FISHERY BULLETIN: VOL. 74, NO. 2 



Table l. — Laboratory experiments of rearing the northern anchovy at the Southwest Fisheries Center, La JoUa, Calif. 



Source 



Hunter (1976) 



Kramer and Zweifel (1970) 

 Lasker et al. (1970) 



Leong (unpubl. data)' 



Paloma (see text footnote 3) 



Schumann-I (G. O. Schumann unpubl. data)^ 



Sctiumann-ll (Kramer and Zweifel 1970) 



Scfiumann-lll (G. O. Schumann and A. Saraspe unpubl. data)^ 



Theilacker and McMaster (1971) 



'Pers. commun.. Southwest Fisheries Center, La Jolla, Calif. 

 ^Data are on file at the Southwest Fisheries Center, La Jolla, Calif. 



In all of these experiments the fish were from 

 the southern California stock (Vrooman and 

 Smith 1971), reared at laboratory ambient water 

 temperature, and not subjected to experimental 

 treatment or excessive handling. All fish sampled 

 for measurements were sacrificed. The length 

 measurement is standard length. 



TREATMENT OF DATA 



The age of anchovies reared by Schumann-II 

 and Schumann-Ill were known because the an- 

 chovies were hatched from eggs at the start of the 

 rearing experiments. In Leong's {pers. commun.) 

 experiment, the exact age of his fish was not 

 known because juvenile fish of average length of 

 88.3 mm were used at the start of the experiment. 

 We estimated the age of Leong's fish from data 

 from Schumann-I in which anchovies were reared 

 for 97 days from an average length of 18.0 to 81.9 

 mm (Table 1), and data from Schumann-Ill which 

 indicated that an 18.0 mm fish, raised from eggs, 

 was about 30 days old. Our age estimate is 4 mo. 



Several mathematical models describing 

 growth of organisms are available (e.g., Parker 

 and Larkin 1959; Richards 1959; Laird 1969). 

 The commonly used models in fisheries are the 

 exponential, the von Bertalanffy, and the Gom- 

 pertz models (Beverton and Holt 1957; Silliman 

 1969). The Gompertz model was selected for our 

 study because it was shown by Kramer and 

 Zweifel (1970) to be better than the exponential 



model for describing growth of laboratory-reared 

 anchovy larvae and because it generally de- 

 scribes the growth of fishes fairly well. Also, pre- 

 liminary analysis of our data indicated that the 

 von Bertalanffy model poorly described the 

 growth of young fish. 



The Laird version of the Gompertz growth 

 model (Laird 1969) describes an asymmetric sig- 

 moid curve of the form. 



L, = L exp {C [ 1 - exp i-at )] } 



where Lq = length at zero age or hatching 

 C = a constant 

 a = rate of decay of exponential 



growth 

 t = age in months. 



This model was fitted to our data using an itera- 

 tive least squares procedure (Conway et al. 1970). 

 Our goal was to describe growth on a coarse 

 time scale, i.e., monthly rather than on a fine 

 time scale, i.e., daily. 



GROWTH FROM HATCHING TO 

 JUVENILE STAGE 



The Gompertz growth model and an exponen- 

 tial growth model were applied to data of Schu- 

 mann-II by Kramer and Zweifel (1970). Both 

 models described the data from Schumann-II 

 reasonably well, although the Gompertz model de- 



272 



