OCONNELL and ZWEIFEL: FEEDING OF PACIFIC MACKEREL 



Table 2. — The average fish weight and the mean (Z) 

 and standard error (SE) of the standardized number of 

 Artemia in the digestive tract for each trial. 



Density 



Number/ 

 lifer 



Feeding 

 duration 



Number 

 fish 



Average 

 fish weight 



Number in 

 digestive tract 



SE 



22 



66 



112 



min 

 2.5 

 5 



10 



30 



60 



90 



2.5 



5 

 10 

 30 

 60 

 90 



2.5 



5 

 10 



30 

 60 

 90 



2.5 



5 

 10 

 30 

 60 

 90 



2.5 



5 

 10 

 30 



S 



125 

 130 

 136 

 134 

 135 

 150 



149 

 148 

 149 

 123 

 157 

 149 

 136 

 159 

 141 

 135 

 165 

 152 



159 

 139 

 158 

 174 

 158 

 168 



145 

 155 

 139 

 147 



149 



299 



572 



2,280 



3,430 



5,620 



254 



695 



1,530 



4,636 



5,790 



11,325 



745 



2,013 

 4,399 

 10,893 

 15,664 

 18,079 



1,502 

 4,041 

 5,232 

 9,863 

 12,676 

 15,723 



3,079 



3,617 



7,629 



15,845 



60 

 168 



86 

 373 

 342 

 216 



37 



81 



163 



583 



684 



1,204 



263 



101 

 270 

 790 



648 

 1,143 



197 

 409 

 858 

 990 

 484 

 605 



936 



447 



982 



2,600 



Nf = number of organisms in the di- 

 gestive tract after t minutes of 

 filter feeding, 

 D = the nominal density of food in 

 numbers per liter, 

 Nx = asymptotic number of organisms 

 in the digestive tract at full ca- 

 pacity, 

 a = specific feeding rate, and 

 f3 = instantaneous feeding rate. 



Fitting all of the data in the two low density 

 series with equation (1) and those in the three 

 high density series with equation (2) resulted 

 in a satisfactory fit for the low density group 

 but not for the high density group. Calculated 

 values tended to be lower than trial values for 

 the 22 Artemia/liter series, and higher than trial 

 values for the 66 Artemia/liter series, particu- 

 larly for the longer time periods. The difl^culty 

 arises from the fact that average quantities in 



the digestive tracts were lower for the 66 Ar- 

 temia/\iter level than for the 22 Artemia/liter 

 level in the 30-, 60-, and 90-min trials. The rea- 

 son for this is not known, but examination of the 

 quantities in the anterior and posterior halves 

 of the intestine for all trials (Table 3) offers a 

 plausible explanation. 



Table 3. — Average number of Artemia in anterior (A) 

 and posterior (P) halves of intestine for each trial. 



Time 



Density 



1 



22 



66 



112 



2.5 A 

 P 



5.0 A 

 P 



10 



30 



60 



90 



A 

 P 

 A 

 P 

 A 

 P 

 A 

 P 





 



0.3 

 



1.8 

 

 50 

 



0.7 

 

 141 

 4 



15 







15 







50 







51 







165 



45 



209 



339 



9 

 



108 

 



57 







320 



36 

 830 

 720 

 890 

 385 



33 







101 







215 



10 



1,077 



327 



569 



714 



937 



722 



300 







175 







467 



59 



1,265 



245 



At 22 Artemia/Uter the maximum quantity 

 in the intestine is reached at 60 min, and at 66 

 A/"^ew/a/liter the maximum is reached at 30 min. 

 Fluctuations in the two halves of the intestine 

 thereafter suggest posterior movement of ma- 

 terial and intermittent elimination. This is cor- 

 roborated by visual observations made during 

 the trials. The earliest detected defecations were 

 at 50 min for the 22 Artemia/liter level and at 

 about 30 min for the 66 Artemia/liter level. 

 No defecation was detected in the trials at 112 

 Artemia/liter, which did not go beyond 30 min. 

 The quantities in the intestine suggest that rate 

 of movement of material into the posterior part 

 of the intestine approached the maximum at 66 

 Artemia/liter and that defecation might not 

 start any sooner at 112 Artemia/liter than at 

 66 Artemia/liter. 



From these data it is reasonable to suppose 

 that beyond 30 min the time-specific losses by 

 defecation would be greater for the 66 Artemia/ 

 liter series than for the 22 Artemia/liter series, 

 and negligible for the two lower density levels. 



On the supposition that the greatest under- 

 estimates of total amounts consumed occurred 

 in the three longest time periods of the 66 Ar- 

 temia/liter series, equation (3) was refitted to 



977 



