We amplified and sequenced the mitochondrial control region from 23 species representing six families of teleost fish. The length of this segment is highly variable among even closely related species due to the presence of tandemly repeated sequences and large insertions. The position of the repetitive sequences suggests that they arise during replication both near the origin of replication and at the site of termination of the D-loop strand. Many of the conserved sequence blocks (CSBs) observed in mammals are also found among fish. In particular, the mammalian CSB-D is present in all of the fish species studied. Study of potential secondary structures of RNAs from the conserved regions provides little insight into the functional constraints on these regions. The variable structure of these control regions suggests that particular care should be taken to identify the most appropriate segment for studies of intraspecific variation.

We constructed a second-generation linkage map of tilapia from the F 2 progeny of an interspecific cross between Oreochromis niloticus and Oreochromis aureus. The map reported here contains 525 microsatellite and 21 gene-based markers. It spans 1311 cM in 24 linkage groups, for an average marker spacing of 2.4 cM. We detected associations of sex and red color with markers on linkage group 3. This map will enable mapping and selective breeding of quantitative traits important to the economic culture of tilapia as a food fish and will contribute to the study of closely related cichlids that have undergone explosive adaptive radiation in the lakes of East Africa.

Olive flounder, Paralichthys olivaceus, is an important fish species in Asia, both for fisheries and aquaculture. As the first step for better understanding the genomic structure and functional analysis, we constructed a genetic linkage map for olive flounder based on 180 microsatellites and 31 expressed sequence tag (EST)-derived markers. Twenty-four linkage groups were identified, consistent with the 24 chromosomes of this species. The total map distance was 1,001.3 cM based on Kosambi sex-average mapping, and the average inter-locus distance was 4.7 cM. Linkage between the loci was identified by an LOD score of ≥3. This linkage map may be used to map quantitative trait loci associated with important traits of the species and may assist in breeding programs.

We obtained 282 base pairs of sequence for the mitochondrial control region of 70 individuals of Korean striped ®eld mice Apodemus agrarius coreae and Apodemus agrarius chejuensis to determine the levels of genetic divergence between these morphologically distinct taxa. The DNA sequences showed more genetic diversity (p) in A. agrarius coreae (2.98%) than in A. agrarius chejuensis (1.86%). Our data do not support the current concept that the two morphotypes are dierent species, but phylogenetic analysis indicates that animals of A. agrarius coreae with large body size from Wan Island cluster with the large-bodied A. agrarius chejuensis, and should be included in that taxon. As currently accepted A. agrarius coreae is not strictly monophyletic, because the large-bodied samples cluster within the range of mitochondrial variation of A. agrarius chejuensis. The fact that the two morphotypes do not share mitochondrial haplotypes (v 2 66, P < 0.001) suggests that there is little gene¯ow between them. A molecular clock estimate suggests that the two subspecies might have been isolated at the time of separation of the islands from the mainland.

For the ¢rst generation of a selective breeding programme, it is important to minimize the possibility of inbreeding. This mostly occurs by mating between closely related individuals, while proper mating can provide an opportunity to establish the base families with wide genetic variation from which selection for subsequent generations can be more e¡ective. Genotyping with microsatellite-based DNA markers can help us determine the genetic distances between the base populations. The genetic markers further facilitate the identi¢cation of the correct parents of the o¡spring (parentage assignments) reared together with many other families after hatching. We established a genetic analysis system with microsatellite DNA markers and analysed the genetic distances of three farmed stocks and a group of ¢sh collected from wild populations using eight microsatellite markers. The averaged heterozygosity of the farming stocks was 0.826 and that of the wild population was 0.868. The hatchery strains had an average of 8.6 alleles per marker, which was less than a wild population that carried an average of 14.3 alleles per marker. Signi¢cant Hardy^Weinberg disequilibrium (HWDE) was observed in two farming stocks (Po0.05). Despite relatively low inbreeding coe⁄ciency of the hatchery populations, the frequency of a few alleles was highly represented over others. It suggests that the hatchery stocks to some extent have experienced inbreeding or they originated from closely related individuals. We will develop a selective program using the DNA mar-kers and will widen the usage of the DNA-based genetic analysis system to other ¢sh species.

