New measurements of the solar oblateness have given a value for the fractional difference of equatorial and polar radii of (5.0 ±0.7)x 10~5. A corresponding discrepancy of 8 ( of the Einstein value for the perihelion motion of Mercury is implied.It is generally believed that the observed classical excess motion of the perihelion of Mercury agrees 1 with the Einstein value with an accuracy of 1 %. This assumes negligible contribution from a possibly oblate sun. However, the solar quadrupole moment can account for an appreciable part 2 of the excess. The suggestion 3 ' 4 that the whole of the excess might have this origin is in conflict with planetary observations, 3 ' 5 but as much as a 20% contribution might be accommodated. 2 This is a serious uncertainty, for the precession of Mercury's orbit is presently the most interesting of the three classical tests of general relativity. In fact the gravitational red shift 6 is predicted by a wide range of theories, and the gravitational deflection of light is but poorly known. 7 According to the scalar-tensor theories of gravitation, 8 "* 10 the precession of Mercury's perihelion should be less than the Einstein value, by about 10% if several independent observations have been correctly interpreted. 11 " 13 We are reporting measurements of the solar oblateness which indicate that 8 % of the precession may be due to a solar quadrupole moment. This implies an 8% discrepancy in the Einstein value, and lends support to the scalar-tensor theory.The sun may have a sizable quadrupole moment due to an internal rotation with a period of 1-2 days, for the magnetic torque induced by the solar wind on the sun's surface is believed to be great enough to keep the connective surface layer rotating slowly, balancing the viscous torque caused by a rapidly spinning core. 1 ? 14 How is the quadrupole moment of the sun, due to the unseen interior, to be measured? It is determined to a good approximation by the shape of the observed surface of the sun, in a manner independent of the details of the interior. 2 With negligible shear stresses in these outer "seen layers" of the sun, the surfaces of constant P, T, p, and $ all coincide. 15 Here $= cp-^r 2 u) s 2 sin 2 9 is the gravitational potential supplemented by the centrifugal force term of surface rotation. The observed solar oblateness is essentially that of a surface of constant density, hence of constant <£>. The gravitational potential cp is thus determined over this surface and hence, through Laplace's equation, exterior to the sun. This exterior solution determines the quadrupole moment of the sun. The surface layers of the sun do contain weak shear stresses due to turbulence velocities, differential rotation, and magnetic fields. These are observable, and their effects appear to be small. Details will be given elsewhere.The observed oblateness of the sun is designated A = (^eq.~rpole.)/ r -Tne oblateness of a surface of constant gravitational potential at the solar surfaces is A =A-4w 2 r/g :where g s is the gravitational accelerati...

The behavior of the atomic hydrogen maser is analyzed for both stationary and transient operation. An expression for noise in the signal from the maser oscillator is derived by applying the previously developed theory of Shimoda, Wang, and Townes. A variety of relaxation phenomena are analyzed, including effects of chemical reaction with the surface and magnetic field inhomogeneities. Several mechanisms leading to frequency shifts in the maser are also analyzed, including cavity pulling and the Doppler effect.

Up to the present it has not been possible to produce maser oscillations with gaseous atoms due to the weakness of the magnetic dipole radiation matrix elements. However, with sufficiently long interaction times with the radiation field, oscillation can be achieved. Such increased inter

The atomic-beam separated, oscillatory-field resonance technique has been used to study the hyperfine frequency of cesium which is perturbed by collisions with storage box walls. With a wall coating of long straight-chain saturated hydrocarbons, resonances are observed after as many as 200 wall collisions. A theory of the effect of wall collisions on the hyperfine frequency which is in qualitative agreement with experimental results is described. The shape of the resonance curve is analyzed by a detailed consideration of the statistical nature of the wall collision.

No abstract