750 
Fishery Bulletin 95(4), 1997 
Mature herring were captured during or just prior 
to spawning at all sites, sorted by size, and artifi- 
cially spawned. Capture gear included gill net, cast 
net, and purse seine. Fish were chilled immediately 
after capture and transported within two hours to a 
field laboratory, except Seymour Canal fish, which 
were transported directly to Auke Bay Laboratory 
(ABL). To approximate the different age classes 
present, fish were sorted by sex and size (usually in 
10-mm increments; e.g. 220-230 mm fork length). 
Six or more size classes were usually identified at 
each site. From each size class, 25 females were ar- 
tificially spawned with males of the same size; gen- 
erally 3 males contributed sperm for all 25 crosses. 
Size classes of fish were processed at random. Each 
fish was assigned an identification number, mea- 
sured to the nearest mm (fork length), and weighed 
to the nearest 0.1 g (wet weight). To determine age, 
three scales were removed from the left side of each 
spawned fish near the posterior margin of the dorsal 
fin, placed on a glass slide, and covered with a sec- 
ond slide. 
For spawning, testes were removed, sealed in a 
plastic bag and maintained in chilled seawater until 
use; ovarian membranes were cut longitudinally, and 
eggs were removed with a hydrocarbon-free stain- 
less steel spatula similar to that used by Brown. 4 
From each female, approximately 150 eggs were de- 
posited with a gentle swirling motion onto a 25 x 75 
mm glass slide placed on the bottom of a shallow plas- 
tic dish filled with seawater. Each slide was then 
placed in a staining rack and suspended in its own 
1-L beaker of seawater. Milt was prepared from col- 
lected testes by cutting sections from each into small 
segments; segments plus a small amount of seawa- 
ter were mixed with a spatula. A few milliliters of 
the milt mixture were added to each beaker contain- 
ing eggs. Eggs and milt remained in contact for 5 
min; the milt was then poured off, and the eggs were 
gently rinsed in seawater. Slides were kept in stain- 
ing racks and maintained in ambient seawater with 
constant aeration until they were transported to ABL 
by air. To transport the eggs, staining racks were 
placed in plastic seawater-filled containers, which 
were then placed in coolers with blue ice. 
Slides with eggs from each site were randomly dis- 
tributed among twelve 600-L tanks with flow-through 
seawater. Slides were suspended from monofilament 
line attached to a pivoting overhead framework de- 
signed to cause slow egg movement (1 rpm) through 
the water. During the first 16-18 days of incubation, 
4 Brown, E. D. 1995. Alaska Department of Fish and Game. 
Commercial Fisheries Management and Development Division, 
PO Box 669, Cordova, AK 99574. Personal commun. 
all slides were maintained in the seawater bath. A 
few days before hatching, each slide was isolated in 
a 1-L glass jar that contained seawater and that was 
surrounded by flowing seawater. Lighting was natu- 
ral, supplemented by overhead fluorescent light dur- 
ing daylight hours. Seawater flow was approximately 
1 L/min at 3.9°C, warming to 7.1°C with normal sea- 
sonal change. Salinity was 32 ±1 ppt. 
Reproductive success of female herring was defined 
as the production of physically and functionally nor- 
mal larvae. Key reproductive parameters included 
hatching success and larval viability, swimming abil- 
ity, and spinal abnormalities. These four parameters 
were sensitive to oil in laboratory studies (Carls et 
al. 5 ). Other parameters examined included fertility 
and hatching times. Fertility was not considered a 
key parameter because it may have been negatively 
influenced by unavoidable handling conditions at the 
different sites and by variable periods in the storing 
of gametes prior to spawning. Hatching times were 
not considered a key parameter because they were 
strongly influenced by seasonal increases in water 
temperature. 
Fertilization success and stage of development 
were determined 1 to 10 days after spawning. Ex- 
cess eggs were removed from all slides by scraping — 
i.e. those along slide margins susceptible to mechani- 
cal damage and clumps of eggs not directly exposed 
to water. This process was accomplished in water 
with minimal exposure to air. 
Isolated eggs were inspected every two days to 
determine onset of hatching. Once hatching began, 
larvae were counted and assessed daily for swimming 
ability and gross physical deformities. Without ex- 
posing eggs to air, we changed the seawater in each 
jar every two days prior to hatching and daily after 
hatching began. All hatched larvae were collected, anes- 
thetized with tricaine methanesulfonate, and preserved 
in 10% phosphate buffered formalin. Approximately the 
first and last 10% of larvae hatched from each female 
were preserved in separate bottles. Live larvae were 
preserved separately from dead larvae. After hatching 
was completed, remaining eggs were inspected; infer- 
tile eggs and dead embryos were counted. 
A subset of preserved larvae was scored for yolk- 
sac edema, pericardial edema, and yolk volume. Ten 
females from the 1989 year class were randomly se- 
lected from each site, and 10 larvae per female were 
randomly selected from the central portion of hatched 
eggs for analysis. At Fish Bay, only five females from 
5 Carls, M. G., D. M. Fremgen, J. E. Hose, S. W. Johnson, and S. 
D. Rice. In prep. Effects of incubating herring (Clupea 
pallasi) eggs in water contaminated with weathered crude 
oil. Auke Bay Laboratory, National Marine Fisheries Service, 
NOAA, 11305 Glacier Hwy., Juneau, AK 99801. 
