62 



NATURE 



\_Nov. 2 1, 1889 



It thus appears that the most striking characteristic 

 which may accompany the formation of haze is an un- 

 usual dryness of the air, and that a total absence of haze 

 is often observed when the air is unusually charged with 

 vapour. It does not follow that haze, or a light fog 

 much resembling it, is not also seen in a damp state of 

 the air, or that a saturated air is always free Irom haze ; 

 indeed, something much resembling a dry haze does occur 

 with sudden changes of temperature in all ordinary hygro- 

 metric states in our climate. But the very condition to 

 which haze in England is commonly, and in a certain 

 sense correctly, attributed — namely, atmospheric humi- 

 dity — is, if sufficiently uniform and extended, least favour- 

 able to its manifestation. A constant moisture-laden 

 westerly breeze would give a climate nearly as clear as 

 that of the south-west corner of France. 



Two principal factors go to the production of ordinary 

 haze : the first, a rather large amount of vapour between 

 the earth and a great altitude, say 60,000 feet ; and the 

 second, a mixture of two heterogeneous masses of air. 

 Evidence of the correctness of this proposition is to be 

 found in the geographical distribution of haze and the 

 state of the winds when it occurs. 



The causes of fog are either radiation of heat from the 

 earth into space and cooling of the overlying humid strata 

 of air to a temperature below the dew-point, or else the 

 mixture of two winds, differing in temperature and other 

 conditions, one of the currents being usually near its 

 point of saturation previous to contact with the other. 



If the above-mentioned statement of the causes of 

 haze be correct, we shall be enabled to account for the 

 appearance of haze in certain conditions, which have 

 been given, and for its absence in others. Taking them in 

 order — 



(i) A gentle wind from east to north-east inclusive 

 is favourable to haze, especially if it extends to no very 

 great height. Often the approximate depth or height of the 

 easterly current is difficult to ascertain ; but, in general, 

 if it be of short duration, it is shallow, and sometimes 

 upper clouds from a westerly direction may be observed. 

 In these cases especially haze prevails. Considering the 

 shallowness of lower winds compared with their extent — 

 an easterly wind, for instance, which has travelled 300 

 miles beneath a westerly wind only four miles above the 

 earth's surface — it is quite certain that a very large ad- 

 mixture of the two currents must take place. And 

 we may be sure that in. the majority of cases the easterly 

 surface wind has above it an upper current from a westerly 

 direction. Mr. William Stevenson {Edinburgh Philo- 

 sophical Magazine, July 1853) observed the cirrus cloud 

 at Dunse, Berwickshire, for eight years, and from his 

 summary of the direction of the motions of that cloud 

 we derive the following figures : — 



Direction of motion of cirri from between south- 

 west and north-west inclusive 



Direction of motion of cirri from between north 

 and east inclusive 



Other directions 



Direction of wind at surface of the earth from 

 south-west to north-west inclusive 



Direction of wind at surface of the earth from 

 north to east inclusive 



Other directions 



75-2 



10 



14-8 



54-6 



32-4 

 13 



Thus there remains a difference of over 20 per cent, excess 

 of westerly upper current over westerly surface wind, and 

 at the level of the cirrus a wind between north and east 

 only prevails once to every three occasions of a surface 

 wind from that quarter. The significance of these figures 

 is not seriously affected by the idea, first suggested by 

 Admiral Fitzroy, that visible cirrus is less likely to form 

 in the polar than in the equatorial current, and any 

 careful observer can easily satisfy himself that westerly 

 winds are more common and easterly winds less common 



at the cirrus level than on the surface. Mr. Buchan 

 (" Handy Book of Meteorology," p. 230) remarks that, as 

 the north-west current advances into southern latitudes, 

 the increasing heat of the sun will tend to dissolve the 

 cirri which mark its course, and he therefore thinks that 

 the north-west upper current is the most prevalent in 

 Great Britain. The actual numbers obtained by Mr. 

 Stevenson during the eight years were 243 for north-west, 

 and 256 for south-west direction of cirrus. 



Mr. Ley ("Laws of the Winds," Part I. p. 154) re- 

 marks :■ — " The fact, indeed, that the observed westerly 

 upper currents prevail over the observed easterly upper 

 currents, even more than the westerly surface winds do 

 over the easterly surface winds, has been admitted by 

 most of the observers who have investigated the subject 

 in different parts of Western Europe ; and the same 

 phenomenon is noticed in similar latitudes of North 

 America. ... Be this as it may, the theory of prevalent 

 polar upper currents derives no support from our own 

 collection of examples. Again, the results of the obser- 

 vations classified in Table IV. appear altogether adverse 

 to the supposition that an easterly upper current is 

 common over the northern portions of those depression 

 systems whose westerly winds are the strongest at the 

 earth's surface. . . . Instead of easterly upper currents, we 

 find a great preponderance of southerly currents." 



Out of nine balloon ascents recorded in Glaisher's 

 " Travels in the Air," in which the wind at starting from 

 the surface was easterly, there was not one in which a 

 different current was not encountered at a moderate 

 elevation. The changes were as follows : — 



On one occasion — January 12, 1864— the temperature 

 from 3000 to 6000 feet was higher than on the surface, but 

 at 1 1,500 feet it was more than 30^ colder — namely, 11°. 

 A large number of balloon ascents show not only a variety 

 of currents, but large and sudden variations of tempera- 

 ture within a few thousand feet. 



Thus we may confidently assume, in the majority of 

 cases of east wind, and especially when this wind is of 

 brief duration, local, or gentle, that a westerly wind flows 

 above it at no great distance from the surface of the 

 earth. Considering the perpetual rapid interchanges 

 (hardly to be called diffusion) going on in the atmosphere, 

 the lower wind must be largely mixed with air of a dif- 

 ferent condition derived from the westerly current. If a 

 cold dry east wind be permeated by patches and fila- 

 ments, however minute, of moister and warmer air, they 

 must be cooled by contact with the polar wind, and a 

 slight deposition of vapour may take place. Or the 

 countless invisible dust particles may, by increased radia- 

 tion towards space through a drier air, either cause a 

 slight deposition of moisture upon themselves or collect 

 still smaller particles together, as dust is known to collect 

 on cold surfaces in a warm air. If deposition of moisture 

 take place, the dryness of the air prevents the water 

 particles from growing to anything like the size of the 



