TABLE 868.— STELLAR RADIATION MEASUREMENTS* 759 



Radiometric magnitude of any star = visual (or photographic) magnitude of a spectral 

 class A star giving the same radiometric deflection. If »;,-, m P ,-, and m pg are, respectively, 

 radiometric, photovisual, and photographic magnitude, then Color Index, CI = (m pg — 

 m pv ) ; heat index, Hl pv = >n P v — m r ; Hl pg = m pg — m r . Spectral class: Henry Draper, 

 revised by D. HofHeit (DH); by W. W. Morgan (WWM). All measures reduced to 

 zenith at Mount Wilson ; two reflections from fresh silver ; zinc-antimony black thermo- 

 junction; rock salt window. Stars of known or suspected variability are rejected from 

 this list. 

 All the stars were in both the Mount Wilson and Harvard observing programs. 285 

 The reduction of the Mount Wilson and Harvard data to a common basis has been 

 rather difficult. The following are the principal factors that differ between the Mount 

 Wilson and Harvard observations. 



(1) The Atmosphere. — There was more water vapor over Oak Ridge than Mount Wil- 

 son ; hence, early-type stars would be too faint at Oak Ridge. 



(2) The thermocouple blacking. — Probably the surfaces were equally "black" in the 

 ultraviolet and visible regions; the Harvard surfaces were blacker in the infrared; 

 hence, late-iype stars would be too faint at Mount Wilson. 



(3) The cell window. — Rock salt was used at Mount Wilson ; fluorite was used, at Har- 

 vard. These are equally good throughout the ultraviolet, visible, and infrared to 

 the region of 6 to 8 microns. For longer wavelengths, rock salt is better. The effect 

 of this difference is in the opposite direction to the thermocouple blacking in (2) 

 above. However, the very small percentage of stellar energy beyond 8 microns and 

 absorption bands in the earth's atmosphere means that the difference in the cell 

 windows has a very much smaller effect than the thermocouple blacking and, there- 

 fore, (2) above dominates. 



A systematic difference exists between the Mount Wilson and Harvard observations 

 which follows a pattern predicted in accordance with factors (1) and (2) above. There- 

 fore, corrections which are usually less than 0.1 magnitudes have been applied. The largest, 

 0.16 magnitudes, is for 51 Gem. This correction brings the two sets of data into better 

 agreement but there remains an apparent difference in zero-point of about 0.13 magnitudes. 

 Since it is impossible to determine which of these two sets of observations is in error, the 

 mean of the Mount Wilson and Harvard data has been taken, corrected as indicated for 

 factors (1) and (2) above. These mean values are the data given in the m r column. 



* Prepared !>y R. M. Emherson, Research and Development Board, Washington. D. C. 



2HG Pettit and Nicholson, Astrophys. Journ., vol. 56, p. 295, 1922; vol. 68, p. 279, 1928; vol. 78, p. 320, 1933. 

 Stern and Emherson, Astrophys. Journ., vol. 94, p. 412, 1941. 



TABLE 869.— NONGALACTIC NEBULAE 



Some 400 considered. Distribution of magnitudes appears uniform throughout sequence. 

 For each stage in the sequence the total magnitude (Mr) is related to the max diameter 

 (d) by the formula : Mr = C-5 log d. When minor diameter is used, C approx constant 

 throughout sequence (C=10.1). Mean absolute visual magnitude — 15.2. The statistical 

 expression for distance in parsecs is log I) = 4.04 -f 0.2 Mr. Masses appear to be of the 

 order of 2.6 X 10 s X our sun's. Apparently nebulae as far as measured are distributed 

 uniformly in space, one to 10 17 parsecs 3 or 1.5 X 10~ 31 in cgs units. 



Corresponding radius of curvature of the finite universe of general relativity is of order 

 of 2.7 X 10 10 parsecs, about 600 times the distance at which normal nebulae can be detected 

 with the Mount Wilson 100-inch reflector. 



SMITHSONIAN PHYSICAL TABLES 



