H 



NATURE 



\Nov. 4, 1875 



hand, where the column is narrowed to a very great 

 extent, and where the velocity of the gyrations is exces- 

 sive, the obstacles met with exercising little influence, are 

 instantly reversed or overcome. Waterspouts acting by 

 the lower extremity, at a distance from their funnel-shaped 

 top {embouchure), undergo no change ; but cyclones will 

 not withstand the forces brought into play so easily. 



Let the modifications thus induced by external causes 

 be what they may, the preceding theory shows that they 

 are possible, and that the rigorously circular movement 

 enunciated by the authors of the " Laws of Storms " 

 allows of perturbations more or less local, and more 

 or less marked, for the simple reason that the 

 whirling movements, which in the case of cyclones 

 are reduced to their upper portions and are there- 

 fore little more than mere discs, are very easily 

 modified. It is this, moreover, which explains the de- 

 viations from the rule that are found in even the earliest 

 writings of the authors of the " Laws of Storms," as, for 

 example, in Fig. i (vol. xii, p. 401), representing the Cuba 

 hurricane, where, notwithstanding the general agreement 

 of the arrows showing the wind with the purely circular 

 theory of storms, there also occur several local deviations. 



what can these perturbations be ? How can a gyra- 

 tory movement be changed under the influence of a given 

 external cause ? What happens if the velocities of the 

 generating current undergo local changes ? It would be 

 as difficult to answer these inquiries a priori z.?, it would 

 have been to foresee, before the development of the 

 mechanical theory of solid bodies, the astonishing results 

 of an external force brought to bear on them ; but the 

 study of other whirling movements more within the reach 

 of observation, and directed to the sun, has shown that 

 a cyclone is not arbitrarily deformed in any manner 

 whatever. Segmentation, or breaking up of the cyclone, 

 is the last term of the alterations which it can undergo. 

 Then the fragments into which it is broken up tend to 

 assume the form of cyclones, each as perfect as the one 

 from which they were formed, and they follow routes dif- 

 fering but little from each other and describing nearly 

 the same trajectory, but at a distance from each other. 

 This segmentation of cyclones occasionally occurs in the 

 case of the thunderstorms which advance on France from 

 the Bay of Biscay, of which the thunderstorm of the 9th 

 of March, 1865, so well described by M. Marid-Davy, 

 may be cited as an example. A like process of segmen- 

 tation cannot take effect unless the primitive cyclone in 

 some part and for some time deviate from its rigorously 

 circular form. The tendency to keep this form maintains 

 the ascendency sometimes, but if it begins to give way in 

 a large cyclone, the result is a breaking up of the cyclone 

 itself into segments. 



It would be easy to adduce numerous cases in which 

 whirlwinds, tornadoes, and cyclones appear in groups 

 about a given point, or at least follow each other with 

 rapidity. The evidence all goes to show that they are 

 most frequently the result of the phenomenon of segmen- 

 tation, so called from the term used in natural history to 

 designate the process by which some of the lower animals 

 are divided into segments each of which soon becomes a 

 complete animal of itself. But it is in solar cyclones 

 where this mysterious operation can be best followed step 

 by step. Thus a circular sunspot may be seen gradually 

 undergoing the process of deformation, then breaking up 

 into parts, and ending by giving birth to other spots, 

 which precede the original one in a row and at some dis- 

 tance, proceeding at the same pace and reproducing on a 

 small scale the features and behaviour of the primitive 

 type. 



Conclusion. 



The laws of storms, the statement of which in absolute 

 terms ignores the modifications wc have indicated, are 

 therefore in reahty only an approximate enunciation, just 

 as are Kepler'b laws, to which we have mo:e than once 



compared them. Kepler's laws would be rigorously exact 

 if we could leave out of account the action of the planets 

 on each other and on the sun ; but this being impossible, 

 these laws are not an adequate expression of the truth. 

 The same holds good with respect to the laws of storms. 

 They would also be exact if the currents of the atmo- 

 sphere never exerted any disturbing action, and as the 

 laws take no account of these disturbing actions, and do 

 not give the means of foreseeing them, or at least of 

 measuring their effects, it would be a mistake to apply 

 them blindly in practical affairs. 



It was not by substituting the cassinoide for Kepler's • 

 ellipse that science made progress ; in like manner it will 

 not be by the substitution of centripetal diagrams of 

 storms for circular diagrams that navigation will be ren- 

 dered safer. If we have succeeded in giving the true 

 theoretical interpretation of these laws, it must be granted 

 that the time has not come to abandon them, but rather 

 to make them more complete. 



To sum up, there are no centripetal waterspouts, 

 whirlwinds, typhoons, or cyclones. The moveable forces 

 of aspiration formed, as is said, over the heated ground 

 of the tropics, do not transport themselves with their 

 accompanying updraught to a distance of 700 or 800 

 leagues over the cold soil of high latitudes, and they have 

 never determined the whirling movements of our atmo- 

 sphere. The Laws of Storms are in general accord with 

 the mechanical theory of these movements. The rules 

 of navigation which are deduced from them, merit in 

 ordinary cases the confidence sailors have had in them 

 for the past thirty years. The exceptions should be only 

 regarded as mechanical disturbances of the gyratory 

 movement, the further study of which seems destined to 

 complete a first and happy approximation to the truth. 

 The discovery of the approximate laws of storms is one 

 of the finest scientific conquests of this century, and if a 

 closer approximation is to be made, it will be by a more 

 careful study of solar cyclones. 



Formerly whirling movements played an important part 

 in our general conceptions of the universe. Fallen into 

 disrepute by a very natural reaction from a false idea, 

 they have been too much forgotten ; therefore when at a 

 later period a gyratory character was recognised in the 

 great movements of the atmosphere, an effort Avas made 

 with one consent to connect them with totally different 

 causes. Geometricians seemed to class them among 

 those irregular movements of which nothing could be 

 made. We see, however, that movements of the cyclonic 

 order constitute in truth a vast series of regular and 

 stable phenomena, of which their perturbations even 

 exhibit a behaviour in accordance with geometric prin- 

 ciples. This seiies which begins with simple eddies in 

 our streams of water, embrace the most singular as well 

 as the most dreaded phenomena of the atmosphere, to- 

 gether with the vast movements which observation reveals 

 in the sun, and extends even to the nebulte, the structure 

 of which Rosse's telescope has proved to be characterised 

 by whirling movements. It is therefore most desirable 

 that the theory of these moverrents should be again in- 

 cluded in the domain of applied mechanics. The first step 

 to this end is an empirical investigation of their laws, and 

 this work the eminent authors of the " Laws of Storms " 

 accomplished thirty years ago. 



NOTES 



Two members of the British Ornithologists' Union, Messr 

 Harvie-Brown and Henry Seebohm, have recently returna 

 from a most successful expedition into Northern Siberia. Lea^ 

 ing this country early in the spring of this year, they arrived at 

 Ust Zylma, on the Petschora River, ia the middle of April, after 

 travelling overland fr>,m Archangel. They remained tliere 



