KENT. — SPECTRA OF THE ZINC ARC AND SPARK. 97 



General Description and Arrangement of Apparatus. 



An echelon spectroscope and a constant deviation spectroscope of 

 the Hilger pattern were ordered of A. B. Porter of the "Scientific 

 Shop," the echelon having 33 plates, a 1 mm. step, 34 mm. height of 

 plate and about 15 mm. thickness, and the lenses of the constant 

 deviation and echelon spectroscopes being of If" and 2" diameter 

 and 17" and 20|" focal length, respectively. The constant deviation 

 prism proved to be of insufficient aperture to fill the echelon, and was 

 therefore sent to Hilger for a new prism. ^^ 



The echelon itself finall}^ appeared to be a poor instrument and 

 wholly unfitted for first-class work; for, upon final adjustment, the 

 green mercury line X 5471 showed a false pattern and there also ap- 

 peared in certain zinc spark lines a distinct pattern which the writer, 

 in view of the false satellites in the mercury line, at first deemed 

 likewise spurious, inasmuch as a smaller and less powerful echelon 

 made by Petitdidier, and kindly loaned by Professor Goodwin of the 

 Massachusetts Institute of Technology, did not show it. This was 

 later identified with Nutting's "peculiar fluted appearance, character- 

 istic of spark lines ".^^ 



The Porter instrument was finally sent to Petitdidier for over- 

 hauling. Three plates were taken out and all were adjusted so that 

 the step was more uniform. The instrument again showed both 

 patterns, the mercury line pattern being false. Many months were 

 thus lost with these various difficulties. At length it was decided to 

 continue the work with the borrowed Petitdidier echelon, an excellent 

 instrument, although of only 20 plates, total aperture 27 X 15f mm., 

 step I mm. and 14| mm. thickness of plate. 



The apparatus generally emplo^'ed was, then, the Petitdidier 

 echelon and Porter constant deviation spectroscope with a prism 

 fitted by Hilger. 



The spark was generated by a Holtzer-Cabot motor-generator set, 

 the alternator of 4.5 K. W. giving 60 complete cycles per second and 

 feeding a 5 K. W. transformer (of ratio of transformation 110 to 30,000) 

 in the secondary of which was a condenser of 0.0226 microfarads, 

 which discharged, at times through various inductances, over a spark 

 gap generally set horizontal. 



Two methods of producing the arc were employed, one giving what 



15 Professor Porter died within a short time after the instiniment was de- 

 livered. 



16 Nutting, Astrophysical Journal, 23, No. 1, Jan. (1906). 



