EARLY RHODE ISLAND HOUSES I SHAM AND BROWN -fV-V WM- EARLY RHODE ISLAND HOUSES Digitized by the Internet Archive in 2008 with funding from IVIicrosoft Corporation http://www.archive.org/details/earlyrhodeislandOOisharich Early Rhode Island Houses HISTORICAL AND ARCHITECTURAL STUDY NORMAN M. ISHAM, A. M. INSTRUCTOR IN ARCHITECTURE, BROWN UNIVERSITY ALBERT F. BROWN ARCHITECT Providence, R. I. PRESTON & ROUNDS 1895 M^ 7Z3S Copyright, 1895 by preston & rounds COMPOSITION AND I'KESS-WORK liV E. L. FREEMAN & SON, PROVIDENCE, R. I. PLATES MADE BY RHODE ISLAND PHOTO-ENGRAVING CO., PROVIDENCE, R. TO William Carpenter, John Smith the Mason AND the rest of THE early craftsmen of Rhode Island iW243'?52 PREFACE HE present book is the result of much observation and study of the early colonial work in Rhode Island. Yet it can hardly be said to have exhausted even so small a subject as this would seem to be. Newport and the Narragan- sett country each deserve a book by themselves, and each needs more complete exploration than we have been able so far to give them. As far as the book goes, however, we claim for it consid- erable accuracy, and whatever may be the reader's opinion of the theories put forth in the text, he may rest assured that the draw- ings are veritable historical data. Every plan, elevation and section is based upon measurements of the house it illustrates ; and the perspectives are made — two from pencil sketches made on the spot, the rest of the number from photographs. We have thus personally examined, sometimes from garret to cellar, every house described in the text, and our thanks are most heartily tendered to the courteous owners and occupants who al- lowed us to explore, measure and sketch at our leisure, and often shared our enthusiasm. Every house in the catalogue in Chapter IX has also been seen or examined either by ourselves or by Mr. Edward Field, Record Commissioner of the city of Providence, who has worked much with us and to whom our thanks are due, not only for his exploration but for the keenness with which he has 6 EARLY RHODE ISLAND HOUSES. run to earth the documentary evidence for the dates of the Prov- idence houses. These dates we have merely stated, leaving the authorities upon which they are based for him to pubHsh. In the cases of the Newport and Narragansett houses we have no documentary evidence to show for the dates. Both the New- port and North Kingstown ' records are in such a condition that little can be learned from them. We have given conjectural dates for all these houses based upon the Providence work and upon the date of the Smith house, which we believe to be 1678-80. Perhaps a word may be necessary on the meaning of some geographical names which are now much narrower than they were two hundred years ago. Providence in the sense in which it is used in this book means the whole northern part of the State, practically the present Providence' county west of the Seekonk and Blackstone rivers. Warwick included the present town of Coven- try ; it was really the strip of land between Warwick and Gaspee points, running twenty miles inland. The territory which now forms the towns of Cumberland — the old Attleborough Gore — East Providence, Warren, Bristol, Tiverton and Little Compton, was part of the Plymouth Colony. We hope that this work will be a help to the future historians of New England and that it will promote the collection of scien- tific data about the oldest houses in the original New England colonies, so that the vague descriptions of too many of our town histories may be supplemented by accurate measured drawings. ' If we include the " Pawtuxet Purchase." CONTENTS CHAPTER I. Introductory CHAPTER II. The Houses of the Earliest Period, 1636-1675 - - . - ig 1. The Roger Mowry House, Providence. 2. The Arthur Fenner House, Cranston. CHAPTER HI. The Houses of the Second Period, 1671^-1700 ----- 30 The Thomas Fenner House, Cranston. The Edward Manton House, Manton. The Thomas Field House, Field's Point. The Eleazer Whipple House, Lime Rock. The Eleazer ArnoUl House, Moshassuck. CHAPTER IV. The Houses of the Third Period, i 700-1 735 ----- 4^ 1. The Epenetus Olney House, North Providence. 2. The Benjamin Waterman House, Johnston. 3. The John Crawford House, Providence. 4. The James Greene House, Buttonwoods. 8 EARLY RHODE ISLAND HOUSES. CHAPTER V. Newport 55 CHAPTER VI. Narragansett - 6i CHAPTER VH. Construction ----------- -68 1. Stone-work and Brick-work. 2. The Frame. CHAPTER Vni. Relation of Colonial Architecture to English Work - - - 83 CHAPTER IX. List of Old Rhode Island Houses 90 Index of Names and Places ---- 98 LIST OF PLATES. LIST OF PLATES. Front ispie Tailp ece a Plate I. (( 2. k( 3- (( 4- a 5- a 6. " 7- a 8. (( 9- a lO. 11 1 1. i 1. 12. 11 13- bi 14. a 15- a 16. it 17- (( 18. 11 19. ( ( 20. (( 21. a 22. ii 23- (( 24. a 25- a 26. ti 27. ece. Thomas Fenner House. Gravestone of one of the Fenners East Room, Smith House Development of the Rhode Island Plan Map of Seventeenth Century Providence Roger Mowrv House . - - Arthur Fenner House Thomas Fenner House Edward Manton House a li Thomas Field House (( n Eleazer Whipple House Page 29 " 67 Present Exterior Restored Plan Restored Section Restored Exterior Framing Exterior Ruins Measured Plan Restored Plan First Story Plan Second Story Plan Section Present Fireplace Ancient Fireplace Restored Framing Restored Exterior Present Exterior Plan Section Restored Exterior Exterior in 1894 Plan Section Present Exterior Plan 10 EARLY RHODE ISLAND HOUSES. Plate 28 29 30 31 32 33 34 35 36 37 38 39 40 4' 42 43 44 45 46 47 48 49 50 51 52 53 54 55 50 57 58 59 60 Eleazer Arnold House Epenetus Olney House Benjamin Waterman House John Crawford House James Greene House .k (( 4; Governor Henry Bull House Sueton Grant House it (( a Richard Smith House Phillips House - - - Arthur Fenner House ^ Thomas Field House ' Benjamin Waterman House Epenetus Olney House John Crawford House Framing Details — Rafters " " Overhangs Map of Early Rhode Island Present Exterior Plan Cross Section Longitudinal Section Restored Exterior Present Exterior (1894) Plan Cross Section Longitudinal Section Framing Present Exterior First Story Plan Original Second Story Section East Elevation North Elevation Second Story Plan Cross Section Longitudinal Section Present Exterior Plan Plan Plan Restored Exterior Plan Chimney Details Details Details Details First Story Details Second Story Details In pocket, rear cover > > J > J -s > w EARLY RHODE ISLAND HOUSES. CHAPTER I. INTRODUCTORY. ARLY NEW ENGLAND was not without poHtical his- tory of a very active kind. Several different colonies existed within its narrow territory, and as a result of the relations between these colonies we find several more or less clearly marked schools of architecture. Massachusetts had one style ; Connecticut another, slightly, if any, different ; and Rhode Island a third, which seems quite clearly separated from the other two. With the internal differences in the two former schools we have little to do.. In the present limits of Rhode Island we find a difference between Providence and Newport, and again between these . and the old "King's Province," the "South County" of fa- miliar language. * The causes of these differences lie, to a great extent, as indi- cated above, in the political history of early Rhode Island, which was not entirely peaceful. In 1636 Roger Williams purchased the Providence Plantations. Gorton, in 1643, bought Shawomet ; and 12 EARLY RHODE ISLAND HOUSES. between the two, in November, 1642, John Greene had settled at Occupasnetuxet, now Spring Green. Of the islands in Narragan- sett Bay, Patience and Prudence were owned in partnership by Williams and Gov. Winthrop ; while Aquidneck and parts of the others of the lower group came into possession of those who settled Newport and formed the colony of Rhode Island. In 1638, or very near it, Major Richard Smith settled in the Narragansett country at what is now Wickford. He and others held what was known as the Atherton Purchase, the strip of coast northward to Gorton's southern line. West of them was the Fones Purchase of about the same size. Southward, still on the east or shore side of the Great Swamp, the Pettaquamscut and other purchases continued the line of settlement ; while pioneers from Newport, landing at Westerly, spread gradually up the river valleys into the region westward of the Swamp, the present townships of Hopkinton, Richmond, and the rest. The people of the different settlements were of different characters, and these communities sometimes quarrelled with each other and sometimes fought fiercely in disputes at home. Now with this state of things within — and the picture is rather mildly drawn — must be combined the greed and machinations of Massachusetts on the east and of Connecticut on the west, the one hating the idea for which the new colony stood as well as desiring its territory, the other not actively hostile, but anxious to extend its jurisdiction to the shore of Narragansett Bay. There can be little doubt that these two colonies fomented the controversy aroused by William Harris and his claims ; while the eighteen years of Massa- chusetts jurisdiction over Pawtuxet, the Gorton episode, and the leaning of the Smiths toward Connecticut, show that the unity for which, fortunately, the greater minds in the colony did not cease to struggle, must often have seemed a desperate matter. INTRODUCTORY. 13 Since the early history of the colony was such as we have de- scribed, we may now perhaps see why there should be, not only in New England but also in the narrow limits of Rhode Island, differ- ences in building corresponding to the different political divisions ; why the earliest buildings of Providence find little analogy in the Bay or in Connecticut; and why Newport, scarcely thirty miles away, should exhibit small architectural affinity, except at the very outset, with its sister colony. In fact, Newport work seems to lean some- what toward the school which prevailed at Hartford, a fact hard to account for except through the influence of trade along the shore of the Sound. The two types, Providence and Connecticut, seem to meet there, as also at Wickford. Below Wickford in the South County the examples of very early date are wanting, but the older houses are sometimes of the Connecticut type,' sometimes very near akin to the work in Providence. The character of the architecture of the early colonies depends, also, very closely on the early artisans and the training they brought with them from the old world. Each colony had its craftsmen, and only the first log huts can have been built by the settlers them- selves. Sawyers are mentioned very early in the records of the Bay Colony, as well as carpenters, masons, brick -layers, and thatchers." All these men had learned their trades of English or Dutch masters in England or in Holland, and to their apprentices they taught the methods they had learned in their youth. Their ideas were some- what limited. They were of the humbler classes, and each followed, so far as we can judge, the style of building which prevailed in the district whence he came. The early types disappear as the old ' In Plate i, B is the type of house prevalent in Hartford, while E was the type in Salem. '■'Colonial Records of Massachusetts, Vol. I., p. 74. See, also, Contract for building fort, in I'l)-- mouth Records. 14 EARLY RHODE ISLAND HOUSES. craftsmen and their immediate apprentices die out, but the three original New England Colonies are never the same in their colonial architecture. How much Dutch influence there was in the early work in Providence it is hard to say. There was much intercourse with New Amsterdam, which was apparently friendly, and which no doubt furnished a ready market for furs, while its people possibly sold the colonists tools and other manufactured articles which could not be bought in Boston. There were also Dutchmen in the town itself. Christopher Unthank was one, and the unhappy John Clawson an- other, and the latter was by trade a carpenter. In addition to these sources of influence, we may also consider what Dutch traditions there may have been in the colony of Plymouth, whence many of our Providence settlers came. When we look at actual work as it remains to us, we find noth- ing which cannot be thoroughly English, and due to the English training of the original settlers. One method in use in Providence, which it seems to share only to a limited extent with the other colonies, at least in houses, is that of boarding vertically. Even this, which certainly was used in the Low Countries, may be Eng- lish also, for it was used at Hildesheim in Germany, and hence was not confined to Holland. Still, even if it came from England, it might have been a Dutch importation, if we are to credit those industrious immigrants with what may have been common to all Mediaeval Europe. The obscurity of early Rhode Island history is well known. There was no historian, not even a diarist of any account, and the separation of church and state deprives us of church records. The public records are rather brief and fragmentary, and are full of gaps; and the meagreness of their references to building — for the INTRODUCTORY. 15 inventories do not begin till comparatively late — makes it difficult to gather from contemporary evidence of what kind were the houses of the earliest settlers. When we say of any house now standing that it was built before King Philip's War, we cannot stand pre- pared to prove our statement with documents signed, sealed, and witnessed. Nevertheless it is fairly well known of what sort these early houses were, and we shall make the First Period of the three into which we intend to divide the chronology of the subject extend to 1675, the date of the Indian War. The Second Period extends from [675 to 1700. The Third Period, the last of our divisions, brings us down to 1725-30, when the old forms of construction were abandoned, or rather were transformed, and the pre-revolution- ary style began — a style more easily recognized as "Colonial," and closely akin to that of the great houses which from 1 750 to the end of the century gave its peculiar architectural character to the Atlantic seaboard. These periods are not so arbitrary as thfey^ look. The war with King Philip was one of extermination on both sides. Its success- ful end marked a great step in colonial progress. Security was assured ; the Indian question was settled in Eastern New England. From now on the outlying settlements in the Plantations grew stronger. Again, it was about the beginning of the i8th century that occurred the significant change in the habits of the good towns-people which turned them from agriculture to sea-traffic and brought in the wealth and the wider ideas which, acting with the weakening of the old traditions under successive apprentices, de- stroyed the almost medieval types of the old craftsmen, and sub- stituted, not all at once, of course, work akin to the classic models with which Chambers afterward made men so familiar. The first houses of Providence, built around the spring near 16 EARLY RHODE ISLAND HOUSES. where St. John's Church now stands, were no doubt of logs halved together at the corners, and contained but one room roofed with other logs, or with bark or thatch on poles. The chimney, if there was one — for the settlement was made in the early summer when cooking could be done out of doors — was probably also of logs, on the outside of the house, at one end, and like the house was plas- tered with clay. But these huts were only temporary. The news of the founding of a new settlement soon attracted those who through tenderness of conscience or through contumacious disposi- tion could no longer dwell with peace and comfort in the Colony of the Bay. With the immigrants came craftsmen, if indeed they were not among the original few; and perhaps as winter came on the new plantation of Providence began to have more substantial dwellings, akin to what Roger Williams called an "English house.'" The houses which succeeded log huts did not differ from them in plan. They contained only a single room," the " Fire Room," one end of which was almost entirely taken up by a huge stone chimney with its cavernous fireplace.' Beside the fireplace, in the corner of the room was the staircase — little, if anything, better than a ladder — which led to the "chamber" above; for few of these houses were more than a story and a half high. A glance at Plate I, A will show this arrangement, and also will give, in B, the plan in vogue in the colony of Connecticut, and, in E, that common in Salem. It will be noticed that a Rhode Island plan is just the half of one of those in vogue among our early neighbors of Connecticut, and so it remained until nearly 1730. Nor did this iR. I. Hist. Coll., Vol. IIL, p. 166. ' Providence fireplaces are larger than those of Connecticut. ^H. C. Dorr, Planting and Growth of Providence, p. 24. We reached our conclusion, however, before seeing his work. INTRODUCTORY. 17 earlier form, which we have just described, undergo, as far as can be seen, any modifications except the lean-to, before King Philip's War. More than that, it lived on for some years side by side with a later form, which we shall next describe. After the close of the Indian War some of the burned houses were rebuilt on the same primitive lines, but in a few years the increased sense of security and the greater wealth which now pre- vailed brought about a change. The older houses were added to, partly by the lean-to, partly by lengthening at the end away from the chimney. This probably took place very early, and was not confined to any one time. But the main characteristic of the sec- ond period is the construction, under one roof, with a lean-to and with or without additions at the end away from the chimney, of houses whose plan is given at C, Plate i ; and the difference which marks the third period, which sometimes contains houses of the plan of the first, but mostly of the second, is that the houses are often of two full stories and the chimneys are partly or wholly of brick. Beyond the third period, or beyond 1725, the transition is rapid, but it takes two directions. From C, which with brick chimney is practically the plan of the old Crawford house at the corner of North Main and Mill streets, the step is easy to the plan at D, which is that of the Brown house, all of brick, on the grounds of Butler Hospital. It is also a very common disposition of an end- chimney, when an old house has been lengthened, as mentioned above, at the end opposite the original chimney. But this form seems never to have been developed further. Another, and a more convenient, supplanted it, and it was only after many years that it reappears, and then it does not have its original form ; it is rather a fresh discovery. 