by
J.A. Pereiro
Spanish Institute for Oceanography
Madrid, Spain
To assess the effects of a change in mesh size from 40 to 60 mm and to estimate the sustainable yield corresponding to different assumptions regarding the exploitation rate, a yield per recruit computation was carried out. Por that purpose the following parametric values were used in a linear Ricker model:
Growth: The two growth curves presented by Messrs Guerra and Hatanaka to the Working Party (section 3.1) were successively used:
1. total length = 130.19 [1  e ^{.423 (t +.435)}] presented by Hatanaka for the growth of the Cape Garnett spring recruits:To convert mantle length into total length and viceversa, the following equation was used:2. mantle length = 40 [1  e ^{.72 (t.34)}] presented by Guerra for the growth curve of Cape Garnett stock.
mantle length = 14 x (total length)^{1.08}The following length/weight relationships were also utilized:
W =.00497 (total length)^{2.9259 }Selectivity: A selection factor equal to 1.7 was adopted (Guerra, Appendix 6). With this factor, the lengths at first capture corresponding to a 40 and a 60 mm mesh size are 7 and 10 cm, respectively.
W =.976 (mantle length)^{2.691}
Natural mortality: A value of M =.2 on a yearly basis was used. This assumption was based on the observations made in section 3.2.1 regarding the likely rate of octopus exploitation phase and according to which every female would die after her first spawning season, and would remain low during the short exploited phase.
It was assumed as well, from sexratio studies (Hatanaka, section 3.2.1) that every female dies when reaching a mantle length of 18.5 cm. Hales, on the contrary, were supposed to suffer a natural mortality of M .2 along all their life span.
Exploitation rate: Production models seem to indicate that the exploitation rate is very high. Similar computations made by H. Hatanaka (pers. communication) lead to the same conclusion. With M =.2, the corresponding figures for F are.5, 1.0 and 2.0 respectively.
The final yields per recruit are given in tables1 and 2; they are expressed in percentages of the lower result obtained with each growth curve. Estimates given in table 1 correspond to computations made with Guerra's growth curve; those in table 2 correspond to the Hatanaka's curve. In both cases highest yields per recruit are obtained for an exploitation rate of 8 and a 60 mm mesh size. However, for all combinations of input parametric values, a gain is to be expected on the longterm by increasing the mesh size from 40 to 60 mm.
Considering the lack of accuracy of the various parameters used in the computation, these results should however be considered as provisional.
CAPE GARNETT OCTOPUS YIELD PER RECRUIT (LINEAR RICKER MODEL)
Table 1 Guerra's growth equation
E 
.7 
.8 
.9 
Mesh size (mm) 



40 
117 
123 
100 
60 
123 
138 
111 
E 
.7 
.8 
.9 
Mesh size (mm) 



40 
155 
1_{5}0 
100 
60 
178 
198 
162 