REPORT OF COMMITTEE ON OBSERVATORIES 1 39 



All the objects found that are suspected to have a fairly large 

 parallax would be taken up by telescope c. They would be further 

 investigated only in those cases in which a first plate confirmed the 

 large parallax. 



On the plates furnished by this telescope only the principal object, 

 with five or six well chosen comparison stars, would be measured. 

 The work of measuring and reducing would be very moderate, 

 therefore, especially as a reduction with three constants would be 

 sufficient in nearly every case. 



Instrument c would serve for the parallaxes of the brighter stars 

 (say o — 6) and, together with instrument d, for the further inves- 

 tigation of objects of certainly measurable parallax. 



The observatory ought further to be fitted out with (say) ten 

 measuring machines. It is ver}' probable that the observers, though 

 their labors at the telescope would take up only a small fraction of 

 their time, will be unable to make the measurement and reductions 

 keep pace with the production of the photographs. It will be nec- 

 essary, therefore, to procure assistance for them. This, however, 

 will be a question of cheap labor. (See P. S.) 



I have dwelt thus long on the subject because it is practically a 

 new one. On the others I have only a few words to say. 



(2) Fundamental Determination of Right Ascension and Declination. 



The great importance of a fundamental determination of star 

 positions, with extension of these observations by secondary methods 

 to include every star brighter than the seventh magnitude south of 

 — 20°, is evident to any one who has made some study of stellar 

 motion. 



I feel very warmly for the plan of transporting to the southern 

 hemisphere, for a short term of years, one of the reversible meridian 

 instruments of the northern hemisphere. The observations of a con- 

 siderable number of stars with the same instrument at a northern 

 and at a southern observatory cannot but lead to a material reduc- 

 tion of the influence of systematic instrumental error and error of 

 refraction. 



It would seem to me that it would be advantageous to make the 

 determinations of right ascension and declination by two separate 

 instruments, a transit (for which the instrument also used for 

 parallaxes may serve) and a vertical circle. Besides other well 

 known reasons, there is this one : If for the determination of decli- 

 nation we bisect the image of a star by the horizontal wire, its 



