and prepelvic length of larval and juvenile 

 sardine . In comparing relatively small samples 

 from southern California and Lower California 

 he found: (1) the head of larvae and juveniles 

 from Lower California tends to be longer, deep- 

 er and have relatively bigger eyes than those 

 from southern California. This tendency is 

 eventually reversed, so that the adults from 

 southern California have relatively longer heads 

 than adults from Lower California. (2) The 

 predorsal length is shorter in young fish from 

 Lower California, but about the same in adults 

 from the two localities . (3) There is only a 

 slight tendency for fish from southern Californ- 

 ia to have a longer preanal distance. (4) The 

 prepelvic length is slightly greater in sardines 

 from Lower California. These findings are of 

 interest, but more intensive work is needed to 

 determine whether or not these differences may 

 be characteristic of different subpopulations . 



Dr. Wilhelm Harder, Institut fllr Fischer- 

 eibiologie, Universit&t Hamburg, is engaged in 

 a comparative study of the visceral morphology 

 of clupeoids . While these particular findings 

 have not yet been published, he has kindly in- 

 formed me that there are certain differences in 

 sardines from different localities. Most out- 

 standing among these is the length of the small 

 intestine in relation to standard length. At fish 

 sizes of 100-150 mm. the small intestine is 

 longer in sardines collected off British Columbia 

 than in those collected off southern California 

 and Lower California. At fish sizes of 175-225 

 mm. sardines collected off British Columbia 

 have shorter small intestines than do fish from 

 more southerly localities. In other words, al- 

 though the small intestine of British Columbia 

 sardines is longer at fish size 100 mm., its 

 rate of increase relative to fish size is less 

 than in fish from southern California and Lower 

 California . 



The cause of these differences and their 

 physiological implications are as yet unknown . 



A revision of the genus Sardinops is now 

 in progress by R. W. Wisner, Scripps Institu- 

 tion of Oceanography. This study employs the 

 conventional form of analysis of morphometric 

 and meristic characters. It does not now seem 

 likely that this revision will include the study of 

 possible subpopulations of S . caerulea . 



Meristic studies: (1) Vertebral numbers: 

 Prior to the work of Schmidt (1917), average dif- 

 ferences in meristic characters, such as vertebral 

 numbers, were regarded as indicative of genetic 

 differences between subpopulations of fishes. 

 Following Schmidt's experiments and observa- 

 tions in nature by Hubbs (1934) and others, it 

 was realized that while these characters undoubt- 

 edly have genetically determined limits they 

 show great variability within these limits. This 

 variability is presumably influenced by variations 

 in environmental conditions, among which tem- 

 perature has been most frequently mentioned. 



Although this general problem would seem 

 almost ideally susceptible of experimental 

 analysis, very little progress has been made, 

 owing in part at least, to the difficulty of rearing 

 marine fishes. The recent experiments of Taning 

 (1944, 1946 and 1952) on sea trout, Salmo trutta 

 trutta, will give impetus to such studies. His 

 work indicates that the influence of temperature 

 on vertebral number in sea trout is not a simple 

 one and that other factors, such as oxygen pres- 

 sure and carbon dioxide pressure, may also 

 influence vertebral number . 



Similar experiments on sardines are 

 needed and will be carried out when methods of 

 rearing marine fishes are available. It seems 

 a reasonable a priori assumption that the varia- 

 tion in vertebral number exhibited by sardines 

 will prove to be largely phenotypic . Evidence 

 for this is offered by Clark (1947) and by McHugh 

 (1950). 



A number of studies, or continuing 

 studies, have been made of variations in vertebral 

 numbers of sardines. These include the worK of 

 Hubbs (1925), Thompson (1926), Hart (1933, 1934), 

 Clark (1936, 1947), Taranets (1937), McHugh 

 (1950), and California Marine Research Commit- 

 tee (1950). As early as 1934, Hart (1934:H67) 

 concluded that "In view of the fact that the re- 

 sults are inconclusive in spite of the large amount 

 of data collected, i t would appear that other 

 methods of studying this problem must be 

 developed." 



Clark (1947) has gathered and reported 

 upon all available data (slightly more than 50,000 

 counts made on specimens from various local- 

 ities from Alaska to the Gulf of California) . 



110 



