in the five great hurricanes of the 19th century. Descriptive 

 na-terial of a f^cneral. nature indicates that the levels of flooding 

 in these five rtorr?s pi'obably did not exceed the levels experi- 

 enced in 1?38 and 195U. Using the three recorded high-water 

 marks attained in recent hurricanes and the estimated flood levels 

 in five hurricane occurrences during the 19th century, an eleva- 

 tion-frequencj"- curve has been prepared. In preparing the ciorve, 

 account has been ta'cen of the historical hurricanes of 1635 and 

 I63B, their frequency beincr established on the basis of the period 

 from 1631? to 1955» and their elevations of flooding being asstaned 

 as higher than the 1938 flood level. 



VJith respect to seasonal variation of hurricane occurrences 

 in southern New England, the period of greatest activity extends 

 from early .Aueust to the end of October. However, records indicate 

 occurrences as early as the middle of June and as late as the 

 middle of December, 



16 



\ 



