500 



NATURE 



\OcL 7, 1875 



ascending air, when it has reached the elevated regions 

 of the clouds, could not be diverted above by upper cur- 

 rents so as to assume the form represented in Fig. 8 ; for 

 these upper currents could no more displace the focus of 

 aspiration than they could move a locomotive by deflect- 

 ing the column of smoke which issues from it. (2) The 

 mechanical effects will be very limited, because the aspir- 

 ing force being measured by a few millimetres of mercury, 

 were the end of the suction-tube to be plunged into a 

 river or the sea, the water would be raised there a few 

 centimetres. Moreover, at the instant when the extre- 

 mity of the tube reaches the ground or the water, the air 

 ceases to flow into it and fails any longer to keep the 

 ascending column together, and thus all mechanical 

 action ought to disappear at this moment. Further, it is 

 evident that if the phenomenon has its origin in a per- 

 fectly calm stratum of air Avhere not a breath of air is 

 felt, the element of mechanical work, that is to say the 

 force or the motion, fails altogether, or becomes reduced 

 to a feeble ascensional tendency in any stratum of air 

 that may have acquired over the place an abnormal 

 excess of a few degrees of temperature. 



Compare now with the facts, these two conclusions drawn 

 from the theory. It would be difficult to find a disagree- 

 ment more complete. Everyone is aware of the ravages 

 produced by hurricanes, typhoons, tornadoes, and even 

 simple waterspouts and whirlwinds— ravages which imply 

 an enormous development of mechanical force. Then, 

 everyone knows that the peculiarity of all cyclones is to 

 possess a movement of translation, often very rapid, which 

 the theory of centripetal aspiration denies to them. Of 

 all waterspouts hitherto observed, only one instance of a 

 stationary one has been recorded ; and even the sta- 

 tionary character in this exceptional case may have been 

 not real but only apparent. As regards tornadoes, all 

 those that traversed the United States since 181 1 were 

 propagated onwards with a speed varying from four or 

 five to twenty metres a second. The well-known water- 

 spout of Monville, in France, swept over a league in less 

 than four minutes, or at a rate of about twenty metres 

 per second. At such rates typhoons and cyclones, without 

 exception, also advance ; their movement of translation is 

 usually increased as they proceed into higher latitudes, 

 and varies from three to eighteen nautical miles an hour, 

 or from two to ten metres per second. 



It has been said by the advocates of the theory of aspi- 

 ration as applied to hurricanes, that if the converging 

 currents are stronger on one side than on the other, the 

 centre of aspiration, that is the base of the waterspout, 

 will be displaced in the direction indicated by the stronger 

 currents, as shown in Fig. 9. But why this difference of 

 speed, especially over the sea, where there are no ine- 

 qualities of surface over which the different winds blow ? 

 The velocity with which air free to move rushes into a 

 suction-tube is determined by the amount of the suction 

 force ; if the movement be impeded on one side of the 

 orifice, the air will enter by the other with a determined 

 velocity, [but not with a velocity tripled or quadrupled. 

 Moreover, in order that an excess of velocity of twenty 

 metres per second on one side of the centripetally flowing 

 currents could communicate a like velocity to the onward 

 march of the waterspout, it would be necessary that a 

 wind of the force of a terrible tempest blew in that direc- 

 tion exceeding by a velocity of twenty metres per second 

 the contrary wind. This is scarcely compatible with the 

 absolute calm which ordinarily prevails round water- 

 spouts, tornadoes, and even typhoons. 



Fig. 10 represents the appearance of a waterspout 

 whose base is represented as driven forward by a sup- 

 posed excess of velocity of the inflowing horizontal cur- 

 rents, whilst the top of the ascending column is retarded 

 by the resistance of the air. Now the real figure is that 

 represented in Fig. 1 1, and it agrees neither with Fig. 9 

 nor with Fig. 10. 



It will be seen on reflection that under all these 

 attempts at explanation there lies a settled conviction 

 which Pliny has aptly expressed in these words : " Quum 

 spissatus humor rigens ipse se sustinet," — the idea, in fact, 

 which was naively reproduced in the experiment at the 

 foundry at Washington, in which it was tacitly assumed 

 that the column of a waterspout or tornado is composed 

 of some rigid material, and that it may be displaced 

 bodily by a force acting on its lower part. In truth, the 

 force which could so act is not to be lound. The expla- 

 nation suggested by Prof. Mohn, that the movement of 

 translation of storms is determined by a difference in the 

 average pressure in the front as compared with the rear 

 of the storm, caused by the condensation of vapour 

 which takes place in front, is insufficient, because we see 

 waterspouts and tornadoes marching onwards, from which 

 not a single drop of rain falls. 



No navigator has ever shown that there is in a cyclone 

 the least indication of a decided movement of ascension 

 to which the essential cause of the phenomenon is attri- 

 buted. Everyone speaks about ascending currents, but 

 no one has seen them, or seems even to have had the 

 idea of verifying their existence in the case of their 

 assumed hurricanes of aspiration. The whole thing is 

 taken for granted, and preconceived notions, whose origin 



we have traced, have complete control over tneir thoughts. 

 As regards waterspouts, no manner of doubt is enter- 

 tained, for the water is seen whirled up their columns sky- 

 wards. If this were really the case, waterspouts and 

 tornadoes might draw up even to the sky the thousands of 

 trees which they uproot, and a little afterwards furnish 

 the spectacle of a whole forest tumbling from the clouds, 

 it being evident that thousands of trees may be trans- 

 ported as easily to the clouds as thousands of tons of sea- 

 water. Eye-witnesses are not wanting to prove that 

 branches of trees have ascended beyond the clouds, they 

 having been seen lying at some distance on the ground, 

 covered with hoar-frost in the middle of summer. 



What remains then of the theory of hurricanes which 

 is based on centripetal aspiration .? It starts from a pre- 

 judice, sacrifices at the outset the simplest notions of 

 mechanics, and does not take the trouble to represent a 

 single characteristic trait of the phenomenon. Is it then 

 on this theory we are to rely for the rectification and 

 completion of the laws of storms 1 Shall we borrow from 

 it, in order to correct the diagrams oi Reid and Pid- 

 dington which are perhaps in some cases too absolutely 

 circular, the hypothesis of centripetal currents suggested 

 by it. Especially shall we sacrifice to it the practical 

 rules of navigation followed during the past thirty years ? 

 Unhappily there is some cause for fear, for sailors them- 

 selves have long since been prepared by the tales and 

 narrations of the forecastle, for these ideas of aspiration 

 affirmed regarding tornadoes, typhoons, and cyclones. If 

 then they are told that in a particular case one of the 



