36o 



SCIENCE. 



[Vol. XVII. No 438 



The temperature rises, and the region no doubt by degrees be- 

 comes warmer than the adjacent regions or the high air above it. 



Presently the time comes when the high cirrus rapidly thiclsens, 

 the cigar-shaped masses of cirro-stratvis or cirro-cumulus appear at 

 lower levels, and soon a tumultuous mass of cumulo-stratus clouds 

 develops far below. The latter frequently envelop the top of 

 Pike"s Peak, so they are from a mile to a mile and a half above 

 the plains. These clouds soon coalesce into a continuous sheet, 

 which develops fringes and festoons on its bottom and outer mar- 

 gin, and thus continues to descend. At this time there is usually 

 hut little surface wind. Sometimes the storm reaches this stage and 

 then clears up again. When the cloud has nearly reached the 

 plains there is a sudden rush of wind at the surface, bearing snow 

 or rain. Usually the storm is inaugurated by a succession of 

 squalls or hail-storms, — sometimes from the north. These squalls 

 are often electrical. As squall follows squall the festooned outer 

 border of the storm-cloud can be seen to enlarge laterally and sink 

 to lower levels. The surface temperature rapidly falls, and the 

 local storm-areas become connected by a great but not wholly 

 homogeneous cloud of precipitation, which rashes either south or 

 north over the Arkansas-Platte Divide. When the lower wind is 

 at first from the south, it usually swings around to the east, then 

 north-east, and finally north. Thi-i usually completes the storm, 

 but not seldom the cycle is repeated. Often a rush of north wind 

 for several hours is followed by a south wind, and then by a north 

 wind again. During all this time there is more or less precipita- 

 tion. Usually we are enveloped in the clouds of precipitation, but 

 often there are small rifts in these clouds, through which the up- 

 per air movements can be observed. In this manner I have ob- 

 served in almost every storm the higher clouds (mostly cirroid) 

 coming rapidly from the south- west for one or two days after the 

 under-rush of cold saturated air began transverse to their direction. 

 No matter whether the under- currents are going north, south, or 

 west, the storms usually continue till the upper cirrus comes from 

 the north; then the surface wind soon turns into the north, and 

 the storm clears off cold. The lower cloud from which the 

 precipitation occurs is seldom homogeneous in structure. In 

 almost all cases it consists of a series of squalls, the local storm- 

 areas being connected by stratus. This seems to be the general 

 law of the Great Basin also. After general storms I have seen, 

 both on the Wasatch and Rocky Mountains, great variations in 

 the depth of snow on plains and mesas that could not be accounted 

 for by differences of topography and altitude. In a recent rain- 

 storm that covered a large part of Colorado east of the mountains 

 these local storms were unusually well differentiated The general 

 storm began as a series of small thunder-storms, affording both 

 haU and rain, each electrical area showing massive black cumulus 

 clouds, which could easily be distinguished from the leaden and 

 rather homogeneous stratus which extended from one of the local 

 storm-areas to another. At one time three of these local storms 

 could be seen in different directions. The development of the 

 storm was 'signalized by a great fall in temperature. All the 

 clouds afforded rain, but the fall was much more rapid from the 

 electrical areas. 



On June 9 a storm occurred as follows. The upper cirrus had 

 been coming from the south-west for about three days, and hot 

 winds from the same quarter had prevailed much of the time at 

 the surface. During the night of June 8-9 there had been a heavy 

 dew, a rare occurrence on the plains. Early June 9 a series of 

 broad tracts of cirro-stratus formed along the eastern base of the 

 mountains. Their western edges were situated a little east of the 

 mountains. So near as tlie eye could estimate, these clouds occu- 

 pied the same position all day. The separate flocks and fibres 

 could be seen moving rather rapidly from the south-west. Evi- 

 dently new cloud- fibres were being formed at the western edges 

 of the cloud masses as fast as those already formed moved north- 

 eastward . 



