4. iS75] 



NATURE 



ithis general dilatation in the torrid zone takes place under 

 the vertical rays of the sun. In this manner the centre 

 of gravity of the lower strata rises vertically ; these raise 

 the strata above them, which being specifically lighter, dry 

 and transparent, are consequently less sensitive to the 

 sun's rays. All the strata in succession, thus thrust up- 

 ward above their surface of normal height, tend to flow 

 with accelerated motion along these surfaces in the direc- 

 tion of the two poles, where the temperature is relatively 

 low. This effect is still further increased by the peculiar 

 march of the aqueous vapour which is principally con- 

 densed about the poles, whence it returns to the equator 

 by another way than that of the atmosphere, viz., along 

 the surface of the earth in the liquid state. 



The atmosphere cannot exactly follow the diurnal 

 rotation. A half of its mass, or from about 30° lat. S. to 

 » 35° lat. N., lags somewhat behind, since all the molecules 

 in this region being thrust upward describe circles con- 

 • tinually increasing in size with the linear velocity from 

 the lower level from which they started in their ascent. 

 To this retardation must be superadded that of the sur- 

 face trades resulting from their general flow towards the 

 equatorial region. Beyond the tropics, on the contrary, 

 in the temperate zones where the air advances into paral- 

 lels ef latitude continually diminishing in size, the other 

 half of the atmosphere flows in advance of the earth's 

 rotation. Towards the polar circles this advance is con- 

 verted into a circling movement round the two poles from 

 west to east. 



The unequal distribution of land and water over the 

 globe modifies this general aerial curient, so that it does 

 not flow on in one current, but is broken up into many 

 currents — the equalities of the surface throwing the 

 current of the counter-trades into several currents 

 more or less distinct from each other. We can easily 

 imagine the behaviour of the counter-trades by com- 

 bining their march toward the poles with the two oppo- 

 site transverse tendencies of which we are about to 

 speak. Between the tropics, the resulting currents do not 

 blow straight to the equator, but wear round more toward 

 a westerly direction. Beyond the tropics, they do not 

 blow directly toward the poles, but take a course inclined 

 more to eastward. The two following figures will explain 

 our meaning : — 



Fig. 13 represents the whole upper currents for both 

 hemispheres on a projection of the meridian ; and Fig. 

 14 for the northern hemisphere on a projection of the 

 equator. The dotted arrows mark the surface counter- 

 currents ; in other words, the trade-winds blowing 

 obliquely towards the equator, making nearly a right 

 angle with the upper trades of the torrid zone. A slow 

 whirling movement may also be seen around both poles 

 resulting from the counter-trades. These really exist, for 

 the meteorologists of the United States have recently 



described them under the name, a little fanciful, perhaps, 

 of polar cyclones. 



The aiirial rivers which are marked out in the midst of 

 these great movements, by which the equilibrium, inces- 

 santly disturbed, tends constantly to re-establish itself, 

 exhibit then precisely the course which we have recognised 

 as a peculiarity of the trajectories of cyclones, whilst the 

 surface-trades have no relation to these same cur\'es show- 

 ing the courses of cyclones. This agreement is a further 

 proof that cyclones must have their origin in the upper 



Fig. 14. 



regions of the atmosphere, and thence descend even to 

 the ground, and in doing so traverse strata of air either 

 calm or in motion, in such a way as to be totally inde- 

 pendent of the cyclone — a state of things incomprehen- 

 sible on any other hypothesis that has yet been advanced. 

 As to the direction of rotation of cyclones, it results from 

 it that, in these currents strongly recurved, the velocity 

 goes on diminishing transversely from the concave side 

 to the convex side. The zone of calms would then no 

 longer accord with the phenomena of an ascending up- 

 draught, but with a maximum of dilatation to the right 

 and to the left of the place where the movement toward 

 the poles commences. Lastly, the mean velocity of these 

 currents, feeble at first in the neighbourhood of the 

 equator, would go on accelerating just as the velocity of 

 translation of our cyclones. 



Segmentation of Cyclones. — Whatever may be thought 

 of these opinions regarding the march of the upper trades, 

 of which the sur ace-trades are the counterpart, it is im- 

 possible to doubt that cyclones take their origin from 

 these currents. Let us then look more closely at these 

 gyratory movements. If the maximum height of these 

 trades be from 33,000 to 40,000 feet, and the lower dia- 

 meter of the gyratory movements, or cyclones, where 

 they meet the ground, from 120 to 180 nautical miles, it 

 will be seen that cyclones must have a figure very dif- 

 ferent from waterspouts and tornadoes, whose proportions 

 are altogether different, since the height of these last 

 is enormously disproportioned to their lower diameter. 

 Piddington was therefore right in comparing cyclones to 

 mere whirling discs. It would however be more correct 

 to fegard them as waterspouts reduced to their upper 

 funnel-shaped portion, or deprived of their slightly conical 

 column, which descends even to the ground. As it ad- 

 vances, the generating current is lojvered a little ; the 

 vertical height of the cyclone is thereby as much dimi- 

 nished, its section enlarges by contact with the ground, 

 and the disc becomes even more flattened out. 



This being granted, if any whirling movement en- 

 counters resistances, or if the general current exhibits 

 differences of velocity in different places, it is in the 

 upper regions of the atmosphere especially that these dis- 

 turbing causes will act most powerfully on the phenomenon, 

 because the velocities are there less, and the distances 

 traversed by the currents enormous. Below, on the other 



