326 



NATURE 



[August 2, 1900 



on the glass, and a photograph of the spark was taken. It was 

 found that an explosion occurred at each change in direction of 

 the spark, or at each fork in it. The two effects are shown in 

 Fig. 4 and Fig. 5 ; and you will see that even the sinuosities are 

 reproduced by a rent in the paper. In the case of a thunder- 

 storm, may not the peculiar rolling of the thunder be due to the 

 successive explosions along the path of a single discharge some 

 hundreds of feet apart ? 



But I will not dwell longer upon the fascinating study of 

 lightning in a laboratory ; for I wish to call your attention to 

 larger fields of inquiry which the possession of this great 

 battery opens. One of the most promising is that of 

 .spectrum analysis. In connection with the bktlery I have three 

 hundred glass plate condensers, one-eighth of an inch thick, and 

 about ten by eighteen inches coated surface ; this condenser is 

 charged in multiple to a potential of twenty thousand volts. 

 The glass of the thickness of one-eighth of an inch stands this 

 stress ; but I can not use my full voltage of forty thousand, 

 for the glass plates are immediately pierced. To utilise this 

 voltage it will be necessary to employ plates a quarter of an 

 inch in thickness. This is an interesting proof of the large 



Fig 



surface density furnished by the battery. The noise of the dis- 

 charge from this condenser is like the report of a pistol. 



Here is an example of its great heating effect. An iron wire 

 was stretched across the spark terminals. This was deflagrated 

 (Fig. 6), while at the same time a spark passed between the 

 terminals. The surrounding air was filled with the scintillating 

 sparks of iron. This shows that it is not impossible that sparks 

 may be formed inside a metallic cage or enclosure ; for we can 

 conceive of such an enclosure as a multiple circuit around a 

 spark gap. 



By means of the discharge of these condensers charged to a 

 difference of potential of forty thousand volts, I can produce 

 probably the highest degree of instantaneous temperature which 

 has been reached. I have been obtaining instantaneous photo- 

 graphs of the spectra of gases and of the vapour of metals. One 

 discharge, with a Browning direct vision spectroscope, will give 

 a photograph of the spectra of hydrogen, and ten or twelve dis- 

 charges are sufficient when a short focus grating is employed with 

 a fairly fine slit. I find it desirable to use a peculiar end-on 

 tube for the study of hydrogen. It is of the nature of a Crobkes' 

 tube, one end being blown into a very thin bulb ; this tube can 

 be heated to a very high temperature during the process of 

 exhaustion to drive out the water vapour. 



NO. 1605, VOL. 62] 



I have thus submitted hydrogen to a higher temperature than 

 it has been possible to reach before, and the study of the spectra 

 promi.ses to have a bearing on stellar spectra. The advantage 

 of an intense source of light, and consequently of a short time 

 of exposure, is very great ; for a large amount of fog is thus 

 escaped, and faint lines come out which escape observation by 

 the comparatively long exposures hitherto necessary. 



There is another direction in which this battery can be used, 

 which promises to be of importance in surgery. It furnishes a 

 new source of the X-rays. 



The greatest need in the scientific study, and also in the 

 employment of the X-rays in surgery, is a steady source of them. 

 All the methods now in use give a light which is far from con- 

 stant ; the electrical impulses which produce the rays are unequal 

 in strength, and even when they are equal they are generally 

 alternating in character. This to-and-fro action tends to pro- 

 duce a blurring of the shadows, for fluctuating electrical impulses 

 sent through an X-ray tube are apt to give a shifting radiant 

 point. 



The ideal method of producing the rays is by the employ- 

 ment of a large storage battery, and I have been working toward 

 this much-desired end during the past two years. 



Traces of the X-rays can be obtained with a steady current 

 at a voltage of five thousand ; and they are strongly produced 

 at twenty thousand. When forty thousand volts are used with 

 a steady current, the exhibition of rays is surprising ; a fluor- 

 escent screen is lighted with extreme brilliancy, and marvellous 

 >hadows of the bones of the hand are obtained. A steady cur- 

 rent is undoubtedly the ideal current f<ir the production of the 



Fig. 6. 



X-rays ; for the radiant point of the rays does not fluctuate, and 

 there is no to-and-fro or oscillatory motion which tends to pro- 

 duce what may be called X-rays ghosts. It is well known that 

 these ghosts are often puzzling to the surgeon. 



In my experiments I was surprised at the small amount of 

 current necessary with a voltage of forty thousand to produce a 

 strong development of the X-rays. The use of ten milliamperes 

 was dangerous to the tube ; the anode grew white hot, and the 

 Crookes' tube resembled an enclosed arc lamp. It was interest- 

 ing also to notice that the usual fluorescence ceased to be 

 noticeable ; and although the tube was of a milky white hue, 

 the X-rays were extraordinarily brilliant. In my first experi- 

 ments the fall of resistance in the tube was so rapid that the 

 anti-kathode was melted. In the case of all the tubes with 

 which I have experimented, the fall in resistance advances very 

 rapidly with the degree of reddening of the anti-kathode. When 

 this becomes red, or when a red spot appears on it, the differ- 

 ence of potential between the terminals of the tube does not in 

 general exceed twenty thousand volts. 



It would seem therefore uneconomical to continue the use of 

 a high potential machine when this critical point is reached. 

 At this point, moreover, the rays seem to be given off most 

 vigorously, and at this stage a quantity machine giving a com- 

 paratively small voltage could be substituted to advantage for a 

 coil or other apparatus giving six to eight inch sparks. A large 

 storage battery makes it possible to regulate the strength of the 

 current which is at any moment exciting the tube. I accomplish 

 this at present by means of a liquid resistance, which enables me 

 to graduate the strength of the current to any extent. This 

 advantage is a very great one, and is not possessed by any other 

 method. It seems possible, by carefully regulating the strength 



