THE ARRIVAS HOUSE RESTORATION DRAWINGS
During the process of investigating this house, over the course of this
summer, a number of attempts have been made to visualize its condition at various times.
Floors and sub-floor walls have been dated archaeologically and suffice to show basic
outlines at various periods. Maps and their brief descriptive keys indicate the
number of stories (though not with great consistency), the type of materials, the
owner and the general condition. Photographs are helpful, but only after about 1860
and for the parts of a structure that they show. Inevitably the most convincing
visualization results when all of these, together with a sense of historic occur-
rences and ownerships, are related to the remaining fabric of the structure. In the
case of this house, a large portion of the remaining fabric is a wood frame second
story. The most authentic restoration of this house would, therefore, attempt to
represent its condition when this story was in its original state.
For a time it was felt that the majority of this second story was the one
referred to by Mariano de la Rocque in the description accompanying his map of 1788,
although he described it as "a house of masonry, of a story" which corresponds with
his description of other two story masonry houses. It was felt that the brevity of
his descriptions were subject to misinterpretation, that there was evidence of some
inconsistency on his part in other instances, that the lower masonry walls showed
no evidence of or capacity for a masonry second floor, that the second story
framing corresponded with early British practice, and that the evidence of a pro-
jecting balcony over the street coincided with 1760-80 descriptions of the city.
However, all of these feelings were put in proper perspective by the discovery that
much of the stucco lath was secured to the original frame with cut nails of a more
recent origin. The exact date of the particular type of cut nail used has not been
determined but is presumed to be around 1830. Fred Gjessing's report will provide
further enlightenment on this.
The determination of the probable date of the second floor was a distinct
disappointment to all who were concerned with this project. Further efforts were
made to construct a visualization which would have a masonry second story and could
be justified by some physical evidence. None of these succeeded in being sufficiently
convincing to merit serious comparison with the evolving knowledge about the nature
of the house with its wood frame second story. These efforts seemed to underscore
the recommendation of restoration professionals, as expressed by Fred Gjessing and
Earle Newton, that the best evidence should be used.
The drawings represent a serious attempt to show the building as it existed
when the most evidence was relevant. At that time it contained many parts of older
structures just as it does today. It is felt, however, that the period for which we
have the most evidence, and to which it would be restored on the basis of these
drawings, was also the period of its highest development as a home for St. Augustinians.
As indicated by the archaeological investigations and the character of the
second floor framing, the first floor consisted of four interior rooms (with plastered
coquina masonry walls and partitions), an arcade or loggia onto a south patio (with
coquina piers or arches), a porch or loggia onto a north patio (with wood posts),
and a similar porch to the west. The floors to all of these areas except the west
porch and south patio were what is now the top-most layer of tabby. Due to grade
conditions the west porch floor was probably tabby as well, though none was found
in this area and wall remnants there seem to indicate the existence of a wood frame
and floor at some period. The absence of tabby there is presumed to be the result of
destruction caused by numerous pipe trenches and other occurrences which have left
the ground in a very disturbed state. The wood frame and floor are presumed to have
existed during the time of the three west rooms shown on Rocque's 1788 map. A well
was found just west of this area. It is believed that the kitchen in use during the
subject period will be found under the present office and south of the well. A
building shows in this location on the 1893 insurance map, and people still living
remember "the old stone kitchen" south of the well. Undoubtedly gardens and out-
houses existed to the west beyond the well.
The east wall of the house and the patios was located at the present street
curb line. This was determined by excavation at the northeast corner and by early
photographs. The placement of the curb has destroyed this wall to such an extent that
further excavation, to attempt to locate door openings, was not considered feasible
at this time. Door and window openings have been located by projection from photo-
graphs and adjustment to the earlier room uses (resulting in the abandonment of a
central door which shows in the earliest photograph). The size and type of these
windows have been made to match others which remain in the building and this also
seems to correspond with photographic evidence (although the photographs are not clear
enough in this area to show convincing detail).
