It is consequently necessary to have a duplicate collection in frames to make sure of preserving a stock of these disappointing subjects. Of course there is the alternative of not growing these at all, and it is advisable for gardeners not to do so as a rule; but it is only fair to bear in mind that the collections trotted out from time to time in 'The Garden' have an attached stock in frames, and sometimes houses of a more expensive character.

Turning now to what your correspondent has to say as to his experience with tender bedders, and the figures he has so kindly placed at the disposal of your readers, a criticism of which he challenges, permit me in the first place to point out the absurdly large number of plants he employed to fill the border he discusses under the two systems. Allow me to quote the whole of what he says on the numbers of plants raised: "We had to vary the style of the border every year; but on the average, the labour and cost were about the same. Sometimes the border was ribboned and sometimes panelled, etc. Our plan, then as now, was to plant autumn-struck Geraniums ' the length of the trowel' apart each way, or about 9 inches - not more; and such things as Lobelias, Calceolarias, Verbenas, etc, were planted about 6 inches apart; while succulents were set nearly touching each other. One way and another, front and back, at least 2400 Geraniums were required, and more when the groundwork was composed of Mangles's variegated Geranium, or other kinds, besides other 3 or 4 rows perhaps.

In addition to these, at least 1200 Calceolarias were used, as Aurantia multiflora and Aurea floribunda, and as much Verbena venosa between these and the Geraniums. This left still near the half of the border to fill, which was perhaps filled up with Lobelia panelled with Geraniums, or vice versa, etc. The Lobelias, we know, came to about 8000, and in front of these are sometimes planted a row of Echeveria secunda glauca, Crimson King Verbena, or any other suitable edging at the time. The total number of plants generally used was about 13,000." The above statement, though somewhat mixed up and wanting in plainness of expression, obviously means that the border which we were told in a previous sentence was 360 feet long by 10 feet wide, took 13,000 plants to fill it, or nearly 4 to every square foot; that, occasionally, after more than half of the border was filled, 8000 Lobelias were panelled with Geraniums in a portion of that front space, and that a suitable edging was used in addition ! That all ordinary bedding-plants were set out at 6 inches apart each way, with the exception of Geraniums, which were planted 9 inches apart each way. Now, let us analyse these figures.

A border 360 feet in length will require 720 plants at 6 inches apart in the row; 18 such rows will require 12,960 plants. Allowing a space of 4 3/4 inches next the box to the first line, and a space of 4 3/4 inches between the back edge and the backmost row of plants, there remains, of the 10-feet border, 17 spaces of 6 1/2 inches each between the 18 rows. Next let us take the instance of the 8000 Lobelias (which were panelled with Geraniums in "near the half of the border "), and the suitable edging. " Near the half of the border " would be a little less than 5 feet wide, perhaps 6 inches less, but 5 feet will do. Planted at 6 inches in the row, or 720 plants, these 8000 Lobelias would fill 11 rows of 6 inches wide, and cover a total space of 360 feet by 5 feet. There remains this query for more ingenious arithmeticians than I can pretend to be, to find where the Geraniums which were panelled with these 8000 Lobelias were put, as also the precise position in the border the "suitable edging" was placed. Then there remains the question of distances between the plants - 9 inches for Geraniums, and 6 inches for other kinds of plants.

Here again I confess to be somewhat at variance with the above mode of arithmetic. "Where is the utility of planting Geraniums which are 9 to 15 inches across, at a distance of 9 inches from each other, and allow them six weeks in which to meet; or of planting Calceolarias of an average size of 6 inches across, at 6 inches each way, and allow them ten weeks to do it in ? Of course the plants for that particular border might have been grown on a principle of great refinement, and were not allowed to grow beyond a certain size before being planted out; but I do not think that gardeners at all act on that principle in a general way.

Before leaving the 13,000 bedders, we will compare the number of hardy flowers required to stock the same border. For hardy plants 35 were spent, or "close upon" that sum. "Most of the plants came from Mr Parker's, of Tooting, at the prices marked in the catalogue." Sixpence per plant used to be the lowest price quoted in Mr Parker's list, and we may take the average at ninepence each. That gives, allowing a percentage of "gratis" plants, 950. "The most" was made of these, and with the addition of what was "propagated" from "our own small stock," and the further "addition of a few annuals," the border which swallowed up such an unconscionable lot of tender bedders was complete, though why annuals should be classed as herbaceous plants it is difficult to say. It will also be noted, as the Editor did on the appended notes to "J. S., W.'s" article, that all mention of hardy bedders is omitted. This is hardly fair, when such plants as Violas, variegated Grasses, Polemonium variegatum, the dark-leaved Ajuga, Golden - feather Pyrethrum, Sedum spectabile, and others, are to be found in almost every garden.

