WINOOSKI RIVER. 133 



width in Essex. It extends nearly to the north line of Colchester, almost connecting with the highest ter- 

 race of the La Moille river. Its wide extent is represented on the general map, Plate II. On the south 

 side, this fourth terrace is very extensive. Commencing quite wide on the Winooski marble ridge, it nar- 

 rows and divides at Burlington, part running west, and the rest passing on east of the University. East of 

 Burlington it expands again to a width of two or three miles, and narrows out gradually beyond Williston , 

 becoming a point at Richmond. At Painesville this terrace is 240 feet above the lake. It is not very dis- 

 tinct in the village of Burlington, on account of the encroachments of man and of the lake upon it. The 

 third terrace in Burlington is the flat just north of the common ; the fourth between the common and the 

 University, which is upon a hill of modified drift. We should not expect the terraces in Burlin gton to be 

 well marked, because they are on the lee side, so to speak, of the terraces on the river. Whatever terrace- 

 like forms appear ought perhaps to be considered as lake terraces, and belonging to Lake Cham plain. 

 Northeast of Winooski village, seven very distinct terraces may be seen rising above each other in suc- 

 cession, and extending two or three miles to the northwest. The highest of them corresponds in height 

 with what has been called the fourth before. This great terrace, No. 4, we think may be regarded as a 

 delta terrace, extending from Richmond to Lonerock Point. When the lake was at the level of its summit, 

 this terrace must have occupied, besides its present situations, the whole of the valley now occupied by the 

 Winooski below Richmond. So that, had the water dried up suddenly, there would have been presented 

 to the eye a vast unbroken plain of sand from Lonerock Point to Richmond, and from Mallett's Bay, Col- 

 chester and Milton, on the north, to Williston village and South Burlington, on the south. Since that 

 period, the Winooski has cut out its present channel and formed the lower terraces ; and also tributary 

 streams have done their part in excavating their own beds, and forming their terraces, while the level of the 

 lake and of the channel of the Winooski were gradually sinking. 



We stated that oceanic deposits were connected with these fluviatile accumulations. We shall hereafter 

 state the particulars respecting these deposits (the Champlain clays), and will only anticipate a little. 

 Fifty feet below the plain of Burlington, and in the whole of the railroad excavation in the village, there 

 may be seen strata of clay containing fossil marine shells, such as are now found living upon our coasts. The 

 same may be found underneath the sand, over the whole of this high terrace. This shows us that it was 

 probably the ocean in which this delta terrace was formed, and that the shells were mixed with the deposits 

 brought down the Winooski. Hence this plain was formed by the joint action of the river and of the ocean ; 

 both the terraces and the clays are consequently of the same age. These shells have been found at a height 

 of 300 feet above the lake, and therefore we may expect to find them in the valley of the Winooski as far 

 as Waterbury. We have not examined the clay banks for them in this valley, for want of time. Should 

 they not be found in future, we think the cause may be ascribed to the presence of a fresh water current. 

 Marine animals cannot generally live in fresh water. Perhaps this may explain the reason why we have 

 not found these marine fossils in the valley of the Connecticut and other streams below tb.3 height of 500 

 feet above the ocean. On the Connecticut we can proceed to the Twenty-miles Rapids in Waterford, 

 before we reach this altitude, but we know of no case between Waterford and Long Island Sound where 

 these fossils have been found. So on the Hoosac and Battenkill Rivers. 



Also the fact that sand only is found at this delta at the height of 240 to 260 feet above the lake, shows 

 that the current besides being too fresh was also too strong for the deposition of clay or silt in which these 

 shells are found. The character of the deposits on the Connecticut is generally sandy or gravelly, and where 

 clay abounds it is possible that some of these were formed in basins separate from the sea, as they occur in 

 the lowest portions of the valley, and are often filled with terrestrial fossils, such as wood, fruits and nuts. 

 Upon this interesting sub-basin near Burlington, we will adjoin some remarks of the late Prof. Thomp- 

 son, written for the Geological Survey in 1847, in a manuscrip"t entitled Geology of Chittenden County. 

 They will illustrate positions already advanced by us respecting the extent of the highest terrace, and the 

 distribution of the sands and clays. 



" Sands. The deposits of sand of this period (older pleistocene) are quite extensive in Chittenden 

 County, particularly in the towns of Milton, Colchester and Burlington. They are for the most part super- 