The south wall of the south patio was incorporated into the north wall of
the Rodriguez-Watkins House in accordance with wall rights granted in 1760. It may
be possible to determine something more of the character of this patio when the present
office structure is removed. The arcaded loggia portion of the house which faces
this patio is based partly on the memory of living relatives of the Arnaus, who lived
in the house during most of the 19th century, and partly on physical evidence. The
second pier from the east, which now supports the second floor fireplace, is a
newer construction than the other piers and has been deleted in the drawings. The
other piers show evidence of having been cut square above a certain level. This
level has been used as the spring-line of the arches. A pier to replace the deleted
one has been placed in a logical position to form reasonably regular arches. A pier
to replace the deleted one has been placed in a logical position to form reasonably
regular arches. It is possible that this pier was also sculptured similar to the
other interior pier as it exists today. This has been considered relatively unlikely,
however, due to the loss of strength this would entail and to the fact that stucco
remains on the sculptured portion correspond with that on later additions in other
parts of the structure (such as the west chimney base).
The location of stairs was a considerable problem. For a long time it seemed
that there was no evidence of stairs, other than those which had been recently
built over the interior fireplace and a curious floor framing condition at the south'
east corner of the loggia. Parts of this framing seemed to be very recent. We
pursued this clue, however, and found (upon removing stucco on the adjacent pier)
the imprint of a stair carriage where the mortar of masonry in-filling had been ex-
truded against it. We then looked more closely at the earliest known photograph
and were able to detect the railing around the well for this stair. This has been
further confirmed by Arnau descendants. We felt that there must have been a stair up
to the west porch second floor also. The 1893 insurance map indicated one on the
outside of the porch area leading up to the north. One of the Arnau heirs seemed
positive that it went down straight west from the stairs to the attic. No doubt
both of these existed, but neither are shown, for several reasons. The first is
that the south wall projected somewhat to the west of the west wall of the house,
and until a fairly recent date as no stucco is to be found on the broken ends remain-
ing. How far this projection extended is unknown. It is felt, however, that some
functional object must have existed near it in order to justify its intrusion into
the porch area. The stairs shown seem to be ideal for this. The second reason
is that sheltered access to the upper rear area is preferable (certainly for
modern occupancy by the Commission) and that the possible sacrifice of authenticity
for this element is not a serious consequence.
The door and window openings from the arcade into the south rooms are all
justified by physical evidence. The eastern-most of these appears to have been in
continuous use since the first coquina structure. The central door opening has
been squared-off in recent times but evidence of its former beveled jambs is still
visible. The westernmost opening was originally a door and was later converted
into a window. By correlation of the various coats of plaster it has been decided
that it became a window either before or during the time of the remodeling which
included the construction of the wood frame second story.
The west wall openings also appear to have been in use since the earliest
coquina structure adjacent to them. Both have been reformed several times and are
now shown with beveled jambs and sizes which conform to the remainder of the first
floor. The fireplace and chimney have been determined to be a later addition and
have been deleted in the drawings. This determination was partly based on the exist-
ence of stucco on the original wall against which the chimney base was built and
partly on the condition of the wood frame and stucco behind the second floor portion
of the chimney.
The door openings now shown in the north wall also appear to have been
in use since the time of the walls in which they exist. This is borne out by the
archaeological excavations at the easternmost door. This door has been widened
beyond its earlier condition and is now shown with its beveled jambs and probable
original size restored. The westernmost door has not been studied below its present
sill level, is curious in its proximity to the northeast corner and the west door,
and also shows strange inconsistencies in the shaping of its east jamb. Nevertheless,
it has been accepted, both as a necessary access to the north patio during the
existence of the w est rooms and, later, as a convenience which was maintained. The
irregularities in the east jamb are assumed to have occurred as a result of careless
cutting and of the reconstruction of the upper portion of the wall. The central
opening, now shown as a window, actually exists, at present, as a door. However,
the previous window opening and frame are visually evident. The way in which some
of the coquina, around the window opening, are cut indicates that the window too
was probably built into an existing wall. This would have been consistent with the
need for light after the addition of the west rooms (prior to 1788). In any event,
it seems to be logical as a window for the house as remodeled with the second story
The central fireplace was built against existing coquina walls and on top
of the latest tabby floor (apparently British period). There are a number of factors
which indicate that it too was a part of the remodeling which included the construc-
tion of the wood frame second floor. One of these is the angle at which the begin-
ning of the flue is constructed. As may be seen in the drawing of the reconstructed
fireplace group, this angle is ideal for directing the smoke from this fireplace into
an upper cluster of flues serving other fireplaces as well. Also, the placement of
this fireplace in the corner is less than ideal for this room, but does locate it so
that a fireplace above will be centrally located on the interior wall of the main
upper room. The existence of such an upper fireplace, together with two smaller ones,
has also been verified. The upper closet and roof framing also correspond with these
assumptions. The trim indicated for all fireplaces presumes that they were built at
- 7 P
the same time and that the trim used would correspond with other trim used at that time,
particularly for the second floor doors and windows. The extent of this fireplace
trim could be determined fairly accurately from smoke smudges and other markings on
The minor second story fireplaces were determined from the location of the
attic stair partition and from indications of their hearths. This evidence primarily
occurs on the uncovered floor planking, but is further substantiated by the charred
condition of some lath and the ceiling framing for the attic stairs. The trim for
these two fireplaces is more speculative than for the others but seems reasonable.
