SANTIAGO OBSERVATIONS. 



xlix 



the trapezium in the sword-handle of Orion was distinctly seen whenever sought at rea MI able 

 altitudes; it separated unequal and close double stars, as A Octantis, with great precis., >n and 

 sharpness, and defined the millions in the gorgeous cluster surrounding 47 Tucanse with won- 

 derful effect. An aggregation of star-dust, as appears this most extraordinary object, a power 

 of 235 almost separates each brilliant particle from the rest. 



LATITUDE. 



Twelve hundred and sixty observations have been discussed for the latitude of the meridian 

 circle, from which only nine observations of the twenty-two selected stars have been excluded, 

 and on each of these occasions there is doubt of the true circle-reading. 



Bessel's refractions, as constructed by Professor Coffin, U. S. N., for the Washington Obser- 

 vatory, have been used in the computations. A special table for barometric corrections, ren- 

 dered necessary by the elevation of the observatory above the sea level, has been so formed 

 that all the logarithms remain positive on subtracting .02306 from Log. R of Table II. 



All the observations were reduced to mean places of the stars January 1st, 1851, and the 

 declinations at that date have been derived from the British Nautical Almanac for 1858, by 

 application of the annual variation in declination during 7 years. 



From quick moving stars. 



No recent observations appear to have been accessible from which to correct the mean places 

 of * Eridani and < Argus subsequent to the publication of the Nautical Almanac for 1851, 

 (in 1847,) and as one certainly is, and the other may, very probably, be in error, they have 

 not been incorporated in the mean. Giving to each of the other stars weight proportionate to 

 the number of observations, the latitude resulting would be S. 33 26' 25". 27 with a probable 

 error of 0".042. 



H 



