Nov. 



1886] 



NA TURE 



75 



\fany striking individual peculiarities were noticed in the 

 course of the experiments, which these general averages fail to 

 show. Three of the male observers were able to detect one part 

 of prussic acid in about 2,000,000 parts of water. Two of these 

 were persons engaged in occupations favouring the cultivation 

 of this sense. Careful chemical tests failed to show the presence 

 of prussic acid in several of the more dilute solutions, in which 

 it could he detected by the sense of smell. We found some of 

 both sexes who absolutely could not detect prussic acid even in 

 sohitions of almost overpowering strength. There were several 

 instances of the same peculiarity as regards bromine. Again, 

 our averages show that the sense of smell is in general much 

 more delicate in the case of male than of female observers. 



Edward L. Nichols 

 E. H. S. Bailey 



University of Kansas, November 4 



Tidal Friction and the Evolution of a Satellite 



Adverting to the correspondence in Nature (vol. xxxiv. 

 p. 2S6), I think that Mr. Darwin has not, ?o far, fully realised 

 the results that would follow from the circumstance that the 

 Martian satellite's period would be affected many hundred times 

 more than th.at of the planet's rotation, as explained in the 

 correspondence referred to. He argues that, the moon's mass 

 being great, she should recede to an enormous distance before 

 there will be a reversal of the direction of her tides on the 

 earth ; while the satellite of Mars, being very small, need only 

 to recede a short distance before a similar tidal reversal ensues. 

 No mention being made of any other supposed ditference in the 

 systems at the starting-point, it must be inferred that other 

 things are supposed about equal. But, as a matter of fact, the 

 present position of the Martian satellite is incompatible with an 

 initial rotation of its planet anything like so great as that ascribed 

 to the earth at a like stage. If Mars be supposed to rotate ten 

 times while the satellite, at its present distance, makes nine 

 revolutions, the satellite's period would still be affected or 

 lengthened much more than would that of the planet's rotation. 

 The difference between the periods of revolution of the planet 

 and satellite would increase quickly at first, but more slowly 

 as the satellite receded a certain distance, till at a certain 

 time there would be no increase, after which there would be 

 a decrease, and finally a reversal. When the satellite would 

 have receded to a short distance, where she would revolve in the 

 same period as Mars now rotates in, the planet would have lost 

 but little of its original rapid rotation. Now, supposing the satel- 

 lite tide to go round in the same time as the solar one, the period 

 of the satellite would be affected about thirty times as much as 

 that of the planet's rotation. Allowance being made for the 

 comparative slowness of the satellite's tides, the satellite's period 

 would still be changed more than ten times as much as that of 

 the planet. It would be only when the little body got further 

 out, and the planet's rotation slower than it now is, that there 

 coi:ld be a reversal of the direction of the satellite's tides. 

 Wherever started, the satellite must either go directly into the 

 planet, or go out a short distance and back into the planet, before 

 the rotation-period can have been mucli changed by solar tides ; 

 or else the satellite must go far out — as when it gets a fair start 

 — and could not possibly turn back until the rotation of Mars be 

 slower than now. Hence it seems that under no conditions could 

 the rotation of Mars, at the birth of her moon, have been twice as 

 rapid as now, and the evidence is very strong that the rota- 

 tion-period could not have been changed more than a very few 

 hours, if so much. Then, if the rotation of Mars was so slow in 

 the beginning, and so little changed during the whole existence 

 of the satellite, the circumstance does not support the view that 

 the earth's rotation was very rapid in the beginning and so much 

 changed during her past history, but rather inclines the other 

 way. 



Respecting the statement that two heavenly bodies cannot 

 revolve about their centre of inertia as parts of a rigid body 

 with their surfaces nearly in contact, unless one be smaller and 

 denser than the other by a certain amount, I can only say, at 

 the present time, that such was the conclusion at which I arrived 

 when investigating the results of the tidal effects of two bodies 

 on one another at close quarters. Without going far into the 

 question, it can be seen that if the rule holds when the two 

 bodies are of the same size and density, it will hold throughout. 

