ON WOOD AND PLANTATIONS. 89 



the most striking part or parts of the prospect (near the 

 house for example) to answer this end. This should be 

 done, not by planting a continuous, uniformly thick belt of 

 trees round the outside of the whole ; but by so arranging 

 the various outer groups and thickets, that when seenfro?n 

 the given points they shall appear connected in one whole. 

 In this way, there will be an agreeable variation in the 

 margin, made by the various bays, recesses, and detached 

 projections, which could not be so well effected if the 

 whole were one uniformly unbroken strip of wood. 



But where the house is so elevated as to command a 

 more extensive view than is comprised in the demesne 

 itself, another course should be adopted. The grounds 

 planted must be made to connect themselves with the 

 surrounding scenery, so as not to produce any violent 

 contrast to the eye, when compared with the adjoining 

 country. If then, as is most frequently the case, the lawn 

 or pleasure-ground join, on either side or sides, cultivated 

 farm lands, the proper connexion may be kept up by 

 advancing a few groups or even scattered trees into the 

 neighboring fields. In the middle states there are but few 

 cultivated fields, even in ordinary farms, where there is 

 not to be seen, here and there, a handsome cluster of 

 saplings or a few full grown trees ; or if not these, at 

 least some tall growing bushes along the fences, all of 

 which, by a little exercise of this leading principle of 

 connexion, can, by the planter of taste, be made to appear 

 with few or trifling additions, to divaricate from, and 

 ramble out of the park itself. Where the park joins 

 natural woods, connexion is still easier, and where it 

 bounds upon one of our noble rivers, lakes, or other large 

 sneeis of water, of course connexion is not expected ; foi 



