Colin Mills, compiler of the Hortus Camdenensis, died in late November 2012 after a short illness. As he always considered the Hortus his legacy, it is his family's intention to keep the site running in perpetuity. It will not, however, be updated in the near future.

Letters on the Culture of the Vine Part 10: The Wine Cellar

Letter XVIII – Description of the Cellar

In a former paper I promised to describe a cellar more complete in its design than the one there briefly mentioned. I propose, therefore, to devote this, the last of the series, to a consideration of the subject, cautioning the reader that I have not had so much experience practically in the construction of this description of buildings, as with the majority of the details, upon which, I have endeavoured to communicate information. The great desiderata in the site and arrangement of a cellar for an extensive vineyard, appear to be facility of access with the loads of grapes to that portion of the building in which they are to be crushed; immediate communication between this part and the one in which the fermentation is to be conducted; and perfect separation between this last and the cellars in which the wine of preceding years is to be stored, the whole to be combined under one roof, in such manner, that, without unnecessary sacrifice of space, there may be sufficient room for the different requisite manipulations, free ventilation where it is required, and on the other hand, perfect exclusion from the external air, and an equable temperature in those parts of the building where such conditions are of importance.

With these objects in view, the most convenient site would seem to be the side of a steep bank or hill (such in fact is the description of site adopted at Camden) the declivity being, say from one foot in five to one in ten; and so circumstanced, that a cart road can be formed without difficulty from the vineyard to the upper side, or rather the upper end of the building. The form I prefer for the ground plan, supposing the building to be complete, and to be required for a very large extent of vineyard is nearly that of the Greek cross, (see Plate 3, Fig. vii. [this can be seen in Part 9]) of which, the centre limb A, I propose should consist of two floors, or rather a floor over the cellar, and to be built with its length parallel to the slope, or, in plain English, directly up and down hill. The entrance to be at the upper end, the floor over the cellar, at this end, being nearly flush with the natural level of the ground. As this is the portion of the building which will be the soonest required, it may be first completed, leaving the construction of the others until a later period. I will assume a length of from sixty to seventy feet, with a breadth of thirty to thirty-five feet, to comprise space sufficient in the area of the floor over the cellar, for the different operations of the vintage, and if properly husbanded, room for a very large quantity of wine, say from twenty to thirty thousand gallons, in the cellar itself. In excavating for this latter portion, it will be most convenient to commence with the lower end, because a great part of the earth may thus be removed down hill. Although it is desirable that a height of at least eleven or twelve feet should be allowed from the floor of the cellar, to the under side of the beams of the floor above, it may be practicable, if the declivity be great to obtain this depth, at the upper end, without commencing below the level of the soil at the lower. In which case it is obvious that only one half of the entire cubic space contained in the cellar will require to be excavated, the soil which occupied this portion, being banked up outside the lower half, as the walls are built; so as when finished to render the whole cellar, for all purposes of utility, as completely under ground, as if the entire area had been excavated. Supposing such a building to be built of stone or brick, the walls ought not to be less than eighteen inches thick; and, a solid foundation for them having been found, I would advise that the lower half should be commenced at the same time as the excavation, and be gradually extended towards the upper end, as the latter is proceeded with, care being taken that the mason or bricklayer always has his “work racked back,” and not “toothed.” As I propose that a large door-way should be left in each side-wall, about the centre of its length, to serve as a communication with the other portions of the building, which, when finished, will complete the form of the cross above mentioned, it may perhaps be more convenient to extend the walls, in the first instance, as far up hill as the jambs of these door ways, and not to rack back until the height of their lintels is attained. At the bottom of the lower end wall, near the centre, an opening should be left, closed with a grating, to communicate with a drain, to be formed at the commencement, and extending a sufficient distance down hill, to project beyond the base of the embankment when completed. This drain may prove very serviceable, as well in carrying off moisture, as in providing for the escape of the carbonic acid gas, in case it should occupy the cellar during the fermentation; the great weight of the latter, as we have seen in a former paper, rendering it more easy to expel it by means of openings made at the lowest levels, than by any other methods.

