Junk iS, 1896J 



NA TURE 



149 



and burning the exhaust as fuel ; but wIkii in flight, the force of 

 the wind extinguished the tlame. 



It was clearly seen by the experiment iliat lor practical com- 

 mercial success of this class of steam apparatus an air condenser 

 is essential, as the weight of water used in a few minutes' run 



Fig. 2. — Same engine and boiler as Ijefore, attached to two inclined wing 

 planes, and tail. From tip to tip of wings ii feet; total surface of 

 wings and tail about 22 square feet. Total weight of whole apparatus 

 3J lb. Steam raised to 50 lb. per square inch, and started. Length 

 of flight about 100 yards on level ; maximum height during flight, about 

 20 feet. The propelling screw is seen in front and above the frame. 



equals the total weight of engine and boiler. Without a con- 

 denser, the length of flight must necessarily be limited to a very 

 few miles, and it would seem that the chief problem that workers 

 in this field have to solve, is to obtain an efficient and light 

 dry air condenser. Chari.es A. Parsons. 



Heaton Works, Newcastleon-Tyne. 



Experiments on Rontgen Rays. 



After examining the fluorescent and photographic action of 

 the rays (X„) emitted on strongly /;t'ii//«;,' a " focus tube," and 

 finding them different to the rays which have been hitherto 

 noticed (.\i), in that the relative transparency of flesh, bone, 

 aluminium and glass differs for the two kinds of rays, it seetned 

 desirable to try the eflTect of cooling the tube. Solid carbon 

 dioxide and ether, and then solid carbon dioxide alone were 

 employed, with the result that in both cases the fluorescence of 

 screen and tube very rapidly died out and the current apparently 

 failed to pass through the latter ; as the tube gradually grew 

 warm again, the fluorescence in it returned, not gradually, but 

 very suddenly, at a temperature not very far below that of the 

 room, the glass lighting up brilliantly, and the shadows of the 

 bones .showing on the screen with increasing distinctness. The 

 emission of X] rays reaching a maximum at about 12° C. (a 

 rough guess). On further heating X„ rays begin to be evolved, 

 judging from the increasing opacity of the flesh, whilst at the 

 same time the fluorescence excited on the screen grows rather 

 blighter, until the state recorded in my paper of June 4, is 

 reached. As the condition for the maximum of Xj rays probably 

 varies to a certain extent with the diflerent forms of tube, and 

 even different specimens of the same kind of tube, with the 

 degree of exhaustion, i&c. , it seems to follow from the.se experi- 

 ments that in some cases warming the lube slightly might be 

 useful in photographing the bones, whilst in others moderate 

 cooling would be better ; and from the accounts of various 

 operators such would seem to be the case, though, as will be 

 seen in the light of subsequent experiments, the particular 

 method of heating or cooling is an important factor in the result. 

 Solid carbon dioxide seemed very op.ii|ue to the rays when its 

 low den.sity is considered, but the efl'ect may have been partly 

 due to the frost condensed upon it from the air. 



Wishing to verify for myself the results of other experiment- 

 alists, I next examined the tube for its action on a well-insulated 

 brass plate, at first by connecting it with a sensitive electroscope, 

 and afterwards with one of the standard Thomson's quadrant 

 electrometers. As the experiments were all carried out with great 

 care, it may be worth while to state that, using thin aluminium 



xo. 1390 vol.. 54] 



plate well earthed by soldering to a gas-pipe to screen the plate 

 from all ordinary inductive action, the results were that the rays 

 after penetrating it," of aluminium, discharged the plate com- 

 pletely, whether electrified positively or negatively, leaving no 

 charge ; and that if the plate were uncharged to begin with, it 

 remained uncharged. This result is contrary to that observed 

 by some ; but from the delicacy of the instrument used, and the 

 great distance between mirror and scale, there is little room left 

 to doubt the accuracy of this result. 



