copings and their finish. This method is shown in the accompanying

cut:

a

FIG. 3.

b

The simplest method of stopping or buttressing a flight of steps is by a continuous plinth, with the top line running parallel with the slope, as in Fig. 4. This is, however, liable to two objections. In the first place,

it does not, unless each buttress is made of a single stone, have a proper constructive look; for it gives the idea that it is set on a slope, and this suggests a slippery and unstable foundation. Again: the slope and the buttress, by this method, are on the same angle of inclination; and this repetition is to the last degree tame and monotonous to the sight, there being no contrasting forms whatever, the slight breaks of the steps not being sufficient to counteract the effect of their general inclination. The Greek method, and the one followed by all the great builders, is to buttress the steps with blocks laid in horizontal courses, and broken down to conform to the steps, as in the following figure. The buttress at the top of

FIG. 5.

the steps should be set up above the ground, so as to mark the point where the steps commence to a person approaching them from the upper walk.

Some modification of this method (and the variations can be quite numerous) is far preferable to the sloping buttress for a simple flight of steps without any decorative features, as it gives an effective contrast to the line of the slope, and has, besides, a substantial and constructive appearance. Figs. 6 and 7 show two designs, in one of which the contrast desirable between the slope and the buttresses is obtained by horizontal buttresses

FIG. 6.

FIG. 7.

with a slope between, while the other is a modification of the simple horizontal method. Either is in good taste, and may be used with or without decoration.

Should it be desired to ornament such a flight of steps, the vases or

FIG. 8.

sculpture should be placed at the top of the flight, unless in extremely exceptional cases. They will thus be seen from the walk above as well as

from below, and will serve to give distance to the objects behind them; whereas, if they are placed at the bottom, they will either come against the bank in a diagonal view, or will cut against the upper line of the slope, half relieved against the bank, and half against the objects beyond the upper walk, and thus lose their full effect from either point of view.

The diagram, Fig. 8, gives the simplest way of decorating such a flight of steps, shown in profile. The upper step should be widened sufficiently to correspond to the plinth under the vase; and this latter should be set low, so as to harmonize, and not contrast violently (as in Fig. 9) with the

FIG. 9.

steps and bank. The effect of this latter from below is shown in the following cut:

If the house and all its adjuncts are rich and elegant, there is no objection to decorating even a simple flight of steps with sculpture, if the flight is of sufficient magnitude so that the objects on either side will

not look crowded together. The difficulty of doing so satisfactorily is entirely one of judgment of position and harmonious combination. Statues in such a situation should not be placed on high pedestals; and great care should be taken to give them a proper point of view and a simple and harmonious background. It should be borne in mind that they will almost invariably be best seen from the lower level. A very long flight of steps, to correspond to very rich surroundings, may be decorated both at the top and the bottom; but in this case the upper portion should be the principal one, in accordance with the reasons already given.

It has already been remarked, that there are two kinds of terraces: the architectural decoration of the first, or earth-slope, we have just been considering; the other is entirely an architectural creation. This has no slope whatever. The edge of the terrace is bounded and sustained by a wall; and, this making a perpendicular descent, a guard of some kind becomes indispensable to prevent persons from inadvertently stepping off. This necessity, so simple, and easily provided for, has been made the foundation of some of the most superb works of mere decorative architecture, exercising all the invention and skill of the architect in their design and execution; for, if it is necessary to have a parapet, why not an elegant one? If steps and approaches are required, why not make them on a grand scale? With the earth-slope, no parapet or balustrade is required. There is no danger of walking off it; and, as there is no more danger of this on the steps than at the upper edge of the slope, no balustrade is required at the steps; and it would be quite out of harmony to place one here, and omit it at the upper edge of the slope. There are other reasons why the earth-slope should not have a parapet or balustrade. built, it must rest upon a wall. This wall must be solid, and be set deep enough in the ground to avoid the effect of frost. But why, if you have a wall, have also a slope? Again: the sods are always dropping away from a wall, where they are laid up to it on a slope; and the line of junction becomes ragged, uneven, and disagreeable.

If

Hammatt Billings.

FIELD-MICE.

No group of our indigenous quadrupeds possesses greater interest, in an economical point of view, than that great subdivision of the rodentia, the muridæ, in which are comprehended the sub-families dipodinæ, or jumping-mice; murinæ, rats and mice proper; and arvicolinæ, or fieldmice and a short article here on their general habits will not be out of place, particularly as this is the season in which they are especially destructive in the orchard and nursery, the field and garden.

Of the dipodinæ, the genus Faculus is the only representative in this country. Its most prominent characteristics are a long, slender body, sharp nose, very long tail, and greatly lengthened posterior limbs, adapted to leaping.

THE JUMPING-MOUSE (Faculus hudsonius) is the most familiar of these little animals, and it seems to be generally distributed throughout the country. It is known in different sections by the names "kangaroo-mouse,"