on our icnowleugic of spectrum analysis. 275 



§ 2. On the Influence of Teju'Eratuke and Pressure on the Spectra 

 OF Gases. By Dr. Schuster, F.B.S. 



A study of the changes which may be observed in the spectra of gases, 

 under varying circumstances, is of great importance, both from a practical 

 and a theoretical point of view. We are here chiefly concerned with the 

 practical side ; and it is clear that a full investigation of all spectroscopic 

 variations attending changes of physical conditions, will ultimately lead to 

 a science which will aim, not only at a merely qualitative analysis, as the 

 original spectroscopy did, but which will enable us to determine the exact 

 physical state of a luminous body, at whatever distance from us that body 

 might be placed. 



There is some difficulty in arranging the great quantity of partially 

 unconnected facts which we shall have to place before the reader. We 

 shall endeavour, for clearness' sake, to arrange our material under five 

 difiereut heads. We shall first discuss what changes we have a right to 

 expect in the appearance of a spectrum, if the quantity of luminous 

 matter is increased, or if the temperature is raised, the absorbing pro- 

 perties of the gas remaining unaltered. We shall next speak of the 

 widening of lines, which, as we shall see, often accompanies an increase 

 of pressure. Then we shall treat of the different spectra given by one 

 and the same body at different temperatures, and we shall see how far 

 satisfactory explanations have been offered for their existence. 



So far our road will be clear, but we shall find that these spectra of 

 different orders, as they have been called, are only extreme cases of con- 

 tinuous changes which are nearly always going on. Very often we can 

 refer these continuous changes to a gradual displacement of one spectrum 

 by another ; but often we shall not be able to prove the existence of a 

 second spectrum. There is, a priori, nothing impossible, or even improb- 

 able, in the view that the relative intensity of different lines may be 

 different at different temperatures, and often when we observe a variation, 

 we may equally well explain it by assuming the gradual appearance of a 

 new spectrum, or an alteration only in the relative intensities of the lines. 

 It becomes then a matter of extreme difficulty to decide which of the 

 two suppositions is correct. In doubtful cases we may often be able to 

 obtain important information by means of a method which is little under- 

 stood, even by spectroscopists. It is the method which has first been 

 extensively used and investigated by Mr. Lockyer, of projecting an image 

 of the luminous source, spark, arc, or flame on the slit of the spectroscope, 

 and thus localising the spectra which are thrown and confused together^ 

 if the luminous source is examined directly without the interposition of a 

 lens. We shall see how, by means of this method, we shall often at a 

 single glance be able to tell how the body will behave at different tem- 

 peratures and under different pressures. Many facts which have been 

 quoted as remarkable might have been ft>retold by means of this method. 

 Our fourth chapter will be devoted to it. In our last chapter we shall 

 have to give an account of some changes which have not found a place 

 under the pi'evious heads. 



I. Influence of Thickness of Radiating Lmjer on the Spectra of Gases. 



Let a be the coefficient of absorption for a certain wave-length of a 

 layer of gas, of thickness and density equal to unity. Let e be the 



T 2 



