Gorfine et a I,. Two methods for estimating abundance of Haliotis rubra 
445 
Table 3 
Analysis of variance for interannual differences in abundance estimates of H. rubra for each of Victoria’s fishery management 
zones, made from radial transect surveys, during 1992-94 (type TV SS). ns = nonsignificant; * = PcO.lO; ** = P<0.05; *** = P<0.01. 
( o 2 = relative magnitude of variance estimate (%). 
Management zone 
Source 
df 
MS 
F 
P 
of 2 
Significance 
Central 
Year 
2 
515.43 
4.08 
0.0222 
3 
** 
Site 
29 
378.00 
5.04 
0.0001 
54 
*** 
Diver 
10 
93.24 
1.24 
0.2611 
1 
ns 
Diver x site 
123 
47.16 
0.63 
0.9989 
21 
ns 
Year x site 
56 
126.36 
1.69 
0.0022 
18 
*** 
Year x diver 
5 
31.23 
0.49 
0.7842 
1 
ns 
Year x site x diver 
35 
64.17 
0.86 
0.7066 
2 
ns 
Residual 
479 
75.00 
Eastern 
Year 
2 
130.23 
0.87 
0.4312 
0 
ns 
Site 
14 
801.99 
5.33 
0.0001 
38 
*** 
Diver 
5 
1170.54 
7.78 
0.0001 
21 
*** 
Diver x site 
46 
274.05 
1.82 
0.0021 
23 
*** 
Year x site 
28 
150.21 
1.00 
0.4728 
0 
ns 
Year x diver 
4 
301.05 
5.22 
0.0057 
11 
*** 
Year x site x diver 
18 
57.69 
0.38 
0.9899 
7 
ns 
Residual 
242 
150.48 
Western 
Year 
2 
36.54 
2.30 
0.1214 
1 
ns 
Site 
12 
261.00 
7.14 
0.0001 
41 
*** 
Diver 
7 
100.08 
2.74 
0.0096 
7 
*** 
Diver x site 
40 
47.61 
1.30 
0.1201 
7 
ns 
Year x site 
24 
147.51 
4.03 
0.0001 
40 
*** 
Year x diver 
6 
65.70 
1.31 
0.3016 
1 
ns 
Year x site x diver 
18 
50.04 
1.37 
0.1201 
4 
ns 
Residual 
225 
36.54 
divers’ mean transect collections differed significantly 
from the remaining nine divers during three annual 
abalone stock surveys. During routine monitoring, 
two divers (4 and 5) consistently collected less aba- 
lone than the others and frequently aborted dives 
after having failed to complete all their allocated 
transects. Reasons given for terminating their dives 
included problems with middle ear equalization, dif- 
ficulty coping with the surge and backwash created 
by waves, and fouling of lines and air hoses due to 
entanglement. These observations concur with Shep- 
herd (1985) who found that surge and kelp density 
were two significant factors affecting research diver 
efficiency in underwater censuses of abalone popu- 
lations. In contrast to divers 4 and 5, diver 8 consis- 
tently collected more abalone than the other research 
divers. However, diver 8 participated only in those 
parts of the study conducted in the eastern zone, 
where abalone stocks are generally more abundant. 
Also, as an ex-abalone diver, diver 8 could reason- 
ably be expected to possess a superior ability in per- 
forming abalone searches and collections that may 
have biased his estimates. 
These observations suggest that some divers may 
not possess the aptitude required to conduct under- 
TabJe 4 
Mean number (least squares) of H. rubra collected per 
transect by each diver during 1992-94 Victorian abalone 
stock surveys. Vertical bars in “ns" column indicate groups 
of means that were not detected as significantly different 
by Ryan’s test (a=0.05). Is = least squares; SE = standard 
error of the least squares means. 
Diver 
n 
Is mean 
± SE ns 
8 
36 
19.48 
1.92 
9 
147 
16.03 
0.95 
1 
467 
12.62 
0.53 
12 
38 
12.00 
1.87 
3 
408 
11.26 
0.57 
11 
66 
10.74 
1.42 
7 
6 
10.50 
4.71 
6 
17 
10.03 
2.80 
2 
119 
10.02 
1.06 
10 
15 
9.73 
2.98 
4 
107 
7.83 
1.12 
5 
12 
3.58 
3.33 
water censuses and highlight the need for under- 
standing the nature of research being conducted. The 
ability to work at sea in often arduous conditions, 
discipline in adherence to sampling protocols, and 
