08 On the Production of Spectra by the O.ryhydrogen Flame. 



excess of hydrogen at the high temperature of the flame seemed 

 to reduce the oxide of calcium, and an atmosphere of vapour 

 of metallic calcium partially enveloped the cylinder, and hence 

 the double spectrum. 



Following the suggestion, I soon found a simple way to 

 eliminate entirely the continuous spectrum, leaving the other 

 remarkably bright, sharp, and persistent. I accomplished this 

 by taking the light from a point above or quite outside of the 

 incandescent point of lime. After various experiments, the 

 method adopted for producing these results most satisfactorily 

 was to prepare artificial cylinders perforated centrally, the 

 opening being J inch in diameter ; this was slipped on the jet, 

 which was a double one, the two jets of gas playing upon oppo- 

 site sides of the inner surface of the hollow cylinder. This 

 threw a flame directly upwards, from which the spectrum was 

 obtained. 



The incandescent points being within the cylinder, no con- 

 tinuous spectrum was possible ; and the results produced by 

 this device were in the highest degree satisfactory. 



It was evident, if lime could thus be reduced and the 

 metal volatilized, that other oxides could be similarly treated ; 

 I therefore had hollow cylinders of strontia prepared, from which 

 I obtained the strontia spectrum with far greater brightness 

 and persistency than is possible by ordinary means. Indeed, so 

 far as my experiments have gone, I see no reason why this 

 method is not applicable to most of the metallic oxides. In 

 experimenting, however, with an easily fusible oxide, like that 

 of copper, I have found it necessary to combine it with some 

 very refractory substance, like alumina, in order to prevent 

 fusion. 



In applying this method to the projection of spectra, the 

 results were very satisfactory, and they warrant the expecta- 

 tion that the projection of most spectra may be effected in this 

 way. My experiments in this direction have been limited for 

 want of time. 



An attempt to photograph the lines projected to a distance 

 of 2 or 3 feet was sufficiently successful to satisfy me of the 

 practicability of the operation. 



I should say that having called the attention of my friend 

 Professor A. K. Eaton to these facts, he has willingly lent his 

 aid in the prosecution of these experiments, particularly in the 

 line of projection. 



Hoboken, New Jersey, Dec. 16, 1875. 



