M I IT HOROLOGY. 



449 



il circulation >t' tin- atmosphere from the 

 r to the p<>le can onlv take place l>\ virtue o|' 



tlint tin- air \vhii-li overflow > from the equa- 

 i from the tropical heights actually loses it* 



( tin rely diminishes in temperature) and eon- 



- it cools. Palliation is the only clliciclit pn>- 

 0688 Capable of accounting fur tile great lo.ss of heat 

 that the atmosphere experiences. Although \\ e HIV 



not yet alilc to state the average umount and law of 

 radiation as dependent on temperature and pre-Mire, 

 yet it is ca^v to -ce that this must be known, and must 

 Miter into 1'iir equations of condition, before \ve can 

 fully account for the phenomena observed in the gen- 

 eral circulation of the atmosphere. The passage of a 

 .-.tcady atmo>j>hcne eiirreiit over a mountain raiiire, 

 iiiL,' ram and snow on the windward side, but, 

 linir as dry air on the leeward side, i_'i\cs rise 

 to s standing wave, such as may be seen above any 

 obstacle in a rapid river. Thus kinetic energy is con- 

 verted into static pressure; and therefore on the lee- 

 ward side, under the summit of the wave, down to 

 the earth's surface, there is a somewhat higher press- 

 ure than there would be incase no such current c\- 

 isted. The clear, descending air by its dryness has 

 also a slightly -.Teatcr den.-ity than before at the same 

 temperature and pressure, which adds somewhat to 

 the barometric excess. The coolness by radiation an- 

 nuls approximately the warming by compression, and 

 again gives increased density. These three factors, 

 therefore, conspire to increase the pressure at the 

 earth's surface, and this is again further increased 

 largely by the influence of the earth's rotation and 

 the southward How, as explained by Ferrel and llclm- 

 holtz. This is the best explanation I have yet been 

 able to frame of the formation of the high areas and 

 cold waves that move southeastward over Canada and 

 the United States. I believe that I first stated my 

 conclusion as to this mechanism when, in 187<>, 1 hud 

 occasion to urge the importance of maintaining me- 

 teorological stations in Alaska. The reports from 

 these stations as subsequently established, as well as 

 the international maps of the Signal Service, served 

 to confirm that view. The original current on the 

 Pacific side of the Rocky mountains may be due either 

 to a special cyclone, or to the general circulation of 

 the atmosphere, resulting in a standing wave whoso 

 .summit is over the Mackenzie river. When the cur- 

 rent temporarily ceases, the summit and the high area 

 die away ; and when it is strongest, the pressure is the 

 greatot. Generally the original current may be con 

 sidcivd as a temporary overflow from northern Sibe- 

 ria along the arctic circle to Alaska. 



Among the points to which the attention of 

 meteorological observers is called by a pamphlet 

 on " Meteorological Work for Agricultural Insti- 

 tutions," are problems of temperature, such as 

 the differences that occur in quiescent air be- 

 tween places that are close together, and the sub- 

 ject of protection from frosts; moisture in the 

 :iir. especially measurements of evaporation and 

 the transpiration of plants ; condensation and 

 precipitation of moisture, including an accurate 

 record of the amount of dew, and the density of 

 fog which should be recorded on some uniform 

 plan, such as the distance at which a slender 

 pole can be seen, and local weather predictions 

 independent of the daily weather charts. 



M. Angot's report on temperature observations 

 made during 1890, at three different heights on 

 the Eiffel Tower, shows that during the night 

 the temperature increases up to a mean height 

 of about 500 feet, then decreases, slowly at first, 

 and afterward more rapidly; at about 1,000 feet 

 tlic moan decrease of temperature is about 1-4 

 per 328 feet (100 metres). During the day t lie 

 temperature decreases constantly from tin- 

 ground upward ; in the lower strata the decrease 

 VOL. xxxn. 29 A 



is slower in winter than in .summer. In the 

 later season it amounts to 2'5 per 82b feet; but 

 almve 5t)0 feet the rule of decrease doe-, nut show 

 a decided annual variation ; the amount is about 

 1-6 per 828 feet. It is worthy of remark that 

 at a height of 984 feet C500 mctr<- in open air 

 the decrease of temperature is extremely rapid 

 both during the night and during the day, and 

 nearly approaches the theoretical value of the 

 law of the adiabatic expansion of ga 



From an examination of the st iti>tics of the 

 frosts of the present century M. A. Lanca-ier 

 finds that a cold winter has never been followed 

 by a very hot summer, and that in the great 

 majority of cases the summer following a severe 

 winter has been cold. The same opinion has 

 been expressed by Humboldt, in his "Cos-mos," 

 and by other writers. 



Clouds. The average heights of some of the 

 principal clouds at Mr. A. L. Notch's Blue Hill 

 Observatory were found by II. II. Clayton to be: 

 Nimbus, 412 metres; cumulus (base), 1,512 me- 

 tres ; false cirrus, 6,500 metres ; cirro-stratus, 9,652 

 metres; cirrus, 10,135 metres. The cumulus was 

 highest during the middle of the day. Obser- 

 vations at Upsala, Sweden, show that the base of 

 the cumulus increases in height until evening, 

 but neither of these conclusions applies to the 

 observations at Blue Hill. The average velocity 

 found for the cirrus (82 miles an hour) is twice as 

 great as that found at Upsala. The extreme 

 velocity was 133 miles an hour. A comparison 

 between wind and cloud velocity shows tnat be- 

 low 500 metres the wind velocity is less than the 

 cloud velocity. Above that the excess of the 

 cloud velocity increases up to 1,000 metres, 

 and then decreases again tilr about 1,700 me- 

 tres, after which it steadily increases. This 

 decrease between 1,000 and 1,700 metres is prob- 

 ably due to the fact that the clouds between 

 700 and 1,000 metres were mostly observed dur- 

 ing the morning, when the cumulus moves most 

 rapidly, and that the clouds between 1,000 and 

 1,700 metres were mostly observed during the 

 afternoon, when the cumulus moves slowest. 



Observations of the phenomena of the " lu- 

 minous clouds" were again made by Herr O. 

 Jesse, of the Royal Observatory, Berlin, in the 

 summer of 1890. They were visible between 

 May 26 and the beginning of August, or for 

 about four weeks before and four weeks after 

 the summer solstice. The statement that the 

 time when the phenomenon appears in the south- 

 ern hemisphere has a corresponding relation to 

 the summer solstice there, is confirmed. One 

 hundred and eighty photographs of them were 

 obtained of the clouds at German stations, of 

 which 75, having been secured at the same time 

 in at least two different places, were suitable for 

 the determination of height. Thirty of them, 

 having been taken at proper intervals at the 

 same place, may be used for determining the 

 speed and direction of the movements of the 

 clouds. The phenomenon was again leas bright 

 than it had been in the preceding year. Only 

 when the atmosphere was exceptionally trans- 

 parent was there an approach to the former 

 brilliancy. The aggregations of the ma- 

 particles seem t<> lie becoming thinner, as is also 

 to be perceived from the more distinct appear- 

 ance of certain relations of structure. It has 



