ATMOSPHERE OF GREAT BRITAIN AND ON THE CONTINENT. 655 



is a large amount of vapour present, the air will be clear, with even a slight wet-bulb 

 depression if the number of particles be few, but it will be greatly hazed if the number 

 be great. If now the temperature of the same air be raised so that there is still the 

 same amount of vapour present, but a considerable wet-bulb depression, the vapour will 

 have a much less hazing effect. In other words, the hazing effect of the humidity is 

 proportional to the relative, not to the absolute, humidity. It should be remarked that 

 from some observations in Part I. it would appear that the absolute humidity has also 

 a slight effect, and that very hot days are hazed generally more than cold ones. This 

 may result from two causes : one the higher tension of the vapour enabling the dust 

 to condense more moisture, the other the irregular mixing of the highly heated air causing 

 a kind of hazing effect. 



We shall now examine the results of this investigation, as shown in these tables, on 

 the effect of the humidity on the transparency of the air, and compare them with the 

 conclusions we arrived at on this point, from observations made on the hazing effect of 

 atmospheric dust, in the paper already referred to. In that paper it is shown that the 

 clearness of the atmosphere, for each direction of wind, depends on its dryness, becoming 

 about 3 '7 times clearer when the wet-bulb depression was 8° than when it was 2°. That 

 is, for a given amount of impurity the clearness was nearly proportional to the wet-bulb 

 depression. If we examine Tables IX., X. and XL, we shall see that this conclusion is con- 

 firmed by the results of the present investigation. In the previous investigation we had 

 observations as low as 2° of wet-bulb depression, but in this one we have no tables under 4° ; 

 we have therefore not such a wide range of humidities to deal with. The mean wet-bulb 

 depression of all the observations in Table IX. is only about half the mean of those in Table 

 XL The value of the hazing effect will be inversely proportional to C. From this number 

 in the two tables we see that the damper air has a little more than double the hazing effect 

 of the drier. Perhaps this is more clearly shown in Table XIV., where it is seen that a 

 complete haze is produced in the damper air by little more than half the number of 

 particles required to produce the same effect in air dry enough to give twice the wet-bulb 

 depression. It was impossible to separate the observations in the tables rigorously 

 to the wet-bulb depressions at the time, as the amounts varied a little during the periods 

 selected. It may, however, be taken that most of the observations in Table IX. were not 

 much under 4°, while those in Table XL were about 8°, so that the observations in the 

 latter table will have fully twice the wet-bulb depression of those in the former. The 

 conclusion, therefore, pointed to by these tables is the same as that previously come to, 

 namely, that the transparency of the air increases with the dryness, and for a given 

 number of dust particles it is nearly but not quite proportional to the wet-bulb 

 depression. In the former paper we came to this conclusion from observations made 

 on the transparency of the air in winds from different directions at different humidities, 

 and in this investigation the same conclusion is arrived at by finding that when the 

 air gives twice the wet-bulb depression it requires nearly double the number of particles 

 to produce the same amount of haze as the damper air. Table XIV. shows the relation 



