Return/retour

Dormers

by Linda Hoad

In Historians,
Preliminary Architectural Studies,
Volume 02, Unpublished Report HG 02
(Fortress of Louisbourg, 1972,
Report Number H G 02 02 05)

This study of dormers is based on the Domestic Architecture index, secondary sources,, and a study of the Louisbourg map collection and the related maps, plans and views. Several Toisés are also included, although this material is not yet included in the Domestic Architecture index. The emphasis in this report is on structural details of dormers. The number and placing of dormers has not been considered, since they are related to the floor plan and room use of each individual building.

Unfortunately, there is little specific information, documentary or otherwise, concerning dormers. The available information will be considered under the following headings: location, shape, size, framing, covering, flashing, and general data. The final section of the report will present the general conclusions and recommendations.

Only two copies of this report have been provided with the reader printed enlargements of the plans and views used in making the analysis. These are the draughting copy and the copy filed in the Archives. The Xeroxed material is attached to each copy.

(1)

LOCATION

It would appear from the plans and views that there is no right or wrong location for dormers. Many are flush with the exterior face of the wall (718-1, 724-3, 725-4a, 733-9, N.D.-34); a considerable number are flush with the interior face of the wall (731-3a, 733-11, 743-1, N.D.-72); some are set well back on the slope of the roof (725-8, 724- 12 726-1); others appear to be in the middle of the wall (743-3, Plate III); and some are a continuation of the wall, necessitating a break in the roof (Plate IV).

There is only one documentary reference to the location of a dormer. In an estimation of repairs and additions to a house and storehouse in 1752, it is specified that the dormers were to be built on the wall plate of the storehouse [NOTE 1].

(II)

SHAPE

The majority of the dormers illustrated in the plans and views are gabled. The only exception is 725-4a., which is hipped. According to the secondary sources, a hipped dormer is fairly common, and is called a lucarne à la capucine (See Plate VII, fig. 1723; Plate VIII, fig. 107). However, this type does not appear to have been extensively used in Louisbourg.

The majority of the dormers have a rectangular opening, with the exception of those shown on plans 718-1 and 725-8.

Doyon and Hubrecht state that the general rule for dormers is that the slope of the dormer roof should correspond to the slope of the roof on which it is situated. However, the hip on a dormer may be steeper in order to accentuate the feature [NOTE 2].

A rather curious feature of the Louisbourg plans and views is the lack of an overhang at the front of the dormers. (See plans 725-8, 743- 1, N.D.-34, Plate VI.) An overhang does appear on plans 733-9, 733-11., 743-3 and N.D.-7a. It seems impractical to construct a dormer without an overhang, and in fact, Doyon and Hubrecht insist on the need of a good overhang [NOTE 3].

(III)

SIZE

There is considerable documentary evidence for the size of dormers, and the analysis of the Toisé material should provide even more when it has been completed. Thus, no attempt has been made to scale the plans and views.

(A) SHUTTER DIMENSIONS

(1) Chateau St. Louis: Large - Height: 2P 9p x Width: 1P 7p; Small - Height: 1P 9p x Width: 1P 5P [NOTE 5]

(2) Hospital: Height: 2P 9p x Width: 1P 7p

(B) SASH DIMENSIONS

(1) Engineer's House: Large - Height: 3P 2p x Width 4P 6p; Small - Height 1P 5p x Width: 1P 7P [NOTE 5]

(2) Magasin de Vivres: Height: 1P 6p x Width: 2P [NOTE 6].

(Also note that the heights and widths seem to be
reversed in some cases. The figures were given in this way in the Toisé,
but it seems more usual for the height to exceed the width.)

No exact sizes are known for domestic structures, but the number of glass panes is occasionally mentioned, and the sash size can be calculated from this. The number of panes mentioned are 12, 6, and 4, and the size of the glass, given in two instances, is 7 by 8 pouces [NOTE 7].

Plan 753-1 seems to indicate that a double sash is possible in a dormer, although there is no other corroborating evidence.

(IV)

FRAMING

The size of the framing is given in the Toisés as 4 by 4 or 5 by 5; that is, for the ridge, plates, uprights and braces. The sash frame for one dormer is given as 6 by 6, as opposed to the other members which are 4 by 4 [NOTE 8].

The secondary sources indicate that the common rafters serve as valley rafters for the dormers, although Chabat suggests that the rafters supporting the dormer should be larger than the others, and that trimmers should be added, if necessary, to support the purlins. (See Plate VII., figs. 1722, 1723.) The sill is not mentioned in the documents, and it does not appear on most of the detailed plans. However, it is mentioned in Blondel, 9 and appears in Plates VII, VIII, and IX.

The best views indicating framing are 733-9. 733-11, 743-3, 752-8, and Plates II and III. A king post is quite common, and mortise and tenon joints seem to be the rule.

Marcus Whiffen states that dormers in Colonial Williamsburg were framed "to the upper surface of two rafters, over the boarding to which the shingles are nailed." [NOTE 10]. This meant that dormers could be added or removed more easily. There does not seem to be any parallel for this in the French sources, although it is certainly a valid method for constructing an added dormer.

Plate 1 would seem to indicate that dormers do not have to be framed between two common rafters.

(V)

COVERING

Toisé information for dormer covering is vague. The slate and shingles is always estimated for dormers, and therefore is difficult to interpret. The views indicate that both the roof and the cheeks are covered with the same material as the main roof. (See plans 725-8, 731-3a, 743-1.) One exception to this is Plate IX, fig. 150 bis which shows a colombage dormer with visible framing, and a slate roof.

