Lowe et at: Geographic variation in genetic and growth patterns of Pleurogrammus monopterygius 
503 
An analysis of morphological and meristic data by 
a Russian scientist indicated separate populations 
in the Gulf of Alaska and the Aleutian Islands. 3 The 
meristic study compared the number of dorsal, anal, 
and pectoral fin rays, total number of vertebrae, and 
number of gill rakers on the first gill arch from a 
sample of 100 fish collected off Kodiak Island in the 
Gulf of Alaska and the Rat Islands in the Aleutian 
Islands. The morphological study consisted of a sta- 
tistical comparison of various partial fish body 
lengths as a percentage of fork length, by area. Lee 
(1985) also conducted a morphological study analyz- 
ing the covariance between four partial fish lengths 
and fork length by area and sex from samples taken 
from the Bering Sea, Aleutian Islands, and the Gulf 
of Alaska. The data showed some differences (al- 
though not consistent by area for each characteris- 
tic), suggesting a certain degree of reproductive iso- 
lation. On the basis of an analysis of variance of Aleu- 
tian Islands growth data with year, area, and sex as 
factors, Kimura and Ronholt ( 1988) found significant 
differences in weight and length-at-age of Atka mack- 
erel from six different areas in the Aleutian Islands, 
indicating potential stock differentiation. Kimura 
and Ronholt (1988) suggested therefore that Atka 
mackerel appear to be distributed in localized groups 
or assemblages once they assume the more demer- 
sal phase of their life history. Recent analysis of Aleu- 
tian Islands Atka mackerel growth data by current 
fishery management areas shows a distinctive size 
cline, with length at age smallest in the western Aleu- 
tians and largest in the eastern Aleutians 1,2 
Differential growth by area can be an indication 
of stock delineations. The presumption for Atka 
mackerel, based on the morphological and growth 
analyses, was of at least a discontinuous distribu- 
tion throughout the Aleutian Archipelago. Genetic 
tests are necessary to confirm separate stocks and 
help to further our understanding of the life history 
and distribution patterns of Atka mackerel. Infor- 
mation about the extent and nature of stock differ- 
ences is also critical to improve stock assessments for 
long-term management of the Atka mackerel fisheries 
in the Gulf of Alaska and Aleutian Islands. To date, 
there have been no population genetics surveys of this 
species to determine the existence of discrete stocks. 
A preliminary survey of allozymes from 40 indi- 
viduals indicated a sufficient number of polymorphic 
loci to conduct a full study of the genetic population 
structure of this species (Winans et al., 1995). We 
report here 1) length-at-age and length-weight rela- 
3 Levada, T.P. 1979. Comparative morphological study of Atka 
mackerel. Pac. Sci. Res. Inst. Fish. Oceanogr.5(TINRO ), 
Vladivostok, U.S.S.R. Unpubl. manuscript, 7 p. 
tions derived from samples taken throughout the 
Aleutian Archipelago and 2) a genetic survey of Atka 
mackerel from samples taken from four locations in 
the Aleutian Islands ranging from 169°W to 172°E 
(Fig. 1). Our null hypothesis is that there are no dif- 
ferences between samples taken along the Aleutian 
Archipelago, from Umnak Island in the east to Attu 
Island in the west. 
Materials and methods 
Age , weight, and length data 
Biological samples and length and weight informa- 
tion were collected during National Marine Fisher- 
ies Service (NMFS) triennial trawl surveys, June to 
August of 1993 and 1994, from the Gulf of Alaska 
and Aleutian Islands region, respectively. The sam- 
pling coincided with the summer spawning period of 
Atka mackerel (July to October; McDermott and 
Lowe, 1997). Random samples of Atka mackerel from 
the trawl survey catches were sorted by sex, and in- 
dividual weight and fork-length data were collected. 
Length was estimated to the nearest one centime- 
ter, and weight was estimated to the nearest gram 
with a platform scale when weather conditions al- 
lowed (Martin and Clausen, 1995). Age structures 
(sagittal otoliths) were collected by using a length- 
stratified sampling scheme of five fish per sex, per 
centimeter length category. An attempt was made to 
distribute the otolith collections over the entire sur- 
vey area. Otoliths were placed in vials with 50% etha- 
nol, and age was determined by personnel in the 
NMFS Age Determination Laboratory. 
A total of 510 otoliths were collected and aged 
(Table 1). Otoliths were prepared by snapping each 
along the dorsal-ventral plane and passing the bro- 
ken surface over a flame. The burnt cross-section was 
examined under a dissecting microscope and illumi- 
nated by reflected light (Anderl et. al, 1996). 
Growth parameters were estimated by fitting the 
age-length data to the widely used von Bertalanffy 
(1938) growth equation: 
/ t = L 00 fl-e~ K<t ~ t ° ) J, 
which expresses length at age t ( l f ) as a function of 
three parameters: L m , K, and t Q . As age increases, 
length approaches L^, which is the mean asymptotic 
length. The slope of the von Bertalanffy curve con- 
tinuously decreases with increasing age as l t ap- 
proaches the asymptotic length. This rate of decrease 
is described by K. The parameter t 0 is the theoreti- 
cal time a fish would have been zero length. The 
