Harris et al.: Otolith morphometric analysis for species discrimination of Sebastes melanostictus and S. aleutianus 
235 
et al., 2006). In this research, the 2 species, which were 
referred to as Type I and Type II, have a fixed difference at 
the Sma6 microsatellite locus (Gharrett et al., 2005). Orr 
and Hawkins (2008) subsequently resurrected S. mela¬ 
nostictus for the blackspotted rockfish following a detailed 
morphological study. Orr and Hawkins (2008) also devel¬ 
oped a discriminant model for identifying the 2 species 
with 97.8% total accuracy. Unfortunately, this method 
relies on careful caliper measurements of several morpho¬ 
logical features, such as the gill rakers, dorsal spines, and 
pelvic fin rays. As such, this method has not been widely 
adopted for use with large-scale stock assessment surveys, 
and a quick and accurate method for distinguishing these 
species in the field remains elusive. 
At sea, blackspotted rockfish can often be distinguished 
by black markings on the body, but these spots vary in 
number and intensity. Rougheye rockfish often have 
some minor spotting or dark coloration, such that there 
is considerable overlap in the color patterns of the spe¬ 
cies. Experienced field scientists can separate them with 
a moderate level of accuracy, but fishery observers and 
other personnel do not attempt to identify these species. 
The genus Sebastes has been found to have several cryp¬ 
tic species of similar appearance, but they have different 
habitat preferences and require different management 
measures (Hyde et al., 2008). Consequently, in some areas 
the fishery catch is predominantly blackspotted rockfish 
(Orr and Hawkins, 2008). If the catch is not evenly distrib¬ 
uted, managing these species jointly runs the risk of caus¬ 
ing detrimental effects to the more vulnerable population 
(Hawkins et al., 2005). 
Otoliths are found as 3 paired sets of calcium carbon¬ 
ate structures in the inner ear of teleost fish species and 
function in hearing, balance, and orientation. In rockfish 
species, the sagittae are the largest of the otoliths 
and are most frequently collected for age deter¬ 
mination. Age estimates are obtained by exam¬ 
ining sectioned otoliths under a microscope, and 
ages obtained this way have been validated by 
using radiometric dating in a number of species 
(Hutchinson et al., 2007; Kastelle et al., 2008). 
Because otoliths from different populations may 
have unique shapes, the analysis of otolith data is 
a popular method for distinguishing similar spe¬ 
cies (Sadighzadeh et al., 2012) or different stocks 
within a species (Campana and Casselman, 1993; 
Hicks et al., 2003). 
The purpose of this study was to develop a 
mathematical method that accurately separates 
the rougheye and blackspotted rockfish species on 
the basis of their otolith morphometries and other 
data that are collected in the biennial fishery- 
independent surveys conducted by the NOAA 
Alaska Fisheries Science Center (AF3C) in the 
Gulf of Alaska. Because diagnostic genetic mark¬ 
ers have been developed (Gharrett et al., 2005), it is 
possible to correctly identify a specimen to species 
and to use these genetically identified specimens 
to construct and validate a discrimination method 
based on otolith morphology. This model will be used to sep¬ 
arate the 2 species in AFSC fishery-dependent collections 
and to verify field identifications made on stock assessment 
surveys. An otolith-based discrimination method would 
give scientists access to over 30 years of AFSC archived 
otolith collections that were not separated by species until 
2006 and would greatly increase the data for each species. 
Materials and methods 
Sampling 
Otoliths were collected from rougheye rockfish and black¬ 
spotted rockfish sampled from the Gulf of Alaska in the 
summers of 2009 and 2013 during AFSC survey cruises on 
contracted fishing vessels (Fig. 1). During 2009 and 2013, 
AFSC scientists collected 945 and 902 fish, respectively. 
At the time of capture, each sampled rockfish was mea¬ 
sured and weighed, its sex was determined, and its otoliths 
were removed and stored in 50% glycerol-thymol solution. 
Fin clips were taken from each fish and stored in 20% 
dimethyl sulfoxide and 80% 0.25-M EDTA solution. 
Genetic analysis 
Fin clips were sent to the AFSC Auke Bay Laboratories in 
Juneau, Alaska, for genetic analysis. These samples were 
genotyped according to the methods published in Wim- 
berger et al. (1999). The Sma6 locus has been shown to 
clearly delineate rougheye and blackspotted rockfish spe¬ 
cies, with each species homozygous having a unique allele 
in more than 99.7% of all specimens (Gharrett et al., 2005). 
The field identifications were compared with the genetic 
