Range of Meteorological and Magnetic Phenomena. 63 



of Iceland*, Christiansand, on the west coast of Norway, 

 Greenwich, and the stations in the Netherlands. 



Much importance must be attached to the fact that at 

 Stykkisholm, on the west coast of Iceland, this minimum is 

 not at all perceptible, and hardly, if at all, at Brest. And 

 some importance also to the fact that it sets in first at the two 

 Scotch stations, next at Valencia and the Faroer. 



If you consider these facts I think that you will own that 

 here is an anomaly which must have its origin at the coast of 

 the west of Scotland, and probably at no great distance. For 

 if this origin were at a greater distance from the Scottish 

 coast, it would be hardly conceivable that its influence should 

 die out so rapidly on the two straight lines Thorshavn, 

 Berufjord, Stykkisholm, and Scotland, England, Brest 



Thus far 1 believe my conjecture is backed by what con- 

 stitutes a certain amount of proof. But the road, or rather 

 roads, which this minimum subsequently follows through 

 part of Europe is decidedly bewildering. Brussels shows it, 

 Paris most decidedly; so do Lyons, Montpellier, Triest, and 

 Klagonfurt; while, curiously enough, south and north of the 

 line uniting these stations, both in the German and the 

 Italian stations, I cannot find a trace of it. 



But in the North of Europe also there are decidedly some 

 traces of it. Both Copenhagen and Haparanda show them, 

 so do Baltischport and Kem on the western side of the 

 White Sea. In this part of Europe the boundary beyond 

 which this feature is not traceable is a line passing w r est of 

 Konigsberg, Petersburg, and Archangel. Some traces of 

 this minimum are also to be found in a group of stations of 

 which Warsaw is the northernmost. 



There is another fact which may be connected with the 

 strange distribution of this minimum. Nearly in the whole 

 of Russia and the North of Europe the highest point of the 

 whole curve has a strong tendency to occur in August, in 

 the very beginning for the eastern stations with a single 

 maximum, while those with a double summer maximum show 

 the highest one about the middle of August. This may, 

 for instance, be seen in the curves for Thorshavn, Berufjord, 

 and Christiansand. In the west and south of Europe, on the 

 contrary, the July maximum, on an average, is decidedly, 



* I owe these northern stations to the kindness of Dr. A. Paulsen, 

 Director of the Meteorological Observatory at Copenhagen. I owe thanks 

 to a great many more gentlemen for the kindness with which material, 

 even manuscript, was placed at my disposal ; but their names will come 

 more naturally when the time shall ultimately have come to publish the 

 whole of this investigation. 



