March ii, i8S6] 



NA TURE 



4j9 



paper which many Englishmen have often the opportunity of 

 seeing ; and second, because the article in question occasioned 

 considerable discussion. 



With this reference to the relationship of my experiments to 

 those of Mr. David Stevenson, and I will furnish others if re- 

 quired, I think it would only be just for Prof. Piazzi-Smyth and 

 Mr, D. A. Stevenson to withdraw their accusation of July 2. 



I must now refer to a second point which Mr. Stevenson has 

 raised. In your issue of August 6 Mr. Stevenson appeals to 

 the results of my experiments as showing the value of his aseis- 

 matic tables. All my experiments with aseismatic tables, whether 

 used for seismographs or buildings, when approximating in form 

 to Mr. Stevenson's joint were, for a variety of reasons, unsatis- 

 factory. Under compulsion I have stated that Mr. Stevenson's 

 joint itself, as applied to the lighthouses of Japan, has been 

 unsatisfactory. I did so for the following reasons : — 



(i) Shortly after erection the free motion of the tables occa- 

 sioned so much inconvenience, that the European engineers 

 then on the Japanese service had them clamped. For this 

 reason Mr. Stevenson's arrangement was not adopted in light- 

 houses which were subsequently erected (see Brunton on "The 

 Japan Lights," Institute of Civil Engineers, No. 1451, p. 9). 



(2) I learn from the Lighthouse Uepaitment that in 1SS2, 

 wishing to give Mr. Stevenson's tables another trial, several of 

 them were put in working order. The result has been that on 

 March n, 1S82, at Tsurgisaki, a number of lamp-glasses on 

 the burners on the aseismatic tables were overthrown. 



Some time afterwards a second shock produced a similar 

 effect. At neighbouring lighthouses, of which there are several 

 (two being within 8 miles), not provided with aseismatic tables, 

 no damage was sustained. 



The shock of March 11, 1882, which was felt for at least 300 

 miles along the coast, was severe, and its effects at Yokohama 

 ■\nd Tokio, which are no great distance from Tsurgisaki, were 

 i-arefully recorded. I am not aware that any small articles like 

 lamp-glasses, bottles, vases, &c. , were overthrown inside ordin- 

 ary houses (see Trans. Seis. Soc, vol. vii. part ii. pp. 41-44). 

 The fact that no ill effects occurred at other lighthouses provided 

 with Mr. Stevenson's tables, like those on the Inland Sea and 

 near Kiushiu, must not be regarded as an argument in favour 

 of the tables, inasmuch as the earthquake referred to was not 

 felt in those districts. 



I may here remark that several of the aseismatic tables are 

 at places where earthquakes are almost as rare as they are in 

 Britain. Mr. Stevenson tells the readers of N.\Tt;RE (June 26) 

 that his lighthouses suffered when the aseismatic tables were «o/ 

 in working order. I have shown that they suffered when they 

 were in good working order. 



Speaking generally about Mr. .Stevenson's descriptions of his 

 aseismatic arrangement, he invariably refers tn it as a great 

 success. Where it was applied, earthquakes have produced no 

 effect, but where it was suppressed, evil effects have resulled. 

 After inquiries at the Lighthouse Department, which is a branch 

 of my own department, I find that the facts adduced by Mr. 

 Stevenson are exactly ilie reverse of the facts which have been 

 placed at my disposal ; and from what I learn, and from my 

 own experiments, I conclude that thus far Mr. Stevenson's tables 

 have been a failure. 



As a further support to my conclusions I will quote the fol- 

 lowing translation of a report from the chief lightkeeper at the 

 Tsurgisaki Lighthouse : — 



" Sir, — On October 15, 1884, at 4.16 a.m., very severe shocks 

 of earthquake were felt. The aseismatic table was in working 

 order, but the shocks were so violent that fifteen lamp-glasses 

 out of twenty-one in use were upset and broken. The lamps 

 thus stripped of glasses began to smoke. The milled heads of 

 the wick-holders being shaken off, and besides the revolving 

 machine being in motion, we had some difficulty in replacing 

 the glasses promptly ; however, we managed to put them all in 

 proper order again by 4.21 a.m. — I am. Sir, your obedient 

 servant, &c. , 6:c.'' 



