These aquaculture- and conservation-oriented commentaries are not abstracts written by the original authors.  They reflect the opinions of someone else -- usually Roger Doyle.  Direct quotations from the papers or abstracts are marked with inverted commas.

345.  Fish evolve to defeat the fishermen
         Sustaining fisheries yields over evolutionary time scales. 2002. Conover, D.O., and S.B. Munch. Science 297 (5578):94-96.
         This is a direct experimental proof that when you selectively harvest the larger fish in a population, the population will evolve in a way that reduces your catch. The experimental fishery consisted of a number of lab populations of the Atlantic silverside (Menidia menidia) which were selectively fished for four years.
         In one pair of populations all but the smallest 10% were harvested (mimicking a heavily exploited fishery), in another pair all but the largest 10% were harvested, and in the third pair the harvest was random with respect to size. This is, of course, just a selection experiment and the fish responded in the expected way: when only the small fish were left to breed the harvested "yield to the fishery" grew less and less as the generations passed. When the large fish were left to breed the yield increased.
         The idea that size-selective fishery (the usual practice) will reduce yields has been around for decades and indeed was included in Palumbi's "cost of evolution" estimate (Dec 2001 #262). Analogous situations are notorious in aquaculture -- in Chinese carps and Macrobrachium, for instance. However, this is the first time it has been proved to happen in an experimental model of a wild fishery. dconover@notes.cc.sunysb.edu

344.  Pacific salmon require special Ne calculations
         Effective size of fluctuating salmon populations. 2002. Waples, R.S. Genetics 161:783-791.
         The salmon species referred to are those which only breed once as individuals but mature in several different years as a cohort, so that generations overlap even though each year brings out an entirely new set of spawners. That includes chinook and chum, but not coho. Since there is a lot of variation among years in the total number of spawning adults (from several overlapping generations), how should we calculate the long-term effective population number, Ne? 
         The author explores this problem analytically and by simulation, and finds that variation in the number of breeders among years within a cohort enters the calculation as the harmonic mean, just like variation among generations in the well known non-overlapping generation model. The result is that Ne can be smaller even than the usual harmonic-mean estimate, and much smaller than the average census count. This is useful to know when evaluating the genetic status of endangered populations of Pacific salmon.
         The detailed calculations presented here may be about as far as one can go in estimating Ne in Pacific salmon solely from demographic data. However, it is likely that DNA marker data will soon be adequate to permit some sort of direct pedigree determination, or pedigree sampling, in non-experimental populations of salmon and other fish. When that happens, inbreeding and heterozygosity-by-descent can be calculated exactly from the pedigree and the loss of QTL diversity estimated, in even the most complicated situations, by gene-drop simulation. Who then will care about Ne? [To this facetious question the author has replied: "I think Ne will still have a lot of general interest as the "currency" of evolution, so that you make predictions about processes in unstudied systems"].   robin.waples@noaa.gov

343.  Growth rate and disease resistance negatively correlated
         Genetic variation for growth rate, feed conversion efficiency, and disease resistance exists within a farmed population of rainbow trout. 2002. Henryon, M., A. Jokumsen , P. Berg, I. Lund, P.B. Pedersen, N.J. Olesen, and W.J. Slierendrecht. Aquaculture 209 (1-4):59-76.
         An excellent experimental design for aquaculture genetic studies was used here -- an incomplete factorial in which each male was mated with two females and each female with two males. Each full-sib family was grown out separately, with periodic random culling to reduce the biomass density in the tanks. Feeding was restricted, with the ration determined by the anticipated growth rate.
         The results of the genetic (animal model) analysis of size-at-age and feed conversion efficiency were somewhat odd, in that the genetic variance and covariance components differed between the successive stages of growout, but the differences didn't show any easily interpreted pattern. Growth and feed conversion efficiency were observed to be positively correlated, which the authors ascribe to the dual effect of the same genes on both traits. Correlation would have been expected anyway, however, because of the way in which feeding rates were fixed. The authors discuss this problem. There were also problems with the genetic variance-covariance matrix which may (in my opinion) have been the result of competition among individuals in a tank. The statistically independent (individual) effects are not really independent at all. This problem would be exacerbated by restricted feeding but it may turn out to be insoluble with current genetic models and estimation procedures even when fish are fed ad libitum. Which is not to say that trout will fail to show a realized response to selection for increased growth.
         Additive genetic variation for survival was low in these experiments but overall survival was high except in the disease challenge experiments.
         The analysis of the viral haemorrhagic septicaemia (VHS) challenge experiments is an admirable model for aquacultural geneticists. The "Survival Kit" software of Ducrocq and Sölkner http://www.boku.ac.at/nuwi/software/softskit.htm  was used to estimate individual breeding values and population-level genetic variances and covariances for "frailty", a technical term which can be interpreted intuitively as the probability that an animal will die from the disease. Additive genetic variance for VHS resistance did exist and the estimated heritability was 0.13. This could be important in practical selection programs. "Only 63% of the fish from the sire and dam with high resistance were predicted to die after 21 days following challenge with VHS. By contrast ... all of the fish from the sire and dam with low resistance were predicted to die after only 12 days following challenge."
         Most interestingly, genetic variation for resistance to VHS tended to be negatively correlated with genetic variation for growth. This is potentially very important for the design of broodstock improvement programs, as the authors point out. The cautions expressed above about the growth-rate variance estimates also apply to the estimates of covariation between growth and VHS resistance, however. Mark.Henryon@agrsci.dk

