% 
420 Transactions of the Boyal Society of South Africa. 
When it was raised to a Hne opposite the centre of the fan — 
10 inches from fan 890 feet per minute 
22 „ 876 
34 „ 760 
46 „ 512 
Evidently, therefore, the current of air thrown out by the fan is not 
a simple cylinder in shape, but spreads out, say, conewise from some point 
in the axis, or axis produced, of the fan. Hence it comes about that the 
loss by evaporation is so much less in the comparatively sheltered position 
of the cup nearest the fan. 
In further experiments the gauges were placed upon a shelf, the fan 
being placed upon a box beyond the end of the shelf at such a height that 
its axis was about 3 inches higher than the water-level of the gauges. 
By this arrangement the lower edge of the fan was slightly below the edge 
of the shelf. For the purpose of reading the velocity of the current of air 
the anemometer was placed with its axis nearly level with the water. 
The first set of experiments under these conditions gave the following 
mean results : — 
Current velocity 7 inches from fan 1,054 feet per minute 
21 „ 866 
35 „ 694 
,, 49 ,, 539 „ 
the average of the temperatures of the water at first being 79 "9°. And by 
dividing the separate experiments into two sets, according as the relative 
humidity of the air of the room was greater or less than the mean of all, 
we get the following : — 
Dry bulb 80-7° 78-3° 
Dew-point 33-9 40-7 
Humidity, per cent 18"5 26*3 
Final temperature of water in still air 76*4 75*9° 
,, „ 9 inches from fan 70*4 70-4 
23 „ 70-3 70-3 
37 „ 71-0 70-7 
51 „' 70-9 70-8 
Loss by evaporation in still air 17'3 gr. 16-0 gr. 
,, 9 inches from fan 110*7 84*5 
23 „ 110-7 84-2 
37 „ 91-8 73-3 
51 „ 77-1 61-8 
