Schueller et al.: Using a projector and microscope to read scales from Brevoortia tyrannus and B. patronus 23 
samples with agreed upon ages from experienced 
readers, ages that are representative of all ages 
and the monthly seasonality typical in the fishery. 
The reference collections are used to assess aging 
consistency between readers and among estimates 
from a reader over time. The second reader (reader 
2) estimated ages with scales on the Eberbach pro- 
jector but only for Atlantic menhaden sampled in 
2013, for Gulf menhaden sampled in 2005, and for 
both reference collections (SEFSC’). From 1969 
through 2015, the same reader (reader 2) esti- 
mated ages of menhaden by using scales, and 
those estimates from the database at the Beau- 
fort Laboratory (SEFSC°) compose the historical 
record of ages that was used in this study. Com- 
parisons of ages determined with an Eberbach 
projector were made between reader 1 and reader 
2 for each species for the sample years that could 
be compared. 
A subsample of 200 scales each from the full set 
of scale samples from Atlantic and Gulf menhaden 
was used to estimate ages a second time by reader 
1 with both the Eberbach projector and the micro- 
scope. The best scale for each sample was selected, 
and its position on the slide was recorded so that 
the same scale could be used for aging on both 
devices. Subsampling provided 2 age readings on 
each device for each scale, allowing comparison of aging 
errors for each instrument and between them. This subsa- 
mpling allowed errors to be classified as random reading 
errors versus errors due to the effect of the device used. On 
average, for reader 1, 10-19 months separated the first and 
second readings for scale samples from Atlantic and Gulf 
menhaden on the Eberbach projector and 6 months sepa- 
rated readings with the microscope. 
Measurements of scales were taken for all subsamples 
(Fig. 1). A sonic digitizer pen fitted on the Eberbach pro- 
jector was used for measuring distance from the focus of 
the scale to each annulus and to the edge margin. Mea- 
surements were recorded in Eberbach units and later 
converted to millimeters. The same distances were mea- 
sured in millimeters on the calibrated image taken with 
the microscope, by using the image analysis software. On 
each image, an origin was placed on the scale’s focus. The 
point tool was used to mark each annulus from the point 
of origin up to and including the edge. A new embedded 
image was created that included the scale image and 
the recorded measurements, and measurements were 
exported for analysis to Microsoft Excel 2016 (Microsoft 
Corp., Redmond, WA). In addition, while the Eberbach 
projector was being used for aging, a ruled blue card was 
placed on top of the projector’s screen to manually mea- 
sure each annulus and edge measurement of a scale image, 
as was done during the early years of data collection by 
° SEFSC (Southeast Fisheries Science Center). 2018. Unpubl. data. 
Menhaden Biostatistical Database. Beaufort Lab., Southeast 
Fish. Sci. Cent., Natl. Mar. Fish. Serv., NOAA, 101 Pivers Island 
Rd., Beaufort, NC 28516. 
Anterior field 
Posterior field 
Figure 1 
Image of a scale from an Atlantic menhaden (Brevoortia tyrannus), 
collected in 2019 off the Atlantic coast of the United States, showing 
2 annuli in the anterior field. Scale measurements were taken for 
all subsamples and taken from the focus to each successive annulus 
and to the edge. 
the Beaufort Laboratory. A comparison of measurements 
for each scale was used to verify if the same annuli were 
counted on each device. This process allowed comparison 
of measurements taken by hand, with a sonic digitizer 
pen, and with image analysis software, temporally across 
the age data from the Beaufort Laboratory for both species 
of menhaden. 
Both the full sample and the subsample of scales were 
chosen to reflect the seasonality of the fishery by month 
and the spatial spread of the fishery by location of the 
processing plants. In addition, samples were selected so 
that the length distribution of fish from which the sam- 
ples were taken reflected the length distribution of all fish 
sampled from landings during a given year. Samples were 
chosen such that an adequate number of fish with younger 
and older ages were selected and such that the age com- 
position matched that of the fishery samples taken in a 
given year. 
Some scale samples were more than a decade old, and 
the quality of scales was inconsistent because of storage 
conditions. Although some scales remained in a quality 
good enough to age, others had debris or had degraded 
with time. Scales that were deemed unreadable by reader 
1 on either of the instruments or that were missing or lost 
were excluded before analysis. First, we compared the 
number of excluded scales between readers and between 
devices. Second, we compared age readings for those scales 
that were legible. Finally, we compared successive annu- 
lus measurements between the Eberbach projector and 
the microscope. 
Basic plots of ages by reader and device were created for 
each of the comparisons listed in the previous paragraph 
