CONTRIBUTIONS TO METEOROLOGY. 61 



90. If now wp compare the preceding results with those hei'etofore found for the West India 

 cyclones, while pursuing a westerly course, and for the cyclones of the liay of Bengal and China 

 Sea, for the same part of their course, we shall liave a view of the movement of storm areas under 

 a great variety of conditions. The average results for these five districts for the entire year are as 

 follows : 



Miles per hour. 



For tlie United States 28. 4 



Middle latitudes of the Atlantic Ocean 18. 



Europe 1(1. 7 



Neigbborbood of the West Indies 14. 7 



Bay of Bengal and China Sea 8.^^ 



The velocity here given for the West India cyclones is the mean of all the determinations in 

 Tables III and IV, and the velocity given for the Bay of Bengal and China Sea is the mean of 

 all the determinations in Tables VII and IX. 



Thus we see that the average rate of [)rogress of storm centers over the Atlantic Ocean is 

 about the same as over Europe and is double the rate of progress for the China Sea, and the rate 

 of progress for the United States is more than three times the rate for the China Sea. These 

 results are derived from so large a number of observations that they must be accepted as 

 substantially correct, and the.y demand a clear explanation. 



91. 1 have endeavored to determine how far these difl'erences may result from a ditference in 

 the mean velocity of the wind for these several districts. For thi.s purpose I determined the 

 average velocity of the wind for that portion of the United States within which the storm centers 

 are most frequently iouud, viz, that portion north of the parallel of 40° and east of the meridian 

 of lOOo from Greenwich. A slight examination of the observations sliows that at stations near the 

 Atlantic Ocean or near one of the Great Lakes the velocity of the wind is greater than at stations 

 in the interior of the country. I have therefore divided the observations into two groups, one 

 including the stations near the ocean or one of the Great Lakes, and called coast stations, the 

 other group including the remaining stations, which are called inland stations. Table XXXVI 

 shows for each month of the year the average monthly movement of the wind in miles for these 

 two classes of stations, according to the Annual Report of the Chief Signal Officer for 1883. 



Column 2 shows tor each station the number of years of observation, and at the bottom of 

 each group of stations is given the mean hourly velocity of the wind for that group. In the 

 succeeding line is given the mean between the velocities of the two groups; the next line shows 

 for each month the rate of jirogress of stoiiii centers as given in Table XXIX, and the last line 

 shows the ratio of the velocity of storm centers to the mean velocity of the wind. 



We see that this ratio is not the same for all months, but for that month in which the rivte 

 of progress of storms is greatest the ratio is sensibly the same as for that month in which the 

 rate is the least. This coincidence seems to indicate that the rate of progress of storms is in some 

 degree dependent upon the mean velocity of the wind, but the considerable inequalities in the 

 value of the ratio show that the rate of progress of storms cannot depend solely on the average 

 velocity of the wind. 



92. I next determined, as well as I was able with the means at my command, the average 

 velocity of the wind for that part of Europe within which storm centers are most frecpieiitly 

 found, viz, between the parallels of 50° and 6(1°. Table XXXVII shows the results which I 

 have obtained, the velocities being expressed in meters per .second, and the observations are 

 divided into two groups, as in Table XXXVl. 



Column 4 shows the number of years of observations from which the velocities are derived. 

 At the bottom of each group of stations is given the mean of the observations for that group- 

 The succeeding line shows the average between the results for the separate groups; the next line 

 shows the mean velocities expressed in miles per hour; the next line shows the average rate of 

 progress of .storm centers, as given in Table XXXIII; and the last line shows the ratio of the 

 numbers in the two preceding lines. These ratios are quite different froui those found for the 

 United States, and the corres]tondence between the rate of storm movements and the movements 

 of the wind is not as distinctly marked. Nevertheless some degree of corresi)ondence can be 

 detected, and it is noticeable that the change in the wind's mean velocity for the different mouths 



