691 
Estimating live standard length of 
net-caught walleye pollock 
( Theragra chalcogramma) larvae using 
measurements in addition to standard length* 
Steven M. Porter 
Annette L. Brown 
Kevin M. Bailey 
Alaska Fisheries Science Center 
National Marine Fisheries Service, NOAA 
7600 Sand Point Way NE 
Seattle, Washington 98115 
E-mail address (for S M Porter): steve.porter@noaa.gov 
Accurate measurement of larval fish 
lengths from ichthyoplankton samples 
is critical in the estimation of growth, 
birthdate and mortality. It is well 
known that larvae shrink in length 1) 
when caught in plankton nets (Thei- 
lacker, 1980; Hay, 1981), 2) between 
the time of collection and preserva- 
tion, and 3) afterwards from the effects 
of preservation in chemicals or from 
freezing (Theilacker, 1980; Fowler and 
Smith, 1983; Yin and Blaxter, 1986). 
Shrinkage caused by the plankton net 
(the net can damage a larva’s integu- 
ment; Holliday and Blaxter, 1960) and 
by delays in preservation can be up 
to 40% of the initial live length (Hay, 
1981). In some cases, measurements of 
larval body length may be unsuitable 
for growth estimation (Jennings, 1991 ). 
In other cases, algorithms have been 
developed to estimate live lengths from 
preserved larvae (Bailey 1982, Hjorlei- 
fsson and Klein-MacPhee, 1992; Fey, 
1999) and from net-caught and pre- 
served larvae (Theilacker, 1980; Thei- 
lacker and Porter, 1995; Fox, 1996). 
It has also been noted that constant 
shrinkage-correction factors should be 
applied cautiously (Pepin et al., 1998). 
Shrinkage of the standard length of 
fish larvae is species-specific ( Jennings, 
1991; Fey, 1999), size-dependent (Fowl- 
er and Smith, 1983; Theilacker and 
Porter, 1995; Fey, 1999), solution-de- 
pendent (Hay, 1982; Tucker and Ches- 
ter, 1984), and may also depend to 
some degree on the ambient tempera- 
ture at the time of preservation (Hay, 
1982). Other body measurements may 
shrink due to effects of collection and 
preservation. In fact, many corrections 
for larval shrinkage have been devel- 
oped with respect to estimating mor- 
phometric-dependent condition factors 
(Theilacker, 1980; McGurk, 1985; Yin 
and Blaxter, 1986). 
Measurement of otoliths has been 
proposed as a method to correct for 
shrinkage in the length of fish larvae 
(Leak, 1986; Radtke, 1989), but this 
method may not be generally appli- 
cable because of systematic variations 
in the relationship between fish size 
and otolith size (Neilson and Campa- 
na, 1990). In addition, this relationship 
can be nonlinear and unpredictable. 
Another approach to obtain shrinkage 
correction factors is by controlled ex- 
periments in which larval shrinkage 
is recorded from the time of death, or 
for the period of time larvae are in a 
plankton net, and again after preserva- 
tion. The problem with this approach 
is that during most sampling at sea, 
the time larvae enter the net to the 
time of their preservation is unknown 
and can vary from several minutes to a 
half hour or more, depending on their 
depth of capture and the duration of 
the tow. 
In our study, we used measurements 
of other body parts as ancillary data 
to estimate live lengths of preserved 
fish larvae caught at sea. We report 
on two common preservatives: ethanol 
and formalin. To avoid the problem 
of duration of time in the net, shrink- 
age correction equations were formu- 
lated by pooling data of known-length 
larvae that were treated for varying 
durations in simulated plankton tows 
in the laboratory and then preserved. 
Morphometric measurements were col- 
lected on preserved larvae that could 
be used in equations to obtain more 
accurate and precise estimates of live 
length. Rather than use alternative 
body part measurements as a substi- 
tute for length, we hypothesized that 
other body parts, nonshrinking or oth- 
erwise, could be used to correct for 
shrinkage, in spite of how long larvae 
had been in nets. 
Materials and methods 
Rearing protocol 
Spawning walleye pollock ( Theragra 
chalcogramma ) were collected by trawl 
in the eastern Bering Sea by the NOAA 
ship Miller Freeman during April 1999. 
Fertilized eggs were transported to 
the Alaska Fisheries Science Center, 
Seattle, Washington, where they were 
incubated at 6°C in the dark in 4-L 
glass jars filled with 3 L of filtered sea- 
water at a salinity of 33%<r . Larvae were 
reared in 120-L circular, black fiber- 
glass tanks (62 cm diameter, 43 cm 
deep) filled with 90-L filtered seawater, 
and maintained at a temperature of 
6°C. Two replicate tanks with approx- 
imately 1000 eggs each were main- 
tained. A 16:8 hour light:dark cycle 
was started at hatching and the amount 
of light at the surface of the water 
varied from 3.0 to 3.5 p mol photon/ 
m 2 /s. For the first two weeks of feeding, 
prey consisted of rotifers, Brachionus 
plicatilis , at a concentration of approxi- 
mately 10/mL, and wild zooplankton (a 
mixture of Acartia sp. copepod nauplii 
and copepodites, and gastropod and 
polychaete larvae; at approximately 
* Contribution FOCI-0400 of the Fisheries 
Oceanography Coordinated Investigations 
(FOCI), NOAA, 7600 Sand Point. Way NE, 
Seattle, WA 981 15. 
Manuscript accepted 19 March 2001. 
Fish. Bull. 99:691-696 (2001 ). 
