April, 1913. 



KNOWLEDGE. 



129 



their common " centre of gravity," with a third 

 fainter and more remote body, whilst e Lyrac con- 

 sists of two pairs of stars, and the multiple star 

 6 Orionis, not far from the centre of the great 

 Nebula, consists of four principal stars and two 

 minute companions very close to two of the brighter, 

 " to perceive both which is one of the severest tests 

 which can be applied to a telescope " (Sir John 

 Herschel), but three of the four brighter were detected 

 by Huyghens, in 1656. Though, as we have just 

 seen, the number of double stars known to modern 

 astronomers, and visible through the telescope, 

 is thus considerable, yet the angular distance 

 of the components is too small to admit of their 

 detection by the unaided eye in pre-telescopic days, 

 so that though a few clusters were known to the 

 ancients, the first double star which attracted 

 attention seems to have been £ Ursae Majoris, Mizar 

 in the Great Bear, which was noted as double by 

 Riccioli about the middle of the seventeenth century 

 (Lewis). This was also the first star to be photo- 

 graphed as double, by G. P. Bond, in 1857, and also 

 the first " spectroscopic binary," a class of objects of 

 which we shall hear more later on. It is curious, 

 too, that there is comparatively close to Mizar, which 

 is of the second magnitude, a faint star, Alcor, just 

 visible to the unaided eye, and it is said that the 

 Arabs considered its detection as a test for keen 

 eyesight. This star is said to be sometimes known 

 as " Jack bv the Middle Horse," Mizar being thus 

 the " Middle Horse " pulling " Charles' Wain." 

 Dr. Hooke, in 1665, discovered that y Arietis con- 

 sisted of two fourth-magnitude stars, eight seconds 

 of arc apart. During the eighteenth century the 

 well-known doubles, a Centauri, y Virginis, Castor, 

 61 Cygni, /3 Cygni, e Lyrae, a Herculis, and f Cancri 

 were added to the list. The latter star was discovered 

 by Christian Mayer, a Jesuit priest living at 

 Mannheim, and shortly before his death, in 1781, he 

 published a catalogue of all double stars known up 

 to that date, including his own additions, making a 

 total of eighty-nine pairs. For the first time he 

 hazarded the suggestion that some at least of these 

 pairs must be physically connected, a suggestion 

 fully confirmed a few years later. As already 

 mentioned, when first discovering and observing 

 hitherto unknown double stars, Herschel hoped to 

 employ them to determine parallaxes, but continued 

 the work with other ends in view. His first catalogue 

 contained two hundred and sixty-nine pairs, and its 

 examination by the well-known philosophical writer, 

 the Rev. John Michell, the " ingenious Mr. Michell " 

 as he is called by Mr. Lewis, led him to make the 

 following remarks. He says: "Thevery great numbers 

 of stars that have been observed to be double by Mr. 

 Herschel, if we apply the doctrine of chances, cannot 

 leave a doubt with anyone that by far the greatest 

 part, if not all, of them are systems of stars so near 

 to each other as probably to be liable to be affected 

 sensibly by their mutual gravitation, and it is, there- 

 fore, not unlikely that the periods of some of these 

 about their principals (the smaller ones being upon 



this hypothesis considered as satellites to the others) 

 may some time or other be discovered." We may 

 assent to the conclusions, or rather to the probable 

 meaning of their author, without committing our- 

 selves to approval of his language. It was not, 

 however, till 1803 that Herschel made the definite 

 statement that some of these combinations were 

 indeed binary, in a paper which he contributed 

 to the Philosophical Transactions of the Royal 

 Society. This he justified by his examination of 

 the measures of Castor, y Leonis, y Virginis, 

 S Serfientis, and e Bobtis. Of seven hundred and 

 two double stars contained in Herschel's two 

 catalogues, the members were divided into six 

 classes according to their angular distance apart. 



Class I contained 97 pairs separated less than 4". 



Since Herschel's day the whole subject of double- 

 star astronomy has been vastly extended, and our 

 knowledge greatly increased by the labours of a host 

 of professional and amateur workers. Amongst the 

 foremost of these must be placed the name of 

 F. G. W. Struve, of Dorpat, whose classic work, 

 familiarly known as the " Mensurae Micrometricae," 

 still remains the standard authority " for method and 

 arrangement." This catalogue, published in 1837, 

 contains measures of two thousand six hundred and 

 forty pairs of stars, three fourths of the ten thousand 

 four hundred and forty-eight measures being made 

 by Struve himself without assistance, but during the 

 latter part of his work he had the assistance of his 

 son and other observatory assistants, " who entered 

 the readings and turned the dome, but made no 

 measures." These measures have been collected, 

 compared and discussed with much other information 

 relating to double stars in a volume by Mr. Thomas 

 Lewis, F.R.A.S., of the Royal Observatory, Green- 

 wich, published in 1906 (Memoirs of the Royal 

 Astronomical Society, Vol. LVI). Struve's work 

 comprised a general survey of the sky between the 

 North Pole and 15° -5 declination, about 0-63 of the 

 whole sky. His plan of work was (1) to discover 

 and catalogue double stars ; (2) to make micrometer 

 measures of them ; (3) to estimate the magnitude 

 and note the colours ; (4) to fix the places by 

 meridian observations. Between February 11th, 

 1825, and February 11th, 1827, he examined one 

 hundred and twenty thousand stars from the first to 

 the third magnitude, and found three thousand one 

 hundred and twelve double stars, whose distance 

 apart did not exceed 32". 



Many of the double stars exhibit the remarkable 

 phenomenon of contrasted colours, but it has been 

 remarked (Proctor, " Old and New Astronomy," 

 p. 783) that this is never the case when the two 