18 • EARLY RHODE ISLAND HOUSES. In F, Plate i, we have this supplanting form. The chimney is now in the middle or nearly so, as for a time at least the original single room is larger than the room which was added at R. In many old houses, notably in the Tillinghast house on South Main street, built probably about 1730, the "great room," the descendant of the single room in the ancient houses, as they were the descend- ants of the old English " hall," has two windows, while that on the opposite side of the chimney has but one. The staircase, it will be noted, has not changed its place — it is still next to the chimney in the same relation to the old room as before ; and the door, which in some old houses opened upon the stairs, has been brought na- turally into the centre of the new front, without changing its old location. Soon the two rooms became equal, with the chimney still in the centre, and now nothing except detail distinguishes the Rhode Island house from those of the neighboring colonies. Next, each room had its chimney {G, Plate i) and the hall ran through the house. Finally, at the end of the eighteenth and the beginning of the nineteenth century, each of the four rooms has its fireplace ( H, Plate I ) and these are in the wall again. This broad classification generalizes the architectural history of Rhode Island. With its later forms this essay has nothing to do, except incidentally. We shall now go back and take up the earli- est houses, giving in each period a particular account of each note- worthy structure. This done, we shall study carefully the materials, the methods of framing and other details of construction. In clos- ing, we shall attempt to trace the relation of the early colonial craftsmen to the work in Old Ens^land. CHAPTER II. THE HOUSES OF THE EARLIEST PERIOD, 1636-1675. HERE are three sources of evidence as to these oldest houses. We have certain traditions handed down, some- times in famihes, sometimes from one owner of the prop- erty to another. We have also a meagre amount of documentary evidence, partly contemporary, partly later in date, but of such a character that inferences can be drawn from it with good critical results ; and finally, the surest source of all, we have the houses themselves, both those of this period and those of the next — for here, just as in the case of the documents, the examination and comparison of the later houses are necessary in the study of the scanty remains of the earlier work. Tradition claims that there were several houses which survived the Indian attack, whatever it was, of March, 1676. Tradition, of course, must be critically examined, and, even if it cannot be dis- proved, must be given only as tradition to which the reader must be allowed to give his own weight. It is curious to observe, how- ever, that one of the two points mainly urged against the statement that any house now exists in Providence built before King Philip's War, is also a tradition — one most tenaciously clung to by the 20 EARLY RHODE ISLAND HOUSES. older writers, that Providence was pretty well destroyed by the In- dians — a tradition which, with some of the poetic adorning it has received, contained no doubt some exaggeration. The second point brought against the early dates is the result of an investigation made about sixty years ago. At this time, when, in spite of the fact that men were then living whose grandfathers had seen the second generation of the settlers, almost nothing was known scien- tifically of colonial architecture, a number of antiquaries thoroughly examined, it is said, every house on the " Towne Street," and they reported that nothing remained of the ancient settlement' Their opinion, the best possible at that time, can, however, hardly be con- sidered as final. It must, like those expressed to-day, be constantly subject to revision, and we think that in the light of later evidence, overlooked by the older antiquaries, it must be revised. The documentary evidence necessary to critically test tradition is, in the earliest period, very meagre. The wills and the inven- tories attached to them begin at a time when alterations and ad- ditions had to a certain extent been made. They are still very valuable, but they cannot take the place of those which have been lost — no doubt irrevocably — in the missing First Book of Wills. Notices in letters are few. The records of the Plymouth and Massachusetts Bay Colonies are valuable as showing what our neighbors possessed, but allowance must be made to adjust these accounts to our own work. Our own records say nothing about buildings. The deeds are vague, the boundaries so indicated that it is now impossible in many cases to identify positively a tract of land." * H. C. Dorr, The Planting and Growth of Providence. *. Many deeds were not even recorded. THE HOUSES OF THE EARLIEST PERIOD. 21 I. The Roger Mowry House. When we turn to the existing house and remnant of a house which claim the long descent from the middle of the seventeenth century, we find that the houses of that date were, as far as these examples show, all. single-roomed, story-and-a-half structures with a huge stone chimney at one end. We have only two examples to appeal to, and of these only one is now standing. In the case of this one, however — the so-called Whipple or Abbott house on Abbott street near the North Burying Ground in Providence — tra- dition, the documents, and the testimony of the house itself seem to unite in the statement that it was built as early as 1653, per- haps earlier. It belonged undoubtedly to Roger Mowry, and as his tavern played a very prominent part in the affairs of the early colony.' We shall therefore refer to it hereafter as the " Roger Mowry house." This house as the visitor approaches it along Abbott street, up the hill from North Main, gives no impression of its age. From above it, looking back, we see the old stone chimney ( Plate 3 ), which though topped out with brick is almost exactly in its ancient condition, and which shows, on its sides, the shoulders or slopes which mark the position of the rafters of the original roof. It is inside the building, however, that its age can best be appreciated. The plan (Plate 4) indicates the difference between the old and the new portions of the house, which like all these old homesteads has been greatly altered. It originally consisted of the single "Fire Room" shown in black on the plan, which gives the additions in cross-hatching ; and in that room the original framing is almost ' The town council met there, and tradition says Williams held prayer meetings in it. 22 EARLY RHODE ISLAND HOUSES. intact. The four corner posts PP, QQ, the side girts BB, which connected each pair of them -at the level of the second floor, and the "summer"' or large middle beam A, which spanned the room lengthwise from the chirmiey girt C, connecting the two posts PP, to the end girt E, connecting QQ, are all in place ; and those of them which are not cased show their ancient chamfers with their mediaeval stops. The end girt E is cut out curiously under the end of the "summer," as is shown in the sketch on the restored section (Plate 5). This cutting away, which is quite common as an after- thought, is here probably original — a view which is favored by the holes for the pins which held the tenons of studs or posts at the sides. These posts could hardly have been put in after the sill and girt were in place, for it was the custom to tenon them at top and bottom. The framing of the girts into the corner posts is also shown in a sketch on the section ( Plate 5 ), while on the plan will be found a note of the manner in which the old sill pro- jected into the room. The present arrangement of the room would not lead the visitor to suspect the size or even the existence of the old stone fireplace. There is a fire-board behind the stove, and on each side of the fire-board a closet. Opening one of the closet doors, however, will reveal the stone cavern wherein, when the Town Council met, Roger Mowry burnt the logs of "this dales fireing," for which, and for the " house roome," we read the Town Treasurer was ordered, on January 27, 1657, to pay him one shilling and sixpence.' Some idea of the size of this ancient heating apparatus may be gained from the plan and from the section, which shows that it was nearly ' Derived from French sonu/iier, Latin sagmaritis, a pack-horse. ' The Early Records of the 7'own of Providence, Vol. L, p. no. THE HOUSES OF THE EARLIEST PERIOD. 23 as high as the underside of the chimney girt. It will be seen that one side of the chimney is out of doors, while the other did not, and does not now, reach the outer wall of the house. It was in this space between chimney and outer wall that the stairs, or the ladder which served instead, were placed. This is abundantly proved in other houses, some of which still retain a flight of stairs in that very location. Upstairs, in what the old inventories call the " Chamber," there is at present a large high room. None of this framing can be original above three feet or so from the floor where the original posts stop, as can be proved by sounding the casing of the posts as they show in the room. The shoulders, also, on the chimney, shown in Plate 3, and the positions of the old shelves or water- tables vS^'^' (Plate 5), which were made to project a couple of inches to prevent the rain-water from running down the chimney face into the house, leave no doubt of the original position of the rafters. Their evidence may always be relied on, whatever the position of the modern roof. The original house, then '( Plate 6), was no doubt such as the restoration shows. An examination of the perspective view of the framing of this type of house given in Plate 7 will help the reader to understand the more technically drawn plan and elevation. A little patience spent in studying these first figures will be of much value in the later chapters, as the names " sill," " plate," " summer," " side girt," " end girt," " chimney girt," " floor joist," " rafter," " collar beam," and " post " will constantly recur, and a glance at Plate 7 will ex- plain what they are, better than many words of definition. From this description, with a study of the drawings which ac- company it, the reader can form a clear idea of the original house of the settlers of Providence. With such houses as these, with 24 EARLY RHODE ISLAND HOUSES, their gables toward the street and their chimneys toward the hill, the old " Towne Street " was more or less thickly fringed." And of all the old dwellings, this veteran — this old tavern — an eye- witness of the town's history, a sharer in all its early struggles, almost an embodiment of its early life — alone remains. II. The Arthur Fenner House. The second house to which we refer — that of Captain Arthur Fenner — has been destroyed. It stood in Cranston, in the Pocas- set Valley, just south of Neutaconkanut Hill, near the present vil- lage of Thornton. The cellar — of later date than the original house — is still visible, and the masonry of it is excellent. The ruins, as they now exist,' lie lengthwise north-east and south-west. On the north-east is the cellar just spoken of — (see the plan) — next to it the debris of the chimney, pulled down about 1886; beyond that the decaying sill of a very old " leantoe " con- struction which was visible, in ruins, in 1883; and beyond this, in turn, at the extreme south-west of the group, a curious square de- pression, fringed with small trees. This arrangement will be made clear by reference to the illustrations. The view of the house' (Plate 8), taken before it fell into ruin, shows on the right hand the newer house which stood over the cellar, in the middle the stone chimney, with the remains of the roof of an older house about it, and on the left the old lean-to. In Plate 9 — a sketch made in 1884, when the lean-to had fallen and its beams lay on the floor in front of the chimney — the fireplace and the construc- ' H. C. Dorr, Planting and Growth, of Providence, p. 24. « 1894. ^The original photograph is owned by Mr. S, A. Hazard, a descendant of Arthur Fenner. THE HOUSES OF THE EARLIEST PERIOD. 25 tion of the stack are clearly shown. A measured plan of the ruins as they were in 1885, except that the plan of the cellar is given instead of that of the house over it, will be found in Plate 10. These three plates give all the data, except a few details, now ex- tant in regard to this curious old building. As one looks at Plate 8, the question arises — What are the dates of the different parts of the group } The lean-to is probably a remnant of the house which replaced Captain Fenner's original "house in the woods,"' built here about 1655,' and burnt during King Philip's War.' This rebuilding probably included the part of the house over the depression fringed with trees at the left of the picture in Plate 8. There are not wanting signs which make it look as if the lean-to was built of the remnants or was itself a remnant of the earliest building. The chimney belonged, part of it to the original house, part of it to the rebuilding, or to a later period still. The house at the right of Plate 8, over the newer cellar, was, according to accounts, a revolutionary structure of no interest. Whatever view, we may take of the history of the house, we must start, it seems to us, with one fact. The smaller fireplace in the chimney is the older and belonged to the original house, which was probably built by " old Mr. William Carpenter." It is narrow and deep and low, and has, over the opening, an oak beam i6i inches wide by 23 inches deep, beveled on the fire side, and on its lower corner, on the room side, adorned with the most elaborate mouldings in the colony. (See Plate 54). These mouldings, too, though of classic form, betray in their profiles mediaeval tradition ^Letters of Roger Williaiiis, Narr. Club Pub., Vol. VI., p. 374. "^ Early Records of the Town of Providence, Vol. II., p. 14. ^ Letters of Roger Williams, quoted above, p. 379. 26 EARLY RHODE ISLAND HOUSES. and die away at the end in a manner which can be matched in work of the Gothic period. They were no doubt the handiwork of William Carpenter, worked out literally by hand ; and the beam thus ornamented was built into the small chimney, probably about the size of that shown in black in Plate 13, about 1655. This, the original chimney, was, in later alterations, incorporated in the stack which appears in Plates 8 and 9. To this fact, which has only architectural proof, may, perhaps, be added another. It will be seen from Plate 1 1 that the location of the summer, given by measurements taken in 1883 when the second floor could be studied, is unusual. It is not in the centre of the frame, as is the case in every other house. It occurred to us to lay out a plan with the summer in the centre — leaving the summer in the same absolute location as given by the measure- ments, but supposing the original north side of the building to have been at the same distance from the summer as the actual south side was. The result, as shown in Plate 11, did not, of course, coincide with the lean-to plan, but it brought the chimney and fireplace — freed from the mass of masonry at the left side of it, which was added when the larger and newer fireplace was built — into the normal relation with the outside of the house. It also brought the width of the house, 16 feet 2 inches, into almost exact agreement with a curious fragment, in the handwriting of Captain Arthur Fenner himself, lately discovered among the family papers. This runs: "The house is six and thirtie foot longe and 16 foot wide and is 9 foot and od inches between joynts.'" If we consider the plan in black in Plate 1 1 as the original house, and assume, ' By this expression he probably meant the distance between sill and plate, the height, that is, of the posts, showing that he was speaking of a story-and-a-half house. This was just about the height of the posts of the Mowry house ( lo feet). THE HOUSES OF THE EARLIEST PERIOD. 27 what is very probable, that Fenner, before the Indian War, dug a cellar toward the west and lengthened his house, and that the western end of the depression gives the limit of this lengthening, we reach the " six and thirtie foot " of which the old Captain speaks as the length of some house.' Let us assume that this was the original length of the house, and that it was burnt in the Indian War. We have then to ex- plain why Fenner, when in rebuilding he widened his frame to the size given by the plan of the lean-to in Plate ii, still left the summer in its old position — a position which brought it out of centre in the new work. Perhaps enough of the original frame, including the summer, remained to lead him to add to the earlier construction instead of beginning anew. This explanation, based on economy — a powerful factor in all alterations, as architects well know — is probably the true solution of a question, which, with the ruins before us as they existed twelve years ago, could very likely have been satisfactorily answered. Fenner, then, according to our conjectural history, rebuilt his house after the war about four feet wider than it was originally. If we are asked why only four feet wider, we answer by asking why the house which stood on the eastern cellar was only four feet and eight inches wider than the lean-to and the depression. The building, then, was as long as the old one, with one chimney — the original one — at the eastern end. Whether there was ever another at the western extremity we do not know ; if there was, it has been gone many years. This lengthening of houses was no uncommon thing in colonial times. Some time after the war, per- haps-about 1685 or even 1700, possibly even after the old pioneer's ' Prov. Town Papers, No. 17,649. We admit that the document may have no relation to this house. 28 EARLY RHODE ISLAND HOUSES. death,' the cellar at the eastern end of the group was dug, a new house built, wider than the other, and the new and much larger fireplace constructed. It may be, of course, that the cellar is con- temporary with the newer house — said to date from about 1790 — and that the house which this replaced, that was built, according to our conjecture, from 1685-1703, had no cellar, or only a small one, and simply continued the lines of the lean-to frame. At any- rate, the chimney stack, as we see it, was built with the eastern fireplace against and over the original flue. The house must by the time of the Revolution have become quite ruinous ; for, as we see it in Plate 8, a new house has arisen over the eastern cellar." Later on, as we see, the part over the depression has utterly disappeared, and that between it and the chimney has been patched up in the makeshift which we know as the lean-to, and which the last inhabitants of the old house used as a kitchen. For the fact that this lean-to part was once a two- story house is proved by the mortise for the second -story summer which existed in the second-story chimney -girt carrying the gable which appears in Plates 8 and 9 above the lean-to roof. There are other hypotheses upon which the house could be re- stored and its history conjectured, but we shall leave them to the ingenuity of the reader. He has our data before him, and if our conjectures — which are, after all, in the case of this house only conjectures — are not satisfactory to him, and they are not alto- gether so to ourselves, he is welcome to try his skill on the most puzzling problem in the architectural history of the colony. ' October lo, 1703. ■■'Of course, it is possible that this house was the addition of 1685-1703 with a later roof — for, unfortunately, no inside measurements were ever taken, so far as we know ; and we cannot remem- ber whether it had a summer or not. THE HOUSES OF THE EARLIEST PERIOD. 