the axis. (iv) The highly irregular magnetic field in the magnetosheath region contributes an additional pressure that is not included in the usual gas-dynamic analysis. Until further analysis, and perhaps further observations, have been made, these suggestions must be regarded as speculative, and the dynamic soundness of our model must be regarded as uncertain. Other models are possible but unattractive. Most of the features seen near Venus could be regarded individually as features of the

more »

... rplanetary medium even though some of them are very unusual. However, it seems extremely unlikely that all these features would have been observed during the short period in which Mariner was near the planet and that they would have been organized purely by coincidence into the observed pattern. The magnetopause and bow shock observed around Earth can be scaled so that the shock passes through point 1 and very near to point 5. The scaled magnetopause is then in the upper atmosphere at the subsolar point, and the trajectory would be inside the magnetosphere between points 2 and 4. The corresponding scaling of the dipole moment gives a value for Venus about 1/700 that of Earth. Although the presence and position of the shock is an attractive feature of this model, the presence of solar wind inside the scaled magnetopause and the very low value of \BI near event 3 seem to rule out the model completely. An upper bound can be placed on the magnetic moment of Venus, but it depends for its precise value on the details of the model used. The unsuitability of the scaled geomagnetic analogue summarized above requires that the actual ratio of dipole moments be significantly smaller than that derived from the model. We estimate a reasonable limit to be within a factor of two of 10-3 times that of Earth. Obviously, there is no way to tell from such analyses whether or not Venus has an intrinsic field smaller than the upper bound. Nothing that we report here is inconsistent with the available information on the recent Russian observations which appear to give an even lower limit. The interaction of the solar wind with Venus differs from its interaction the axis. (iv) The highly irregular magnetic field in the magnetosheath region contributes an additional pressure that is not included in the usual gas-dynamic analysis. Until further analysis, and perhaps further observations, have been made, these suggestions must be regarded as speculative, and the dynamic soundness of our model must be regarded as uncertain. Other models are possible but unattractive. Most of the features seen near Venus could be regarded individually as features of the interplanetary medium even though some of them are very unusual. However, it seems extremely unlikely that all these features would have been observed during the short period in which Mariner was near the planet and that they would have been organized purely by coincidence into the observed pattern. The magnetopause and bow shock observed around Earth can be scaled so that the shock passes through point 1 and very near to point 5. The scaled magnetopause is then in the upper atmosphere at the subsolar point, and the trajectory would be inside the magnetosphere between points 2 and 4. The corresponding scaling of the dipole moment gives a value for Venus about 1/700 that of Earth. Although the presence and position of the shock is an attractive feature of this model, the presence of solar wind inside the scaled magnetopause and the very low value of \BI near event 3 seem to rule out the model completely. An upper bound can be placed on the magnetic moment of Venus, but it depends for its precise value on the details of the model used. The unsuitability of the scaled geomagnetic analogue summarized above requires that the actual ratio of dipole moments be significantly smaller than that derived from the model. We estimate a reasonable limit to be within a factor of two of 10-3 times that of Earth. Obviously, there is no way to tell from such analyses whether or not Venus has an intrinsic field smaller than the upper bound. Nothing that we report here is inconsistent with the available information on the recent Russian observations which appear to give an even lower limit. The interaction of the solar wind with Venus differs from its interaction with either Earth or Moon. In the case of Moon, the plasma ions are albsorbed by the lunar surface, and no shock develops (10). Moreover, Moon 29 DECEMBER 1967 with either Earth or Moon. In the case of Moon, the plasma ions are albsorbed by the lunar surface, and no shock develops (10). Moreover, Moon 29 DECEMBER 1967 appears to be a sufficiently good insulator to allow the interplanetary field to be convected nearly unchanged through it (11). Except for a region close to or inside the lunar shadow, the plasma flow near Moon is unaffected by its presence. In contrast, the plasma flow near Venus is bounded by an anemopause. The shock around Venus resembles that around Earth except in scale, but conditions inside are quite different because the anemopause around Venus is probably supported by the ionosphere, whereas that around Earth is supported by the geomagnetic field. The plasma appears to expand into the cavity on the downwind side of Venus, whereas, behind Earth, inward expansion is prevented by the magnetic field in the geomagnetic tail. appears to be a sufficiently good insulator to allow the interplanetary field to be convected nearly unchanged through it (11). Except for a region close to or inside the lunar shadow, the plasma flow near Moon is unaffected by its presence. In contrast, the plasma flow near Venus is bounded by an anemopause. The shock around Venus resembles that around Earth except in scale, but conditions inside are quite different because the anemopause around Venus is probably supported by the ionosphere, whereas that around Earth is supported by the geomagnetic field. The plasma appears to expand into the cavity on the downwind side of Venus, whereas, behind Earth, inward expansion is prevented by the magnetic field in the geomagnetic tail.