METEOROLOGY. 545 



publications of the Army Signal Office, Hoffrneyer, Looinis, and the 

 Deutsche Seewarte. He seems not to have access to the similar charts 

 prepared by the present writer for the statistical atlas of the United 

 States Census Bureau and published in 1875. Koppen's charts show 

 seven principal centers through which storm tracks are most likely to 

 pass. Of these, two are in America on the parallel of 45°, and the cen- 

 tral over Lake Superior and New Brunswick, respectively. Three be- 

 long to the North Atlantic, near the parallel of 65°, and central, respect- 

 ively^ in Davis's Straits, southwest of Iceland and northwest of Norway. 

 The principal European center is central over Denmark and Southern 

 Sweden. For all these six centers thirty or more barometric minima 

 occur on the daily morning charts in the course of the year for each 

 square of 5° in latitude and 10° in longitude. [This statistical presenta- 

 tion of the frequency of storm centers seems imperfect in that it takes no 

 account of the movement of the storm center from one morning until 

 the next; it is simply a summation of what appears on the daily morn- 

 ing maps. The charts of the United States statistical atlas, on the con- 

 trary, were based upon actual storm tracks whose paths could be confi- 

 dently laid down by means of the three or more tri daily maps of the 

 Army Signal Service ; this atlas therefore presents the total frequency 

 for the whole year and the whole day. A new edition of these charts, 

 embodying all the work of inany years, has been prepared by Finley 

 and is now in press.] In some remarks on his charts Dr. Koppen states 

 that the minima which pass from America to the English Channel re- 

 quire about six days to travel from the 70th to the 10th meridian of 

 longitude, whereas the transatlantic storms require nine or ten days, 

 but the irregularity in the rate of storm movement is very great, both 

 by the ocean and the land. The greater part of the storms of America 

 pass over Greenland and Iceland, and daily weather telegrams from 

 islands and borders of the North Atlantic Ocean would afford to Euro- 

 pean meteorologists a practically useful synopsis of the condition of the 

 weather for the guidance of the navigators. Koppen also remarks that 

 the decided excess in the number of storms passing to the north of 

 Europe over those passing to the south not only affects the climate by 

 the characteristic warm and damp south and west winds, but is also 

 the foundation of the so-called Dove's law of the rotation of the winds 

 according to which, in Europe, they change most frequently in the order 

 east, south, west, north, or, as frequently expressed, shift with the sun, 

 namely, in the direct and not the backward order of rotation. In Green- 

 land, on the other hand, where the observer is located on the left-hand 

 side of the storm-path the change of wind is in the opposite direction, or 

 they are said to back against the sun or from the west through the south 

 and east. (Z. 0. G. 3L, xvn, p. 257.) 



J. Spindle* has published a collection of paths of typhoons in Chinese 

 and Japanese waters, compiled for the years 1858 to 1878. He finds 

 the turning-point in their parabolic paths about 30° north latitude. 

 H. Mis. 69 35 



