MR. S. A. HILL ON THE WINDS OF NORTHERN INDIA., AND THEIR 
Table II.—Comparison of Mean Wind Velocities at Places on Lines running 
W.N.W. to E.S.E. 
Station. 
Wind velocity in 
miles per diem. 
January. 
May. 
July. 
October. 
Jeypore . 
92 
181 
154 
92 
Agra. 
82 
144 
121 
72 
Benares . 
68 
119 
111 
61 
Patna. 
50 
113 
87 
47 
Neemuch. 
168 
298 
284 
175 
Sutna. 
94 
194 
223 
103 
Hazaribagh. 
114 
208 
204 
120 
Malegaon. 
96 
257 
286 
111 
Akola. 
81 
206 
200 
78 
Raipur. 
58 
171 
275 
87 
The onlv instance in the Table, where the velocity during May increases in going 
eastwards, is between Sutna and Hazaribagh. The latter station is, however, the 
higher by nearly 1000 feet, but, being situated on a broad plateau, not an isolated 
peak, this would probably not make much difference. At Hazaribagh, however, we 
are on the border of the region over which these hot westerly winds prevail, the excess 
of observations in the direction of the westerly resultant being, in May, only 15 per 
cent., and the increase of velocity at that station is probably the effect of the frequent 
incursion of sea winds from the south-east. 
A suggestion regarding the origin of these hot winds may be derived from their 
excessive dryness, above referred to ; in fact, it was from this circumstance that I was 
first led to think of the key to their explanation, which I believe will be found in the 
following pages. This dryness is such that the relative humidity of the air over the 
Bundelkhand plateau, south of the Ganges, frequently falls in the middle of the 
afternoon to 5 per cent, or less. Indeed, the humidity deduced by the usual Tables, 
(founded on August’s formula with Regnault’s constants) is occasionally zero or 
apparently negative.* On the Gangetic plain it never falls quite so low, but in all 
* It must be recollected that the formula is founded on the convection theory of the psychrometer. 
The quantity to be deducted from the tabular pressure of saturated vapour, at the temperature of the 
wet bulb, in order to get t he vapour-pressure actually present, is ^—— - 1 -. The true value of the 
factor s/d, specific heat of air divided by relative density of vapour, is about ’38, but, to suit the formula 
to the usual condition of nearly still air, Regnault empirically increased it to '4S. Hence, during a 
high wind, the quantity to be deducted is too great when Regnault’s formula is employed, and if the 
actual humidity be very low the result may come out negative. 
