One year of magnetic field data from the geostationary spacecraft ATS 6 have been analysed for effects associated with the equatorial plane components of the interplanetary magnetic field (IMF). It is shown that perturbation fields in the Y (dawn to dusk) direction appear in association with the Y component of the IMF, in agreement with previous theoretical suggestions. On average a fraction 0.28 ± 0.02 of the IMF Y field appears at geostationary orbit, such that the average ATS 6 B y field is 1.9 ± 0.4 nT larger when IMF B y is positive than when it is negative. The perturbation field magnitudes are also found to depend strongly on local time, however, with largest effects appearing in the midnight and dawn quadrants, where the average perturbation fields are nearly half the simultaneous IMF B v . field. At noon this fraction drops to one fifth, and no average effect occurs in the dusk quadrant. Both the daily mean perturbation fields and the diurnal modulation are also found to depend upon the level of magnetic disturbance as measured by K P , or equivalently upon IMF B z , and upon season of the year. Overall stronger daily mean perturbation fields occur when K P is low or when IMF B z is positive, than when K P is high or when IMF B z is negative. This effect is not linear, however, and there is also a trend in the data towards increasing perturbation fields with IMF B z negative and decreasing. On dividing the data according to season, increasingly strong daily mean effects are found in the order winter, summer and equinox for both quiet and disturbed magnetospheres. Diurnal modulations of the perturbation field magnitudes for low K P (IMF B z > 0) take the form of large amplitude quasi-sinusoidal variations about mean values which are very marked in the equinox data, are present to a lesser degree during summer and are absent during winter conditions. When K p is high (IMF B z < 0) significant deviations from mean perturbation field values occur generally only during nightside hours and little seasonal dependence is evident. Finally, it is shown that the highest correlation between the IMF data and the ATS 6 perturbation fields occurs with zero time delay between the two data sets, showing that a prompt response to IMF conditions occurs at geostationary orbit within the 1 h time resolution available in this study. Although many details of the above ATS 6 response remain to be understood, these results overall demonstrate in a very direct manner the magnetically “open” nature of the Earth's magnetosphere.