As to cost of production, "J. S., W.," will not lose anything at 5s. per 100. Geraniums are the only expensive article to produce, and a great deal less than 8 per 1000 will produce these. Calceolarias are more than covered at 15s. per 1000; Lobelias, Ageratums, and other tender plants, at 20s. per 1000. The cost of hardy bedders, in most cases, depends on how many pieces a man can break up in a day. Now I do not want to make out a case either way, as I think just as much of a border of mixed hardy herbaceous plants as I do of a bed of Geraniums; but from a little calculation I have made, I find that "J. S., W.'s" border could be planted with 6500 plants, and that their cost would not exceed 10 for the lot, or 3s. per 100. Take the whole quantity of plants required for a flower-garden, and much less would suffice for even an extravagant "bedding man".

Now let us take a nurseryman's prices, and see what the cost of the two systems would be were a stock of each to be bought. I have a quotation before me for 5000 plants, 1500 of them to be Geraniums - of which 500 are gold and silver variegated, - the remainder "Calceolarias, Lobelias, Verbenas, Ageratum, etc." These, furnished next May, fine plants out of single pots, are 4, 10s. per 1000; out of store pots, 2, 5s. per 1000. Remember, these are all good plants. Well, then, here is a quotation from a nurseryman, who makes hardy plants a specialty; his rate is 3, 10s. per 100 in quantity, Ms own selection, and does not include the best sorts. A selection of finer sorts is 7, 10s. per 100, and of course does not include such things as Paeonias, the rarer hardy Lilies, etc. If we follow the directions given by the persons who have taken hardy flowers under their own special care, a hardy border must be planted so thickly that the taller plants will support each other, and all bulbs will have a carpet of dwarf plants through which they are to spring. It would take nothing short of "J. S., W.'s" 13,000, taking the above mode of planting, and allowing from three to twelve bulbs for a clump, to fill the ground the first year as completely as the same number of bedding-plants would do.

These gentlemen will perhaps be able to say something on the cost of furnishing a border under the two systems, allowing 13,000 plants in each case, the one at a nurseryman's price of 4s. 6d. per 100, and the other at 66s. 8d. per 100. Before closing this question of cost, I would just say that I know the cost of bedding-plants in one of the largest establishments in the three kingdoms, and where some 200,000 plants are used, is reckoned at one farthing per plant, or 2s. 1d. per 100. I have taken up too much of your space already, but allow me just to say further, that I consider the cost of production, as raised by "J. S., W.," has very little to do with the matter from a gardener's point of view. To us the question is whether we should rid ourselves entirely of ordinary bedders, and only find room for hardy flowers, or whether we should adopt the common-sense plan of finding a place for each, no matter whether the cost of production or the cost of keep should in either case be more than we like. Strawberries in March or new Grapes in April, French Beans and Cucumbers in February, are not worth the cost of production. In the same sense it does not pay to mow grass once in ten days, or hoe walks once a fortnight, or sweep up autumn leaves every morning.

The whole of "J. S., W.'s" argument is a question of the same kind.

R. P. Brotherston.

[We do not think it advisable to devote more space to this discussion, for, after all, cost is no argument against a system of flower-gardening, or any other phase or branch of horticulture owners of gardens may choose to spend their money on, as being to their taste, and most suitable for their places - just as one may derive most pleasure from, and spend money on Orchids, and his neighbour's fancy may lead him to spend on Alpines or Auriculas, etc. It is the fact that so much has been said of the expensiveness of the one, and the wonderful cheapness of the other, that led to our remarks in November. It entirely depends on how either system is gone about. If hardy herbaceous flower-gardening is to be done as well as tender bedding has generally been done, and a constant succession of bloom to be kept up over a given area, in conjunction with the absence of untidiness for five or six months of the year, then we maintain it cannot be any cheaper than tender bedding, while its first cost, if the same area be planted, is many times more costly.

To stick a few hundreds of hardy herbaceous plants widely apart into a border, and make up the spaces between them with tender hardy and half-hardy annuals, biennials, and other plants, and call it hardy herbaceous gardening, is not correct, and very different from a border kept gay by means of a mixture of hardy herbaceous plants alone, and should not be called a herbaceous border. To call it the mixed border of hardy herbaceous and tender plants would be its proper denomination. Some writers would appear to wish for nothing so much as the total extinction of the bedding system, and have it replaced with an ideal system not yet to be found. Now what we advocate is a curtailment of the one, and a slight extension of the other - the extent in either direction depending on the time the owner resides at his country-seat; and we think nine-tenths of our readers will agree with our idea of the matter. Let us have both systems, in proportions depending on circumstances, and away with the furious tirade against flowers in masses. - Ed].