Cessation of X-ray Pulsation of GX 1+4

We report results from our weekly monitoring campaign on the X-ray pulsar GX 1+4 with the {\em Rossi X-ray Timing Explorer} satellite. The spin-down trend of GX 1+4 was continuing, with the pulsar being at its longest period ever measured (about 138.7 s). At the late stage of the campaign, the source entered an extended faint state, when its X-ray (2-60 keV) flux decreased significantly to an average level of $\sim 3 \times 10^{-10} ergs cm^{-2} s^{-1}$. It was highly variable in the faint state; the flux dropped to as low as $\sim 3 \times 10^{-11} ergs cm^{-2} s^{-1}$. In several observations during this period, the X-ray pulsation became undetectable. We can, therefore, conclude conservatively that the pulsed fraction, which is normally $\gtrsim$ 70% (peak-to-peak), must have decreased drastically in those cases. This is very similar to what was observed of GX 1+4 in 1996 when it became similarly faint in X-ray. In fact, the flux at which the cessation of X-ray pulsation first occurred is nearly the same as it was in 1996. We suggest that we have, once again, observed the propeller effect in GX 1+4, a phenomenon that is predicted by theoretical models of accreting X-ray pulsars.