164 PROCEEDINGS OF THE AMERICAN ACADEMY 



method. In my own experience at Chicago upon the great zones I 

 had uo chronograph, but did not find that a serious drawback. Arge- 

 lander was on this point ultra-conservative ; but I do not think there 

 are many observers by the new method whose work I would take in 

 exchange for his by the older. The temptation with mechanical 

 methods of observing is to undertake more work than can be reduced ;> 

 which is a bad practice. 



In determination of time, Dollen has suggested using the transit in 

 the vertical of Polaris. I have often used this method on a first 

 night's work, but, when the chronometer error is roughly known, can 

 never resist the temptation to bring the instrument at once pretty 

 close to the meridian. The pupil should be thoroughly practised in 

 the minutiae of this process ; for I have seen even a good observer 

 badly vexed with it, when an instrument new to him, especially a poor 

 one, was employed. 



In the determination of latitude by Talcott's method, I think a de- 

 vice of Mr. Rogers is likely to be useful. He replaces the micrometer 

 by a system of parallel lines oblique to the meridian, so that each star 

 must pass two sets of three each in its transit. The lines are very 

 beautifully and exactly ruled by his process, and they save the time 

 required to turn the micrometer screw. If this has to be moved 

 largely, a star might often be lost ; which is the more troublesome, as 

 in doubtful weather, when there are flying clouds, the pairs are often 

 spoilt by losing one of the stars. It is quite probable that Mr. Rog- 

 ers's improvement will enable us to go farther from the centre of the 

 field, and thus help the choice of star-places. 



The stars should always be so chosen that the positive and negative 

 distances from the centre of the field [^ (tJ-|-5') — (jr] may pretty 

 nearly balance, so as to give the means of determining the micrometer- 

 values from the latitude observations themselves ; otherwise the 

 resulting latitudes could not be as accurate as the star-places and 

 observations would permit. 



The foreign astronomers use Talcott's method very little. Their 

 objection to it is, that the star-places must necessarily be worse than 

 for Bessel's or Struve's, or for the employment of a vertical circle 

 or portable meridian circle. These latter instruments, however, are 

 not used in America, principally because our methods were formed 

 independently of the modern Germans, partly because we have no very 

 good dividing-engines, and because they are too delicate to stand 

 transportation over our frightful ^Yestern roads. On the other hand, 

 Talcott's method has greatly helped our astronomers by furnishing a 



