We investigate the two-dimensional, steady state distributions of solar wind speed and 5-GeV cosmic ray flux with respect to the heliospheric current sheet and their variation with the solar cycle from late 1964 through mid-1982. Synoptic K coronameter data are used to locate the current sheet, taken as the center of the ''band of coronal streamers,'' during years of low and intermediate solar activity. We confirm the conclusion that the profile of the solar wind at 1 AU during these years consists of (1) a minimum in mean speed of

~400 km s-1 at the heliospheric current sheet, (2) a rise to a mean speed of ~600 km s-1 at an angular distance from the current sheet of 20¿ to 40¿, and (3) a plateau of high mean speed of ~600 km s-1 over the polar coronal holes. During those years of moderate solar activity when the current sheet provides a simple organization to the heliosphere, the mean velocities at the current sheet and in the plateau are approximately independent

of phase in the solar cycle. The principal change in the profile is that the breadth of the minimum decreases with subsiding solar activity and reaches its narrowest at sunspot minimun. The profile of the flux of 5-GeV cosmic rays shows a maximum at the current sheet, with a decrease of 2 to 3% out to a displacement in latitude from the current sheet of ~45¿. There is no evidence for change in this gradient with either the level of solar activity or the overall modulation of cosmic ray flux within a given solar cycle. The data are unable to demonstrate the existence of any dependence of either the sign or the magnitude of the gradient with the polarity of the large-scale heliospheric magnetic field. The observed distribution in cosmic ray flux is compared to the predictions of several theoretical models in an effort to resolve the question

of the role played by particle drifts in cosmic ray propagation. Although drift-dominated models with inclined current sheets naturally produce negative gradients, which could be made to coincide with the observations by a suitable choice of parameters, such models also predict a dramatic dependence of the magnitude of the gradient upon the polarity of the large-scale heliospheric magnetic field. This dependence is not apparent in the currently available data.

Diffusion-convection models adjusted to yield a constant radial gradient of cosmic ray flux in the ecliptic also produce a negative gradient with displacement in latitude from the current sheet. However, these latter models require considerable arbitrary ''tuning'' of the latitudinal dependence of the propagation parameters to reproduce the observations. The models which explain recurrent fluctuations in cosmic ray flux at the in-ecliptic consequences of corotating interaction regions (rather than those of a gradient in displacement in latitude from the current sheet) are shown to produce effects quite contrary to the observations. We conclude that a satisfactory, quantitative explanation of the gradient of cosmic rays with respect to the current sheet is not yet available.