Book II. DIRECTION OF ROADS. 513 



3301. The most perfect line, according to Marshal, is that which is straight and level. 

 But this is to be drawn in a country only which is perfectly flat, and where no obstruc- 

 tions lie in the way ; joint circumstances that rarely happen. Where the face of the 

 country, between two points or places to be connected by a road, is nearly but not quite 

 level, by reason of gentle swells which rise between them, a straight line maybe perfect, 

 may be the most eligible under these circumstances. But where the intervening 

 country is broken into hill and dale, or if one ridge of hill only intervenes, a straight line 

 of carriage road is seldom compatible with perfection. In this case, which is nearly ge- 

 neral, the best skill of the surveyor lies in tracing the midway between the straight and 

 the level line. The line of perfection, for agricultural purposes, is to be calculated, by 

 the time and exertion, jointly considered, which are required to convey a given burden, 

 with a given power of draught from station to station. On great public roads, where expe- 

 dition is a principal object, time alone may be taken as a good criterion. 



3302. A regular method of finding out the true line of road, between two stations, where 

 a blank is given, where there is no other obstruction than what the surface of the ground 

 to be got over presents, is to ascertain, and mark at proper distances, the straight line j 

 which is the only certain guide to the surveyor. If the straight line be found to be ineli- 

 gible, each mark becomes a rallying point, in searching, on either side of it, for abetter. 

 If two lines of equal facility, and nearly of equal distance from the straight line, present 

 themselves, accurate measurements are to determine tlie choice. If one of the two best 

 lines which the intervening country aifords is found to be easier, the other shorter, the 

 ascent and the distance are to be jointly considered j the exertion and tlie time required 

 are to be duly weighed. 



3303. The nature of tlie ground, the source of materials, and the comparative expense of 

 forming the road, by two doubtful lines, as well as their comparative exposure, are also 

 to be taken into consideration. Although, in some places, Paterson observes, it may be 

 of little consequence, either to the traveller, or to the public in general, which way the 

 bendings are turned, provided the level is nearly obtained, yet a great deal may depend 

 upon those turns or bendings for the real benefit and advantage of the road. In bend- 

 ing it one way, you may have no metals that will stand any fatigue, unless at a great dis- 

 tance and expense ; while, in turning it the other way, you may have metals of the very 

 best quality, in the immediate vicinity. In the one way, too, you may be led over ground 

 of a wet bottom, where even, with twelve or fourteen inches deep of metals, there would be 

 difficulty in keeping a good road ; while, in the other, you may have such a dry bottom, 

 that the road would be much easier upheld with seven or eight inches of metals. So that 

 the tract that may appear most eligible to the eye, at first sight, may not always be the 

 one that should be adopted. " A combination of all the requisites I have already men- 

 tioned, should be studied, as far as possible ; and where these cannot be found all to unite, 

 the one possessing the most of these advantages, and subject to no other material objec- 

 tion, should, of course, be adopted." {Treatise on Roads, p. 19.) 



3304. Roads, Edgeworth observes, should be laid out as nearly as may be, in a straiglit 

 line ; but, to follow with this view the mathematical axiom, that a straight line is the 

 shortest that can be drawn between two points, will not succeed in making the most com- 

 modious roads ; hills must be avoided, towns must be resorted to, and the sudden bends 

 of rivers must be shunned. All these circumstances must be attended to, therefore a per- 

 fectly straight road cannot often be found of any great length. It may perhaps appear 

 surprising, that there is but little difference in the length between a road that has a gentle 

 bend, and one that is in a perfectly straight line. A road ten miles long, and perfectly 

 straight, can scarcely be found any where, but if such a road could be found, and if it 

 were curved, so as to prevent the eye from seeing further than a quarter of a mile of it, in 

 any one place, the whole road would not be lengthened mor^ than one hundred and fifty 

 yards. It is not proposed to make serpentine roads merely for the entertainment of tra- 

 vellers ; but it is intended to point out, that a strict adherence to a straight line is of 

 much less consequence than is usually supposed ; and that it will be frequently advan- 

 tageous to deviate from the direct line, to avoid inequalities of ground. It is obvious, 

 that, where the arc described by a road going over a hill, is greater than that which is 

 described by going round it, the circuit is preferable ; but it is not known to every over- 

 seer, that within certain limits it will be less laborious to go round the hill, though the 

 circuit should be much greater than that which would be made in crossing the hill. 

 Where a hill has an ascent of no more than one foot in thirty, the thirtieth part of the 

 whole weightof the carriage, of the load, and of the horses, must be lifted up, whilst they 

 advance thirty feet. In doing this, one thirtieth part of the whole load continually resists 

 the horses' draught ; and in drawing a waggon of six tons weight, a resistance equal to' 

 the usual force of two horses must be exerted. 



3305. A perfectly level road is not always the best for every species of draught. Slight 

 and short alternations of rising and falling ground are serviceable to horses moving 



