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.

297.  News flash: "supportive breeding" need not increase inbreeding and drift
         Genetic effects of multiple generations of supportive breeding. 2001. Wang, J., and N. Ryman. Conservation Biology 15 (6):1619-1631.
         In 1991 Ryman and Laikre published a paper which demonstrated that the effective population number (Ne; Feb 2000 #18, Aug 2001 #212) of an endangered, wild population would actually be decreased by supplementing it with hatchery stock (1991, Conservation Biology 5:325-329). This bad news has strongly influenced the thinking of resource managers and has helped fuel doubts about the wisdom of stock supplementation. The 1991 paper also triggered an interesting, and ultimately helpful, debate about the relationship between effective number as it affects genetic drift and effective number as it affects the accumulation of inbreeding.
         The 1991 paper dealt with the first generation of stock supplementation. In this follow-up paper, Wang and Ryman show, both analytically and by simulation, that the reduction in effective number (both drift and inbreeding) is quickly reversed in later generations as long as the census size of the population is increased by supplementation. (Census size is the actual number of animals, including those breeding in the hatchery and the wild.) This proviso will usually be met because supplementation would be pointless if it doesn't make the population bigger.
         The authors also consider an important practical decision: should hatchery breeders be chosen at random, or should they be selected from those whose parents bred in the wild, or should they be selected from those whose parents bred in the hatchery? The second course is usually recommended and followed, on the grounds that it helps prevent unwanted adaptation to the hatchery environment. The third course is one I recommended years ago, more or less tongue-in-cheek, on the grounds that since we seem to have taken permanent control of the environment we should accept our responsibilities and help fish get used to it: Doyle et al. 1995. In Uses and Effects of Cultured Fishes in Aquatic Ecosystems, edited by H. L. H. Schramm and R. G. Piper: American Fisheries Society, pp. 205-211.)
         Anyway it turns out that whichever selection procedure is followed the effective population number will increase. This should be very good news for conservationists. Of course effective population number only measures one of the problems potentially caused by supplementation -- domestication selection and ecological problems remain.  nils.ryman@popgen.su.se 

296.  Successful selection for higher growth rate in shrimp
         Selection for better growth of Penaeus stylirostris in Tahiti and New Caledonia. 2002. Goyard, E., J. Patrois, J.M. Peignon, V. Vanaa, R. Dufour, V. Viallon, and E. Bédier. Aquaculture 204 (3-4):461-468.
         This important paper shows that a straightforward mass selection program can produce useful genetic gains in growth rate in shrimp. Selection intensities, which ranged between 4% and 18%, were not especially high. Furthermore the strain is known to have lost some diversity during the early years of domestication and to be rather inbred (June 2000 #71). Despite these potential disabilities the program produced a 21% increase in growth rate by the fifth generation, for a realized heritability of about 0.11.
         The authors also ran experiments aimed at determining the correlation between early and late selection. These experiments involved grading shrimp at various sizes and ages and using fluorescent elastomer tags. The authors conclude that selection of fast-growers at an early stage, when all the shrimp are small (averaging about one gram) is likely to be effective in increasing the size at harvest. If is confirmed this is important information. Selection at harvest size is impracticable for commercial shrimp farmers in Tahiti because of space limitations.
         The authors do acknowledge, however, that their size-at-age experiments were phenotypic and required some "unnatural" manipulations like artificially holding back the growth rate of some individuals. The actual selection experiments which can measure the genetic correlation between early and late selection are apparently underway. egoyard@ifremer.fr

