NORTH AMERICAN SALMONIDAB 



Table 21. — Summary of gill-raker count of Xorth American Salmoniilac 



[First gill arch, left side) 



253 



' Standard deviation of 1.6 multiplied by 2.1. McCrimmon (1949) gives 

 1.6 as standard error of mean for salar and 6.01 as standard error of mean for 

 tmlla. The first must be standard deviation, the second is improbably 

 small since standard deviation would be only 0.06. 



' Assuming same interperceutile range as for S. gairdneri above. 



3 Eastern Canada. 



' Karluk River, Alaska. 



6 Anadromous stock, Aloser River, Nova Scotia. 



fl Resident stock, Moser River, Nova Scotia. 



' Three small brooks in Nova Scotia. 



TSHMVrrSCHA 



6 KAULOOPS 



O UALUA 



o NAMAYCUSH 

 O ALPINUS 



o FONTINAUS 



TRUTTA 

 SALAR 



LATERAL LINE SCALES 



ro'oci 



LO OTMI 



3 rOCDSTER AnO W» 

 JTHER AUTHORS 

 I LATERAL LINE 5ENS0RT PORES 



SENSORT PORES CONTINUED ONTO CAUDAL 

 ^ 80-20 PERCENTILE 



NUMBCR OF LATERAL LINE SCALES 



Figure 7. — Number of lateral-liue scales. 



lateral line from the gill aperture to the adipose 

 fin and, then, to continue the count at a lower level 

 from the adipose fin to the caudal. Tlie five meth- 

 ods are brieflj- summarized as follows, in the order 

 of usually increasing count : 



1. Nmnber of sensory pores on lateral line. 



2. Number of scales on lateral line. 



3. Number of diagonal scale rows in the hori- 

 zontal row just above the lateral line. 



4. Niunber of diagonal scale rows from top of 

 gill aperture to caudal. 



5. Number of diagonal scale rows from top of 

 gill aperture to caudal, counting on a lower hori- 

 zontal row posterior to adipose fin. 



ilost investigators terminate the count at the 

 base of the caudal fin (standard length), but some 

 count the scales that extend on to the caudal fin. 



Available counts of lateral-line scales (methods 

 1 and 2) are sunmiarized in table 22 and in figure 7. 



It is obvious from figure 7 that the variation 

 between the mean numbers of lateral-line scales 

 from different localities (and perhaps between 

 counts b\' different investigatoi-s) is so great that 

 only a few of the species can be separated by this 

 character. However, there is a general trend 

 with species of Oncorhynchus the highest, and 

 fontlnalk; salar, and trutta the lowest counts. 



