444 



NA TURE 



[August 30, 1906 



How closely the physical cause of this discontinuity 

 resembles that in the case of a carbon arc is still in doubt, 

 though investigations bearing on this question are under 

 way. With the iron arc there seems to be no sharply 

 defined crater, for each electrode terminates in a viscous, 

 incandescent globule of what seems to be magnetic oxide 

 of iron, from which the discharge takes place. Thus we 

 have to do, properly speaking, not with an arc between 

 iron electrodes, ijut with one between electrodes of Fe,0,. 

 Even when the arc is hissing strongly, the discharge seems 

 to take place from only a small area on the surface of 

 the globule. Moreover, a large increase in diameter of 

 electrodes is accompanied by only a small increase in the 

 value of the critical current, which varies between o-8 

 ampere and 1.5 ampere over a wide range of values of 

 length of arc and thickness of electrodes. On the other 

 hand, I have found no positive evidence that the discon- 

 tinuity is not due to the presence of oxygen around the 

 anode. A test with an exploring electrode showed that 

 the effect is confined mainly, if not entirely, to the anode. 

 Given an arc burning on the quiet stage in the neighbour- 

 hood of the hissing point, the hissing can be precipitated 

 by shortening the arc, just as in the case of the carbon 

 arc. 



.\fter the current has been increased somewhat beyond 

 the hissing point, the arc begins to rotate rapidly, so that 

 on the anode a ring instead of a spot of light appears. 

 This is accompanied by a high-pitched squeak or whistle, 

 which, as the current is still further increased, degenerates 

 into a sputter, and this in turn into a steady, strong hiss, 

 the ring meanwhile having disappeared. At the beginning 

 of the " whistling stage " the arc has a curious tendencv 

 to jump back into the quiet stage, so that for an instant 

 the hissing ceases, the current falls abruptly, and the P.D. 

 rises several volts. If one begins to diminish the current 

 immediately after one of these abrupt changes, the quiet 

 stage can sometimes be maintained steadily, even though 

 the current is far greater than that at which hissing 

 normally occurs. It is not impossible that slight irregu- 

 larities in the supply E.M.F. may in certain circumstances 

 serve to precipitate the change from the one stage to the 

 other, even though the current be not that at which the 

 change normally takes place. 



In conclusion, the question may be raised whether 

 Lecher's observation of the discontinuous nature of the 

 arc discharge between iron electrodes was not made on 

 the hissing stage alone, and whether, as with the carbon 

 arc, the discharge may not be perfectly continuous when 

 the current is made sufficiently small. It is planned to 

 repeat Lecher's experiment, making tests on both the 

 quiet and the hissing stages of the iron arc. 



Middletown, Conn., August q. W. G. Cady. 



Volcanoes and Radio-activity. 



In the Popular Science Monthly for June Major Dutton 

 has an interesting article on the above subject, which was 

 noticed in a recent issue of Nature. Having been occupied 

 lately with the study of volcanoes in connection with a 

 more general inquiry into the cause of earthquakes, it 

 occurs to me to point out that Major Dutton has over- 

 looked the recognised distribution of volcanoes about the 

 sea coast, which seems completely to invalidate his theory. 

 If radium, which the researches of the Hon. R. J. Strutt 

 have shown to be so abundant in typical rocks of the 

 earth's crust, such as granite, were an exciting cause of 

 volcanic activity, we should expect to find an abundance 

 of active volcanoes in the interior of continents, such as 

 the United States, Europe, Asia, Africa, Australia, and 

 Brazil, which is contrary to observation. 



T. J. J. See. 



Naval Observatory, Mare Island, California, August 10. 



The Radio. activity of the Chemical Elements. 

 In connection with the emission, from the radio-active 

 elements, of corpuscles with velocities below the critical 

 velocity necessary for the ionisation of gases, it has occurred 

 to me that such a form of radiation is possibly a fairly 

 general property of the chemical elements. It is, I think, 



NO. 1922, VOL. 74] 



usually accepted that " 7 " radiation always accompanies 

 the projection of " (8 " particles, and the extreme penetra- 

 tion of the " 7 " rays seems to be directly due to the very 

 high velocity of the average " /3 " particle. As the 

 efficiency of the " X " rays is due to the sudden negative 

 acceleration of the unit electrical charges (i.e. the cor- 

 puscles) as they strike the anti-kathode, it appears quite 

 possible that corpuscles, moving with comparatively low 

 velocities, may yet be capable of causing a form of " 7 " 

 radiation of feeble penetrating power. The fact that the 

 kathode stream, which can hardly penetrate the glass of 

 the tube, is still able to set up very penetrating X radiation 

 when given a sudden negative acceleration by impact with 

 the platinum anti-kathode may perhaps be given as an 

 instance in support of this idea. It seems probable that the 

 photographic action of a beam of corpuscles (deviated away 

 from the " 7 " radiation by a magnetic field) may be 

 chiefly due to a form of " 7 " ray set up on contact with 

 the plate itself. The several mysterious instances of the 

 fogging of photographic plates left in certain conditions 

 for considerable periods may be caused by a very feeble 

 form of " 7 " radiation set up by the impact of slow- 

 moving corpuscles on the surrounding matter. Such 

 evidence of these slow-moving corpuscles may be somewhat 

 meagre and doubtful, but I think that, so far as the 

 ordinary chemical elements are concerned, the emission of 

 such corpuscles may be very much greater than the 

 measured activities would lead us to suppose. 



. C. W. Raffety. 

 Streatham Common, August 25. 



THE OXIDATION OF ATMOSPHERIC NITRO- 

 GEN IN THE ELECTRIC ARC. 



IN the year 1775 Priestley published his " Experi- 

 ments and Observations on Various Kinds of Air," 

 in which he showed that when a series of sparks was 

 passed through air, the air became acid. The experi- 

 ment was carried out by means of a glass tube, having 

 one end closed with wax through which a wire was 

 fixed, the open end being placed over a solution of 

 blue litmus. Sparks were passed between the solution 

 and the wire, and in a short time the blue litmus 

 turned red. He further noticed the important fact 

 that the water gradually rose up towards the wire. 

 The observations of Priestley were shortly afterwards 

 substantiated by Cavendish, and in 1893 Lord Ray- 

 leigh, with better apparatus and appliances, repeated 

 the experiments which ultimately led him to the 

 discovery of argon. Priestley attributed the acidity 

 to the formation of carbon dioxide, but Cavendish, on 

 repeating the work, proved it to be due to the forma- 

 tion of nitric and nitrous acids. 



After the successful experimental work of Lord Ray- 

 leigh, attention was turned towards the production of 

 nitric acid from atmospheric nitrogen. But it was 

 undoubtedly due to Sir William Crookes, who as 

 president of the British Association in 1898 directed 

 attention to the gradual depletion of the world's store 

 of nitrogenous products, that the importance of the 

 fixation of atmospheric nitrogen was recognised by 

 the scientific and commercial world. At the present 

 time about i'~, million tons of Chili saltpetre are 

 annuallv exported, but those who have studied the 

 question consider that at this rate of exportation the 

 Chilian beds will be, at the latest, depleted by 1940. 

 But as the population of the world increases, the 

 quantity of nitrogenous material required for fer- 

 tilising purposes advances in equal ratio. Sir William 

 Crookes pointed out in 1898 that the world's growth 

 of wheat was about 163,000,000 acres, which at the 

 average of 12.7 bushels per acre gave 2,070,000,000 

 bushels. " But thirty years hence the demand will 

 be 3,260,000,000 bushels. . . . By increasing the 

 present yield per acre to twenty bushels, we should 

 with our present acreage secure a crop of the requisite 



