The distribution of the Oxlip Primula elatior (L.) Hill in Essex
eaten. The scatter diagram for these results is shown in Figure 4. Correlation coefficients were
calculated for all of the woods in which the deer survey was carried out, and also for those woods
containing > 1000 Oxlip plants. The results from the latter had a smaller error, suggesting that as a
predictive tool this system will work better on sites where there are more plants more evenly
distributed throughout the wood. A higher range in the results when all sites arc used may reflect a
higher variation in grazing levels in large woods with small or fragmented Oxlip populations.
When deer find a small population of plants, the percentage of inflorescences eaten may be very
high; conversely, if it is not found , the grazing level will appear very low. Given a long enough
period of time the results will become more even since the deer will have sufficient time to find all
of the Oxlip colonies.
Combined deer scores were produced, encompassing all species of deer. Whilst it is relatively easy
to differentiate between the activity and damage resulting from Fallow and Muntjac Muntiacus
reevesi Ogilby, it is far more difficult to differentiate when Roe Capreolus capreolus L. or Red
deer Cervus elaphus L. are added to the equation.
Fig. 4. Relationship between the percentage of Oxlips eaten against the deer activity score
(Based on figures from 16 woods all containing an Oxlip population >10()() plants. Note that some
sites have identical scores, hence some points on the graph equate to two or three sites)
What is less well understood in the deer/Oxlip relationship is the indirect effect on the plant resulting
from deer browsing coppice re-growth. It is well known that deer cause enormous damage to the
young tree shoots (Rackham 1975; Cooke 1993; Cooke & Farrell 1995; Tabor 1993), and that in
severe conditions trees arc killed (Rackham 1975; Cooke 1998). If this happens when a neglected
and understocked coppice is first cut after many years, the death of many of the stools prevents the
quick canopy closure that is an essential part of the coppice cycle. The shading resulting from early
canopy closure after coppicing controlled the growth of the light demanding species that would
otherwise take over. Typical of these species are pendulous sedge, tufted hair grass Deschampsia
caespitosa (L.) P. Beauv, and bramble (Rubus spp). A number of woods were visited during the
survey in which coppice stools, particularly ash Fraxinus excelsior L., had been killed by deer with
the result that a mixture of grasses, sedge and bramble had achieved dominance in the newly cut
coppice plots at the expense of Oxlips, amongst others. Figures for Hempstead Wood, for example,
showed the numbers of Oxlip plants per hectare in recently cut coppice very badly eaten by deer
was 481 (SE+/-124), significantly less (P< 001 using T test) than uncut coppice with 4375 (SE
Essex Naturalist (New Series) 19 (2002) 131