290 Scientific Proceedings, Royal Dublin Society. 
loc. cit.), were mostly of a very feeble character; “the exposure 
was one hour.” ‘‘ No attempt was made to purify the coal-gas, as 
the object of examining this spectrum was to determine the origin 
of any lines which might be caused by hydrocarbons in the oxy- 
hydrogen flame.” 
The oxyhydrogen blow-pipe was used in subsequent experi- 
ments for obtaining metallic spectra. It was not intended that the 
lines observed in the oxy-coal-gas flame should be considered as 
having been accurately measured, it is quite obvious indeed that 
the conditions necessary for their observation rendered the usual 
degree of accuracy impossible, on account of—first, the small 
dispersion, particularly of the less refrangible rays; secondly, 
the “ somewhat wide slit ’’; and thirdly, the feeble spectrum.! 
The wave-lengths assigned to the lines and edges of bands are 
expressly stated to be merely approximations, which were to serve 
for the recognition of the same lines and bands, in case they were 
observed in photographs taken subsequently under other condi- 
tions. It could hardly have been anticipated that this would be 
misunderstood if the original publication was consulted, seeing 
that at the head of the description it is so prominently stated that 
the measurements were approximations. 
Messrs. Eder and Valenta, in the following words, state what is 
decidedly, though no doubt quite unintentionally, misleading :— 
“Hartley glaubte die Kayser und Runge’schen Cyanbanden 
A= 4215, 4208, &e., des elektrischen Kohlenbogenlichtes in 
der Oxygen-Leuchtgasflamme zu finden. Dagegen ergeben 
unsere Messungen zweifellos, dass die mit X = 4216 beginnende 
violette Cyanbande mit der violetten Leuchtgasflammen-Spectral- 
bande (Z) gar nichts gemein hat, und nur ungenaue Messungen 
konnen zu solchen irrigen Schlissen fihren.” 
Now, between wave-lengths 3893 and 3487°8 there are eighteen 
bands, and opposite to each is the wave-length of a band measured 
by Kayser and Runge placed under the head of “ Remarks.” 
There are also three other measurements, each described as a 
1The following quotations give some idea of the definition of lines and bands on 
the photographs, p. 170 :—‘‘ The edges of the bands are as sharp as they are generally 
seen in the spectrum of a Bunsen flame, and the lines of which the bands are com- 
posed are somewhat wide,” p.173. ‘Then appears a series of beautiful, very fine, 
and closely-adiecent lines.”’ 
