INTERESTING SCIENTIFIC WORK IN THE ARCTIC 207 



seen in especially favorable circumstances. The pole star, even 

 if visible in winter, is of no practical utility. After the observa- 

 tion is taken certain corrections must be applied to the observed 

 altitude to convert it to the true altitude. Of these the correc- 

 tion for refraction is the most important. 



The irregular appearance of the disc of the sun near the hori- 

 zon when rising or setting is well known. In the Arctic the 

 sun is always close to the horizon, and its rays in traversing the 

 atmosphere are strongly refracted. This, if not allowed for, is 

 sufficient to cause an error of from two to thirty-six miles, but as 

 its amount is known and tabulated for ordinary conditions such 

 errors are not very likely to occur. 



As the pole is approached the observations become very simple. 

 At the pole the altitude of any heavenly body the sun for in- 

 stance is the same as its declination or distance north of the 

 equator as given in the astronomical tables. An explorer, then, 

 approaching the pole, needs only to take an observation of the 

 sun at noon, and after correcting the observed altitudes for errors 

 of instruments, refraction, etc., to compare his correct altitude 

 with the declination given in the tables. The difference between 

 the declination and the altitude is the distance he is from the 

 pole. As he draws nearer to the pole it becomes of continually 

 lessening importance that his observation be taken exactly at 

 noon, and when at the pole it may be taken at any time and will 

 be the same as the given declination at that time. 



It thus becomes a very simple matter to determine one's posi- 

 tion at the pole. As one's distance from the pole increases the 

 computation requires slightly more time, but still remains very 

 simple. The following method can be employed at any time, but 

 various other methods are equally available: 



The time of the observation is taken by the pocket chrono- 

 meter, and by a comparison with the other chronometers the ex- 

 act time of Greenwich is determined. To this is applied the Ion- 