The drain being completed, and a few feet in height of the wall at the lower end finished, the earth of the excavation, as it is brought out, should be placed against the outside of the latter; care should be taken to add it in thin layers, and to ram it thoroughly, for the breadth of four or five feet, with properly formed wooden rammers, shod with iron. This part of the business must be carefully attended to, that there may be no danger of an after settlement in the soil so embanked; which if not carefully guarded against, might force in the wall. By proceeding in this manner, the walls of the lower half of the cellar may be completed by the time that the labourers have finished the excavation of the upper end; and the wall of this half can then be completed also, the door-ways above alluded to being bricked up until required by the further extension of the building. In carrying up this last portion of the walls, as they will be close to a high bank of solid earth, I would recommend that a few inches of space, sufficient to admit a rammer, should be left between the wall and the bank, so that a little loose soil may be thrown in from time to time, and made quite solid with the rammer as the work is completed. Without this precaution, the security of a high wall, built against a bank of earth, might be endangered, during long continued rains. The walls of the under portion being finished to the height of twelve feet: to support the floor, stout beams should be thrown across, at short intervals, which, for greater convenience, as the span is considerable, may be in two lengths, scarfed in the middle, and, to render the floor perfectly solid, should be supported by a row of stout uprights, from the cellar below, down the centre, and parallel to its length. The beams are to receive joists also framed at short intervals, and these last battened over, to receive a floor of bricks or flags; a counter ceiling having been previously formed of stout laths and rough mortar, upon which is to be laid several inches thick of dry saw-dust, so as to fill up the space between it and the flagging. Above the floor, the walls may be again continued to the height of ten feet, but if of brick, with a substance diminished, say, to fourteen inches. In this upper story a sufficient number of windows should be left to admit abundant light and free ventilation; as well as a spacious door-way at the end next the hill to receive the grapes when brought home to be crushed; the whole to be surmounted with a carefully framed roof, which, if the gable-ended form, which I prefer, be adopted, should consist of about six pair of principal rafters, with king and queen posts, purlines, &c. The tie beams, for greater convenience in obtaining the timber, to be scarfed in the middle, and to be strengthened by diagonal horizontal braces, extending from nearly the centre of the first tie beam, to the extremities of the second, and from the ends of the second, to the centre of the third, and thus to the other end. In cutting the rafters, I think it desirable to allow for a considerable projection at the eaves, as by this means, the rays of the mid-day sun are prevented from acting upon the walls.

We shall now proceed to consider how the area of this extensive building may be most advantageously laid out; commencing with the cellar. The great height allowed to this portion is with the view to admit of the use of large vats or tuns. The economy in the expense of construction, in the labour and attention they require when full, in the loss of wine from evaporation, but, chiefly in the space requisite for a given quantity of wine when contained in them, being such, when compared with ordinary casks, that no person possessing a large vineyard, should omit to have them provided, if it be practicable without too great an outlay. At the time the cellar at Camden was in progress, not being able to procure either timber suitable for such large vessels, or a workman conversant with their construction, a more durable, but expensive, material for the vats was substituted: viz,, a compact, close grained sandstone, impervious to moisture. These vats, of which I will not here enter into a particular description, answer their purpose completely, and require no trouble to preserve them in a fit state for use, a few buckets of water with a brush being all that is necessary to render them thoroughly clean after being emptied; and a coat of thin white-wash to preserve them sweet until again required; besides which, they will probably endure without repairs for generations. They are so constructed that wine can be preserved in them for an indefinite period, almost hermetically sealed. Vats of masonry appear to have been frequently adopted on large establishments in France, previous to the Revolution, and are strongly recommended by Chaptal, from whose work the idea of building them here was taken; but I cannot discover that they are now to be met with in Europe. Ours are of two sizes, which contain respectively, 900 and 1,700 gallons; and we use them, as well to ferment in, as to store the wine in afterwards. But although, from the experience acquired in making those above mentioned, I could now undertake to construct vats of stone, equally as good, at far less cost, I apprehend that vessels of wood, will, in most cases, be more convenient to adopt; I shall, therefore, confine myself to some account of their form and construction.

