‘poorly 
INVESTIGATION OF THE WHITESHEET DOWN ENVIRONS 1989-90 175 
In most features the upper layers contained the 
greatest abundance of charred wood and the widest 
range of genera, while neither oak nor Prunus 
occurred at the base of the sampled features. The 
paucity of these two genera must, however, be 
balanced with the relatively small overall quantities 
of charcoal recovered from the lower feature fills. 
Land clearance prior to construction of the 
causewayed enclosure may have extended into the 
immediate vicinity, allowing for early secondary 
regrowth. The numerous deposits of hazel nut 
shells indicate that shrubby, fruiting forms of this 
species were plentiful, suggesting an open aspect 
perhaps not far from the site. At least one member 
of the Pomoideae family was relatively abundant, 
possibly growing as scrubby hawthorn recolonising 
some of the cleared areas and/or as whitebeam. 
Blackthorn may also have been competing in this 
situation. Clearance may have involved the felling 
of oak/ash woodland. Regeneration from the 
stumps of immature trees would have occurred 
unless they had been removed or killed. Such 
regrowth would have produced numerous shoots 
that if regularly cut would have resulted in coppice 
stools. Hazel also coppices easily and would have 
fruited well in this situation, especially if managed 
on a cycle of more than six or seven years. 
Stem material of ash and hazel was present in 
several of the pits, lending support to the potential 
for coppice woodlands. Oak, however, was very 
represented in the charred samples 
examined, suggesting that it was either not 
available or deliberately not selected. While oak, 
and indeed ash, may not have survived on the 
downlands around the enclosure, it would be likely 
to have grown on the clay soils at the base of the 
hill. 
Pollen analysis 
by Robert G. Scaife 
A sequence of 31 samples was taken from the fill of 
the recut enclosure ditch (1331) to elucidate the 
character of the vegetation and land-use during the 
second phase of the monument. Pollen preservation 
is always marginal in alkaline soils of rendzina type, 
especially on chalk. Initial assessment showed that 
pollen was preserved, although not as well as that of 
other sites studied elsewhere in southern England 
(Scaife 1984; 1995; Scaife in Cleal and Allen 1994, 
79). Samples were thus selected for analysis. 
Seven samples were selected for pollen analysis 
with standard techniques used for extracting pollen 
and spores (Moore and Webb 1978). Because of the 
low absolute pollen frequencies encountered, 
relatively large sample sizes of up to 10ml were 
used. In addition to ‘normal’ chemical procedures, 
micromesh (104m) sieving techniques were also 
used for removal of inorganic fractions. Pollen 
identification and counting was carried out under 
normal and phase contrast illumination at 
magnifications of x400 and x1000 using an 
Olympus biological research microscope. 
Pollen preservation in calcareous soils 
It is a common misconception that pollen is not 
preserved in the highly aikaline and oxidising 
environments encountered on archaeological sites 
on chalk. Pollen preservation is usually poor but it 
has been proven that valuable information 
regarding palaeoecology of chalklands can be 
obtained (Scaife 1984; 1995; Dimbleby and Evans 
1974; Dimbleby 1978). Such analyses have to take 
into account the poor preservation of the pollen and 
the possibility of skewed data resulting from the 
differential preservation of pollen taxa. 
At Whitesheet Hill the fills of the recut are 
relatively chalk-free silty loams and are 
undoubtedly derived from the typically worm- 
sorted upper humic soil horizons (Ah) on the edges 
of the ditch. These fills most probably derive from 
surrounding soils very shortly after the re-cut and 
may provide evidence of the vegetation growing at 
this time. As in Dimbleby’s analysis of postholes at 
Ranscombe Camp, Sussex (Dimbleby 1986 and 
pers. comm.) such were analysed in the hope of 
providing data on the local environs. 
The pollen data 
Absolute pollen frequencies were measured in 
samples from 78, 96 and 100cm. Raw pollen data 
and absolute pollen frequencies (where ascertained) 
are presented in Table 8. A total of seven samples 
produced pollen and spores. In the case of samples 
from 54cm and 64cm pollen was extremely sparse. 
Pollen sums from the remaining samples range 
from 100-200. 
The data (Table 8) clearly shows that 
Liguliflorae (Jaraxacum type) is the dominant 
taxon in all samples analysed. This is highly 
characteristic of pollen in downland soils where 
differential preservation favours this robust pollen 
type. Similarly, the relatively high values of certain 
fern spore types must also be noted. The seven 
pollen levels analysed are all characterised by 
relatively few pollen grains of arboreal taxa 
