geographical restriction to the western part of the island, an area with which they were less familiar. In fact, the earliest explicit reference to the use of cedar in fencing is in evidence given to the Land Commission of 1875, where one witness called it ”a valuable timber for fencing posts”47‘, while another informed the Commission that ”it is the most valuable kind of wood in that part of the country”, adding that "a cedar post is worth on the spot cut about 8 cents”472. Also, two other witnesses indicated that the fencing of land with

cedar increased the land's value“.

The amount of wood required for fencing — In comparison with other types of fencing, the ubiquitous snake fence required an enormous amount of wood.474 With the aid of basic geometry and arithmetic it is possible to estimate the number of rails (or poles) required for the snake fences on a typical island farm: if the rails were 14 feet in length, then five sections of a snake fence would have taken up a 49 foot stretch of ground.475 If we assume that eight rails were used in each section“, then to fully enclose the boundaries of the typical island farm of 100 acres with its 220 chains of bounds (Le. a 10 chain frontage and 100 chain depth)477 would have taken an amazing 11,856 rails and 2,964 bracing poles.478 However, if, as was often the case, the

‘71 Land Commission 1875: evidence of George W. DeBlois for

Lot 9.

472 Land Commission 1875: evidence of Donald McPhee, Lot 9.

473 Land Commission 1875: Peter Doyle of Lot 7 proudly stated that his farm was “nearly all fenced with cedar”; while Joseph Mooreshed of Lot 10, referring to a particular area of land under discussion, said that “if the land was mussel-mudded and fenced with cedar it would be worth more“.

‘7‘ Williams 1989, pp. 69-71. "5 Johnstone (1822) (p. 99, not extracted) gives the rail lengths as ‘about 14 feet'; Lewellin (1832) as ‘14 to 15 feet’; and near the end of the twentieth century, Knox (1980) (p. 22) gives values of ‘13 or 14 feet‘ from one source, and ‘about 12 feet' from another. I have therefore taken the length as 14 feet, and have also assumed an overlap between the rails of one foot at each end, and have set the rails at an obtuse angle of 110 degrees with each other.

‘75 The number comes from Lewellin (1832) - see footnote 456.

477

See footnote 305. ‘78 If the angle between the rails was 90 degrees (as it may have been in some fences), rather than 110 degrees, then these numbers would rise to 13,672 rails and 3,418 bracing poles. The long narrow shape of the ‘typical‘ island farm of 100 acres results in a very high exterior boundary length of 220 chains. If, however, a farm of the same acreage had a 20 chain frontage and a 50 chain depth, as some did, it would reduce the exterior bounds to 140 chains.

72

back half of the farm remained unfenced, then the wood requirements for the external boundaries would have been reduced by one half; and if there were neighbouring farms on each side to share the fencing load, then the number required would drop further, to 3,240 rails and 810 bracing poles. However, there are still the internal field boundaries of the farm to be considered, and for these there would have been no help from neighbours. Though in future it may be possible to obtain estimates of the length of the internal boundaries of a large sample of farms by extrapolating backwards from field boundaries visible in the 1935 aerial photographic survey of the island, for the moment, an estimate based on the field boundaries of a single farm, shown in the 1880 ’Meacham’s Atlas’, will have to suffice. The farm is the 60 acre farm of Abraham Gill of Little York in Lot 34, which I calculate to have had about 140 chains of internal field boundaries.479 To fence these would have required 7,544 rails and 1,886 bracing poles480, which, when added to the fences on the farm boundaries, gives a total of 10,784 rails and 2,696 bracing poles. If we use the 12 to 15 years that Johnstone (1822) gives for the longevity of spruce and fir rails, then we get an average turnover, or replacement rate, for the rails of between 719 and 899 per year, and for the bracing poles, of between 180 and 225. However, if both the rails and the bracing poles were of cedar, and if cedar lasted twice as long (say 30 years), then these numbers would reduce

‘79 ‘Meacham's Atlas' (p. 54) shows an oblique ‘aerial view' of a

good part of Abraham Gill’s farm with all of its subdivison fences, many of which are shown as snake fences. What is shown is almost all cleared farming land, and so might be taken as equivalent to the cleared portion of a typical 100 acre island farm. Visible are five cross boundaries each of 10 chains length (i.e. 50 chains), while the fencing seen running back through the farm property is equivalent to about one and a half property lengths (Le. 90 chains), giving a total of about 140 chains of internal boundaries requiring fencing. I note that the ‘Plan of Lot 34' (p. 83 of Meacham‘s) shows two farms marked ‘Abraham Gill 60 [acres]’, directly beside each other, each with a frontage of 10 chains on the Little York Road, and a depth of 60 chains. Examination of the aerial view of the farms in the 1935 aerial photographic survey (Plate 5063-85) suggests that it is the more southerly farm that is shown in the atlas. (If the house shown in the ‘view' is taken to have a frontage of 42 feet, then the width of the farm shown in the view (including a projection for the part out off on the left) works out at 640 feet or 9.7 chains, which is near enough to the 10 chains that it should be, to indicate to me that the ‘view’ shows only one of the two farms.)

“w This is, as before, assuming eight rails per section of fence, and 110 degree angles between the rails. If the angles were reduced to 90 degrees, then 8,696 rails and 2,174 bracing poles would be required.