October i, 1908J 



ATA TURE 



551 



the year, which had been supposed to be nearly complete 

 ai S or 9 kilometres altitude, does not exist, but, on the 

 contrary, sudden changes of temperature occur with the 

 passage of cyclones and anticyclones which would furnisli 

 to an observer in those regions the chief evidence of the 

 changes occurring at the surface. 



Causes oj the Isothermal Layer. — The summary of the 

 observed phenomena has led me to this conclusion, that 

 the cessation of the temperature diminution is due to the 

 fact that there is at these heights no considerable vertical 

 convection. 



The fact that one meets with layers of air thousands of 

 metres thick where the temperature increases and decreases 

 rapidly, and others where it is stationary, is incompatible 

 with the existence of motion of the air accompanied by 

 pressure variations, which always tend to produce a vertical 

 temperature gradient more or less near that for the 

 adiabatic state. It does not follow that the movement in 

 ihi- isothermal layer must be horizontal, but that it takes 

 place along the isobars without crossing these surfaces 

 nearly in the manner in which a body rolls on an inclined 

 plane. 



These ideas have been developed in several communica- 

 tions, in • particular at the Conference d'Airostation 

 scientifique at St. Petersburg in September, 1904. 



Dr. Shaw, in the absence of M. L. Teisserenc de Bort, 

 opened the discussion. He explained what was the main 

 feature of the phenomenon, and showed how it had been 

 corroborated by balloiis-sondes ascents made in England. 

 The temperature of the air decreases in the lower layers 

 on the average at 5° C. or 6° C. per kilometre up to a 

 height of about 10 kilometres. Above this height the 

 temperature ceases to fall rapidly and falls very slowly 

 indeed, or remains constant or in some cases increases. 

 It had been suggested that the phenomenon might be due 

 to a change in the composition of the air at great heights. 



M. L. Teisserenc de Bort had succeeded in sending up 

 balloons carrying vacuum tubes, which were opened and 

 re-sealed electrically at a height of 14 kilometres. The 

 samples of air so obtained were examined spectroscopically, 

 and the examination showed that there was no change in 

 the composition of the air sufficient to account for the 

 cessation of temperature diminution. 



Mr. Rotch said the only haUom-sondes which have been 

 sent up in America were those dispatched by him. Since. 

 1904 seventy-six rubber balloons have been launched from 

 St. Louis, and all but one have been recovered. The 

 majority of those which rose higher than eight miles 

 (12,870 metres) entered the stratum of relatively high 

 temperature. 



.All the ascents occurred after sunset, so that there can 

 be no question as to the effect of solar radiation. The 

 instruments used were of M. Teisserenc de Bort's construc- 

 tion, and were verified for low pressures and temperatures 

 before and after the ascents. The warm stratum, which 

 was not isothermal, but became warmer witli increased 

 height, was at its lowest level in summer, having a mean 

 minimum temperature of — 54°-6 C. at 12,000 metres. 

 During the autumn of 1907 the warm stratum of tempera- 

 ture was penetrated eight times, the mean minimum 

 temperature of — 6o°.5 C. occurring at 12,370 metres. 



The changes in the level of the minimum temperature 

 from day to day were large. Thus the minimum of 



— h7°.i C. at a height of 14,500 metres, on October 8, was 

 followed two days later by a descent of the minimum of 



— 62°.2 C. to 12,000 metres. In the first case, the tempera- 

 ture at the highest point reached, viz. 16,500 metres, was 



— 58°.! C, and in the second case, when 15,000 metres 

 was attained, — 56°.o C. On November 6 the minimum 

 temperature of — 52°.2 C. occurred at 9700 metres, but 

 the place of occurrence of the minimum of — 63°-i C. had 

 risen to 14,250 metres on November 8. The temperatures 

 at the highest points reached were — 50°.5 C. at 10,000 

 metres and — 6o°.2 C. at 15,380 metres respectively. 