29 If this must be the verdict on its architectural history, however, the civil record of this house is one of the clearest with which we have to deal. It has been in the possession of Arthur Fenner's descendants till within the memory of men not yet old ; and, but for neglect and deliberate destruction, the stronghold of the col- onial captain might still be a monument to his memory. ^"f CHAPTER III. THE HOUSES OF THE SECOND PERIOD, 1675-1700. HE Indian War marks, as we have said, an epoch in the history of the colony. There must have been an in- creased feeling of security. Then, too, the second and even the third generation was growing up ; the outlying allotments were being settled ; wealth was increasing, and the effect is soon visible in architecture. The house at first retains the same plan as before the war, but now it more often has two stories. Alter- ation and addition are no doubt liberally made, and a new form of house comes into fashion — that with the two fireplaces side by side in one enormous chimney at the end of the house. The number of houses of this period — 1675 -1700 — is quite large — larger probably than many are aware. We shall study five of the most typical examples, referring to the others only for par- ticular features. For a complete list as far as now known to us. Chapter IX may be consulted. The five houses to be treated are : I. The Thomas Fenner House, Cranston, 1677. II. The Edward Manton House, Manton, circa 1680. III. The Thomas Field House, Field's Point, c. 1694. IV. The Eleazer Whipple House, Lime Rock, c. 1677. V. The Eleazer Arnold House, Moshassuck, c. 1687. THE HOUSES OF THE SECOND PERIOD. 31 I. The Thomas Fenner House. According to the tradition of the Fenner family, the house which for years has been called the " Sam Joy " house was built by Captain Arthur Fenner for his son Thomas. The date found, it is said, upon the chimney during some repairs, seems to corrob- orate the statement. It is now painted in black letters on the whitewashed stone -work, " 1677," and its accuracy can hardly be questioned. It has not the character of an artificial date, and it marks just the time when the elder Fenner, rebuilding his own house, would provide for his son an establishment also, near by;' for this house is not a mile over the hill from the old " Fenner Castle." It has always been in the Fenner family, though not in the name, and is now possessed by Mr. S. A. Hazard, one of the descendants. It is still inhabited, as indeed are all save one or two of these old Providence dwellings. The house faces the east, with the chimney at the north — no doubt as a protection against the coldest winds — and is on a gentle slope, not far from a stream, though not so near as is the Arthur Fenner house, which is so close to Ocquockamaug Brook that the ancient palisade whereof tradition speaks must, if it ever existed, have enclosed the rivulet. In general, indeed, these old houses were very skilfully placed, aside from those in the " Towne Street," where there was often little to admire in the location of them. In the outer settlements they are always well situated on rising ground, near the inevitable brook ; near wood, of course, and 'The inventory of the estate of William Harris, Early Records of Providence, Vol. VI., p. 86, gives among debts due the testator, "of Clabord nayles lent to Capt: ffenner 1500, to be payd in nayles againe." The date of the inventory is 1681. 32 EARLY RHODE ISLAND HOUSES. often in places from which there is now at least, whatever might have been the case then, a fine view. A brook means both water and meadow -land where the settlers obtained "meadow hay." To return to the house. The plan of the first story (Plate 12) will show the original arrangement of the house, which is given in black. We have here the same heavy corner posts ; the same " summer " lengthwise of the room, which is about seventeen feet square ; and the same stone chimney, with a fireplace which was originally ten feet wide. The resemblance to the Mowry house is complete, except that this house was built with two stories. The chimney is in the same place, and one side of it — the western — was originally, like that in the house just mentioned, outside the wooden frame. In the second story we find the summer (Plate 13), which sup- ports the third floor, running across the house, at right angles to the direction of the summer carrying the floor below — an arrange- ment adopted to form a tie across the building at the foot of the rafters, which all have collar beams besides. The section will, with the plan, make clear the framing spoken of, as well as the peculiar cutting off of the three middle rafters just below the peak of roof, of which no one can now say whether it is the result of accident or of design.' There are two doors at the right of the fireplace, one of which is marked W in Plate 15. It will be noticed that one of these, as the plan shows, leads to the present pantry, and that the other (marked W) is now closed. Where did it lead.? In the floor of the pantry is shown a square patch which gives the answer to this ' This is not sliown on the perspective of the framing, which may be considered as a typical two-story house, while it follows the dimensions and form of this dwelling. THE HOUSES OF THE SECOND PERIOD. 33 question, for investigation in the cellar made it clear that this marked — not the repairing of a worn spot in the floor, but the filling of a trap -door. In the corner between the chimney founda- tion and the east wall of the cellar the framing changed. Though still of oak it was much lighter than that in the rest of the cellar, which was of 7x12 inch white oak beams laid flatwise, four inches apart. Now, this lighter framing was arranged for a trap-door which had been filled up — and filled up recently, for the timber used in it was not oak but spruce. Here, then, was the original trap -door to the cellar.' Again, in the second story, also in the corner between the chimney and the east wall of the house, and so just above the pantry, there is in the floor a series of oblong holes, now filled up. These were the holes for the balusters at the head of the old staircase (Plate 13), probably not the original one, but one which has long been done away. It was to this staircase that the long unused door led, while the other led to the trap- door and the ladder to the cellar. Neither of these doors, however, though they are of a very old form, nor the partition which con- tains them, are part of the original house. In the frontispiece we have a drawing of the house as it appears to-day (1895). The arrangement of the cellar makes it appear that the lean-to was not part of the original house. It was, perhaps, the earliest addition — a fact vouched for by the ancient windows, now boarded up, which still exist in its west wall. It is said that the windows now in the house took the place of these small ones throughout the building, except in the third story next the chimney. No doubt these sash were leaded — either with diamond panes or ' The present stairs to the cellar are new, and descend under those, also new, which go to the second story. 34 EARLY RHODE ISLAND HOUSES. with square lights Hke those in the sash of the Coddington house now in the Cabinet of the Rhode Island Historical Society. The brick chimney- top is also modern, but is probably panelled in a way somewhat similar to the original stone stack. We have so many examples left that it is easy to restore it, as we have done in Plate i8, which represents the house as it was built. The next addition after the lean-to was at the end away from the chimney. This, or its successor, became ruinous and was pulled down and replaced by the present southern end about 1837. The cellar is of good masonry, with small windows which may possibly be original. The curious thickening of the wall, when the new cellar at the south end was built, resulted from the filling -in of the ragged interstices at the back of the old foundation wall when this wall was uncovered by the new excavation.' In the chimney foundation and in the west wall of the cellar are niches, which were probably shelves. The steps from the new cellar to the old are formed, as the section shows, of old, finely chamfered oak beams — probably the summer, cut in two, of the old south end. The original fireplace, of which we give a restoration in Plate 16, is one of the largest which has come down to us. It extends from a point a little to the right of the present fireplace opening — K, in Plate 15 — to the point marked / on the left of the same drawing. The fireplace which there appears, with the fire- board taken down, is of brick ; and is built, as the plan ( Plate 1 2 ) will show, within the ancient stone opening. The framing of the house, which is the best example of a two- story house of its date now standing, and which we have had very ' In one house the ragged backs of such a stone wall are left visible in the newer cellar. THE HOUSES OF THE SECOND PERIOD. 35 good opportunities for measuring, is explained by the perspective view of the original frame in Plate i^ The corner braces appear | )") now only in the upper part of the house. A comparison of this drawing with the corresponding view of the framing of the Roger Mowry house ( Plate 7 ) will be instructive, and will show very clearly the change from the first period. II. The Edward Manton House. On the " Goddard Road," as it is called, just north of the Old Killingly Road, and a little way beyond the end of the Manton car track, stands the house which once belonged to Edward Manton — if not to his father, Shadrach, sometime " Towne Clarke " of early Providence. It is inhabited and in good repair. As will appear in the drawing (Plate 21) the house has now a long sloping roof toward the north. This is a later addition, as is proved by the existence of the old sill, projecting into the room, above the floor, on the north side of the original house. A little study of the figure will make this clear. We have, then, in this house, a survival in the second period of the type of the first ; for this house belonged to Edward Manton, the son of Shadrach, and was probably built at the time of his marriage, in 1680. It may, of course, be earlier; for — as a comparison of it with the Roger Mowry house will show — it has all the marks of the very early houses, and it may have been Shadrach Manton's " house in the woods," though of such a house we find no trace in his papers.' ' Shadrach Manton owned land in " Secessacutt," which was on the Woonasquatucket, near the site of this house, if it did not include it, as early as 1661. — Early Records of Providence, Vol. I., p. 91. 36 EARLY RHODE ISLAND HOUSES. We have, therefore, assumed the later date, though we assure our readers that the architectural evidence would place it earlier. The house, as it stands, is an excellent example — the least changed of any of the single- room, story -and -a-half houses that have come down to us. It has a few peculiarities. The projec- tion of the sill into the room — which we noticed in the Mowry house — occurs here; and, as it is confined to these two houses, it gives a strong impression in favor of an earlier date than 1680. The chimney is rather roughly built and shows signs of patching, though not where we should expect a patch, under the lean-to roof, which, as its pitch is very much flatter than that of the original roof, must have called for an addition to the chimney to fill up the space between it and the old rafter. This would make the lean-to a part of the original house but for the unanswerable evi- dence of the old sill. The addition to the chimney, then, was no doubt made more dexterously than the later patchings ; and, in- deed, as all the patching which can be seen is comparatively recent, the joints between the old and new work have, no doubt, been covered by some new pointing. A glance at the section ( Plate 2 1 ) which should be compared with the section of the Field house (Plate 25) will show that the low angle made by the collar beams with the lean-to rafters can hardly have been original.' On account of the low pitch of the lean-to roof and the near- ness of the northern eaves to the ground, we were enabled to make exact measurements of the chimney- top, and, further, to look over into the flues. There are two of these separated by a partition or "with," as it is called, of flat stone, about li inches thick, set on edge. The chirnney-top outside has a flat pilaster on each wide See also the Waterman house, Chapter 1 \^ THE HOUSES OF THE SECOND PERIOD. 37 face and none on the ends. This fact — that there are two flues, and that there was a fireplace in the second story — is the strongest evidence for the date 1680. The part of the house toward the road — that is, toward the east, for the chimney is at the west end — is a small addition of late date. The last few feet in length of the main house were also added, as the plan shows — though this was done quite early. The present dormer replaces an older one. Whether there was a dormer on the original roof it is impossible to say ; but, as they were common elsewhere before 1700, there is some possibility that there was. III. The Thomas Field House. Up to the autumn of 1894 a picturesque and ruinous old house stood on a knoll at the head of the "cove" at the north of Sassa- fras Point. This was once the house of Thomas Field, son of William Field, one of the early settlers of Providence. It lay east and west, and consisted of two houses — the newer of which, that at the western end, which is still standing, cannot, from the evi- dence of its construction, its end chimney of brick and its ancient framing with summer and girts, be later than 171 5. The eastern half was evidently older yet, and was probably built at the time of Thomas Field's marriage in 1694. It was one of the few examples remaining in the Plantations of a house built with a lean-to; for the section (Plate 25) shows that the lean-to here cannot have been an addition, but was an integral part of the framing. The old house, therefore, consisted of a "fire-room" which is the small- est of which we have record, and a lean-to at the north of it, with no cellar. The chimney, at the west end, was of stone up to the 38 EARLY RHODE ISLAND HOUSES. level of the second floor. Whether it was of stone originally above that point we cannot say, for there were the remains of a brick construction, with two flues, starting where the stone -work stopped.' A few years after the house was built its owner added at the east- ern end a room with a cellar under it. A study of the section (Plate 25) will show how the lean-to roof was managed. The pitches are the same or nearly so on both sides. To put the plates of the fire -room frame {P Q in the sec- tion) at the same level would bring the eaves of the lean-to — there was no cornice in the classic or later colonial sense in these older houses — lower than the top of the side -girt at the second floor, an arrangement not to be tolerated. The southern plate, then, was put at the level usual in a story-and-a-half house, and the northern one allowed to come above it. But, if this northern plate was allowed to come up high enough for the lean-to rafter to rest upon it, it would be higher than the old carpenter desired. He therefore compromised. He raised his frame with the north- ern plate high enough to cut one inch into the collar beams of the roof. Then he framed his roof trusses together on the ground, with the collars mortised and pinned into the rafter, and cut "gains" or notches one inch deep and the width of the plate Q in length, out of the under sides of the collars. To hoist each truss up till the collar beam rested on Q was the next step, after which he moved it along till the "gain" in the collar slipped over Q, and the feet of the rafters or " spars " as he probably called them, fell into the cuts made for them' or at the places marked for them in the plate P in the wall of the fire -room, and on the plate R ' It is probable that the brick-work of this cliimiiey was original, and was the earliest work of the kind in Providence. ''He called the collar beams "couples" very likely. THE HOUSES OF THE SECOND PERIOD. 39 in the wall of the lean-to, which was at the level of the side -girt B of the fire -room. As it stood just before its destruction, the old house, in spite of its gruesome surroundings and its tenantry of hens, was well worth a visit from anyone who was curious to see how our fore- fathers lived. Unlike any other house we know,' it had never been plastered either on walls or ceiling, and posts, girts, summer and floor joists were visible as clearly as they were in all the old houses before lathing came into fashion. When the newer house was built, about 17 15, as an addition to the old one, its brick chimney was backed up against that of the original house. This is proved by the rough mortar projecting from the joints on the back of the later chimney, which was not connected with the earlier one except in the first story, where an oven was opened from the back of the old fire-place. The later house has the same framing as the older one, though the sticks are smaller. It continued westward on the plan the lines of the first building, but it seems to have been a two story building from the start, though this is not certain, and it may possibly have at first continued the old lean-to, and have been built up later. At any rate the present roof is not original. This house is plastered, both stories, and the beams in it are cased with white pine boards I of an inch thick, as are those in the Thomas Fenner house and some others. This casing, and the plastering, are probably contemporary with the house, for lathing was in use at this date and even earlier, as is proved by the inventory of Benjamin Beers, taken July 5, 1714, which mentioned "an old lathing hammer."' ' Except the Arthur Fenner. '^ Early Records of Providence, Vol. VI., p. 260. 