The only form of aseismatic joint that I have found at all 

 practical is one where something more like a layer of cast-iron 

 sand rather than a bed of cannon-balls is used to break the con- 

 tinuity between a structure and its foundations. I arrived at 

 this after spending much time and, I may add, a considerable 

 sum of money, and although the method involves the same 

 principle zs Mr. Stevenson's tables, I hardly think he is justified 

 in claiming my successes to back up what I cannot but feel have 

 been his failures. 



If the ways and means permit, I hope to make experiments 

 upon a small building resting on a bed of sand or fine gravel. 

 Should results of any value be obtained, surely Mr. Stevenson 

 will not expect me to do more than I have done already — 

 namely, to state the relationship which may exist between these 

 experiments and those which he carried out at the expense of 

 the Japanese Government. 



If everything connected with earthquake investigation which 

 involves the same principles as are involved in Mr. Stevenson's 

 lighthouse tables are to be regarded as his creations, he cannot 

 avoid claiming the rolling sphere seismograph, the rolling 

 cylinder seismograph, the horizontal pendulum seismograph, the 

 conical pendulum seismograph, and in short, a very large pro- 

 portion of the work accomplished by the Seismological Society 

 of Japan. To this I cannot assent. All that Mr. Stevenson 

 can be accredited with is a particular method of applying a prin- 

 ciple, and this method has to my mind been a failure. 



The question of the principle involved in Mr. .Stevenson's 

 device is one that has been repeatedly discussed in Japan. As 

 an example of these discussions I will refer to the Transactions 

 of the Seismological Society of Japan, vol. iii. p. 9, where Prof. 

 J. A. Ewing is speaking, amongst other things, about a rolling 

 sphere seismograph, the invention of Mr. Thomas Gray. Prof. 

 Ewing says that Mr. Gray's contrivance was an application of 

 "the method of supporting a mass by a movable piece in such a 

 manner that the connection with the earth was at the centre of 

 percussion of the movable piece, the mass being at the cor- 

 responding centre of instantaneous rotation, while at the same 

 time the supporting piece was arranged so that its movements 

 did not introduce any disturbing force due to the action of 

 gravity upon the mass. This kinetic property, common to all 

 these instruments, he believed he might fairly claim to have 

 introduced into seismomeliy." 



If Messrs. Stevenson and Smyth see fit to comment upon 

 these notes, I trust that they w ill distinctly state whether they 

 yet consider that I have attempted "to get round the letter" of 

 Mr. David Stevenson's invention, and whether they were justi- 

 fied in publishing the objectionable personalities about "a B. A. 

 man " on July 2. 



This is the main point at issue, and if they choose to neglect 

 it, the discussion may be considered as ended. 



In conclusion I may remark that it was not I who commenced 

 this controversy. John Milne 



Tokio, January 6 



On the Velocity of Light as Determined by Foucau't's 

 Revolving Mirror 



A FEW years ago Lord Rayleigh raised an interesting question 

 as to the quantity actually determined by our experiments and 

 observations on the velocity of light. There can be no difficulty 

 as long as the medium transmits different wave-lengihs with the 

 same velocity, but whenever the medium possesses the property 

 of dispersion the velocity with which any one crest of a wave 

 travels ( y) is different from the velocity with which a group of 

 waves is propagated (6^) ; hence the question arises in each 

 particular case, whether it is f or i/ or something depending 

 on both quantities that we measure. 



In his first article on the subject (Nature, vol. xxiv. p. 382J 

 Lord Rayleigh states that, in experiments with Foucault's revolving 

 mirror, the group-velocity, U, is determined ; but subsequently 

 (vol. XXV. p. 52) he corrected this statement and gave y-/(7 

 as the quantity measured. A paragraph was added, however, 

 in which the remark is made that, if a convex lens is interposed 

 so that an image of the slit is formed on the fixed mirror, the 

 rotation of wave-frunt, caused by the different velocities ot dif- 

 ferent wave-lengths, and acquired on the outward journey, is 

 neutralised during the return, so that in this case we should 

 measure F. 



Gouy {Comptes rcndns, ci. p. 502, 1885) dissents from Lord 

 Rayleigh's second view, and gives i/as the quantity determined ;. 

 without, however, giving sufficient reason in support of his 

 opinion. 



Finally, Michelson has performed the experiment with bi- 

 sulphide of carbon, and obtained a result in close agreement 

 with U. In a discussion of Michelson's measurements in the 

 American Journal of Science, by J. W. G. , his result is said to 

 give " no countenance " to the theory which would make the 

 velocity observed K-/ U. 



It is the object of this communication to support Lord Ray- 



