(542 MR JOHN AITKEN ON THE NUMBER OF DUST PARTICLES IN THE 



of the sky as regards cloud at the hour the dust and other observations were taken. 

 These cloud observations have been entered in the last column of the Tables L, II. and 

 III., the amount of the sky covered with cloud being entered. The cloud observations 

 have also been entered in Diagrams I., II. and III., the condition of the sky being- 

 indicated in the diagrams by a black line underneath the base line of the dust curves. 

 The thickness of this line represents the amount of cloud at the time. When this black 

 line occupies the whole space up to the base line of the curves, it means that the sky was 

 entirely clouded ; the line is absent when the sky is cloudless ; and its thickness at any 

 point indicates the proportion of sky clouded. 



If we now examine the relation between the amount of cloud and the readings taken 

 in N.W. winds we shall at once see that on the days the sky was entirely clouded 

 the numbers showed no tendency to rise as the day advanced, but that whenever 

 there was clear sky — especially in the morning — the number of particles increased, and 

 generally increased very much in proportion to the amount of clear sky. On all the 

 days, with the exception of one, the numbers showed a decided tendency to become 

 great, and the increase was inversely proportional to the amount of cloud. The evident 

 conclusion is that these abnormal readings in N.W. winds are directly connected in 

 some way with the amount of sunshine, and that so long as the sky remains clouded 

 all over they never appear. The observations made in 1890 are found also to agree on 

 this point with those of the last three years. On examining the records for that year, 

 I find that the days on which abnormal readings were obtained were all days of more 

 or less sunshine, while the days of steady low numbers were clouded days. 



Under the black lines representing the amount of cloud in the diagrams will be 

 noticed short lines hanging from the cloud line. These lines represent that rain was 

 falling at the time the observations were being made. So far as can be gathered from an 

 examination of these lines, rain or the immediate effects of rain do not play any part in 

 producing the abnormal readings. It will be seen that, though there was rain on the 

 afternoon preceding the 9th July 1891, and also on the afternoon preceding the 14th 

 July 1893, both days of very abnormal readings, there was no rain from the 13th to the 

 15th July 1893, the other day of very abnormal readings. 



An investigation has also been made to see if there is any relation between these 

 abnormal readings and the conditions as represented by the isobars at the time, with the 

 view of finding out whether these high readings have any relation to cyclonic or 

 anticyclonic circulation. The observations made on the 26th June and 13th July 1893 

 have to be rejected, owing to the circulation on those days being doubtful, and it being 

 impossible to say whether it was cyclonic or anticyclonic. After these are rejected, there 

 are left in the tables for the last three years 37 observations taken when the wind was 

 north- westerly, and the result of the examination is : — 



On 23 days the numbers remained low all day: 



on 19 of these the readings were taken in cyclonic areas, 

 and on 4 days the readings were taken in anticyclonic areas. 



