This paper discusses the archaeological value of small pollen sir:es which
reflect patchy local vegetation conditions typical of prehistoric floodplai
ns. It is argued that both floristics and vegetation structure can only rel
iably be gained from macrofossil analysis and/or local pollen assemblages w
here the problem of the mixing of pollen from different communities is mini
mised and site to site differences are maximised. This exploits the pollen
recruitment characteristics of small sites and differences between the poll
en transport curves between canopy and understorey/groundcover species, man
y of which are known to be reliable indicators of vegetation structure and
management. The data used to illustrate this approach are taken from six ar
chaeological and environmental sites in the Soar and Nene valleys. The data
available at present suggests small-scale temporary clearances of the Neol
ithic floodplains with regeneration. It is not until the Bronze Age that la
rger scale deforestation occurs and it appears to be associated with ritual
use of the floodplain. By the middle-late Iron Age the floodplains are alm
ost total ly deforested and used for grazing. A common feature of the sites
is an expansion of Pteridium (bracken) as clearance occurs and its use as
bedding in the Iron Age is indicated by macrofossil finds at one site. Give
n the small scale of the Neolithic clearances (however caused), including p
ermanent changes to the herb curves it is suggested that this may reflect a
forest farming (grazing) system. The coincidence of the landscape-scale de
forestation of the floodplain and the construction of funery monuments rais
es the question as to whether the deforestation was primarily a ritualizati
on of the landscape, agriculturally driven,or both.