Febeuary 21, 1902.] 



SCIENCE. 



295' 



middle of the plate, enclosed by the com- 

 parison-spectrum ; (3) a mere subjective 

 tendency to set the cross-hair relatively 

 farther to the right on a dark line in a 

 white field than on a bright line in a dark 

 field. The conclusion is reached that the 

 third cause alone really operates, or at least 

 that neither the curvature of the lines nor 

 the relative position of star-spectrum and 

 comparison affect the case. 



Four New Spectroscopic Binaries: "W. W. 



Campbell. 



Four stars have recently been observed 

 with the Mills spectrograph to have vari- 

 able velocities in the line of sight. These 

 are <f Persei, f Herculis, a Bquulei and » 

 Andromedag. The first of these is an inter- 

 esting bright-line star. The second has a 

 large radial velocity, and is moreover an 

 interesting visual double star whose period 

 is about 33 years. This therefore affords 

 another connection between visual and 

 spectroscopic binaries. In the case of the 

 third star, a composite spectrum was ob- 

 served a few years ago by Miss Maury at 

 Harvard College Observatory. Thirty-two 

 spectroscopic binaries discovered with the 

 Mills spectrograph in the past three years 

 had previously been announced, thus bring- 

 ing the number up to 36. On the list of 

 suspected binaries are 14 stars, awaiting 

 confirmation. Before the discovery of 

 these binaries, 3 had been found in the 

 same list of stars by Belopolsky, making 

 about 40 spectroscopic binaries in 325 stars 

 observed. The proportion is therefore one 

 spectroscopic binary for every eight stars 

 observed. The variable velocity of our sun, 

 due to its revolving planets, has a double 

 amplitude of only a few hundredths of a 

 kilometer. As the work progresses and the 

 degree of accuracy attainable increases, we 

 shall probably find that there is a regular 

 gradation of double amplitudes from' that 

 of our sun up to those of the spectroscopic 



binaries already discovered, and it is pos- 

 sible that the star that is not a spectro- 

 scopic binary will prove to be the rare ex- 

 ception. This field of investigation is one 

 of extreme richness. 



Discovery of Five Hundred New Double 



Stars: W. J. Husset. 



While observing the Otto Struve double 

 stars the writer discovered new companions 

 to five of them and also picked up several 

 new pairs in the vicinity of others. This 

 led to the notion of making an extended 

 search for new pairs, which was begun as 

 soon as it was possible to do so without in- 

 terfering with the work in progress. In the 

 spring of 1899 search was commenced in a 

 tentative way and in July of the same year 

 it was taken up regxilarly, and since then 

 has been conducted in a systematic manner. 

 While only a part of the writer's time has 

 been devoted to the consideration of new 

 double stars, he has now discovered and 

 measured five hundred new pairs having 

 distances under five seconds. The work has 

 been done with both the twelve- and thirty- 

 six-inch telescopes of the Lick Observatory. 

 Many close and difficult pairs, some of 

 them having distances less than a quarter 

 of a second, have been found with the 

 smaller instrument, but nearly all the 

 measures have been made with the great 

 telescope. The classification of the new 

 pairs with respect to the distances between 

 their components is as follows: 



0".25 or less, 



.26 to 0".50, 



.51 to 1 .00, 



1 .01 to 2 .00, 



2 .01 to 5 .00, 



37 pairs, 



96 pairs, 



112 pairs, 



112 pairs, 



143 pairs. 



Seventy-one per cent, of the total num- 

 ber have distances under two seconds; 49 

 per cent, under one second; 27 per cent, 

 under half a second; and seven and one- 

 half per cent, under quarter of a second. 



