60 PROFESSOR W. N. HARTLEY OH SPECTRUM PHOTOGRAPHY 
Borates and silicates .—When soluble borates or boracic acid, soluble silicates, 
silicofluorid.es, or bydrofluosilicic acid are submitted to the spark, line spectra of the 
elements boron and silicon result, as I have already shown in a paper submitted at 
a recent date to the Royal Society. I then showed reason for suspecting that several 
lines attributed to carbon by Professors Livetnu and Dewar (Proc. Roy. Soc., vol. 
xxxiii., p. 403) are in reality lines of silicon. In their communication, to which I 
have had to refer for a description of the cyanogen spectrum, I tind ( loc. cit., vol. 
xxxiv., p. 123) a list of lines seen in the arc spectrum which they assign to carbon. 
I give for comparison with these the wave-lengths of the most prominent lines in the 
spectrum of silicon. 
Approximate wave-lengths 
Wave-lengths of 
of carbon arc-lines 
silicon lines 
(Liveing and Dewar). 
(Hartley) . 
2434-8 
2435-5 
2478-3 
. « . 
2506'G 
2506-3 
2514-1 
2513-7 
2515-8 
2515-6 
2518-8 
2518-5 
2523-9 
2523-5 
2528-1 
2528-1 
288F1 
2881-0 
From this it seems probable that the only carbon line in the ultra-violet arc 
spectrum is the strongest of the series with wave-length 2478’3. 
On spectra obtained from dilute solutions and on alterations caused^ by prolonged 
exposures. 
The effect of prolonged exposure of the sensitive plate to the spark produced from 
weak solutions is a lengthening and strengthening of the metallic lines. It is 
reasonable to suppose that the density of the lines, or otherwise the intensity of 
chemical action, is a function of the concentration of the liquid and the period of 
exposure in all cases, but with very dilute solutions this certainly appears not to 
be exactly the case. Thus a solution containing six per cent, of calcium chloride 
exposed for half a minute, yields a spectrum with the calcium lines, showing a 
greater density than that produced by a solution containing one per cent, of the salt 
exposed for three minutes. Probably the conducting surface of the electrodes is not 
large enough to be kept constantly moistened, and so maintained in a condition 
capable of presenting a sufficient quantity of the solution to the spark. It has been 
observed in ail cases that the effect of diluting solutions is to weaken tire metallic 
lines; and, further, with one or two exceptions, to shorten even the longest and 
strongest lines until they finally disappear. 
