396 



NA TURE 



{Feb. 27, 1890 



The conclusion come to from the consideration of all 

 the observations is that the dust in the atmosphere begins 

 to condense vapour long before the air is cooled to the 

 dew-point. It seems probable that in all states of 

 humidity the dust has some moisture attached to it, and 

 that, as the humidity increases, the load of moisture 

 increases with it. i 



Another method of testing the condensing power of 

 dust for water-vapour is then described. In working this 

 method the dust is collected on a glass mirror, and its 

 condensing power is determined by placing the mirror 

 over a cell in which water is circulated, in the manner of 

 a Dines hygrometer. The temperature at which con- 

 densation takes place on the dust and on a cleaned part 

 of the glass is observed. The differenf^e in the two 

 readings gives the condensing power of the dust. One 

 kind of dust artificially prepared was found to condense 

 vapour just at the dew-point, while another condensed it 

 at a temperature 17° above the saturation-point. The 

 atmospheric dust was collected on the mirrors on the 

 same principle as that used in the thermic filter described 

 by the author in a previous paper, the dust being deposited 

 by difference of temperature, the necessary heat being 

 obtained by fixing the collecting mirrors on a window- 

 pane. Dust was also collected by allowing it to settle on 

 the plates. The atmospheric dust was found to condense 

 vapour at temperatures varying from i''"8 to 4'''5 above 

 the dew-point. This condensing power of dust explains 

 why glass such as that in windows, picture frames, &c., 

 often looks damp while the air is not saturated ; and in 

 part it explains why it is so necessary to keep electrical 

 apparatus free from dust, if we wish to have good 

 insulation. 



The constitution of haze is then considered. It is 

 shown that in many cases it is simply dust, on which there 

 seems to be always more or less moisture. But as what 

 is known as haze is generally seen in dry air, the effect is 

 principally due to dust. 



Some notes from the Rigi Kulm are given, where 

 " glories " and coloured clouds were seen. The condition of 

 the transparency of the lower air as seen from the top of 

 the mountain is discussed with the aid of the observations 

 made by observers at the lower levels. These observa- 

 tions were kindly supplied by M. Bilwiller, of the Swiss 

 Meteorological Office. The difference observed at the top 

 of the mountain in the transparency of the air in different 

 directions is shown to have been caused by a difference in 

 the humidity of the air in the different directions. The 

 variation in the number of particles on the top of the 

 mountain is considered, and it is shown that the great in- 

 crease in the number which took place on the sejondday 

 was probably due to the valley air being driven up the 

 slopes, reasons being given for this supposition. The 

 colouring in clouds, and on scenery at sunrise and sunset, 

 as seen from the tops of mountains and valleys, is re- 

 marked upon, and it is shown that there is reason for 

 supposing that when seen from the lower level the colours 

 will generally be the more brilliant and varied. 



The relation of the amount of dust to the barometric 

 distribution is then investigated — as to whether cyclonic 

 or anticy clonic areas have most dust in them. It is 

 shown that there is most dust in the anticyclonic areas. 

 The interpretation of this, however, is shown to be that 

 the amount of dust depends on the amount of wind at the 

 time, and as there is generally little wind in anticyclonic 

 areas, there is generally much dust. Diagrams are given 

 showing by means of curves the amount of dust on each 

 day, and also the velocity of the wind. The curves are 

 found to bear a close relation to each other — when the one 

 rises the other falls. The only exceptions to this are 

 when the stations where the observations were made are 

 not equally surrounded in all directions by sources of pollu- 

 tion. In that case, even with little wind, if it blows from 

 an unpolluted direction the amount of dust is not great. 



The increase in the dust particles which takes place 

 when the wind falls, seems to point to a probable increase 

 of the infection germs in the atmosphere when the weather 

 is calm. As, however, the conditions are not quite the 

 same, the organic germs being much larger than most of 

 the dust particles, and settling more quickly, it may be 

 as well, while accepting the suggestion, to refrain from 

 drawing any conclusion. 



In all the fogs tested, the amount of dust has been 

 found to be great. This is shown to be what might now 

 be expected from a consideration of the conditions under 

 which fogs are formed. One condition necessary for the 

 formation of a fog is that the air be calm. But when the 

 air is calm both dust and moisture tend to accumulate, 

 and the dust, by increasing the radiating power of the air, 

 soon lowers its temperature and causes it to condense 

 vapour on the dust and form a fog. The thickness of a 

 fog seems to depend in part on the amount of dust 

 present, as town fogs, apart from their greater blackness, 

 are also more dense than country ones. The greater 

 amount of dust in city air, by increasing its radiating 

 power, it is thought, may be the cause of the greater 

 frequency of fogs m town than in country air. 



At the end of the paper some relations are pointed out 

 between the amount of dust and the temperature at the 

 time the observations were made, showing that when 

 there was a large amount of dust there was also a high 

 temperature ; and some speculations are entered into as 

 to the effect of dust on climate. But it is at the same 

 time pointed out that the observations are far too few 

 and imperfect to form a foundation for any important 

 conclusi n on that subject. 



In a short appendix is given the result of some tests made 

 between January 23 and 29 of this year at Garelochead. 

 During the gale on Saturday, the 25th, the number was 

 rather under 1000 per cubic centimetre. On Monday, 

 though the wind was still high, the number fell to about 

 250 ; and on Tuesday, when the wind had fallen and 

 veered to th.- north, the number fell lower than had been 

 previously observed. The number varied from a little 

 over 100 to about 90 per cubic centimetre. On this day 

 the air was remarkable for its clearness, the sun was very 

 strong, and the evening set in with a sharp frost. 



John Aitken. 



P.S. — The author of the paper also showed at the same 

 meeting of the Society the apparatus which have just been 

 constructed from his designs for the Observatory on Ben 

 Nevis. The apparatus has been constructed by the aid 

 of a Government grant, obtained by the Council of the 

 Scottish Meteorological Society, for the purpose of carry- 

 ing on the investigation on the dust in the atmosphere at 

 the top of Ben Nevis. Two complete sets of apparatus 

 were shown. The one is the large laboratory form of the 

 dust-counter, and is to be fixed, m the meantime, in the 

 tower of the Observatory ; the air being taken in to it by 

 means of a pipe. The other is the small portable form of 

 instrument, to be used when the direction of the wmd is 

 such as to bring the smoke of the Observatory towards the 

 tower. This latter instrument has for a short time been 

 in the hands of Mr. Rankin, one of the Ben Nevis 

 observers, who has been practising with it near Edin- 

 burgh before beginning regular work at the Observatory. 



A UNIFORM SYSTEM OF RUSSIAN 

 TRANSLITERA TION. 



P to the present time no one system of transliterating 

 Russian names and titles into English has been 

 generally adopted. Some of those most interested in the 

 cataloguing and recording of Russian scientific literature 

 have therefore arranged the following scheme in order to 

 secure the general use of a system which will enable 



U 



