belonged to the family which bought it of the Whitfields (1769), that Whitfield's house was then standing. And we have the testimony of the Massachusetts Historical Society, given when the property had passed to the owner whose descendants held it until 1900, that the same house had meanwhile been "handsomely repaired."* The identity is beyond question, even if it only extends to the fireplace and chimney.

But when, precisely, did Henry Whitfield build the house, or, at any rate, the chimney? It needs no contemporaneous documents, testifying explicitly to the date of erection, to prove that the date is earlier than 1650, when, as such documents prove, Mr. Whitfield left Guilford finally. It needs no contemporaneous papers to prove that he built his house, like other settlers, as soon as possible after the town was settled, and documents amply prove that Guilford was settled in 1639, and that the settlers, with Henry Whitfield at their head, had begun to occupy the lands in some way "as planters" before September twenty-ninth of that year, when the formal transfer was made by the Indians. They had probably already built some temporary houses or cabins to shelter them during the winter, and those who could do so must have completed more permanent houses in the course of the next year. Mr. Whitfield, the richest of the planters, with a wife and eight or nine children, had no doubt substantially finished his stone house in 1640, though we are assured that it could not have been finished in 1639. But since it was intended, according to Mr. Ruggles, that is according to the children and grandchildren of the settlers, to serve as a fort, and such a defense of the settlement would be secured as early as it could be, it is hard not to believe that it was begun in 1639, and so far finished that it could be used as a place of refuge in case of attack, and could be made a comfortable abode. for Mr. Whitfield and one of his sons, or some friends, until spring. And during the recent changes already referred to a break was found in the west or front wall of the cellar which suggested to the architect that two parts of that wall had perhaps been built at two different times. And on the east or

* Ruggles' Hist. of Guilford, Collections of Massachusetts Historical Society, 1st series, vols. iv and x.

opposite side of the house a vertical joint, discovered where the north wall of the wing or ell met the wall of the front part near the north end of the house, and signs of plaster on the latter behind the end of the other wall, suggested that the ell was built later than the front part. On these grounds the architect thinks it "somewhat" likely that the house was at first a single room. It might have been a square structure with the south end filled in with timbers, while the great north fireplace made it habitable. We cannot affirm positively that this fireplace and the adjoining walls date from 1639, but several facts point that way and create a strong probability that that is the true date. That the house, as a whole, dates from 1640 is fairly well proved.

But is the Whitfield house the oldest in New England, or, as has often been affirmed, in the United States? It seems impossible to be sure that it is, unless one can know the age of all the oldest houses in the country (leaving out of consideration the much earlier Spanish settlements), and that, I fear, we can never know. A very partial inquiry gives, however, a result interesting as far as it goes. In Virginia no dwelling house now standing is known to have been built before 1654. In Rhode Island one house is variously assigned to 1640 and 1639, making it as old as ours. But good authorities think this assignment less than probable; that house is "possibly" as old. In Massachusetts there are, or were not long ago, about a dozen houses for which a date earlier than 1640 is claimed, and there is an antecedent probability that such claims can rightfully be made in territory settled nearly twenty years before Guilford. And yet it does not appear that any one of these dates has been so conclusively established as to be accepted by local historical students as a body. That is, it seems that the precise dates depend on tradition only. At least that is the only inference which I can draw from what I have learned about the matter. It remains entirely possible, if not probable, that Massachusetts does now contain dwellings older than the Whitfield house; it is possible that there are such in Virginia, settled in 1607. But it seems fair to say that

in the three states mentioned there are none for which a date

୮

as early has been so nearly proved. I believe that as much could have been said if the inquiry had been extended to other states of English origin as long settled as Connecticut.

Before dealing with the question as to the probable appearance of the house in Whitfield's time, it is better to say something of the changes known or believed to have been made during the long interval, now approaching two centuries and threequarters, with a brief mention of the owners who made them. The story of the successive owners is an interesting one and enriches the house itself with some memorable associations, but I can give little more than names and dates.

