These aquaculture-oriented commentaries reflect the opinions of GCL associates -- usually Roger Doyle -- and are not the original abstracts of the papers.  Direct quotations from the papers or abstracts are marked with inverted commas.

19. Competitive ability depends on the neighbours
Group size and relative competitive ability: geometric progressions as a conceptual tool. 2000. Humphries, S., G.D. Ruxton, and N. B. Metcalfe. Behavioral Ecology and Sociobiology 47:113-118.
An ingenious mathematical and experimental study shows that there is no simple, general relationship between the relative competitive ability of two individual fish and the size of the group of fish they are living in. Nevertheless, group size does make a tremendous difference in relative competitive ability, and the optimal group size for each fish will depend on its relative size. (It has to be said that this study might be enhanced by a game-theoretic analysis of what is optimal.) In any case, for aquaculture, it suggests that even more attention should be paid to size-grading and tank size. For aquaculture genetics it gives additional support to the idea that competitive interactions have totally fouled up conventional heritability estimates and classically-designed selection programmes based on related "independent errors" assumptions. For biological conservation, it is one more thing to think about when trying to design a hatchery programme for supplementing wild stocks that won't domesticate the fish. s.humphries@bio.gla.ac.uk

18. Unlimited effective population size
Marker-Assisted Selection to Increase Effective Population Size by Reducing Mendelian Segregation Variance. 2000. Wang, J., and W.G. Hill. Genetics 154:475-489. Everyone knows that the effective population size of a broodstock (the size that determines the rate of inbreeding and genetic drift) is a "virtual" number. When you select an equal number of offspring from every family the effective population size is approximately twice the actual size. It had been supposed that this as good as you could do to minimize inbreeding in genetic conservation, aquaculture broodstock management etc. (Numerous mating schemes have been proposed which do somewhat better than this, but at such tremendous rates of initial inbreeding that they are impractical.)
The authors of this highly technical paper have now shown by theory and simulation that the effective population size can be made indefinitely large with the aid of marker-assisted selection. The procedure should work best when the number of offspring is enormous, as in aquaculture. Unfortunately, it is completely impractical at the present time. In 15 years or so, when marker maps have been filled in for the major species and gene-chip technology is cheap, our inbreeding problems may be solved. jinliang.wang@ed.ac.uk

17. Better to know about house flies than to know nothing at all
Experimental tests of minimum viable population size. 2000. Reed, D.H., and E.H. Bryant. Animal Conservation 3:7-14. Another experiment with house flies. Some populations were maintained at a constant size, some put through a severe bottleneck (down to 5 individuals) then allowed to expand rapidly. The bottlenecked populations recovered quickly from their initial inbreeding depression. "However, they were less adaptable to environmental stress than constant size populations, suggesting that populations founded with few numbers may do well within a single environment but may do far less well if they are reintroduced to natural environments or exposed to rapid environmental changes." This has obvious consequences for genetic conservation and perhaps also for aquaculture. Also, perhaps, for fisheries stock-enhancement programs.

16. Markers for the traits that matter
A logistic regression based extension of the TDT for continuous and categorical traits. 1999. Waldman, I.D., B.F. Robinson, and D.C. Rowe. Annals of Human Genetics 63:329-340. I've always argued that within-family selection is the best way to go in aquaculture. These authors have come up with a procedure for finding statistical associations between markers and continuous traits such as growth and fecundity, a procedure that works within families. (Other within-family procedures work only with categorical yes-or-no traits.) I think it is about time research stations or even farmers should start collecting and storing fin-clips from fish within families. Just on a speculative basis. When marker maps are adequate, these samples and their associated size and rudimentary pedigree data might be extremely valuable 15. Inbreeding hits fitness first
A comparison of inbreeding depression in life-history and morphological traits in animals. 1999. DeRose, M.A., and D.A. Roff. Evolution 53:1288–1292.
Traits related to fitness (in aquaculture these would be growth, survival and fertility) show greater inbreeding depression than other traits (shape, colour etc.). This re-analysis of data from many published research papers demonstrates large amounts of dominance variance in fitness traits (see also #10), which is significant for broodstock improvement. It is also another indication that inbreeding can cause big problems in small populations. Perhaps by the time aquaculturists begin to notice development (shape) abnormalities and deformities which they ascribe to inbreeding, they are already losing much more production due to mortality and slow growth than they realize.

