NEILSON and GEEN: CHINOOK SALMON OTOUTH INCREMENT FORMATION 



Figure 3. — Change in otolith microstructure in a Oncorhynchus tshawytscha alevin 

 transferred from a 24-h period temperature cycle (4°C amplitude) to a 12-h period tempera- 

 ture cycle (4°C amplitude). 



50 ^tm 



temperature regimes (P < 0.05, Student- 

 Newman-Keuls test). 



The top two horizontal strata of Figure 4 consti- 

 tute a 3 X 2 factorial design and were examined 

 with a two-way analysis of variance. The effects of 

 time of feeding with respect to the diel tempera- 

 ture cycle, ration level, and their interaction were 

 examined in relation to mean otolith increment 

 width. The effect of time of offering with respect to 

 the diel temperature cycle on mean increment 

 width was significant (P < 0.001), whereas ration 

 level was not (P > 0.05). The interaction of time of 

 offering in relation to the diel temperature cycle 

 and ration level on otolith increment width was 

 also significant (P < 0.001). 



Fish in treatments receiving 2 feedings/24 h or 

 fed once per 24 h and exposed to a 10-min bout of 

 activity produced growth increments whose aver- 

 age widths were significantly less than those of 

 fish in treatments fed the same ration once per 24 

 h. Treatments in which fish received either rations 



of 8% or 4% with 1 or 2 feedings/24 h comprise a 

 2x2 factorial design, and were analyzed with a 

 two-way analysis of variance. Increased feeding 

 frequency significantly reduced mean increment 

 width (P < 0.001), although ration level did not (P 

 > 0.1). The interaction of feeding frequency and 

 ration level was not significant (P > 0.1). 



Widths of otolith increments formed when fish 

 were fed 4% B.W./24 h and subjected to a 10-min 

 bout of activity were not significantly different 

 from widths of increments in fish which received 

 two feedings equivalent to the 4% B.W./24 h ration 

 level ( ^-test, P > 0.05). However, fish fed a ration of 

 8% B.W./24 h with two feedings produced incre- 

 ments whose average width was significantly 

 greater than the latter two treatments (analysis of 

 variance and the Student-Newman-Keuls test, P 

 < 0.01). Mean increment widths in fish fed 4% 

 B.W./24 h and exposed to a constant water- 

 temperature regime were compared with incre- 

 ment widths in fish receiving the same ration plus 



95 



