40 CYPRINTD^ OK EASTERN (CANADA. 



To one \Yhc) has been afield, examined liundreds of speci- 

 mens, and noticed the wide range of variation in this respect 

 among stock of the same or neighboring ])laces. the founding of 

 species on such small, inconstant, and largely' individual dif- 

 ferences must api)ear as little else than designating the extremes 

 of variation in a given s})ecies. The Avriter has found iihimheus 

 to vary from 9-55-6 or 7 to 12-70-8; and hence to comprehend 

 within its limits the three above si)ecies, at least as far as this, 

 the chief feature, is concerned. 



While individualism may be of much interest to the nat- 

 uralist or }ihysiologist as marking a reversion or the extent of 

 variation between present and ancestral types, it can be of no di- 

 rect value in classification, uidess it be found to spread out. so 

 to speak, and become a general character of stock occupying a 

 wide range. Then it ceases to be individualism, and rises into 

 varietal or specific value, according to its significarce and the 

 degree of structural modification it re[)resents. Among New 

 Brunswick Couesii the scale formula ranges from 10-()0-7 to 11- 

 67-7 or 8, but as the type hovers around 10-65-7, the extremes 

 must be attributed to individualism, for they are seldom or never 

 associated with any other dift'erence. It is interesting, however, 

 to note how individualism here becomes a general characteristic 

 of stock in remote and isolated districts. Nova Scotia fish, col- 

 lected by the writer from the Annapolis and La Have rivers, 

 show a marked tendency towards a reduction oi' the fornjula 

 and .approximation to the minimum standard of New Brunswick. 

 9-55-6 or 7 being frequently met with, while 10-60-7 is seldom 

 exceeded. In Gaspe. on the other hand, the common average is 

 higher than the maximum, especially in the Grand Cascapedia, 

 Nouvelle (New Carlisle) Lake, Grand Pabos, and streams east- 

 ward, where it balances at about 12-70-'8. 



From the.3e facts it must be apparent the scale formula can 

 afford no sufficient ground for specific difi'erentiation within 

 the limits noted ; yet the extremes occuj)y certain geographical 

 ranees, are more or less isolated, and may evidence certain 

 physical and biological forces, making for uniformity and sta- 

 bility in definite directions. As yet, however, the series is too 

 conii>lete. anv two consecutive links in the chain too mucli alike. 



