January q, 1902] 



NA TURE 



by the Emperor William, a very complete establishment 

 for the exploration of the upper air has been installed 

 there. The'first balloon acquired was the Humboldl, which 

 made its first voyage on Maich i, 1893. It was fired by 



by Berson, Assmann's ventilation thermometer, with 

 special mounting to avoid the car's interference, was 

 used, and every precaution was taken to make the circum- 

 stances otherwise comparable with. those under which 

 Glaisher made his memorable ascent from the Crystal 

 Palace. M. Berson carried these precautions even to the 

 length of making an ascent from the Crystal Palace itself 

 on September 15, 1898, while a simultaneous ascent was 

 made at Berlin, to make sure that the observed 

 differences were not due to climatic differences between 

 the air over Germany and over England. The day was 

 remarkably hot in England, the temperature being 

 10 C. above that of Berlin, but the zero isotherm was 

 found within a few hundred metres of 6000 metres at 

 each station. Further trial was made by reproducing 

 C.Iaisher's arrangement from his description and com- 

 ])aring its readings with the Berlin arrangements. 

 The results were only accepted as conclusive after careful 

 I onsideration of all the measurements of temperature 

 obtained from the numerous flights of manned and un- 

 manned balloons. In the end Berson was satisfied that 

 the difference of the observations was to be laid to the 

 account of instrumental errors in Glaisher's observations. 

 The comparison of Glaisher's and Berson's results for 

 the rate of fall of temperature with height, as given by 

 Assmann, is as follows : — 



an electric spark and destroyed on landing after its si.\th 

 voyage. Through the Emperor's generosity it was 

 replaced by the Plionix. Prof. Assmann, in the work 

 already referred to, gives some particulars of the 

 arrangements and results, but the subject is more 

 fully dealt with in a work consisting of three handsome 

 volumes published last autumn and entitled " Wissen- 

 schaftliche Luftfahrten ausgefiihrt vom Deutschen 

 V'erein zur Fbrderung der Luftschiffahrt in Berlin" 

 (Braunschweig : F. \'ieweg und Sohn). This contains 

 the account of seventy-five voyages and a number of 

 flights of unmanned balloons, together with the material 

 collected in the course of the expeditions, maps of 

 the regions traversed and a volume 

 of results. The meteorological in- 

 terest of these voyages is \ery great, 

 but the work is too elaborate for brief 

 summary. A single example may 

 give some idea of the possibilities of 

 investigation of this kind. < )n July 6, 

 7894, the /^/(li'w-v started from Berlin 

 at 6.32 p.m. and travelled north-west 

 to Jutland in nineteen hours ; the 

 unmanned balloon Cirrus was started 

 at the same time from the same 

 place and was carried by an upper 

 current to Bosnia. 



The primary meteorological ques- 

 tion to be determined by balloon 

 ascents is the rate of variation of tem- 

 perature with height. For some time 

 after GlaisheHs celebrated voyage of 

 1862, which concluded his balloon 

 work, the matter was regarded as 

 settled. The Berlin work has re- 

 opened it on the ground that the 

 thermometers used by Glaisher were 

 not sufficiently ventilated or screened 

 from radiation. Glaisher was aware 

 of the necessity for precautions, 

 although he discarded special appa- 

 ratus for ventilation, and his im- 

 mense experience in the use of thermometers might 

 enable him to obtain results, as, for e.xample, with 

 Daniell's hygrometer, where others of less skill and 

 experience would fail. In the balloon investigations 



NO. 1680, VOL. 65] 



3. — Neighbourhood of the Belle AI 



photographed from a b:\lloon. 



It will be noticed that whereas Glaisher's observations 

 lead to the conclusion that a constant temperature is in- 

 dicated at no very great height, Berson's numbers show 

 increasing rate of fall with height, so that the difference 



