HURRICANES INTO NEW ENGLAND — BROOKS 245 



velocity was thereby considerably increased on the east side and 

 moderately increased on the west side; around the center the high, 

 rotary velocity continued. The mean wind velocity computed from 

 the combination of the general and the cyclonic gradients on the east 

 of the storm as its center passed across Massachusetts should have 

 been 102 miles an hour if no friction had reduced it. This 102-mile 

 velocity, expected from pressure gradient, was reduced by friction 

 to 80 or 90 miles an hour along exposed coasts and a general average 

 of 60 miles an hour inland. There were, however, occasional puffs 

 of extraordinary velocity, apparently due to the penetration of pro- 

 jectiles of freely moving air from a height of a few thousand feet 

 through the friction-hindered surface layer. These puffs had veloci- 

 ties of more than 100 miles an hour at the time of greatest intensity 

 and may have engendered certain eddies, whicli still further aug- 

 mented the local velocity and destructiveness. Such eddies were 

 surmised by Farrar, who described the effects of the tempest of Sep- 

 tember 23, 1815, in New England as follows : ^^ 



It was very violent at places separated by a considerable interval from each 

 other, while the intermediate region suffered much less. Its course through 

 forests in some instances was marked almost as definitely as where trees have 

 been cut down for a road. In these cases it appears to have been a moving 

 vortex and not the rushing forward of the great body of the atmosphere. 



Similar features occurred in the recent hurricane, but most of them 

 seem to have been caused by local topographic funneling or by local 

 weakness within a forest that permitted the entry of the destructive 

 wind into the forest along a lane down wind from the initial break. 



The same explanation for the violent gusts of a West Indian 

 hurricane is offered by Vazquez." Eddies, at least on a small scale 

 on both sides of the great gusts, were present in the storm of 1938, 

 for the wind direction during the passage of each gust varied con- 

 siderably. An eyewitness told me that, as the gusts came across a 

 meadow, eddies could be seen plucking and wliirling grass into the 

 air. Spray was picked up off the water in the same fashion. There 

 seems to be no evidence of tornadoes in this storm, such as have 

 been observed in one hurricane in Florida and in two or three at 

 Charleston, S. C. No incipient funnel clouds were observed from 

 Blue Hill, though the lowest cloud layer was visible through the mostly 

 rainless air for a distance of several miles. The turbulently mixed 



1" Mem. Amer. Acad. Arts Sci., vol. 4. Quoted by Henry Piddlngton in The sailor's 

 hornbook for the law of storms, 2d ed., p. 3, Loudon, 1851, and by Channing, loc. cit, 

 p. 5. 



" Vazquez, P. E., Naturaleza de las rachas cyclonicas In Nueva 0rientaci6n en los 

 estudios cicl6nicos. Obs. Col. Montserrat, Cienfuegos, Cuba. pp. 58-66, 3 figs., 1940 ; also, 

 not quite so complete, in Bol. Observatorio Nacl. [de Cuba], ser. 3, vol. 1, pp. 113-127, 

 1936. Vazquez' diagram of "tornados" in the cyclone is reproduced by P. J. Darling- 

 ton, Jr., in Quart. Rev. Biol., vol. 13, p. 280, 1938. 



