birch) "clothed the highlands of the interior”, while "a heavy growth of spruce and hemlock, fir and pine” was found ”along the shore and extending nearly two miles inland" (i.e. in the Sea View / Park Corner area). As for the other two land-types, Stewart (1806) said that there were ”not many swamps of any extent [on] the island" and ”no extensive pine barrens", which is echoed by MacGregor's (1828) statement that the ’swamps’ and the ’barrens’ made up ”but a small proportion [of] the whole surface of the island". Similar to these is Samuel Hill's (1839) comment that "the districts where the soil is very sandy and hardly at present worth cultivating”, and which were covered with ”scrubby spruces and other firs of stinted growth” were ”not extensive” in area. However we must bear in mind that both Stewart and MacGregor — though not Samuel Hill — would have been far less familiar with the part of the island west of the Malpeque-Bedeque isthmus, if indeed they had ever visited the area. There are few indications in the early records of the geographical distribution of the three forest- types, that is so evident from modern studies.‘°'5 We have already noted the association of the barrens with the St. Peters area and with west Prince County. In addition, Hill (1839) noted that there were ”swampy lands of considerable extent in some districts", though he did not record where these districts were. It is really only at the end of the nineteenth century with the writings of Francis Bain (1890), that useful if generalized written information emerges on the perceived broad geographical distribution of the various forest-types, though the picture has to be carefully teased out by reading the ‘Soils and Botany’ section of his school textbook in parallel with the ’Historical Geology' section. In doing so, we observe that he considered the prevailing forest-type of the island, ”the deciduous forests of beech, birch and maple”, to specifically occur in western Queens County including the large area of what we now know as the Central Hill Lands, and in Kings County in the South-eastern Hill Lands and in the north-east peninsula, as well as in other unspecified parts of 105 See Sobey & Glen (1999) (Figs. 10 and 11) or Sobey 8: Glen (2004) (Figs. 8 and 9) for the predominance of hardwood forest in the central part of the island, and of the wetter woodland types in the east and the west. 17 the county.106 These are all areas where the remnants of such forest occur today‘°7. The only area that he failed to include for this forest-type is Prince County, where remnants still occur in parts of the west, and in much of the county east of the Malpeque-Bedeque isthmus.108 Bain is less comprehensive on the geographical distribution of the two other forest-types that he mentions, ”the swamps and the barrens”. He says that the ”very light sandy soils”, that were ”originally clothed with shrub pines and birches and the prostrate arbutus” occurred ”in parts of King’s and Prince Counties" — he must be referring here to the 'barrens’ even though he does not use the word at this point; and he also says that the "heavy clay soils” that supported an ”often swampy” vegetation that was originally composed of "black spruce, larch, ash and willow”, were found in the low areas around parts of Hillsborough Bay (he specifies ”the low lands about Pownal, Orwell and Nine Mile Creek”), and in ”the extended plains on the west of the island”. He thus considered that both of these wet and poor forest-types were largely confined to the eastern and western counties, which is in broad agreement with the distribution of their remnants today‘°9. The relative abundance of the different trees — All of the early records for the distribution and we The detail on which this summary is based is as follows: Bain considered the deciduous forest to be associated with what he called the "Boulder Clay formation". which he said overlay "the rolling districts of the Trias” (i.e. “the extensive range of hills stretching from Bonshaw to New London [i.e. what is termed by MacDougall et al. (1988) as the ‘Central Hill Lands‘ of the island], eastward to Wiltshire and Rustico", plus “some areas in Kings County"). The same deciduous forest also occurred on what Bain called the “upper Permian", which he said comprised “the rolling district around Charlottetown Harbour and its river estuaries, and most of the Island to the eastward“), and it also occurred on the sandstone tracts ofthe Lower Permian (i.e. comprising “the bulk of Tea Hill“, the area “south of the West River, the hills of Belfast and Upper Montague [these are termed the South-eastern Hill Lands by MacDougall et al. (1988)], and the hilly tract from Souris to East Point”). ‘°7 See Sobey & Glen (2004) (Figure 3), which is the same figure as in Sobey & Glen (1999) (Figure 5). '08 It is not that Bain was unaware of the forest-types of the western part of the island, as is evident from the short article describing a railway journey from Charlottetown to Alberton, that he published in the Daily Examiner in 1883, where he contrasted the swampy uncultivated land between Miscouche and O'Leary, with the “magnificent deciduous forests“ between O‘Leary and Alberton ([Bain] 1883). Rather, his textbook presents really only a brief generalised overview for the island‘s school pupils. ‘°9 See Sobey & Glen (1999) (Figures 6 and 7) or Sobey 8. Glen (2004) (Figures 4 and 5).