FISHERY BULLETIN: VOL. 74, NO. 1 



similar observations led Neave (1955) to com- 

 ment that migrating chum and pink salmon fry, 

 failing to reach the ocean in a single night, hide 

 during the day and resume migration at night- 

 fall. Hiding behavior disappears in coho salmon 

 fry at time of complete yolk absorption but is re- 

 tained for several days at high light intensities 

 (Hoar 1958); this suggests a threshold intensity 

 for the avoidance response which increases as the 

 alevin stage proceeds. 



Concurrence between change in numbers of fry 

 observed in the choice chambers, a collective re- 

 sponse, and the accumulated number of emergent 

 sibs could reflect either a sudden shift in photo- 

 response of individual fry or gradual erosion of 

 the negative response occurring simultaneously 

 in all fry. The sudden shift alternative is best 

 supported by three patterns of behavior noted in 

 the choice chambers. Individual fry were ob- 

 served to spend considerable time in the light 

 compartment upon entering it, alternately 

 swimming about slowly and remaining locally 

 quiescent. Positions were commonly adopted with 

 the head projecting into the light compartment 

 (Figure 1), or entrance, and departure was rapid, 

 irrespective of the presence or absence there of 

 other fry until the last few days of testing when 

 aggression was observed (Figure 6). 



Despite near depletion of vitellus at time of 

 emergence, the shift in photoresponse did not ap- 

 pear to be due to starvation because the response 

 was not altered significantly by feeding. This is of 

 interest as Smith (1952) reported marked meta- 

 bolic changes in rainbow trout, S. gairdneri, ale- 

 vins a few days prior to emergence, suggesting 

 that these physiological events signified the onset 

 of starvation. The change in photobehavior ap- 

 pears to be an ontogenetic behavioral change 

 normally associated with emergence from the 

 redd rather than one instigated by nutritional 

 deficiency, premature feeding, or light experi- 

 ence. It remains unclear as to whether or under 

 what conditions such stimuli can modify this 

 change significantly; however, under hatchery 

 conditions, Harvey (1966) found that sockeye 

 salmon fry took food 2 wk after hatching but that 

 emergence of fry from a simulated redd coincided 

 with complete yolk absorption some 3 wk later. 

 Heard (1964) noted that most emerging sockeye 

 salmon fry trapped from natural redds in an 

 Alaskan stream contained little or no yolk, re- 

 mained photonegative, and emerged primarily 

 during hours of darkness. 



The timing of the photoresponse change rela- 

 tive to emergence and yolk reserves may vary 

 within common limits for most stream salmonids 

 and differences may reflect species-specific adap- 

 tions of value to fishery biologists. As in the fry 

 emerging from the simulated redds, the yolk re- 

 serve of coho salmon fry emerging from natural 

 redds averaged 7% (unpubl. data). Stuart (1953) 

 observed a definite change in photoresponse of S. 

 trutta when yolk neared depletion, and the photo- 

 response change was employed by Gray (1929b) 

 to denote the conclusion of incubation when 

 measuring the effect of temperature on alevin 

 size at time of yolk depletion. Woodhead (1957) 

 disagreed with Stuart as to the timing of the 

 photoresponse change in S. trutta, and asserted 

 that it occurred coincident with maximum activ- 

 ity of the alevin 15 days after hatching when yolk 

 reserve constituted 70% of the dry weight of the 

 fry. This considerable difference in timing re- 

 mains unresolved. 



Denying the photoresponse fry streambed ex- 

 perience during the last few weeks of the alevin 

 stage had no apparent effect on the final size of 

 the fry, probably due to their advanced stage of 

 development prior to application of treatment dif- 

 ferences. Marr (1963, 1965) has shown that de- 

 velopmental efficiency is reduced by exposure to 

 natural light or lack of substrate contour which 

 stimulate locomotor activity at the expense of 

 growth. However, marked effects on locomotor ac- 

 tivity were only measurable until development 

 was 75-80% complete. The weight disparity be- 

 tween experimental and control groups of fry 

 (Table 4) which was the outcome of weight loss or 

 reduced weight gain is presumed to be an out- 

 come of reduced feeding opportunity. 



In summary, the present results show that coho 

 salmon fry underwent a definite shift (sudden or 

 otherwise) from a strong to a weak negative 

 photoresponse. This shift was accompanied by a 

 positive response to water current leading to pre- 

 ferred movement upstream. The emerging fry 

 was an actively feeding animal yet to fill, or in 

 the process of filling, its air bladder, fed in the 

 gravel prior to emergence, and emerged when av- 

 erage yolk reserves declined to 7% of total dry 

 weight. In contrast to fry emerging from natural 

 redds (Koski 1966), later-emerging fry were 

 larger than those emerging earlier and may have 

 derived from larger eggs. Because first-emerging 

 fry held ecological advantage over later-emerging 

 fry in stream aquaria (Mason and Chapman 



174 



