September 27, 1894J 



NA rURE 



53' 



OUR. ASTRONOMICAL COLUMN. 



The AccuRAi V or Astronomical Observations. — At the 

 tenth general meeting of the Asfuciation Gcodesique Inter- 

 nationale, Prof. Cornu read a paper on the necessity of intro- 

 ducing addilicjiial precautions in astronomical observations 

 requiring great accuracy. The subject of the paper was sug- 

 gested by the inierprelatlon put upon observed variations of 

 latitude. Prof. Cornu fir^t remarked that, while carrying 

 out their experiments on the earth's density some time ago, 

 J. B. Bailie and himself found that the constant of gravitation 

 appeared to undergo an annual variation, being a little greater 

 in spring than in autumn. A minute examination of the experi- 

 mental conditions showed the investigators that this periodic 

 anomaly was purely au apparent phenomenon, and that it was 

 caused by an annual variation in the temperature of the room 

 in which the apparatus was placed. It is impossible not to be 

 struck with the analogy of these results and those of the 

 variations of latitude. In both cases the period is approxi- 

 mately an annu.il one, and the maxima and minima occur in 

 spring and autumn respectively. The question arises, therefore, 

 whether astronomers have been careful to eliminate all the 

 meteorological causes affecting their results, and whether their 

 claims for marvellous accuracy are not, to some exent, ex- 

 aggerated. Prof. Cornu has applied the reasoning of the 

 physicist to astronomical observations and instruments. 

 Beginning with the meridian circle, he points out that, on a 

 divided circle one metre in diameter, one-tenth of a second of 

 arc corresponds in round numbers to -2^T<T,Trr,T, of a metre, or 

 o 0005 mm., that is, a semi-micron. But numerous metro- 

 logical experiments have shown that, even under the most 

 lavourable conditions, a semi-micron is the limit of precision in 

 dilTerential linear measures on scales one metre long ; and to 

 attain this limit, it is necessary to put the scales side by side in 

 a bath having a practically constant temperature. But the 

 circles of meridian instruments are subjected to all sorts of 

 variations, hence ii seems affectation to assume that observa- 

 tions made with them are true to one-tenth of a second of arc ; 

 yet that assumption is made in the discussion of observations of 

 latitudinal variations. When the spirii-level, the telescope, and 

 the micrometer are criiicised from this physical point of view, 

 they are found wanting in the extreme accuracy usually ascribed 

 to them. And, in adilition to the innate defects in the parts of 

 a meridian instrument, there are the temperature variations 

 which almost bailie estimation. Several additional precautions 

 should be taken to reduce this vitiating cause. In the first 

 place, the distribution of temperature around and inside the 

 instrument should be frequently determined, so that the proper 

 corrections (or lelraction could be made. The conductivity of 

 the parts of the instrument ought also to be increased, and 

 made as nearly equal as possible, in order to reduce flexional 

 and torsional effects produced by inequality of temperature. 

 Another improvement would be to reduce the quantity of heat 

 emitted by light-sources in observatories, and, finally, 

 attempts sitould be made to give the air in the telescope 

 tube the same temperature as that outside ; so that systematic 

 errors of retraction might be eliminated. Prof. Cornu thinks 

 it is only by having recourse to precautions of this kind that 

 definite results on the variability of latitudes can be obtained. 

 In the actual state of the observations, he says, two purely 

 physical objections exist against the reality of the pheno- 

 menon. They are: — (I) Is it certain that observations of 

 latitude by the Talcott method are free from periodic annual 

 errors due to meteorological influences, particularly to the 

 variation of temperature during different seasons? (2) Can it 

 be demonstrated that ihe astiunomical data used in the pre- 

 paration of the catalogues of the stars employed in these 

 observations are also free from the errors referred to ? 



Liverpool OiisERVAroRy. — In a small pamphlet, published 

 by order of the Mersey Docks and Harbour Board, we have 

 placed before us some extracts from the report of the Director 

 of the Ob-iervatory, Mr. \V. E. Plummer, to the .Marine Com- 

 mittee. The excessive briefness of the extracts in question 

 curl ail our remarks very considerably. The transit instrument, 

 for time determinations, has been used on every possible 

 occasion, and the normal clock has been maintained more 

 accurately during cloudy weather by the adoption of separate 

 clock errors and rates lor the Bond and .Molyneu.x clocks. The 



NO. 1300, VOL. 50] 



