306 (3L S. Hume — Stratigraphy and Geological Relations 



during Niagaran time, and the only possible connection 

 seems to be by way of the Arctic. Megalomus has been 

 reported from Alaska (Kindle 1907) and also occurs on 

 the island of Gotland with TrimereUa. These occur- 

 rences probably mean that the Guelph sea connected with 

 the Arctic ocean, but that the waterway must then have 

 been restricted to the western or Cordilleran geosyncline, 

 while in Lockport time it transgressed more widely over 

 the Canadian shield. Pycnostylus guelphensis has also 

 been found in Manitoba, but none of the other typical 

 Guelph fossils occur here, and hence it is not certain 

 whether Guelph is actually present or whether this species 

 comes from the equivalent of high Lockport beds such as 

 the Eramosa beds of Ontario. It is possible that with 

 further field work over suitable areas in the northwestern 

 part of Canada, a Guelph fauna will yet be found and fur- 

 nish a connecting link between the Arctic and the interior 

 Guelph basin of Ontario and New York. 



Devonian. — From the known paleogeography, it would 

 be expected that rocks of late Middle Devonian time 

 should occur in the Timiskaming area, but none are now 

 present. It is possible that erosion may have completely 

 removed them, but if so, that erosion must have taken 

 place prior to the faulting that gave rise to the present 

 fault-scarp along the west side of Lake Timiskaming. 

 The downthrow side of such a fault would be a favorable 

 place for preserval, but the highest rocks now found be- 

 long to the Lockport formation. 



Post-Glactal Clay Deposits. 



The clay deposits of Lake Barlow, which form a part of 

 the northern Ontario Clay Belt south of the Continental 

 Divide, have already been spoken of in connection with 

 terraces. The clay is very evenly stratified in alternate 

 light and dark bands each less than one half inch in thick- 

 ness. Each two bands represent one complete cycle of 

 deposition, and on the basis of De Geer's results in 

 Sweden, are regarded as of one year. Over a great part 

 of the Clay Belt, no thickness over 30 feet has been pre- 

 viously reported for the clay. However, at the head of 

 Lake Timiskaming, where the clay has accumulated in the 

 depression formed on the downthrow side of the north- 

 west and southeast fault, a much greater thickness occurs. 

 Even above the present surface level the clay is over 150 



