88 Dr. W. M. Watts on the Spectrum 



II. On the Temperature of the Flame, and the differences observed 

 at different times. 



It is in the flame obtained on adding the highly manganiferous 

 Spiegeleisen that the manganese-oxide spectrum is most fully 

 developed. In the experiments at Crewe a marked difference 

 was almost always observed between the ordinary Bessemer- 

 spectrum and that of the Spiegel, arising from a difference in the 

 relative intensity of the different lines. The difference was so 

 marked that it was not at first perceived that the two spectra 

 were in any way the same ; but at Barrow this difference does 

 not exist, the Spiegel-spectrum being identical with the Bessemer, 

 only more intense. The ordinary Bessemer-spectrum at Barrow 

 is in fact identical with the Spiegel- spectrum as obtained at 

 Crewe. 



A reference to fig. 2, Plate V. (which represents the two 

 spectra on the scale of a one-prism spectroscope), will render 

 this difference clearer. 



It will be observed that in the Barrow spectrum there is a 

 more marked division of the separate lines into groups having a 

 general resemblance, the group y (65 to 67) being the brightest; 

 while in the Crewe spectrum the brightest portion is nearer the 

 red (58 and 59), and the spectrum does not extend so far into 

 the blue. This difference is probably due simply to difference 

 of temperature ; but it can be distinctly connected with the kind 

 of iron employed : the steel at Crewe was intended for the axles 

 and tires of wheels ; that made at Barrow was to be employed for 

 rails. The same difference was remarked in observing two 

 " blows " at the Bolton Iron and Steel Works. The first was of 

 iron of No. 1 brands, and the second of inferior iron. The 

 spectrum in the first case was the same as the Crewe one, but 

 the Spiegel-spectrum was the same as that observed at Barrow ; 

 in the second " blow " the spectrum was nearly that of Barrow. 



An attempt was made to determine the temperature of the flame 

 by observing whether wires of gold, platinum, and of an alloy of 

 90 per cent, platinum and 10 per cent, iridium were melted 

 when held in the flame. It was found that towards the middle 

 aud end of the blow the gold was always melted ; but on no oc- 

 casion was either the alloy or the platinum melted, even when 

 the wire was kept in the dame for several minutes. At the 

 beginning of the blow, however, gold does not melt. A wire 

 held in the flame from the commencement was not melted when 

 the sodium -line appeared flashing across the continuous spectrum, 

 but did melt about the time that the sodium-line became con- 

 stant, and before any of the Bessemer lines proper made their 

 appearance. If we take the melting-point of gold to be 1300° C. 