295.  An Indian carp species shows "reverse domestication"
         Growth and survival of six stocks of rohu (Labeo rohita, Hamilton) in mono and polyculture production systems.
         2002. Reddy, P.V.G.K., B. Bjarne Gjerde, S.D. Tripathi, R.J. Jana, K.D. Mahapatra, S.D. Gupta, J.N. Saha, M. Sahoo, S. Lenka, P. Govindassamy, M. Rye, and T. Gjedrem. Aquaculture 203 (3-4):239-250.
         Growth and survival of a domestic Rohu stock and five wild stocks were compared in two farm environments: alone (monoculture) and in polyculture with other Indian carp species. Polyculture is the usual farm environment in India. The domestic stock, with a long farming history, performed worse than the five wild stocks in both environments. The probable explanation, as the authors say, is the long history of inbreeding and negative domestication selection in Indian broodstocks.
         Figs. 1 and 4 in the paper reveal that there were many changes in rank among stocks when they were grown in the different environments. The genotype (stock) x environment interaction was significant (p<.01) and highly significant (p<.001) in the two year classes studied. Not only was the growth of the domesticated stock generally worse than the others, but the proportional decrease in polyculture, relative to monoculture, appears greatest for the domesticated stock. The absolutely worst-growing stock in the whole experiment was the farm stock growing in the environment to which it had been habituated for more than thirty years.
         Growth x environment interaction therefore adds support to the idea that domestic evolution has been messing up in India.
         Paternal as well as maternal half-sib families were generated in the experiment but only the full-sib data (families with one sire and one dam) were analysed. The publication of genetic inferences based solely on full-sib data, when half-sib data are available, is puzzling, especially when the work was done on carp.
         "The substantial and highly significant full-sib effect [i.e. variation among families] ... indicate a significant additive genetic variance within the stocks." "The significant full-sib effects ... suggest [growth as well as survival] can be improved through selective breeding." "The interaction between production system and stock on growth and survival was significant but low, particularly compared to the significant full-sib effects." And finally, a conclusion so valuable that it may justify the means of analysis, "Therefore, the development of specialized varieties for each of the two production systems is not required." Reddy e-mail: cifa@ori.nic.in

294.  How to compare the growth of strains in aquaculture
         Comparison of growth performances of three French strains of common carp (Cyprinus carpio) using hemi-isogenic scaly carp as internal control. 2002. Vandeputte, M., E. Peignon, D. Vallod, P. Haffray, K. Komen, and B. Chevassus. Aquaculture 205 (1-2):19-36.
         This paper is oriented towards an evaluation of the internal control technique for comparing strains. No strain differences were actually found. The internal control is a "reference strain" which is visibly distinguishable from the test strains and which is included in all the experimental units such as cages, tanks, ponds etc. The growth of the test strain is compared with the growth of the reference strain in the same units, which eliminates much of the environmental variance.
         Data were analysed in various ways by linear statistical models, and a power analysis was also performed. The discussion of the statistical analyses is particularly clear and thorough. The authors found that, "The use of the internal control divides the number of replicates needed to have a good statistical power by 5 at 5 weeks [of age] and by 8 at 2 summers [of age], when compared to separate testing without control".
         They put this in a practical context by saying"... this type of design with an internal control is more efficient than a classical design by blocks. It is also much more versatile because if we consider that the environment estimation by the control is correct, this allows the use of designs ''in the field'' without replication in a given location, or even with some strains not present in all locations, without hampering the possibilities of correctly estimating environment effects. This would be of special interest when there is no experimental facility with a large number of standardized ponds."
         There are, of course, some difficulties and problems which are also well discussed, mainly concerning the necessity of exactly matching the size and condition of the control and test lines. mvande@jouy.inra.fr

293.  Successful selection for growth rate in tilapia
         Response to within family selection for body weight in Nile tilapia (Oreochromis niloticus) using a single-trait animal model. 2002. Bolivar, R.B., and G.F. Newkirk. Aquaculture 204 (3-4):371-381.
          The experiment, conducted in the Philippines and analysed here after 12 generations, involved within-family selection, rotational mating, and preliminary, and pre-selection size grading ("collimation", Aquaculture 57:27-35, 1986) of the selected stock." A realized heritability estimate of 0.14 was obtained based on the regression of mean breeding values on cumulative selection differentials after 12 generations. The inbreeding coefficient was 6.3% after 12 generations with an average inbreeding of 0.525% per generation. The family rotational mating used to propagate the families was effective in keeping the inbreeding rate to a minimum even at high selection intensities." Within-family selection permitted high selection intensities of about 2%. The result seems to have been that the size of the fish aged 16 weeks more than doubled after 12 generations. rbolivar@mozcom.com

