In the most recent assessments (Rademeyer and Butterworth, 2006) of the South African hake resource, Merluccius paradoxus and M. capensis are treated as two separate stocks, but are assessed simultaneously within a single assessment framework and for the south and west coasts combined. This simultaneous assessment is necessary because much of the data is available in species-aggregated form only. Thus the model is one of two species and two spatial strata (see Fig. 1) with differences in the distributions by age within each stratum handled by allowing for stratum-specific commercial (in principle, though not in this particular implementation) and survey selectivities, rather than explicitly modelling movement. This follows the recommendation from the January 2004 BENEFIT/NRF/BCLME workshop (BENEFIT, 2004), though the further recommendation of that workshop to extend to four spatial strata (two by depth as well as two longshore) has yet to be implemented. The only data available which are explicitly disaggregated by species are those from research surveys that have taken place from 1986 to the present. However the framework does admit implicit disaggregation of data from the commercial fishery as summarised below.