446 



NATURE 



[September 5, 1895 



MKRi> OF SwiFl^ ^. ■Mri.— The following elements 

 cineris of Swift's comet, the reap]>earance of which was 



„..i ujt'd last week, have been computed by Dr. Bcrberich, 



and are published in Edinburgh Circular, No. 45. The ele- 

 ments are deduced from the observations : — Mount Hamilton, 

 August 21 ; Nice (M. Juvelle), August 24 ; Hamburg, August 



25- 



.\nother ohser\-ation of August 23, made by Mr. J. Witt at 

 the L'rania Observatory, Berlin, is closely represented by the 

 cphemeris. Dr. Berberich thinks the comet w ill possibly belong 

 to the group of periodic comets with short revolution. 



EUmeuts. 

 T = 1895, Sept. 3-3630, M.T. Berlin. 



» = 179 37 '941 



il = 172 59'65J- Mean Equinox iSgs'O. 



»■ = 4 38-55) 

 1<^ ,j = 0-16537. 



Ephemerii for Berlin Midnight. 

 I .<!-. R..*. Decl. log. A. log. r. 



117 

 I-I8 



bright- 

 ness. 



+ 6 17-3 

 6 iS-o ... 9-7100 ... 0-1656 

 6 17-5 



6 15-9 ... 9-7071 ... 0-1664 

 6 13-2 



6 9 6 ... 9-7062 ... 0-1679 .. I'lS 

 6 51 



5 59-7 - 9-7073 - 0-1700 1 10 

 5 536 

 22 ... J 29 49 ... +5 469 ... 9-7107 •■• 01727 ... 116 

 The brightness at .August 21-5 has been taken as unity. 



Comets .\xd the Sus-si'OT Period. — Since the discovery 

 of the [leriodicity of the sun-spots, investigations have shown 

 that many terrestrial phenomena are, and others may be, closely 

 allied to it. These are generally looked upon as results due to 

 the variation in the .sun. If it were found that comets had 

 an eleven-yearly jK-riod, we should have the question before us 

 as to whether this [wriod be the result of this period, or whether 

 the (wriod dcjiended to a certain extent on this periodical 

 cometar)' influx. If the sun, as has been supposed, were fed, so 

 to speak, with cometary matter, then the spot period would 

 naturally lie de|x:ndcnt on some external source of supply such 

 as this. But since the solar atmosphere has a circulation which 

 seems now to have lieen fairly shown to be the cause of this 

 perio<Iicity, such an outward supply of energy is not thought now 

 to l)e of such importance as would have been the case some 

 years ago. This <lo<;s not take away the interest, however, from 

 Herr J. Unterwcger's investigation concerning the connection 

 of sjxits and apjitarances of comets, but would rather instigate 

 it. The author h.a.i, by a strict examination of the elements of 

 the larger perimlical comets, obtained a function which can be 

 repre.sente^l mathematically by a formula, and from which an 

 eleven-yearly |ierir«l since 1740 can Ik- recogni.sed. From the 

 year 1833, also, the maxima and minima points fall together, with- 

 out exception, with those of the sun-spot curves. In determining 

 the length of the period, the amplitude of the jwriod was set 

 for each series as a function of the length of the ]>eriod, and 

 then each value for the duration of the perio<l a.scerlained. The 

 calculation wa.s so arrange<l that two neighl>ouring values, which 

 made the amplitude a minimum, were als<j determined. The 

 values for the function came out as 8682, 1 1 '2 26, 13-365 years, 

 those for the scries showing the relative number of .sun-spots 

 lieing 8-721, 11-254, 13-424 years. 



To determine also whether the points of maxima and minima 



for the function were coincident with those for the series showing 



the -un-'.fHii numl>crs, the curves drawn from the values derived 



' , extent th.-il a sccondar)' maximum could be 



1 them in .similar |Hisilions. 



. h the investigator indicatc*s as having been 



..hown art : -Tin- identification of the ihirty-five-yearly sun-sp^il 

 |>crl<Kl, the fiiri'li'in giving larger values in 1778, 1S16, 184S 

 and 1H82, and smaller minima in 1764, 1806, 1834 and 1867; 

 the tim<- brinecn two succe-s.sive maxima lieing in the mean 

 J4" 



I iwriod 1764-1806, with maximum at 1777-80, 



coil - a secular maximum of sun-spots and a large 



gteltehemerjlon which K-gan in 1 768 and ended about 1785. The 



NO. 1349. VOL. 52] 



1806-1834 interval, with a maximum at 1S16, corrc>|Kinil> lo a 

 maximum of sun-spot and to an intense gklSihcn't-rslors from 

 1S14 to 1824. The third and fourth |)eriods are also likewise 

 explained. 



