CONTRIBUTIONS TO METEOROLOGY. 



65 



Colmini 4 shows tlie number of years of observations employed, and the velocities are expressed 

 in miles per hour. The line marked mean siiows for each moiitii the average of tlie numbers for 

 the sever;il stations; the next liue shows the rate of progress of storm centers as derived from 

 Tables III and IV for the mouths August, September, and October. These are the oujy months 

 for which the tal)les iurnish more than a single observation, with the exception of June, for which 

 month there are three observations. In determining tlie average rate of progress of storm 

 centers I have rejected the velocity given for No. 4 of Table III, because it dift'ers widely from 

 all other velocities recorded in the two tables, and because it was derived from insufficient data. 

 The last line of the table shows the ratio of the numbers in the two preceding lines. 



its. I next endeavored to determine the mean velocity of the wind for that part of the 

 Atlantic Ocean in the neighborhood of the usual tracks of storm centers, and have adopted the 

 results contained in No. 3 of the Mittheilungen aus der Norddeutsche Seewarte, as exhibited in 

 the pamjihlet No. 5 (non-official), published by the British meteorological committee. The first 

 line of the following table presents a summary of these results for the four seasons of the year, 

 the force of the wind being estimated in units of Beaufort's scale (1-12). 



The second line shows the velocities denoted by Beaufort's numbers reduced to miles per 

 hour according to the table prepared by the British meteorological committee ; the third line 

 shows the average rate of progress of storms according to article 89, and the fourth line shows 

 the ratio of the numbers in the two i>receding lines. 



96. If we group together the results now obtained we shall have the following summary for 

 the average rate of progress of storm centers, the average velocity of the winds, and the ratio of 

 these two velocities. 



This table appears at first view to present a discouraging medley of anomalies, but some of the 

 anomalies may appear less formidable after a careful examination. It seems highly probable that 

 the slow progress of storm areas in Southern Asia is partly due to the small velocity of the winds 

 of that region. It is not obvious why storms should travel more rapidly near the West India 

 Islands than in the China Sea. It is possible that this anomaly may disappear when the mean 

 velocity of the wind has been determined by a more extensive series of observations. 



It seems to be established that over the North Atlantic Ocean the mean velocity of the wind is 

 considerably greater than the rate of progress of storms. This inequality is strikinglj- exhibited 

 in numerous cases. Over this ocean we frequently find an area of low pressure 2,000 miles or 

 more in diameter, with a pressure of about 28 inches at the center, attended by winds blowing with 

 hurricane violence (see Nos. 14, 29, 32, 41, and 44 of Table XXXV), while from day to day the 

 center of the low area makes little or no progress eastward, showing that the movement of the 

 atmosphere which corresponds to the average system of circulation is almost entirely interrupted 

 over this ocean. 



The most noticeable anomaly shown in the preceding table is, however, presented by the 

 United States, where the mean velocity of the wind is only one-third as great as over the Atlantic 

 S. Mis. 154 9 