40 EARLY RHODE ISLAND HOUSES. IV. The Eleazer Whipple House. What is now best known as the "Ben Mowry" house, the home- stead built in 1677 by Eleazer Whipple/ stands in what was known as Louisquisset, on the edge of the present village of Lime Rock. The present dwelling is a long structure lying east and west, and the eastern part is quite new, probably not earlier than 1825. The roof is also entirely modern. The space in which the stairs now exist is an eastern extension of the original single room of which, like the Thomas Fenner house, the ancient dwelling consisted ; but in this case it looks as if it might have been part of the original building. The old or western half of the house consists really of two inde- pendent houses, each with its complete frame, summer and all, and independent fireplaces. The southern room is probably the orig- inal. The northern was built later against the old one, as the plan (Plate 27), and the small section drawn on it, will show. Here we see two girts side by side — a very interesting arrangement, recall- ing that soon to follow, if it did not already exist, where two rooms are built as parts of the same building with one girt between and no summer in the side room. The original house was probably almost exactly like the Thomas Fenner, with the summer running crosswise in the second story, and stone -topped chimney with only one fireplace. When the second house was added the chimney of the new fireplace was patched on to the older stone -work. The original stone top is now replaced by a brick chimney which prob- ably is, like the roof, comparatively new. 'He was a carpenter or " housewright. " — Austin, Geneal. Dictionary of Rhode Island, p. 222. THE HOUSES OF THE SECOND PERIOD. 41 The liking our forefathers had for fine views is nowhere better illustrated than in the position of this house. It stands on the western brow of the ridge which separates the Blackstone valley from the valley of the Moshassuck, and commands a wide prospect over the rolling country to the west and southwest. It faces the south, and was probably protected by the woods from the north winds which now have little mercy on its exposed situation. The peculiarity of this building — where as a later addition we find the type which a trifle later is characteristic of the second period — will appear in a stronger light when we have studied the next house which we have to consider. V. The Eleazer Arnold House. This picturesque dwelling — built in 1687 by Eleazer Arnold — stands on the old North road,' half a mile this side of the Butter- fly Factory, and about a mile west of Lonsdale. It is very well placed on rising ground, near a brook, and not far from the bank of the Moshassuck river. The house differs from any we have thus far studied. It was originally built, as the old slope of the chimney shows (Plate 28), with a lean-to like the Field house. But here the lean-to is not a sleeping -room or a mere store-room. The chimney, as can be seen from the perspective (Plate 28) and the plan (Plate 29), extended across the whole end of the house on the outside; the lean-to became the kitchen and had its own fireplace like its more aristocratic neighbor, the old fire -room. Here, then, we have the plan which is characteristic of the second period — that of two fireplaces side by side in two different rooms of the ' Lonsdale Avenue. 42 EARLY RHODE ISLAND HOUSES. same house. In this house the lengthening shown on the plan (Plate 29) at the end opposite the chimney was not an addition,' but was, like the lean-to, a part of the original building; and, a marked peculiarity in a Providence house, there was a gable on the side which now faces the road. Plate 32, which gives a restoration of the original building, will explain these statements, while the sections (Figures 30 and 31) will show the transition from the ancient framing to the construc- tion of the present roof. The house, which is the oldest example we have of a two -story house built with a lean-to, was originally framed, as the section shows, with the ordinary two -story construc- tion in the front room, while on the rear the framing stopped with the level of the new side -girt. The original side -girt of the Thomas Fenner house has here become a sort of second summer (Plate 30). The plate on the front carries the rafters of that side of the roof. The second story chimney girt, X, the second story end girt Y, the additional end girt Z and the second story sum- mer, which here, as in the Thomas Fenner house, runs across the house, are notched down upon the plate over the second summer (corresponding to the plate on that side in the Fenner house), and, projecting beyond, are tenoned into the rafters in the rear of the roof, which run down and frame into the lower plate (marked W in Plate 30). We have thus four trusses united by purlins," which are framed into the principal rafters P and Q. These trusses are all original, and the absence of any stud mortises in that over the girt which usually formed the end of the house, prove that the present length was that of the original building. ' The chimney in this addition is new. "The purlin is the horizontal beam framed between the trusses (see Plate 31) to carry the small rafters on which the roof boards are nailed. THE HOUSES OF THE SECOND PERIOD. 43 When the modern roof was put on, the rear of the house was built up and a new plate put on at the level of the third floor, as the section shows. The old tie beams were taken out and replaced by others spanning the second story rooms and supporting the old rafters, or were spliced so as to accomplish the same purpose, and the new rafters were sustained in the middle by struts from the old trusses. In the longitudinal section (Plate 31) can be seen two slanting beams cutting across the trusses, and interrupting the common rafters (T) supported by their purlins. These are the valley rafters of the old gable which once existed on the front of the house. The fact that the rafter on which they meet does not run down to the plate, and never did run down, is proof of this. We know it never ran down because the collar beams (Plate 31) run through to the roof and the rafter is tenoned into the collar, not the collar into the rafter. Further, the rafters (7", Plate 31) now filling" in the space between the two valleys are newer, and are nailed to them, a thing not dreamed of by the ancient carpenters. Finally, the mortises for the gable purlins still exist in the valley rafters — which are laid flatwise and halved into the truss rafters' — and, by the angle they make with the face of the valley, bear invincible testimony to their character, and to the existence of the gable. How common an occurrence this gable on the front of a house was we have no means of knowing. Though this is the only instance of it in Providence, we know from the work in Newport and in the other colonies that it cannot have been unfamiliar. ' Note the distinction in the section (Plate 31) between the principal rafter, which is part of the truss, and the common rafter which the truss carries by means of the purlins. The truss consists of these principal rafters, — one on each side — the collar beam which is a tie, and of the tie beam formed by the summer in the attic floor. 44 EARLY RHODE ISLAND HOUSES. The dwelling-house in the Providence Plantations has now passed through several steps, which may be roughly indicated as follows: First, the single -roomed, story -and -a -half structure with one fireplace; second, the two -story house with one room on each floor, still with a single fireplace'; third, the story- and -a -half house with a lean-to, but with only one fireplace; fourth, the story-and- a-half house with the lean-to and two fireplaces, like the building we have just considered. Though these steps do not follow each other chronologically — for all these types appear together in the second period — they still show progress toward a larger dwelling. In the third period we shall find houses of two full stories, a form which, as we saw, occurs in the Eleazer Whipple house, in the shape of an addition to a single -roomed, two -story house. ' During the first period all lean-tos were probably additions. CHAPTER IV. THE HOUSES OF THE THIRD PERIOD, 1700-1725. HE houses of the third period do not differ in plan from those of the second, nor do they always differ in con- struction. The characteristics of the period are the pre- dominance of two -story houses and the increased use of brick, of which, finally, all the chimneys are built, while they retain exactly the same forms as those of earlier times. The date of the first use of brick in Providence is difficult to fix. William White, bricklayer, was in the town as early as 1665,' but, in a deed of 1671,' he is expressly named as "Of Boston," so he did not remain. The earliest mention of bricks of which we are aware is found in the inventory of Epenetus Olney, Senior, who died in 1698.' These brick were probably made near the house, though possibly they came from Taunton or Nayatt, where the clay pits may have been worked since very early times. The brick in the chimney of the Greene house at Buttonwoods, built about 17 15, are said to be of clay dug and burnt on the shore of the salt water cove on which the dwelling stands. Bricks were probably made, then, in Providence, as early as 1690, or thereabouts ; though the ^ Early Records of Providence, Vol. III., p. 72. ' The same. Vol. IV., p. 9. ^This inventory, allowed July 12, i6g8, mentions "A percell of sawne bords," and "A percell of Brickes." 46 EARLY RHODE ISLAND HOUSES. first actual proof of the existence of the industry is the grant by the Town Council, in 1725,' of the right to "dig clay at Waybaus- sett Hill for to make bricks." At any rate, whatever the date of their introduction, the use of brick became general in this period, and the stone chimney, though it lingered very late in the South County, disappeared from Providence. In this period we shall discuss four houses : I. The Epenetus Olney House, North Providence, c. 1700-05. n. The Benjamin Waterman House, Johnston, c. 1700. HI. The John Crawford House, Providence, c. 1715. IV. The James Greene House, Buttonwoods, c. 1715. I. The Epenetus Olney House. This stands on the bank of the Woonasquatucket river between Allendale and Lymansville. It is untenanted and is rapidly going to pieces. It has always been and still is (1895) in the Olney family, and was no doubt built by Epenetus Olney, the second of the name, probably about 1700 or 1705. The plan and sections will show the arrangement of this house, which is like that of the Arnold house except that here there are two full stories and no lean-to. The perspective of the framing (Plate 37), without the roof which is not old, will show, with the details in Plates 56 and 57, Chapter VII., the scheme of the build- ing. It is easier to study the frame of this house than that of any other, because of its very dilapidation, which renders the work accessible and enables it to be studied at leisure. 'Records quoted by H. C. Dorr, " The Planting and Growth of Providence" pp. 130-31. THE HOUSES OF THE THIRD PERIOD. 47 The chimney at the west end of the house, toward the river, is partly of stone and partly of brick. The stone work, which rises to the level of the plates, is the best of its kind now stand- ing in Rhode Island, a worthy example of a lost art, a proof that some apprentice of old John Smith or his son — by the way there was relationship between the Olneys and the Smiths — had bettered his old master's instruction. Above the level of the top of the plate the chimney was built of brick, one of the earliest instances of the use of this material in Providence.' The bricks (Plate 33) are very artistically handled; the course immediately above the stone -work consists of headers. Above, for some distance, perhaps originally for the whole gable, the courses are alternately of stretchers and of headers, which, in their turn, are alternately red and dark grey blue, making a very artistic arrangement. The three fireplaces in the house, two on the first floor of stone, and one on the second of brick, required three flues here as in the Arnold house, but in this chimney the third flue falls back into the main stack before the roof is reached, and instead of a T- shaped top we have, or had as the chimney was originally built, a stack with three pilasters on the wider faces and none on the nar- rower, an arrangement which can be traced in the chimney top as rebuilt in its present form, and which we have shown in Plate 37. The brick gable has been altered to suit the change of pitch, when the new roof was .put on the old house, to match that of the new house which was added at the east, and the windows at each side of the chimney probably date from this change. It will be interesting to compare this house with that of Thomas Fenner, as we can do by means of the cross sections and the per- ' As has been said, the chimney in the Field house is probably an earlier instance. 48 EARLY RHODE ISLAND HOUSES. spective views of the framing of each building. In both houses the stairs were in the same place, at the side of the chimney, and in this house one entrance to the old part still leads to a passage- way which, with a modern pantry, fills up that corner. In the second story, however, though there is now a closet over the first story pantry, the step-ladder, or ver}'^ steep rough stair leading to the garret, is or was still in place. It will be noticed that this house is exactly like the Thomas Fenner house' with what we may call the half of another house added to it at the side. One of the side -girts is the same in both stories, so is the summer. The other side -girt of the Fenner house is now, as in the Field and Arnold houses, a kind of second summer, carrying the ends not only of the floor joists over the main room, but of those over the side room as well. There is here, however, a new side -girt and a new plate in the outer wall, and the rafters now span the whole house with their pitch equal on both sides." This house, unlike most if not all the others of its period, was not lengthened in early times at the end opposite the chimney. The new house which was added at that end was built about 1812, with money earned, it is said, by boarding the workmen who built the dam and mill just above at Allendale. It has under it a cel- lar connecting with that which extends under the eastern half of the older building.' In this building we have reached the final form of the early houses. From now to the abandonment of the heavy summer and ' Except that the arrangement of the second story summer in the Fenner house is different, and that the relation of stairs to chimney is reversed. * Note the letters on the posts in Plate 37. They refer to details in Plates 56 and 57. This occurs elsewhere only in the old Sayles house, on Westquadomeset or Sayles Hill. ' Some settlement in the foundations has curved the floor of this house, as is shown in the sec- tion (Plate 36). THE HOUSES OF THE THIRD PERIOD. 49 its small floor joists we have the same system as that here before us. A glance at the dimensions of this house, one of the largest which has survived, and a comparison of them with those of the Roger Mowry, or even of the Edward Manton house, will show the changes which in sixty years the increase of w^ealth has wrought in the colony. II. The Benjamin Waterman House. In Plate 39 we give a plan of the so-called "Nick Waterman" house, which stands in Johnston on the cross-road between Hughes- dale and the Hartford Pike. It is in the valley of the Pocasset, perhaps two miles above Captain Arthur Fenner's Castle, and under the lee of the hill which lies parallel with Neutaconkanut, about two miles from the latter. It was probably built by Benja- min Waterman, about 1 700. In this plan, as in the others, the original work is in full black, while the later additions and changes are only cross-hatched. It will be seen, then, that the stone portion at the west on the plan (shown also in Plate 38) is later than the rest of the house, which, as the section will show, was built like the Field house, with a lean-to. Indeed the resemblance of this house to the Field house is very marked, as a comparison of the drawings of the two will prove. The workmanship, however, of this house is better, and it never had any western extension. Its chimney, like that of the Field house, is of stone up to the level of the second story. Above that it is of brick, and shows the same blue headers which we met in the Olney house. This artistic use of these brick, which are very large, about four inches by nine, and two and a half inches thick, shows that this was built as an outside chimney, and that 50 EARLY RHODE ISLAND HOUSES. the tradition as to the later date of the western addition is to be relied on. Only the side walls of the first story have been plastered in this house, and the summer and its joists still show in the room. The side -girts, which have not been cased, are cut into over the window heads, as in the Mowry house above its ancient doorway ( Plate 5 ). Here, however, the arrangement is probably an after- thought. In the second story the floor is laid in two thicknesses. There is sheathing instead of plaster in the attic, and the rafters (Plate 55) and the collar beams, which appear in the room, are framed as no other work is framed that is now standing in Rhode Island. The ruined lean-to of the Fenner Castle alone could show anything equal to it. They are planed and chamfered, and their joints are close and true even now. Although the plans and sections of this house show that it is a survival in this period of an older scheme, that it is quite unlike the Olney house, and, as we have just said, very much like the Field homestead, it yet has a striking peculiarity which distin- guishes it from the earlier mansion. The use of brick in the chimney does not make the difference, for the upper part of the Field chimney was very likely original. But in the second story (Plates 40 and 41, also Plate 55), at the north side of the chim- ney, the side opposite to the old staircase, which was no doubt where the passage or entry is into which the outside door now opens, are what remain of the studs of the old gable, for here the chimney does not fill the whole space up to the rafters on this side. There is nothing unusual in this use of studding, which was the regular way of filling the old Providence gables. What is unusual about this example of it, however, what is in fact unique in this part of the colony, is that the clapboards are nailed to the THE HOUSES OF THE THIRD PERIOD. 