The geographical position of the defensive works connected with these remains indicates, as has been often remarked by writers on this subject, a pressure from northern hordes which finally resulted in driving the inhabitants of the fertile valleys of the Miami, Scioto, and Muskingum, southward, possibly into the Gulf States, where they became incorporated with the tribes of that section. [Footnote: Force: "To what race did the mound-builders belong?" p. 74, etc.] If this is assumed as correct it only tends to confirm the theory of an Indian origin. But the decision is not left to mere assumption and the indications mentioned, as there are other and more direct evidences bearing upon this point to be found in the works of art and modes of burial in this region. That the mound-builders of Ohio made and used the pipe is proven by the large number of pipes found in the mounds, and that they cultivated tobacco may reasonably be inferred from this fact. The general use of the pipe among the mound-builders is another evidence of their relation to the Indians; while, on the other hand, this fact and the forms of the pipes indicate that they were not connected with the Nahua, Maya, or Pueblo tribes. Although varied indefinitely by the addition of animal and other figures, the typical or simple form of the pipe of the Ohio mound- builders appears to have been that represented by Squier and Davis [Footnote: Ancient Monuments of the Mississippi Valley, 1847, p. 179.] in their Fig. 68; and by Rau in Smithsonian Contributions to Knowledge, No. 287. [Footnote: 1876, p. 47, Fig. 177.] The peculiar feature is the broad, flat, and slightly-curved base or stem, which projects beyond the bowl to an extent usually equal to the perforated end. Reference has already been made to the statement by Adair that the Cherokees were accustomed to carve, from the soft stone found in the country, "pipes, full a span long, with the fore part commonly running out with a short peak two or three fingers broad and a quarter of an inch thick." But he adds further, as if intending to describe the typical form of the Ohio pipe, "on both sides of the bowl lengthwise." This addition is important, as it has been asserted [Footnote: Young Mineralogist and Antiquarian, 1885, No. 10. p. 79.] that no mention can be found of the manufacture or use of pipes of this form by the Indians, or that they had any knowledge of this form. E. A. Barber says: [Footnote: Am. Nat., vol. 16, 1882, pp. 265, 266] The earliest stone pipes from the mounds were always carved from a single piece, and consist of a flat curved base, of variable length and width, with the bowl rising from the center of the convex side (Anc. Mon., p. 227).