We have constructed a genetic map for a tilapia, Oreochromis niloticus, using DNA markers. The segregation of 62 microsatellite and 112 anonymous fragment length polymorphisms (AFLPs) was studied in 41 haploid embryos derived from a single female. We have identified linkages among 162 (93.1%) of these markers. 95% of the microsatellites and 92% of the AFLPs were linked in the final map. The map spans 704 Kosambi cM in 30 linkage groups covering the 22 chromosomes of this species. Twenty-four of these linkage groups contain at least one microsatellite polymorphism. From the number of markers 15 or fewer cM apart, we estimate a total map length of ~1000–1200 cM. High levels of interference are observed, consistent with measurements in other fish species. This map is a starting point for the mapping of single loci and quantitative traits in cichlid fishes.

We compared the variation in mitochondrial DNA among species of cottontail rabbits in the northeastern United States to (i) assess the effects of historic transplants of eastern cottontails (Sylvilagus floridanus) on subspecific status, (ii) examine the extent of hybridization between invading eastern cottontails and declining populations of endemic New England cottontails (S. transitionalis), and (iii) evaluate the recent reclassification of S. transitionalis into two sister-species, New England cottontail and Appalachian cottontail (S. obscurus). Sequence variation in the tRNAPro and the first 310 base pairs of the control region supported a separation of the two species S. floridanus and S. trunsitionulisl S. obscurus. However, geographic structuring of haplotypes was not detected for either group. Genetic similarity among populations of S. floridanus indicated that subspecific designations are not warranted for this species in the northeastern United States. Comparisons of sequence information from populations of S. transitionulislS. obscurus that were never sympatric (Maine), recently sympatric (New Hampshire), or sympatric with eastern cottontails since the last glaciation of eastern North America (Virginia, West Virginia, Maryland, North Carolina) indicated that hybridization is not occurring between eastern cottontails and New England or Appalachian cottontails. Limited variation in mitochondrial DNA does not support the recent reclassification of S. trunsitionulis into sister-species. However, karyotypic and morphological differences between northern and southern populations should be considered during any efforts to restore declining populations of this species.Resume : Nous avons compare la variation de I'ADN mitochondrial chez les diverses espkces de Sylvilagus du nord-est americain dans le but (i) d'ivaluer les effets des introductions de Lapins a queue blanche (Sylvilagus floridanus) sur l'existence des sous-espkces, (ii) d'examiner l'importance de l'hybridation entre les Lapins a queue blanche introduits et les populations endimiques en declin du Lapin de Nouvelle-Angleterre (S. transitionalis) et (iii) d'evaluer le nouveau statut de S. transitionalis, maintenant reparti entre deux espkces soeurs, le Lapin de Nouvelle-Angleterre et le Lapin des Appalaches (S. obscurus). La variation des sequences d'ARNtPr" et les 310 paires de bases de la (region contr6le) justifient la separation des deux espkces S. floridanus et S. transitionalislS. obscurus. Cependant, aucune structuration geographique des haplotypes n'a it6 dicelie chez l'un ou l'autre des deux groupes. La similariti ginitique entre les populations de S. floridanus ne justifie pas la reconnaissance de sous-espkces dans le nord-est amiricain. La comparaison, par examen des skquences, des populations de S. transitionalislS. obscurus qui n'ont jamais it6 sympatriques (Maine), qui sont sympatriques depuis peu (New Hampshire) ou qui vivent en sympatrie avec des Lapins a queue blanche depuis la dernikre glaciation (Virgine, Virginie occidentale, Mary...

The complete nucleotide sequence of a sea lamprey (Petromyzon marinus) mitochondrial genome has been determined. The lamprey genome is 16,201 bp in length and contains genes for 13 proteins, two rRNAs, 22 tRNAs and two major noncoding regions. The order and transcriptional polarities of protein-coding genes are basically identical to those of other chordate mtDNAs, demonstrating that the common mitochondrial gene organization of vertebrates was established at an early stage of vertebrate evolution. The two major noncoding regions are separated by two tRNA genes. The first region probably functions as the control region because it contains distinctive conserved sequence blocks (CSB-II and III) common to other vertebrate control regions. The central conserved domain observed in other vertebrate control regions is not found in the lamprey, suggesting that it is a recently evolved functional domain in vertebrates. Noncoding segments are not found in the expected position of the origin of replication for the second strand, suggesting either that one of the tRNA genes has a dual function or that the second noncoding region may function as the second-strand origin. The base composition at the wobble positions of fourfold degenerate codon families is highly biased toward thymine (32.7%). Values of GC- and AT-skew are typical of vertebrate mitochondrial genomes.