51 studs without any intermediate boarding. Again, the walls in the first story, which are plastered, are not lathed on the outside boarding, but on studs which run from sill to girt, and which probably, as the measurements seem to show, have the clapboards nailed directly to them, as in the gable. That is to say, we have here an abandonment of the regular Providence system of vertical boarding, and the adoption of the studs covered with clapboards, which were in fashion in Connecticut. Why an example of this construction should occur here, where everything else shows so little trace of Massachusetts or Connecticut influence, and so much stubborn individuality, was to us a riddle until we examined the Greene house, at Buttonwoods, which we shall describe a little later. Ill, The John Crawford House. On the west side of North Main street, the old "Towne Street" of Providence, near the corner of Mill street, stands the house which Gideon Crawford probably built for his son John when the latter was married, in 17 15. The level of the street has been raised so that the door from the sidewalk opens into the second story. On the side of the house toward the river its original height can be seen. The plan of this building is almost identical, except for the po- sition of the original stairs at the right instead of the left of the chimney, with that of the Olney house. Here, however, as in the Greene house, the chimney is entirely of brick, and it is the oldest example on the old "Towne Street" of the use of that material.' The bricks were probably made in Providence. They are about 2 ' Neither the Field nor the Olney house was in the "Towne Street." 52 EARLY RHODE ISLAND HOUSES. inches thick, 4 inches wide and 8 inches long, and are laid in alternate courses of headers and stretchers. No blue headers ap- pear.' The chimney ends in a triple stack which recalls old Eng- lish work. The course of headers as dentils under the string below the separate flues can be very closely matched in the old country.' The old house is interesting for more than its chimney or its framing. It is a landmark, because we can date it very closely, and it is the last Providence house in which we shall see the ancient summer, with its chamfered edge and its small clean-cut floor joists. We know this because we have another landmark, the brick house in the Butler Hospital grounds, at the turn of the road, near Swan Point. This was built in 1730.' Its chimney is at one end, but the two fireplaces have been brought together, so that the main room and that which was in the Crawford house a smaller one at the side are equal, and the fireplaces are in the corners, while the old summer has disappeared. It still exists in the ceiling, but it is no deeper than the floor joists, which are now larger than of old, and it is plastered over flush with them. Head room has been gained, and the new fashion of plastering has made the room look more "elegant" perhaps. No doubt it has made it warmer. The old deep beam, the mark of the seventeenth century, the honored sign of colonial date, has been improved away, and with its dis- appearance, which occurs between the dates of these two houses, the last period of early colonial architecture in Rhode Island comes to an end.' ' The whole chimney except the top has been painted, so that there may have been blue headers below the roof lines. '' There is a stack at Tenbury, in England, which is almost exactly like the one under discus- sion, except that it has two flues instead of three. ^According to documents in the possession of Mr. S. S. Rider, of Providence. See Chapter L of this book. * Note the letters at the posts on the plan, they refer to details in Plate 58. THE HOUSES OF THE THIRD PERIOD. 53 IV. The James Greene House. This, the best known house in the State, stands on a little stream which runs into the head of Brush Neck Cove, in Button- woods. The top of the chimney and the rather low pitched roof, which can be seen in Plate 47, are new. The rest of the house was built almost exactly as we see it today, somewhere about 1 71 5.' The wood -shed against the chimney, and the lean-to on the north, are the only additions the house has received. The plan of this dwelling is like that of the Olney house, ex- cept that the chimney is wholly of brick. The original stairs too exist, and are in the corner diagonally opposite to that in which we generally expect them (see Plate 48). That they are original is proved by the narrow stone stairway descending to the cellar under them. It is on this cellar stairway that can best be seen the studs of the outer wall, with the filling of brick' between them. This, then, like the Waterman house, has studded walls; and here, too, where the old clapboarding remains in the north- east corner, as we can see in the second story of the lean-to, the clapboards are nailed directly to the studs. Where we can see the inside of this old clapboarded portion of the wall, on the stairs at the third floor level, we find it simply daubed with plaster, without any brick filling. The east end of the house is at present boarded outside of the brick filling, but this boarding was put on at the same time as the new clapboards with which the whole house, ex- cept where hidden by the lean-tos, seems to be covered. ' It seems to us that the date assigned by tradition to this house, 1687, really applies to the earlier one, of which the site is still pointed out, and that this building cannot be earlier than the date we give. There can be little doubt that it was built not by James Greene, but by Fones Greene his grandson. ■^ Some of these bricks are crumbly, and appear almost as if sun-dried, though of darker eolor. 54 EARLY RHODE ISLAND HOUSES. Now none of this work resembles the general type which pre- vailed in Providence,' but it does, as far as the stud system goes, resemble some of the work in South County, which in turn was copied from originals at Newport. Again, the stone stairs to the cellar are unknown in Providence so far as we have observed, but not rare in South County and Newport, while the sawed balusters over the door -heads are exactly like those found in the two southern colonies. Notice further the fireplaces on the plan (Plate 48), and observe the straight sides, not splayed at all as are all those we have heretofore seen, and the rounded corners. These rounded corners occur in the Spencer house,' in Newport, and the coincidence is striking to say the least. If we consider all these Newport ear -marks," it will be hard to resist the conclusion that some Newport craftsmen wrought this house.* The studs of the Waterman house, though we can so far only suspect the existence of brick in them, seem also to point to the hand of an Aquidneck carpenter. ' This house was in Gorton's purchase of Warwick. The plan is of the Providence type. * See Chapter V. ^ The walls of some of the houses in Newport are no doubt filled with brick. * The old homestead has never been out of the Greene family. The present owner, Mr. Henry W. Greene, courteously shows the house to visitors. He has preserved the main fireplace intact, with the hangers used before cranes were made. CHAPTER V. NEWPORT. JS we have said already, the early colonial architecture of Newport differs considerably from that of Providence. The type of house with the end chimney, which, as we have seen, prevailed to the exclusion of any other in the northern colony, does appear in the Island settlement ; but it had to dispute supremacy with the central chimney type which belongs to Con- necticut, and, if both styles have survived in the same proportion, the latter must have been victorious. There are at least three houses in Newport of early date with end chimneys. There are more than double that number with central chimneys which can claim to be ancient. Although King Philip's War passed by the Island, so that we are not prevented by traditions of burning from carrying the date of a house far back into the seventeenth century, yet unfortunately, for some reason or other, the numerous old houses in Newport are of so late a period that it is more difficult even than in Providence to say of what sort the earliest houses were. The late George C. Mason in his work, '' Reminiscences of Newport," gives a view of 'Page 138. The date is given by H. C. Dorr, " The Planting and Growth of Providence" p. 27, as 1650. This view is taken apparently from the same source as that in Palfrey's ^' New Eng- land."' 56 EARLY RHODE ISLAND HOUSES. the Governor Coddington house, now destroyed, which stood on Marlborough street. If the date assigned to this house — 1641 — is authentic, we might reason from it that the end chimney type to which it belongs was at least as early as the other. If it is so early, then the other might reasonably be considered as the later type, and as the result of the increase of wealth brought about by the trade of the port. Against this view might be brought what is said to be the oldest house in the State, the Governor Henry Bull house on Spring street,' for which the date is given as 1 638-1 640. But it is almost certain that very little of that house as it stands goes back to the date assumed, if, indeed, any of it does. Tradition, indeed, asserts that the southern end of the building, that at the top of the plan (which we give in Plate 49), is the older. This statement is borne out by the fact that the summer runs, Salem- fashion, across the room, as the plan will show. It will be noticed, however, that this summer does not reach the outer wall on the east, but is framed into a girt several feet away from that wall. That means either that the eastern wall at the south end of the building was original, or that there was on the line of that girt an earlier stone wall which was taken down to enlarge the house, or that the wall on that side was once of wood. We hesitate be- tween the first view and the last, though we think that the Gov- ernor Carr house, which till recently stood on Conanicut Island, would bear out the last opinion.'' If we adopt the last view, we must say that the whole eastern and northern stone wall of the house is late, and so is the wooden ' Number 20, near the northern end of the street. ■^ See Mason, quoted above, page 407. The house was ruinous and was recently rebuilt. It is nothing like its original form. NEWPORT. 57 part of the west wall, and the stone part of this nearly to the door. The chimney in the room where the two summers cross each other — a late and uncommon arrangement — is probably quite modern, though this room is no doubt older than the eastern wall; and almost certainly the other chimney, though quite old, is not the original one, which was of stone, and, as in the Governor Carr house, formed part if not all of the north end of the house. The present gambrel roof replaced an older roof of the same form, which itself replaced the original sharp -pitched covering. If we hold to the first opinion, we still assume a stone chim- ney with a strip of wall between it and the eastern side, and this wall may have been either of stone or of wood. Any attempt, however, to restore this house is rendered very difificult and unsatis- factory by the changes which must certainly have been made, though they are almost impossible to trace. If, then, the end -chimney type was what the early settlers of Newport brought with them, it survived well down into the early eighteenth century, as did its counterpart in Providence. On the corner of Marlborough and Duke streets stands a house with an end-chimney of stone. It is late in date, perhaps 1670-80, and the roof on the side toward Marlborough street has been raised. Other- wise it greatly resembles the Arnold house, at Moshassuck. An- other house on Duke street is also of this type, though, as the chimney is of brick, it is probably later than the other. Further up Marlborough — one has to go a little down Branch street to see the house well — is a dwelling with a fine pilastered brick chimney at one end. The main interest of Newport, however, is in the fact, which we have often reiterated, that very many of its old houses, built towards the end of the seventeenth century or early in the eigh- 58 EARLY RHODE ISLAND HOUSES. teenth, have a central chimney. There is one house — the Sueton Grant — with a stone chimney, all the others, and there are several, have chimneys of brick. The Sueton Grant house stands in Hammett's court, about a hundred feet back from Thames street. It was once owned by the famous Newport merchant whose name it now bears, though he can hardly have built it, for he came hither from Scotland in 1725, and the house is at least fifty years older than that date, and possibly more. We have selected it for special study as a type of the imported Newport house, because it is probably the oldest of that type now standing, and because, unlike any other house now standing in the State, so far as known to us, it retains the over- hang. The house originally consisted, as the blacked part of the plan (Plate 50) shows, of two rooms, one on each side of a large stone chimney, within which the stairs are built, and at the back of which was the fireplace of the lean-to kitchen. The summers in both stories were lengthwise of the house, from the end -girt, that is, to the chimney -girt. The outside walls are studded, no doubt, and thus thickened to make room for shutters in some of the rooms." The arrangement of the stairs is peculiar. The two piers at each side of the flight come together above the stairs in the sec- ond story. Under the stairs in the first story are the steps to the cellar, which is very interesting, for the foundation of the huge chimney instead of being a square mass of stone, is cut into 'on three sides by deep recesses, which are arched over to support the masonry above them. The recess on the fourth side is occupied ' The walls are very likely filled with brick. , NEWPORT. 59 by the stairs from the first story. We more than suspect that these arches were turned by the same mason who built the Old Stone Mill. This arrangement occurs outside of Newport only, so far as we know, in the Lippitt house in old Warwick, of which the chimney is of brick. In the garret the principal rafters of the old trusses remain in the end walls, and in the side toward the street, showing that the lower pitch of the gambrel, on Hammett's court, is original. The patches made by filling up the roof, when the original gables on the front were removed, are also plainly to be seen. As the res- toration shows, the original roof was a high gambrel, with a very narrow and flat upper roof. The authority for this restoration is the relation of the original rafters to the old water-tables on the chimney. This was, no doubt, one of the oldest gambrels. How common the overhang was in ancient Newport, we have now no means of knowing. The Governor Coddington house had one, according to the published drawing' of it, which seems to have been a copy of a sketch made before the house was pulled down in 1835. An old cut of the Paine house on Conanicut Island, long since destroyed, gives that an overhang. It was very common — probably almost universal — in Massachusetts and Connecticut, as old prints, and old examples still standing, testify.' Very likely it was also the prevailing method of building in Newport. Whether it was or not, this single instance in the Sueton Grant house is all that has actually come down to us. In this case the over- hang seems to have existed only on the front and only in the second story. In Connecticut and in Massachusetts, overhangs in 'First published in Palfrey's ''History of New England,'' Vol. II.. p. 62. ' These overhangs were a tradition of Mediaeval and Elizabethan England. It is hardly neces- sary to say they had nothing to do with Indian fighting. 60 EARLY RHODE ISLAND HOUSES. gables exist in the same house with that on the front. In Provi- dence, too, the overhanging gable is quite common — but it is late, and did not exist, so far as can be ascertained, in the ancient houses. Of the other dwellings, the Spencer house, on the west side of Thames street, just north of Marlborough, is interesting for its fine chimney, which still shows above the present gambrel the marks of the original steep roof. The fireplace in the north room — which has two summers crossing each other in the ceiling — has the rounded corners which we noticed in the Greene house at Button- woods. In the cellar the chimney is carried on two piers with an arch, or rather a tunnel -vault, between them. The house is of a late date — probably as far down as 1720-25. The Wanton house, on West Broadway, is probably of the same time. It has as fine a chimney, and retains the sharp pitched roof on the side toward the street. A little later, perhaps, even than these is the Arnold house {}) on Hammett's wharf, which is probably about the last example in Newport of the use of the summer. Here the logic of the carpenter led to an interesting result. He saw that the beams had been altogether too heavy for the work required of them. So, as he had two rooms side by side on each side of the chimney, sep- arated by a partition under the second summer, as in the Eleazer Arnold house in Providence, he did not think it necessary to put a post under this second summer, but framed it into the end -girt exactly as he did the main summer. In the west end of the house, therefore, which was the only one which could be examined on the first floor, there are only two posts — those at the corners of the building. CHAPTER VI. NARRAGANSETT. HE name Narragansett applied in general to the southern mainland of the State below the present town of Coven- try. It was the country of the tribe from which it took its name; but — because the white inhabitants were at first only a handful — to go to Narragansett meant to go to Richard Smith's trading post near the present town of Wickford. One of the oldest roads in the State, the Narragansett trail, or Pequot path, as it was called, runs from Providence through Pon- tiac, Apponaug, East Greenwich and Belleville, a mile or so west of Wickford, down along Tower Hill, and then, making a great curve to the westward, skirts the Atlantic shore and passes on in- to Westerly, Stonington and New London. It is along this path, the " Old Post Road," that we find the greater part of the inter- esting houses of South County. It was on this ancient highway, close to the shore of the cove just north of the present harbor of Wickford that Richard Smith put up " in the thickets of the bar- barians, the first English house amongst them." ' Smith seems to have used his house at Cocumscussuc — as the land he had purchased was called — only as a trading post until 1659. About this original house we know almost nothing, though '^Letters of Roger Williams, Narr. Club Pub., Vol. VI., p. 399. 