618 



SCIENCE. 



[N. S. Vol. III. No. 69. 



give the instrument greater freedom the 

 clamp arm was detached from the pier, ex- 

 cepting for a few of the earlier observations. 

 Screens of fine brass wire were used to re- 

 duce the apparent magnitude of the brighter 

 stars so as to make them more comparable 

 with the fainter stars. The screens were 

 mounted on a frame travelling north and 

 south and entirely separate from the instru- 

 ment. They were used not for fear of the 

 errors arising from momentary uncertainty 

 on the part of the observer, but for fear of a 

 systematic change in his habit of noting the 

 bisections of the brighter stars. Such a 

 change might come about gradually during 

 the six months' interval between two suc- 

 cessive epochs of observation and would 

 enter directly into the concluded parallax. 

 To prepare each observation of a group of 

 three stars for combination with other ob- 

 servations of the same stars a simple reduc- 

 tion is made. The differences of the ob- 

 served times are corrected for deviation of 

 the instrument from the meridian and for 

 proper motion of the stars so far as they are 

 known, and then reduced to a common 

 equinox. The effect of the clock rate in well 

 selected groups of stars is rarely apprecia- 

 ble. The solution is then made so as to de- 

 termine three unknown quantities, namely, 

 the normal difference in time between the 

 middle star and the point exactly midway 

 between the first and third stars, the resi. 

 dual correction for proper motion and the 

 parallax. 



The method has certain distinctive ad- 

 vantages and disadvantages to be foreseen 

 which may here be noted. The former are 

 as follows: 1. The absence of any known 

 systematic effect of refraction, thus avoiding 

 any refraction term whatever in the reduc- 

 tions. 2. The simplicity of the observations 

 and reductions and the rapidity with which 

 the former may be secured. 3. The great 

 freedom allowed in the choice of comparison 

 stars as regards distance from the principal 



star in zenith distance. 4. The stability of 

 the instrument and the fact that it is un- 

 touched at the moments of actual observa- 

 tion. 5. The ease with which the condition 

 may be secured that all observations on a 

 given star shall be made with the same 

 position of the instrument and of the ob- 

 server. As compared with one or another 

 of the modern, refined methods of measur- 

 ing stellar parallax, the following advan- 

 tages may also be given : 6. A large di- 

 mension of the parallactic orbit is always 

 measured. 7. All observations are made 

 at the same place in the field of the ej'e- 

 piece. 8. The attention of the observer is 

 directed to one point only at a time. 



The disadvantages are as follows : 1 . 

 Limitation to meridian passages, so that 

 observations at the time of maximum effect 

 of parallax are in general impracticable 

 through one-half of the year. 2. Limita- 

 tion in the choice of comparison stars, since 

 brighter stars must be selected on account 

 of the smaller apertures of meridian instru- 

 ments. This necessitates moreover greater 

 intervals between the stars allowing more 

 time for disturbances to occur affecting the 

 transits of the stars. 3. The necessity of 

 moving the entire telescope in passing from 

 one star to the next, sometimes requiring 

 a change of several degrees in the pointing 

 of the instrument and incurring the risk of 

 inducing strains among its parts. 4. The 

 fact that the instants of obsei'vation of any 

 two stars cannot be made simultaneous. 



The present observations were made on 

 an illuminated field. In making up the 

 star groups I gave the preference to sym- 

 metry of position over that of magnitude. 

 The observing list has seemed too crowded 

 in some places, but the influence of this and 

 of any other adverse circumstances will be 

 better determined by the final discussion. 

 In order to secure, if possible, a fair num- 

 ber of observations at each epoch, I have 

 continued the observing in general on poor 