Mr. Whitfield died in England, in 1657, leaving his Guilford property to his wife, Dorothy Sheaffe. When their son, Nathaniel, sold it for his mother two years later (1659), the purchaser was Major Robert Thompson, a Puritan merchant of London, who had become also a landowner, and who is to-day represented by his descendant, Sir Francis Astley-Corbett, owner of Major Thompson's country house, Elsham Hall, in Lincolnshire. Four descendants in the male line, all living in England, held the property until 1772, when another Robert Thompson sold it to Wyllys Eliot of Guilford, a fictitious lawsuit in New Haven being necessary to break the entail. The house itself belonged to Mr. Eliot (of the family of the famous apostle to the Indians, still represented in Guilford and else where) less than a fortnight, and was sold in November, 1772, to Joseph Pynchon, the solitary owner, after Whitfield, who ever lived in it; it is, therefore, a "Pynchon House" by a much better title than the "House of the Seven Gables." In 1776, Mr. Pynchon sold it to Jasper Griffing of Guilford, with whose descendants it remained until the sale to the state of Connecticut in 1900. The last individual owner, Mrs. Sarah Brown Cone of Stockbridge, descends from a first cousin of Mrs. Whitfield's, Joanna Sheaffe, the wife of William Chittenden, and therefore shares the blood of the first mistress and second owner.

In indicating the changes which have made the house what it is, I follow a guide to whom I have already referred, the architect who planned and carefully watched the latest changes, whom I may now introduce as Mr. Norman M. Isham of Providence, and to whom my obligations are greater than I can easily tell you. In a book on "Early Connecticut Houses," of which he was one of the authors, and which some of you probably know, he assigns by conjecture to one of the Thompsons, near the beginning of the eighteenth century, the division by a floor and partitions of a long, high hall, of which tradition makes the whole front part of the house to have consisted originally, and which Mr. Isham is strongly inclined to believe in. It is likely to have been divided very early by a floor, and just such a change had been made in a multitude of such halls in England in the sixteenth century. This change may have been Major Thompson's own contribution to the convenience and comfort of his American tenants. Another Thompson very likely removed in the eighteenth century a chimney which there is good reason to believe stood in the beginning at the south end of the hall but which did not exist forty years ago. This removal was made, doubtless, to render it possible to put windows into the south upper room and perhaps the attic, but the south wall was weakened by doing it. Almost certainly after Jasper Griffing became owner repairs were made which may have included the building of a second fireplace at the north end, in front of the old one. Early in the nineteenth century the outer wall was plastered. Finally, in 1868, during the ownership of Mrs. Cone's mother, Mrs. Chittenden, when the house had become uninhabitable, and at least the south wall was insecure, it was largely rebuilt. This was done to save it, and those most concerned earnestly desired to preserve everything that was capable of preservation. The extent of the changes, however, has been variously reported and there has long been a strong desire to know positively how much of the old work was left, a desire which it was hoped that the late changes would. to some extent gratify. The latest fireplace in the north chimney was now introduced and the south chimney was restored.

A sad loss, apparently inevitable, was that of the old roof, the curious construction of which, as far as can be inferred, might have been intended to provide gables in front, containing windows for the original garret, otherwise quite in darkness.

The observations made by Mr. Isham, in 1903, were limited by the object then in view. This was not to solve problems, but to provide a convenient room for the principal collections of the Museum, of course without injury to what was most precious of them all, the remnants of the original house. Less was learned than had been hoped, therefore, about the building as it was at first and about the portion of that which remained after 1868.

The most interesting results concerned the fireplace and chimney, which, if I understand my guide, were probably unique in New England in virtue of two or three features which have survived on the other side of the Atlantic. One of them is shown in plans of Shakespeare's birthplace, and consists of two pilaster-like buttresses, one on either side of the wide opening of the fireplace, and each extending backwards two feet from the opening and projecting into it, one a little less, the other a little more, than one foot. They must have borne part of the weight of the huge timber lintel or manteltree (long since removed and then bearing marks of fire) on which the masonry forming the front of the chimney rested. And a mass of new masonry, now covered by panelling and approximately a rude triangle, indicated the space from which the old work had fallen. Another peculiar feature, such as is seen in photographs of old English fireplaces, is a depression in the back of the chimney four and a half feet wide and twelve inches deep, beginning about thirty inches above the present floor, and disappearing nine or ten feet higher, through the gradual contraction of the flue to the width of the depression. Finally, it was found that the chimney contained two flues, the partition beginning seven or eight feet above the hearth. These were inserted, Mr. Isham believes, to carry off the smoke from two small fires to be built, one or both, in mild weather at opposite ends of the fireplace, where was found, beneath each flue, an iron bar for supporting pots and kettles. In cold weather