No. 715]  RACIAL AND INDIVIDUAL VARIATION       121 
this work Heincke observed that the differentiation of the 
brackish types of unrelated species had been parallel. 
Several American zoologists worked on variation 
within fish species before the close of the nineteenth cen- 
tury. We may refer to Eigenmann's work (1895) on the 
geographical variations in number of anal rays in 
"Leuciscus" balteatus; to Moenkhaus's papers (1894, 
1896, 1898) on the darters and to Rutter's study (1896) 
of Gasterosteus. 
A great advance in the study of variation in fishes 
came with the appreciation of the importance of race 
analysis in fisheries biology. In fact, the real foundation 
of race investigations in fishes is the classic work of the 
pioneer Heincke (1898), .on the herring of Europe. Many 
other investigators have since worked on this species, for 
example Johansen (1919, 1924), Lea (1919, etc.), Ehren- 
baum (1929), Strubberg (1930), Schnakenbeck (1931) 
and others. Thompson (1917), Hubbs (1925), Fujita 
and Kokubo (1927) and Rounsefell (1930) have made 
similar investigations on the Pacific herring, and several 
Russian ichthyologists are now publishing racial studies 
of both forms. These authors have carried the analysis 
of herring races into significant detail, but many ques- 
tions remain debatable. 
European investigators have analyzed the variation in 
other species of commercial importance. The literature 
is large, but much of it has been reviewed by Redeke 
(1902, 1912, 1915). The work of Duncker (1900, 1913, 
etc.) on flatfishes is among the best, but this author's 
most elaborate variational study (1908) was based on a 
pipefish, Siphonostoma. Fage, to mention only one 
other, has studied variation in Mullus (1909) and in 
clupeoids (1911, 1920). 
A splendid example of parallel raciation was found by 
Thning (1918), in statistically comparing the several spe- 
cies of lantern fishes in the Mediterranean and the At- 
lantic. Ege (1930) obtained similar results in comparing 
the races of two species of Paralepis in the same region, 
as did Chabanaud (1929) with a sole. 
That the remarkable homing habits and consequent