FISHERY BULLETIN: VOL. 76, NO. 2 



Figure 14. — Relationship between 

 pectoral body depth and standard 

 length of Trachurus symmetricus lar- 

 vae which were fed (open circles), 

 starved 1 and 2 days (dots), and starved 

 3 days (x's). 



6 

 E 



bJ 

 Q 



>- 

 Q 

 O 

 CD 



< 



cr 

 o 



I- 



UJ 

 Q. 



2.5 



30 35 40 



STANDARD LENGTH (mm) 



45 



HL, and 11) bd-2/HL. Standard length was used in 

 ratios and not as a unit to allow discrimination 

 between fed and starved larvae of the same length. 

 To discriminate between the two groups, the 

 analysis selected a set of five variables, listed in 

 the order of selection; 1) bd-2/SL, 2) bd-2/HL, 3) 

 bd-l/HL, 4) ed/SL, and 5) bd-1. With these five 

 variables, 83% of the fed and 86% of the starved 

 larvae in the select size class were correctly clas- 

 sified. When all the larvae, 69 fed and 48 starved 

 (Figure 14), were included in the analysis and the 

 discriminant functions derived from the above five 

 variables were used, 87% of the fed and 94% of the 

 starved were correctly classified. 



Comparison of Histological and 

 Morphological Techniques 



All larvae classified using morphometric 

 characters were also analysed histologically (Ta- 

 ble 1). The histological grading system classified 

 83% of the larvae which were fed into the healthy 

 group and 83% of the larvae deprived of food for 3 

 days into the starved group. The histological 

 grades of 1- and 2-day starved larvae were not as 

 well defined; 19% of the cases were graded as heal- 

 thy, 44% were graded intermediate, and 37% were 

 graded starved. This outcome was similar to re- 

 sults of the initial histological study of 2-day 



starved larvae reported under Histological Re- 

 sults. 



In the morphometric S WD A (Morphological Re- 

 sults), all the starved larvae (1-, 2-, and 3-day 

 starved) were entered into the analysis as one 

 group (n =48). However, from the above histolog- 

 ical grades of the starved larvae, the class should 

 be divided into two groups, early starvation (in- 

 cluding 1- and 2-day starved larvae which re- 

 ceived all three histological grades) and prolonged 

 starvation (including 3-day starved larvae which 

 were mostly graded as starved). Preliminary 

 laboratory experiments supported this decision. 

 Food was offered to 1-, 2-, and 3-day starved larvae 

 and some 1- and 2-day starved larvae did initiate 

 feeding. No 3-day starved larvae ate food. The 



Table l. — Histological classification oi Trachurus symmetricus 



larvae. 



410 