292.  Fluctuating asymmetry measures environmental stress, not genetic stress
         Relationship between heterozygosity and asymmetry: a test across the distribution range. 2001. Kark, S., U.N. Safriel, C. Tabarroni, and E. Randi. Heredity 86 (2):119-127.
         Fluctuating asymmetry (FA: random developmental differences between the left and right sides of an organism) may be induced by genetic stress such as inbreeding and/or lack of genetic diversity, or environmental stress such as high temperature. In either case FA is, potentially, a useful indication that something is going wrong in a hatchery or conserved population (see brief explanation in the June 2000 list #70). Recent papers reviewed here have shown FA is not related to fitness components in Drosophila (June 2000 #70), is related to inbreeding in ungulates (July 2000 #82) and not itself a heritable trait in Chinook salmon (February 2001 #167).
         This work describes the relationship between the FA of a morphological trait and allozyme heterozygosity in a species of partridge "across a sharp climatic gradient in Israel from the arid periphery, through the Mediterranean–desert ecotone towards the Mediterranean areas located further away from the range boundaries. Genetic diversity, as estimated using both observed and expected heterozygosity, was not associated with asymmetry at either the population or at the individual level. We argue that whereas asymmetry may serve as a useful tool for estimating changes in environmental stress, it may not be widely applicable for estimating genetic stress." salit@stanford.edu

291.  Hybrids between locally-adapted populations are not always unfit
         High larval performance of leopard frog hybrids: effects of environment-dependent selection. 2001. Parris, M.J. Ecology 82 (11):3001-3009.
         Are natural hybrids unfit in nature? The theoretical answer depends on a number of factors including the degree of genetic differentiation between species, environmental specificity of adaptations, habitat variation, genetic architecture of isolating mechanisms, genetic architecture of the trait itself (e.g. compatabilities, dominance) ... and the list goes on. Thus carefully-collected, empirical evidence is very useful.
         In this experiment two frog species and their F1 hybrids were grown in experimental enclosures in three habitats characteristic of the parental species and their naturally-occurring hybrids in central Missouri. "There was no evidence of reduced F1 hybrid larval performance in any environment.... Thus, hybrid genotypes performed equivalently to or better than their parental species in three aquatic habitat types. Transplant experiments such as these are critical for directly measuring genotype-by-environment interactions and for assessing the role of ecological factors in determining the adaptive potential of natural hybridization."
         One should note that the risk that hybridization poses to the genetic purity of the parental species also depends on the nature of pre-zygotic isolating mechanisms, if any, and on the degree of fitness decrease (if any) in F2 and backcross generations. Risks may also be hiding in the genetics of correlated fitness traits which were not studied. mparris@memphis.edu

290.  Success (but potential problem) selecting shrimp for growth and survival
         Selective breeding of Pacific white shrimp (Litopenaeus vannamei) for growth and resistance to Taura Syndrome Virus. 2002. Argue, B.J., S.M. Arce, J.M. Lotz, and S.M. Moss. Aquaculture 204 (3-4):447-460.
         Two vannamei lines selected at the Oceanic Institute have produced very interesting results. The growth rate of the line selected for growth increased dramatically in one generation (21%, giving a realized heritability of 1.0!). The survival of the line selected mainly for survival increased by 7 percentage points (18% of the initial value), but the growth rate of this line declined by about 5%. Similarly, survival in the line selected for growth was slightly, but non-significantly, lower than the control.
         This experiment certainly gives additional scientific weight to anecdotal evidence that growth and survival (to TSV challenge) respond well to selection, but that there is some sort of negative genetic correlation between these traits. Genetic correlations are notoriously difficult to estimate precisely because of error variances which are intrinsically large. (Also see #289, below.)
         The splitting of a selected broodstock into two lines with different selection criteria is a good way to estimate realized correlations but may not be the best way to increase overall economic value in a commercial broodstock. For that purpose, properly weighted selection of both traits in a single line is usually recommended. argueb001@hawaii.rr.com

289.  Life-history trade-offs were not found, although looked for very carefully
         Testing life-history pleiotropy in Caenorhabditis elegans. 2001. Knight, C.G., R.B.R. Azevedo, and A.M. Leroi. Evolution 55 (9):1795-1804.
         Managers of captive populations maintained for aquacultural or genetic conservation purposes are concerned about the practical effects of "trade-offs" among traits affecting fitness. One type of tradeoff is caused by pleiotropic genes, i.e. genes that directly affect two or more different components of fitness (maternal body size and fertility in this study; growth and survival in other studies such as #290, above). There are other types of tradeoffs too, such as environmental tradeoffs involving genes which diminish aggression and may confer greater fitness in captive environments while reducing fitness in nature.
         It is the first sort of trade-off, genes which affect two components of fitness in opposite directions, that is the focus of this experimental study.
         Attempts to actually find such genes are so rare that this work on the model organisms C. elegans (a nematode) should be of considerable interest to population managers who are interested in long-term effects of captivity and selection. Especially since the results were negative.
         Three QTLs affecting fertility and two affecting body size were found. None of them had pleiotropic effects on both traits, One fertility QTL was closely linked with a body-size QTL, however, so it is easy to imagine a situation in which a small population selected for one trait would respond in both traits. The point, though, is that this correlated response is not an intrinsic property of the genes and is unpredictable; it could be either positive or negative depending on accidents of sampling and gene frequency at start-up. a.leroi@ic.ac.uk 

