DYNAMIC METEOROLOGY. 419 



move ; the location where it is now raining, as shown by the map, is 

 usually near the present location of the center of disturbance and in ad- 

 vance of the center of low pressure due to the movement of wind. 

 Loomis finds that in the United States the rain area extends farther in 

 front than in the rear of the storm ; its mean extent is 500 miles in front, 

 and for every increase in this distance the velocity of the storm move- 

 ment is increased. He also finds that within the United States much 

 more rain fiills in front of the area of low pressure than in its rear; a 

 great rnin-fall in the rear retards the motion. 



(e) The barometric changes, being the direct result of horizontal 

 winds and only the indirect result of the changes in temperature, mois- 

 ture, vertical movements and condensations sum up as it were the re- 

 sult of all preceding disturbances. Loomis finds that the velocity of 

 progress of a storm is independent of the amount of the central depres- 

 sion of the barometer. The average velocity of movement of centers 

 of low pressure in the United States is about 20 miles per hour; the 

 pressure rises faster in the rear of fast-moving storms ; thus, if the 

 pressure in the rear rises 20 per cent, faster than the mean, the storm 

 moves 7 miles per hour faster ; if it rises 20 per cent, slower, the storm 

 moves 7 miles per hour slower. This is evidently a direct result of the 

 fact that the center of indraft up toward the clouds is propagated (not 

 moved bodily) forward faster than the lower winds can follow. 



(d) J. Elliot gives the following as the favorable conditions for the 

 formation and growth of cyclones in the Bay of Bengal, aiul these may, 

 with slight modifications, be applied to the Gulf of Mexico and the West 

 Indies: 



1. The invariable antecedents of cyclones are approximate uniformity 

 of pressure over and around the coast of the Bay of Bengal, light vari- 

 able Windsor calms over a considerable portion of the bay, and little or 

 no rain-fall over the coast region, as well as in the bay ; the weather is 

 therefore sultry and the sea smoother. The amount of aqueous vapor 

 accumulates and finally gives rise to peculiar sky effects. The immedi- 

 ate antecedent is heavy rain-fall concentrated over a portion of the bay 

 accompanied by a strong indraft, most marked from the Indian Ocean 

 at the entrance of the bay. This indraft fiom the Iiulian Ocean gives 

 rise to strong winds and heavy rains at the stations on the south and 

 west coasts of Ceylon. 



2. The source of the energy is the condensation of vapor and precipi- 

 tation of rain ; or the primary cause of cyclone formation is the i)rodu(;- 

 tion and ascent of a largequantity of \ ai)or, which is condensed with the 

 liberation of its latent heat over the ])lace of production, instead of be- 

 ing carried away to some distant region. (See .1. Eliot's rei)ort on the 

 Madras cyclone of May, 1877, and similar conclusions in his account of 

 the southwest monsoon storm of lS8o.) 



(2) rretlictivnof locdl storms. — The tornado is an exaggeration of the 

 mild thunder storm, especially in that the revolving cloud that is some- 



