FISHERY BULLETIN: VOL. 78, NO. 3 



and ammonia concentrations that do not produce 

 stressful conditions. 



On 16 September 1971, pink salmon eggs were 

 collected from spawners in Sashin Creek on 

 Baranof Island, southeastern Alaska. The eggs 

 were immediately fertilized, water hardened, and 

 placed in Heath^ trays. On 4 November, we trans- 

 ported the eyed eggs from Sashin Creek to Auke 

 Creek near Juneau, Alaska, and on 16 November, 

 the eggs were placed in eight gravel incubators at 

 five seeding densities— 0, 1,600, 6,400, 12,800, 

 and 25,600 eggs /incubator (Table 1). 



Each incubator (inside measurements, 30 cm x 

 30 cm X 30 cm, Bailey and Heard 1973) contained 

 0.015 m^ of gravel. A 25 mm layer of bird's-eye 

 gravel (particle size, 2-4 mm) covered the sides and 

 bottom. The remainder of the gravel was larger 

 (particle size, 13-32 mm). We installed airtight 

 covers on the incubators to prevent exchange of 

 gases between atmosphere and water. Water was 

 introduced into each incubator from the bottom in 

 an upwelling flow of 0.8 1/min (apparent velocity, 

 53 cm/h). 



Numbers of eggs were estimated by displace- 

 ment (Burrows 1951). Precision of the seeding 

 densities, given by ±2 times their estimated stan- 

 dard error (Table 1), was based on appropriate 

 expansion of variation in egg counts from ten 100 

 ml samples. In previous studies of incubation at 

 this hatchery (Bailey and Taylor 1974), the eggs 

 hatched in late December or early January, 100- 

 120 d after fertilization; and the fry emerged in 

 April, 205-225 d after fertilization. In this study, 

 we expected the eggs to hatch and the fry to 

 emerge at similar times. 



Oxygen and total ammonia concentrations were 

 measured weekly between 3 December 1971 and 

 11 April 1972. Dissolved oxygen concentrations 

 were measured with the Winkler method to the 

 nearest 0.01 mg/1. Total ammonia (NHg -I- NH4^) 

 in the water was measured with an autoanalyzer 

 using a procedure modified from that of the U.S. 



Environmental Protection Agency (1974). Our 

 modification used a larger capacity heating bath 

 and measured total ammonia to within 0.004 ppb. 

 If temperature and pH are known, the amount of 

 NH3 can be calculated from tables by Emerson et 

 al. ( 1975). When calculating periodic estimates of 

 oxygen uptake and total ammonia production per 

 individual, we corrected for the number of fry that 

 had left the incubator. We assumed the number of 

 alevins in the incubator equaled the final total of 

 alevins emerging from the incubator less the 

 number of alevins already emerged. Tempera- 

 tures of incubator effluents were measured daily 

 to the nearest 0. 1° C , and pH was measured twice a 

 week with a standardized Corning model 112 pH 

 meter. Confidence intervals were calculated for 

 each estimate and displayed graphically. These 

 confidence intervals were computed under the as- 

 sumptions of normality of variation and 

 homogeneous variance, the latter holding both 

 among incubators and over observation times. 



The fry were sampled and counted daily (Feb- 

 ruary through April 1972) as they voluntarily 

 emerged from incubators. Samples of fry were pre- 

 served in 5% Formalin for 6 wk. Later we selected 

 three samples of 50 fry from the daily samples of 

 each incubator to represent the days when 

 cumulative fry emergence was 25'7f , 50^7^ , and 75% 

 of the total emergent fry for each incubator. Fry in 

 these selected samples were measured to the 

 nearest millimeter (fork length) and weighed to 

 the nearest milligram after they were blotted with 

 a damp paper towel. Developmental index iK^) 

 was computed to determine efficiency of yolk con- 

 version [Kj) = 10( weight, milligrams)' '/(length, 

 millimeters). Bams 1972]. The /C^ index was com- 

 puted for unfed fry at the time of emergence. A 

 high K,-i indicates a large amount of unabsorbed 

 yolk, whereas a small Kd indicates a small 

 amount of yolk and a more developed fry. The sam- 

 ple of fry at 25% emergence from the incubator 

 seeded with 25,600 eggs was lost. 



^Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



Table 1. — Survival of pink salmon from eyed egg to migrant fry 

 in gravel incubators seeded with indicated number of eggs 



(±2SE). 



Eyed eggs 

 per incubator 



Survival 

 (%) 



Eyed eggs 

 per incubator 



Survival 



1.600 ± 76 

 6,400 ±302 

 6,400 ± 302 



100 

 94 

 92 



6,400 ± 302 

 1 2,800 ± 604 

 25,600 ± 1 .209 



92 



100 



50 



TEMPERATURE, pH, AND 



TOTAL AMMONIA IN 



INCUBATOR EFFLUENT 



Temperature of the water source decreased as 

 the experiment proceeded. Temperature was 

 about 8° C when the eggs were fertilized 16 Sep- 

 tember 1971 (day 0, Figure 1), remained above 7° C 

 until 14 October (day 28), and then gradually 

 dropped to 3.6° C (range, 3.5°-3.8° C) by 16 



650 



