This paper further develops an integrated assessment for Patagonian toothfish in Division 58.5.2. It updates the model used at WG-FSA in 2006 using data from the 2007 season as well as 2006 data not available for WG-FSA in 2006. It also includes the following refinements: (i) estimation of the coefficient of variation (CV) for length given age, (ii) use of non-informative priors for year class strength parameters, (iii) separate selectivity parameters used for the pre-2006 compared to the 2006-2007 fishing seasons for the main trawl ground, (iv) separate selectivity parameters for the late (within-year) season compared to the combined early (within-year) seasons for the main trawl ground, and (v) the use of an improved method of determining effective sample size for commercial catch-at-length data. The estimated long-term yield was 2500 tonnes with depletion probability of 0.081 and escapement probability of 0.505. As expected, the assessment was sensitive to the inclusion of different datasets and to the choices of parameters used in both the stock assessment and projections. It is concluded that until the difficulties with the use of mark-recapture data are resolved, recruitment surveys provide the best means of establishing current stock status as an absolute index of abundance.