444 



NATURE 



{March 13, 1890 



meteorology. The movement of the daily barometric 

 oscillations from east to west is only quasi-tidal, being 

 quite different from the manner in which the tides of the 

 ocean are propagated from place to place over the earth's 

 surface ; these oscillations being, undoubtedly, directly 

 generated by solar and terrestrial radiation in the regions 

 where they occur, and it is thus only that the striking 

 variations in the curves of restricted districts compara- 

 tively near each other are to be explained. These 

 peculiarities do not occur over the open sea. 



As illustrating these variations, reference is made to 

 the retardation of the time of occurrence of the morning 

 maximum, which is delayed as the year advances, the 

 latest retardation being in June ; and the curves of 14 

 stations are given, these stations being situated in the 

 middle and higher latitudes, and in localities which, while 

 strongly insular in character, are at the same time not 

 far from extensive tracts of land to eastward or south- 

 eastward. These barometric curves for June present a 

 graduated series, the two extremes being Culloden, where 

 the morning maximum occurs at 7 a.m., and Sitka, where 

 the same phase of pressure is delayed till 3 p.m., there 

 being thus eight hours between them. Another set of 

 curves is given from lower latitudes, showing the diurnal 

 variation in mid-ocean from the Challenger observations, 

 together with a series of land stations representing the 

 influence of a land surface in increasing the amount of 

 the variation, which reaches the maximum in the driest 

 climates. Latitude for latitude, the maximum daily varia- 

 tion occurs in such arid climates as Jacobabad on the Indus, 

 and the minimum over the anticyclonic regions of the 

 great oceans. At Jacobabad the variation from the 

 morning maximum to the afternoon minimum reaches 

 01 87 inch, whereas in the South Pacific it is o-036 inch, 

 and in the North Atlantic only 0014 inch. 



The following are some of the other types of barometric 

 curves discussed — the curves at high-level stations on 

 true peaks, and down the sides of the mountain ; the 

 curves in deep contracted valleys ; those in high latitudes 

 in the interior of continents where the morning minimum 

 disappears ; and those in high latitudes over the ocean 

 where the afternoon minimum disappears. In the two 

 last cases, the curve is reduced to a single maximum 

 and minimum, which as regards the times of occurrence 

 are the reverse of each other. 



The atmosphere over the open sea rests on a floor or 

 surface, subject to a diurnal range of temperature so 

 small as to render that temperature practically constant 

 both night and day; but notwithstanding this, the diurnal 

 oscillations of the barometer occur over the open sea, 

 equally as over the land surfaces of the globe. Hence 

 the vitally important conclusion is drawn that the diurnal 

 oscillations of the barometer are not caused by the 

 heating and cooling of the earth's surface by solar and 

 terrestrial radiation and by the effects following these 

 diurnal changes in the temperature of the surface, but 

 that they are primarily caused by the direct heating by 

 solar radiation and cooling by terrestrial radiation of the 

 molecules of the air and of its aqueous vapour, and the 

 changes consequent on that cooling. It follows that 

 these changes of temperature are instantly communicated 

 through the whole atmosphere, from its lowermost stratum 

 resting on the surface to the extreme limit of the at- 

 mosphere. There are important modifications of the 

 barometric curves affecting the amplitude and times of 

 occurrence of the principal phases of the phenomena, 

 over land surfaces, for example, which are superheated 

 during the day and cooled during the night according to 

 the amount of aqueous vapour present in the atmo- 

 sphere. But it is particularly insisted on that the baro- 

 metric oscillations themselves are independent of any 

 change in the temperature of the floor of the earth's 

 surface on which the atmosphere rests. It scarcely 

 requires to be added that these results of observation 



will necessitate the revision of all theories of the 

 diurnal oscillations of the barometer that have assumed 

 a diurnal change of the temperature of the sur- 

 face on which the atmosphere rests as a necessary 

 cause of these oscillations. The theory of the diurnal 

 oscillations of the barometer submitted by Mr. Buchan 

 may be thus stated : Assuming that aqueous vapour, in 

 its purely gaseous state, is as diathermanous as the 

 dry air of the atmosphere, it is considered that the 

 morning minimum of pressure is due to a reduc- 

 tion of tension brought about by a comparatively 

 sudden lowering of the temperature of the air itself by 

 terrestrial radiation through all its height, and by a 

 change of state of a portion of the aqueous vapour from 

 the gaseous to the liquid state by its deposition on the 

 dust particles of the air. The morning minimum is thus 

 due, not to any removal of the mass of air overhead, but 

 to a reduction of the tension by a lowering of the tem- 

 perature and change of state of a portion of the aqueous 

 vapour. 



As the heating of the air proceeds with the ascent of 

 the sun, evaporation takes place from the moist surfaces 

 of the dust particles, and tension is increased by the simple 

 change from the fluid to the gaseous state ; and as the dust 

 particles in the sun's rays rise in temperature above that of 

 the air-films in contact with them, the temperature of the 

 air is thereby increased, and with it the tension. Under 

 these conditions the barometer steadily rises with the 

 increasing tension to the morning maximum j and it is 

 to be noted that the rise of the barometer is not oc- 

 casioned by any accessions to the mass of air overhead, 

 but only to increasing temperature of the air itself and 

 change of state of a portion of its aqueous vapour. 



By and by an ascending current of the warm air sets 

 in, and pressure gradually falls as the mass of air over- 

 head is reduced by the ascending current flowing back as 

 an upper current to eastward — in other words, over the 

 section of the atmosphere to eastward whose temperature 

 has now fallen considerably lower than that of the region 

 from which the ascending current is rising ; and this 

 continues till pressure falls to the afternoon minimum. 



The back flow to eastward of the current, which has 

 ascended from the longitudes where pressure at the time 

 is at the minimum, increases pressure over the longitudes 

 where temperature is now rapidly falling, and this atmo- 

 spheric quasi-tidal movement brings about the evening 

 inaximum of pressure, which occurs from 9 p.m. to 

 midnight according to latitude and geographical position. 

 As the early hours of morning advance these contributions 

 through the upper currents become less and less, and 

 finally cease, and the effects of terrestrial radiation now 

 going forward again introduce the morning minimum as 

 already described. It is during the evening maximum 

 that the diurnal maximum of periods of lightning without 

 thunder and of the aurora take place , it being during 

 this phase of the pressure that the atmospheric conditions 

 result in an abundant increase of ice spicules in the upper 

 regions of the atmosphere, which thus serve as a screen 

 for the better presentation of any magneto-electric dis- 

 charges that may occur. 



It is interesting to note, in this connection, that the 

 amount of the diurnal barometric tide falls conspicuously 

 to the minimum, latitude for latitude, within the anti- 

 cyclonic regions of the great oceans, where, owing to the 

 descending currents which there prevail, deposition from 

 the aqueous vapour is less abundant on the dust particles. 



From a discussion of the whole of the two-hourly 

 observations of the wind made during the cruise, sorted 

 into those made over the open sea and those made near 

 land, it is shown that the velocity of the wind is greater 

 over the open sea than at or near land, the difterence 

 being from 4 to 5 miles per hour. The most important 

 result is that there is practically no diurnal variation in 

 the wind's velocity over the open sea. But as respects 



