116 



SCIENCE. 



[N. S. Vol. VII. No. 161. 



On the other hand, they all occur within 

 less than four feet of the surface, in the 

 zone in which the sand may repeatedly have 

 been disturbed by up-rooting of trees, bur- 

 rowing animals and Indian burials. ISToth- 

 ing of structure was seen in the sand itself 

 by which this crucial question could be 

 positively determined. The ' red clay films' 

 observed at various intervals in the sand 

 are not, in my opinion, lines of stratifica- 

 tion at all, nor are they strongly clayey. 

 They are rather zones or bands of infiltra- 

 tion and deposition of ferric oxide which 

 has somewhat cemented the sand grains. 

 Since they are not lines of stratification, 

 the fact that they are continuous above the 

 specimens is not necessarily conclusive 

 proof that the latter are in situ. Neverthe- 

 less, in spite of the absence of decisive evi- 

 dence pro or con, I am inclined to the 

 view that the artifacts are w situ and not 

 intrusive. 



The deposit in which they occur is, in my 

 opinion, dune-sand, accumulated after the 

 river had partially or completely excavated 

 its trench below the level of the Trenton 

 terrace. The reasons for this conclusion in 

 brief are as follows : 



1. The location. The trenches are all 

 ■ within 100 or 150 feet of the edge of the 

 terrace, which here overlooks a broad sandy 

 flood plain. According to the testimony of 

 those who have explored most thoroughly, 

 the artifacts are found most abundantly in 

 the sand near the edge of the bluff. The 

 location is one peculiarly favorable for the 

 accumulation of wind-blown sand driven 

 by southerly and westerly winds and de- 

 rived from the steep face of the terrace be- 

 fore it was covered by vegetation. As the 

 river eroded its channel below the terrace 

 level and left bare the freshly cut bank of 

 sand and gravel, the prevailing winds un- 

 doubtedly swept sand on to the terrace. 

 Naturally, some would accumulate along 

 the edge of the blufF. I have observed 



wind-blown sand at many points in exactly 

 similar positions farther north along the 

 Delaware. 



2. Tlie topography. In the immediate 

 vicinity of the trenches the surface of the 

 terrace is slightly irregular, being diversi- 

 fied by low swells and saucer -like depres- 

 sions. The surfaces of the swells are more 

 sandy than of those parts of the terrace 

 where the undulations are not present. Oc- 

 casional large boulders occur on the surface 

 of the terrace, but none were noted on the 

 sandy knolls. Certainly none occur in the 

 immediate vicinity of the trenches. In say- 

 ing that the surrounding topography is at 

 least suggestive of wind action I am not over- 

 stating the facts. In this connection, too, it 

 should be noted that the present flood-plain 

 is marked by low dunes now in the process 

 of formation, and the similarity of surface 

 is thus clearly brought out. 



3. Tlie deposit. Beneath the layer of 

 forest loam, measuring six to ten inches in 

 thickness, there is loose yellowish sand 

 absolutely without structure lines, but 

 traversed by two or three more or less dis- 

 tinct films, which, as noted above, are prob- 

 ably due to infiltration of ferric oxide. 

 Beneath the yellowish sand, which has a 

 maximum thickness of less than three feet, 

 there is a reddish layer of sand, eight or ten 

 inches thick, grading downward into the 

 cross- bedded sand and gravel, which all are 

 agreed is of glacial age and in which the 

 artifacts are not found. In the artifact- 

 bearing sands there is no evidence that it 

 was water-deposited. That it is a local de- 

 posit is shown by the fact that it does not 

 occur on the gravel seen in large open pits 

 a few hundred yards distant from the 

 trenches. It seems to be best developed 

 along the edge of the terrace. Its texture 

 is not unlike that of wind-blown sand ob- 

 served elsewhere, and it is decidedly unlike 

 the sand-beds exposed in the gravel pits. 

 This latter fact, however, does not neces- 



