CONTRIBUTIONS TO METEOROLOGY. 55 



would have been fouud. In No. 17 the form of the isobars was quite siuuUir to those shown on 

 Plate II. 



From tlie last colninn in Table XXX we see that a considerable number of these storms were 

 immediately succeeded by au area of high pressure of unusual nuiyuitude. In thirfeen of the cases 

 the pressure exceeded 30.50 inches; in thirty-two of the cases the pressure was as great as 30.25 

 inches ; but in six of the cases it was as low as 30.12 inches. These facts seem to indicate that an 

 area of high i)ressure, immediately succeeding an area of low pressure, is favorable to the rajjid 

 progress of the latter, but a very high pressure is not essential to riii)id progress. 



79. We perceive that about four fifths of the storms included in Table XXXI occurred between 

 the jMississippi River and the Rocky Mountains, and they occurred most frequently in the 

 neighborhood of Fort Sully and Bismarck. This region, therefore, appears to be especially 

 favorable to the slow movement of areas of low pressure. A careful examination of the Signal 

 Service mai)S shows that in the cases which occurred in the region above mentioned a pressure 

 below 30 inches extended to a considerable distance westward, generally as far as the Pacific 

 Ocean, and the low center did not make much progress eastward until an area of increased 

 pressure, coming from the west or northwest, began to be felt on the east side of the Rocky 

 Mountains. Nos. 1, 2, 3, 7, 9, 12, 13, 15, 18, 19, 24, 25, 26, 27, 28, 29, 33, 35, 36, 37, 38, 42, 44, 

 45, 46, 49, 51, and 52 were apparently examples of this kind. 



In Nos. 5 and 11 there was an area of moderately high pressure prevailing at the time over 

 the Eocky ISIountains, but this high area remained sensibly stationary during the day in question. 

 As soon as the high center began to advance eastward the low center advanced also, and at about 

 the same rate. 



In se\eral instances the low areas appeared to have been filled up by a slowly increasing 

 pi-essure on the north side, until the depression was so inconsiderable that it could not be 

 satisfactorily traced. Nos. 8, 10, 20, 21, 22, 39, 40, 41, 43, and 48 were apparently of this kind. 



In some of the cases the low center vibrated to and fro within the limits of a few hundred 

 miles for two, three, or four days. At length an increasing pressure on the northwest side 

 either drove the low center eastward or filled up the low area so that it could no longer be 

 satisfactorily followed. Nos. 20, 35, 37, 41, 45, 46, and 51 were of this kind. 



Apparently the reason why these areas of low pressure lingered so long in the neighborhood 

 of Fort Sully and Bismarck was the absence of a sufficient pressure on the west side, and the 

 Eocky Mountains api)areutly formed the barrier, which ju'evented the air from flowing in freely 

 on the western side. 



SO. The amount of rain which accompanied these low areas was extremely small, and it is 

 surprising that the winds within them acquired so great velocity. The average amount of 

 rainfall in eight hours for the forty eases which occurred in the northwestern part of the 

 United States was 0.06 inch on the east side of the low center and 0.02 inch on the west side. 

 A part of this rain fell near the borders of the low area, where the pressure was but little less than 

 30 inches, and it probably had but little intiueuce upon the movement of the low center. In the 

 twenty-two cases which occurred nearest to Fort Sully there were only nine in which a drop of 

 rain fell at that station during the days in question, and much of the time the sky was reported 

 as either clear or fair. 



81. It seems probable that the direct heat of the sun, acting ui)on the dry surface of the 

 barren plains, between the meridian of 97° and the Eocky Mountains, supplied a large part of the 

 moving force which maintained the velocity of the winds. Of the forty cases which the table 

 enumerates for the northwestern ])art of the United States two occurred near the end of 

 March, three occurred in A{)ril, ten in May, three in June, six in July, five in August, nine 

 in September, and two in October. During five months, including the colder part of the year, no 

 case occurred, and at. the time of nearly all the cases here enumerated the heat was unusually 

 great. Table XXXII shows for three successive days the maximum temperatures observed at 

 Fort Sully at the time of the twenty-two cases which occurred nearest to that station. Column 1 

 shows the number of the case as recorded in Table XXXI; column 2 shows the maximum tem- 

 perature two days before the date in the table; column 3 shows the maximum temperature one 

 day before the given date, and column 4 shows the maximum temperature on the given day. 