62 EARLY RHODE ISLAND HOUSES. traditions not few have been handed down about it. The timber is said to have been floated from Taunton, where Smith Hved for a time. The site of it is said to have been a few rods southwest of the present house and partly over Cocumscussuc brook, which runs through the farm. Again, it is said to have been a block house; and this may very likely have been true. Certain it is that it served as a rendezvous for the colonial army during the campaign which ended in the Swamp Fight. To this day the spot where were buried the soldiers killed in that bloody action is shown to the visitor. This small plot of ground is covered with grass which the cattle will not touch, strangely enough, as it is the famous " blue grass " of Kentucky. The infuriated Indians burned the trading houses soon after their defeat in the Swamp ; and, in 1680, Richard Smith, the Younger, constructed the present building partly from the materials of the old "garrison." We give in Plate 52 a plan of this house. As will be seen at once, it is almost exactly like the Sueton Grant house in Newport. The walls, which in the first story at least are quite thick, are no doubt filled with brick. We did not discover whether the chimney was of stone below the attic floor. Above that floor it is of brick, but is probably new, for the whole roof is later than the original house. There were gables on the front of the house, it is said, but there seems to have been no overhang. In each of the large rooms of the house there are, as the plan will show, two summers crossing each other at right angles. The summer which carries the joists is that which runs from the end- girt to the chimney -girt, parallel, that is, to the front of the house. The other was probably put in for ornament, as ■ it, with the first, divided the ceiling into four large squares. This arrangement is poor constructively, as the beams have either to be halved together NARRAGANSETT. 63 in the centre, or tenoned — the first into the second. As all the beams are cased, this cross summer might be considered a mere built-up affair of thin pine boards; but we meet the same arrange- ment at Newport, in the Spencer house, where there can be no doubt of the genuineness of the beams. This last example, also, by the bolt — the head of which appears below the intersection of the summers — gives convincing testimony to the danger of this kind of framing. No symptoms of such trouble, however, occur in the Smith house. One of these large rooms must have been the "hall" mentioned in the inventory of Richard Smith, Junior, in 1692. This docu- ment specifies the goods contained in the following buildings and rooms : " warehouse, shop, kitchen in great house, store house cham- ber, hall, dairy room, kitchen chamber, porch chamber, hall cham- ber, lean-to chamber, etc., etc."* Of these, the kitchen, hall, kitchen chamber, porch chamber, hall chamber, lean-to chamber, and possibly the dairy room, belong in this house. The evidence of the cellar wall, which shows very plainly where the new excavations which have been made have ex- posed the back of the old foundation wall, has been relied on for the statement of the drawing that the room at the north-east is an addition. Under what we have assumed to be the original wall of the house on the north-west there is now no wall in the cellar and no signs of any jointing in the chimney foundation. A large beam spans, in the cellar, the distance from the outer stone wall to the chimney, and, as the post shown in the northwest wall stands upon this stick, it seemed to us that it was the original sill of the house, and that the room on the northwest was also an ad- ' Austin, Genealogical Dictionary of Rhode Island, p. 185. The inventory is recorded in Bos- ton Probate Office, Suffolk, XIII., 29. Op Dyck Genealogy, p. 82. 64 EARLY RHODE ISLAND HOUSES. dition. It may not be, however; and, again, both it and the room in the northeast may have been added before Smith's death in 1691 ; at any rate, it is very difficult to make the inventory fit the house, to which there can be no doubt it refers. The cellar is quite interesting. Some of the stones in the walls of it are very large bowlders split in two, with clear division sur- faces. The foundation of the chimney, as the plan will show, is extremely large. The original steps to the cellar, under the front stairs, are of stone as in the Grant house, Newport, and in some other houses in Narragansett. This house, the oldest in the South County, shows the influence of Newport in the plainest manner. It is a house of the Connecticut fashion, like most of those now standing in the Aquidneck colony. We shall meet many like it in the Narragansett country, though with none so large nor so fine. Nor have any other houses, so far as we know, the crossed sum- mers which mark this house as one of a class rare in all New England. Opposite the Smith house, on the other side of the Post Road, stands a very old house with a stone chimney, a veritable stone - end house, it seems, of the Providence type. It is known as the Palmer Northup house. Further deponent saith not. Were it not that we should be promptly confronted with the fire-brands of King Philip's War we should say that this was the very house which Roger Williams deeded to Richard Smith when he sold to him his trading post and his two big guns, or "murderers," to ob- tain money wherewith to go to England after the charter. About a mile further north, on the same side of the Post Road as this house, that is, the west side, stands an old chimney which once formed the outside of a small house belonging to some family of Browns. NARRAGANSETT. 65 In Belleville, still on the Post Road, stands the Phillips house, known as " Mowbra Castle." It was probably built about 1695- 1700 by Michael Phillips, who came from Newport. Its plan somewhat resembles that of the Arnold house at Moshassuck, but the chimney is nearly square, and the fireplace in the side room is at an angle of ninety degrees with that in the main room. On the Post Road, below Wakefield, toward the road to Matu- nuc, stands the Watson or Congdon house, with a stone chimney in the centre. It may date back to 1690-1700; it is framed of cedar, and the summer runs Massachusetts fashion, that is, across the room, parallel with the chimney-girt. This is the case, also, with the Robert Hazard house in Charlestown, near the Champlin farm. Here there either is no .fireplace in the room which con- tains the summer, or it has been filled in with stone. The date of this house is perhaps 171 5. Its chimney is of stone up to the roof. The General Stanton house consists of three parts, a Connec- ticut type house with centre chimney, and with a summer in each room, an eastern addition to the length of the house, also contain- ing a summer, and a gambrel roofed ell likewise with a summer. This last is probably the oldest portion of all. The Welcome Hoxsie house, with a centre chimney, is also interesting, as is the Church house with its stone chimney at the end of the main room. This chimney, however, is not really an end chimney, for it has a fireplace at the back of that in the main room with the summer, in a one story lean-to. All these houses are on the Post Road except the last, which is on the road further toward the shore, parallel with the Post Road below Perryville. The only stone -end houses in all South County, so far as we have explored or have received trustworthy reports, appear to be 66 EARLY RHODE ISLAND HOUSES. the two near Wickford, the Palmer Northup house and the Brown house, of which only the chimney remains. Most of the old chim- neys are stone, but they are in the centre of the house, or where, as in many cases they are at the end, they are covered by a wall of boarding. Near East Greenwich are two remarkable houses which show strong Newport influence. They are quite late and somewhat peculiar, having the summer in one end and not in the other, and a stone chimney in the centre. These are the Coggeshall house, c. 1715-20, on the Post Road, a mile and a half below the village, and the Payne house, c. 17 10-15, somewhat nearer the village, on a road parallel to the Post Road, but further west. The Payne house is unique in having been widened towards the front instead of by a lean-to in the rear. The Coggeshall house has stone stairs to the cellar, and the sawed balusters which are characteristic of Newport. The string of the stairs has a huge cove moulding, almost exactly like that in the Sueton Grant' and the Spencer houses. Down in the Stanton Purchase, between West Kingston and Shannock, in Richmond, is the Stanton house, a good centre- chimney house, abandoned and going to decay. It exhibits a mixture of rudeness and elegance, which is one of the marks of the Narragansett country. Poverty, especially in these inland town- ships, is stamped plainly upon the life of the early settlers by the appearance of their dwellings. Here and there a house rises out of it, especially along the Post Road, which after 1715 must have been lined with dwellings about a mile apart, for nearly its whole length. ' The string is probably not original in this house. NARRAGANSETT. 67 The Carmichael house at Shannock is a late building, about 1715-20, with a stone chimney at the end of a single room, though the stone -work does not appear on the outside. The summer and beams are cased with the upper edge of the casing boards left rough where they were to be concealed by plaster, which was never applied. There does not seem to be much if any old work in Westerly and its northern neighbors, though that whole district waits further exploration. -Smith Ga£ri3QM Ha-cocunscu-sisuc CHAPTER VII. CONSTRUCTION. ET us now examine more in detail the construction of the houses which we have described in the foregoing chapters. In so doing we shall take up the principal parts of the building, which hitherto we have seen in relation to the whole, and discuss them separately. The old craftsmen's solu- tions of the problems before them were generally so simple and logical that they can hardly fail to be interesting even to the un- technical reader. I. Stone-work and Brick-work. Stone: — The bowlders and fragments of bowlders scattered over the soil of the colony furnished the early masons with a plentiful supply of this material. These field stones seem very often to have had at least one naturally flat smooth face, for many of the granite or gneiss bowlders show to this day lines of cleavage akin to those along which they themselves were split off from the original ledges. A kind of sandstone also occurs with good faces, apparently natu- ral. These stones had long ago deposited their water of crystalli- zation, or "sap," and thus had acquired a sort of outer crust or skin which resisted fire splendidly, something which modern quarried CONSTRUCTION. 69 granite will not always do. It is certain, therefore, that in the fire- places, at least, the old masons used the field stones as they found them. In Providence they selected the stone quite carefully, while in South County they do not seem always to have wanted a flat face, but to have been content in some instances with round stones. We do not know how much quarrying the colonists did in Rhode Island. They may have used somewhat the out -cropping ledges of granite, but it is probable that they contented themselves mostly with breaking up the bowlders at hand when they could not find flat -faced fragments. For the fact that many walls are built of split stone is almost incontestible. Some of the surfaces are marvellous, and show that the old craftsmen knew accurately the cleavage lines of the stone. They seem to have possessed only the heavy mason's hammer,' for none of the faces, so far as we have observed, show any marks of the drill or of the chisel, yet the stones in the Arthur Fenner cellar can hardly have been picked up in their present condition on the surface of the ground, and look like blocks quarried along the cleavage lines. Often we find stones which show an irregular fracture, as if they were broken by a heavy blow. Sometimes we find bowlders of conglomerate, "pud- ding stone," split as smooth as if sawn apart, and yet with no sign of a tool -mark. Foundations : — These, where there were no cellars, were of field stone laid on the ground, without, probably, any trenching, since all traces of them have in some cases disappeared. Where there is a cellar the work is also of field stone, sometimes used just as it was picked up, sometimes split, as we have said above, and generally, if not always, laid in mortar. • The inventory of Thomas Olney, Senior, presented Oct. 17, 1682, mentions "a stone hammer, or small sledge." Early Records, Vol. VI., p. 94. 70 EARLY RHODE ISLAND HOUSES. The Chimney was built of stone until about 1700. The best stones were used in the fire -room and on the outside of the build- ing. Small stones were used for filling, and many rough stones were used under cover. Sometimes part of the inside was laid dry, as in the Olney house, but in most cases mortar is used in- side and out.' The fireplaces are very large, 9 or 10 feet in width and from 2 ^ to 3 feet in depth, with a splay of about 6 inches on each side. The height varies a little, but was probably very nearly the height of the room. In the Field house the under side of the wooden beam, 13 x 14 inches, over the fireplace opening, is level with the bottom of the beam which crosses the room in front of the fire- place. The original fireplace in the Fenner house was much lower. All the beams are of oak, with a bevel on the back which continues the slope of the flue.' The hearths were slabs of stone, chosen for their flatness. There were andirons at that time, but we have never encoun- tered any — except, perhaps, in the Greene house, Buttonwoods — which seemed to go back to those early days. Bars across the flues, with " trammels " as in the Greene house, served instead of the crane of later times.' The old houses had fireplaces in one story only. The second period shows us fireplaces on the second floor. ' Some of the bowlders or split stone used in the fireplaces may have been of gneiss, which is a good fire stone. ' This bevel on the back of the beam across the fireplace opening was always cut before the stick was put in place, and was never the result of burning, though the blackening by years of ex- posure to smoke may deceive some observers. ^ There is a very large crane in the garret of the Eleazer Arnold house, which evidently belonged in one of the old fireplaces, but its date has not been ascertained. In the Greene house at Button- woods the fireplace in the main room is as it always was. The andirons here are no doubt also original. CONSTRUCTION. 71 It is in the cap of the chimney that the old mason showed whatever artistic ability he possessed. The chimney starts at the first floor level as a rectangle nearly as long as the width of the house. On the outside this length is maintained till the stone- work strikes the underside of the roof, the pitch of which it then follows. On the inside the chimney narrows as it rises, leaving the outside as a kind of wing on each side, about i6 inches thick, and finally goes through the roof as a much smaller rectangle, perhaps 3 feet by 4 feet, or in the T- shape which we find in the Arnold house. Just above the point at which the stone- work clears the shingles, on the sides toward the slope of the roof, a thin course of stone projects like a shelf about 2 inches to prevent the rain from fol- lowing down the chimney into the house. It serves instead of our flashing of lead, a metal which the colonists probably did not pos- sess in abundance for such a use — since bullets were current as money. Above this projection the chimney rises plain for 2 feet or so, when another projection of about the same size goes quite around it. From this string course to the top, most of the old chimneys are ornamented with pilasters (Plate 53), one on the end and two on the sides, or one on the side and none on the end. In the Arthur Fenner chimney (Plate 9) there were three on the side and one on the end. The caps of these picturesque stacks, which well deserve modern imitation, were made of several projecting courses of flat stone, which imitate fairly well the effect of the mouldings of the old English chimney (see Plate 53). One chimney at least, that of the old Fenner castle just mentioned, which was a beauti- ful stack, the best of them all, had a necking around it just under the cap. This adds immensely to the character of the work — as 72 EARLY RHODE ISLAND HOUSES. both necking and cap courses follow the projections and recesses formed by the pilasters. The oldest chimneys, as in the Mowry house and in the house opposite the Smith Garrison at Wickford on the Pequot path, are plain, and as they have lost their tops we cannot say just how they were finished — probably by one or more thin courses. The brick chimneys were almost all finished with pilasters, like their stone predecessors, as in the Epenetus Olney house, (Plate 37), and in the Smith house — now the Gushing — in Wans- kuck. These are the only brick -end houses known in Providence with pilastered chimneys ; but such a treatment must have been the fashion, for the Tillinghast house on South Main' street, the house at the corner of South Main and Sovereign streets, and that num- bered 295 on North Main street — all later than 1730 — have pilas- ters. There does not seem to be a single pilastered brick chimney in Warwick or in Narragansett, while in Newport there are several fine specimens — though no house with an outside brick chimney occurs. The Grawford house stack in Providence, already spoken of, is the only one of its kind — if, indeed, there were any others — which has been preserved. This curious arrangement (Plate 43) is evidently, like the old stone and brick pilasters, a survival of Eng- lish traditions. On the whole the old stone mason showed no falling off in his work up to 1700. The Olney house, as has been said, will rank as good masonry anywhere. We use the word "masonry" here in its old sense — for in those times a mason was a man who built in stone," and, as the records show, working in brick was a separate 'North and South Main streets are the old " Towne Street" of Providence. '^ This is still the English sense of the word. CONSTRUCTION. 73 trade, and those who followed it were called, not masons but brick- layers/ Mortar: — The earliest mortar of the colony is what is called shell mortar. Perhaps the best known specimen of masonry built with this material is the stone mill at Newport, built by Governor Benedict Arnold somewhere about 1670. This mortar is identical with that of the Bull house' and of some other buildings. It is composed of "pulverized shells, clay, sharp sand, and fine gravel." This sort of cementing material was used in the other settlements of the colony and seems to have lingered until quite late. At the Greene house at Buttonwoods several lumps of it, quite hard, with very large fragments of shell, and with seemingly strong traces of clay, are shown as what they no doubt are — fragments of the orig- inal house, built probably about 1687. The chimney of the Arthur Fenner house was built with shell mortar, which also occurs on the top of the old cellar wall under a sill which has now disappeared. The whole subject of this mortar, however, is far more intri- cate and puzzling than it might seem. Just what the process of mixing the ingredients was, and just how each one acted, no one seems to know. We do know that it was a good hard mortar, and that it resisted the weather splendidly. Mr. Uriah Cummings has discussed the subject in an article in the Brickbuilder, to which we have already referred. He seems to be uncertain as to whether the shells were burnt or not, and it can be inferred from what he says that he thinks that after the wet clay had been added to the pounded shells the whole mixture must have been burnt. We are ^ Early Records, Vol. IV., p. g. ''J. P. MacLean in American Antiquarian, quoted by Uriah Cummin s, Brickbuilder, July. 1895, Vol. IV., p. 151. ''Cummings, in article quoted above. 