168 WINDS OP THE NORTHERN HEMISPHERE. 



DEDUCTIONS AND REMARKS. 



1. Plates VII., VIII., and X. disclose a system of winds on each side of the Atlantic 

 Ocean possessing monsoon features. If we represent the mean annual tracks by 

 drawing straight lines from one extremity of each annual curve to the other, we 

 perceive that on the western side of the Atlantic, the actual track falls south of 

 these lines in the fore part of the year, and north of it in the latter part; and 

 that on the eastern side the curvature is generally in the opposite direction. Out 

 of thirty-five curves on the western side, in British America, the Eastern United 

 States, and the western half of the Atlantic, there are but two excejDtions; and both 

 of these are between the parallels of latitude 31° and 33°, just on the limit which 

 divides the equatorial winds from the westerly ones. 



On the eastei^n side of the Atlantic, there is a general similarity in the form of 

 the curves, yet by no means so great as on the western side. At sea, we perceive 

 it only between the parallels of latitude 15° and 40°; but on land, all the curves 

 show it, more or less, except that for St. Petersburg on the north, and that for 

 Home and Naples on the south. The opposite curvature of the latter, also that for 

 Madeira, and the two at sea, south of latitude 10°, will be adverted to hereafter. 



2. On the western side, the monsoon character of the winds is much more strongly 

 marked iiear the sea-coast than in the interior of the country. Thus, on Plate VIII., 

 the curvature is greater in the New England States (No. 28) than in the State of 

 New York (No. 30) or Pennsylvania (No. 32). Compare, also, the curve for Pom- 

 pey, in the interior of New York State (No. 29), with the curves east of it, all of 

 which are for places nearer the sea-coast; or No. 40, which is derived from obser- 

 vations taken mostly at Nashville, in Tennessee,' with Nos. 33, 34, 35, 36, 37, or 

 41 near the coast. No. 37 is remarkable — almost equal to the monsoons of India, 

 as may be seen by comparing it with the latter on Plate VII. In Ohio (Nos. 57, 

 58, and 59), the monsoon feature does not appear to exist at all, though there seems 

 to be slight traces of it still ftirther west (Nos. 61, 62, and 63). 



3. On the western side of the Atlantic, there appears to be considerable uniformity 

 in the time of the year when the curves cross the mean annual path, particularly 

 in the zone of westerly winds. Starting from the 1st of January, all the latter, 

 both on sea and land, fall to the right or south of the line that represents the mean 



Five years at Nashville, two at Chapel Hill, in North Carolina, and ten months at other places. 



