192 SMITHSONIAN ISIISCKLLANEOUS COLLECTIONS VOL. 75 



tious and designate them by appropriate names if they have not been 

 named elsewhere in tlie same geological iMH)vince. 



The break in the record may be fonnd in an ajiparently solid layer 

 of rock, as shown by a block to which I called attention in 191 2.* 

 The lower part of the layer cai"ries typical Upjier Cambrian, the 

 npper part typical Upper Chazyan fossils, and the strata and faunas 

 of the intervening- Canadian formations are absent. Elsewhere the 

 missing strata and faunas appear, and the two faunas of the solid layer 

 of rock are found to be separated by several geological formations, 

 which proves that a long time interval elapsed between the deposition 

 of the lower and upper portions of the layer. A critical examination 

 of the block also showecl a marked dilTerence in the composition or 

 structure of its lower and upper portions, which is not apparent on a 

 hasty inspection. This illustration is given here in order to show how 

 futile are lithological characters and broad generalizations, based on 

 imperfect knowledge, in determining the limits of geological forma- 

 tions and even systems. 



Every field geologist working in sedimentary formations should 

 have sufficient knowledge of the faunas that he may encounter to 

 identify the lu)rizon in which they occur, or else have a trained paleon- 

 tologist with him to collect the fossils and post him from day to day 

 on the stratigraphic position of the various strata he is studying. Less 

 than this means inaccurate maps, structure sections, and historical 

 records. 



I insert the above observations here as the Sawback formation is a 

 good illustration of the difliculties a geologist will meet with in the 

 great Cordilleran and Appalachian Mountain ranges, and the broad 

 intercontinental areas. Thanks to Ulrich, Schuchert, and their associ- 

 ates, much has been done to elucidate the history of the continent from 

 the beginning of Paleozoic time to the present. Much remains to be 

 done especially in the western United States, Canada, and Alaska. 



DEVONIAN 



The first to recognize and name the Devonian formations of the 

 Bow Valley section of Alberta was R. G. McConnell of the Canadian 

 Geological Survey, who in 1887 published, under the heading of 

 BanlY limestone, the " BantT series." He says : 



" The BantY limestone series has a total thickness of about 5,100 

 feet [ 1,554-5 "l], and is divisible into a lower and upper limestone, 

 and into lower and upper shales : " ^ 



* Smithsonian Misc. Coll., \'ol. 37, No. 9, 191J, pp. 253- J54. 

 ° Geo!. Surv. Canada, Report for 1886 (1887), Pt. D. p. 17 D. 



