TWENTY-EIGHTH BIENNIAL REPORT. 



59 



tOOOibs 



300OUH 



lOCOibi 



60° 



i5° 



50' 





Co/th e^ oibo^ore Jufy 24 to Ochib^r f2 J$/S oon^parm<t rififh aat/f immmum t*rnp*rotvre af Santa 

 Barbara for /tie torn 9 p*rto^ 



Fig. 12. The catch of albacore varies with the temperature. In this 

 graph, temperature data have been set back three days so that that 

 for July 27, for instance, appears on the same ordinate as does 

 the catch recorded for July 24. 



tiiation.s occur, whctlier in abundance or in (lualitv; {?>) the foretelling 

 of these fluetnation.s ; (4j the analysis of whatever migrations occur, in 

 order that the interdependence of fish of different regions may ])e 

 known, and (5) the foundation of measures for protection in case 

 dei)letion should occur. 



Of these purposes, the second, although naturally not of greater 

 importance, must precede all the others. It would he impossible to 

 detect depletion, were the natural changes unknown; antl, as is made 

 very clear in the law under which we work and in our published papers, 

 the conservation of our fisheries must be set above all else. The 

 thorough understanding of the natural changes leads to their foretelling 

 and to the as.sigTinient of the proper causes for instances of decline in 

 abundance.! The analysis of the migrations is necessary primarily in 



Fig. 13. Comparison of the amount of barracuda, white sea bass, and yellow- 

 tail caught by purse seiners and that taken by other boats. Black : 

 Weekly landings by purse seiners at San Pedro, 1922. White: Weekly 

 landings by other boats at San Pedro, 1922. 



order that the extent of the effect of overfishing or of natural over- 

 abundance in a given locality upon other localities, may be .determined. 

 It is nevertheless true that the discovery and correct definition of the 

 great natural fluctuations in abundance is an end in itself. Thev mast 



