(346 MB JOHN A1TKEN ON THE NUMBER OF DUST PARTICLES TN THE 



wind was blowing at high level, which caused the numbers to rise high at that station. 

 The impure S.E. wind does not seem to have penetrated the valleys on the 26th and 

 27th, the air being calm at low level and the dust remaining low on both these days. 

 On the 28th and 29th the circulation was again northerly and numbers low at both 

 stations, except for a short time. On the 30th the wind changed at both stations to 

 S.E. and E., and the numbers at both levels became very high on this and on the 

 following day. On the 2nd and 3rd, however, the wind at low level was again northerly, 

 and the numbers fell, but from the 4th to the 12th the wind again went south-easterly 

 at both stations, and on most of these days very high readings were obtained at high 

 and low levels. From the 13th to the 17th the wind was northerly and the numbers 

 low, except the abnormal readings already referred to. The readings for Ben Nevis on 

 the 14th and 15th are marked "Doubtful" by Mr Rankin, owing to there being a calm 

 on the Ben. A high reading was got at Ben Nevis on the 16th, — this was probably due 

 to wind being unsteady and southerly. The number at low level was not so high, but 

 the number is entered as doubtful, owing to the numbers being very irregular at the 

 hour of testing. On the 1 7th the wind was northerly and the numbers again low at both 

 stations, except the abnormal afternoon readings at low level. From the 18th to the 

 close of the observations the general circulation was south-westerly to north-westerly, 

 and the numbers at both statioris remained low. 



These three years' observations entirely confirm the conclusions arrived at in Part II. 

 regarding the relative purity of the winds from different directions. The impure 

 condition of the south-easterly winds is best marked in the 1893 observations, because 

 in that year there were more south-easterly winds than in any of the others. The effect 

 of the impure winds is well marked by the height of the dust curves on the 30th June, 

 1st and 2nd July, on the 6th, 7th and 8th, and again on the 10th and 11th July, the 

 other high parts of the curve on the 13th, 14th and 15th July being the abnormal 

 afternoon readings. 



If we examine the figures and curves for the different years to see if there are any 



indications of high readings in sunshine while the wind blows from other than the N.W. 



quarter, it is not easy to get a very definite answer ; but if there is any rise with other 



winds, it is not well marked and it certainly does not take place on anything like the 



same scale as with N.W. winds. The wind never seems to have settled long in any of 



the other directions, and the change of numbers is often due to change of wind. On 



some days it will be seen there was a tendency to rise on some sunny afternoons, but on 



other days, under similar conditions, there was no rise ; and again, a rise sometimes took 



place with clouded skies, so that on this point no very decided answer can be given, 



other than, if it does occur, the rise is slight and its presence is not manifest in the 



diagrams. 



Dust and Transparency at Kingairloch. 



In Parts I. and II. we have frequently referred to the relation which this investi- 

 gation has shown to exist between the number of dust particles in the air and its 