It was felt that the later fireplace on the south wall, which still exists, was built
to match these, and its trim has been used as a model.
Much of the interior plaster is original and serves as a guide to the
original room arrangement as shown. Marks on the ceiling joists indicate that it too
was finished with wood lath and plaster.
The original flooring, of wide wooden planks, also still exists in most
areas, though covered in some places by later flooring materials. Not only did this
help in locating fireplace hearths, but it also served to locate the doors between
the west rooms. Under the flooring, in the west central area, beams were found which
had beaded edges and mortise cuts indicating a probable earlier use as exposed beams
under a flat roof. It is believed that they were cut from mulberry. None are now
in the original position and many have become unserviceable due to decay.
The southeast corner room and the east balcony are shown as indicated on
the earliest known photograph. This is consistent with evidence found in the walls,
with railings which still remained on the north porch, and with much of what is known
of St. Augustine practice around the beginning of the 19th century.
The north balcony, or porch, size was determined from the condition of the
ends of the continuous beams which supported it. These also indicated the existence
of an edge beam which was needed to provide additional support from columns or
posts. These wood posts were made to correspond with those which would have occur-
red above by following the pattern shown in the early photograph.
The west porch is assumed to have corresponded with the north. At the
southwest corner, there are markings on the corner-post which indicate the terminus
of a railing and which show the location of an eave closure beam. The height
of the eave beam corresponds with that on the south side of the east balcony. By
employing the photographic evidence as to roof slope and by assuming a ridge location
over the central interior beam, the west eave projection then corresponds with the
depth of porch also determined for the north side. It has been assumed that wall
support for the floor beams was provided by the large wood plate, though no physi-
cal evidence was found to justify this.
Most of the interior trim and many of the doors and windows were deter-
minted, by their paint coatings, to be original materials. Where they were not of
this nature, the drawings now show them restored to this condition or one which
ATTIC AND ROOF
The existence of a former attic was suspected from the amount of space
which would be available under the roof indicated in early photographs and from
the evidence of nails and planking on the upper side of the ceiling beams in much
of the eastern portion. This was later confirmed in conversations with the Arnau
descendants and by evidence of the stAir to this attic. The location of partitions
is a speculation based on these conversations and the demands of logic. There is
little physical evidence observable at this time, though this may alter after the
existing roof is removed. The window locations are strictly speculative. As has
- 9 -
been noted, the form of the roof has been derived from photographic evidence.
The covering of wood shingles also derives from this, from the practices then
current and from the 1893 insurance map. The framing method is one which was fairly
common except in the way in which the eaves were framed. The extended ceiling
beam rafter connection has no known counterpart in this area. It is, however, the
only way to justify the absence of any indication of the use of another method and
maintain structural stability and consistency with other details. The collar beam
above is a detail frequently found in old St. Augustine structures and accounts
for the lack of evidence as to any other kind of central support. It also creates
the kind of attic space that corresponds most closely with obtainable descriptions.
