424 C. PIAZZI SMYTH ON MICROMETRICAL MEASURES OF 



linelet, in proceeding from the red to the violet end, and of decreasing distance 

 at the same time between the leading lines, or rather pairs of lines in each 

 band, — as could not have arisen from mere accidental error, or blind fancy; 

 and did seem to testify to a considerable portion of the interesting arrangements 

 of Nature, in this branch of her handiworks, having been secured on this 

 occasion in linear record. 



If the leading lines too in each band, notwithstanding their faintness and 

 haziness, have been represented by me as double, though single to all former 

 observers, we may find strong confirmations thereof in our next subject, viz., 



Advantages of Electric- Lighted Gas Vacuum Tubes. 



These tubes are usually illumined for observation by that variety of electri- 

 city contained in the spark from an Induction coil, actuated by a Bichromate 

 Galvanic battery. Such was the kind employed here, and was capable of 

 giving sparks from 2 to 4 inches long in the open air : the size of the immersed 

 portions of the plates being 4 //- 2 by 4"'6 ; and the number of cells 12, but only 6 

 of them having fresh exciting flnid each observing night. The coil was by Mr 

 Apps, said to be a 6 inch spark coil if used with five quart-sized Bunsen, or 

 Grove, cells. It was also furnished with a quantity, as well as an intensity, 

 primary ; one or other to be employed alternatively ; but after many trials with 

 little or no difference on spectra, I settled down to the intensity arrangement 

 alone, and to carefully attending to the state of the spring brake ; its freedom 

 from oxidation ; or from becoming self-soldered, and being in the best state of 

 strain for illuminating sparks, without stopping dead. 



Employing these sparks then on vacuum tubes, they may of course be 

 expected to show higher temperature effects than mere blow-pipe flame. But 

 they have other advantages over both that method of incandescence, and 

 induction electric sparks in the open air, whether taken direct from the 

 coil or with the interposition of a condenser, in the shape of " the jar dis- 

 charge." 



To illustrate the nature of these advantages for our particular purpose of 

 Micrometrical measures of precision, I submit on Plate LII. four views of the 

 salt lines (D 1 and D 2 of the Solar spectrum) rendered incandescent in as many 

 different modes, and viewed under high dispersion. 



No. 1 is the effect of burning a solution of Na (chloride of Sodium) on a 

 small spiral of iron wire in the flame of a Bunsen burner of coal-gas and air, in 

 the usual manner of all ordinary Spectroscopists. The effect will be seen to 

 be broad, dull and hazy in the extreme ; there is much continuous spectrum 

 paling the lines, and the outer envelopes of vapour of the flame produce on 

 each line an inversion of the direct action; or cause a black line to run down 



