282 



NATURE 



[July 23, 1896 



vacuum was fairly high. The resistance, however, was low. The 

 experiment was not really a successful one, for there arose some 

 trouble from either small leakages or vapour pressures. The 

 next form was Fig. 7. This gave really bright flashes on a 

 sensitive screen, and the resi-stance was low. Still another form 

 is that of Fig. S. The anode now is a hollow cylinder with one 

 end open. The total area of this anode is considerably greater 

 than that of Fig. 3, but the latter gave much the more jiowerful 

 radiation. It appears, therefore, that both the size and the shape 

 of the anode have an important influence on the radiating power 

 of the apparatus. 



The form which gave the most powerful radiation was that of 

 Fig. 3. This sent a powerful radiation through 3 feet of 

 solid timber. The rays on emerging were received on a 

 fluorescent screen made of about fifteen shillings' worth of 

 potassium-platino-cyanide, and the area of which was 36 

 square inches. This screen was considerably affected by the 

 rays after having traversed ihe 3 feet of timber, and gave 

 sufficient light to see very small objects in. But the hand, when 

 placed between the screen and the timber, cast no shadow 

 whatever. 



The next observation on tlie power of the radiation was to 

 take the screen to a distance of 30 feet from the source. At 

 this distance the bones of the hand could be seen, but not the 

 flesh. Even the bones cast no deep and sharp shadows at this 

 distance, not owing to lack of fluorescence — for the screen was 

 really bright — but owing probably to the turbidity of the 

 intervening 30 feet of air. 



The source was afterwards placed in position at one end of 

 the laboratory, and the screen taken to the opposite end, or 



facturing the various parts. The parts are easily enough made, 

 but manufacturers seldom care to attend to single articles except 

 at their own convenience. 



In conclusion it may be stated, though it is unnecessary to do 

 so, that the instrument just described owes its existence to the 

 teaching of Prof. Lodge. Benjamin Davies. 



THE ROBOROVSKY EXPEDITION. 



"D OBOROVSK\' and KozloflT, the two explorers who ac- 

 ■'■^ companied Prjevalsky in his last journeys, and for two 

 years continued his work of exploration of Central Asia, are 

 now back at St. Petersburg ; and they have returned literally 

 loaded with zoological, botanical, and geological collections, 

 together with the results of meteorological observations and 

 extensive surveys, as well as of numerous astronomical determina- 

 tions. The chief interest of the collections will certainly be 

 centred round the specimens of the wild horse (Eijuus 

 Przcwahkii) and wild camel which they have secured, as well 

 as in the great numbers of new species of plants and insects 

 which have been systematically collected by M. Kurilovitch, 

 who stayed at well-chosen .stations, while Koborovsky and 

 Kozloff, mostly accompanied by one man only, made the most 

 adventurous "excursions" — that is, journeys three and four 

 hundred miles long — into the unknosvn highlands of the Nan- 

 Shan. Great privations were endured by the two explorers 

 during these journeys, which were made without guides, during 

 the winter, when the thermometer stood at - 25" to - 35° C. , and 

 fearfid snovv-storms blew away the tent, while the sand borne 



rather to the end of the corridor leading to the laboratory — a 

 distance of 62 feet from the source. Even here the screen 

 fluoresced with some energy, but the hand was observed to cast 

 no perceptible .shadow. When this apparatus was working, 

 there was no place within the large room where the screen did 

 not fluoresce, the rays passing through masses of timber and 

 tables with surprising penetration. 



This experimental tube, however, with its rubber joints and 

 ebonite insulation, is not a lasting concern. Although a good 

 vacuum can be maintained for hours together when not in work, 

 it will not last more than half an hour or so when in continuous 

 use, after which tnore pumping is necessary. The current 

 evidently produces some change in the rubber and ebonite, dis- 

 engaging a gas which slowly destroys the vacuum. 



In the final instrument the joints are of mercury, and the in- 

 sulation of porcelain. The joints are first ground and polished, 

 and then flooded with mercury. Except the porcelain, the 

 entire apparatus can be made in the lathe, which is a great con- 

 sideration. A longitudinal section of the instrument is given in 

 Fig. 9. At the end of the j)orcelain is an arrangement for 

 focusing, which can be manipulated while the instrument is 

 working, .so that a point source can be obtained definitely and 

 easily by trial. This figure, which is reproduced from an early 

 picture, has a spherical anode. This should be replaced by a 

 circular plate resembling the anode of Fig. 3. 



This last form of instrument, though designed in the middle 

 of May, has not yet been built, owing to the delay in manu- 



NO. 1395, VOL. 54] 



by tlie wind sharply stung the travellers" frozen faces. These 

 lirivations, of which the Russian travellers speak so light- 

 heartedly, seem, however, to have ruined the health of the chief 

 of the expedition, Roborovsky. And when the expedition made 

 its last journey into the highlands which separate the Nan-Shan 

 highlands from the valley of the Yellow River, Roborovsky, 

 who had already endured pleurisy and erysipelas, was laid 

 down with a stroke of paralysis, which deprived him of the 

 use of all the right part of his body ; while the Tangut robbers, 

 who people these mountains, gathered in bands round the small 

 encampment. The expedition was already at the western foot 

 of the high snow-clad chain of peaks of the Alma-machin, 

 which rise on the left bank of the famous bending of the 

 N'ellow River ; a few days' journey only separated them from 

 the yellow waters of the Hoang-ho ; but in such conditions 

 they were compelled to return — the Tanguts immediately 

 taking advantage of the retreat to attack the caravan. They 

 were only repulsed, Roborovsky writes, after "a great loss in 

 men and horses.*' 



Notwithstanding this failure, even the purely geographical 

 results of the expedition, to say notliing of its scientific 

 collections, are very important. From Lake I.ssyk-Kul the 

 small ])arty proceeded eastwards, exploring the highlands and 

 the jjlateau of Yulduz, to Karashar, near Lake Denghiz, or 

 Bagrach-Kul, and thence to Turfan and the oasis of Hami. 

 Having now to cross the great desert of the Ilashun Gobi, 

 before reaching the Nan-Shans, Roborov.sky and Kozloft" took 



