Sutherland and Richards: Aging Lophius americanus based on length-mode progression of a strong cohort 17 
Marginal increment analysis 
Marginal increment analysis was conducted only for 
illicia because it had been done previously for vertebrae 
(Armstrong et al., 1992). Samples were limited to those 
collected during May 2016—May 2017 and for which 
the ring count matched the known age. Furthermore, 
only samples that had at least one band inside the edge 
(i.e., a minimum age of 1+; the notation 1+ indicates the 
presence of additional material outside the first ring) 
were included because relative increment width cannot 
be calculated on younger fish, whose illicia lack rings 
inside the edge. 
Illicia were imaged and measured with knowledge of 
known age but without knowledge of fish length or cap- 
ture date. The order in which they were examined was 
randomized by row. Images were taken at 150x mag- 
nification with an Olympus BH2 microscope, an Olym- 
pus DP25 camera (Olympus Corp.), and 64-bit TWAIN 
Twacker software (vers. 2.0; TWG, 2008). ImageJ soft- 
ware (vers. 1.49v; Schneider et al., 2012) and the ObjectJ 
plugin (vers. 1.03s; University of Amsterdam, available 
from website) were used for image processing. 
Diameters of the settlement check, first ring, last ring 
inside the edge (F,,), and total illicium diameter (D) were 
measured to the nearest 0.0001 mm along the axis with 
the largest settlement check diameter (Fig. 3A). Rela- 
tionships between these measurements (settlement 
check, first ring, and total diameter) and total length 
were examined by using linear regression on untrans- 
formed data with the regression analysis add-in for 
Microsoft Excel 2016 (Microsoft Corp., Redmond, WA). 
Regression results were used to test the assumption 
that the diameters of the settling check and first annu- 
lus were constant across the range of fish lengths. Total 
illicium diameter was expected to increase linearly with 
fish length. 
These measurements were then used to calculate the 
relative marginal increment (MIR) (Sun et al., 2002): 
Dale, 
MIR = ———*_, 
R, = Rai 
(1) 
where R,,_, = the penultimate ring. 
If the last ring is on the edge, D—R,,=0. For fish with only 
one ring inside the edge (i.e., fish with estimated ages of 
1+ or 2 years), R,_, is not present, simplifying the calcu- 
lation to the following equation (Vilizzi and Walker, 1999): 
I (2) 
R 
n 
Average MIR was calculated for each month, and minima 
in the monthly MIR values were considered indicative of 
the timing of ring deposition. 
Results 
Survey length composition and sample collection 
The length mode representing the 2015 year class 
(Table 1, Fig. 4) was first seen in the scallop dredge 
Table 1 
Known age, mean length with standard deviation (SD), and growth increment of the 
dominant length mode of goosefish (Lophius americanus) collected during surveys 
conducted by the NOAA Northeast Fisheries Science Center from June 2015 through 
September 2018 along the Atlantic coast of the United States from Virginia to Georges 
Bank. Mean length was estimated by using normal curves fit to the stratified mean 
number per tow at length. Growth increment was calculated as the change in length 
divided by the number of days since the previous survey. Surveys include bottom trawl 
surveys conducted in spring and autumn and scallop dredge surveys conducted each 
June. TL=total length. 
Known 
Average age 
Survey survey date (years) 
Scallop 2015 7-Jun-2015 
Fractional 
Autumn 2015 
Spring 2016 
Scallop 2016 
Autumn 2016 
Spring 2017 
Scallop 2017 
Autumn 2017 
Spring 2018 
Scallop 2018 
Autumn 2018 
20-Sep-2015 
1-May-2016 
29-May-2016 
27-Sep-2016 
2-Apr-2017 
11-Jun-2017 
25-Oct-2017 
7-Apr-2018 
1-Jun-2018 
22-Sep-2018 
WWWNNNRFRFF OS 
Growth 
increment 
(cm TL/d) 
Mean 
age length (SD) 
(years) (cm TL) 
10.7 (1.8) 
21.6 (3.5) 
27.2 (3.6) 
26.2 (3.8) 
36.8 (3.7) 
40.9 (4.2) 
41.7 (6.2) 
51.5 (6.4) 
50.7 (8.1) 
53.0 (6.7) 
53.0 (6.1) 
