Schobernd ei al.: A comparison of numbers of fish larvae extruded from plankton nets of different mesh size 
245 
12 3 4 
Length (mm) 
Figure 2 
Length-frequency histograms of body length by total larval abundance in samples collected in the Gulf of 
Mexico during 2005-2007 with nets of 2 different mesh sizes (0.202 and 0.333 mm) for all larvae, unidentified 
larvae, and larvae of the suborder Percoidei. Larval abundance was measured as the total number of larvae 
caught in a sample per the volume of water filtered during the tow multiplied by the maximum depth of the 
tow in meters and the factor 10 (total number of larvae under 10 m 2 of sea surface). Only body lengths up to 
10 mm are displayed in this figure. Note the different range in body length for each category of larvae. 
tion (r 2 )=0.841; y=1.0615x+57.643, r 2 =0.8076, for the 
samples collected with nets with 0.202-mm and those 
collected with 0.333-mm mesh, respectively). Plots of 
tow duration by larval abundance for each mesh size 
revealed that higher levels of abundance occurred in 
shorter tows at shallower, inshore stations where lar¬ 
vae are more concentrated than in longer tows at deep¬ 
er, offshore stations. 
Larval abundance in samples collected with the dif¬ 
ferent mesh sizes varied widely among the 6 taxa and 
the unidentified group chosen for analysis (Table 3). 
Overall, clupeid larvae were captured in the greatest 
numbers followed by sciaenids and the category for un¬ 
identified larvae. Lutjanid larvae were the least numer¬ 
ous taxon collected. Disparities between samples from 
nets with the 2 mesh sizes were greatest for sciaenid, 
unidentified, and percoid larvae, and least for lutjanid, 
engraulid, and clupeid larvae. These disparities were 
especially evident for unidentified and percoid larvae 
for which total larval abundance in samples from nets 
with 0.202-mm mesh was an order of magnitude great¬ 
er than in samples from nets with 0.333-mm mesh (Ta¬ 
ble 3). Although mean abundance varied significantly 
only for 2 of the groups examined (unidentified larvae 
and Percoidei), K-S tests revealed significant differ¬ 
ences in length-frequency distribution for larvae under 
10 mm for all taxa examined (Table 3). 
Larvae of taxa that spawn during late summer and 
early fall predominated in collections from the nets 
of different mesh size because of the preponderance 
of sampling in the months of September and October 
(Table 4). This temporal coverage resulted in greater 
availability of the smallest, least developed sciaenid 
larvae in samples taken with both mesh sizes. Scom- 
brid larvae were equally prevalent in spring (May) 
and late summer (September) samples. The presence 
of both spring-spawning taxa (Auxis spp. and Thunnus 
spp.) and protracted-spawning taxa, including the lit¬ 
tle tunny (Euthynnus alletteratus), king mackerel, and 
Spanish mackerel ( S. maculatus), increased the avail¬ 
ability of the smallest size category of scombrid larvae. 
Standardized abundances (number of larvae un¬ 
der 10 m 2 ) of all fish larvae taken in the nets with 2 
different mesh sizes during this study, combined and 
grouped by 0.1-mm size classes, indicate that larvae <3 
mm were consistently found in greater numbers in the 
nets with 0.202-mm mesh than in the nets with 0.333- 
mm mesh (Fig. 2). This was also the case for the cat¬ 
egories of unidentified larvae and larvae identified to 
the suborder Percoidei, for which most specimens were 
<2 mm in length (Fig. 2). Among the smallest larvae 
identifiable to 1 of the 5 target families, all but clupeid 
larvae were abundant at sizes <2 mm, and larvae in 
that size category were found in greater numbers in 
the finer-mesh net than in the coarser-mesh net (Fig. 
3). The smallest, most abundant size classes of clupeid 
larvae present in study samples ranged from 2 to 3 
mm, and those larvae were also found in greater num- 
