Detailed studies of the genetics of speciation have focused on a few model systems, particularly Drosophila. The copepod Tigriopus californicus offers an alternative that differs from standard animal models in that it lacks heteromorphic chromosomes (instead, sex determination is polygenic) and has reduced opportunities for sexual conflict, because females mate only once. Quantitative trait loci (QTL) mapping was conducted on reciprocal F 2 hybrids between two strongly differentiated populations, using a saturated linkage map spanning all 12 autosomes and the mitochondrion. By comparing sexes, a possible sex ratio distorter was found but no sex chromosomes. Although studies of standard models often find an excess of hybrid male sterility factors, we found no QTL for sterility and multiple QTL for hybrid viability (indicated by non-Mendelian adult ratios) and other characters. Viability problems were found to be stronger in males, but the usual explanations for weaker hybrid males (sex chromosomes, sensitivity of spermatogenesis, sexual selection) cannot fully account for these male viability problems. Instead, higher metabolic rates may amplify deleterious effects in males. Although many studies of standard speciation models find the strongest genetic incompatibilities to be nuclear-nuclear (specifically X chromosome-autosome), we found the strongest deleterious interaction in this system was mito-nuclear. Consistent with the snowball theory of incompatibility accumulation, we found that trigenic interactions in this highly divergent cross were substantially more frequent (46 Â ) than digenic interactions. This alternative system thus allows important comparisons to studies of the genetics of reproductive isolation in more standard model systems.

Like all organisms, sponges allocate resources to life functions such as growth and reproduction. Additionally, some sponges are defended by secondary metabolites that deter potential predators. Assuming resources are limiting, species that produce defensive metabolites should allocate fewer resources to growth and reproduction. To test the hypothesis that there is a trade-off between chemical defense and growth, predator exclusion experiments were conducted to compare the growth rates of 7 common Caribbean sponge species with similar branching morphologies: chemically undefended species Callyspongia armigera, Iotrochota birotulata and Niphates erecta, and defended species Amphimedon compressa, Aplysina cauliformis, Aplysina fulva and Ptilocaulis walpersi. Thirty field experiments lasting 124 to 195 d were performed over a 9 yr period on a total of 1158 sponges. A 3-factor ANOVA was used to compare the effects of chemical defense (undefended/ defended), treatment (uncaged/caged) and season (summer/winter). Despite the fact that different sponge species could be allocating resources differently to reproduction, thereby potentially obscuring the interaction between growth and chemical defense, growth in cages was significantly greater for undefended than defended sponges (110.4 versus 65.8% growth yr -1 ). While the growth of chemically defended sponges inside and outside of cages was not different, growth of undefended sponges inside cages was significantly greater than outside cages, confirming that palatable sponge species were grazed by sponge-eating fishes. Growth during winter months was significantly less for both undefended and defended sponges compared to growth during summer months, but again, growth was greater for undefended sponges than for defended sponges. Differences in growth rates demonstrate that sponge species have evolved alternative allocation patterns to cope with resource constraints, as well as predation, and provide evidence that there is a metabolic cost associated with chemical defenses.

Aim Previous work on the tidepool copepod Tigriopus californicus revealed a curious case of incipient speciation at the southern end of the species' range in Baja California, Mexico. The present study expands on the geography of this pattern and tests for congruence between reproductive and phylogenetic patterns.Location The Pacific coast of North America, from central Baja California to south-eastern Alaska (27-57°N), including the full range of T. californicus.Methods Primary techniques included mating experiments (> 4000 crosses), phylogeny reconstruction (mitochondrial cytochrome c oxidase subunit I) and screening of single nucleotide polymorphisms (SNPs, 42 loci). Analyses used > 8000 copepods for the mating experiments, 86 copepods for the phylogeny and 41 copepods for the SNP assays. Phylogenies were constructed using Bayesian, maximum likelihood and maximum parsimony methods. ResultsPopulations were found to fall into three reproductive groups: northern and southern groups that were reproductively isolated from each other, and an intermediate group that could serve as a conduit for gene flow. The northern and intermediate populations fell into one clade while all southern populations fell into a second clade. These two clades are now separated by less than 12 km at latitude 29.35°N. Nuclear SNP data for a subset of locations confirmed striking divergence between populations on either side of this boundary. The second (southern) clade was further subdivided into two clades separated by the lagoon region of Guerrero Negro (latitude 28°N).Main conclusions Reproductive assays and molecular data (both mitochondrial and nuclear) reveal a sharp break at 29.35°N, a region with no obvious barriers to dispersal, with no evidence for mixing across this narrow transition zone. Results also showed a milder break at the Guerrero Negro Lagoon (28°N), a location where breaks have been reported for other taxa.

Local adaptation has been understudied in marine systems, but might be expected to be pronounced in the tidepool copepod Tigriopus californicus, which has a broad geographic range and extremely restricted dispersal. Tolerance to temperature and salinity was assessed in 14 populations over a 20° latitudinal range. Adaptive differentiation to temperature and salinity was found at scales as low as 5.6 km. Latitudinal clines were significant, with northern populations being more tolerant of low salinity and less tolerant of high temperature and high salinity. Both temperature and salinity tolerance were more closely associated with long-term thermal maxima than with long-term precipitation data. Hyperthermal and hyposmotic tolerance were inversely correlated, a pattern that could potentially slow adaptation to future conditions. Together, these studies of intraspecific geographic patterns in resistance to multiple stressors are important in predicting how environmental change may effect range shifts and local extinctions.