THE FLOWER GARDEN. 



IOC 



their colom-s are so charming-, and their habit of 

 growth so dense and spiral, that they deserve to be 

 noted as amongst the finest of shrubs for bedding- 

 out pm-poses. It will be noted that No. 1 is a 

 Box edging, and that there is a boundary hedge 

 of Cicpvessus Zaicsoiiii, - 'No. 17. Both of these 

 might be dispensed with ; and if the bed is on 

 turf, No. 1 must be so, while No. 17 is only suit- 

 able for positions where it is desirable to conceal 

 a fence, road, or — as in this case — the vegetable 

 garden, it being an actual arrangement on each 

 side of the central walk in the kitchen garden, 

 and repeated, almost in toto, of shorter length, as a 

 boundary border at the end of a terrace garden, the 

 tine spread of tm-f in fi'ont there being infinitely 

 preferable to the gravel walk. 



Though we have thought it best to set down all the 

 plants actually used, were it intended to repeat the 

 design we should exclude the Lobelia in No. 13, 

 using only Yiola, solely on account of the failure of 

 the former before the end of the season. 



The marking out of this plan preparatory to 

 planting is almost self-explanatory, the most par- 

 ticular part being the central line running through 

 Nos. 6 and 9, by which the panels, circles, and ob- 

 longs are formed. Measure oif the correct distance 

 from the centre of one circle to that of another, and 

 the size of intervening ground-work ; and with two 

 strings reaching to these points and running round 

 the pegs, the circle is formed by the shorter string, 

 and the oblongs by the longer. It then only re- 

 mains to rub out the straight lines enclosing the 

 panels on the space for ground-woi-k plant. No. 2, 

 and the design is completed. 



Design Fig. 2. — This is an excellent design for 

 placing on each side of a long gxassy glade, and 

 equally so on but one side of a walk, if the opposite 

 one be furnished with shrubs ; or it may form one of 

 the piincipal groups of beds on the x>arterre, the 

 planting and size being varied according to taste and 

 the position it is to occupy. It is designed with a 

 view of showing how to combine two very opposite 

 sections of plants — flat-growing foliage, and taller- 

 growing flowering kinds. A further combination 

 being, that whilst (like the preceding plan) it may be 

 of any length, it may just as appropriately be either 

 of one, two, or three panels (circles) in extent. 

 " Sentinel" or dot plants, except in the very centres, 

 have been purposely excluded from this design, for 

 the reason that the heights of the plants themselves 

 which are employed in it are sufiiciently varied in 

 that respect, and also because the narrow ovals, Nos. 

 o and 6, are not good shapes for them, as they do 

 not look well if they overhang or exceed the bounds 

 of the panels in which they are placed. 



In Fig. 2, the edging, No. 1, is raised four inches 

 above the tui'f , the mode of doing which we shall deal 

 with pi-esently ; meanwhile, it must sufiice to say that 

 om' opinion is, that all edgings should be thus raised. 

 The enclosing band. No. 2, of Veronica repens 

 (green), whilst not objectionable near thetui-f, owing 

 to the upright edging of grey and pm-ple being be- 

 tween, renders the design adaptable to being sur- 

 rounded by gravel. The ground-work, No. 4, en- 

 closing the circles, being of a greenish-white in 

 colom', and running thi'ough colom-s of blue, white, 

 and dark pm-ple, we get a combination of colom-ing, 

 showing both hannoni/ and contrast in the same bed, 

 which must be seen for its beauty to be fully realised. 

 Then the plants in Nos. 3, 5, and 6, being of higher 

 growth than Nos. 2 and 4, if kept even in outline, 

 and tapering to the centres of their respective plots, 

 the imdulations caused by Nos. 2 and 4 are of the 

 most pleasing description. It will be observed that 

 the plants in Nos. 1, 2, and 4 are quite hardy ; and 

 also the middle plants, Nos. 8 and 3, may be arranged 

 with the hardy British plant, Ajuga rcptans purpurea ; 

 then, when frost cuts oif the flowers, a very mode- 

 rate amount of ingenuity will be able to either 

 wholly or partially fill out the beds for the winter. 



The marking out of this design is just about as 

 simple as the previous one. Run a line through 

 the middle of the border ; then draw circles to the 

 enclosing horizontal lines : divide these circles into 

 fom- parts at the outer edges ; from each of these 

 four points draw other circles, and thus is formed 

 the half-circle on each side of the design. 



Design Fig. 3. — This is of a less elaborate 

 nature, but arranged on the same principle as the 

 foregoing, and is a very suitable arrangement for 

 placing in an isolated position where it is desii-ed to 

 have a bed of a formal description. We have seen 

 and admired this pattern on the plateau of a \illa, 

 and have also made use of it in a large garden. 

 In this latter connection, its most telling position 

 is between beds of similar size which are filled 

 ^\'ith plants of one or two coloui's only, and which 

 need some assistance in the way of refinem.ent and 



dressiness,'''' other than the extreme of that which 

 regular " carpet-bed " patterns produce. As regards 

 size, that may be of any extent proportionate to the 

 place it is to occupy ; but if very large we should 

 di\T.de it into six or even eight parts, instead of four, 

 and thus have six or eight beds ; and the middle 

 circle would also, of com'se, have to be enlarged. 

 The eye would be the best judge of proportion for 

 any particular spot it is to occiii^y. 



Design Fig. 4 is a still simpler design, but none 

 the worse for that ; for, as a rule, the least complicated 



