ECLIPSING BINARY STARS AS TESTS OF GRAVITY THEORIES - NEW SOLUTIONS FOR AS CAMELOPARDALIS

AS Cam is a double-line spectroscopic and eclipsing binary system whose physical properties and orbital parameters have been previously studied, both spectroscopically {Hilditch [PASP, 84, 519 (1972a)]} and photometrically {Hilditch [MRAS, 76, 141 (1972b)]; Padalia & Srivastava [Ap&SS, 38, 87 (1975)]; Khaliullin & Kozyreva [Ap&SS, 94, 115 (1983)]}. It serves as an important test case for gravity theories {Maloney et al. [AJ, 98, 1800 (1989)]} because the expected general relativistic contribution to its apsidal motion (omega-GR = 8.5 +/- 0.2 deg/100 yr) is both large and comparable to the expected classical contribution (omega-CL = 35.8 +/- 5.8 deg/100 yr). However, its measured apsidal motion (omega-OBS = 15.0 +/- 5.3 deg/100 yr) is significantly less than that predicted from the theory (omega-CL + GR = 44.3 +/- 5.8 deg/100 yr). AS Cam joins DI Herculis, and perhaps several other such systems, in having observed apsidal motions in conflict with the theory. These systems are calling attention to possible problems with our understanding of the classical or general relativistic gravitation at work in close binary star systems {Carroll et al. [BAAS, 22, 1217 (1990); Phys. Rev. Lett. (submitted)]}. Recently, the values (0.14 < e < 0.17) previously obtained for the orbital eccentricity of AS Cam have been questioned by Krzesinski et al. [Inf. Bull. Var. Stars No. 3495 (1990)], with the suggestion that should e lie closer to 0.10, the discrepancy between the observed and the predicted values for the apsidal motion would disappear. In this paper we examine this possibility analyzing the available photometry and spectroscopy with the Wilson-Devinney method of differential corrections described by Wilson [ApJ, 234, 1054 (1979)]. From these analyses we find a reasonably consistent set of orbital characteristics and stellar parameters for AS Cam. Of particular importance are the radii of the components, as the value of omega-CL depends critically upon them. We conclude that while the upper limit for the eccentricity should probably be closer to 0.16, there is no evidence within the timings of minima, the photometric solutions to the light curves, or the spectroscopic solutions to the radial-velocity data for AS Cam to support a value of e significantly outside the range 0.14 < e < 0.16. Finally we discuss in an appendix the failure of the method employed by Krzesinski et al. (1990) to determine the apsidal motion for AS Cam.