46 H. F. Blanford—Rainfall Frequency at Calcutta. [No. 1, 
actions being to a certain extent mutually antagonistic in their effects on 
the formation of rain, it would be impossible to foretell, in the absence of 
direct observations made in the cloud-forming strata, when and how these 
effects would mutually balance, and in what measure and at what epochs 
one or the other would become predominant. 
In the dry and hot season the diurnal course of rainfall variation is 
very different from the above. The diurnal epoch of minimum is not very 
distinctly indicated, but would appear to fall about sunrise. There is, how- 
ever, but little variation from midnight up to 9 or 10 a. M.; and after this 
only a slow rise up to 2 P. M., when the increase becomes more rapid. 
About two hours before sunset there is a sudden rise of about 50 per cent., 
and the hour of maximum raininess occurs between 7 and 8 Pp. m., the num- 
ber of recorded falls being then six times as great as at sunrise. ‘This very 
striking feature of the hot season is due to the well-known evening storms, 
commonly called Worth- Westers, which are closely analogous to the thunder- 
storms of the European summer ; and, whether as rain or hail-storms or 
simply as dust-storms, are characteristic of the dry season more or less in all 
parts of India. In Lower Bengal they are especially frequent, and the 
favouring conditions appear to be, the presence of a certain moderate 
supply of vapour brought by the coast winds, a high temperature at and 
near the ground surface, and a dry westerly wind from the interior of the 
country, which in Lower Bengal blows chiefly as an upper current from the 
plateau of Western Bengal, but during the hottest hours of the day, when it 
is at its greatest strength, produces a marked effect on the mean wind 
direction at Calcutta, and is sometimes felt there directly as a hot surface 
wind. It is when this wind slackens towards sunset, and that from the 
direction of the coast gains in prevalence, producing a calm in the interval, 
that North-Westers chiefly occur. They receive their name from the fact 
that the storm-cloud most commonly originates in the North-West, and 
advances or rather forms up with great rapidity from that direction, the 
formation of the nimbus overhead being speedily followed by violent gusts 
of wind from the same direction, which raise clouds of dust and occa- 
sionally exert pressures comparable with those of a.cyclone. Immediately 
before the onset of the storm, the barometer rises rapidly, sometimes more 
than 0:1 inch ; and, as Mr. Eliot has shown from a study of the autographic 
records of the Alipore Observatory, the subsequent fall coincides with the 
onset of the stormy winds, and a great and sudden fall of temperature and 
vapour pressure. Frequent casual observations of the motion of the dust 
and cloud margin in advance of these storms, have led me to conclude that ° 
the stormy wind which blows out from under the storm-cloud is a great 
horizontal eddy, the impulse of which is furnished by the air dragged down, 
