August 3, 1900.] 



SCIENCE. 



189 



G. C. Comstock. If 12th magnitude 

 stars are to be used in the comparisons, can 

 provision be made to secure the observation 

 of the same stars at the different stations ? 



W. W. Campbell. A photogi-aphic plate 

 can be secured from which the accurate 

 positions of these stars can be ascertained. 



G. C. Comstock. Is there time for this 

 before the campaign begins ? 



"W. W. Campbell. I think there should 

 be a carefully prepared plan. 



S. J. Brown. In relation to the obser- 

 vation of differences of right ascension and 

 declination, all the observations will have 

 to be made before we can determine the 

 corrections, because the changes of right 

 ascension and declination will not be in pro- 

 portion to the elapsed time in consequence 

 of the rapid motion of the asteroid. We 

 need to eliminate the systematic errors. 

 You mark your time, and your measurement 

 does not correspond with the time you have 

 marked. That leads to a systematic error. 

 If we can eliminate this I think that obser- 

 vations of great accuracy can be made. In 

 this case it is proposed to select stars which 

 shall not be more than 30 seconds in time 

 away from the planet. In micrometric 

 measures the suggestion is 60", which added 

 to the displacement of the planet amounts 

 to 90". Above that distance there would 

 be a rapid increase of probable errors. The 

 main dif&culty appears to be in the selection 

 of stars down to the 11th or 12th magni- 

 tude. 



W. "W. Campbell. It seems to me that 

 the method of observation of transits is 

 objectionable. With the equatorial tele- 

 scope there is a tremendous personal equa- 

 tion in the observation of stars and comets. 

 Undoubtedly there will be a similar per- 

 sonal equation in the observation by tran- 

 sits of an asteroid and a faint comparison 

 star differing 3 or 4 magnitudes from the 

 asteroid. There are many things that may 

 occur ; the refraction may be changed, or 



the position of the instrument may be 

 changed if we wait a minute or two for the 

 comparison star to come along. Our ex- 

 perience in the observation of comets shows 

 that very much better observations can be 

 secured if we throw overboard the method 

 of observing differences of right ascension 

 and declination. 



G. W. Hough. I disagree as to the time 

 required for the observations, and I have 

 made thousands of observations of differ- 

 ences of right ascension and declination. 

 But one measure of the position angle is 

 better than anything you can get from dif- 

 ferences of right ascension and declination. 



S. J. Brown. We have to shift the mi- 

 crometer 90° in taking the position angle. 

 Suppose the planet to move 1" per minute, 

 the change in the angle is not proportional 

 to the time, and the change in the distance 

 is not proportional to the time. I have 

 come to the conclusion that differences of 

 right ascension and declination can be de- 

 termined with greater accuracy than differ- 

 ences of position angle and distance. One 

 point we wish to agree upon is a uniform 

 method of measurement. Observations 

 cannot well be compared if one observation 

 is made upon one plan and another obser- 

 vation upon a different plan. I have no 

 doubt that the difference in magnitude is a 

 very important factor in determining the 

 position. In my observations of Titan I 

 reduced my observations by the method of 

 least squares, and all the equations gave 

 positive residuals except one ; and that 

 could only be explained upon the theory of 

 a variable personal equation from difference 

 of magnitude. Even with faint stars it 

 would be difficult to use differences of right 

 ascension observed with the chronograph or 

 eye and ear. I think we ought to have the 

 experience of those who have had experi- 

 ence with stars of the magnitude to be used 

 as to their method of treating them. The 

 photographic method I think is subject to 



