corresponding to pollarding and regrowth (39). Those I have seen in Epping Forest indicate a pollarding cycle of about 13 years; in Hainault there is a variable cycle of 18-25 years. Records show that most of the produce was faggots, which would have been produced on the shorter rotation; larger sizes of firewood — billets, shides, long and short talwood — are mentioned as the product of boilings chopped up by landlords or timber trees illicitly felled. There is no evidence that pollarding was ever arranged on a regular cycle, which would have been difficult to organize and unnecessary. Even in Monk Wood, over which the landowner had complete control (see later), the cycle was variable. The 1608 survey records 36,294 "firewood trees" and 1810 "decaying oakes'' on a third of the Crown lands, themselves about a third of the Forests. If we assume that both these categories refer to pollards the total number of pollards in the Forests was about 340,000 or 31 to the acre. Since about a third of the Forests consisted of plains the density of pollards in the wooded parts must have been about 45 to the acre. There was some turnover of pollards. The Crown and other landowners felled boilings, usually to pay for the upkeep of buildings and roads. Conversely they repeatedly complained of commoners "topping young Trees to make Pollards of them". Small fines were sometimes imposed for the first pollarding of oaks, which would otherwise have become timber trees, but this had little effect; throughout the Forests 5% - 10% of the present pollards are oaks. The system worked well with neither party gaining the upper hand. The wood-rights at this time belonged to the occupiers of particular houses; sometimes the chimney of a demolished house might be left standing to preserve the right (46). By 1793 there were 34 such rights in Barking and 39 in Dagenham; each was supposed to amount to five loads each of a hundred faggots (27). There were 32 rights in Waltham (including Upshire) and 33 in Sewardstone (24). These four manors covered about a third of the Forests, and if similar rights existed everywhere the total would have been about 500 rights. If we assume that a sledge-load was a ton we infer that the commoners cut 1/4 million faggots, or 2500 tons of wood, a year. The Warden had a further 10,000 faggots (100 tons) annually. This is a maximum estimate, for we have assumed that all the rights were exercised and have taken the highest reasonable figure for a sledge-load, yet it amounts to only a third of a ton per acre per year from the wooded parts of the Forests. Even allowing for timber, the growth of wood on the Forests can hardly have been less than three times this figure. We conclude that woodcutting by landowners, although not appearing systematically in the records, must have accounted for at least two-thirds of the annual production. The clearest example of woodcutting by landowners is the Monk Woods, which were expressly excluded from the commoners' lopping rights (14), yet are full of pollard trees just like the rest of the Forest; the present trees were cer- tainly lopped well into the nineteenth century. In 1848 the owners of Higham Bushes in Chingford and of other parts of the Forest were said to derive '' a profitable return'' from lopping. 45