248 



ANNUAL, REPORT SMITHSONIAN INSTITUTION, 19 3 9 



occurred in the vortex near its eastern inner margin, where the rotary 

 velocity and forward movement coincided. Thus the track of the 

 "center" passed west of New Haven and Hartford, while that of great- 

 est destruction passed northward 60 to 70 miles farther east. How 

 far north the eye lasted I do not know, but Northfield, Vt. — in a deep 

 valley, to be sure — had a light wind for 18 minutes as the direction 

 shifted from northeast to south; and on Mount Whiteface, N. Y., 



FiGCKB 3. — Weather map for September 21, 1938, 4 p. m., eastern standard time. This 

 weather map is based on the airways observations of the U. S. Weather Bureau in 

 the area, kindly sent to the author on request. Each long barb on the wind markers 

 Indicates two units on the Beaufort scale ; a large dot at the end indicates moderate to 

 heavy rainfall, a small one, light or intermittent rain. The closeness of the isobars 

 suggests the great velocity of the wind. On account of the high wind velocity and 

 great turbulence, the frontal boundaries between different air masses were not at all 

 sharp. The isobars and fronts on this map were drawn by Raymond Wexler. The corre- 

 sponding map by Pierce (op. cit., fig. 18) bulges the (northern) warm front westward 

 near the center and the eastern cold front northward over Rhode Island, the latter 

 apparently without justification. Pierce did not find a sufficient contrast south of the 

 center to Justify him in marking another cold front, as in Wexler's analysis. Neverthe- 

 less, he marked a trough of low pressure in this area, marking, apparently, a strong 

 secondary vortex about 100 miles SSW. of the main center and moving northward at 

 the same speed, causing marked changes in pressure, wind direction, and velocity from 

 New Jersey to central Connecticut. 



there was a 10-minute calm between hurricane winds from the North- 

 east and southwest. 



As the hurricane entered the trough, it traveled along the general 

 north-south front between the tropical air on the east and the modi- 

 fied polar air, 20° F. cooler, on the west. The rapid rotation began 

 to turn the fronts : the warm front, originally toward the northeast, 

 turned toward the north ; the cold front, originally toward the south, 

 turned toward the east as the storm passed over southern New Eng- 

 land. As the tropical air passed up the warm-front slope, its fairly 