I also tried the effect on a radiometer, making similar experi- 

 ments with a lanipblacked Leslie's cube at about 94° C. , and a 

 candle, to see how far radiant energy from these sources was 

 capable of penetrating the large aluminium screens used, in case 

 any heat action might interfere with or mask the effect, and 

 found, as others have, that when perfectly screened from all 

 other action, the radiometer is entirely unaffected by the 

 Rontgen rays, whether they be from a hot or a cold tube. The 

 X-rays are also without action on the blackened face of a 

 thermopile connected with a very sensitive galvanometer. 



When the ordinary inductive eflects of the tube were not 

 screened, I found that the space external to the tube was at a 

 high positive potential, which increased up to contact with the 

 glass of the tube, and was of the same sign all over the bulb part. 

 The cold bulb acted like a positively charged conductor whilst 

 the discharge was passing, and attracted the radiometer arms 

 just as any other charged body would, the effect lasting some 

 time after the discharge stopped. The unscreened cold bulb also 

 partly discharged the insulated plate, if the latter were strongly 

 positively charged, it more rapidly discharged it when negatively 

 charged, and left it with a positive charge of the same magnitude 

 to begin with as that left when the plate was strongly positively 

 charged, and lastly in the uncharged plate it developed a positive 

 charge, again of the same magnitude ; and this is what would 

 naturally follow from the combination of the Rontgen rays effect 

 with that of ordinary induction. 



The Xj rays seem to discharge a charged plate whether 

 positively or negatively charged, but of this I cannot at present 

 feel quite sure. Aluminium seems so far opaque to them that it is 

 doubtful whether, when a screen is u.sed, any of the rays get 

 through, and when a screen is not used, one cannot feel certain 

 that the eflect observed is due to the Xj rays either wholly or 

 partly. After heating the tube and turning on the current, the 

 whole tube is filled with a whitish, lavender-coloured light, 

 which comes to a focus on the glass behind the kathode, above 

 or below it ; and whilst in this state and giving little or no 

 fluorescence on the screen, the tube does not charge an un- 

 screened insulated plate, but it does rather rapidly drain it of a 

 previously communicated charge, either positive or negative. 

 As the tube cools the lavender light retreats more and more 

 from the kathode till at last it reaches the upper edge of the 

 rectangular anode, when the positive charge, mentioned before, 

 begins to be given to an uncharged insulated plate, but very 

 slowly ; as soon as the centre of the anode is bare of the 

 lavender discharge, the potential of the unscreened plate very 

 rapidly rises, and by the time the whole anode is clear of 

 the lavender light the normal positive charge is re-established 

 on the insulated plate. When the lavender glow retires from 

 the kathode, it leaves behind it a space full of almost invisible 

 light, which excites whitish green fluorescence on the glass of 

 the bulb, and it is during this time that most of the rays are 

 .\.,, as is shown by the fluorescent screen, and photographs of a 

 hand. I have not yet followed out the changes, relative and 

 absolute, of potential of the anode and kathode, but it would 

 seem almost certain that during the life of the lavender glow 

 the whole tube acts as a relatively good conductor up to the 

 time when the lavender glow cro.sses the centre of the anode — 

 or the point where the axis of the kathode mirror cuts the anode, 

 when there is an abrupt decrease in conductivity. I intend to 

 investigate this point as soon as possible, and meantime pass on 

 to what I imagine will prove of great use and interest to all 

 who work with these tubes. 



During seme experiments on the tube with an 18" Wimshurst 

 machine, I noticed that the X, rays, i.e. those showing the 

 bones best, seemed to be emitted or not according as a brush 

 discharge occurred on the wire leading to the anode of the tube, 

 close to the tube, or not ; and that the "electric wind" which 

 blows from the tube as from all charged bodies, seemed also to 

 vary in intensity with the X, ray flashes, the X, rays being most 

 copiously emitted apparently at the moment when the brush 

 discharge ceased or the wind moderated. I soon found that by 

 making a small brush on the wire near the anode, or drawing 