288.  Host genetic diversity reduces herbivore and disease damage
         Does host genotype diversity affect the distribution of insect and disease damage in willow cropping systems? 2001. Peacock, L., T. Hunter, H. Turner, and P. Brain. Journal of Applied Ecology 38 (5):1070-1081.
         The authors planted willow tree genotypes having different susceptibilities to beetles and rust fungus in pure monocultures and mixtures. In all cases the damage to the plants was greater in the monocultures than in the genotype mixtures. What is especially interesting  is the fact that the arrangement (location) of the genotypes within the mixture also made a big difference to the magnitude and distribution of damage. lori.peacock@bbsrc.ac.uk

287.  Sorting out the origins of mixed stocks
         Admixture analysis and stocking impact assessment in brown trout (Salmo trutta) estimated with incomplete baseline data. 2001. Hansen, M.M., E.E. Nielsen, D. Bekkevold, and K-L.D. Mensberg. Canadian Journal of Fisheries and Aquatic Sciences 58:1853-1860.
         The authors of this interesting paper try out a Bayesian method of analysing the genetic contribution of hatchery fish to an artificially stocked trout populations when there was no information about gene frequencies prior to stocking. The method (the STRUCTURE program of Pritchard, et al. 2000. Inference of population structure using multilocus genotype data. Genetics 155: 945-959) appeared to work well, but, "The results showed almost complete absence of stocked, domesticated trout in samples of trout from the [Danish] rivers." Evidently the stocking didn't contribute much to long-term rehabilitation of the population. See November 2001 #260 for another paper which comes to the same conclusion. mmh@dfu.min.dk

286.  Valuable local adaptation not well indicated by neutral markers
         Local adaptation across a climatic gradient despite small effective population size in the rare sapphire rockcress. 2001. McKay, J.K., J.G. Bishop, J-Z. Lin, J.H. Richards, A. Sala, and T. Mitchell-Olds. Proceedings of the Royal Society (Ser. B.) 268 (1477):1715-1721.
         A lot of controversy and uncertainty attends the issue of whether on not a population at risk of extinction is actually worth saving, given that resources are limited and we can't save everything. Genetic uniqueness and genetic diversity are often cited as the proper criteria. However, as noted before in these Genecomp listings, the neutral markers with which genetic diversity and uniqueness are measured may have no correlation with the quantitative genetic variation which is the actual basis of adaptation (short term adaptation anyway; see May 2001 #194, Sept. 2001 #237, Jan. 2002 #281).
         The authors of this paper found that small, peripheral populations of their rare plant are genetically adapted to local microclimates. "Local adaptation occurs despite (i) the absence of divergence at almost all marker loci and (ii) very small effective population sizes, as evidenced by extremely low levels of allozyme and DNA sequence polymorphism. Our results provide empirical evidence that setting conservation priorities based exclusively on molecular marker diversity may lead to the loss of locally adapted populations."
         The authors performed transplantation (common garden) experiments on plants living at various altitudes in western Montana. Adaptive traits included water use efficiency, carbon isotope ratios and root mass ratio. jmackay@selway.umt.edu

285.  On-line help with genetics, virology etc.
         National Center for Biological Information resource for molecular biology information. 2002. URL.
         The NCBI has begun to provide full-text, on-line access to a number of comprehensive standard genetics texts. The background material needed to understand most of the population, molecular genetic and bioengineering research papers noted on this GCL website can be found in these books, which include: Molecular Biology of the Cell. 3rd ed. B. Alberts et al.(1994); Introduction to Genetic Analysis. 7th ed. A.J.F. Griffiths et al. (1999); Modern Genetic Analysis. A.J.F Griffiths et al. (1999); Molecular Cell Biology. 4th ed. H. Lodish et al (1999).
         NOTE: to get information about a topic you merely enter a word or phrase into the "search box" at the top of the first page of the NCBI website. The rest of the website appears to be trying to sell you the books. This URL has been added to the "Dictionaries" page. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=Books