by half, and if it lasted three times as long, they would reduce by two-thirds.481

The other type of fence, the post-and—rail fence, following, as it did, a straight line, and having fewer rails per section than the snake fence, required far less wood: if there were five rails per section, 5,080 would have been required for the whole of our farm instead of the 10,784 of the snake fence.482 However, though the number of posts required was less than the number of bracing poles in the snake fence (2,032, compared with 2,696)4a3, the post-and-rail fence had the major disadvantages of not only requiring much greater labour in its construction on account of the digging of the post-holes, but also the posts when in the ground decayed more quickly, and thus had a shorter life-span.484

Of course, the first rough fences of logs or stumps may have served their purpose along the farm boundaries for many years. However, when the farm began to pass beyond the forest clearance stage, and the farmer began to construct his fences out of wood specifically collected for the purpose, he then became committed to the regular — most likely annual — collection of fencing materials in order to maintain his fences. For most farms, especially from the 18505 when much of the land on the island suitable for farming was beginning to fill up, this supply of fencing materials would have had to come from the farm’s own woodlot. And in fact fencing materials are given equal status with firewood as a reason for

retaining an area of woodland on each farm‘“,

‘31 l have no numerical estimates for the length of time the rails of

a cedar snake fence would last; Knox (1980) (p. 21) says vaguely that they could last for ‘several generations’.

‘82 See footnote 460 for a discussion of the number of rails in a post-and-rail fence.

‘83 l have arbitrarily assumed, in the absence of any contemporary evidence, that five rails were used per fence section, and l have used the same criteria as for the snake fence - i.e. that only the front half of the farm was fenced, and that the fencing of the side boundaries was shared with the neighbouring farms. With these assumptions the number of rails and posts required for our 100 acre farm is 1,525 rails and 610 posts for the external boundaries, and 3,555 rails and 1,422 posts for the internal, making 5,080 rails and 2,032 posts in total.

‘8“ The ‘straight rail fence’ was also recalled in the 19705 as requiring very sturdy posts in its construction, being “liable to blow over in the first strong gale" (Knox 1980, p. 73). Knox also reports that those who built straight rail fences were sometimes regarded as poorer farmers, having fewer rails for the job.

‘85 Land Commission 1860: William McGowan, referring to 'fence rails' as a 'necessary', stated that “a man must procure firewood and fence rails off his farm".

73

with, in several statements the annual fencing requirement being mentioned in tandem with the annual need for firewoodm.

Was the fencing requirement able to be met? — If we extrapolate the figures for our ’typical’ island farm to the 11,241 ‘occupiers of land’ counted in the 1861 census of the island, the annual requirement for fencing materials over the whole island becomes enormous: if all of the fences were snake fences of spruce or fir, then between 8.1 and 10.1 million new rails and between 2.0 and 2.5 million bracing poles would have been required each year‘”; if all of them were post-and-rail fences then the requirements would have been less: between 3.8 and 4.8 million rails and between 1.5 and 1.9 million posts. If all of the ‘longers’ had been cut as ’poles’ (rather than from split rails), then these numbers would also have been the number of trees that would have had to be cut down each year for the purpose. Of course, the use of cedar for fences in the west of the island would have reduced somewhat these numbers.

Rather than use such astronomical numbers, which are difficult to graSp, it is easier to return conceptually to the level of the individual farm. At that level, using the longer 15 year replacement rate, what my calculations reveal is that in order to sustain the fencing requirements of a 'typical’ farm, the farm woodlot would have had to produce each year enough spruce, fir or poplar trees to provide 719 rails and 180 shorter bracing poles. (If thicker stems were available for splitting into rails, then these numbers could reduce by a half — or more, where more than two rails could be split out of a tree stem.) And of course if our ’typical farm’ was in an area where cedar for fencing was available then these numbers would fall by a half or more.

The question is whether there would have been enough young poplar, spruce or fir trees on a farm woodlot of say 50 to 70 acres, to support this annual demand? Most of the farms had been created in areas that had supported upland hardwood forest, and in such areas all three of

‘86 Anon. 1836; Land Commission 1860: evidence of W,

McGowan, representing tenants of Lots 44 and 45; B. LeLacheur of Lot 64; and G. W. DeBlois, agent for the Cunard estate.

“7 The actual calculated numbers come to 8,082,275 rails at 15 years longevity and 10,105,659 at 12 years; and 2,023,380 bracing poles at 15 years longevity and 2,529,228 at 12 years.