1888.] occurred at Dacca on April 7th, 1888, 207 



the line AB (PL XXVII., Fig. A.) than to those on the right. This was 

 also the direction in which it originally started, but the continued resis- 

 tance on the left no doubt helped to force it more and more to the right. 



The storm passed well to the right of the pucka house of Babu 

 Kailash Chandra Das, a Municipal Commissioner, which was not dis- 

 turbed, and, at the Elephant ghat below the Philkhana, the vortex was 

 actually down in the old bed of the river. The road from the Phil- 

 khana to the Elephant ghat here crosses the track of the tornado at a 

 right angle. On the west side of this road, on the old bank of the river, 

 is a small Hindu temple, and there stood a tall Jagarnath Car ; along 

 the west side of the road was a brick wall. The brick wall and the 

 Jagarnath Car were thrown down to the west, and the east corners of 

 the temple were torn away, and the bricks thrown to the west into the 

 compound, clearly showing that they had been caught by the retrograd- 

 ing lateral radius, and that the vortex was therefore to the right or south 

 of them as in the diagram, PI. XXVII., Fig. C. At this point of its 

 course, the Khedda Sergeant's house was beyond the influence of the 

 tornado, and the houses on either side of the road leading from Hazari- 

 bagh were undisturbed, but, on a spur of land lying to tiie south of 

 the old river bed, the branches of trees and the plantains were broken 

 and lying eastwards, as they had been caught by the advancing lateral 

 radius of the whirlwind. 



At this point of its course, the tornado bade fair to pass out into 

 the open maidan lying to the south of Nawabgunj, that is to say, in 

 the direction of least resistance, its vortex being already in the old river 

 bed. But it is evident that a great barometric depression had formed 

 to the north of its course. This was no doubt due to the constant 

 sucking action of the wind forces on the retrograding (left) lateral and 

 the posterior radii. It is clear, I think, that there must always be in- 

 creased barometric pressure to the front of a tornado and on its advanc- 

 ing radius, and a barometric depression outside the retrograding and 

 posterior radii and behind it in its track. However that may be, it is 

 evident that, immediately after passing the Elephant ghat, there was a 

 great barometric depression to the north of the whirlwind, for the 

 vortex suddenly moved to the left (north), and at the same time a great 

 hurricane from the north crashed through the trees, from a point to the 

 east of the Philkhana, and joined in the revel of the tornado, the vortex 

 of which was now near, if not on the main road through Nawabgunj. 



The evidences of this great indraught are quite distinct. As you 

 drive from the Lalbagh to the main gate of the Philkhana (Elephant 

 depot) by a road which is roughly parallel, but 600 yards to the north of, 

 the path of the tornado, there are all along signs of a high wind which 



