It has been argued that there was a period of prolonged ENSO conditions between 1990--95 so anomalous that it is highly unlikely to be due to natural decadal-timescale variation [Trenberth and Hoar, 1996>. This conclusion follows from their study of the Darwin sea level pressure anomaly record, which found that the 1990--95 period would occur randomly about once every 1100--3000 yrs. Taking into account the uncertainty in number of degrees of freedom in the Darwin time series, we find that conditions like those of 1990--95 may be expected as often as every 150--200 yrs at the 95% confidence level. Student's-t, ARMA, and Bootstrap/Monte Carlo tests of the time series all yield similar results. We therefore suggest that the 1990--95 period may plausibly be an aspect of the natural variability of the tropical Pacific.¿ 1997 American Geophysical Union