determination of the longitude (not longlitudt, as is twice 

 printed) of the Observatory has not yet been completed, as the 

 repeti; ion of the signals has not yet been effected, bat the in- 

 quiry at present indicates a change from previous values, the 

 Observatory acquiring a new position some 400 yards west of 

 the old one. The latitude observations, by the method of 

 transits of zenith stars over the prime vertical, although said to 

 give this quantity with great exactness, are likely to be dis- 

 appointing when used for discussing the small variation in lati- 

 tude, for the instrument and its mounting have shown signs of a 

 slight instability. Besides the observations of comets, stellar 

 parallaxes, by the method of chronographic record of meridian 

 transits, have been investigated. The meteorological observa- 

 tions have, as usual, been continued, and the present report 

 contains numerous tables of the results obtained. 



The Variaiile R Lvr/e. — The variability of this star was 

 first pointed out by Baxendall in 1S56, and its period was 

 stated to be one of forty-eight days. At a somewhat later date 

 .Schonfeld, from a few more observations, deduced a forty-six- 

 day period. Generally the observations have been somewhat 

 scarce, but more recently their number has been considerably 

 increased Vendell, Sawyer, Plassmann, and Knopf having been 

 chiefly responsible for these. That the time had arrived when 

 a more thorough investigation of this " lichtwechsel ' might 

 be attempted, is the opinion of lierr .\. Pannekoek, and the 

 details of his work will be found in AslrLmomischeNachrichtcn, 

 No. 3252. In the examination of the observations it was soon 

 found out that the period did not appear to remain constant. 

 In the earlier epochs, commencing in 1SS7 and continuing up 

 to the past year, the numbers indicated clearly a period of 

 about forty-six days, but recently they have somewhat in- 

 creased. In general, also, it has been noticed that the time for 

 the increase in the light curve is shorter than that for the de- 

 crease, llerr Panr.ekoek concludes, however, from his work, 

 that R Lyras does not vary its brightness so irregularly as is 

 supposed. He suggests that the observed apparent deviations 

 recorded can be attributed to special errors of observation. He 

 is of opinion, however, that perhaps when many more ob- 

 .servalions are at hand, we shall be in'a far better position to 

 investigate the subject more thoroughly, and to state more 

 definitely through what amplitudes these light variations swing. 



The Cleaning oi-' Object-glasses. — Owners of telescopes 

 (refractors) will no doubt be glad to have a few words of prac- 

 tical .advice, from one who can speak from long experience, 

 with regard to the cleaning of object-glasses. The advice in 

 question is extracted from an article, by Mr. Brashear, in the 

 September number of Pofu'ar Aslronomy, and we do not hesi- 

 tate in helping to distribute it. The reader may be rather sur- 

 prised to hear that the use of fine chamois-skin, tissue-paper, 

 or an old soft silk handkerchief, or '"any other such material to 

 wipe lenses, as is usually advised," are not advocated. The 

 reason for this is, not that the materials themselves do the mis- 

 chief, but that the chief enemy to an object-glass, " dust par- 

 ticles," and these most likely of a siliceous nature, must not on 

 any account be rubbed on the glassy surface. The receipt, in a 

 few words, may be summed up as follows: — If the lenses be 

 dirty or dusty, a tuft of CDtton or a camel's-hair brush may be at 

 first applied, but pains should be taken that no pressure be 

 given to either. For further cleaning, a woodcit bowl, previously 

 washed out with soap and water, should be filled with clean 

 water of approximately the same temperature as the objective. 

 .K little ammonia (quantity, a teaspoonful to half a pail of 

 water) should be added to the water. "Cheese-cloth " is 

 snongly recommended as a means of applying the soap to the 

 glass; but this, first, should be "thoroughly washed with soap 

 and water," and thrown away when done with. Plenty 

 of water must always be used. .-V third or fourth cloth should 

 be used to wipe the objective dry. " Vigorous rubbing will do 

 no harm if the surfaces have no abr.ading materi,il on them, and 

 I have yet to injure a glass cleaned in this way." If the objec- 

 tive be not taken from the cell, the camel's-hair brush and the 

 soap-andwater process can be still used, and the work finished 

 with a dry cheese-cloth. Mr. Brashear gives good reasons for 

 taking the lenses apart from time to time, and giving them a 

 thorough clean ; and he adds that everyone who owns and uses 

 a telescope should be so familiar with his objective that he can 

 take it apart and put it together just as well as the maker of it. 

 In moist climates particularly this should be done frequently. 



