ON LAYING OUT A FLOWER GARDEN. 35 J 



to cross at right angles, or to have parallel lines, as it 

 divided or laid down by a mathematical scale, which is 

 too formal for the diversification of nature. All 

 walks through these pleasure departments should be 

 winding and enlivening, not continuing any length in 

 one direction.* The continuous view of a straight 

 walk is dull and monotonous. The divisions should be 

 highest about the centre, that whatever is planted 

 therein may have effect ; and to make a Flower-garden 

 fully interesting, and render it a source of natural infor- 

 mation, where free scope might at all times be afforded 

 to employ the leisure hours in mental improvement, 

 there should be a good system of arrangement adopted. 

 The Linnean system is the most easily acquired. A 

 small compartment laid out in beds might contain 

 plants of all the twenty-four classes, and a few of all the 

 hardy orders, which do not exceed one hundred. Or to 

 have their natural characters more assimilated, the 

 Jussieuean system could be carried into effect by laying 

 down a grass plat, to any extent above one quarter of 

 an acre, and cut therein small figures to contain the 

 natural families, which of hardy plants we do not sup- 

 pose would exceed one hundred and fifty. The diffi- 

 culties of this arrangement are, that many of the cha- 

 racters are imperfectly known even to the most 

 scientific. Mr. John Lindley has given additional light 

 on the subject by his last publication. All the large 

 divisions should be intersected by small allies, or paths, 



* Since writing the above we have seen the Flower-garden of J. B. Smith, Esq. 

 and consider it a beautiful specimen, finely illustrating the taste of that gentle- 



