Vol. 6, 1920 INTERNATIONAL ASTRONOMICAL UNION 
383 
state of the question it seems undesirable that any particular set of stand- 
ard magnitudes be designated by the adjective "international." Such 
an action on the part of an International Committee would necessarily 
convey the impression of an authoritativeness and finality of decision which 
it would be difficult to justify in view of the present outstanding differences 
in the results of different observers. 
This does not mean that there should be no international standards. 
On the contrary, standards to which this designation may properly be ap- 
plied should be established. But this can be attained only through a 
cooperative effort of the type specified under (2). The question will be 
reverted to later. An additional remark should be made here, however, 
as a supplement to the above considerations, namely, that the provisional 
character even of any act of "international" standards should be empha- 
sized, say, through some such designation as I. C. P. 1920 (International 
Committee on Photometry, Standards of 1920). 
We believe, too, that much conservatism should be displayed in the 
matter of standardization of methods and equipment. As an illustration, 
it is well known that photometric results obtained by visual methods de- 
pend upon the color perception of the eye of the observer. Thus for the 
Harvard durchmusterung {Harvard Annals 4 5) and the durchmusterung 
of Muller and Kempf (Potsdam Pub. 9), both catalogues of visual magni- 
tudes, Schwarzschild finds (Gottingen Aktinometrie B 16) 
H - P = -0.27 - 0.04(P - 6.5) + 0.23 (G - P) 
in which 6^ — P is the color-index obtained by comparing Gottingen photo- 
graphic with Potsdam visual magnitudes. The last term of the equation 
represents the differences which are to be attributed to peculiarities of color- 
perception. In visual photometry naturally there can be no standardiza- 
tion in the sense in which the term is used here. Each observer must 
work with the eye with which he has been provided. 
Of late years, however, photovisual magnitudes have been much used 
as a substitute for those derived visually. An isochromatic plate exposed 
behind a suitably chosen yellow filter affords results which are at least 
approximately the same as those obtained visually. But photovisual 
magnitudes are obviously dependent upon the color-sensibility of the par- 
ticular kind of plate and filter employed, and in this connection it is fre- 
quently suggested that a standard plate and filter be universally adopted 
and used for all photovisual observations. 
Unfortunately the difficulties are not so easily to be avoided. The 
suggestion overlooks a third contributing factor whose significance is not 
generally appreciated, namely, the telescope. The selective absorption 
by the glass of the objective or by the silvered surface of the reflector 
plays a very, important part in determining the color-sensibility of the 
equipment. Moreover, it is not merely a question of refractors as against 