These observations, made near latitude 35° N., show the 

 warm stratum to be at a distinctly higher level than in 

 northern Europe, whereas the results obtained by the ex- 

 pedition sent jointly by M. Teisserenc de Bort and the 

 author to explore the atmosphere over the tropical .Atlantic 



NO. 2031, VOL. 78] 



in 1906-7 show that it was there considerably higher. In 

 fact, the observations obtained over the equator up to 

 15,000 metres show no reversal of temperature, and a 

 lower temperature than exists at a corresponding height 

 in northern latitudes. 



Mr. Cave said that during the last week in July he was 

 able, by means of theodolites, to follow four balloons into 

 the isothermal layer. From these observations it appeared 

 that the wind velocity increased to a maximum just below 

 the isothermal zone, and decreased rapidly above. The 

 wind velocities were very high, and most of the balloons 

 went out to sea; one, sent up on July 28, was recovered. 

 From the record of the meteorograph it appears that the 

 isothermal layer was entered at 11,500 metres; the theo- 

 dolite observations indicated that this was the height of 

 the maximum wind velocity ; above this the velocity 

 dropped to eight miles per hour at 13,000 metres. 



Mr. W. H. Dines said that he knew there had been 

 some doubt expressed about the e.xistence of the isothermal 

 layer, and possibly there were still some wlio thought that 

 the results obtained were due to instrumental errors. Such 

 a view was now quite untenable, for about seventy ascents 

 had been made in the British Isles during the last eighteen 

 months, and the results entirely confirmed those previously 

 made on the Continent and in -America, although the 

 instruments used for recording the temperature were of a 

 totally different pattern. These ascents had mostly been 

 made at about the time of sunset, so that no possibility 

 of solar influence might be present, but in every case 

 (about sixty), when suflicient height had been reached, 

 the temperature gradient had become negligible or of 

 opposite sign. .After calibrating many instruments he was 

 convinced that the temperatures recorded were, with but 

 few exceptions, trustworthy within two or three degrees 

 centigrade. 



The results, however, were most remarkable, and it was 

 not surprising that doubts about their accuracy were 

 expressed. It had been found that over places only a few 

 hundred miles apart, and at the same time, the tempera- 

 tures might be widely different, and within the same week 

 and over the comparatively small area of the British Isles 

 differences of 30° C. had been recorded, namely, —40° C. 

 at 15,000 metres, at Limerick on July 27, —60° C., at 

 Pyrton Hill, Oxon., on the same date, and —09° C, at 

 Pyrton Hill on July 29 and again on July 30. Very 

 similar differences between Manchester, Ditcham Park, and 

 Pyrton Hill had been noted on previous occasions. 



The absence of any temperature gradient in the air is 

 definite proof of the absence of any vertical circulation, 

 but this alone did not present any difficulty. He (Mr. 

 Dines) had always thought that the vertical circulation 

 was chiefly due to the heat set free w'hen aqueous vapour 

 was condensed to water, and since it was known that the 

 relative humidity was small at great heights, it might 

 well be that above to or 12 kilometres there was no 

 aqueous vapour, and therefore no vertical circulation. 

 The difficulty was how large temperature differences could 

 e.xist at small distances apart without producing convec- 

 tion currents. In a mass of gas at rest under a con- 

 servative system of forces the isobaric or isothermal 

 surfaces must be coincident. In this case the tempera- 

 ture observations led to two contradictory results — they 

 showed that there was no circulation and also that the 

 isobaric and isothermal surfaces were not identical. At a 

 height of 15 kilometres a very small change of pressure 

 would produce a large adiabatic change of temperature, 

 but it was difficult to see how, with so small a mass of 

 air left above, changes of pressure could be produced. 

 The accelerations produced by curvilinear motion of the 

 air particles and by the effect of the earth's rotation an 

 a moving body appeared to be far too small for the pur- 

 pose. Was it possible that the upper air could carry a 

 sufficiently strong electric current to be influenced by the 

 earth's magnetic field, and so produce forces comparable 

 with gravity? Prof. Schuster had suggested some such 

 origin for the daily variation of the magnetic declination. 



Mr. Gold said that any explanation of the existence of 

 the isothermal layer must tal<e into consideration the 

 effect of atmospheric radiation. On the assumption that 

 the radiation per unit area from a layer of gas was pro- 

 portional to the mass of gas in the layer, and that the 



