ON VARIATIONS IN DIRECTION AND MOTION OF THE AIR. 269 



years, thus sorting, as it were, all the winds. The mean amount of these is 

 then taken for each day, as represented in Table III., showing the daily 

 amount and duration of south wind during the month of January. 



From these Tables diagrams are drawn (see Plates VII. & VIII. ), repre- 

 senting to a scale the comparative number of miles of air that passed from 

 each point of the compass for every day in the year, taken from a mean of 

 seven years, the length of the line being in proportion to the length of the 

 current from each point on each particular day. An approximate meteor- 

 ological division of the year is thus obtained, having reference to the currents 

 that pass each station. 



When examining these diagrams it is important not to be misled by the 

 extent of a current that was due to one or two years being mistaken for the 

 type of a period of years. A space is therefore allotted under each day, 

 which is divided into seven parts, representing the seven years. One of these 

 divisions is then marked to indicate the year in which each wind occurred, 

 the size of the mark having reference to its duration on its particular year. 

 By these means the prevalence or otherwise of any wind at any period of the 

 year is at once rendered conspicuous. 



The diagrams explain this mode of illustration. Thus at Wrottesley 

 (Plate VII.*), commencing with the north-easterly and easterly winds, a re- 

 markable cessation is observable towards the end of January and beginning 

 of February, again at the end of June, also at the beginning of August and 

 September, and in the early part of December, while the amount towards 

 the end of February, and the prevalence in April, and more particularly in 

 May, often extending to the early part of June, is very clearly shown. 



The S. to the W .S.W. winds have their maximum in January, increasing 

 greatly towards the end of the month. They also prevail in the early part 

 of April, during the latter half of July, and through the month of August, 

 and more than any other current in September. 



The winds from the W. and N.W. prevail in March, towards the end of 

 May and in June, especially in the latter half of that month. During the 

 last four months of the year currents from these points are exceptional. 



On reference to the diagram representing the diurnal variations of the 

 wind at Liverpool (Plate VIII.), it will be observed that there is the same 

 cessation of the easterly and north-easterly winds at the end of January and 

 beginning of February as at Wrottesley, also at the end of June, &c. The 

 times of prevalence also occur about the same periods. The S.E. and S.S.E. 

 winds, however, correspond to the periods of the S. and S.S.W. at Wrot- 

 tesley, and the S.S.W. and S.W. at Birmingham. 



The W. and W.N.W. winds prevail with considerable regularity in March, 

 also at the end of April and beginning of May, again at the end of May, and 

 particularly at the end of June, extending into the early part of July, and 

 at the end of August and first portion of September. The diagrams show 

 better than any written description the features of each current. 



The last row of diagrams in each of the Plates VII. & VIII. represents 

 the mean amount of horizontal motion of the air, without reference to direc- 

 tion, for every day in the year — projected on the same scale as the separate 

 winds. From these it will be observed that a considerable amount of mo- 



* In Plate VII. a, the portions of Plate VII. which exhibit the duration of each wind 

 on every day throughout the seven years, are brought together in such close juxtaposi- 

 tion, as to give a comprehensive view of the periods of prevalence of each wind during 

 that term of years at Wrottesley. The points from N.N.E. to S.S.E. are repeated. 



