Prof. R. Bunsen's Spectral- Analytical Researches. 527 



are similarly dealt with. It is worthy of notice that in the Rev. 

 George Peacock's celebrated treatise on Arithmetic in the Ency- 

 clopaedia Metropolitan^ p. 499, the decimal equivalent for -1 7 - is 

 given as •05882*352941 (having the portion -05882 finite, 

 although 17 is a prime number), whereas the process above in- 

 dicated gave the value as T V= '05882352941] 7647. The latter 

 value has a repeating period of sixteen figures, and is found to 

 be correct. 



This theorem applies to all recurring decimals. 



74 Brecknock Road, N. 

 October 1875. 



[To be continued.] 



LXI. Spectr al- Analytical Researches. By R. Bunsen *, 



[With Three Plates.] 



[Continued from p. 430.1 



II. Spark- spectra, Flame-spectra, and Absorption-spectra 

 of the Elements. 



ff^HE appearance presented in the spectroscope by a con- 

 -*- tinuous spectrum is chiefly dependent upon the breadth of 

 the slit. Homogeneous light appears as a sharply defined line 

 of one colour, whose apparent breadth increases in proportion 

 with the breadth of the slit. If, however, the rays of light 

 proceed from several neighbouring sources, then they appear, 

 arranged in the order of their colours, as a more or less broad 

 band, whose apparent extension is not proportional to the 

 breadth of the slit, but to this breadth plus a constant. Such 

 bands therefore increase in a less ratio than does the breadth of 

 the slit. 



The influence exercised upon the appearance of these bauds 

 by the gradation and maxima of light is also to a great degree 

 conditioned by the breadth of the slit, because the images pro- 

 duced by the widening of the slit become superimposed upon 

 those which are already present. Groups of new lines, which 

 entirely change the habitus of the original spectrum, may there- 

 fore become visible by gradually narrowing the slit : thus in 

 the spark-spectrum of yttrium the peculiar bands appearing 

 in the red are resolved, on narrowing the slit, into a great 

 number of sharply defined lines, which, from the peculiarity of 

 their position and intensity, serve as tests for the presence of 

 yttrium. 



* From Poggendorff's Anndlen der Physik und CJiemie, vol. civ. pp. 

 3G6-384, translated by M. M. Pattison Muir, The Owens College, Man- 

 chester. 



