The Essex Naturalist 105 bufferland are descended from black deer, and how many from common? Even if all the black fallow were descended from black parents, there still must be a large influx of common deer from elsewhere in Essex to account for the high common to black ratio. This belief is backed up by Hollins and Carroll (1997) who stated that deer numbers were increasing in agricultural areas in the south- east, by the Corporation of London ecologist and Gordon Wyatt of English Nature. From this fieldwork 30% of the fallow deer on the bufferlands were black. Taking the 1996 count of 513 deer as representative then at 30% the number of black deer on the bufferlands was 150. There are estimated to be between 50 and 70 deer remaining in the forest itself (Langbein, 1996). Therefore if we assume that all the black deer derive from the forest population and if every deer remaining in the forest is a black one, there are still twice as many black deer on the bufferlands as there are in the forest. Looked at in this way, the evidence supports the historical hypothesis statement that the black forest deer have been driven out of the forest and into the bufferlands. In summary, the work on the numbers of black deer found in the bufferlands and the increasing number of visitors to the forest supports the historical hypothesis while the location of deer around the Wake Arms is ambiguous because of deer access to Sable Shaw. Conversely, the work on the numbers and location of deer in the Lower Forest, and the nitrogen analysis do not support the historical hypothesis. Evidence for the behavioural hypothesis In every case where fallow deer were found, there was also found evidence of regular use of access points, particularly those at Rookery Wood and the Lower Forest. The slots usually pointed both to and away from the fields. This strongly suggests that the deer were passing from the woodland to the fields and back again, as the behavioural hypothesis states. On all occasions where deer slots had been found at the access points and the observer remained undetected, the expected crossing movements were observed. Fallow deer slowly emerged from the forest and began to feed in the fields. The one exception to this case was the Lower Forest, and an explanation for the lack of crossing observed here is given when the results of the traffic survey for that area are studied. On the evening of the traffic survey (and indeed on all the evenings when the forest was observed), it appeared there was not a time when the deer could cross at dusk when they were being observed, because of the sheer volume of traffic passing along the road. Yet the evidence of the slots at the access points show that they do cross which by a process of elimination must be late at night and the return in the early hours of the morning. This is a far cry from Chapman and Chapman's (1975) description of undisturbed fallow feeding in the afternoons. With this restriction on movement it is hardly surprising that the deer abundance is higher on the agricultural and buffer lands with the mixed habitat of fields for feeding