354 M. F. Zollner on the Temperature and 



to pass through Nicol's prisms the sections of which are per- 

 pendicular to one another, so that the two spectra in the field of 

 the spectroscope would consist of lights polarized perpendicu- 

 larly one to the other. A NicoPs prism brought into the ocular 

 of the spectroscope would then alter, according to the known 

 law, the ratio of intensities of the two spectra proportionally to 

 the square of the tangent of the angle of rotation. The method 

 indicated, however, is applicable only on condition that equally 

 great portions of the apparent surfaces of the bodies to be com- 

 pared send their light into the spectroscope, and also that the 

 rays from the two bodies do not suffer different degrees of ab- 

 sorption before they enter the slit. The first condition, as 

 already remarked; will be satisfied by sufficient apparent 

 magnitude of the body; the second, on the contrary, will in 

 general only be realizable for terrestrial substances, because 

 the light of the heavenly bodies only reaches our apparatus after 

 the loss of the light' absorbed by the atmosphere, and, with the 

 exception of the sunlight, is radiated from objects of vanishingly 

 little apparent magnitude. 



To remove this limitation of the method, I may briefly remark 

 as follows. Suppose that in the field of view of the above-men- 

 tioned apparatus there are two spectra close together, the light 

 of which is emitted from bodies at different distances, and the 

 rays from which on their way to the spectroscope undergo differ- 

 ent degrees of absorption. So long as the latter is not elective 

 (that is, effected by coloured transparent media), the intensity- 

 ratio of homologous parts of the two spectra can only be altered, 

 by the circumstances referred to, in an equal degree for all kinds 

 of rays. If, therefore, in the apparatus the brighter spectrum 

 be so far enfeebled that the brightness of any kind of rays 

 whatever corresponds with the homologous brightness in the 

 other spectrum, then all the other homologous parts must agree 

 photometrically in the two spectra, provided that both bodies 

 have the same temperature. If the temperature is different, this 

 accordance does not take place, but, when the two spectra have 

 been made photometrically equal for a definite kind of rays 

 (e. g, those corresponding to the line D), the more refrangible 

 rays of the spectrum belonging to the hotter body will prevail 

 over the homologous ones of the cooler body. The intensity- 

 ratios which here take place are temperature-functions which 

 must result from KirchhofPs function. But even without the 

 knowledge of the latter, this method, applied to the stars, would 

 permit us to determine, at least qualitatively, their tempera- 

 ture-ratios — that is, to decide which of two stars possesses 

 the higher temperature. With my astrophotometer, before the 

 ocular of which a small direct-vision prism was placed, I have 



