358 
Fishery Bulletin 1 13(4) 
124°13'55.56"W) from 9 April to 25 June 2012 under 
the annual scientific collecting permit issued by the 
Oregon Department of Fish and Wildlife (ODFW) for 
the collection of samples by their employees (DWW 
and TNF). Specimens were frozen for transport, 
thawed, and photographed. Samples of tissue from 
the trunk musculature were removed and archived, 
and 35 of these freshly collected samples ( S . mysti- 
nus [n=15], S. diaconus [n= 20]) were genotyped during 
microsatellite analysis. Voucher specimens were fixed 
in 10% formalin, preserved in 50% isopropanol, cata- 
loged, and deposited in the Oregon State Ichthyology 
Collection (OS). More detailed information on catalog 
number, collection locality, and size of each examined 
specimen appears in the species descriptions that fol- 
low. We examined 134 specimens for comparisons of 
meristic and linear morphometric data: S. mystinus 
(n= 68), S. diaconus (n= 58), S. ciliatus (n=6), and S. 
variabilis (n= 2). 
Genetic sampling 
Muscle tissue was used to extract DNA from 15 blue- 
blotched and 20 blue-sided individuals with DNeasy 2 * 
blood and tissue kits (Qiagen, Valencia, CA). Micro- 
satellite loci were amplified by a 3-primer polymerase 
chain reaction (PCR) protocol (Schuelke, 2000). To test 
the assignment of the color morphotypes with the pre- 
viously described genetic groups, 16 type-1 and type-2 
individuals previously identified and analyzed by Bur- 
ford and Bernardi (2008) were included in microsatel- 
lite amplification and analysis. 
Amplification of microsatellite markers and DNA 
sequences 
We amplified 6 anonymous microsatellite markers 
(Seb37, Roques et al., 1999; Spi4, Gomez-Uchida et al., 
2003; Sra7-7, S?'al5-8, Westerman et al., 2005; Ssc69, 
Yoshida et al., 2005; and KSsl8A, An et al., 2009). Two 
microsatellite loci ( Sra7-7 and Sral5-8) were used by 
Burford and Bernardi (2008) in the original description 
of the 2 lineages. The 3-primer amplification protocol 
was used to run in 15-pL volumes with the follow- 
ing reaction concentrations: lx AmpliTaq PCR Buffer 
(Thermo Fisher Scientific, Inc., Waltham, MA) buffer, 
2.5 mM MgCl 2 , 0.4 mM of each dNTP, 2.7xl0- 4 mg/mL 
BSA, 0.3 pM reverse primer, 0.3 pM fiuorescently la- 
beled M13 sequence (5'CACGACGTTGTAAAACGAC3') 
with dye labels (FAM, VIC, NED, PET; Thermo Fisher 
Scientific, Inc.), 0.07 pM M13 5'-end labeled forward 
primer, 0.2 units of AmpliTaq DNA polymerase, and 
5pL DNA (5-20 ng/pL). We used the following ther- 
mal profile: initial denaturing step at 94°C for 5 min, 
followed by 30 cycles of 94°C for 30 s, 52°C for 30 s, 
and 72°C for 45 s. The incorporation of the M13 prim- 
2 Mention of trade names or commercial companies is for iden- 
tification purposes only and does not imply endorsement by 
the National Marine Fisheries Service, NOAA. 
er required 15 cycles of 94°C for 30 s, 48°C for 30 s, 
and 72°C for 45 s. A 7-min extension after the final 
cycle completed the thermal profile. Amplified products 
were run on an Applied Biosystems 3100 automated 
sequencer with GeneScan 500 LIZ Size Standard and 
genotyped with GeneMapper, vers. 3.7 (Thermo Fisher 
Scientific, Inc.). 
Analysis of microsatellite markers and DNA sequence 
We used the software program Structure, vers. 2.3 (Fa- 
lush et al., 2003; Pritchard et al., 2000) to determine 
the correspondence of blue-sided and blue-blotched 
morphotypes with the type-1 and type-2 genetic groups 
described by Burford and Bernardi (2008). Parameters 
in Structure were set to produce posterior probabilities 
with 500,000 replicates recorded after a burn-in period 
of 50,000 steps that were discarded. Default settings 
were used with the admixture option in Structure. Be- 
cause the optimal K value was previously identified as 
2 (Burford and Bernardi, 2008), we ran simulations 
according to these parameters to identify the corre- 
sponding genetic groups of each morphotype. We also 
simulated structure runs with K values within a range 
of 2-4 and identified optimal K value with the online 
program Structure Harvester, vers. 6.92 (Earl and von- 
Holdt, 2012). 
Measurements and meristics 
Measurements were taken to the nearest 0.01 mm with 
digital calipers. Morphometries (32 measurements) and 
meristics (17 counts) followed Orr and Hawkins (2008), 
except as follows. Symphyseal knob length was mea- 
sured on the ventral side of the lower jaw from the 
posterior margin of the symphysis to the anterior tip of 
the knob. Head depth was measured along the vertical 
bisecting the eye. An additional point-to-point measure- 
ment was taken from the dorsal-fin origin to anal-fin 
origin. We augmented the meristics used by previous 
authors with counts of dorsal-fin spines, of branched 
and unbranched pectoral-fin rays on both sides of the 
fish, of transverse lateral scale rows, and of posterior 
and anterior gill rakers from the left side of the fish. 
Dorsal-, anal-, pectoral- and pelvic-fin rays and spines 
were counted from preserved specimens. Vertebrae and 
caudal-fin rays were counted from a subsample of spec- 
imens (n=68) via film radiographs. 
Morphometric analysis 
We created a size-standardized morphospace for 120 
specimens (6 specimens were excluded because of dam- 
age) with the Allometric Burnaby technique (Burnaby, 
1966) implemented in the software PAST, vers. 2.17 
(Hammer et al., 2001), which log transforms the 32 
linear measurements and projects them orthogonal- 
ly to the first principal component. These data were 
then used in a principal components analysis (PCA). 
For statistical analyses, putative species membership 
