NA TURE 



[January 2, 1908 



necessary to enter; but certainly Prof. J. J. Thomson has 

 recognised the important part played by the residual gas, 

 for example, in his lecture reported in Nature of March 

 22, 1906. So far as I have yet been able to see, the 

 action of the residual gas in the passage of large currents 

 is much the same whether a VVehnelt kathode or an 

 ordinary electrode is used. There appears to be a definite 

 pressure at which the conductivity is a maximum, and a 

 steady diminution of conductivity when this pressure is 

 either increased or diminished. I do not think, with large 

 currents at least, there is a sudden drop in the discharge 

 current when the pressure is gradually reduced. Of 

 course in the experiment I quoted there was a sudden 

 cessation of the current, but then there is a sudden 

 .ibsorption of the residual gas when the calcium 

 volatilises. The practical difficulties in the way of main- 

 taining a high vacuum in an apparatus containing in- 

 candescent bodies by ordinary methods probably led to the 

 earlier view that it was possible to make a vacuum a 

 good conductor by use of an incandescent kathode. 



F. SODDV. 



British Association Seismology. 



An ancient Bogie has again appeared upon the scene. 

 The object it has in view is to show that the seismological 

 work done by the British Association is practically use- 

 less. It owes its existence to Prof. Bruno Weigand, of 

 the Strassburg Kais. Hauptstation fur Erdbebenforschung, 

 and it was introduced to the notice of seismologists as a 

 " Begriissung " at the second International Seismological 

 Conference. On this occasion, as a welcome to the British 

 and other delegates, Dr. Weigand dilated at considerable 

 length upon the defects of the instruments used by those 

 who had kindly cooperated with the British Association. 

 At Strassburg a Milne pendulum did not record so many 

 earthquakes as a pendulum of the Reubeur-Ehlert type, 

 and the records from the first of these instruments 

 indicated very late commencements. 



This statement, which month after month and year 

 after year Strassburg has published in the form of 

 registers, has just been emphasised in two new works on 

 seismology, " La Science s^ismologique," by Comte de 

 Montessus de Ballore, and "Earthquakes," by Prof. 

 W. H. Hobbs. These gentlemen, who I do not think 

 claim to have any practical acquaintance with seismology, 

 candidly tell us that their information came from others. 

 What Dr. Weigand has said about his experiences at 

 Strassburg is no doubt absolutely correct. It is not, 

 however, sufficiently complete ; he ought to have added 

 that, although records were not being obtained at Strass- 

 burg, they were being obtained from similar instruments 

 in Britain and at stations in other parts of the world. As 

 to whether a seismograph gives satisfactory results or not 

 depends upon its foundation, its adjustment, and, amongst 

 other things, upon the light which is used for photographic 

 purposes. If the light is too high, halation steps in, and 

 all small vibrations are eclipsed in a broadened line. 



As Dr. Weigand correctly points out, a factor of great 

 importance connected with the working of a seismograph 

 is the speed with which record-receiving surfaces move. 

 In the early form of the Milne instrument, unless great 

 care was exercised with regard to the light and adjust- 

 ment of the boom, in consequence of the slowness with 

 which the paper moved, very small earthquakes or pre- 

 liminary tremors might be lost by the thickening of the 

 line. In the modern form of record-receiving surface, now 

 in use for some years, where the paper moves at 24 cm. 

 per hour, nothing of this sort lakes place. The seismo- 

 graph remains exactly as it was first designed, and with 

 its new recording surface yields results as useful and 

 accurate as those obtained from any other type of instru- 

 ment, and this it does at a cost of less than 3/. per 

 annum. An instrument of higher sensibility th.in the 

 Milne pendulum cannot be used on soft ground, whilst one 

 which uses a very large quantity of photographic paper per 

 year is beyond the means of small observatories. I am 

 quite content to continue with the instrument I now 

 possess, and, in spite of all criticism, I have no desire to 

 change th,e same. John Milne. 



Shide, December 30, 1907. 



