22 



MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 



No. 10.-1878, September 3-8. 



No. 12.— 1878, September 24-29. 



September 24. September 25. 



Sun Juan I 29.91 NE. 6.. 29.87SE.il.. 29.95 SE.l.. 



Navasea ' 29. 95 NW. 12 29. 85 NNE. 18 29.83 E.8... 



Santiago deCul>a.. 29.96 N. (1) .j 29.88 NNE. a.i 29.85 NNE. (1) 



September 26. Srptembei 27. 



September 28. 



September 29. 



29. 92 SE. 6 . . . . 29. 91 SE. 4 . . . . 29. 88 SE. 4. 



29. 82 NW. 12 . . 29. 70 S. 25 29. 7H S. 15. 



29.79 N.(l).... 29.71 N\V.(1) . 29.72 SW.(l). 



No. 21.— 1879, October 10-15. 



San Juan 



Navassa 



Santiago de Cuba. 



King.ston 



Nassau 



October 10. 



Havana 29.98 ENE.IO 



29. 97 SE. 4 

 29. 87 NE. 5 

 29. 92 N. 7 . 

 30.11 calm 

 30.02 NE.(3). 



October 11. 



29.89SE. 7... 

 29. H2E. 10.. 

 29. 86 N. 2 ... . 

 30.(11 calm .. . 

 29.97 NE. (1). 

 29.90E. 6.... 



October 13. 



29.96SE.0 ... 

 29.76 E.20 ... 



29. 83 SE. 6 . . , 



30. 04 SE. 18 . 

 29. 91 NE. (3) , 

 29. 76 E. 12 ... 



October 14. 



29. 99 SE. . . 

 29. 92 SE. 15 . 

 29. 90 SE. 8 . . 

 30. 11 SE. 6 ... 

 29. H8 E. (2) . . 

 29. 64 E.20 .. 



October \'a 



30. 01 SE. 0. 

 29. 95 SE. 18. 

 29. 96 SE. 6. 

 30. 17 calm. 

 29.95 SE.(l). 

 29. 75 SSE. 18. 



It will be seen that at several of these statious the fluctuation of the barometer was small; the 

 winds were feeble, and their cyclonic character was indistinctly marked; but at those stations where 

 the tlnctuation of the barometer was greatest, there was a decided change in the direction of the 

 wind about the time of least pressure. In two-thirds of the cases the passage of the low center 

 was immediately preceded by a northerly wind, and in every case (but two) the passage of the low 

 center was immediately followed by a wind from the SE. These two cases occurred in storm No. 

 12, whose progress was almost exactly toward the north, and the low center was followed by a 

 wind from the south at Navassa and by a wind froiu the SW. at Santiago. In all cases, the SE. 

 wind showed indications of being a steady wind, resulting from causes of a more permanent 

 character than the storms here considered, for it generally continued for several days ; and at certain 

 stations, where the fluctuation of the barometer was sn)all, the wind blew from the SE. daring the 

 entire six days here represented. 



34. I have endeavored to determine the average direction of the wind for the three months, 

 August, September, and October (which months include nearly iill the tropical cyclones before 

 enumerated), tor that i)arl of the Atlantic Ocean in which these cyclones have most frequently 

 occurred. Table V is derived from Maury's Pilot Charts of the North Atlantic, and shows the 

 number of times the wind was observed to blow from the diflferetit points of the compass, in each 

 of the five-degree S(juares from latitude 15° to 25° N. and longitude .50 to 75° W. from Greenwich. 



The 11th, 22d, and 33d horizontal lines of the table show the sum of the observations of each 

 month for each of the sixteeu wind directions, and the last column of the table shows the average 

 direction of the wind computed from these numbers. 



35. Table Vt gives the results of all the observations collected by the U. S. Hydrographic 

 Oflice, including Maury's charts and the charts of the British Meteorological Oflice. The numbers 

 represent the percentage of winds from sixteeu points of the compass for each of the five-degree 

 squares. When the sum of the numbers in any horizontal line is less than 100 the difference 

 represents the cases of calms and variables. 



This table makes the average direction of the wind somewhat more northerly than Table V. 