342.  Initial advantage, terminal disadvantage to immigrant genes
         Heterosis and outbreeding depression in descendants of natural immigrants to an inbred population of song sparrows (Melospiza melodia). 2002. Marr, A.B., L.F. Keller, and P. Arcese. Evolution 56 (1):131-142.
         Does immigration help a small population by reducing inbreeding depression, or does it do harm by bringing in genes that are poorly adapted to local conditions? We hardly ever know the answer to this question in the real world. Therefore this new information on the sparrows living on the tiny island of Mendarte is very interesting. The resident population includes several hundred birds of each sex. The annual immigration rate is less than 5%.
         The effect of an occasional immigrant on genetic diversity and inbreeding depression of the population has been published earlier (Sep 2001 #235). Now it appears that the F1 hybrid offspring of residents and immigrants are somewhat more fit than the local residents in the same generation, a fact which the authors attribute to heterosis. The F2 generation, however, was much less fit than would be predicted merely from the breakdown of F1 hybrid vigour due to segregation of loci. The authors found "large performance differences between F1s and F2s, which suggested that either heterosis was associated with epistasis in F1s, that F2s experienced outbreeding depression, or that both phenomena occurred". The authors prefer outbreeding depression as an explanation – i.e. the breakdown of locally adapted gene complexes. amarr@interchange.ubc.ca

341.  Rapid immigration of genes caused by hybrid vigour
         A selective advantage to immigrant genes in a Daphnia metapopulation. 2002. Ebert, D., C. Haag, M. Kirkpatrick, M. Riek, J.W. Hottinger, and I. Ilmari Pajunen. Science 295 (5554):485-488.
         Should tiny, endangered populations of West Coast salmonids be kept pure and uncontaminated by stock from elsewhere? More generally, is immigration good or bad for small populations from a genetic point of view? Good, presumably, if the population is suffering from inbreeding depression and has drifted away from its fitness optimum. Presumably bad if the population is uniquely well adapted to its local environment. Complicated, if inbreeding depression purges the population of deleterious recessive genes and immigration then brings them back (e.g. Sep 2001 #235).
         This paper reports on a genetic study of Daphnia magna populations in small shoreline pools in Scandinavia. Every now and then one of the pools goes extinct and is recolonized from some other pool(s). Sometimes, when pools are recolonized by only one individual, there is a period of clonal reproduction before sexual reproduction takes place between identical, but not homozygous, individuals. This is a classic metapulation in which inbreeding and loss of genetic diversity are major evolutionary considerations.
         The authors experimentally manipulated immigration rates both in natural pools and in laboratory populations. They found that the rate of introgression of immigrant genes was enhanced by hybrid vigor of the offspring in almost all cases. That is, fitness increased, first at the individual and then the population level. Since different immigrant clones were used in each experiment the authors conclude the effect was due to hybrid vigour not to the selective superiority of particular alleles.
         "One key effect of hybrid vigor is an increase in effective gene flow ...[in these experiments] the effective rate of gene flow is about 35 times larger than would be predicted by the number of immigrants alone. ... An important effect of hybrid vigor is the "genetic rescue" of populations from extinction .... In this light, gene flow is an essential component for the persistence of metapopulations. Thus, our study gives clear empirical support for the need to maintain gene flow in the management and conservation of subdivided populations." dieter.ebert@unifr.ch

