FISHERY BULLETIN: VOL. 75, NO. 1 



number of days representing each cruise is large 

 and spawning almost certainly was not uniform 

 within each cruise period. This may account for 

 the more than fivefold difference in biomass esti- 

 mated during the 2 yr. On the other hand it is 

 possible that biomass did differ greatly between 

 the 2 yr. This is especially possible because the 

 eastern Gulf may be an open-ended system with 

 regard to round herring habitat. Round herring 

 eggs and larvae were abundant in the north- 

 central Gulf (Fore 1971) indicating that a large 

 adult population is present there. If a single popu- 

 lation of round herring inhabits the Gulf, the part 

 found in the eastern Gulf might vary from year to 

 year. 



The area adjustments that had been made for 

 two 1971-72 cruises, to account for part of the 

 spawning area not being sampled, affected the 

 biomass estimate in that spawning season. With- 

 out adjustments the biomass estimate was 

 685,273 metric tons rather than 717,815 metric 

 tons. The effect of adjustment was to raise the 

 estimate by more than 32,500 metric tons. This is 

 only a 4.7% increase in estimated biomass. 



It is unlikely that round herring biomass is as 

 great as 1 million metric tons in the eastern Gulf of 

 Mexico, but it probably is considerably in excess of 

 100,000 metric tons. Confidence limits, at the 0.95 

 probability level, based on the annual spawning 

 variance estimates (Table 5) placed the probable 

 range of round herring biomass between 517,470 

 and 918,160 metric tons in 1971-72 and between 

 45,430 and 216,840 metric tons in 1972-73. 



Method II 



The daily spawning estimates for each of the 

 three cruises during 1971-72 and 1972-73 were 

 plotted against their cruise middates (Figure 8). 

 Areas under the resulting polygons were deter- 

 mined and were equated to annual spawning (Ta- 

 ble 6). This method is like that outlined by 

 Simpson (1959). Biomasses were calculated using 

 Equation (7). 



Biomass estimates were 698,045 metric tons in 

 1971-72 and 130,995 metric tons in 1972-73 (Table 

 6). These estimates are similar to those obtained 

 by Method I. 



Method III 



If spawning follows a normal distribution dur- 

 ing the period 15 October to 31 May, then each 



15 Oct l5Nov l5Dec l5Jan l5Feb l5Mar l5Apr l5May 

 CRUISE MIDDATE 



FIGURE 8. — Round herring egg abundance estimates in the 

 eastern Gulf of Mexico based on three cruises in 1971-72 and 

 1972-73. Each symbol represents the estimated daily spawning 

 at the middate of a cruise. The area enclosed by the polygons is 

 an estimate of the total spawning by round herring during each 

 of the seasons. 



TABLE 6. — Annual spawning and biomass estimates for round 

 herring from the eastern Gulf of Mexico during 1971-72 and 

 1972-73 spawning seasons. Estimates are based on the method 

 described by Simpson ( 1959). 



1971-72 



1972-73 



1 .034.852 



194.200 



698,045 



130.995 



cruise within that 229-day period can be rep- 

 resented as some proportion of the area under a 

 normal curve with standard deviation of 38.17 

 days. Saville (1956, 1964) discussed use of the 

 technique for a single cruise near the peak of the 

 spawning season, but I have applied it (Equation 

 (8)) to eight representative cruises during four 

 round herring spawning seasons (Table 7). The 

 observed variation within a season on annual 

 spawning and biomass estimates is great. Al- 

 though spawning is heaviest near the middle of 

 the spawning season (Figure 8), it probably does 

 not follow the normal distribution closely. It seems 

 that in most years spawning intensity increases 

 rapidly to near peak level during late November 

 and then gradually decreases during spring 

 months. Deviations from normality would cause 

 large estimating errors, especially for cruises that 



78 



