100 



74. Detailed analysis from particular view points should 

 also be made of the great mass of statistics entombed in 

 the records of the fish- curing yards ; among others I would 

 suggest that the weekly catches of sardines for all years 

 available be tabulated separately in the case of three or four of 

 the principal yards on the West Coast, that a comparative 

 (statement be made out showing the date of appearance in 

 quantity during normal seasons of the different species at the 

 selected centres both on the East and West Coasts, and that a 

 similar one be prepared showing the dates they leave the coast 

 on their minor migration to deeper water. Statistics of 

 the total catches of sardines on the West Coast must be 

 inclusive not only of the quantities used for food but also of the 

 very variable surplus turned into fertilizer and oil, without 

 which the figures do not exhibit sufficiently emphatically 

 the enormous fluctuations so frequently characteristic of this 

 fishery. In making out tables the official year beginning 1st 

 April will be most satisfactory as well as convenient, as the 

 sardine season ends about this date ; statistics based on calendar 

 years are much less useful in the casj of sardines, as large 

 catches are often made after the end of December, which, 

 under this system of tabulation, are wrongly credited to the 

 subsequent season tending to reduce the disparity shown when 

 an unsatisfactory season follows a highly successful one. 

 Pending the preparation of such statistics, I give below a table 

 showing the fluctuations in the catch of sardines (all species) 

 and also for mackerel for the years 1896 to 1907 as indicated 

 bv the quantities cured in Government fish vards. Unfor- 

 tunately it suffers from the defect alluded to above, being 

 made out for calendar years and not for fishing seasons ; neither 

 does it take account of the quantities manufactured into 

 fertilizers, defects which will be remedied in future tabulations. 

 As a consequence the sardine fluctuations are less clearly 

 depicted than they are in actuality. The statistics for mackerel 

 indicate annual fluctuations more precisely as it is rare for this 

 fish to be used otherwise than as food ; here again the 

 character of successive fisheries will possess more value if the 

 tabulation be concerned with fishing seasons and not calendar 

 years. 



V 



75. The enormous drop in catch from the huge totals of 

 ] 896-1897, long remembered as the last of a series of abundant 

 seasons, is brought out clearly ; a long period of more or less 

 lean years succeeded, lasting till 1907 Avhen the shoals returned 

 once more in enormous abundance. It would be rash to endea- 

 vour to draw deductions from the curve shown pending the 



