already acquainted with agriculture. One of them was discovered lying flat two inches below the ancient level.

Of especial interest are the mortar and grinding-stone or muller of Plate IV. The mortar or metate is of sandstone, about 911⁄2 by 11 inches on its upper surface, 6 inches thick at its highest corner and 2 to 3 at its lower end. It constitutes a sloping, not a hollow, grinding surface, one part of which rises at an abrupt angle to a higher level than the other. The muller is of coarse gritty sandstone, 4 by 311⁄2 by 134 inches. It was found a hundred feet away from the mortar, but the manner in which it is worn leaves practically no doubt that it was used with the latter. It will be noticed that it has one side worn at a particular angle. When it is turned over, the flat face turned uppermost in the illustration exactly fits the bed of the mortar, and the worn edge exactly fits the abrupt upward slope which divides the mortar into two portions lengthwise, showing that it must have been rubbed back and forth repeatedly in that position. A second mortar was found of about the same size, but with a very shallow hollow on its upper surface.

The object shown in Group C, row two, at right-hand end, is a lump of pure graphite weighing an ounce and a half. The nearest places at which this may have been gathered are Mansfield, Mass., and Cranston, Cumberland, and Tiverton, R. I. A few minute fragments of red ochre have also been detected at various scattered places during the excavations.

The gouge is stained on one edge with sulphate of iron, and Mr. Willoughby tells me that he believes that this is due to its having once lain close to a firestone of iron pyrites in a grave. No other signs of graves have been discovered on this site. Of one broken arrow-head, not illustrated here, Professor Brown remarks: "Heating has destroyed the possibility of recognition of the variety of stone. It seems to have been cracked and charred by fire." In one place, near which were found the graphite, some ochre particles, the gouge and mortars and numerous other objects, I uncovered what I thought may have been a pavement or hearth of stones laid compactly together, measuring about two by four feet. These indications of the use of fire give some slight support

to the evidence offered by the considerable variety of objects and the domestic character of some of them, that this site was actually dwelt upon at a time when its level stood high enough to be beyond the reach of the tides. At one time I thought that certain deep holes in the soil filled with decayed organic material were evidence of ancient stake-holes, but I do not now believe them to have had any connection with human activities.

The chief value of these observations is that they give some clue to the length of time during which New England has been inhabited. It is not very definite, and sets only a minimum limit at best, but so far as I know it is the only indication yet reported. This surface now lies under six feet of water at highest tides, and must have been at least two or three feet above them if it ever was the site of an encampment. In such case there must have been a subsidence of the land of at least nine feet since it was occupied by human beings as a dwelling-place. It is well known that north of Maine the land has risen since it was relieved of the weight of glacial ice; and there are many who believe that there has been a compensatory subsidence to the southward. The rate at which it occurred is still uncertain, and there is difference of opinion as to whether it still continues. Charles A. Davis, who studied the salt-marsh formations near Boston, estimates that the rate of downward movement is the same as that at which the peat growths are built up,5 and remarks that this "has not yet been determined, but is probably slow, perhaps less than a foot in a century." One observation of my own is pertinent. In front of a house on Assonet River, that was built early in the nineteenth century, I cleaned out a 12 to 15 inch deposit of peat from the beach, and found underneath it numerous fragments of oldfashioned crockery and glass that must have been thrown there

4 Economic Geology, 1910, vol. 5, p. 623.

These grasses can grow only within certain limits of depth of tidal waters. If a coast-line is sinking too rapidly, the water deepens too fast and peat will not form. Where the latter occurs, therefore, the rate of subsidence cannot have been greater than the rate at which peat can grow; and the argument of this paper, that a certain time at least must have elapsed since the surface was habitable, will hold. If the rate was slower, this time would be longer.

by the early occupants of the house while the beach was bare.6 This would indicate a peat-growth of perhaps a foot per century, as nearly as can be estimated, which agrees fairly well with the suggestion made by Davis. This is probably, therefore, fairly near to the maximum rate at which subsidence in this vicinity can have taken place, but it does not determine whether it is still going on. Douglas W. Johnson, reviewing evidence bearing upon the matter, comes to the conclusion that there has been no recent subsidence of the coast. The downward movement if it occurred at all ceased some time ago. In the vicinity of Assonet Neck this seems to be true, for the location and acreage of the neighboring salt-meadows was described nearly three hundred years ago, and so far as I can judge by careful estimates there seems to have been no change. Unfortunately none of these data are very exact, and thus we have as yet no sure knowledge as to when the movement of subsidence ceased, or at what rate it had been previously going on. One possibility, however, is that there has been no subsidence since 1600 at least, since the meadows do not seem to have changed since then, and that, if previous movement was at the average rate of about one foot in a century, the habitable character of the ancient level of Grassy Island on and beneath which these relics lie ceased some nine hundred years before that date.

This estimate, it must be acknowledged, will not be accepted by all of the best authorities. John R. Freeman, for example, believes that bench marks and other data in the vicinity of Boston afford conclusive proof that subsidence is still in progress at the rate of about a foot per hundred years. If this be a correct interpretation of the facts, then the minimum time of our calculations would be somewhat shortened. On the other hand, Johnson argues that all of the supposed proofs of land sinking within historic

The beach was bare at first, I assume, because the current naturally sweeps strongly against the shore at this place; but a wharf was built at one end of this beach, probably not far from the year 1800, and a division wall of large boulders at the other end, thus producing conditions favorable for the growth of the grass.

7 Scientific American Suppl., No. 2168, July 21, 1917; and Geogr. Rev., 1917, iii. 135.

* Report of the Committee on Charles River Dam, 1903, pp. 529ff.