438 CAPTAIN W. DE W. ABNEY AND DR. T. E. THORPE ON PHOTOMETRIC 
The canvas should be previously cut to the sizes required, viz.:— 
3 pieces 28 r long. 
6 ,, 6' 6" long. 
2 „ 6' 6" cut diagonally. 
The canvas is fastened to the hut by means of fillets (as shown in the drawing). 
These should be 2" X V', cut into the following lengths : 18 pieces 7' 6" and 
32 pieces 6' 6", making in all 343 feet. The studs and rafters should be fastened 
by 3-inch French wire nails (say, 3 lbs.), and the fillets over the Willesden canvas 
by 1-inch French nails (say, 1 lb.); these are easier to drive by unskilled hands than 
the ordinary kind, and are less liable to split the wood. 
The climatic conditions at Fundium were wholly different from those with which 
we had to contend at Grenada. At the latter place our chief difficulties were due to 
the frequent rains and constant humidity. 
At Fundium, on the other hand, we had excessive dryness ; at noon there was 
often a difference of 20° F. between the wet and dry-bulb thermometers, and with the 
exception of the evening before the eclipse, there was not even a trace of dew at 
night. This excessive dryness, combined with the high temperature—it occasionally 
rose to 110° or 112° in the huts—was very trying to the woodwork. But our main 
trouble was the dust, which was excessively fine and light, and deposited itself as 
an impalpable powder over the apparatus, and greatly interfered with the proper 
running of the clocks. However, by covering the base of the huts with layers of 
the shells of some variety of Cardium, which were found in large numbers near the 
beach, and by frequently watering the ground, we to some extent kept down the 
cloud which every foot-fall otherwise raised. 
On the afternoon of the 14th the weather changed slightly for the worse. Up to 
that time the wind had been mainly in the east, and the sky almost unclouded. On 
that day the wind went round to the west, the temperature fell considerably and 
there was more cloud and haze in the sky and a certain amount of dew r in 
the evening. 
On the 16th, the day of the eclipse, the conditions w r ere slightly better as regards 
the amount of cloud, but the haze and general opalescence of the air was not less 
distinct. The morning was bright and clear, but the effects of the westerly winds 
were to be seen in the milky colour of the sky, and, as the sun rose higher, in the 
grey appearance of the heavens. During the whole period of the eclipse, however, 
the sky was cloudless, except for a few thin wispy cirri near the horizon, and 
although, of course, the temperature fell considerably as the phase of totality 
approached, there v r as not the slightest appearance of condensing moisture and there 
was no increase of cloud although a considerable increase of wind. In this respect 
the conditions were entirely different from those in the West Indies. At Grenada 
all the observations show, in the clearest manner, that the results were affected by 
