1880.J and May 1888 in the Poah and Eohilkhand. 1G3 



Fressims at 9,000 feet 4 p. m. 



Station. 



30tli April. 



1st May. 



Station. 



30th April. 



1st May. 



Simla 



21-494" 



21-505'' 



Jeypore 



21-613" 



31-602" 



Ohakrata 



21-492 



21-499 



Meerat 



21-473 



21-346 



Mnssooreo 



21-516 



21-541 



DdM 



21-524 



21-378 



llaiiikliet 



21-505 



21-542 



Agra 



21-540 



31-614 



Pithoragarli 



21-550 



21-551 



Lneknow 



21-GOl 



21-614 



Katmandu 



21-579 



21-643 



Allahabad 



21-610 



21-663 



Ludhiaua 



21-425 



21-362 





21-585 



21-641 



Dohra 



21-509 



21-523 



Ghazipur 



21-536 



21-634 



Roorkoo 



21-4G3 



21-411 



Patna 



21-415 



21-563 



Baroilly 



21-541 



21-397 



Gya 



21-521 



21-614 



Goraklipnr 



21-425 



21-491 



Sutna 



21-625 



21-655 



Durbhanga 



21-317 



21-467 



Now gong 



21-649 



21-664 



Laliora 



21-287 



21-288 



Jhansi 



21-633 



21-640 



Sirsa 



21-5(3 



21-494 



Noomuch 



21-654 



21-651 



Sanibhar 



21-570 



21-560 



Ajmcro 



21-590 



21-587 



The cbarts iudicate very clearly that on both afternoons there "was at 

 the cloud-level a low pressure region overlying the Punjab and another 

 over Beliar, and that these two regions were connected by a long narrow 

 trough of low pressure extending over Rohilkhand and the north of 

 Oudh. The conditions were therefore both days such as were likely 

 to generate eddies by the interference of the easterly and westerly 

 currents along the low-pressure axis ; but at 4 p. m. the isobars were 

 more regular and the conditions likely to generate tornadoes better 

 marked on the 1st May than on the 30th, April. On the 30th, at this 

 hour in the aftei'noon, the low-pressure trough over Rohilkhand was 

 almost completely blocked by a region of relatively high pressure near 

 Bareilly, caused apparently by the upheaval of the pressure planes 

 there, on account of the local high temperature (99-8°). Towards sun- 

 set this obstruction doubtless disappeared with the fall of temperature 

 below, and then the distribution of pressure became similar to that 

 of the 1st May. At any rate, whatever may have been the reason, tlie 

 tornadoes of the 30th did not commence until near sunset, whilst on 

 the following day they began in the middle of the afternoon. 



The tornado tracks shown on the two high-level charts have a very 

 definite relation to the isobars. Tliey commenoed a little to the south of 

 the low p-essure axis, and approached slightly or ran parallel to it for 

 a certain distance, hut in no case succeeded in crossing it. These storms, 

 as regards the horizontal components of their motions, appear therefore 

 to have been eddies formed in the strong westerly currents of the lowest 



