. 216 



The general absence of conglomerate from both the bed-rock 

 and drift of the harbor islands and peninsulas is a remarkable 

 fact, the true signification of which appears to be that, whereas 

 on the main land, where the strata are more elevated, the anti- 

 clinal crests have been worn away, exposing the conglomerate, 

 and in many cases the underlying crystallines, under the harbor 

 these are yet substantially intact, and the inferior division of 

 the formation is concealed. And then, again, it is highly 

 probable that as we recede from the ancient shore line the vol- 

 ume of the slate is augumented at the expense of that of the 

 conglomerate, the latter rock having its greatest and the for- 

 mer its smallest development toward the head of the Primordial 

 bay. If the bottom of the harbor could be mapped geologically, 

 the contraction of the conglomerate belts and the expansion of the 

 slate belts eastward would probably be very conspicuous, this rela- 

 tion being due in part to the diminished erosion toward the east, 

 and partly to the increased ratio of the slate to the conglomer- 

 ate in that direction. It will be observed that this view is 

 diametrically opposed to that commonly held, according to which 

 the conglomerate is the overlying rock, and is absent from the 

 harbor while yet remaining on the main land in consequence of 

 the supposed greater deundation suffered by the former district. 

 I conceive that the axes of the folds in our Primordial beds, fol- 

 lowing the ancient sea bottom, dip down toward the east or 

 6eaward. 



Lying in the course of the Painsford Island synclinal, toward 

 the north-east, are George's Island, the Great, Middle, and 

 Outer, Brewster Islands, and Martin's Ledge ; and on all of 

 these, so far as known, the rock is slate. 



The Hyde Park, Mattapan, and Sqaantiim Belt. — The 

 belt of country one to two miles wide extending E. by N. from 

 Hyde Park to the outer end of Squantum, including the valley 

 of the Neponset, and marked on the map mainly as conglomer- 

 ate, is very complicated stratigraphically. In a general way 

 the region indicated is anticlinal, lying between the synclinal 

 last described and the well-defined band of slate on the north, 