During the afternoon there were numerous abortive attempts 

 at storms on the mountains. Just before sunset an electrical 

 storm began near the top of the Arkansas-Platte Divide. It was 

 narrow, perhaps five or ten miles wide, from east to west, but it 

 rapidly prolonged itself to a length of a hundred miles or more 

 from north to south. The most important facts about this storm 



are these; the south to south-west winds which had prevailed at 

 the surface during the day gave place to a violent cold wind from 

 the north at the moment the storm-cloud was formed; moreover, 

 this long, narrow storm was generated, as nearly as the eye could 

 estimate, along the exact north-and-south line where during the 

 earlier part of the day the formation of cloud had been going on 

 along the western edges of the cirro-stratus tracts. The north 

 wind raged at a high velocity for several hours. 



This was peculiarly a plains storm. To the west there appeared 

 but few clouds, and no storm of consequence was visiUe for at 

 least two hundred miles along the mountains. Even Pike's Peak, 

 who insists on dipping his head into every storm that comes into 

 this region, had for once to be content with a few .scattering clouds 

 about his shoulders, and looked on in utter helplessness. 



Summary. — (1) Over the mountain region there has been since 

 January a very great and persistent movement of air from the 

 south-west. (2) Unlike last summer, only a few days have at any 

 time elapsed before halos and sun-dogs have appeared near the 

 sun. They have invariably been followed by a rush of cold at the 

 surface, causing abundant precipitation. (3) During the general 

 storms of the winter and spring, the movement from the south- 

 west continues one to three days after the lower clouds of precipi- 

 tation have been formed in currents which travel hundreds of 

 miles hack and forth in directions transverse or even opposite to 

 the upper movement. (4) The movement of clouds in the high 

 atmosphere from the south-west is in most cases interrupted 

 toward the end of a storm by high currents from the north or 

 north-west ; but in a few cases the movement from the south-west 

 was either not interrupted or almost immediately resumed. (5) 

 The formation of the clouds of precipitation during the general 

 storms of winter and early spring proceeds from above down- 

 wards, and is usually aided by the development of local storms. 

 There is a sudden and often great fall of temperature at the time 

 the surface clouds of precipitation are formed, and this, too, 

 whether the surface clouds go north or south. The general situa- 

 tion, then, is this : before the breaking of the storm the lower 

 mile or two of the atmosphere consists of air from the south-west 

 of a relatively warm temperature, and generally containing con- 

 siderable moisture. The temperature is above the saturation 

 point. When the storm breaks xipon us the temperature suddenly 

 falls below the point of precipitation and there is a great rush of 

 cold air horizontally. 



Several facts deserve special notice in this connection. First, 

 the precipitation continues from five to forty or more miles after the 

 surface under-current which contains the cloud of precipitation 

 has passed the top of the Divide, hence while the air is being 

 warmed by condensation while descending from five hundred to 

 two thousand feet. Here is greater cooling than could take place 

 from rarefaction alone while the air was being forced to higher 

 elevations. Second, the cold under-current affords abtmdant pre- 

 cipitation, often for twenty-four to forty-eight hours, long enough 

 for three hjundred to twelve hundred miles of wind to pass. It is 

 therefore a moist wind. 



Now, no cold wind from either north or south could become 

 colder in sinking from higher to lower levels, nor could it in de- 

 scending to the earth become super-saturated with moisture, 

 whereas it contained no clouds of precipitation at higher levels. 

 We have therefore to look for the precipitated moisture in the 

 lower atmosphere, which in this case is relatively warm up to the 

 breaking of the storm. The most probable interpretation of the 

 facts would seem to he this; the cool under-currents which bring 

 the rain or snow consist mainly of the surface air, much of which 

 is fresh from the Pacific region. This surface air becomes mixed 

 witli considerable bodies of cold air, which descend from above 

 both at the fringed clouds and especially at the local storms. This 

 cold air would he dry, but would receive radiation from the sur- 

 rounding masses of warmer air, and thus cool them, and would 

 partly mix with them. This cooling goes on in spite of the latent 

 heat set free at the condensation of the vapor. 



It is not my present purpose to discuss the mechanism of these 

 movements whereby vast bodies of air leave an ocean warmed by 

 the Japan current and press eastward so persistently over a dry, 

 cold, and elevated plateau, and high range of mountains. That 