Plan 733-9 shows horizontal sheathing on the cheek of the dormer, and plan 743-1 indicates horizontal sheathing, or possibly even shingling on the gable end of the dormers.

(VI)

FLASHING

Flashing is an essential feature of dormers. since they are inclined to leak if not well made. Documentary sources indicate that both lead, soldered with "Bristok" [Bristol ?] pewter,, and plaster were used in Louisbourg. The examples shown in Plate IX are liberally plastered,, including the ridge. Plastering the ridge seems to be associated with the use of tiles and does not appear to have been practised at Louisbourg.

Doyon and Hubrecht recommend the use of the rounded valley [NOTE 11]. This feature is well described in two extracts dealing with slating which have recently been acquired from the Engineering section. 12 (See Plate X.)

(VII)

GENERAL DATA

Dormers can be closed with sashes, or with shutters, or, presumably, with both. There is no concrete information concerning this point, and it seems logical to assume that the use of the room in which the dormer is located, and its accessibility, are determining factors.

The documentary evidence suggests that dormers were added or removed quite frequently in Louisbourg; for example, the repairs to the house and storehouse mentioned above included the addition of two dormers to the attic of the storehouse [NOTE 13], and a tenant was ordered by the court to remove or block in a dormer that he had added [NOTE 14].

Finials are shown on plan 725- 4a, 743-3, and Plate V. The dormer on Plan 725- 4a is hipped, that on 743-3 is gabled, and the shape of the third one cannot be determined.

Doyon and Hubrecht state that the rafters of a dormer should not show, except if the dormer is on a colombage house [NOTE 15]. Plate IX, figs. 150 and 150 bis indicate a slightly projecting moulded plate which receives the rafters. The slates project over this plate to provide the overhang. Unfortunately, none of the primary sources for Louisbourg are detailed enough to confirm this particular feature.

(VIII)

CONCLUSIONS AND RECOMMENDATIONS

The available evidence on dormers does not permit one to draw many conclusions. or to make firm recommendations. However, a few points seem to emerge from the data presented and should be mentioned.

The location of a dormer, and its size, should be determined on the basis of the room in which it is located, and on the function of this room. Sizes, wherever possible., should be based on the documentary evidence. New toisé information will be appended to this report when it becomes available, and this should. provide a good basis for selection.

The most usual shape for Louisbourg dormers should be gabled,, with a rectangular opening. A few examples of hipped dormers, or arched openings would not be out of place. It is unlikely that one would find both hipped and gabled dormers on the same roof. Finials should be used sparingly.

An overhang seems to be necessary, and the solution indicated by Doyon and Hubrecht appears to be appropriate.

Framing should be based on the Toisé sizes and on the detailed sections shown on plans 733-9 and 733-11 as far as possible.

Dormers should be covered with the same material as the main roof, including the cheeks, and possibly the faces. The sheathing should probably be horizontal as indicated by the plans 733-9 and 743-1. Dormers on colombage houses could probably be modelled after the example shown in Doyon and Hubrecht, although this type of construction may prove to be impractical in the Louisbourg climate.

A rounded valley should be used whenever possible.

Flashing is mainly a practical problem,, but on the basis of the evidence,, it would seem that more plaster could be used.

Some attempt should be made to recreate the transitory aspect of dormers. Although no specific information for a particular building has been found, the possibility of added or deleted dormers should be kept in mind when designing domestic structures.

ENDNOTES

[NOTE 1]: 7 août 1752, AFO G2, vol. 208, dossier 475, pièce 72, Reparations dans une maison; [NOTE 2]: G. Doyon et R. Hubrecht, L'architecture rurale et bourgeoise en France,, Paris, Vincent, Fréal et cie., 1958, pp. 218-26; [NOTE 3]: Ibid; [NOTE 4]: 1 septembre 1731, AC C11B, vol. 12, ff. 122-43, "Toisé général et définitif ... "; [NOTE 5]: 30 septembre 1734, AC C11B, vol. 16, ff. 196-213, "Toisé définitif des ouvrages; [NOTE 6]: 1 novembre 1727, AC C11B, vol. 9. ff. 180-92, "Toisé des ouvrages; [NOTE 7]: 26 juillet 1754, AFO G3, carton 2042, no. 69, Devis d'une maison; 28 juin 1757, AFO G3, carton 2045, no. 27, Bail à loyer; 7 août 1752, AFO G2, vol. 208, dossier 475, pièce 72, Reparations dans une maison; [NOTE 8]: 1 septembre 1731, AC C11B. vol. 12, ff. 12-2-43, "Toisé général et définitif ... magasin et angar; [NOTE 9]: J.-F. Blondel, Cours d'architecture, Paris,, chez Desaint, 1791, T. VI, p. 272; [NOTE 10]: M. Whiffen, The Eighteenth Century Houses of Williamsburg. New York, Holt, Rinehart, & Winston, Inc., 1960, pp. 74-75. [NOTE 11]: G. Doyon et R. Hubrecht,, loc. cit.; [NOTE 12]: Report on French methods of slating., 1966; Slate roofing: excerpts from the National Slate Association Publication; [NOTE 13]: 7 aoûst 1752, AFO G2, vol. 208, dossier 475, pièce 72, Reparations dans une maison; [NOTE 14]: 14 février 1757, AFO G2, vol. 209, dossier 505, ff. 2(v)-3(v), Plumitif d'audience; [NOTE 15]: G. Doyon et R. Hubrecht, loc. cit.

----------------------------------------------

Return/retour