92 ANNUAL RECORD OF SCIENCE AND INDUSTRY. 



coming up the Bay of Bengal in the months of May and June, 

 with fresh southwesterly gales, accompanied by torrents of 

 rain and falling barometer, should not press northward until 

 the weather improves, and the barometer rises steadily." 

 And again: "In order to find the bearing of the centre of 

 the cyclone, lay to and watch the barometer and wind, stand 

 with your face to the wind, and measure round to your right- 

 hand side about ten points." (In the southern hemisphere 

 the ten points should be measured round to the left.) Daily 

 telegraphic weather reports are considered to be desirable 

 from six points, mentioned in the report; and signals are sug- 

 gested of three kinds first, a cautionary signal, indicating 

 that bad weather is probable ; second, a warning signal, indi- 

 cating that a cyclone is probably in the course of formation 

 in the bay ; third, a storm signal, indicating that a vortex has 

 been formed, and is probably approaching. The telegraphic 

 connection with the observing stations has been hitherto 

 quite uncertain, especially in stormy weather, when most 

 needed a matter which seems to have been remedied, weath- 

 er reports now having the priority over all other messages 

 in the telegraph office. Report of W. G. Wilson to the Gov- 

 ernment of Bengal. 



THE WINDS OF NORTHERN INDIA. 



Mr. Blandford, of Calcutta, lias presented a paper to the 

 Royal Society on the winds of Northern India, his intention 

 being to trace out the origin and causes of the normal wind 

 currents, and their annual variations, so far as can be discov- 

 ered. Among other things, the author states that the vapor 

 in the atmosphere indirectly greatly influences the baromet- 

 ric pressure by carrying heat from the lower to the upper 

 strata, and by arresting solar and terrestrial radiation, thus 

 equalizing the temperatures of the column of air; but that 

 the influence of vapor in changing the density of the atmos- 

 phere, by reason of the displacement of the heavier air par- 

 ticles, is relatively small, and in some cases unimportant. 

 In general terms, the changes of temperature are the princi- 

 pal causes of the variation of pressure. Inquiry is made as 

 to whether any dynamic heating or cooling of the air can be 

 traced, as is maintained by some must occur, when winds 

 descend to a lower or rise to a higher level ; but Blandford 



