Mr. Pogson on the Madras Celestial Survey. 93 



minute in thickness, so as always to be visible without illu- 

 mination of the field. With this reticle right ascensions are 

 noted to the nearest second of time, and differences of 

 declination to the nearest half minute of arc. Greater 

 accuracy is neither attempted nor desired for the object 

 in view. The usual width of a zone is half a degree, 

 though in certain parts of the heavens in which stars are 

 less numerous than elsewhere, this may be increased to forty 

 or even fifty minutes : while on the other hand, in more 

 densely sprinkled parts, zones of twenty or even ten minutes 

 width may be found most convenient. 



Although the actual number of starlight nights through- 

 out the year is decidedly less at Madras than in England, 

 their equability and superiority in regard to clearness and 

 steadiness of vision is incomparably in favor of the tropical 

 skies. The fine weather comes altogether, from January to 

 May, with a few breaks in the later months of the year ; 

 and it must be remembered that zoning near the horizon is 

 safe and easy here, when in England it would be rarely pos- 

 sible at double the altitude. 



Construction and Revision of Charts.— My own practice 

 in this matter is, I am aware, quite at variance with the 

 views of many who consider duplicate or triplicate zones re- 

 quisite to guard against errors. I can only express my opi- 

 nion founded upon long experience, that such reobservation 

 of the same ground is mere waste of time, and no guarantee 

 of the accuracy of the maps. My plan of proceeding is this. 

 A zone of stars having been carefully observed, including 

 all that time will possibly admit of being recorded as they 

 pass through the field of view, and the reductions to epoch 

 having been obtained from the preliminary catalogues, all 

 stars, new and old alike, are inserted in the manuscript 

 chart in pencil, the estimated magnitude being entered 



