FISHERY BULLETIN: VOL. 73, NO. 4 



1966. Individual food items were measured to the 

 nearest 0.01 mm of body length and diameter with 

 an ocular micrometer to determine volume. Fry in 

 the estuary were collected with a dip net from a 

 skiff and by floating traps anchored near the 

 shoreline. The dip nets and traps were effective in 

 collecting fry less than 60 mm long, which are the 

 subject of this report, but were not effective in 

 collecting larger salmon. The larger fish were able 

 to evade capture by sounding. Most of the fry 

 examined for stomach contents were collected 

 from the estuary during daylight (1100 to 1500). 

 On three occasions, however, fry were collected 

 during nights (0230) when the sky was overcast or 

 moonless and incident light intensity was 0.0 

 footcandle near the water surface. No stomachs 

 were collected during bright moonlight nights, 

 which were rare. 



To estimate the volume of water grazed per day 

 by fry, we measured velocities of water currents 

 close to shore-oriented fry while observing their 

 behavior in relation to the current and food items. 

 Current velocities close to shore-oriented schools 

 of fry were measured by two methods. One method 

 was to record the time it took suspended particles 

 in the water to drift 1 to 5 m along a floating 

 anchored line graduated to 0.1 m. The second 

 method was to measure the velocity by holding a 

 current meter near a school of fish; the meter was 

 attached to the end of a rod about 4 m long. The 

 current meter dial was calibrated to read to the 

 nearest 3 cm/s. Both methods required the ob- 

 server to operate either from an anchored skiff or 

 from shore. Polaroid glasses were used to reduce 

 glare from the water surface and improve 

 visibility. 



Feeding at night by fry was tested in two 

 experiments in an aquarium with known densities 

 of zooplankton. The aquarium consisted of a 7-mil 

 plastic hag suspended in the estuary from a float 

 and containing 76 liters of seawater. The fry were 

 captured in the outer bay and held in a 1-m no. 10 

 mesh (158-ju,m openings) plankton net for 20 h to 

 deny them food and to ensure that their stomachs 

 were empty. The starved pink and chum salmon 

 fry together with a known quantity of zooplank- 

 ton were then placed in the aquarium and held 

 under various light intensities. After tim.ed inter- 

 vals in the aquarium, fry were removed, killed, and 

 their stomach contents removed. 



In the first experiment, fry (length, 32 to 41 mm) 

 in groups of five were placed in an aquarium that 

 contained 240 zooplankters per liter of seawater. A 



cursory examination of the zooplankters revealed 

 that they were predominantly copepods and bar- 

 nacle nauplii. Each group of fish was allowed to 

 feed 13 to 28 min before being removed and 

 preserved in 10*? Formalin solution. The 

 experiment was started in the evening before 

 sunset and continued until the light meter read 0.0 

 footcandle. In the second experiment, 14 fry were 

 placed in an aquarium containing about 260 

 zooplankters per liter; they were kept there for 4 h 

 and 20 min at night before they were removed and 

 preserved in Formalin. The light meter read 0.0 

 footcandle throughout the experiment. 



The time required for fry to evacuate their 

 stomach contents was determined experimentally 

 at 8.5°, 10.0°, and 12.8°C. The procedure was to 

 capture 200 to 300 salmon fry in the estuary and 

 place them in strained seawater in a floating cage 

 of no. 10 mesh plankton net, which prevented 

 entry of prey from the surrounding water. At the 

 start of each test, 10 pink and 5 chum salmon were 

 killed and preserved; at hourly intervals 

 thereafter 5 fish of each species were killed and 

 their stomach contents examined until all 10 fish of 

 two successive samples contained no food in their 

 stomachs. The pink salmon fry examined in these 

 tests ranged from 32 to 57 mm in length and the 

 chum salmon from 34 to 54 mm. Water tempera- 

 tures were recorded by a thermograph to the 

 nearest 1°C. The sensing probe of the ther- 

 mograph was located 1 m below the water surface. 



The zooplankton in Traitors Cove was sampled 

 only in 1965 and 1966 while fry were in the bay. A 

 5-inch Clarke-Bumpus sampler with a no. 10 mesh 

 net was towed at a depth of about 0.5 m until 

 about 50 liters of water (2 to 10 s) had been 

 strained. Seventy-nine samples were collected-in 

 1965, 7 stations in the outer bay were each sampled 

 in 1 day; and in 1966, 10 stations in the inner bay 

 and 14 in the outer bay were each sampled on 3 

 different days (Figure 1). Only one sample was 

 taken at each station. The zooplankton catch was 

 preserved in 5% buffered Formalin solution. The 

 plankton samples were subsampled by two 

 methods for analysis. In the first method, each of 

 the 79 samples was analyzed from 1-ml sub- 

 samples (approximately 1/100 of the sample) 

 placed in a Sedgewick-Raf ter chamber. The kinds, 

 numbers, and size of the various plankters were 

 determined. Volumes of the different plankters 

 were computed from lengths and average 

 diameters, assuming a cylindrical shape for each 

 plankter. In the second method, all 7 of the 1965 



848 