In the first place, I propose that they should be erected upon solid platforms, at least two feet six inches to three feet high from the floor of the cellar, in order that common casks may be filled from them by simply turning a cock. Those of brick, supported upon arches, are the best, after these stone; but timber may also be used, although not so suitable for the purpose. These platforms may be erected all round the cellar, excepting in the centre of the side walls, where a vacant space should be left opposite to each of the large door-ways above mentioned, which are to lead to the other wings of the cellar; besides sufficient room for a stair to communicate with the floor above. They may be built from seven to eight feet broad at the sides of the building, and from five feet six inches, to six feet at the ends. In this manner, room sufficient may be found for sixteen vats, namely, twelve (six on each side) to contain from 1500 to 2000 gallons each, and four from 800 or 900 to 1200 gallons. The vats or tuns of the larger size should be built of the best two-inch oak, unless some colonial timber should be ascertained to be equally suitable for the purpose. The smaller size may be built of timber one inch and a half in thickness. The most convenient form, I apprehend to be that of a stout cask, such as a rum puncheon, for instance, the diameter at the bung and the length being about equal A length of about eight feet inside measurement, with a similar diameter in the middle, will, after allowing for the necessary diminution towards the ends, afford a capacity of about 2000 gallons. Those of six to seven feet, with corresponding diameters, from 900 or 1000 gallons, to 1200. These large tuns are intended to be mounted up on their sides in the position usual with common casks. They are to be sufficiently bound with stout iron hoops, and to have their heads strengthened by pieces of scantling, crossing them at right angles to the joints. Each is to be provided with a man hole, at the bottom of the outermost end, to which is to he fitted a door made either to slide in grooves, and by means of thumb screws, fixed closely to the opening, as in Germany, or on hinges, opening inwards, and secured by a cross bar of wood on the outside, as in France; each method appearing to be effectual in preventing leakage when the vessel is full. This aperture, which need not exceed fourteen inches in height, by ten inches in width, is to be used to admit a man into the interior of the vessel to clean it out when requisite. Another opening, of several inches diameter, is to be made in the usual place for the bung-hole. In the tuns of the largest size above mentioned, supposing the cellar to be twelve feet high, and the platform three feet, this aperture will be within a few inches of the under side of the floor above, through which it is proposed that a small opening should be left, corresponding with the one in the tun. The communication will be rendered more complete by having a stout tube or pipe of wood, with a bore of not less than three inches in diameter, fitted to the aperture in the tun, and of sufficient length to project above the level of the upper floor. The tube will, of course, be made the means of conveying the must into the tun, as well as of carrying up the carbonic acid gas, during the violent fermentation, into the apartment above, where it can be at once freely diluted with the atmospheric air. A stout tap, with a bore of one inch and a half, driven into the outer head, about two inches above the lower part of the chime, will be all that is requisite to complete the vessel. If we suppose it to have been built of well seasoned wood, and secured in its proper position upon the platform, by means of a cradle of stout timber, with room sufficient for a workman to pass between it and the wall, as well as the neighbouring vessels, so as to have access to every part in the event of leakage, it may be considered as being in a fit state for use. In France, for many years past, tuns of this description (termed in French foudres to distinguish them from the open fermenting vats cuves) have been gradually coming into use amongst the larger proprietors of vineyards. They are frequently built with a capacity of four thousand gallons. In a vessel of this size, the annual loss from evaporation is said not to exceed one per cent; in ordinary casks it is rarely less than four, and sometimes as much as eight or ten per cent. Besides, a little calculation will show, that although staves of double the ordinary substance, are to be used in the large tun of two thousand gallons; the number of cubic feet of wood requisite for its construction, will be little more than half the quantity required for twenty casks of one hundred gallons each. In hoops, the saving will be still greater, so will the amount of labour in building it, if the cooper has had experience in this branch of his business. But the greatest saving is in cellar room, and were no other advantages to be derived from the adoption of vessels of large capacity, this circumstance alone should determine us in their favor. It is exceedingly inconvenient in the management of the wine in the cellar to have one tier of casks stowed over another; nothing but absolute necessity should induce this method to be adopted. If ordinary casks, then, are used, it is necessary to have room for them in a single tier, and consequently an immense extent of cellarage when the crop is large. Another advantage will be experienced in the operation of racking. This process is performed with great ease, by means of the bellows and hose, but as the former will require to be worked in the apartment above, the tube above-mentioned to insert into the hung-hole, and to reach to the level of the floor of the tun, will be requisite. Supposing the necessary apparatus to be provided, and it is all very simple, a man would transfer the contents of one large vat into another in little more than three hours. Some difference of opinion has been expressed by practical men with respect to the comparative effect produced by vessels of large capacity, in bringing wine to maturity. I think it probable that a generous wine will ripen more rapidly, at all events after the first year, in small vessels; but that the same description of wine, stored, for a considerable number of years, in large tuns, would ultimately become the best. I should imagine that for the lighter descriptions of wine, the latter must always be preferable. I do not consider it expedient to keep wine stored in ordinary casks in the same cellar that contains the large tuns, although, in case of necessity, there would be room for a row of casks. It is a better plan to devote an adjoining cellar to this purpose.

