728 
Fishery Bulletin 95(4), 1997 
Table 3 
The relations between mouth width (MW) and body length (BL) for the larvae of five teleost species in Wilson Inlet. 
Species 
n 
Regression function 
r 2 
Urocampus carinirostris 
69 
MW = 0.139 + 0.006(£L) 
0.453 
Afurcagobius suppositus 
50 
MW = -15.666 + 14.256(BL) - 4.687(BL 2 ) + 0.671(BL 3 ) - 0.035(£L 4 ) 
0.849 
Pseudogobius olorum 
98 
MW = -0.729 + 0.839(£L) - 0.273 (BL 2 ) + 0.038 (BL 3 ) - 0.002 (BL 4 ) 
0.921 
Favonigobius lateralis 
50 
MW = 0.061 + 0.029(BL) + 0.005(£L 2 ) 
0.963 
Parablennius tasmanianus 
66 
MW = -0.517 + 0.41KBL) - 0.073(£L 2 ) + 0.005(BL 3 ) 
0.889 
Table 4 
The number of cases of significant dietary niche overlap (DNO) for larvae of five teleost species in Wilson Inlet between October 
1988 and April 1989. Co-occurring species were compared only when >10 individuals of each contained food in their guts. The 
number of cases for which pairwise comparisons could be made is shown as n. The number of significant cases are presented for 
DNO calculations I) that used zooplankton concentrations and II) that did not use zooplankton concentrations. Within these 
categories, the number of significant cases were determined a) at P<0.05 from a null distribution derived from bootstrapping and 
b) with an arbitrary cutoff level for significance at an overlap value >0.6. DNO was measured with a modification of Pianka’s 
(1973) symmetric niche overlap coefficient. 
Pseudogobius 
Afurcagobius 
Favonigobius 
Parablennius 
olorum 
suppositus 
lateralis 
tasmanianus 
n (I) (II) 
n (I) (II) 
n (I) (II) 
n (I) (II) 
(a, b) (a, b) 
(a, b) (a, b) 
(a, b)(a, b) 
(a, b) (a, b) 
Afurcagobius suppositus 
5 (0, 2) (0, 0) 
— 
Favonigobius lateralis 
4(1, 2) (1, 2) 
1 (0, 0) (0, 0) 
— 
Parablennius tasmanianus 
8(1, 4) (3, 6) 
1 (0, 0) (0, 0) 
4(2, 4) (3, 4) 
— 
Urocampus carinirostris 
8 (7, 7) (8, 8) 
2(1, 1) (0, 1) 
4(1, 2X2, 3) 
6(1, 4) (2, 4) 
Totals 
n (I) (II) 
(a, b) (a, b) 
43 (14, 23) (20, 28) 
Percent of total 
(32.6, 53.5) (46.5, 65.1) 
prey items consumed by P olorum <4.0 mm BL were 
similar to mouth width, as was also the case with F. 
lateralis of 2. 0-2. 5 mm BL and A. suppositus of 4.0- 
4.5 mm BL. For each of these three gobiid species, 
the maximum prey width of larvae >5 mm BL was 
far less than mouth width. This difference exceeded 
0.20 mm in the larger gobiid larvae. Likewise, maxi- 
mum prey widths for larval P. tasmanianus ap- 
proached mouth width in smaller larvae but were 
much less for larger larvae (Fig. 2D). 
Dietary niche overlap 
Dietary niche overlap between P. olorum and U. 
carinirostris ranged from 0.543 to 0.983 and was sig- 
nificant on seven of the eight occasions in which these 
species co-occurred (Table 4). Although DNO ranged 
from 0.764 to 0.980 on the four occasions that P 
tasmanianus and F. lateralis co-occurred, overlap was 
significant only twice (Table 4). There were a few 
other cases of significant DNO amongst the larval 
fish assemblage; DNO was particularly low between 
A.suppositus and the other species, being significant 
only with U. carinirostris on one occasion. 
Of the 43 pairwise comparisons that could be made 
between the diets of co-occurring larvae of the five 
species, there were 14 cases (32.6%) of significant 
DNO (Table 4). This increased to 20 (46.5%) if zoo- 
plankton data were not included in the calculations 
of DNO’s. If DNO >0.6 had been considered signifi- 
cant, the number of significant cases increased from 
32.6% to 53.5% for calculations which included zoo- 
plankton data and from 46.5% to 65.1% for those 
which did not include these data. 
The magnitudes of the differences for the preva- 
lence of significant DNO found by using bootstrap- 