74 EARLY RHODE ISLAND HOUSES. sure that the shells in the mortar of the Arthur Fenner house have been calcined. The probability is, therefore, that the shells were pounded up, mixed with wet clay, rolled into balls and burnt, and then either slaked in the ordinary way, as is done in the manufacture of hydraulic lime,' or pulverized, much in the way Portland cement is made of chalk and clay. The powder was mixed with sand and water, much as we make cement mortar. The result in either case was a mortar with hydraulic properties, that is, a mortar which will, in a greater or less time, set under water, and which withstood very well the action of the elements. As this hydraulic quality is given to ordinary, or "fat" lime, as opposed to "meager," or naturally hydraulic lime, by the addition of pounded bricks, forge scales or foundry slag, or even cinders, it is possible that our forefathers burned the shells, pulverized them, and then added the fragments to the fat lime which they possessed. For the Providence settlers probably were not long in discerning the rich deposits of limestone which are spread over the north- eastern part of the State. We do not, however, find any recorded evidence on the subject till 1661, when the town voted to Thomas Hackelton liberty to burn lime on the common." Again, in 1665, it was ordered "that those Lime Rockes about Hackelton's lime Killne shal be Perpetually Common."' The Scoakequanocsett re- ferred to as the location of the kiln was no doubt in the southern border of the village of Lime Rock. It can hardly have been the Sockanosset near the Pawtuxet river, as is claimed, for there is no limestone there, so far as we have heard or seen. It is of lime from this ledge that the mortar in the Arnold ' Baker, Treatise on Masonry Construction, p. 51, "^ Early Records of the Town of Providence, Vol. IIL, p. 8. ^ The same, p. 66. See also pp. 22g and 241. CONSTRUCTION. 75 house is made, and some of it, in the chimney in the garret, is beautifully white and very hard. Just when and how far this lime superseded the old shell mortar is a question for further study. II. The Frame. The Summer is the beam which crosses the main room, the "Fire Room" or "Hall," from the end -girt to the chimney -girt (see A in Plates 4 and 5; see also Plates 7 and t^j). The word is itself a relic of the Middle Ages, it is derived from the Norman French "sommier," and finally goes back to the low Latin word "sagmarius," a pack-horse. Its name is well applied, for it does carry half the second floor, the other half of course resting on the side -girts. It is not so deep by two or three inches as the girts into which it is framed (see A, Plate 5), and is nearly 12 inches square. Its edges, as well as those of all the exposed framing, in most of the houses, are chamfered (see Plates 54, 55, 56, 57, 58). The joists which support the second floor and the third floor, are framed into the summer. They are about 3x4 inches, set with their depth vertical, as they are now -a -days, planed quite smooth and not chamfered. They are framed into the large beam in various ways. A (in Plate 54) is the method employed in the old Fenner house, while in Plate 56 is given that used in the Olney house. The dovetail is used to form a tie, and prevent the stick from being pulled out. Both methods are very good. There is no case in Providence of a summer in the first story running parallel with the beam before the chimney which we have called the chimney -girt,' though the arrangement occurs in the second story in the Thomas Fenner, the Arnold, and the Whipple house. ' This does occur in South County. It is also the rule in Salem, Mass. In Connecticut the summer runs as in Rhode Island (see Plate i). 76 EARLY RHODE ISLAND HOUSES. The summer as in the Arthur Fenner house, and probabl}'^ in all others in Providence, was framed into the chimney-girt with a finely cut tusk -and -tenon joint, the old oak still showing the marks of the scratch -awl used by the carpenter in marking out the mor- tise. This joint, now nearly abandoned, was in use till quite lately, and is still shown in old books on framing. The Girts or girders were framed between the posts in the outer walls and across the end in front of the chimney, at the level of the second floor, the tops of the girts coming flush with the tops of the floor joists, which are framed into those on . the side of the building. See Plates 7, 17, and 37. The end and chimney-girts are generally about 8 x 16 inches. The side-girts are smaller, generally about 6 or 7 x ir or 12 inches, so that, as their tops are in the same level, they do not come so far down into the room. They are chamfered like the summer, and are all set edgewise. In Plates 57 and 58 is shown the way in which the end and chimney -girts at the third floor level are framed into the posts which carry them. Plate 57 shows the joint between the post X and the plate, and in the same Plate at L we have the intermediate post L, marked G on the plan of the Olney house ( Plate 34, and in the first story on Plates 35 and 37), showing the way the two girts at the third floor level were carried on the single post. It will be noticed that the chimney- girt in the smaller room at the side is not so deep as that in the main room.' In Plate 58, B shows the same arrangement in the Crawford house. The way the side -girts and the plates were framed into the posts is seen in Plates 56 and 57. The Plates give cross-references, so that the Olney and the Crawford house ' In the Arnold house the chimney -girt does not exist over the side fireplace in the first story. Nor does it in the Crawford. It was parallel with the joists over the side rooms, and not needed. CONSTRUCTION. 77 can be compared. By each post in one house is given its letter in the other. The Sills are generally 8x8, though sometimes much larger, 8 X 12, 12 X 13, in the old Fenner house. They are laid on a low underpinning of stone, where there is no cellar. The posts are framed into them at the corners. The Posts are sometimes of the same size throughout their length, but they often, in the second story, and in story- and -a -half houses in the first, have a projection to receive the end and chim- ney-girts, as is indicated in Plates 57 and 58. This is a mediaeval contrivance, which occurs in old French work. It probably lingered long in the colony, especially in barns. The habit of casing posts drove it out of use in houses about the same time that the sum- mer was abandoned, 1720, The Plates stand in the same relation to the framing of the third floor, when there is one, as do the side -girts to that of the second floor. In addition to the framing of the third floor, and where there is no third floor, the plates perform their original work of carrying the roof. They are generally a little less in size than the girts, and, like the side -girts at the second floor level, are smaller than the third floor end -girt or than the third floor chim- ney-girt, which in two story houses spans the house in front of the chimney (see Plates 17 and 37). The plate in the corner where the stairs are placed, and that at the other side of the chimney are smaller. The Rafters are generally 4x6 inches, placed, in spite of modern ideas as to the wastefulness of the old carpenters, with their depth vertical as it should be, and are spaced about 4 feet on centres. The Collar Beams are 3x4 inches, and are pinned into the rafters. In the Waterman house the rafters and the collar 78 EARLY RHODE ISLAND HOUSES. joists are beautifully planed and chamfered. The feet of the older rafters, or "spars," are cut on the plate in at least three different ways. In Plate 59 the upper left hand drawing shows that in use in the Olney house, where though the old roof has gone, the notches on the plate, combined with the arrangement observed in the Thomas Fenner house, enable us to restore the original scheme. In the upper right hand drawing, in the same Plate, is the arrange- ment used in the Field house. Immediately under these two we give the excellent scheme found in the Greene house, at Button- woods. It will readily be seen that this form of rafter -foot gave no cor- nice such as appeared in the first half of the eighteenth century — that is, such a cornice, or "jet" as carpenters call it, as we are ac- customed to see on a wooden house. It may be of interest to notice how this cornice was managed when it was brought into use. The problem was to sustain, at a projection which in the earliest ex- amples was about a foot or something over, a combination of thin boards built up into the shape of the stone cornice of one or the other of the Classic Orders. This meant that the rafter itself must overhang the plate. We get over this difficulty now -a -days easily enough by notching out the rafter and nailing it to the plate as in the lower left hand drawing in Plate 59. But the old tradition of rafter- footing was too strong for this ; everything must be framed. Nailing did not occur to the old workmen as a means of holding timbers together. So in the Phillips house at Wickford the cornice of the new roof is arranged as is shown in Plate 59. The new tie-beams are halved into the new plate at the top of the posts, and, projecting beyond it, carry the real plate, that on which the rafters are footed. Small struts — exactly like those used on Medi- aeval roofs — are set on the first plate to divide the weight between CONSTRUCTION. 79 the two. Around the second, or projecting plate, the cornice was built. A later form occurs in the Olney house, where the roof is new, and this also is drawn in Plate 59. Square holes are cut through the plate at intervals, and in these holes are inserted stubs, which carry the cornice built around their ends. Boarding : — The sides of the early houses are all, save the Waterman and Greene houses, covered with oak boarding an inch, or a little over, in thickness, nailed vertically from the sill to the plate. This style of boarding for the sides of buildings, or at least of houses, was never abandoned in Rhode Island in colonial times. The stud system, in which the space between sill and girt and be- tween girt and plate was filled with vertical studding on which the boarding or clapboarding was nailed horizontally, seems never to have been used in Providence except in the gables,' where the con- stant employment of it in the old houses shows that it was well understood. In Newport and Narragansett there are houses with studs, though vertical boarding is used in Newport and is the rule in Narragansett. The roofs and gables are always boarded horizontally with oak. The boarding was generally — though perhaps not always — pro- tected with Clapboards or Shingles. The clapboards were often, no doubt, in the earliest work, a foot wide or so, with one edge shaved down a little, put on like the boards of a clinker-built boat. The shingles were probably three feet long, and an inch thick at ' Note that the Waterman and Greene houses — see Chapter IV — are exceptions to the statement. The Waterman clapboards are of the ordinary size — about the same, that is, as those now in use — (see Plate 55). The inventory of Capt. Arthur Fenner, who died in 1703, mentions : " Sum Cedar Clabbords & shingles" valued at i6s. — Early Records of Providence, Vol. VI., p. 233. In the same volume, p. 86, William Harris' inventory — 1681 — mentions " Clabord nayles " and " board nailes." 80 EARLY RHODE ISLAND HOUSES. the butt, though neither shingles nor clapboards of that date seem to exist' It is a question of what the original Floors were made. They were probably either of oak, tongued and grooved, or of white pine like the floor of the Thomas Fenner house, which may not be orig- inal, but which is quite old. Some of its boards are of astonishing width. The floor of the garret in the Eleazer Arnold house seems to be of oak. That of the garret of the Greene house is certainly oak. The under- floor in the Smith house at Wickford is of pine. The Chamfers with their stops, which exist on all the old tim- bers, vary through the forms given in the drawings of the houses. In Plate 54 we give the group of mouldings on the edge of the fireplace beam in the Fenner Castle. In the Cole -Greene house in Warwick is a very elaborate chamfer, consisting of a cyma with fillets above and below. The chamfer stops show a strong Mediae- val tradition. In fact the chamfer itself is a survival of the Middle Age, and seems to have nothing in common with classic ideas. The Windows were very small, partly because of poverty, for they did vary somewhat according to the wealth of the owner and the security of the place he had chosen for his house. Governor Coddington's house, for instance, had sash i foot yi inches by 2 feet 3I inches. Thomas Fenner's sash, with which Coddington's may be contemporary — though perhaps a good deal earlier- — are only 13^ X 2oi inches.' Almost all of the sash were filled with diamond panes, set in lead " calmes." The rectangular panes of Coddington's sash may have been an uncommon form, foreshadow- ing the panes set in wood of the later time. So far as we know ' If Coddington's sash are original and the traditional date of his house is correct, the sash must be one of the oldest specimens in existence. CONSTRUCTION. 81 this is the only example of its kind in the colony which has come down to us. In Plate 54 is shown the sash of the Arthur Fen- ner house, now preserved in the Rhode Island Historical Society.' Glass was no doubt used very early. Balusters were not used till late. The stairs were protected by a hand-rail, with either a second rail below it as in the Stanton house, Richmond, or with the space below it filled in solid with a wide board as in the Greene house, Buttonwoods. Coddington's house had balusters, and from their form they could easily be of early date.' When balusters which we can feel at all sure about come in — and all those before 1725 are in Newport and Narragan- sett — they are shaped like a long old-fashioned / set vertically, and are sawed out of thin stock. The early balusters in Provi- dence are sawed out in the same way, but are of the regular clas- sical baluster outline. This sawed work is found in England. The Overhang. In Plate 60 is given the detail of the over- hang in the Sueton Grant house. By overhang we mean espec- ially an overhanging second story, not considering projecting gables, whether in one or two story houses. The girt runs over the top of the first story post, and the end is finished with a tenon which goes into a mortise in the lower end of the second story post. That this was the method of framing is proved by the drawing of the overhang in the Roger Williams house, in Salem, where the drop has been cut away. 'A sash, filled with diamond panes set in lead, now in the Pilgrim Hall at Plymouth, is i6|^ by 22 J^ inches. ''Two of these are preserved — one by the Rhode Island Historical Society, the other by the Historical Society at Newport. The date of his house is given as 1642. It can neither be verified nor disputed, as the records are lost. The architectural evidences favor it. See Chapter V. 82 EARLY RHODE ISLAND HOUSES. The Partitions are of matched and beaded boards, with mould- ings on the edges of every other board and bevels on the inter- mediates, or else bevels on both boards. They are probably not to be found earlier than 1675. In the Sueton Grant and in the Greene house we have examples of the first kind described. In the Spencer house, East Greenwich, and in the Olney house, we have the second kind. CHAPTER VIII. RELATION OF COLONIAL ARCHITECTURE TO ENGLISH WORK. UR forefathers in Rhode Island were mostly tradespeople. They were carpenters, masons, glovers, surveyors, tanners, printers, and so on, from the cities or the country towns of England. A goodly proportion of them had some knack at preaching, but they were nearly all the descendants of the Mediae- val craftsmen. And it is just the word " Mediaeval " which marks the character of the traditions which our old carpenters and masons held. They were of the class whose ideas change slowly. They had not been greatly affected by the Renaissance; of the classical work of Inigo Jones they knew little or nothing, even the earlier Elizabethans like John Thorpe probably had not influenced them much. They had learned their trade of masters as slow to change as themselves, and these masters had been trained in the Mediae- val fashion. The Mediaeval architecture of England reached its zenith in the Early English work of the thirteenth century. It changed as the centuries advanced until it decayed, died out, and was finally en- tirely superseded by the Classical of the Renaissance. This sup- planting of the native Gothic by the style imported from Italy was a long process, and, as our forefathers left England before it was 84 EARLY RHODE ISLAND HOUSES. finished, and had learned their trades, those of them who were craftsmen, under men imbued with what was left of the Mediaeval spirit, we may be pardoned for examining the Gothic a little closely. What is called Perpendicular Gothic prevailed in England from 1399 down through the reign of Henry VIII., who died in 1546. It was in the beginning of the sixteenth century, about 1509, that the foreign, Italian work first made its appearance in the land. It first obtained a foothold in church sepulchral monuments. Torre- giano, an Italian, finished in 15 16 the tomb of John Young, Master of the Rolls. In 15 16 the same artist, who spoke of "those beasts of English," completed the altar tomb of Henry V^II. and Elizabeth of York, in the Lady Chapel of Westminster Abbey.' From monuments the new movement spread to mantels in houses, and from those and other details it began to affect the houses themselves. Gothic was slowly dying, or these inroads would not have been possible. But, as we have before remarked, it died a lingering death. At first the architects adopted what they thought was best in the Italian manner, and retained what they thought was best in the Gothic. Work on Mediaeval lines was still done in churches, houses and colleges, especially in altera- tions and additions, far into the seventeenth century. Even when the forms of the new style gained ground it was mainly as forms that