748
ASTRONOMY
hemisphere between the equator and the northern celestial pole. It is divided into zones of 20° in breadth. The zones from the equator to 40° of declination have already been published. The most remarkable work now done with the spectroscope is the determination of the motions of stars in the line of sight. The earliest attempts at such measures were those made by Sir William (then Mr) Huggins; and after the practicability of making such determinations was demonstrated by him, the work was taken up at the Royal Observatory at Greenwich. In these early researches only optical methods were used which were not susceptible of the highest precision. The introduction of more sensitive chemicals into photographic practice led to the great improvement of photographing the spectra of stars instead of observing them visually. A twofold advantage is thus gained: fainter spectra can be photographed than can be measured with the eye, and the errors to which the eye observations are liable can be almost entirely done away with. The first extended application of the photographic method to this purpose was made by Dr Yogel at the Potsdam Observatory; he measured the motions of forty-eight of the brighter stars in the line of sight with a precision that could not easily be reached by the optical method. The most interesting results of the new system of observation have been the discovery of binary systems among the stars. In this work, Campbell of the Lick Observatory has been remarkably successful; the results reached by him, Deslandres, and others will be summarized in their proper connexion. Parallaxes and Proper Motions of the Stars. The methods of determining the parallax of a star are of two kinds. In the one the actual parallax is deduced from the annual change in the right ascension or declination of the star produced by the motion of the earth round the sun, and shown by comparing observations with a meridian instrument of the highest precision. Parallaxes determined in this way are termed absolute. It is now found that, owing to the annual and diurnal changes in the instrument and in the air caused by the varying temperature of night and day, summer and winter, it is impossible to determine precisely, even with the most refined appliances of modern astronomy, variations in the absolute position of a star so minute as its parallax. Efforts to determine absolute parallaxes have therefore been abandoned. The method now in universal use consists in referring the displacement of the star whose parallax is measured to small stars lying in nearly the same direction. The numbers thus obtained are the relative parallax between the star measured and those with which it is compared; and if the stars of comparison are so distant as to have no appreciable parallax, the result will give the actual parallax required. How far this is the case we cannot yet say with certainty; but it is probable that, as a general rule, the stars of comparison have rarely any parallax amounting to a hundredth of a second. In recent times the heliometer has been the most powerful instrument for such determinations; and in the hands of Gill, Elkin, Peter, and others it has been applied in making a revision of many results previously obtained with other instruments, and in investigating suspected new cases of parallax. But at the present time it bids fair to be eventually supplemented or even superseded by the photographic telescope. The great advantage of the photographic method lies in the ease with which a plate can be taken and the leisure with which the images of the stars can be measured upon it; the attempts recently made in this direction seem to show that the precision is not markedly less than that attainable by the use of the heliometer.
It is difficult to make a selection of the stars whose parallax can be said to be determined, but, roughly speaking, we may state the number to be 100. Among these are several cases in which the instrument or method of using it was so imperfect that the results are more or less doubtful, and many others in which the apparent parallax is so minute that it is questionable whether the result may not be due to errors of measurement rather than parallax. Such is undoubtedly the case when the result comes out with a negative algebraic sign, as has happened with some apparently good determinations. The following is a list of the determinations which may be regarded as actual, including all which have been made with the heliometer, even when the result is so minute as to be doubtful. The letter g signifies that the result is one of an exceptionally good class reached under Gill’s direction by the heliometer of the Cape Observatory. The probable error in these cases is generally between 0'01" and 0'02". A colon indicates that the parallax is of a rather doubtful character, needing to be re-determined by modern methods, and two colons that it is among the more doubtful of this class. The results given without any indication may be supposed to be affected by a probable error varying from 0'03"to O05". Parallaxes of Stars. Position for 1900. Star. R. A. Decl. Par. h. m. /3 Cassiopese 0 4 + 58-6 15 Gr. 34 0 13 + 43-4 •30 f Tucanse 0 15 -65-5 14 P Hydri 0 20 -77-8 13 g a Cassiopese 0 35 + 56'0 •04: rj Cassiopese 0 43 + 57-2 •20 7 Cassiopese 0 50 + 601 •01: (i Cassiopese 1 2 + 54-4 14 Polaris 22 + 88-8 •06 a Eridani 34 •04 g -57-7 r Ceti . 39 -16-5 •31 g e Eridani 16 -43-5 15 g 502 Persei 2 + 37-8 •04: : o Eridani 11 - 7'8 T7 g a Tauri. 30 + 16'3 11 -45-0 •31 g C. Z. Vh. 243 a Aurigse 9 + 45-9 •09 10 p Orionis - 8-3 •00 g 50 a Orionis •02 + 71 52 P Aurigse + 44-9 •06: 22 a5 Argus -52-6 •00 g 6 39 11 + 43-7 X Aurigse a Can. Maj. 6 41 -16'6 •37 g 6 53 51 H. Cephei + 87-2 •03: a Geminorum + 321 •20:: 7 28 a Can. Min. 7 34 + 5-5 •30 p Geminorum + 28-3 7 39 •06 LI. 15290 + 30'9 7 47 •02 8 52 1 Urs. Maj. + 48-4 13 : 8 54 10 Urs. Maj. + 42-2 •20 LI. 18115 + 531 18 26 + 52-2 6 Urs. Maj. •07 + 43-2 LI. 19022 37 •06 55 20 Leo. min. + 32-5 •06 3 a Leonis + 12-5 ■02 5 Gr. 1618 + 50-0 18 10 22 Gr. 1646 + 49-3 10 10 27 Gr. 1657 •04 + 497 10 38 •01 : LI. 20670 + 47-7 10 58 + 36-6 •46 LI. 21185 11 0 + 44-0 LI. 21258 •22 11 9 15 2 1516 . + 74-0 11 15 + 66-4 O. A. 1167 •27: 11 33 •03 2 1561 . + 45-7 11 40 + 48-2 •02 Gr. 1822 11 47 14 Gr. 1830 + 38-5 11 57 •00: LI. 22632 + 43-7 12 4 •06 LI. 22810 + 40-8 12 21 a Crucis -62-5 •05 g 12 42 -591 •00 g P Crucis