The foregoing pages, I trust, will have given the reader a general idea of the proposed arrangement of the cellar portion of that part of the building marked A [see plate 3 in Part 9]. I shall now, therefore, proceed to describe the floor above. In the centre of this floor it will be necessary to have a large trap, of six feet square, closed by folding doors; over which, from the timbers of the roof, should be suspended a proper tackle and fall, with some description of crane, or windlass, connected with it, to hoist up and lower down casks between this floor and the one beneath. Near the middle of its length, also, but close to a side wall, will be the most convenient position for the stair of communication. This too, should be closed by folding doors. Exactly over the centre of each of the large tuns, should be left the openings mentioned in a foregoing page. They may be made six inches square, trimmed with hard wood, and lined with Clarence River or New Zealand pine, foreign oak, or other clean working wood, not containing astringent sap, or resin. To each a close fitting lid should be provided, capable of being locked.

The screw press which requires to be framed of very stout timbers, may be so put together, without connexion with the building, as to be taken to pieces and set up again, without much labour, in a different position. However, as the one I have in view, occupies with its ground frame, a horizontal space of more than ten feet square, the end opposite to the entrance appears to be the most convenient position for it; in fact, just over the middle of the first girder from that end, having one of the supports from the cellar below, as before described, directly under the centre of the box or trough. Without a corresponding plan or sketch, it will be useless to trouble the reader with the description of the press we have adopted. It is of perfectly simple construction; possessing immense power, the frame is of corresponding strength; it answers its purpose effectually. The treading box has been before described. It is convenient to have it fixed upon a strong frame or tressel, about four feet or four feet six inches high, in such manner, that the gratings being taken out, to diminish the weight, the whole of the rest can be lifted with the tressel, by five or six men, and removed as occasion may require, to any part of the floor. The great height of the tressel or frame, will prove to be convenient in practice whenever it is considered advisable to separate the must from a portion of its ferment previous to the commencement of the fermentation, as is described at page 107 [Letter XIV, Part 8]. For instance, supposing a vat of the capacity of 1500 or 2000 gallons is to be filled with must, and that the process just alluded to is to be adopted; let casks sufficient to contain the entire quantity be ranged, with their heads out, as near as convenient to the opening in the floor which leads to the tun beneath. Let a broad tub, to contain forty or fifty gallons, be placed under the spout of the treading box, and be raised upon some description of temporary support, sufficiently high from the floor to place its bottom nearly upon the same level as the tops of the casks; and let a tap of size large enough to suit the wooden tube attached to the hose (see Page 112 [Letter XV, Part 8]) be fixed in it a few inches from the bottom. As fast as the tub receives the must, (which should be freed from the seeds and hulls by draining through a basket, containing a little clean straw) from the treading-box, it can be made to flow at pleasure, by means of the tap and hose, into each of the casks in succession, until all be full; and when the proper moment for separating it from the ferment has arrived, each cask full can be drawn off into the tun below, by another hose, until all are emptied. The attention of one man is sufficient to carry on this process, even when two thousand gallons or more of must are expressed in one day. When it is not judged expedient to separate the must from any portion of its ferment, the use of the casks will, of course, be dispensed with, and the hose be made to communicate direct from the tub under the treading box to the tun. A large bottle rack, to contain empty bottles, is the only other addition I have to suggest to the arrangements upon this floor, supposing the further extension of the cellar be contemplated. But as the one beneath can at all times be kept quite distinct by being locked, the upper floor can be made to serve as a convenient cooperage. The remaining wings of the cellar may be added or not, according to the views of the proprietor; but if the outlay for this purpose be considered inexpedient, the upper floor of the part I have just been describing, may, in such case, be made to serve as a fermenting cellar. With this view it will be requisite to provide a sufficient number of fermenting vats or tuns, to be ranged round the walls. Whether the open vats or the tuns be preferred, it will be desirable that neither should exceed six feet elevation from the floor: because it would probably become exceedingly inconvenient to have the treading-box raised to a greater height than this, and were the bottom of the latter not to be raised as high as the top of the fermenting vessels, it would no longer be practicable to fill them by means of the hose, and much additional labour would be entailed. In determining the number of fermenting vessels requisite, it should be remembered that each may be used to ferment three or four successive batches of wine in the same season, and that as fast as each batch passes from the tumultuous stage of fermentation, it can be drawn off into the tuns beneath.

