148 N. H. WINCHErX WAS MAN IN AMERICA IN GLACIAL PERIOD 



clase, orthoclase, hornblende, and a rare grain of magnetite, and also of 

 microcline, subangular, angular, and a few rounded. 



3. A sample from the roof of the tunnel is not noticeably different 

 from the sample just above the stratified clay seam. They both contain 

 ramifying calcareous tubules and effervesce freely. 



4. The stratified clay seam which immediately overlies layer number 

 1 is normally very dense and without effervescence, but since its forma- 

 tion calcareous septaria have formed numerously in it, and in its joints 

 calcareous incrustations have been deposited by ground waters that 

 could not readily penetrate it, so that the stratum is mottled with dark 

 and light. In being excavated it crumbles into many polj^gonal masses 

 of the size of half an inch and smaller. In the microscopic mounting 

 are seen mostly grains made up of minute granules which are undeter- 

 minable, apparently unreduced pebbles of the clay itself, and a few 

 angular polarizing grains of quartz and of plagioclase (?). The clay seam 

 is apparently formed from slow wash from number 1 or from the Car- 

 boniferous shale overlying the limestone ledge, or from both, the few 

 crystalline grains and the lime having been derived from the same source 

 as the same in the loess or from the loess itself. 



The transition in the character of the deposit from numbers 1 and 4 

 to numbers 2 and 3 is sudden and remarkable. Certainly the bottom 

 portions of the walls of the tunnel were not derived from the same 

 source nor by the same agency as the upper portions. The bottom por- 

 tions are not loess. The upper portions, so far as their microscopic 

 characters indicate, are loess. 



When we come to consider, however, the megascopic characters of this 

 overlying mass of loess, we see some divergence from the usual loess as 

 seen along the great valleys. It is not finely laminated nor distinctly 

 stratified as a whole. The only sign of the action of water consists in a 

 few rude bands that do not extend the whole length of the excavation.^ 

 These bands blend into the main mass, but are characterized by con- 

 taining limestone pebbles and white spots resulting from the rotting of 

 limestone pebbles. They also contain other pebbles, some of which, 

 being of white quartz, have resisted decay, while others have decayed 

 entirely, leaving only limonitic blotches that show their form and size; 

 but, as already stated, these bands merge into the general loess body, 

 being from 2 to 4 inches in thickness where they show best. 



Throughout the loess mass also, where cut by the trench of Mr Fowke, 

 are similar blotches, both white and limonitic, scattered sparsely over 

 the face of the cut. It appears indeed as if the whole loess mass were 



*Thls observation was made in the cross-tunnel of Mr Fowke, west from the original tunnel 

 perhaps 15 feet. 



