gg ANNALS NEW YORK ACADEMY OF SCIENCES 



Leaving for the present the curves of consecutive means and the con- 

 secutive maps, which will serve later for a better comprehension of the 

 dynamic phenomenon of tiie transformation of one map into another, I 

 will describe these annual departure maps and formulate questions and 

 .suppositions, just as 1 did in my former research into the variations of 

 temperature during the years 1891-1900.' 



The map for 1900 (Fig. 22) is peculiar in that the departures are 

 above the average all over the States. The quasinormal line follows 

 the Ciulf of Mexico from Corpus ('hristi over New Orleans towards 

 Tampa, Fla. At Key West the departure is — 0°.5 F. Porto Rico be- 

 longs to the pleionian area. In Bermuda the mean equals the ten-yearly 

 mean. In the northeast there are negative departures in New Brunswick. 

 The pleion has two centers, one in South Dakota, the other in Penn- 

 sylvania. The highest departures are -|-2°.2 in Valentine, Neb., and 

 -\-2°A in Harrisburg, Pa. The equideparture lines of -j- 1°.5, sur- 

 rounding these centers, are separated by a strip of lower departures ex- 

 tending from Lake Huron towards Oklahoma. ^^ 



The map of 1901 shows a pleion in the west, an antipleion in the east 

 and another pleion having its center south or southeast of Nova Scotia. 

 In Newfoundland the departure is + 0°.8 C. = + 1°.4 F. at St. Johns. 

 It is above -|- 2° in Nova Scotia. The highest departures of the western 

 pleion are + 2°. 2 (Bismarck, N. Dak., and Pierre, S. Dak.), whereas the 

 highest departure in 1900 was -f- 2°. 8. The center of the antipleion is 

 at Macon, where the departure is — 2°.0. Comparing the maps (Figs. 

 22 and 23), it looks as if the antipleion came from the Gulf of Mexico 

 pushing the eastern center of the pleion of 1900 from Pennsylvania over 

 Nova Scotia and New Foundland, while the western center remained 

 stationary. 



The map of 1902 (Fig. 24) is very different from that of 1901. The 

 positive as well as negative departures are smaller; the contrasts between 

 the pleions and antipleions are less accentuated than in 1901. The out- 

 lines of the areas affected by positive and negative departures are com- 

 plicated. The nuip gives the impression of representing intercrossing 

 waves. It is as if the western pleion of 1901 had been cut in two and 

 as if the two centers had been moved nj);u-t: the center of South Dakota 

 northwesterly into Manitoba and the center of Utah southeasterly to- 

 wards Texas. An antipleion separates these pleions and, perpendicularly 

 to this furrow, a ridge of positive departures extends from Louisiana to 

 Nova Scotia. The antipleion of the west also has the shape of a wave. 



=sThe departure of +0°.2 at Chicago is evidently incorrect. Tlie departure — 0°.l 

 of Fresno, Calif., Is Illiewlse in contradiction with the other Californian data. 