I will now proceed to describe the general form and disposition of the two wings, commencing with the one marked B [see plate 3 in Part 9]. This part is intended for the storing, of wine after it has passed its first or second year, either in ordinary casks, or, as we have just seen, more economically in large tuns. Supposing the area allotted to it, to be 72 feet by 27, sufficient horizontal space would be afforded for sixteen tuns of 2000 gallons each, in two rows; or about seventy-four or seventy-six casks, of 100 gallons each, in four rows; in either case, with ample room for every necessary operation connected with the management of the wine. If the large vessels are to be employed, it will, however, be necessary to have the walls of the cellar twelve feet high; but, for ordinary casks, seven feet six inches or eight feet will be sufficient. The wing marked C [see plate 3 in Part 9] is supposed to be intended for the bottling department; that is to contain two rows of casks, (from thirty to thirty-two) in the middle, in process of preparation for bottling, the bins in two tiers, along the side walls. Supposing this wing to be of similar length to the other, there would be room to stow from 3,500 to 4000 dozen (7000 to 8000 gallons) a larger quantity of bottled wine than we can hope to find in one grower’s cellar, at least in our generation. Wine bins are most conveniently constructed of brick, laid in good roman cement. From thirty-eight to forty-two inches is sufficient space to contain one dozen bottles laid transversely, and thirty-six to thirty-eight inches, affords breadth sufficient for four rows endwise, the bottles being placed neck to neck: thus four dozen will compose one tier. In height, they will occupy less room, because the bottles in each succeeding tier, are stowed opposite to the intervals in the layer below. It will not be prudent to exceed twelve tiers in height: these will not require more than thirty-six inches. Therefore, bins three feet six inches long, three feet two inches broad, and three feet high, may be assumed to be a convenient size; each will contain forty-eight dozen - the probable produce of clear wine from a cask of one hundred gallons. A height of seven feet six inches will be amply sufficient for two tiers of bins, the bottom of the lower one being raised from the level of the floor, by the thickness of one brick. In constructing them, the whole of the partitions should first be built four and a half inches, or half brick, thick, each brick being well soaked in water, and carefully bedded in cement. As soon as these are carried up to the required height, (three feet to three feet three inches) very flat arches of bricks, placed edge to edge must be thrown across between the partitions, and filled up at the spandrils with fragments of brick, bedded in cement, and smoothed over with the trowel, to obtain level ?oors for the upper range of bins. A rise of two or three inches will be sufficient to render these arches, if laid in good cement, perfectly capable of supporting the weight of the wine to be stowed over them. [A bin of this description is shown in Part 8]. Each of the wings, B. and C., are designed to have a separate entrance door at the opposite extremity to their point of junction with the centre. These separate entrances should have each a vaulted porch, with an outer door, so as, in a great degree, to prevent the access of the external air to the cellars. I propose that the wall-plates of the two wings should be laid at the same level, and consequently that the door of the wing B. should be sunk four feet deeper than the floor of C., in the event of the former being occupied by large tuns. It is intended that they should have rough ceilings, covered a foot thick with saw dust, or light dry earth, to intervene between them and their roofs. And that the earth excavated from them, should be banked up outside, up to the level of the ceiling. A vaulted roof of brick, plastered with cement, and covered to the depth of two or three feet, with earth, would form the most perfect covering for a cellar, but this plan would involve great expense.

The suggestions which I have thrown out in the foregoing pages, will probably prove serviceable to those who meditate the cultivation of vineyards on a large scale: one or two additional observations may perhaps be acceptable. Large open vats to ferment in, even when provided with covers, are objected to by many French cultivators, because a considerable expense is incurred in constructing vessels which are only serviceable during two or three weeks of each year; whilst, by substituting tuns built in the form of common casks, as above described, it is only necessary to be provided with two or three over the number requisite to contain the entire crop after the wine is made, as they serve to store the wine in afterwards. The only substantial objection to the latter description of vessel, appears to be, when it is required for the manufacture of red wine; because it is very difficult to fit the sort of false lid, described at page 110 [Letter XIV Part 8], which is used to keep the head of the skins beneath the surface of the fermenting wine. A good supply of pure water is an almost indispensable requisite in the vicinity of a cellar. Should no spring or well be easily accessible, it will be advisable to construct a tank, of brick and cement, to receive the water, which, by means of gutters and pipes, may be collected from the roof. I can testify by experience, that by such means an ample supply of excellent water may be provided even in the driest seasons.




Statham & Forster, Printers, Lower George-street, Sydney.