Local environment and fragmentation by drought and damming shape different components of native and non-native fish beta diversity across pool refuges.

Pool refuges are critical for maintaining stream fish diversity in increasingly intermittent streams. Yet, the patterns and drivers of beta diversity of native and non-native fish in pool refuges remain poorly known. Focusing on Mediterranean streams, we decomposed beta diversity of native and non-native fish into richness difference (RichDiff) and species replacement (Repl), and local (LCBD, LCBDRichDiff and LCBDRepl) and species (SCBD) contributions. We assessed the influence of environmental and spatial factors associated with drought and damming fragmentations on beta diversity components and LCBDs, and of local species richness and occupancy on LCBDs and SCBD, respectively. Overall, non-native species showed a more limited occupancy of pool refuges than native fish. RichDiff dominated beta diversity, though it was influenced by drought and damming fragmentations for native fish and local environment for non-native fish. Repl for native fish was slightly influenced by local environment, but for non-native fish was largely driven by drought and damming, albeit with a contribution of local environment as well. LCBD and LCBDRichDiff increased in pools in low order streams for native fish and at low elevations for non-native fish, and with high or low species richness. SCBD was higher for native species with intermediated pool occupancy, but for non-native species with low occupancy. Our results suggest that stream fragmentation may drive native species loss and non-native species replacement in pool refuges, and that environmental filtering may shape non-native species loss. Pools in lower order streams harbouring unique species-rich or species-poor assemblages should be prioritize for conservation and restoration, respectively, and pools at low elevation with unique non-native assemblages should deserve control efforts. We encourage the partitioning of beta diversity and individual analysis of native and non-native fish in intermittent streams, which may be key in stressing the importance of pool refuges in safeguarding native fish diversity.

Strong genome-wide divergence between sympatric European river and brook lampreys.

Lampreys, together with hagfishes, are the only extant representatives of jawless vertebrates and thus of prime interest for the study of vertebrate evolution [1]. Most lamprey genera occur in two forms with divergent life histories: a parasitic, anadromous and a non-parasitic, freshwater resident form [2-8]. The taxonomic status of such 'paired species' is disputed, however. While indistinguishable at larval stages, but clearly distinct as adults, they cannot be differentiated with available genetic data [6,7], which has fuelled speculations that the two forms may in fact represent products of phenotypic plasticity within a single species. Here, we use restriction site-associated DNA sequencing (RADseq) to examine the genetic population structure of sympatric European river (Lampetra fluviatilis L., 1758) and brook (Lampetra planeri Bloch, 1784) lampreys. We find strong genetic differentiation and identify numerous fixed and diagnostic single nucleotide polymorphisms (SNPs) between the two species, 12 of which can be unequivocally assigned to specific genes.

Evolutionary history of lamprey paired species Lampetra fluviatilis (L.) and Lampetra planeri (Bloch) as inferred from mitochondrial DNA variation.

A remarkable trend in the evolution of lampreys is the occurrence in most genera of 'paired species', in which the parasitic anadromous lampreys are believed to have given rise to nonparasitic freshwater resident populations. The present work examines the phylogeography of the European paired species Lampetra fluviatilis and Lampetra planeri, in an attempt to elucidate species pair evolutionary history. We studied sequence variation in cytochrome b and ATPase 6, 8 mitochondrial genes in 63 individuals from 21 localities of the paired species throughout their distribution range. Results from the phylogenetic and nested clade analyses were largely consistent, suggesting the existence of three major evolutionary lineages: lineage I and possibly lineage II are widespread throughout Europe, while the most ancestral lineage III is apparently restricted to the Iberian Peninsula. The high genetic diversity observed in the Iberian Peninsula is probably the result of refugial persistence and subsequent accumulation of variation over several ice ages, whereas the low levels of genetic diversity observed in central and northern Europe should reflect a rapid postglacial colonization. Results suggest that L. planeri originated within at least two distinct evolutionary lineages, rejecting the single origin hypothesis. The observed lack of taxa monophyly within lineage I may be the result of ongoing gene flow if the two taxa are alternate life-history forms of a single species. However, structure within lineage I is also consistent with the hypothesis of divergence of taxa after postglacial dispersion (around 2000 generations ago) with incomplete lineage sorting. Further testing of the alternative hypotheses is warranted.

Simultaneous production of triploid and haploid eggs by triploid Squalius alburnoides (Teleostei: Cyprinidae).

The hybrid minnow Squalius alburnoides comprises diploid and polyploid forms with altered modes of reproduction. In the present paper, we report a cross where a triploid female generated both large, triploid and small, haploid eggs simultaneously, which were fertilized with S. pyrenaicus sperm. Although the large eggs were rarer (15%), they originated offspring with higher survivorship, so that tetraploids were dominant among the surviving siblings. The cross yielded apparently all female progeny. Inheritance patterns were inferred using four microsatellite markers and NORs (Nucleolus Organizer Regions) phenotypes, and suggested that haploid eggs were probably produced by an atypical hybridogenesis, in which the elimination of the unmatched genome permitted random segregation and recombination between the homospecific genomes, while the triploid eggs were clonal. The present results suggest that the occurrence of triploid unreduced eggs may be a new route for the natural tetraploidization in the complex.