Among the boulders of the belt are occasionally found specimens of impure limestone or of limy sandstone that might perhaps be referred doubtfully to some member of the paleozoic series; but on the other hand, might with equal or greater probability perhaps be referred to the similar rocks of the Huronian series. These are quite rare, never forming, so far as my observations go, as much as one per cent. of the series. In the several definite enumerations made to determine the percentage of the doubtful specimens, the result never exceeded a fraction of one per cent. In the most extensive enumeration the result was about one-half of one per cent. Aside from these doubtful specimens there are practically no boulders in the belts that can be referred to any of the paleozoic rocks that intervene in the 500 miles between the parent series north of Lake Huron and the tract over which the boulders are now strewn. Occasionally there may be seen erratics from the paleozoic series at or near the surface, but they are not usually so disposed on the surface as to appear to be true members of the superficial boulder tract. There is, therefore, the amplest ground for the assertion that these boulder tracts are of distant derivation, and that they are essentially uncommingled with derivatives from the intermediate region. 3. The boulders of this series are much more angular than those of the typical till sheets. Some of them, indeed, are rounded, but the rounding is generally of the type which boulders derived by surface degradation and exfoliation present. They rarely have the forms that are distinctively glacial. Quite a large percentage are notably angular, and have neither suffered glacial rounding nor spherical exfoliation. Some few are glacially worn and scratched, but the percentage of these is small. The tracts therefore present these four salient characteristics: (1) the boulders are derived from distant crystalline terranes (400 to 500 miles) and are essentially uncommingled with rock from the intervening paleozoic terranes; (2) they are essentially superficial, and the associated earthy material has a texture differing from that of the subglacial tills; (3) they are notably angular and free from glacial abrasion, except in minor degree; (4) the tracts are so associated with terminal moraines and so related to the topography of the region, that there is no rational ground for doubt that the boulders were borne to their present places by the glaciers that produced the correlative moraines.