212 
Fishery Bulletin 95(2), 1997 
A 
Figure 3 
Diagrammatic representation of prawn-trawl and (A) Nordmpre grid, (B) fisheye, (C) square-mesh panel, and (D) 
extended mesh funnel (EMF) bycatch-reducing devices. T = transversals; B = bars; and N = normals. 
paired comparisons). The blubber chute comprised a 
panel of netting (36-ply, UV-stabilized, high-density 
polyethylene with a mesh size of 90 mm) sewn into a 
funnel (with an anterior circumference of 100 
meshes) located in a 2-m panel of mesh (mesh size of 
40 mm) measuring 150 meshes in circumference (see 
Broadhurst and Kennedy, 1996, for details). The pos- 
terior point of the blubber chute was attached five 
meshes from the end of the 2-m panel. A 30-mesh 
opening (termed the escape exit) was cut immedi- 
ately anterior to this point of attachment. 
As was the case for experiment 1, the position and 
order of each design was randomly determined and 
used in normal commercial tows of 30-min duration. 
Over 8 days, we completed a total of 23 replicate tows 
for each of the five paired comparisons. 
Data collected 
After each tow in each paired experiment, the two 
codends were emptied onto a partitioned tray. All 
organisms were sorted according to species. The fol- 
lowing data were collected from each tow: the total 
weight of prawns; the total weight of bycatch; the 
weights; numbers and sizes of commercially or 
recreationally (or both) important finfish (to the near- 
est 0.5 cm); the numbers of noncommercial or 
nonrecreational species; and the total numbers of 
noncommercial and commercial species in the assem- 
blage. All prawns in a subsample of the total prawn 
catch from each tow in experiment 2 were measured 
in the laboratory (to the nearest 1-mm carapace 
length). Several species were caught in sufficient 
quantities to provide meaningful analyses. These were 
the commercially important school prawns ( Meta - 
penaeus macleayi ) and large tooth flounder (Pseudo- 
rhombus arsius) and the commercially unimportant 
fortesque ( Centropogon australis), narrow banded sole 
( Synclidopus macleayanus), bridle goby ( Arenigobius 
bifrenatus), and catfish ( Euristhmus lepturus). 
Bata from all replicates that had sufficient num- 
bers of each variable (defined as >2 fish in at least 8 
