CUNTRIBUTIONS TO METEOROLOGY. 1 1 



Atlantic (-(uist, we liiul that tli(> aveia<;o direction oftlie loiij^est diameter of llie isobars is sensibly 

 the same for both regions. 



8. The irrej-nlaiit.v in tlie form oftlie isobars abont a low center appears to bo generally dne 

 to the unequal force with which the wind on dilferent sides presses inward towards th(> center of 

 the low area. The depression of the barometer in a low area is considered to l)edue mainly to the 

 detlectiiij; force arising- from the eartli's daily rotation u])on its axis, as will lie more fully explained 

 hereafter. 



If the wind jiressed in with equal force on all sides towards the low center, and there was no 

 disturbance from local causes, we might expect that the isobars would be exact circles, were it not 

 that the deflecting force arising from the earth'srotation increases with the latitude. It seems, then, 

 well nigh impossible that the isobars should ever be exact circles. The magnitude of the average 

 ditierence between rhe greatest and least diameters of the isobars, together with the marked 

 preference which the longer diameter shows for a particular i>osition in azimuth, indicates that the 

 form of the isobars is not wholly determined by an accidental difference between the velocities of 

 the wind on the ditt'erent sides of the low center. 



9. If we examine the cases in which the elongation of the isobars is greatest we may learn 

 something of the causes which produce this elongation, and which determine the position of the 

 longest diameter of the isobars. Plate 1 1 exhibits a case of this kind which occurred March 8, 1877, 

 at 7'' ou" A. M., when a low center was situated between two centers of high pressure not very 

 remote from each other. In cases of this kind, we generally find that the gradients are niuch the 

 steepest in the direction of the high centers, and hence tiiere results a (compression of the isobars 

 ill the direction of a line joining the high centers, and an elongation of the isobars in a direction 

 periieudicular to this line. Cases of this kind are of common occurrence in the United States. A 

 low area is almost invariably ibilowed by a high area, which is generally situated on its northwest 

 side; and tin? low area is generally preceded bj' a high area on its east or southeast side. Such 

 was the case iu the storms represented on Plates I and II. This position of the low center with 

 reference to the high areas causes the longest diameters of the isobars to incline in a direction 

 somewhat east of north. 



10. Over the Atlantic Ocean and also over Europe, areas of low pressure resemble the low 

 areas of the United States in their main features, but exhibit several points of ditierence which 

 ordinarily are pretty clearly marked. Plate III exhibits the isobars for a storm which prevailed 

 over the Atlantic Ocean February 5, 1870. The least diameter of this low area was -!,.380 English 

 miles, and the longest diameter was probably about 3,000 English miles. At the center of this low 

 area the barometer stood at 27.33 inches, and the gradient, where steepest, amounted to 0.71 inch 

 for one degree, and on the southeast side oftlie low center the average gradient up to the isobar 

 of 30 inches was 0.25 inch, and this is nearly double the gradient shown on Plate I. If w e compare 

 Plate III with Plate 1 we perceive that iu the Ibimer the isobars approach nearest to the figure of 

 a circle ; the low area has greater dimensions ; the depression of the barometer at the center is 

 greatest ; and the barometric gradient is the stee]iest. In each of these four particulars we 

 generally find a well-marked difference between the low areas of the Atlantic Ocean and those of 

 the United States. This difleience ajijiears from a comparison of Hoffmeyer's weather charts 

 (1874-1876) with those of the United Stales Signal Seivice. From a comparison of the isobars on 

 Uofluieyer's charts during a period of three years, I have obtained the follo-nung results. The 

 average ratio of the longest diameter of the isobars to the shortest is 1.70: 



Iu 54 Jper tcut. ot the whole number of cases, 

 17 " - - - 



54 ) per eeut. ot the whole number of cases, ( 1.5 

 17 > the ratio of the longest diameter to •? 2 

 1 S the shortest was greater than ( 3 



If we compare these results with those already given for the United States, we jierceive a 

 marked deficiency of very elongated low areas over the Atlantic Ocean. 



11. The longest diameter of the isobars over the Atlantic Ocean may be turned in any azimuth, 

 but (as in the United States) it is most frequently directed towards a i)oint somewhat east of north. 

 The lollowiug table shows the number of cases in a hundred in which the longest diameter of the 