340.  Inbreeding is bad even in good environments
         Relationship between coefficient of inbreeding and parasite burden in endangered gazelles. 2001. Cassinello, J., M. Gomendio, and E.R.S. Roldan. Conservation Biology 15 (4):1171-1174.
         Pedigree records on captive populations of these three species of gazelle enabled the authors to calculate inbreeding levels exactly, rather than estimating them indirectly from heterozygosities (as for example in Apr-May 2002 #320, June 2001 #207, Mar 2000 #32). The species with the highest inbreeding also had the highest intensity of nematode infection. Within that species (but not the other two) there was a positive correlation between an individual's parasite burden and its inbreeding coefficient. The authors conclude that when inbreeding is high, it makes individuals more susceptible to parasites even in a relatively benign artificial environment. cass@mncn.csic.es

339.  Fast growing Frankenloaches
         Accelerated growth performance and stable germ-line transmission in androgenetically derived homozygous transgenic mud loach, Misgurnus mizolepis. 2002. Nam, Y.K., Y.S. Cho, H.J. Cho, and D.S. Kim. Aquaculture 209 (1-4):257-270.
         These fish are truly bioengineered. First a growth-promoting transgene (mud-loach growth hormone fused to mud-loach h-actin regulatory region) was introduced into the loach genome, then sperm from the transgenic loach were used to fertilize eggs of a related fish (common carp) in which the female genome had been destroyed by UV irradiation. Finally, a carefully-timed heat shock was used to double the male (loach) haploid genome. "One homozygous transgenic line displayed consistently faster growth (up to 1.6-fold) than its heterozygous transgenic counterpart. Conversely, growth acceleration in the other two homozygous transgenic lines was limited to the first 3 months of age only, and subsequently grew far slower than their heterozygous transgenic, especially when some physiological and morphological abnormalities began to appear." dongskim@nuri.net

338.  Selection of laid-back trout
         Differences in behaviour between rainbow trout selected for high- and low-stress responsiveness. 2002. Overli, O., T.G. Pottinger, T.R. Carrick, E. Overli, and S. Winberg. Journal of Experimental Biology 205 (3):391-395.
         Is it possible to speed the domestication of farmed trout by selecting a line which is indifferent to stress? Trout which showed either high or low plasma cortisol levels after a standardized stress were selected as breeders. The offspring of these divergent lines also differed in their mean post-stress cortisol, showing that the endrocrinological response to stress was heritable in the parental population. Offspring in the high-cortisol line (sensitive to stress) were more active than the low-cortisol line, but only when an intruding fish was present. "A significantly higher incidence of feed intake was seen in [low cortisol] trout when held in observation tanks (40 % versus 0 % of the fish took food when in isolation), suggesting that these fish acclimated more successfully to the experimental conditions than [high cortisol] fish did." So it appears that one-generation of selection for reduced physiological stress response produced trout that were less excitable and fed more than fish selected in the opposite direction. ooverli@usd.edu

337.  Fluctuating asymmetry important, but maybe not genetically important
         Fluctuating asymmetry is non-genetically related to mating success in Drosophila buzzatii. 2001. Santos, M. Evolution 55 (11):2248-2256.
         Fluctuating asymmetry (FA) is the occurrence of random differences between the left and right side of the body. Since a body usually has the same genotype on both sides, FA reflects random failures of developmental stability. It is known that stability can be decreased, and FA increased, by the non-heritable genetic stress of inbreeding, and by environmental stress (Jun 2000 #70). Because FA is easy to measure it is potentially useful as an indicator of poor genetic quality in aquaculture and genetic conservation. But in addition to inbreeding, is it also a signal of heritable differences in the fitness of animals in particular environments, e.g. in their reproductive fitness?
         In these experiments with laboratory and wild Drosophila populations, males with higher FA did indeed have lower reproductive success, but "the higher FA in single males is most likely due to a poorer average phenotypic condition because there was no evidence of a genetic basis for this trait".
         There is as yet no consensus on the usefulness of FA as a general indicator of genetic quality: Jun 2000 #70 FA not related to components of fitness in D. melanogaster; Jul 2000 #82 FA related to inbreeding and fitness in ungulates; Feb 2001 #167 FA not heritable in chinook salmon; Feb 2002 #292 not related to heterozygosity (inbreeding?) in partridge. This paper makes a useful methodological contribution to estimating FA when the animals also exhibit non-random variation in symmetry ("handedness"). mauro.santos@uab.es

