GEOLOGY OF OGDENSBURG 5I 



fossiliferous, sometimes abundantly so. Above these, separated 

 by an unknown but probably small interval, follow the gray, flinty 

 beds of the Red Mills quadrangle, very unfossiliferous and of 

 unknown thickness. The section shown seems to correlate with 

 division D of the Champlain section. Further work down the river 

 may show somewhere a good section of the higher beds, so that 

 the section may be completed and the full character of the Beekman- 

 town group of this district shown, but the whole valley is so heavily 

 drift-covered that we are not sanguine that this can be successfully 

 done. The formation extends down the river for many miles more 

 before the overlying Chazy beds appear, the Ottawa basin Chazy. 

 It is very doubtful if these beds get over into New York at all. 

 But in any event the indications are that a considerable thickness 

 of Beekmantown beds higher than those in the Ogdensburg region 

 must be present. 



In New York State Museum Bulletin 145 we made certain pre- 

 dictions concerning the age of the Beekmantown formation of the 

 St Lawrence valley based on the results of the Thousand Islands 

 work and of our much earlier reconnaissance work farther east, 

 particularly around Potsdam.^ The results just outlined in the dis- 

 cussion of the Ogdensburg formation seem to us to fulfil those 

 predictions. We argued that the first Beekmantown to appear 

 down the river from the Thousand Islands should not be the lowest 

 division of the formation in the Champlain valley, but of higher 

 beds. The basal beds prove to be of the age of division D, with 

 divisions C and B entirely lacking. There is also a break between 

 them and the beds directly underneath, as we argued should be the 

 case, a break not only shown by a basal conglomerate and sand 

 grains in the basal dolomite, but also by the fact that the beds just 

 underneath vary much in horizon across the district. So far the 

 results of this study accord with expectation. The unexpected 

 results are the continued presence of the Tribes Hill formation 

 underneath, and the entirely new division represented by the 

 Heuvelton beds. Our work has shown an unconformity between 

 the Ogdensburg and the Tribes Hill, and Chadwick's work on the 

 Canton sheet shows a break between the Tribes Hill and the under- 

 lying Heuvelton. We have not demonstrated a break between the 

 Heuvelton and the underlying Theresa, and have mapped them 

 together, but it is quite likely that a break exists. The work of 



Pages 92-96. 



