HEARD: PROBABLE CASE OF STREAMBED OVERSEEDING 



In addition to the increasing oxygen require- 

 ments due to growth and development of Hve em- 

 bryos, Brickell (1971) found that rates of oxygen 

 consumption by dead intact pink salmon eggs ex- 

 ceeded those of early stage live eggs fourfold: 

 0.0018 mg Og/whole dead egg per h versus 

 0.0004-0.0005 mg Oa/T-day-old live egg per h. He 

 noted even gi'eater oxygen consumption for dead 

 eggs when the chorion was pierced or slit or the 

 egg was fragmented: mean oxygen consumption of 

 fragmented dead eggs in constant-flow cylinders 

 was 0.017 mgOs/eggper h, which exceeds the rate 

 Wickett (1954) found for 7-day-old live alevins. It 

 follows that alevins that die shortly after hatch- 

 ing, because of their soft, exposed, and readily 

 oxidizable tissue, would have higher rates of oxy- 

 gen consumption than whole intact dead eggs, live 

 eggs, or early stage live alevins. 



These increases in oxygen consumption upon 

 death of developing pink salmon embryos are the 

 rationale for suggesting a "snowball effect" — 

 rapidly increasing deaths of embryos once lethal 

 oxygen concentrations were approached. With 

 high densities of live embryos already placing ex- 

 cessive demands on the oxygen and each death 

 increasing the demand, the resulting heavy mor- 

 tality could have caused fry production to plunge 

 below that expected from lower initial egg 

 densities — an excellent example of Neave's ( 1953) 

 theory of compensatory mortality. 



Disappearance of Dead Eggs Versus 

 Disappearance of Dead Alevins 



To test the hypothesis that dead alevins disap- 

 pear within the streambed more rapidly than dead 

 eggs, I conducted a small study in Sashin Creek in 

 the winter of 1968-69 to consider the relative per- 

 sistence of dead eggs and alevins in the streambed. 

 A series of Vibert boxes (small plastic perforated 

 containers), each containing a mixture of 

 streambed gravel, 10 dead eggs, and 10 dead ale- 

 vins (all from 1968 brood year pink salmon) were 

 buried in Sashin Creek on 14 December 1968. The 

 boxes were buried about 20.3 cm deep across a 

 riffle in the middle study area. At irregular inter- 

 vals, pairs of the boxes were removed from the 

 streambed and the contents were preserved for 

 examination. 



Alevins disappeared from the Vibert boxes at a 

 much faster rate than eggs (Table 12). Fewer than 

 half of the original number of alevins were still 

 recognizable at the end of 2 wk; after 37 days only 



Table 12. — Contents of Vibert boxes with dead pink salmon 

 eggs and alevins buried in Sashin Creek streambed between 14 

 December 1968 and 14 April 1969.' 



'Each box originally contained 10 dead eggs and 10 dead alevins. Two boxes 

 were removed on each sample date and the contents combined tor reporting. 



^Of all insect larvae recovered, 80°o were Plecoptera. 16°o Diptera, 3% 

 Trichoptera. and 1°o Ephemeroptera, 



^Tentatively identified as Polycelis borealis. a species that Kenk (1953) 

 commonly found in clear cold streams in southern parts of Alaska, 



one box contained identifiable alevins. Although 

 the dead alevins disappeared rapidly, only a few of 

 the dead eggs disappeared. In a study to determine 

 whether certain stonefly numphs were predators 

 or scavengers on salmon eggs and alevins, Ellis 

 (1970) found in one experiment that dead pink 

 salmon alevins buried in Vibert boxes in a stream 

 essentially disappeared within a 2-wk period. 



Concurrently with the rapid disappearance of 

 dead alevins from the buried boxes was a rapid 

 buildup of invertebrates in the boxes. Although 

 invertebrates are commonly found with salmon 

 embryos (Briggs 1953; McDonald 1960; Nicola 

 1968), it is frequently impossible to determine if 

 predation or scavenging is occurring. Although 

 some groups of stonefly nymphs are known to at- 

 tack live salmon embryos (Stuart 1953; Claire and 

 Phillips 1968), Ellis ( 1970) concluded that nymphs 

 of the carnivorous genus Alloper la were basically 

 scavengers rather than predators. 



In addition to various insect larvae, a planarian 

 worm tentatively identified as Polycelis borealis 

 was commonly found in the boxes buried in Sashin 

 Creek (Table 12). Little is known on the biology or 

 life history of this planarian, but under favorable 

 conditions it appears to rapidly increase its num- 

 bers in the streambed, and thus may be particu- 

 larly important in removing dead alevins. I have 

 observed successive seasonal increases in the rela- 

 tive abundance of planarians in samples taken 

 from the Sashin Creek streambed with the hy- 

 draulic sampler in the fall, winter, and spring. 

 Total counts of planarians removed from the 

 streambed with the hydraulic sampler are not pos- 



577 