Cases are also made out for the secular variations in the 

 climate, and a suggestion is thrown out that if we may look upon 

 ** Kometen als stark elektrische Ma-ssen," then at the times of 

 their maximum number and least distances from the e.irth, small 

 induced currents n>ay be set up, which will lie recorded by the 

 magnetic needle : this latter question has ni->t. however, been 

 investigatiil. 



THE SUN'S PLACE IN 2\ATURE.^ 

 XI. 



The Cloik Kate. 



The proper regulation of this clock error and consequent 

 " trail " of the spectrtim across the plate parallel to itself are 

 essential to the .success of photographs taken liy the objective 

 prisms. The S[>ectrum of a bright star must obviously be made 

 to trail more quickly than that of a fainter one, and a shorter 

 exposure is sufficient. Since for the same clock error, and in the 

 same time, a star near the pole will give a shorter trail th.in one 

 nearer the equator, <leclination must also be taken into account. 

 Keeping a constant clock error, equal widths of spectrum for 

 stars of difi'erent declinations may l>e obtained by lengthening 

 the time of exposure for stars away from the equator, but in that 

 case, the stars near the pole would be over-exix>sed in relation 

 to those nearer the equator. 



The exixisure given to stars of equal magnitudes should 

 evidently Ik' the same, no matter in what part of the sky they 

 may be situated, and the clock error should, therefore, be 

 increased in proportion to the secant of the angle of declination. 



The light-ratio of stars being 2-512", where // exprc.s,ses the 

 difi'erence in magnitude, the time of exposure must vary in the 

 .same proportion, and the clock error in inverse proportion. 

 Thus, where 5 minutes' exposure is sufticieni for a first-magnitude 

 star, 31 minutes is required to obtain a fully-exposed spectrum 

 of a star of the third magnitude. This law, however, only 

 applies to photographic magnitudes, and must be mttdified 

 according to the type of spectrum or the colour of the star. 



The red stars, being much weaker in blue and violet rays than 

 the yellow or white stars, require much longer exposures than 

 white stars of equal magnitutle. To obtain a spectrum of 

 (3 Pegasi extending to the K line, for example, at least three 

 times the ex|xisure required by a white star of similar magnitude 

 nuist be given. 



I'or conveniently adjusting the exposures, tables have been 

 constructed which show at a glance the position of the regulator 

 for a star of given magnitude and declination. 



It is obvious that with an instrument of high dispersion, the 

 number of stars it is jiossible to photograph is very limited, as 

 the long exposures required for the fainter stars are impractic- 

 able, and, even if possible, the definition of the lines svould be 

 destroyeil by atmos]iheric tremors. 



Hence, it is at present imly jKissible to photograph the spectra 

 of the faint stars on a very sm.all scale. With an objective of 

 8 inches aperture and 44 inches focal length, and a prism of 13' 

 refracting angle. Prof. Pickering has pholographeil tlie spectra of 

 stars down to the eighth magnitude. These spectra are about 

 I centimetre long, and a millimetre broad, and though they do 

 not show a very great amount of detail, they are sutticicnl to 

 reveal the type of sped rum. 



With an instrument capable of photogra))hing faint stars, a 

 large number of spectra may be taken at one exposure : but, 

 with the instruments of larger dispersion, this is not generally 

 the ca.se, as there are few bright stars of nearly eipial magnitude 

 sufficiently close together. 



The Electrical Control. 



In consequence of the great .accuracy reipiired in ihe driving 

 of the telescope when long exposures are necessary, the 10-inch 

 e(|uatorial has lieen filleil with a simple and inexpensive form of 

 electrical control. This is a modification of that designed by 

 .Mr. Russell, of the Sydney Observatory.^ The exi.sling driving 



1 Rcviwd from «horlh.ind notes of a course of Lectures to Working Men 

 at (he .Mu^uin of Pntclictl Geology during November .ind December, 

 1694. (Continued from pnge 4^5.) 



^ Monthly XotUn. vjI. M. (>. 43, 1890.91. 