NO. 1992, VOL. •]'j'\ 



The Photoelectric Property of Selenium. 



In the letter which appeared in Nature of December 26, 

 Prof. Minchin does not state what kind, of air-pump he 

 used to exhaust the tube containing the light-sensitive 

 selenium " bridge." If he used a mercurial pump, the 

 presence of mercury vapour would explain the great fall 

 in resistance observed. 



In a paper published many years ago (Proceedings of 

 the Royal Society, 1S76-7, vol xxv., p. 22), I showed 

 that mercury combines at ordinary temperatures with 

 selenium, producing a superficial film of comparatively 

 low resistance. This effect is produced with vitreous 

 selenium as well as with the granular or metallic form. 



Selenium converted from the vitreous to the metallic 

 form, in a vacuum free from the vapour of mercury, 

 behaves like selenium rendered light-sensitive in the air 

 in the ordinary way (Draper and Moss, Transactions of 

 the Royal Irish Academy, 1876, vol. xxvi., p. 231). A 

 specimen prepared in vacuo was found to have a high 

 resistance, attributable, no doubt, to its spongy form ; 

 it was, however, highly sensitive to light. 



Richard J. Moss. 

 The Laboratory, Royal Dublin Society, December 27,' 1907. 



Prop. Minchin 's observation (p. 173) that the resist- 

 ance of his selenium cells, or " bridges," when placed 

 in vacuo, becomes enormously diminished is strongly 

 suggestive of internal contact or short circuit, possibly 

 brought about mechanically by the expansion of a small 

 body of confined air in some part of the arrangement. 

 Such short circuits occur not infrequently under ordinary 

 conditions, and may generally be burnt out, and the cell 

 restored to its original state, by the application for a 

 moment of an E.M.F. of 50 volts or 100 volts. 



Selenium cells have often been tested in a vacuum. 

 They are supplied commercially, enclosed in highly ex- 

 hausted glass tubes, by Dr. Ruhmer, of Berlin. 



Shelford Bid well. 



Beechmead, Oatlands Chase, Weybridge, 

 December 30, 1907. 



Echelon Spectroscope. 



Whilst employing a thirty-three plate echelon spectro- 

 scope, constructed by Messrs. Adam Hilger for Prof. 

 Schuster, my attention has been directed to a feature of the 

 instrument which needs to be considered in drawing con- 

 clusions as to the nature of the light. Using an .^rons 

 mercury lamp as the source of light, the bright green 

 line, for example, is resolved by the echelon into a bright 

 and broad principal line with narrower companion lines 

 on either side, and the principal line sometimes shows a 

 narrow dark line not far from its centre. 



Like others who seem to have observed the same effect, 

 I supposed that the principal line showed reversal until 

 I found that, on rotating the echelon slowly about a 

 vertical axis, the dark line moves across the broad bright 

 line, and disappears at the side ; then another dark line 

 appears at the opposite side, moves across, and disappears, 

 like the first; and so on. 



These effects are probably produced by interference bands 

 that have not been considered in the theory of the echelon, 

 but have been described by Gehrcke (" Uber eine Inter- 

 ferenzerscheinung am Stufengitter," Amialen der Physik, 

 xviii., p. 1074, 1905). 



If the echelon is tilted about a horizontal axis, parallel 

 to the plates, these secondary bands become more inclined 

 to the vertical, that is, to the direction of the slit, than 

 the ordinary spectrum lines, and their mutu.nl intersection 

 gives to the spectrum lines a screw-like appearance. 



A further investigation of these bands is being made, 

 and it is hoped shortly to publish additional details of 

 their behaviour that may help to explain their orisin. 



H. Stansfield. 



The L'niversity, Manchester, December 26, 1907. 



A Point in the MatVienT-ticaT Theorv of El-st'city. 



Prof. Pearson has recently made a statement which if 

 correct is of very great importance to engineers. He 

 declares that the distribution of the stresses due to the 

 water pressure on, and the weight of, ■ a dam is entirely 

 different in the case of a thin slab cut from the dam 



