188 Scientific Proceedings, Royal Dublin Society. 
one must distinguish between visible and photographic intensity ; 
and so far as the latter is concerned, the intensity is purely a 
question of what plates are used. Probably the best plan would 
be to adopt the intensities of the gold spectrum as the standard,. 
and compare other spectra with the lines of gold: this would give 
the standard of intensity on the same plate, with the spectrum 
under consideration, and photographed in the same way. It was. 
not possible to deal with the question of intensity in the present 
paper, as it would require a detailed examination of the spectrum 
of each element. The various observers agree very well as to the 
measurements of the wave-lines of the various lines observed ; but 
most serious differences of opinion exist as to the relative intensi- 
ties of various lines; and there can be no doubt that in the spectra 
of many elements, lines have been carefully measured and recorded 
that are really due to impurities. I hope later to investigate a 
number of the coincident, or apparently coincident, strong lines, 
and lines of variable intensity. 
Persistency. 
Professor Hartley was the first to investigate the quantitative 
spectra of the elements, and point out that the persistency of a line 
was much more important than its intensity, and that the most 
persistent lines were not necessarily the most intense. Hartley’s 
method is to take solutions containing 1 part of the element in 100, 
1,000, 10,000, and 100,000 parts of solution, and spark them, 
using graphite electrodes; and he has given complete maps of 
the spectra drawn to a scale of wave-lengths, and also linear 
measurements, showing the successive disappearance of the lines 
of the spectra of magnesium, zinc, cadmium, aluminium, indium, 
thallium, copper, silver, mercury, tin, lead, ‘tellurium, arsenic, 
antimony, bismuth, beryllium, and silicon. 
For purposes of spectrographic analysis of minerals, nothing 
is so necessary as the completion of this work. When we have 
the dilution spectra of all the other elements, it will render the 
spectrographic identification of a substance a matter of rapidity 
and certainty, whether the substance be rare or common, old or 
new, and whether present in large or small quantity. I have 
