THE AID OF THE ACHROMATIC FRINGES. 



19 



and the sensitiveness 8d if 8 (AJV) = io~ 4 cm. is the least micrometer read- 

 ing at 



d= 10 meters 60 = 0.58 cm. 



d= 100 meters 58 cm. 



d = kilometer 5,8 14 cm. 



in view of the small base 52 cm. and small ^ = 4.7 cm. The observed data 

 were well within this. Of course on using the fringes individually these 

 results could be immensely improved. 



The endeavor was now made to work at larger distances d and a larger 

 base b. But the laboratory being surrounded almost on all sides by trees, 

 it was found that only at one upper window was a distance prospect visible, 

 and hence, since it was again necessary to look through a window obliquely, 

 the base was restricted to but 6 = 36.6 cm., not much above a foot. Still, as 

 the values of AJV are proportional to b, the purpose of the experiments could 

 be adequately carried out within a range of about a mile, using the distant 

 hill or horizon for ANo corresponding to d = . Three objects were selected, 

 a school-house gable, a church spire, and a house on the hill. The results 

 were as follows: 



TABLE i. Measurement of larger distances; base 6 = 36.6 cm.; 2^ = 9.4 cm.; i = 45. 



The results for the spire obtained on different days and with different 

 adjustments are as close as the limit of the micrometer (io~ 4 cm.) admit. 

 All were in agreement with the data taken from a city map. With a larger 

 b and a larger R, the accuracy would increase as the product of these 

 quantities, so that the results are entirely satisfactory, notwithstanding the 

 limited range. Measurements from the roof of the laboratory, which would 

 have been very onerous, in view of the improvised apparatus, were therefore 

 abandoned. I may add that by screening off the direct ray (K), the appa- 

 ratus may be adjusted (if out of order) by putting the two images of the 

 reflected (L) rays in coincidence. The fringes with daylight are perhaps 

 not as intense as one would wish and too many are visible, differing in this 

 respect from the case where an intense light and a collimator are used. In 

 these directions further work is necessary. With ordinary plate and verti- 

 cal fringes, the two images will usually be one above the other. Vertical 

 coincidence in such a case is secured by aid of a fine vertical cross-hair in the 

 ocular. Since all coincidences are subject to this, its precise trend is not of 

 importance. Complete coincidence may, however, always be obtained with 

 the use of compensators, as indicated in 10. 



8. Indirect observations. These refer particularly to the actual use of 

 the large angular displacement A0 of the group of fringes in the ocular as 