they did so. The irregularity and the freedom of the Mediae- val scheme of planning were very slowly abandoned, and the Mediaeval methods of work must have persisted with them. It seems to have been nearly the same in France. Renaissance work is, in the beginning of the Revival, grafted upon Gothic forms, and ' Better known as Henry the Seventh's Chapel. W, J. Loftie, Inigo Jones and Wren. RELATION OF COLONIAL ARCHITECTURE TO ENGLISH WORK. 85 supplants them only gradually. The details change first, the plan and elevation next, the method of work probably last of all. For the new style was essentially aristocratic, it appealed to men of culture and refinement, as those words are now artificially under- stood. It was with the Renaissance that the architect began to draw away from the craftsman. He led the way into the new style and the workman followed him, losing his individuality along with his old fashioned notions about building. In some parts of England Gothic lingered longer than in others. One of these was Oxford, where Inigo Jones built for Archbishop Laud the garden front of St. John's College, which, in form, is still a good design. The Chapel of University College was conse- crated in 1665; it much resembles the staircase and entrance gate- way of Christ's College, built possibly by Jones, in [640. In the neighborhood of Bath Abbey, which was finished with- out change of style in 16 16, Gothic persisted strongly. It was possibly the character of the building stone of this part of Eng- land, Wiltshire,' which helped perpetuate the old fashion. At any rate there are several examples here which have the Mediaeval spirit, though built very late. At Corsham is a beautiful little almshouse with a mixture of square and round headed windows, and with Mediaeval gables and chimney, which was built in 1668.' Most if not all old English houses not built of stone, both in the cities and in the country, were of what is called "half-tim- bered" work. This is familiar to almost all from the pictures of old English or Continental streets, for this manner of building was especially common in cities all over northern Europe. The quaint ' William Carpenter, one of the old carpenters of Providence came from Amesbury, in Wilt- shire. ' W. J. Loftie, Inigo Jones and Wren. 86 EARLY RHODE ISLAND HOUSES. fronts with their overhanging stories, latticed windows and carved beams, their framing showing in dark contrast to the white plaster of their panelling, which was often marked with curious patterns, are known far and wide through the drawings of Herbert Railton and the sketches of many an artist in the old countries.' These houses were constructed with posts and studs, the space between which was filled in two ways. According to one fashion, called "wattle and daub," laths were nailed on strips between the studs. These laths were covered on both sides with clay mixed with chopped straw, and the wall thus formed, about five inches thick, was finished on both sides with plaster. The old English plasterers had ways of working their material so that it withstood water. According to the other way, the space between the studs was filled with bricks, and plaster was applied to these. Now the original colonial houses in Massachusetts Bay, in Con- necticut, and, to some extent, in Rhode Island, were many of them, if not all, lined with brick between the studs of their outer walls, as has been already stated, and, further, the clapboards are, in many cases at least, nailed to the studs over the brick, without any boarding. What does this mean? It means one of two things. Either this is a custom brought from the old country, and it does not seem to have been very common there, or it originated here under the following circumstances. The original builders proposed to use the old English half-timbered construction with which they were familiar. They used either the "wattle and daub" or the brick filling between the studs of their houses. To this day can ' The illustration to Besant's London, and other such books or magazine articles, are excellent sources of information on this Old English domestic work. Railton's drawings, many of them are in Coaching Days and Coaching Ways. Rimmer's Ancient Streets and Homesteads of England is very good. RELATION OF COLONIAL ARCHITECTURE TO ENGLISH WORK. 87 be seen, in the garret of the Roger WilHams house in Salem, what seems to be the top of the clay and chopped straw filling of the ancient outer wall on the rear of the house. The outer sur- face of the brick or clay wall of our forefathers, whether that in the Williams house belongs to that primitive time or not, was plastered between the studs, which were left exposed. A few New England summers and winters brought about un- looked-for results. The upright studs shrunk sidewise under the fierce sun, and left along their whole height cracks through which the searching winds of the old-fashioned winter carried their streams of snow. The remedy was found in a sheathing of feather- edged boards overlapping each other, and hence we see the clap- boards nailed directly to the studs. This system lingered nearly a hundred years, whatever may be the truth of this theory as to its origin.' In the old territory of the Providence Plantations, however, this system, as we have often said, did not prevail in early times. The Providence house, and a great number — perhaps the greater number — of the ancient houses in the other settlements of the State are boarded vertically. Now this system is not very com- mon in the old houses in the other colonies. It occurs in the Cobbett house in Ipswich and in some others,' but these cases seem to be exceptions, as the Waterman house is in Providence. Whence came this totally different system ? Why was it used ? The first question can be positively answered only after much ex- ' Some time ago Mr. E. R. Willson, Architect, of Providence, a native of Salem, suggested to us that the old houses of that city were half-timbered, and were built with the bays (many-casement windows) used in Elizabethan houses in England, and that the climate compelled the settlers to abandon these in favor of the smaller sash. ^Br. I. W. Lyon. 88 EARLY RHODE ISLAND HOUSES. ploration in England and on the Continent. It may be English imported by William Carpenter ; it may be Dutch imported by John Clawson. To the second question we are inclined to answer — because of its cheapness. It required no studs. A building with posts at its corners only cannot — if its dimensions be even twelve feet either way — be boarded horizontally with so good results as those attained if it be boarded vertically, for the distance from sill to plate is never more than seven and a half feet at the most — seldom so much. Economy had to be carefully studied in early Providence ; and the pioneers — after finding, before they left Massachusetts, that brick filling required a covering of horizontal boards — did away, first with the bricks and then with the studs which carried these horizontal boards, and which filled in the space between the posts. Really the posts and girts were amply strong without the studs ; and the result has justified the calculations of the framers, except that possibly the houses in Providence show more sagging, twist- ing and other deformation than do their contemporaries in the other colonies, for the studs and the horizontal boarding certainly would help to stiffen the frame. It will thus be clear how closely our ancestors clung to the traditions of their trades as they were practiced in the England of their time, and it will be clear, too, that our claim that these tra- ditions are in their case strongly Mediaeval is well sustained by the examples. These men were good workmen, economical in most cases, skilful in handling their material, though we are too apt to consider them wasteful and clumsy. They were artistic, too, for they solved the problem before them in the simplest manner, with logical use of the material which they had at hand, and with good arrangement of line and mass. Some of the quaintness and charm RELATION OF COLONIAL ARCHITECTURE TO ENGLISH WORK. 89 which belong to their work is of course the effect of the grouping of later additions and of what someone has called the reclaiming touch of nature, softening hard lines and giving undulation and ease to surfaces which must at first have been much nearer geo- metrical planes — but not much nearer, for from the scanty means they had for handling heavy timber, from their rough methods of " raising " a house, where the whole side was put together on the ground and pushed up or " raised " into place by the combined ef- forts of many men and considerable rum, there must have been a certain ease in the lines of the building. Then their pilastered chimney-caps, with moulded tops and irregular sky-line, their sharp- pitched roofs, small latticed windows and large wall surfaces, must have had a pleasant effect even when the work was new. They were simple ; and simplicity, as we are just beginning to see, is the cardinal virtue in architecture. Inside the house the "sanded floor, the blackened fireplace with its volume of roaring and writh- ing fire, the summer and the joists, the posts and girts all frankly shown, all beautifully planed and chamfered with all the care the ancient craftsmen could bestow, must have had a fine effect. They impress us now, when whitewash and plaster and new fireplaces, fire -boards and modern stoves, carpets and wall paper, have done their best to destroy the ancient scheme ; but in the old days, when the original color of the oak, contrasted with the pine flooring above, the white sand below, and the gray stone of the fireplace the effect must have been artistic and extremely effective ; — none the less so because it arose from a frank and simple meeting of the wants of those for whom the house was built. CHAPTER IX. A LIST OF OLD RHODE ISLAND HOUSES. The First Period. The Henry Bull House. Newport, Spring street. Almost wholly of stone. Much altered. Traditional date, 1638. See Chapter V. The Roger Mowry House. Providence, Abbott street. Date c. 1653. See Chapter H. The Arthur Fenner House. Destroyed 1886. Cranston, near Thornton. Date, 1655. See Chapter O. The Palmer Northup House. Wickford, west side of Post Road, opposite Smith "Garri- son." Much added to. Date c. 1 640-50 (?). See Chapter VI. The Sueton Grant House. Newport, Hammett's court. Date c. 1670. See Chapter V. list of old rhode island houses. 91 The Second Period. The Thomas Fenner House. Cranston, not quite a mile beyond Thornton. In excellent preservation. Stone chimney with brick top. Date, 1677. See Chapter III. The Greene House. Now the property of Edward A. Cole. Warwick, on River road, opposite Cole station. Stone chimney gone. Old framing still intact in first story. Date, i676(.?). It may be older. The Eleazer Whipple House. Lime Rock. End chimney of stone, with brick above third floor. Date c. 1677. See Chapter III. The Richard Smith House. Cocumscussuc, about a mile north of Wickford, on Pequot path. Date c. 1678-80. See Chapter VI. The Edward Manton House. Manton, north of old Killingly road. Date c. 1680, or earlier. Stone chimney retains its pilastered top. See Chapter III. The John Mowry, Jr. (.?)or Sayles House. North Smithfield, on Wesquadomeset or Sayles Hill. Date unknown, possibly 1680-90. 92 early rhode island houses. The Eleazer Arnold House. Near Butterfly Factory, about a mile west of Lonsdale. Stone chimney retains its pilastered top. Date c. 1687. See Chapter III. The Benjamin Smith House. Now known as the Cook house. Half a mile north of Arnold house. Pilastered stone chimney at end. May have had a lean-to originally, but has none now. Date unknown; probably 1687. Samuel Gorton, Jr., House. Better known as the Greene house. East Greenwich. Stone chimney with pilastered top. Date c, 1687. East end of house not so old as west end (.''). The John (.'^) Greene House. Warwick. Occupassuatuxet, now Spring Green. Residence of late Governor Francis. Date, i69o(.?). The Thomas Field House. Providence, near Sassafras and Field's Points. Date c. 1694. See Chapter HI. The House. Newport, Marlborough street, corner of Duke. End -chim- ney of stone, topped out with brick. Date c. i69o(.'*). See Chapter V. LIST OF OLD RHODE ISLAND HOUSES. 93 The Valentine Whitman, Jr., House. Lime Rock, half a mile north of the village. Stone chim- ney with brick top at end of house. Plan like Arnold house, but two -story. Date, probably 1694. The Angell House. Lincoln. On road from Wanskuck, near Wenscot reservoir. Cellar and part of chimney remain. Plan like that of Arnold or of Whitman house. Date c. 1685-95. The Congdon House. Now known as the Watson house. South Kingstown. On Post Road, below Wakefield. Date, 1690-1 700 (.''). See Chapter VI. * The Third Period. The Epenetus Olney House. "Sakesakut," between Allendale and Lymansville. Date c. 1700-5. See Chapter IV. The Benjamin Waterman House. Johnston, between Hughesdale and the Hartford Pike. Date c. 1700. See Chapter IV. The Phillips House. Belleville, near Wickford. On Post Road or Pequot path. Date c. 1700. See Chapter VI. 94 early rhode island houses. The Cory House. North Kingstown, Quidnesett. On road from Davisville to State Military Camp. Stone chimney. Date c. 1700-10. The Richard Searle House. Oaklawn. On Main street. End -chimney of stone, topped out with brick. Date c. 1700-10. The John Crawford House.' Providence, North Main street, corner of Mill street. Date, 1 715. See Chapter IV. The Joseph Smith House. Now known as Cushing house. Providence, Wanskuck, on Admiral street. Pilastered brick chimney at end of house. Date c. 1 715. The Lippitt House. Old Warwick, near church. Brick chimney at end. Date c. 1715. The Almy House. Johnston, on the Hartford Pike, about a mile east of Pocas- set. Brick chimney at end of house, but boarded over. Date ^.1718. ' Later evidence secured by Mr. ^^ield makes this house belong to Zachariah Jones, from whom Crawford bought it, and puts its date back to c. 1710. list of old rhode island houses. 96 The James (Fones) Greene House. Warwick, Buttonwoods, at head of Brush Neck Cove. Date c. 1 71 5. See Chapter IV. The Othniel Gorton House. Over a mile west of Oak Lawn. Only chimney and part of cellar remain. Date unknown; probably 1710-20. The King House. On road to Oak Lawn, east of Sockanosset Hill. Stone chimney with brick top. Summer was still to be seen in 1895, near house. Date c. 1718. The Spencer House. East Greenwich. Three miles southwest of villasre. Stone chimney with brick top in centre of house. Date c. 1715. The Coggeshall House. East Greenwich. On Pequot path or Post Road, a mile and a half below the village. Date c. 1715. See Chapter VL The Payne House, East Greenwich. A mile or more northwest from Cogges- hall house, and two miles northeast from Spencer house. Date c. 1 715. See Chapter VL The Stanton House. Richmond. Southeast corner of town, on road from West Kingston to Shannock. Date, 1715-20 (.?). See Chap- ter VL 96 ' early rhode island houses. The General Stanton House. Charlestown. Post Road near South Kingstown line. Date 1715-20 (?). See Chapter VI. The Robert Hazard House. Charlestown. Post Road near ChampHn Farm and road to Watchaug Pond. Date 171 5 (.?). See Chapter VI. The Welcome Hoxsie House. Charlestown. Post Road or Pequot path, a short distance west of General Stanton house. See Chapter VI. The Church House. South Kingstown. On road to Matunuc from Charlestown, below Post Road, south of Perry ville. Date 171 5-20 (.?). See Chapter VI. The Wanton House. Newport, West Broadway. Date c. 1715-20 (.?). See Chap- ter V. The and Houses. Newport, Duke street. Date c. 1720 (.?). See Chapter V. The Spencer House. Newport, Thames street, north of Marlborough street. Date c. 1715-20. See Chapter V. list of old rhode island houses. 97 The House. Newport, Marlborough street, corner of Branch and Farewell. Date c. 1715-20. See Chapter V. The Benedict Arnold (?) House. Newport, Hammett's wharf. Date c. i720-3o(,?). See Chap- ter V. The Arnold House. Quinsnicket Hill, near Butterfly Factory. House in ruins. No summer in second story. Date c. 1720-5. NDEX OF NAMES AND PLACES Almy, 94 Amesbury, 85 Allendale, 46, 48, 93 Angell, 93 Apponaug, 61 Aquidneck, 64 Arnold, Benedict (?), 60, 97 Arnold, Gov. Benedict, 72 Arnold, Eleazer, 41, 92 Arnold, , 97 Atherton Purchase, 12 Beers, Benjamin, 39 Belleville, 61, 65, 93 Boston, 14, 45 Brush Neck Cove, 53, 95 Bull, Gov. Henry, 56, 90 Butterfly Factory, 41, 92, 97 Buttonwoods, 45, 51, 53, 73, 78, 95 Carpenter, William, 25, 26, 85, 88 Chambers, Sir William, 15 Champlin Farm, 65, 95 Charlestown, 96 Church, 65, 96 Clawson, John, 14, 88 Cocumscussuc, 61, 62, 91 Coggeshall, 66, 95 Cole, E. A., 91 Conanicut Island, 56, 59 Congdon, 65, 93 Connecticut, 11, 13,55,59,64,65, 75,86 Cook, 92 Corsham, 85 Cory, 94 Coventry, 61 Cranston, 24, 91 Crawford, Gideon, 51 Crawford, John, 51, 94 Cushing, 72, 94 Davisville, 9.-J^ .4^^ 7;,»^i '^^ EDWARD 1 MAAITOM 19, --\^\ ^yf.—^HM.uu ^WUM^lL^o-r ..'iTa.cc^^W.t^-^^ttx:^^ IQ O O o O e I) s " ?»»„». "»„ .■> ''■', ? ' ^sssss^^s^ -iSssS ^V'vV^V^'^'v'v^ LfllEftDOiTPfl Dlaht ArRAX"I>in^ 5:alc-. =j: n- ALLMTinBER. ISOAIsjrtTHE :::\ j "I I [:--.-: (jlltT Z4. JAO" If.: --.v.-;-.-. 2nD 5PR.YJ , L[A!1T° (PRP&.0Kl6mrtL) I Cellar. UrtOER ^ fARLY ADDlTI?ri Or a J 6 rn PiAt\ or ELEAZER WHIPPLE Hou3E 1618 PE£iE/^T Addition "f i&X5 — • » » J ELEAZEP.-AE/lclLE>-fll=^ •CLrAZE.RARr(q.D M^- -PLAN- ^OAUE _i r*. 5ECT10AI LlEAZER Arnold Mo- JKTIO^: ELE.A7ER APAIOLD Mo- RE5TORATlo/i: •DUMMEK. 7-V 5CALE,,..^ ^ -^ -3 Pr. <• LOAGITUDINAL ^ECTlQ^c - CEAWroW) ^OJ 44 Aill5t. O) m pi IaMES RE E AE" U0\J:)L ' 48. • V^Ama<. Gov. Bull Ho., /Ida/port. «RR!)f ^toi^yPlanq 21 o Q s m o 70 H X IL] ^;ezzzzzzzzzzzz^ 5 Hr^^t:^ 7 Phillips- Ho. cHmAitY, RetPLACE MoULDI/^a K 2^' HiAGE. J4i DETAID OF Arthur Tu3K a/^dTe/^on CHArtree. DffllLl-WAlEmA/i H°- 3x4' ^ Col LA E 5cam raTiV(5f vs/ ixJc OH cr% I 0A6LE VaLL (Ao OUTSIDE BOAR-OlrtO «-'• H-CMK- Ji^ %\ 5iD[ or"C" X.m 5iDE or"C Toward oio3taiK3 ^ -ftAAl/loDLTMU- ri>m\m DcTAiu or i. TnE-OL/^p/-H^- If- . I PI ^^ C4lfEtpo/i|DI/«0 TO '^"'"Ct,,^,, ftft iNALYSSOfJoiNTSX -zfS^^ [ Llhrh #, 'hm Ol^EyMo. l^^nPAoSpfiY m •L- in Ol/^cy novjE 5 fox-C- P°JT 58, •D- J.W.C°R. V^M\m Dcta\l3 59, :°iFinLix WLbini TiAE (Oambrel) goGERYJlLLIAM3 ■^:i^?:!3w\ :,^^^^^^yf^^^^ RETURN CIRCULATION DEPARTMENT T0«^ 202 Main Library ALL BOOKS MAY BE RECALLED AFTER 7 DAYS iirior.ih loans may be fe'ivveO bv '-all ny 64^ 34HS tyear loans may he r-sc^iarqed bv oringing the ijookj; to th» Circulation Desk Renewals and recharges may be made 4 days prior to due date DUE AS STAMPED BELOW w^swnm m OCT 16 198^ CIRCUIATION DEPT, Y-r ■■'"1! LOAN PERIOD 1 HOME USE 2 3 4 5 6 UNIVERSITY OF CALIFORNIA, BERKELEY FORM NO. DD6, 60m, 1/83 BERKELEY, CA 94720 ^^ i I ivi24.'?75'e THE UNIVERSITY OF CALIFORNIA LIBRARY WM.HELBURN.Inc BOOKS 15 EAST 55TH ST, NEW YORK ! § i ^ 1 1 1 1 L ii,^