150 Astronumical Socieli/. 



is stated " till the smallness of the deduced deviation and the op- 

 *' posite direction of it, were thought to be the effect of the errors 

 " of observation; and within narrower limits I have not yet attempted 

 " to bring the instrument, as I have not a catalogue on which I could 

 " rely for the mean place of stars that are well adapted for the pur- 

 " pose, nor a time-piece of sufficient regularity to justify my depen- 

 " dence on it, nor a meridian mark to guide me in the execution of 

 " it." It cannot be expected therefore, that observations made under 

 such unfavourable circumstances can throw much light on the prac- 

 tical solution of that difficult problem, the determination of the 

 longitude ; where the greatest accuracy is necessary. It is under- 

 stood, however, that a new set of instruments has lately been for- 

 warded to the Observatory at Bombay, at the expense of the East 

 India Company ; and the public therefore may soon expect to reap 

 the benefit of so laudable an exertion of patronage and public spirit. 

 This was followed by the reading of a paper by Professor Lit- 

 trow. Director of the Imperial Observatory at Vienna, " On the de- 

 termination of azimuths by observations of the pole-star." The prin- 

 cipal object of this communication is to show that the observations 

 of the pole-star, not merely at the times of its greatest elongation, 

 but at any points in its diurnal revolution, may be advantageously 

 employed in the determination of azimuths. After Mr. Littrow has 

 explained his process of observation he investigates the theorems 

 requisite in the computation. Let t be the mean of the times of ob- 

 servation, T the corresponding sidereal time, the horary angle of 

 the pole-star s = T— apparent ill, p be the apparent polar di- 

 stance, (p the elevation of the pole, A the arithmetical mean of the 

 azimuths of the star read off the instrument, while O is the mean 

 of the corresponding azimuths of the terrestrial object chosen. 

 Then the theorem for the azimuth w of the star, at the time of the 

 middle of the observations, is 



cos ip 



and the azimuth of the terrestrial object isa = w-f (A — O). 



The determination of the values of ow and J^w, which depend 

 partly upon series, is explained in the paper, as well as the mode of 

 determining them by a small subsidiary table easily computed. 

 The application of the method is then shown by some exan^ples 

 from observations in July last year, in which two distinct results of 

 observations with the face of the circle towards the east and towards 

 the west, agree within half a second. 



A communication from G. Dollond, Esq. was then read, giving 

 an account of a singular appearance observed during the solar 

 eclipse on the 29th of November last. The morning was cloudy, 

 but soon after the commencement of the eclipse there was a partial 

 opening in the clouds, through which Mr, D. saw a considerable 

 part of the limb of the moon ivhich had not yet entered on the disc of 

 the sun. Continuing his observations, after a short time as the clouds 

 passed on, he again saw both the sun and a portion of the moon's 

 border ivhich was off" the sun's disc. The sky then became cloudless, 

 and he could no longer discern any part of the moon's limb, except 



that 



