December i6, 1915] 



NATURE 



443 



From these figures it will be seen that, while the 



niiount of suspended combustible matter (carbon and 



i) and mineral matter in the Attercliffe atmosphere 



from 25 to 3 times as great as in that of the other 



^tricts, that of dissolved matter is not more than 



15 times as great. 



These figures furnish no clue to the amount of 

 impurity in the atmosphere, except when rain is fall- 

 ini^; nor is it clear to what extent allowance must 

 1k^ made for the amount of soluble matter dissolved 

 by the same volume of rain when it falls in large 

 drops and rapidly in one district but slowly and in a 

 finely divided state in another. At the best, they 

 represent a minimum value ; they take no account of 

 pollution during rainless periods, and to convey an 

 pproximate idea of the foulness of town atmospheres 

 !i h monthly result would need to be multiplied by 

 ■ nie factor derived from the proportion of rainless 

 to rainy periods in the month. It is doubtful, therefore, 

 whether a comparison of results obtained at different 

 stations in the same observing area, or from different 

 ti^wns, under the Organising Committee's scheme of 

 rain-water investigation, can have much value until 



JULY AUG. SEPT. OCT. NOV. DEC. JAN. FEB MAR. APR. MAY JUNE 



in the Lancet to the end of September, 1915, are given 

 in the following table : — 



40- 



3 



s 



z 



something is known of the extent to which 

 the data are dependent on the duration of the 

 rain per month, or on the number of days 

 on which rain was collected, or on the rate 

 at which the rain came down, or on the direction and 

 velocity of the wind during the dry and wet periods 

 in the month. 



Of all these influences which affect any scientific 

 interpretation of the rain-water data, the only one 

 that can be tested is the rainfall per month. To ascer- 

 tain the influence which monthly variations in the 

 rainfall may exercise on the amount of impurities 

 washed out from the atmosphere by the falling rain, it 

 is desirable to select the case of a town like Malvern, 

 which is practically free from industrial smoke. As 

 the pollution of Malvern's atmosphere may be assigned 

 to one source — household fires — it might be expected to 

 increase with the seasonal fall in temperature : to be 

 greater in the winter than in the summer months. 

 The figures (metric tons per square kilometre) for the 

 p>eriod July i, 1914, to April 30, 1915,' as published 



' The Malvern results for May and June, which have .since been pub- 

 '■shed {Lancet, October g, p. 822 ; November 6, p. 1046), support the 

 ■iclusions drawn from the earlier data. 



NO. 2407, VOL. 96] 



While the figures recorded for insoluble matter, in- 

 cluding tar, are not easy to interpret, those for dis- 

 solved solids and for sulphate seem at first sight to 

 indicate an increase in ppllution with the advent of 

 winter. But they bear another complexion when 

 plotted against the measured monthly rainfall, as 

 will be clear from an inspection of diagram No. 3, 

 and the criticism may be ventured that, for a town 

 like Malvern, the Organising Committee's scheme of 

 chemical investigation might serve 

 better as a means of recording varia- 

 tions in rainfall than as a method 

 for determining the extent of varia- 

 tion of atmospheric pollution. Nor 

 is this criticism disarmed in the case 

 of a manufacturing town like Shef- 

 field, in which atmospheric pollution 

 is due partly to industrial smoke — 

 fairly constant in amount all the 

 year round — and to domestic smoke, 

 varying in amount according to the 

 season, for when a comparison is 

 made of the curves for the rainfall 

 and for the dissolved solids and the 

 sulphate, a similar correspondence 

 in shape can be traced. As an illus- 

 tration, the Attercliffe rainfall curve 

 is shown in F"ig. 2 above the curve 

 for dissolved matter for the same 

 district. 



It seems probable that the chem- 

 ical investigation of atmospheric 

 pollution by the rain-water method 

 could give comparative results 

 only if rain fell at a unform 

 rate, and either all day or at the 

 same period of each day, conditions 

 not attainable in nature. Notoriously, Sheffield's 

 atmosphere is at its worst on a calm day; 

 there were many days during last winter when the 

 University building, 250 ft. above the Don valley, was 

 enveloped in a smoke fog, but as no rain fell the pollu- 

 tion of the atmosphere, so evident to the senses, found 

 no record in the chemical analyses of the rain-water 

 collected on other days than these. Better results 

 might be obtained if a feasible method could be devised 

 for drawing air continuously and at a given rate 

 through water, but whether their value, as evidence 

 of atmospheric pollution, would be commensurate with 

 the cost is a question which experiment alone could 

 decide. If the British Association could see its way 

 to appoint a committee to investigate the comparative 

 merits of chemical and physical methods of studying 

 atmospheric pollution,* new processes might be devised 

 which would furnish evidence of avoidable pollution 

 (the main object of the inquiry into the smoke 

 nuisance) more readily and at less cost than the chem- 

 ical analysis of water used to wash impurity out of 

 the air. 



6 Since thispaper was read, a committee of the British Association has been 

 appointed to report on " Fuel Economy and Smoke Abatement " (see p. 305.) 



