FLOl'NDKKS OF (iEXLS PAHALICHTHYS AND KKLATED GENERA 321 



Gill rakers on upper limb 



Class 3 4 5 6 7 



C'hesaiieake Bay freqiipncies -- -- 31 35 5 



North Carolina frequencies 2 4 29 6 1 



Gill rakers on lower limb 



Class 13 11 15 16 17 18 



Chesajjeake Bay frequencies -- -- 9 28 25 9 



North Carolina frequencies 4 2 9 15 7 2 



Total nuntlnr of ijiU rakers on outer arch 



Class 16 17 18 19 20 21 22 23 24 



Chesapeake Bay frequencies -- -- -- - 6 19 15 23 8 



North Carolina frequencies 22126 15 56 .. 



Anal rays 



Class 61 62 63 64 65 66 67 68 69 70 71 72 73 



Chesapeake Bay frequencies -- __ -_ 13 (i 17 12 6 11 3 6 5 



North Carolina frequencies 11.-21768673---. 



Dorsnl rays 



Class 80 81 85 83 84 85 86 87 88 89 90 91 92 93 94 95 96 



Chesapeake Bay frequencies __ ._ 1 ._ 2 7 5 S 12 12 10 4 3 2 3 __ 1 



North Carolina frequencies 1 1^. 4-- 1 3 4 8 9 4 1 2 2^. 1_. 



The (lata tahulated show that there is one con- The diveijxence of the races of drnfafux is such 



sisteiit and statistically measurable character tliat the popiUation from North Carolina more 



which may be used in racial studies, namely, the nearly approaches that of alhigntta than speci- 



number of jjill rakers. There is only a slight dif- mens from Chesapeake Bay approach that species, 



ference in the frequency distribution of the fin ray When dentatm from Chesapeake Bay is compared 



counts, which is somewhat more pronounced in with alhigvtta there is a comi)aratively wide gap 



the anal ray count. However, while the difference between them as far as the total number of gill 



in the anal count is not very significant, it is of rakers is concerned, and individual specimens of 



considerable practical value in identification, since the two species may be readily distinguished on 



overlapping specimens at the extreme end of the that basis; but this structural gap disappears be- 



frequency distribution disappear in northern tween the two species at North Carolina (compare 



specimens, and denfatm from Chesapeake Bay is t'le tabulated racial data of dentatm with tables 



thus more readily separable from alljUjutta. 2 to 4). To a le.sser extent this is also true of the 



There is practically no difference in the scale count '"'"' ''^•^"^- 



which is, on that "account, omitted from the pre- \ *« ^o be noted also that there is a difference 



ceding tabulations by locality. The racial differ- "; ^^^^ regularity of the frequency distributions of 



. * , ,  X 1 1 the number of gill rakers in the two populations 



ence in proportional measurements, shown in table j, , , ^ rA, i^i i tj i ^■ 



' ' ,. . „ . , ,-, of dentatxis. Ihe Chesapeake Bav i)opulation is 



8, ... g.'o„ps of specimens of app.-oximately ike ^^^^ ^^^^.^^^.^^^ .^^ .^^ structural 'charac'tcrs. the 



sizes, is not pn.nouiued. The No.-th Carolina ^^^^.^^^^^j _^^^^^ ^^^.^^^ ^^ ^,^^ frequency distribution is 



populat.on of dentatuH. m gene.al. is somewhat ,..,^5^^^. compact and regular; while the North 



deeper bodied and has a slightly longer head and Carolina population is more variable, the spread 



maxillary. These differences do not hold in all i„,j,|j, „„)i.e extensive, and the frequency polygon 



the size groups, and it is possible that curves rep- ^-j]] assume a skewed foi-m. 



resenting the .-elative changes with size in these //y?;/7V//.v//(.— The somewhat irregular fivquency 



measurements will have somewhat different forms. distribution of dint't/iis from Xo.lli Carolina may 



bill not enough specimens have been measured to jxissilily be (inc to the p.-escnce of hybrid specimens 



draw any definite conclusions. which ai'e iiilcriuediate in some respects. This 



