FOURTH REPORT OF THE DIRECTOR I907 55 



same author had previously described the beds at Campbellton, N. B., 

 before they were found to contain fish remains, and had pronounced 

 upon their equivalence with the lower part of the Gaspe sandstones. 

 This opinion was based upon evidence of paleobotany, and, having 

 been confirmed a few years afterward by J. W. Dawson, is now 

 generally accepted. Indeed, Logan seems to have entertained simi- 

 lar views as early as 1863 [Geol. Can. p. 450]. As for the plant 

 and fish-bearing beds at Scaumenac bay, on the Quebec side of the 

 Restigouche, these are asserted by Dawson to be " no doubt the 

 equivalents and continuation of the upper part of the Gaspe sand- 

 stones." In the absence of a more precise indication of their age, 

 these beds are commonly referred to as Upper Devonic, and their 

 vertebrate content favors that conclusion. 



In the light of Dr Clarke's coordination of the Gaspe sandstone 

 series w^ith rocks of Postoriskanian age, we are no longer required 

 to look in that direction, nor to the probably contemporaneous Camp- 

 bellton fauna for the origin of the Onondaga fish fauna found in 

 New York State. On the other hand it may be conceded as rather 

 more likely that there was some movement among vertebrate organ- 

 isms in the reverse direction, for such an hypothesis would account 

 for the presence of a typical Onondaga species, Machaeracan- 

 thus sulcatus, at dififerent levels in the Gaspe series (Logan's 

 Divisions i and 6). The genus Cephalaspis is common to both the 

 Gaspe series and Campbellton beds, and together with the majority 

 of forms from the latter horizon is indicative of Old Red sandstone 

 conditions. 



Reverting now to the Hudson Bay Middle Devonic. fauna, we find 

 that, as listed by Whiteaves, it is unmistakably of about the same 

 age as the Onondaga. According to Schuchert, its faunal facies 

 " is more that of the Mississippian type than any other known in 

 America." This similarity is therefore held to indicate that there 

 was at least intermittent connection between the two basins during 

 Onondag^a time, and persisting well into the Hamilton. It is ad- 

 mitted, however, that the question as to how the stream of migration 

 entered the Hudson Bay area during the Middle Devonic is not so 

 easy to answer. Precisely at this point some light is thrown on the 

 Droblem by vertebrate paleontology. A number of specimens of 

 Macropetalichthys sullivanti (=]\[. rap hei do- 

 lab is) are recorded by Bell and Whiteaves frv-^.m the country 

 immediately west and south of Hudson and James bays. This 

 exclusively Onondaga species (Mr Schuchert inadvertently calls it 

 a Hamilton fossil) is most abundant in Ohio and Indiana, and 

 decidedly Icss'common in New York State. The same genus, repre- 

 sented bv some two or three species, occurs also in the Eifclian 

 Devonic, which is equivalent practically to the Onondaga, and in ihe 

 slightlv earlier horizon in Bohemia designated as etagc G^ \o 

 trace of it occurs, however, in the Mcsotlcvonic of the maritime 

 provinces of eastern North America. One may readilv infer that 

 this genus and its various associates arc indigene in Bohemia, a part 

 of the vertebrate fauna from etagc G* being inccjitive in etagc V. 



