912 
Metazoan parasites as 
potential markers for selected 
Gulf of Alaska rockfishes 
Adam IVSoSes 
Jonathan Heifetz 
David C. Love 
Auke Bay Laboratory, Alaska Fisheries Science Center 
National Marine Fisheries Service, NOAA 
1 1305 Glacier Highway, Juneau, Alaska 99801-8626 
E-mail address (for A. Moles): Adam.Moles@noaa.gov 
Rockfishes (family Scorpaenidae) of 
the genus Sebastes constitute some 
of the most important groundfish 
resources in the northeastern Pa- 
cific Ocean. In Alaska waters, 
shortraker rockfish, S. borealis , 
and rougheye rockfish, S. aleutianus, 
are especially prized because their 
large size and red color make them 
valuable on the Asian market. Both 
species reside in offshore waters of 
the upper continental slope, and 
they often co-occur. Commercial 
catches of these two species in the 
Gulf of Alaska (GOA) have aver- 
aged 1700 metric tons per year dur- 
ing this period, and the estimated 
exvessel value of the fishery in 1991 
( the most recent year with good eco- 
nomic data) was $3 million. 
Knowledge of stock structure and 
the degree of mixing among popu- 
lations is important for the ratio- 
nal management of commercially 
important marine species. At 
present, such information is lacking 
for shortraker and rougheye rockfish, 
and questions, such as whether these 
species perform lengthy migrations 
or whether discrete populations 
exist, are unanswered. Traditional 
tag-and-release experiments, which 
are often used to investigate stock 
structure, are not feasible for 
physoclistic fish such as rockfish 
because they cannot survive the 
barotrauma of initial capture. One 
alternative to human-implanted tags 
is the use of parasites as naturally 
occurring tags. Because necropsies of 
rockfishes are time consuming, we 
examined 100 fish of each species 
from the GOA to identify the feasi- 
bility of targeting selected parasites 
as tags for these species. 
Methods 
Adult shortraker and rougheye 
rockfish were captured in the GOA 
in summer 1990 during an annual 
longline survey conducted on the 
upper continental slope by the Na- 
tional Marine Fisheries Service. 
Sampling covered five statistical 
management areas established by 
the International North Pacific 
Fisheries Commission (Zenger and 
Sigler, 1992): Shumagin, Chirikof, 
Kodiak, Yakutat, and Southeast 
(Fig. 1). The survey area extended 
from the Island of the Four Moun- 
tains (52°50'N, 170°W) eastward to 
Dixon Entrance (54°29'N, 134°W). 
Forty-five stations were sampled 
along the upper continental slope 
at depths from 150 to 1000 m. Of 
the 7836 rockfishes of various spe- 
cies caught in the cruise, a sub- 
sample of 20 rougheye and 20 
shortraker rockfish (21 from Shu- 
magin) from each management 
area were frozen on board for later 
necropsy. Total sample size was 101 
shortraker and 100 rougheye rock- 
fish from 21 stations. 
In the laboratory, we conducted 
complete necropsies for metazoan 
parasites on an initial sample of 54 
shortraker and 69 rougheye rock- 
fish. As a result of preliminary 
analysis of the parasite data from 
these fish, we restricted the necrop- 
sies on the remaining 47 shortraker 
and 31 rougheye rockfish to exami- 
nations for the presence of gill and 
fin copepods, monogenetic gill 
trematodes, and visceral acantho- 
cephalans. Identities of representa- 
tive parasite specimens were veri- 
fied by D. J. Whitaker of the Cana- 
dian Department of Fisheries and 
Oceans. Both prevalence (the pro- 
portion of fish with a given para- 
site) and intensity (the mean num- 
ber of parasites per infected fish) 
were calculated for each manage- 
ment area. 
Geographic trends were exam- 
ined by grouping samples into the 
five management areas. Categori- 
cal analysis of variance (SAS pro- 
cedure CATMOD; SAS Inst., 1989) 
was used to determine whether 
parasite prevalence differed among 
areas. For this analysis, the preva- 
lence of a particular parasite was 
used as the dependent variable, 
and area and fish size as indepen- 
dent variables. For each fish spe- 
cies, median fork length was used 
to divide the areawide sample into 
two groups, hereafter referred to as 
large and small. Data on parasite 
intensity for each fish were divided 
by fish length to account for the 
possible influence of fish size on 
parasite intensity. Distribution of 
the intensity data was highly non- 
normal, even after transformation. 
Therefore, intensity data were ana- 
lyzed by using the nonparametric 
Mann-Whitney U test. These analy- 
ses allowed us to account for differ- 
ences in fish size by area while test- 
Manuscript accepted 7 January 1998. 
Fish. Bull. 96:912-916 (1998). 
