Fowler et al.: Distribution and abundance of tuna larvae in near-reef waters of the Coral Sea 
407 
Map of the study area in the Great Barrier Reef Lagoon and Coral Sea, 
near Lizard Island, Australia. Lagoon samples were taken between Lizard 
Island and the outer barrier reefs, within the boundaries indicated by 
dashed lines. Offshore samples were taken in five blocks, A-E, in the 
Coral Sea. The outer reefs are 1) Day, 2) Carter, 3) Yonge, 4) No Name, 
and 5) Number 10 Ribbon. Map adapted from Leis et al. (1987). 
occasionally in block A because of great variation in 
water depth in this area. Samples were fixed in formalin 
(5-10% in seawater) in the field. 
Laboratory procedure 
Larvae from both the port and starboard sides of the 
bongo net were sorted, except for those in the lagoon 
samples from the first and second cruises where the 
catch from only one randomly chosen side was sorted 
because of high plankton volumes. Samples from block D 
were not sorted because of funding cuts to the research 
program. Larvae were removed with the aid of a dis- 
secting microscope and transferred to 70% ethanol for 
storage. Tuna larvae (family Scombridae) were identi- 
fied to species, when possible, by using the descrip- 
tions of Fritzsche (1978) and Nishikawa and Rimmer 
(1987). Larvae of T. alhacares and T. alalunga (albacore) 
<3.2 mm standard length (SL) could not be separated 
and were identified as Thunnus spp. larvae. For larger 
larvae, Richards et al. (1990) advocate also using an 
osteological character, rather than relying solely on 
pigment, when identifying Thunnus larvae to species. 
However, their study was primarily concerned with T. 
atlanticus (blackfin tuna), a species that is not found in 
our study area. Further, the osteological character in T. 
alalunga and T. alhacares seems to vary about as much 
as does the pigment character, and the former cannot 
be used in specimens <6 mm SL (Richards et al., 1990). 
Therefore, we relied on the pigment character to separate 
our T. alalunga and T. alhacares larvae >3.2 mm SL. We 
note, however, that one-third of the Thunnus larvae 
>6 mm SL that we cleared and stained had pigment 
inconsistent with the osteological characters listed in 
Table 1 of Richards et al. (1990). Therefore, according 
to the criteria of Richards et al. (1990, Table 1), at least 
67% of our Thunnus larvae are correctly identified to 
species, and between 0 and 33% may be misidentified 
to species. Given the variability in osteological features 
noted by Richards et al. (1990, their Tables 2 and 3), 
we cannot be more precise about this “uncertain” 33%. 
Larvae of the genus Auxis cannot currently be identified 
to species (Nishikawa and Rimmer, 1987), and adults 
of both A. thazard (frigate tuna) and A. rochei (bullet 
tuna) inhabit the region of the study area (Collette and 
Nauen, 1983). Auxis larvae were therefore all identified 
as Auxis spp. Larvae of Auxis spp. and Euthynnus affinis 
(kawakawa) <2.3 mm SL could not be separated and 
were identified as Auxis-Euthynnus larvae. Notochord 
length and SL were measured to the nearest 0.1 mm for 
preflexion and postflexion larvae, respectively, by using 
a calibrated ocular micrometer. No correction was made 
for shrinkage of the larvae. 
Statistical analyses 
Abundances (no. of larvae/100 m 2 ) were calculated 
by 1) calculating larval concentration (larvae/m 3 ), 2) 
multiplying concentration by the depth (m) sampled, 
and 3) multiplying the result by 100 to obtain appro- 
