SOURCE OF ST. PETEK's RIVER. 179 



banks of the Mississippi, and for which Lake Pepin has 

 long been celebrated. When we consider the easy decom- 

 position of the wacke in which these geodes are imbedded, 

 we cannot be surprised that the latter are always found 

 loose and separated from the imbedding rock. The amyg- 

 daloid was not observed to be stratified, but in some places, 

 it presented a columnar division. We are therefore induc- 

 ed to attribute the columnar appearance which we thought 

 we had seen in the Pate island to the presence of trap 

 rocks at its surface ; it appears to us probable that all the 

 islands which we saw are more or less covered with this 

 interesting formation, which was probably deposited at a 

 period subsequent to that at which the horizontal slate was 

 formed. The examination of the geological structure of 

 the north coast of Lake Superior will probably renew the 

 discussion of the aqueous or igneous formation of the trap 

 rocks. Upon this point we will not dwell, because wc have 

 no new facts to offer. Our visit to this coast was of too 

 transient and hasty a nature to permit us to extend our ob- 

 servations. We, however, confess ourselves unable to dis- 

 cover in secondary trap rocks in general any signs of a Nep- 

 tunian origin. If we were disposed to launch into theory, 

 we might connect the existence of these trap rocks with 

 the evident signs of the action of heat observed upon some 

 of the rocks which we met in Winnepeek river. We might 

 perhaps also attempt to refer to volcanic phenomena on a 

 great scale, the changes to which we have already hinted 

 as having taken place in that country. The rupture of the 

 great barrier which confined the waters of the immense 

 lake might be shown to have been probably produced by- 

 such causes. 



The physical revolutions, of which this part of our con- 

 tinent was the theatre, were too great to attempt to assign 