336.  Selecting temperature-insensitive sex determination in bass
         Temperature effects and genotype-temperature interactions on sex determination in the European sea bass (Dicentrarchus labrax L.). 2002. Saillant, E., A. Fostier, P. Haffray, B. Menu, J. Thimonier, and B. Chatain. Journal of Experimental Zoology 292 (5):494-505.
         The European sea bass is an important aquacultural species. Females grow faster than males and are favoured for cultivation, but unfortunately the sex ratio is biased towards males. In the work reported here 27 sea bass families were produced according in a factorial mating design. Microsatellites were used to sort out the parentage of animals reared together. The sex ratio was affected by temperature, with more males developing when the fish were grown at lower temperatures. (The temperature effect on tilapia sex ratio goes the other way.) There was an interaction between the temperature effect and the developmental stage at which the treatment was applied. " In groups reared at high temperature, both parents had a significant additive effect on the percentage of females, and the interaction between sire and dam was not significant. Genotype temperature interactions were also detected and their existence suggests the interesting possibility of selecting nonsensitive genotypes in breeding programs." esaillant@wfscgate.tamu.edu

335.  Procedure for reducing inbreeding during artificial selection
         Non-random mating for selection with restricted rates of inbreeding and overlapping generations. 2002. Sonesson, A.K., and T.H.E. Meuwissen. Genetics Selection Evolution 34:23-39.
         The authors of this paper have applied (in theory and simulation) the concept of minimal kinship selection to the problem of artificially selecting a trait such as growth. Minimum kinship (= minimum coancestry mating; Dec 2001 #270, Nov 2001 #261) is generally used to reduce long-term inbreeding and genetic drift in a broodstock. One of the problems afflicting aquaculture is the high rate of inbreeding that results from various culture procedures which are hard or expensive to change. Fish in hatcheries conspire to inbreed even when they are not artificially selected. When they are selected the inbreeding problem gets a lot worse.
         This paper shows that when animals are selected on their merits (e.g. by phenotypic selection or BLUP) and the matings among them are arranged to minimize kinship, not only is the accumulation of inbreeding reduced but the rate of genetic gain is increased (consult paper for details). None of this would have been relevant to aquaculture 5 years ago because pedigree records are required to use these procedures. However, we can now use DNA markers instead of pedigrees to minimize kinship (e.g. January 2002 #283) and to estimate breeding values when animals are reared in common pools, as in aquaculture (Jun 2002 #330).
         Breeder populations of moderate size (a few hundreds), having one or few selected offspring per mating, are optimal for combining minimal coancestry with selection, according to this paper. Good news, because it is also the aquacultural setup which would lead to very high selection intensities and great savings in hatchery infrastructure. anna.sonesson@akvaforsk.nlh.no 

334.  Evolution of spawning date in captive Pacific salmon
         Artificial selection and environmental change: countervailing factors affecting the timing of spawning by coho and chinook salmon. 2002. Quinn, T.P., J.A. Peterson, V.F. Gallucci, W.K. Hershberger, and E.L. Brannon. Transactions of the American Fisheries Society 131 (4):591-598.
         Water temperatures in salmon hatcheries in the U.S. state of Washington have been getting warmer over the years. This is an environmental factor that would normally produce a later spawning date in the two species of Pacific salmon studied here. However, breeding and spawning procedures in the hatchery since the 1950s for chinook and 1960s for coho have been selecting for earlier spawning. Given that spawning date is known to be highly heritable, which way has it been moving? Records kept in three hatcheries including one run by Washington State University show that spawning is occurring earlier and earlier as the years go on. "Thus, inadvertent selection at all three hatcheries appears to have resulted in progressively earlier spawning, overcoming selection from countervailing temperature trends". tquinn@u.washington.edu