14. Hatchery salmon contaminating wild salmon
Genetic changes in Atlantic salmon (Salmo salar) populations of northwest Irish rivers resulting from escapes of adult farm salmon. 1998. Clifford, S.L., P. Mcginnity, and A. Ferguson. Can. Jour. Fish. Aquatic Sci. 55:358-363.
Escaped farm salmon did interbreed with wild salmon. The proportion of juveniles of maternal farm parentage in two rivers averaged 7% and reached a high of 70% in an individual sample. "Only a small proportion of 29 000 adult farm salmon that escaped in spring 1992 appear to have bred successfully in the rivers studied." Note that the authors found fish whose mothers came from the farm.

13. Someone really smart should count the dead fish in advance
Effect of Selection Against Deleterious Mutations on the Decline in Heterozygosity at Neutral Loci in Closely Inbreeding Populations. 1999. Wang, J., and W.G. Hill. Genetics 153:1475-1489.
Highly technical, but unusually realistic, computer simulation. Shows that natural selection does in fact "purge" a population of genes that depress fitness when the population is randomly and partially inbred, and that this can be observed at marker loci. The simulation is realistic enough to resemble aquaculture in that not all animals in the same generation have the same inbreeding. The question for aquaculture is, what is the economic cost of all the dead animals that are purged out before they're harvested? Should be possible to estimate and/or simulate this too.

12. Local adaptation made impossible
Gene flow and local adaptation in a sunfish-salamander system. 1999. Storfer, A. Behavioral Ecology and Sociobiology 46:273-279
The adaptation of salamanders in local populations to fish predators (a genetic predisposition to hide) is overcome by gene flow from nearby salamander populations where there are no predators. Analogous to the situation in which huge numbers of salmon escaping from aquaculture may (might perhaps possibly maybe sometimes) reduce the fitness of natural populations.

11. Good environments accumulate bad genes?
Fitness Decline under Relaxed Selection in Captive Populations. 1998. Bryant, E. H., and D.H. Reed. Conservation Biology 13:665-669.
A neat laboratory study on house flies in which the authors relaxed selection on genes that affect fitness late in life by a simple expedient -- killing the flies young. Late-life fecundity declined drastically after a few generations of this treatment. This would seem to be relevant to aquaculture. If natural selection is reduced by improving the environment or equalizing the size of families produced by aquaculture broodstock, there may be a rapid reduction in the ability of these populations to exist under natural conditions. The authors note that this "domestication" effect is independent of population size. It is a genetic problem that occurs because new, bad mutations are allowed to hang around, and is somewhat independent of the deleterious effects of inbreeding and drift.

10. Plenty of selectable genetic variance for fitness
Genetic architecture of fitness and nonfitness traits. 1999. Merila, J., and B. C. Sheldon. Heredity 83:103-109.
It used to be thought that traits related to fitness (survival, fecundity etc.) would not respond to selection because additive genetic variance would be low (mutations with good or bad effects already having been selected to frequencies of one or zero, respectively). Professor Rom Moav used this idea 20 years ago to explain why his carp selection experiment made no progress. It seemed plausible to many people at the time. However, recent studies - none in aquaculture - have completely reversed this view. Fitness traits have more additive genetic variance than other traits, probably because there are a lot more genes involved. However, they also have a lot more dominance variance as well, which reduces heritability.

9. If you're inbred you're going to die of parasites in hard times
Parasite-mediated selection against inbred Soay sheep in a free-living island population. 1999. Coltman, D.W., J.G. Pilkington, J.A. Smith, and J.M. Pemberton. Evolution 53 (4):1259–1267
Small population of wild sheep on a remote North Atlantic island. The individual sheep that are inbred are more susceptible to parasitism by gastrointestinal nematodes at high population density. During periods of high overwinter mortality highly parasitised individuals were less likely to survive. More inbred individuals were also less likely to survive, which is due to their increased susceptibility to parasitism. Dave Coltman's paper is causing a general stir.
Extension to aquaculture would be that the effect of inbreeding should show up under stress, in those individuals that are most inbred, and would be manifested in mortality from parasites and disease?. Point is, this can be investigated on working fish farms by a properly interpreted assay of microsatellite heterozygosity. For future reference fish farmers should  start storing tiny bits of tissue from their morts and healthy fish in 95% alcohol. The depressing effects of inbreeding can be identified.