1884.] 175 [Wadsworth. 



Houghton Falls are situated is the typical one on which the Kewee- 

 nawan Series was founded ; and now when my observations have 

 showed the erroneousness of the evidence said to have been found 

 there, Irving changes the boundary of the series and places it in a 

 covered district some distance below, a locality in which no one 

 before dreamed of imagining the limit of the series to be. Since 

 he has reckoned incorrectly on my method of observation, it will 

 now be necessary for him to move the boundary line still further to 

 the eastward. I protest against such methods, for if a geologist 

 can change the limits of his formation at will, in order to save it, 

 one will have to dig a canal across Keweenaw Point and tunnel 

 under Lake Superior before this question can be settled. 



Another locality in which the eastern sandstone and Keweena- 

 wan Series were shown by myself to be the same formation is on 

 the Hungarian River, flowing into Torch Lake. Irving, following 

 later, has stated that the sandstone and conglomerate often lie hori- 

 zontally, sometimes dipping slightly northwest and at others south- 

 west. From his language and section it may be seen that his 

 observations were made on the more or less undermined blocks on 

 the banks, and he states that he failed to find any evidence of an 

 increasing northwesterly dip but a general horizontality. Also that 

 the sandstone beneath the first lava flow of the Keweenawan Series 

 is probably a fallen block but if it is not, the overlying lava flow 

 certainly is. 1 It now becomes necessary to give the basis of my 

 evidence as published. My observations were made chiefly in the 

 bed 2 of the stream, the water being at that time exceptionally low 

 so that the rocks were traced continuously. It was found here 

 that in general, when the sandstone and conglomerate were undis- 

 turbed, the dip increased from 10° to the northwest up to 15°, to 

 18°, and 20°. Quaquaversal dips were seen and noted but the pre- 

 vailing dip obtained on the undisturbed bed rock was found to be 

 as given above. Instead of the sandstone in the bed of the river, 

 next to the lava, being a loose piece, it was found to extend across 

 the stream at the fall and into both banks ; also it was continuous 

 with the sandstone farther down the stream. Now this sandstone 

 was baked and indurated by the overlying flow of lava at the foot 

 of the fall, the same as the sandstones are elsewhere in the Ke- 



i l. c. pp. 149-151, 153-155. 



2 This Irvmg marks on his section as unexposed. 



