420 



PICTURESQUE IMPROVEMENT. 



BOOK I. 



the bearing posts placed nearer each other, and then the weight 

 of cattle on either end of the cross rails would weigh them 

 down and be frightened away, while the fence would fall down 

 in its former position. It is scarcely necessary to add, that 

 both, for this and the former purpose, the fence would have to 

 be divided into lengths — and that for this last purpose (at least 

 in cases where the excavation was deep) it would be proper to 

 have two fixed rails, or one narrow deal, between each division 

 for ladies to walk over*. 



Another species of invisible fence, also invented by me, and 

 which has most of the advantages of the preceding one, may be 

 formed by making the excavation somewhat deeper, and sowing 

 furze or planting thorns in it, and mowing them when they grow 

 higher than the surface of the surrounding pasture. A section 

 of such a fence is shewn Plate XVIII., fig. 5. In order to admit 



* The only objections that can be made to this fence are those which were 

 made by writers unexperienced in rural affairs when I first announced it *. The sum 

 of their arguments is, that during night cattle may stumble on it and break their legs, 

 and during day they may walk across it. Independently of the success of the trial 

 at Kingswood Lodge, I have to offer, against the first argument, that in fields con- 

 taining uninclosed pits or quarries, cattle are seldom or never known to fall into 

 them ; some striking instances of which there are in Derbyshire, and many parts of 

 Scotland: and in respect to walking across this fence, it is- well known that cattle 

 will not, unless forcibly driven, and scarcely then, leap over a common rail or gate 

 laid flat upon the surface of a grass field: rails and gates are much narrower than 

 this fence, and have no excavation under them. 



* In Observations on Planting, Landscape Gardening, &c. published 1804. 



