436 BEPORTS ON THE STATE OP SCIENCE. — 1919. 



It is probable that the larger particles of soot and dust are deposited com- 

 paratively near their source, and indeed there is some indication of this in the 

 quantity of matter deposited in the central areas of cities as compared with 

 the suburbs. 



Rela\tio7i of Deposit to Rainfall. 



There has }>een a suspicion for some time that a relation would be found 

 between the amount of deposited impurity and the quantity of rain. To 

 elucidate this question the deposit has been divided for several years into soluble 

 and insoluble matter, and these have been plotted for a group of stations against 

 rainfall. Curves for three years bearing on this point were exhibited.' 

 It was found that no regular relation could be discovered between the deposit 

 of insoluble matter and rainfall, whereas from the first there was indication 

 of some relationship between soluble deposit and rain. 



In the curve for the year ending March 31, 1917, the soluble and insoluble 

 matter are both shown plotted over the rainfall, and it will be seen that there is 

 a tendency for the soluble matter to vary with the rainfall, the insoluble varying 

 independently. 



This aspect was further dealt with by reducing the deposit at all stations 

 to a figure showing the quantity per 100 mm. of rain, and it was found that 

 such reduction did not produce any greater uniformity in the figures, but rather 

 the reverse, particularly for the insoluble constituents. 



In the curve for the year ending March 31, 1918, three groups of stations 

 are plotted : Group A is for sixteen stations scattered over the country ; Group B 

 for three London stations ; and C for nine Glasgow stations. In this case the 

 soluble matter only has been plotted, and it will be seen that there is still 

 the same indication of a direct relationship with the rainfall. 



The line ab has been drawn through the points on the curve which fairly 

 represent the relationship, and this is given by the equation shown on the 

 curve, i.e., calling soluble deposit in tons per square kilometre S and rainfall 

 in mm. E, the equation is 



S = 005S R + 2-5. 



For the year ending March 31, 1919, a similar curve was plotted, and a some- 

 what similar relationship appeared. The equation for this year was 



S = 0069 R + 2. 



Again, for a group of four years plotted together an equation 



S = 0081 R + 1-5 

 was found to hold. 



There is a general similarity between all these equations, hut they are not 

 identical ; ail of them indicate, in the absence of rain, that there is a certain 

 amount of soluble deposit, represented by the constant at the end of the equation, 

 and that the total amount is some function of the rainfall 'plus this constant. Of 

 course, the amount brought down must also depend upon the quantity present 

 in the air ; hence it is not to be expected that the relationship would be the 

 same for each year, and unless we could be sure that the conditions as to 

 quantity present were constant, we could not hope to establish a definite 

 relation.ship in this way. 



Suspended Impurities. 



The measurement of the suspended impurities has been the subject of 

 considerable investigation by the Committee, and many methods have been 

 experimented with in order to obtain some suitable means of estimation With- 

 out going through the various steps of the investigation, we shall simply consider 

 the methods as finally evolved up to the present. 



It was found that the quantity of (suspended matter present in the air was 

 so very small as to preclude the possibility of any method involving weighing. 

 We found that in Ijondon on an average day the suspended impurity varied 

 from about 15 to 30 mg. per 1,000 cub. ft. Taking the lower figure — that is, 

 15 mg. per 1,000 cub. ft. for London — ^the method used should be applicable 

 also to the country, where the quantity would be not more than about one- third 

 of this. It was concluded, therefore, that an apparatus to give a reasonably 

 accurate result by weighing would have to handle a volume of air of at least 