It is not known how the attic was finished and we have simply assumed that it too
William A. Stewart
12 September 1960
Recommendations Relative to the Arrivas House Reconstruction
In addition to the nature of the construction team, previously referred to,
the following recommendations are offered for consideration:
I. A general clean-up of the premises should precede any demolition or reconstruction.
This should include:
a. A safety cage over the well;
b. The removal of debris from the archaeological trenches;
c. Refilling the trenches with selected sand, except at a point
adjacent to the south edge of the interior fireplace;
d. Refilling by the fireplace with concrete containing fine aggregate
(3000#, 7" slump);
e. Shoring the main floor beam on this concrete (preferably with
a steel screw column);
f. Reshoring the easternmost interior door arch;
g. The removal of all debris from the premises (being careful
to save large fragments of tabby and coquina, doors, hardware and
other items that may have significance for display or reuse);
h. Cutting weeds and policing the grounds;
i. Posting the property more adequately and providing some means of
restricting access to the interior of the structure;
j. Establishing a method and a place for cataloging and storing
relevant parts of the structure to be disassembled; and
k. Clearly defining a separate place for the temporary collection
of the loose debris resulting from disassembly.
II. Raze the front portion of the present office, being especially careful with
the removal of the floor slab.
- 10 -
III. The entire second floor should be disassembled. The existing roof struc-
ture, and other portions not indicated as being included in the restored
building, should be discarded.
IV. Those portions of the coquina walls which are loose should be removed and
reset. This refers especially to the interior door arches.
V. New materials, to replace those discovered not to be reusable, should be
modern equivalents of the old (rather than imitations of them, regardless
of how carefully contrived), such as:
a. Wood framing of stock mill lumber chemically treated and
dyed for identification;
b. Wood trim, similar to above, but shaped to correspond with that
c. Coquina masonry set in portland cement mortar;
d. Portland cement stucco;
e. Concrete patching for the tabby floors; and
f. Modern glass, glazing compounds and paints;
Note: The question of hardware seems to be the most serious one in connection
with this policy. At the least, the metal should be treated with contemporary
rust-proofing materials. This decision should also be influenced with the probable
necessity of incorporating modern lighting and other conveniences in the building.
It may well be best to be honest and consistent with all of these elements.
VI. Obtain expert advice on the landscaping of the patios and rear yard.
VII. Continue the photographic recording of all new aspects of the building,
both as it is dismantled and as it is rebuilt. The uncovered second story
wall framing will be of special interest.
William A. Stewart
12 September 1960
SUPPLEMENT TO ARCHITECTURAL REPORT
The recent efforts to determine and describe the one story version of the
Arrivas House, as it may have existed around 1730 or 40, are primarily based upon
1. Archaeological studies;
2. Existing coquina walls;
3. Remnants of salvaged beams (re-used in the floor of the later
second story); and,
4. Documentary descriptions (but not of this specific structure).
Further archaeological work, undertaken during this last week, reveals
some items which have not yet been accommodated by the latest drawings. These are:
1. The location of two doors in the east (street) wall; and,
2. The westward extension of the south main wall.
This last item poses a question which is not readily answerable.
Also, no further light was cast on the q question of the character of the semi-enclosed
space at the east end of the loggia. We are not satisfied as to its probable use
or in the manner in which it was roofed.
The absence of fireplaces and of some of the existing masonry columns is
determined by evidence in adjacent floors and walls. Window and door types are a
combination of descriptive documents and apparently logical use. In terms of their
use in this building they are speculative.
The roof is based on the beam remnants and descriptive documents. Its
height, and the height of the parapet walls above it, are speculative. The same is
true of the loggia roofs on the north and south sides.
It is hard to believe that the family we know lived in the house at that
time, could have been contained in the structure shown. In addition to separate
kitchen and out-house buildings, there must have been at least one other structure.
It is possible that this existed across the court to the north, or was a westward
addition of which only a portion of one wall remains. In any event, the building
shown cannot represent the whole story of the facilities present at that time.
In conclusion, it should be pointed out that those who have worked on
this problem are less satisfied with this version than they were with the results
of the previous investigation and the first plans based on them. Mr. Gjessing's
sketches of this "early" version correspond very closely to ours and may be con-
sidered identical in character. Nonetheless, we are less sure of this character,
(and the elements which compose it) and we consider the results to be less dis-
tinguished or desirable. We can only say that they are as honest as they can be
at the present time.
Oct. 24, 1960 Sincerely,
William A. Stewart