3.1. Limits on longterm emission
The dataset for each source has been tested for the presence of gamma ray signals. We find flux limits, based on the maximum likelihood ratio test (Gibson et al. 1982 , Li & Ma 1983), of above 400 GeV for PKS 2005-489 and above 300 GeV for PKS 0548-322. The threshold energy, defined as the peak of the differential triggering spectrum, for the observations has been estimated on the basis of preliminary simulations, and is in the range 300 to 400 GeV for these objects, depending on the object's elevation. The effective collecting areas which have been assumed, again from simulations, are at an energy threshold of 300 GeV and at an energy threshold of 400 GeV. These are subject to systematic errors estimated to be . We have assumed that our current selection procedures retain of the -ray signal, which is subject to a systematic error of %. The preliminary simulations have assumed a source spectrum similar to the cosmic ray spectrum. Deviations from this assumption change the limits found by amounts that are small compared with the systematic errors. A detailed analysis of the spectral response of the Mark 6 telescope awaits the completion of more detailed simulations, which are in progress.
3.2. Search for short bursts
We have subdivided our data sets on two timescales, a dark period of about 2 weeks and on a night-by-night basis, to look for evidence for emission. Following the procedure we have adopted for our analysis of data from PKS 2155-304 (Chadwick et al. 1999), we have normalised our data by dividing the gamma ray excess by the number of off-source hadron initiated showers which survive the cuts.
In Fig. 1 we show the results of such an analysis of our PKS 2005-489 data. We show the normalised excesses binned by dark period (Fig. 1a) and by day (Fig. 1b). We also show a histogram of the day-by-day significances for gamma ray emission (Fig. 1c). These data show no evidence for variable emission on time scales of a dark period or a day. However, it is interesting to note that the distribution of excesses shows 2 nights with excesses in the 2.75 - 3.25 band where none is expected.
In Fig. 2 we show the results of a similar analysis of data from PKS0548-322. Again we conclude that there is no evidence for variable emission on time scales of weeks or days.
3.3. The November 1998 PKS 2005-489 X-ray flare
A multiwavelength campaign on PKS 2005-489 occurred during 1998 September and October. The results of this are due to be reported elswhere (Perlman et al. 2000). During this period there was a large X-ray flare from PKS 2005-489 which lasted until 1998 December (Remillard 1998 , Perlman et al. 1999). At its peak, on 1998 November 10, the 2-10 keV X-ray flux was about 30 times brighter than in its quiescent state (Perlman et al. 1999). VHE observations were made with the Mark 6 telescope at all times when weather conditions and the moon phase permitted during 1998 October. During 1998 November and December, PKS 2005-489 was at too large a zenith angle for meaningful observations to be made.
VHE observations were available for 1998 October 15, 17, 21 and 22. The daily VHE limits, together with the X-ray rates (from Perlman et al. 1999) are shown in Table 1. There is no indication of VHE emission at the sensitivities achievable with the Mark 6 telescope during this flare.
Table 1. Limits () to VHE emission from PKS 2005-489 during the 1998 flare. The X-ray measuements are from the RXTE observations of Perlman et al. (1999).
© European Southern Observatory (ESO) 2000
Online publication: January 29, 2001