96 Address. [Feb. 



Mr. Eliot has during the year also worked out in detail the history 

 of the formation of the cyclone which visited Balasore on May 26th 1887, 

 and caused the loss of the steamers Sir John Lawrence and Retriever 

 at the mouth of the Hooghly. The description of this cyclone is given 

 in a new publication of the Meteorological Department called " Cyclone 

 Memoirs," of which this forms the first part. This publication is 

 intended to contain short descriptions of cyclones in a form suitable 

 for issue to ship-captains, and to be issued as soon as possible after 

 the occurrence of the storms described. The first part also contains 

 some very valuable suggestions to sailors as to the best course for 

 them to pursue in the case of their meeting similar storms. 



Mr. F. Chambers has published during the year, in the Indian 

 Meteorological Memoirs, a full description of a cyclone which occurred 

 in the Arabian Sea in May 1881, in which he investigates very thoroughly 

 the connection between the wind direction in the storm and the position 

 of its centre, and the relation which the direction of the movement of 

 the storms bears to the direction of the normal monsoon wind blowing 

 at the time of its formation. 



Of the Meteorological papers printed in our Journal, the first is a 

 description, by Messrs. Crombie and Pedler, of the severe tornado at 

 Dacca, in April last, in which it appears to be proved that the storm 

 followed Meteorological conditions similar to those which usually precede 

 tornadoes in the United States, in which country the subject of tornado 

 formation has been worked out to the fullest extent by the Signal 

 Services, under Lieut. J. P. Finley, U. S. A. 



The second paper was by Mr. S. A. Hill, on "The Tornadoes and 

 Hailstorms of April and May 1888 in the Doab and Rohilkhand," and 

 in it he discusses the origin of such disturbances, and shows that the 

 tracks of the tornadoes have no apparent relation to the distribution 

 of temperature or pressure at sea level, but the disturbances probably 

 originate in the eddies in the stratum of the atmosphere where cloud 

 formation and precipitation commence, at a level of 9000 to 10,000 feet. 



Mr. Hill has also published two other papers ; one, on " The 

 Annual Variation of the Barometer in India," in the Met. Zeitung for 

 September 1888, in which he points out several features which are 

 still ignored by European meteorologists : notably (1) that the annual 

 variation at all stations north of 20° is accompanied by one of semi-annual 

 period; (2), that the annual variation diminishes in range as the eleva- 

 tion increases, and probably dies out or becomes reversed in direction at 

 14,000 or 15,000 feet ; (3) that the annual range is greater in Central 

 India and along the southern edge of the Gangetic plains than at the 

 foot of the Himalaya. This is very important, as it is the direct cause of 



