GREAT LAKES COREGONIDS 



319 



species (except Jioyi and Iciyi) the AC value frequently is less than 1, while in these 

 species and in clupeaforrnis and quadrilaterale the value usually is more than 1. 



The number of anal rays also is variable and is distinctive in none of the forms. 

 The usual number is 10 to 12 but may vary within the species. Thus, reighardi of 

 Superior has slightly more rays than the typical race. 



PECTORALS 



The length of the pectorals in relation to the pectoral-ventral distance is of 

 systematic importance, but none of the species are absolutely separable by this 

 character, as may be seen in Table 10. The races of Tciyi have relatively the longest 

 pectorals and those of reighardi the shortest. 



Within each species there is enormous variation in this character. Thus, the 

 pectorals of the Huron race of johannx average longer than those of the Michigan 

 race; in zenithicus they are shorter in Michigan and Huron than in Superior and 

 Nipigon; in reighardi they are longer in Superior and Nipigon than in Michigan and 

 Ontario; in nigripinnis they are longer in Huron and Nipigon than in Michigan 

 and Superior; in Tciyi they are longer in Superior and shorter in Ontario than in 

 Michigan and Huron; in hoyi they are longer in Nipigon, Superior, and Ontario than 

 in the other lakes; in artedi the predominant race of Nipigon has longer pectorals 

 than the predominant races in the other lakes; in quadrilaterale the pectorals are 

 longer, on the average, in Superior and Huron than in Lake Michigan. 



Within the same lake there are differences in the length of the pectorals in the 

 case of the species artedi. The manitoulinus form in Huron has much longer pec- 

 torals than the artedi form, and in Lakes Superior and Erie the albus form has longer 

 pectorals than the artedi form. 



The number of pectoral rays is not distinctive, but some species tend to have a 

 lower number, on the average, than others. Within the species, also, the number 

 may vary. Thus, the hoyi of Ontario and the clupeaforrnis of Erie seem to have a 

 lower average number than their relatives in the other lakes; and johannx in Huron 

 has more than in Michigan. 



The shape of the pectoral often has some value as a systematic character. In 

 typical nigripinnis and in johannx the dorsal margin frequently is decurved and most 

 frequently is relatively straight in other species. 



VENTKALS 



The relative length of the ventrals is of more systematic importance than that 

 of the pectorals. (See Table 11.) They are longest in the Tciyi and shortest in the 

 races of quadrilaterale, artedi, and clupeaforrnis, but only the figures for the Tciyi 

 and the quadrilaterale are quite distinctive. The overlapping between the ranges 

 of the Tciyi and the artedi in the same lake usually is very slight, however. 



The same variation occurs within the species as in the case of the pectorals, 

 though it is not so extensive. In zenithicus the Michigan and Huron races seem to 

 have somewhat shorter ventrals than those from other lakes; in reighardi they appear 

 to be somewhat longer in Nipigon than in the others; in the case of nigripinnis they 

 are somewhat longer in Huron; in Tciyi they are shorter in Ontario and longer in 