 There will be no difficulty in seeing that the nde holds so far 

 that when the difference in size between the bodies is as great as 



between any of the satellites and its primary, the small body 

 must be invariably the denser. Now the argument that was 

 supposed to apply in general would at least apply in the case of 

 the solar system. That argument, as explained in my pamphlet, 

 was that, if a rapidly-rotating body were to separate into two, 

 the small body given oft' must be denser than the other to with- 

 stand the tidal disturbance, and that it would be impossible for 

 the small body to be denser than the primary, since the 

 secondary body must be formed from the surface and therefore 

 lightest part of the other body. James Nolan 



Dergholm, Victoria, October 5 



Seismometry in Japan 



I HAVE read, with no small surprise, a paragraph in Nature 

 of November 11 (p. 36), giving a summary of a letter from Prof. 

 John Milne, with reference to an article by me on the seismo- 

 graphs now manufactured by the Cambridge .Scientific Instru- 

 ment Company. Prof Milne is represented as saying that, 

 " with the exception of one or two which have been modified, 

 a set of instruments like those recommended by Prof. Ewing 

 are, so far as Japan is concerned, quite obsolete." His letter is 

 not published, and it is possible that the paragraph inadvert- 

 ently does him an injustice in making him assert what has 

 absolutely no foundation in fact. 



In any case the statement cannot be allowed to pass without 

 contradiction. • My seismographs have been in regular use at the 

 University of Tokio since they were invented ; they are now 

 used for systematic observations by the Japanese Meteorological 

 Bureau ; they were sent last year by the Japanese Government 

 to the Inventions Exhibition in London, where they were 

 awarded the highest diploma among Government exhibits : one 

 of them, the comparatively cheap and simple duplex pendulum 

 seismograph, is employed by many private observers in Japan. 

 In a letter received only a few weeks ago, my friend and former 

 assistant, Mr. Sekiya, now Professor of Seismology in the 

 University, says : — 



"We are going to start a journal called ^e yournnl of the 

 Science College of the Imperial University, yapan. In the first 

 number I will give a paper on ' Comparison of Earthquake 

 Diagrams simultaneously obtained at the same station by two 

 instruments involving the same principle, and thereby proving 

 the trustworthiness of these instruments.' Of course I treat 

 those diagrams recently obtained by two of your seismographs." 



Other letters from Prof. Sekiya are full of accounts of the 

 excellent work he is doing with these instruments, and of their 

 continued and extended usefulness in his very able hands. A 

 paper lately received from him describes a rough but effective 

 form of the duplex pendulum, cheaply made in order to bring it 

 within the reach of private observers, and with reference to this 

 the Japan Mail of February 2, lSS6, says :— 



" The duplex pendulum seismograph designed by Prof. J. A. 

 Ewing, has been employed for earthquake observations in the 

 Tokio Daigaku by Mr. S. K. Sekiya, who has improved 

 many of its details during his long use of the instrument. On 

 account of the simplicity and scientific nature of its construc- 

 tion, and its easy management, it has found its way into the 

 hands of many observers." 



The Mail goes on to mention the name of a native firm by 

 whom the instrument is made and sold. In March last Mr. 

 Sekiya writes: — "The duplex pendulum sells well; some 

 fifteen or twenty of them have been sold." 



So much for the duplex pendulum seismograph, which is one 

 of those described in my article, and now made with the utmost 

 refinement of construction by the Cambridge Company. The 

 other is a three-component instrument, of which the principal 

 part is the horizontal pendulum seismograph — consisting of a 

 pair of horizontal pendulums for recording separately two rect- 

 angular components of the horizontal motion of the ground on a 

 moving surface driven by clockwork. This method of recording 

 earthquakes was introduced by me in 1S80 {Trans. Seis. Soc. 

 Jap., 1S80 ; Free. Koy. Soc., No. 210), and has been in 

 regular use ever since. The instruments made to my designs by 

 native workmen are still doing good service in Prof. Sekiya's 

 hands. Those now made by the Cambridge Company have the 

 advantage of better workmanship and an improved arrangement 

 of pans. As Prof. Sekiya has recently written to me with 

 regard to the purchase of a set of them by the Japanese 

 Government, it is probable that Mr. Milne will before long have 



