NOTE Pepin et a!.: Changes in the distribution of larval fish body length following preservation 
639 
0.45 
0.40 
0.35 
§ 0.30 
(O 
*D 
| 0.25 
0.20 
0.15 
0.10 
0 2 4 6 8 10 12 14 16 
Length interval (mm) 
Figure 4 
Rank correlation (Kendall’s x) of preserved versus 
fresh lengths, within millimeter intervals of fresh 
lengths. Open symbols represent values not signifi- 
cantly different from 0, grey symbols present sig- 
nificant values (P<0.05), and black symbols repre- 
sent highly significant values (P<0.01). 
9 o 
A •• ? i 
• \ 
• 
• 
\ 0 
0 
0 
0 . 
O 
length following preservation. It is important to note, 
however, that although our results show that the 
mean change in length for larvae <6 mm SL was posi- 
tive, there were also numerous individuals in these 
same length intervals that either shrank or showed 
no change in body length after preservation. This 
finding underscores our contention that responses 
to preservation vary significantly at the level of the 
individual as well as among species. 
It is possible that the capture, sorting, and preser- 
vation approaches used in this study may have led 
to variations in the general pattern of changes in body 
length. Following capture, sorting on deck may have 
caused a change in water temperatures. This proce- 
dure may have precipitated the death of individual 
larvae and resulted in contraction of body tissue in 
the “fresh length” measurement and consequently 
may have produced an apparent increase in body 
length following remeasurement. The source of lar- 
vae may also be an important feature to consider. 
Most preservation studies are based on laboratory- 
reared animals rather than on field-caught speci- 
mens (Table 1). Laboratory-reared animals show 
greater changes in length than field caught speci- 
mens (Table 1), although there is some variation 
about this general pattern (e.g. Theilacker, 1986). 
Change in body length of marine fish larvae fol- 
lowing capture and preservation has been attributed 
mainly to a breakdown of the osmoregulatory capac- 
ity of larvae, leading to a net loss of water ( Ahlstrom, 
1976). Simple laboratory-based studies of shrinkage 
have concluded that preservation results in relatively 
small changes in body length (see Hjorleifsson and 
Klein-Macphee, 1992) but that the impact of net-cap- 
ture and handling is likely to have the greatest im- 
pact on changes in body length (Theilacker, 1980; 
Jennings, 1991; Fox, 1996). Most laboratory simula- 
tions of capture used fixed handling times; however, 
under field conditions it is unknown when a larva 
enters the net. Thus correction factors for a fixed 
handling time may introduce substantial error. Our 
sampling methods differed dramatically from those 
normally employed in field situations in both the 
duration of the residence time of larvae in the net 
and in the aggressiveness of capture and handling. 
It is therefore likely that, in comparison with other 
studies, the effect of capture and handling in our study 
was small. However, despite the possible importance 
of capture and handling, the interpretation of the con- 
dition of individual larvae or the distribution of popu- 
lation characteristics must be approached with caution. 
Our results clearly show that the relative position 
of an individual within the size distribution of the 
sample is likely to change considerably following 
handling and preservation. Thus, attempts to cor- 
rect for the effects of preservation or handling on 
fresh larval lengths can, in certain types of analysis 
(e.g. individual-based approaches to the use of mor- 
phometric measurements), introduce significant and 
unpredictable bias in both the data and their inter- 
pretation. There appears to be no simple solution to 
this problem. Conclusions as to the overall vulner- 
ability of a fractional portion of a population to a 
given factor must consider the inherent variability 
in the distribution of measurements caused by op- 
erator, handling, and preservation-induced error. 
Acknowledgments 
We thank Tim Shears and Sarah Donald for providing 
valuable technical assistance. Additional assistance was 
provided by Steph Carter, Dave Bell, and the officers 
and crew of CSS Shamook. C. J. Fox and two anony- 
mous referees provided valuable criticisms of this study. 
Funding for this project was provided by the Depart- 
ment of Fisheries and Oceans (PP) and the Natural 
Sciences and Engineering Research Council (WCL). 
Literature cited 
Ahlstrom, E. H. 
1976. Maintenance of quality in fish eggs and larvae col- 
lected during plankton hauls. In H. F. Steedman (ed.), 
Zooplankton fixation and preservation: monographs on 
