786 
DR. T. R. ROBINSON ON THE DETERMINATION OF 
with the same D at 28 inches distance the tangential velocity = 162, the radial 60; 
with D=22, distance =20 inches, tangential 122, radial = 8 2 with the same irre- 
gularity. It follows from this that the vortex motion is not, as I had supposed, in 
a regular spiral, but in a track of very complicated character, which it is not prac- 
ticable to define. At the same time I tried to verify with this instrument the law 
of inverse squares, which cannot easily be done with the balloons, as it was very 
troublesome to shift them. A scale was fixed on the platform, so that the meter 
could be exactly placed at any distance from the cross. Four distances were taken 
for each value of D, and the machine was whirled for 5 m before beginning to measure, 
and kept in motion till the four were finished. The irregularities are considerable, 
partly owing to the small size of the vanes, for even in real wind the current is far 
from uniform, narrow veins of great intensity being contiguous to others compara- 
tively feeble. The results are given below with their reduction to the cross by the 
factor 
( 108 + d ) 3 
( 108) 3 
Table II. 
D = 25, V = 15 55. 
D = 
30, V = 17-82. 
D = 
35, Y = 19-12. 
Dist. 
Meter. 
Reduced. 
Do. -M. 
Meter. 
Reduced. 
Do -M. 
Meter. 
Reduced. 
Do. -M. 
12-25 
233-0 
288-9 
+ 1-1 
274-0 
339-7 
- 34-6 
291-0 
360-7 
- 1-4 
20-25 
- 38-5 
194-5 
274-3 
- 13-5 
- 57-5 
216-5 
307-3 
+ 2-2 
- 40-5 
250-5 
353-2 
- 9-9 
28-25 
- 15-0 
179-5 
285-7 
- 2-1 
- 24-5 
192-0 
305-6 
+ 0-5 
- 1-5 
243-0 
386-8 
+ 24-7 
36-25 
- 10-0 
169-5 
302-4 
+ 14-6 
- 42-5 
149-5 
267-9 
- 37-2 
- 48-0 
195-0 
347-9 
- 14-2 
287-8 
o-oo 
305-1 
- 17-3 
362-1 
- 0-2 
The irregularities of the differences show that the errors of the formula are merely 
casual. 
(17.) Another question occurred to me: Does W go on increasing for any con- 
siderable length of time 1 To ascertain this a set were recorded during ten successive 
minutes as soon as the anemometer got into full motion ; the b’s are separated by the 
time of one revolution ; the circumstances were very favourable, and the results are 
a good specimen. 
