Table K. — Comparison of mean vertebral counts of juvenile Atlantic menhaden (age 0) from estu- 

 arine nurseries north and south of Long Island, N.Y., with spring and autumn spawners (older 

 ages combined), respectively, of the same year class 



P is less than 0.01. 

 P is less than 0.05. 



involving the 1957 year class) were found to 

 differ significantly (two were significant at the 

 5 percent level and two at the 1 percent level). 

 Differences between means ranged fronn 0.023 

 (1957 year class, autumn spawners) to 0.301 

 (1956 year class, spring spawners). In four of 

 the comparisons, there were increases in the 

 nnean vertebral counts with age; in two, de- 

 creases. 



In most cases, there were statistically sig- 

 nificant differences between the mean vertebral 

 counts of juveniles of a given year class 

 (age 0) inhabiting estuarine nurseries north 

 and south of Long Island, and spring and 

 autunnn spawners of the same year class, 

 respectively, occurring in an adjacent ocean 

 locality. It is, however, evident that the n-iean 

 counts of spawners and of juveniles in the 

 respective localities followed a similar pat- 

 tern; that is, lower mean counts were asso- 

 ciated with spring spawners and juveniles 

 occurring north of Long Island, and, con- 

 versely, higher counts with autumn spawners 

 and juveniles occurring south of Long Island. 

 The observed differences between juveniles 

 and adults could have resulted from (1) changes 

 in mean vertebral counts with age as men- 

 tioned in an earlier section, (Z) recruitnnent 

 of larvae into estuarine nurseries from ocean 

 spawning grounds other than where spawners 

 were collected, and (3) inadequate sampling 

 of the spawning and/or juvenile populations 

 within the respective localities. 



DISCUSSION 



In our studies of the Atlantic menhaden 

 population structure, we are concerned with 

 whether observed differences between groups 

 of fish (for exannple, spring and autunnn 

 spawners) reflect discrete natural biological 



units within the population. Probably the nnost 

 that we can hope to do through morphological 

 studies is to gain sonne insight into the con- 

 sistency of quantitative differences between 

 groups of fish. Such information would lead 

 to a better understanding of population struc- 

 ture. 



Results of studies by June ( 1958) and Suther- 

 land (1963) revealed consistent differences in 

 nnean vertebral counts (as well as other 

 nneristic characters) of juveniles in five suc- 

 cessive year classes (1955-59) inhabiting 

 estuaries north and south of Long Island. 

 Juveniles identifiable with both groups were 

 shown to be living together in Long Island 

 estuaries. June (1958) found differences in 

 nneristic counts of juveniles in Long Island 

 estuaries associated with fish length and 

 hypothesized that the differences reflected the 

 intermingling of progeny from autumn and 

 spring spawnings. Sutherland (1963) noted that 

 fish length was a possible source of variation 

 within samples, but did not consider it in the 

 analysis of his data. Examination of his raw 

 data (table 5) reveals, however, that classifi- 

 cation of individual sannples from Long Island 

 waters, as to northern or southern groups, 

 also can be made on the basis of length. Those 

 samples identified with the southern group 

 comprised fish that were consistently larger 

 than those identified with the northern group. 

 In every instance where both groups were 

 represented in the same estuary, the smaller 

 fish occurred in samples collected late in the 

 season, and, in most cases, there was little 

 or no overlap in the length distributions. 

 There is little evidence, however, that progeny 

 from the autunnn spawning contribute appre- 

 ciably to the juvenile stocks in estuarine 

 nurseries north of Long Island. Numerous 

 attempts to collect larvae and juveniles in 



