254 REPORT—1899. 
January 12 ; March 26 ; ng 15 ; August 16, 17 ; September 16, 17, 22, 
23, 24, 25 ; November 2 2 1, 22, or thirteen nights es all. 
St. Elmo’s Fire was seen on February 6, , 12, 13, 26, 27, 285 
March 14 ; July 2 ; September 18, 28 ; erie e 2, 4, 3; December 7, 9, 
—seventeen times in all. 
Zodiacal Light, not observed. 
Thunder and lightning were reported June 21 ; J aly ie 2; November 3 ; 
December 26. Lightning only, EebeaeTye 2) - October 2 
Solar Halo, January ie 3; February 5, 24 March = ; May 2, 7, 17, 
24, 27; appe 16, (175 July 9; 2805 August 9, 11; September 23 ; 
November 2 22 ; December 8. 
Lunar Halo, August 6,7, 9; December 29, 30. 
Much time continues to be given to the discussion of the hourly 
observations of the two Observatories. The work of reducing and entering 
these observations for every day, side by side, so as to present a direct and 
easy comparison of the two, is far advanced, "being brought down to the 
end of 1897. The number of daily sheets finished is ii 710, and, as each 
sheet contains twenty-two columns, the laboriousness of the work may be 
in some degree appreciated. 
As explained in previous Reports, the rainfall, fog, thunder, lightning, 
halos, aurora, and other phenomena observed at 120 stations on each day, 
are entered on a map of Scotland for that day. The whole of these maps 
are now completed down to December 1898, the number of the maps 
amounting to 2,922. A beginning has been made to enter on these maps 
the gales and storms which have occurred at the sev enty lighthouses round 
the Scottish coasts. _ Care is taken to note the hour of commencement of 
each storm, so that a comparison may be made as to the commencement 
and violence of storms and the related forecasts issued by the Meteoro- 
logical Office in London. 
Storms which strike the Scottish coasts may be conveniently divided 
into these chief classes, viz. : storms which overspread the whole of Scot- 
land ; storms over the west coast only ; storms over the east coast only ; 
and storms more restricted as to the area they overspread, such as only 
from the Tweed to the Tay, from the Tay to the Moray Firth, over the 
Hebrides, and those confined to Orkney and Shetland. As _ regards. 
the intensity of storms, since this depends on the barometric gradients 
formed within the cyclones, these gradients will be specially examined in 
the relations they stand to the vertical gradients of pressure, temperature, 
and humidity formed in the stratum of the atmosphere between the top 
and foot of Ben Nevis. 
The Ben Nevis observations indicate that the great majority of 
cyclones show the winds both at the top and bottom of the mountain 
blowing vorticosely inwards upon the central area of the cyclones. But. 
no inconsiderable number of cyclones passing over Ben Nevis show that. 
while the winds at sea level blow inwards upon the cyclone, the winds at 
the top of the Ben blow outwards from the cyclones. Now the vertical 
gradients of pressure, temperature, and humidity, as disclosed by the two 
Observatories, open up very important lines of inquiry in the investi- 
gation of these different types of cyclones. 
Again, the frequent sudden changes of these vertical gradients suggest 
lines of inquiry of no less importance as to the relation of these changes 
to the manner of: distribution over the stations of the rain accompanying 
