energy ordinarily assumed 

 to act, such as heat, gravitation, and terrestrial rotation, 

 are sufficient, without having recourse to any wild hypo- 

 thesis founded on some unknown function of electricity. 

 The valuable practical hints and suggested modifications 

 of existing rules will do much to avert disasters at sea, 

 the main purpose, doubtless, for which the work was 

 designed, while its thoroughness and comprehensive 

 character will materially help to advance our knowledge 

 of a meteor, which in one form or another comprises 

 almost every condition of the atmosphere included under 

 the term "weather." E. DOUGLAS Archibald 



ON PHOTOGRAPHS OF THE SPECTRA OF 

 THE NEBULA IN ORION* 



FOR about eighteen months I have been giving atten- 

 tion to the nebula in Orion with two objects in view, 

 first to ascertain whether any changes are taking place in 

 that body by making a series of photographs to be com- 

 pared in the future with a similar series ; and second, to 

 photograph the spectrum of the nebula in various parts 

 so as to see whether any new lines could be found, and 

 also whether the composition is uniform throughout. 

 As to the first of these objects I have recently suc- 



1 Read before the National Academy of Sciences, April. 1882, at 

 Washington, U.S., by Henry Draper, M.D. Communicated by the author. 



