of settlement by combining the rudimentary classification schemes of several recorders.

A standard approach throughout the nineteenth century was to divide the island's natural vegetation into three broad physiognomic types, which seem to correspond in part to forest—types. These were: (1) the areas under a normal forest cover (sub-divided by some recorders into la) hardwood forest, and (b) areas dominated by coniferous trees — called ’softwood forest’ by some“), and the areas that were termed (2) the ’barrens’ and (3) the ’swamps’.“2 The latter two habitats evidently also carried, at least in part, a tree cover, which appears to have consisted mostly of conifer species. Further sub-divisions of these forest—types are found in connection with the systems, which will be discussed later“, that many recorders developed for the purpose of using the natural forest as an indicator of the quality of the soils that lay beneath them, and hence their suitability for clearance for agriculture, and these have been incorporated in what follows.

’Hardwood forest’ — Many of the early recorders refer to a forest-type that we would now recognize as ’upland hardwood forest’“. This consisted of the shade-tolerant hardwood species ('beech', 'maple' and 'black birch' [i.e. yellow birch] are usually listed) occurring on well-drained soils, which in general were viewed as suitable for conversion to agricultural land. A number of recorders use the term ’hardwood’ as a descriptive label for this forest-type and at the same time

“ Johnstone 1822, Land Commission 1875 (see footnote 50

for the names ofthe particular witnesses).

‘2 Stewart 1806, Johnstone 1822, MacGregor 1828, Hill 1839, Bain 1890. (Stewart and Hill do not actually use the word ‘barrens‘ in this particular context but the land-type is implied in their descriptions.)

43

See Tables 3 and 4, which are discussed later (pp. 25-29). ‘4 ‘Upland hardwood forest' was a descriptive term used by Erskine (1960) in his classification of the forests of the island, and l have subsequently used it in my classification of the current forests ofthe island (e.g. Sobey 1995, Sobey & Glen 2002, 2004). It is the forest composed of the shade-tolerant hardwood species on well- drained soils, and can occur not just in the hilly or 'upland' landscape of the island but anywhere where the soils are well- drained. The term ‘upland' was also used in exactly this sense by Captain John [MacDonald] (1804) (eg. pp. 14, 18, 26, 36, not extracted) where he used it both as a noun (“the uplands”, and as an adjective (in the phrases “upland farm" and “upland hay") in order to distinguish the part of an island farm on well-drained soil from the part that was based on the use of the salt-marshes.

11

provide a description of its species composition“; others use the label without providing a description“, while still others do not use the term ’hardwood forest’, but describe the tree species composition of a community that is clearly this forest-type“. From these many descriptions we can deduce that the hardwood forest could show considerable variation, but in order to qualify as such, the tree composition had to consist principally of beech, maple (probably largely sugar maple, though there may have been some red maple) and yellow birch. The descriptions indicate that these three species could occur together in any proportion ranging from a full mixture of the three to single species stands. However, the quantitative analysis contained in Appendix 1, as well as the many descriptive comments contained there for each species, indicate that of the three trees, beech was the leading species in the island’s hardwood forests. The make-up of the species mixture seems to have depended on the relative drainage properties of the soils, with yellow birch being the most demanding of the three species in terms of soil moisture, and beech the least.

Two other hardwood species, red oak and white ash, could also contribute to the hardwood forest though both were comparatively uncommon; and there was also frequently a minority conifer element in the hardwood forest, comprising any of three species: ’pine’ (undoubtedly white pine), ’spruce’ (probably red spruce) and ’hemlock’. These conifers seemingly could occur as scattered individuals or in small clusters. Where these conifers were the dominants over extensive areas, which appears to have been so in the ’hollows’ within the hardwood forest, and in low-lying areas along streams“, it is questionable whether the forest of that specific area could still be called hardwood forest.

‘5 Curtis 1775; [Robinson] 0. 1798; Stewart 1806; Johnstone 1822; Stewart 1831; Bain 1868-1884 (in 1873, at Springfield); and some witnesses to the Land Commission (1875).

‘6 Gesner 1846; Palmer 1816; Craswell 8. Anderson c. 1856; Dawson 1868; Bain 1891; Macoun 1894; and many witnesses to the Land Commission (1875).

‘7 Gray 1793; Selkirk 1803; MacGregor 1828; Hill 1839; Lawson 1851; some witnesses to the Land Commission (1875); Bain 1890 — who uses ‘deciduous forest' as an equivalent term.

‘3 Bain (1868-1884) contains a description of such conifers in the ‘hollows' at Springfield in 1873, and he later described a similar distribution of conifers in "ancient forest" near North Wiltshire ([Bain] 1882).