atmospherical refraction at Port Bowen. 
229 
On looking over each individual's observations, it will be 
seen, that great changes in the amount of atmospherical 
refraction took place, without any correspondent change in 
the state of either the barometer or thermometer ; and, 
although the mode of observation adopted by us, is not 
wholly free from objection, inasmuch, as the ray of light 
from a bright star may suffer some degree of inflection, by 
passing over a sharp edge (such as the boards placed edge- 
wise would present, whereby their apparent altitudes would 
not be exactly those of the stars at the time of observation ) ; 
yet we do not consider this circumstance the cause of the 
anomaly alluded to, for we never entertained the slightest 
doubt as to the moment of either of the stars* disappearance, 
both being always instantaneous : and, moreover, when it is 
recollected, that the use of instruments, proper for measur- 
ing altitudes on these occasions, in such a climate, is attended 
with the difficulties already described in this Paper, it will, 
in all probability be admitted, that this mode of observation, 
is at least, calculated to diminish the errors necessarily arising 
from the use of instruments, under such circumstances. 
It is, however, with diffidence that we submit the following 
tabulated results of the preceding observations, for compari- 
son with the various theories, which have from time to time 
been advanced by many eminent astronomers and mathema- 
ticians, to account for all the irregularities which have been 
noticed in the most careful observations on this important 
subject. 
