ScHAW. — Circulation of the Atmosphere. 573 



tion of the atmosphere. The arrows show the directions of 

 air-motion in the successive portions of the great circulation 

 between the equator and the south pole, and also the prob- 

 able meaning of the high and low mean barometric pressures 

 in these various parts of the circulation, the " highs" corre- 

 sponding with downward indications of the motion, and the 

 " lows " with upward indications. 



Fig. 2 shows on a diagram of half the south polar 

 region, extendmg as far as the 30th parallel of south 

 latitude, the horizontal circulations of the air corre- 

 sponding with the vertical circulations shown in fig. 1, 

 The circulations are shown as if they were mathe- 

 matically circular and equidistant ; but we know that in 

 reality they are very irregular, in consequence, no doubt, of 

 the interaction upon one another of the various circulations, 

 of changes in temperature, and of other influences. The 

 diagram illustrates what seems to be the general disposition 

 of the circulations which move eastward round the probable 

 polar anticyclone in alternate belts of cyclones and anti- 

 cyclones. The arrows show the directions of the winds in 

 the various horizontal circulations. The combinations of the 

 vertical and horizontal motions result in spiral motions. 



As regards the origin of cyclones, I have found no fresh 

 light in Mr. Barthorp's logs ; but the light I have found on 

 the whole subject has proved to be so different from what I 

 had supposed that it gives no support whatever to the hypo- 

 thesis, which I ventured to put forward in a former paper, 

 thai; there is some spot in the antarctic regions which is 

 a specially favourable nursery for the birth of antarctic 

 storms, whence they start on a north-easterly course. These 

 cyclonic storms seem to be an integral part of the atmospheric 

 circulation in the Southern Hemisphere, and I see no reason 

 to doubt my former conclusion that the liberation of latent 

 heat by the condensation of water-vapour and the rotation of 

 the earth are the two main sources of their energy and motion. 



With my hypothetic antarctic nursery I must also abandon 

 the idea that the counter-trade winds have anything to do 

 with the eastward motion of the storms, because the counter- 

 trade westerly winds now appear to be themselves only the 

 northerly parts of the cyclonic circulations in latitudes from 

 40° to 50° south. 



The northward or southward deflection which we often 

 observe in tracing the eastward motions of antarctic storms 

 appears to be due to the forms of the anticyclones or land 

 masses they encounter in their progress ; and to the same 

 causes we must attribute their frequent deformation from 

 their normal circular or elliptic shape, as by the projection of 

 long arms northward, so often observed here. 



