B. Winnepenninckx

Performance of 18S rRNA in Littorinid Phylogeny (Gastropoda: Caenogastropoda)

Abstract. In the past, 18S rRNA sequences have proved to be very useful for tracing ancient divergences but were rarely used for resolving more recent ones. Moreover, it was suggested that the molecule does not contain useful information to resolve divergences which took place during less than 40 Myr. The present paper takes littorinid phylogeny as a case study to reevaluate the utility of the molecule for resolving recent divergences. Two data sets for nine species of the snail family Littorinidae were analyzed, both separately and combined. One data set comprised 7 new complete 18S rRNA sequences aligned with 2 published littorinid sequences; the other comprised 12 morphological, 1 biochemical, and 2 18S rRNA secondary structure characters. On the basis of its ability to confirm generally accepted relationships and the congruence of results derived from the different data sets, it is concluded that 18S rRNA sequences do contain information to resolve ``rapid cladogenetic events, provided that they occurred in the not too distant past. 18S rRNA sequences yielded support for (1) the branching order (L. littorea, (L. obtusata, (L. saxatilis, L. compressa))) and (2) the basal position of L. striata in the Littorina clade.

Allozyme diversity in slugs of the Carinarion complex (Mollusca, Pulmonata)

Previous allozyme analyses of the hermaphroditic terrestrial slugs Arion fasciatus, A. circumscriptus and A. silvaticus (subgenus Carinarion) have suggested that in North America these species are each single monomorphic strains. However, new data on 18 putative enzyme loci show that in western Europe the three taxa, respectively, consist of at least three, two and 12 homozygous multilocus genotypes (strains), which regularly co-occur. The current opinion that American and European Carinarion populations are similarly structured, and that colonization events did not affect the population genetics of North American Carinarion, should therefore be readdressed. The present data also provide the first indication of heterozygosity and possible outcrossing in Carinarion. Nevertheless, uniparental reproduction is confirmed as the main breeding system in West European Carinarion, although the high incidence of multistrain populations in A. silvaticus and A. fasciatus appears at variance with the current model of population genetic structuring in selfing terrestrial pulmonates. Finally, the systematic status of the three Carinarion spp. is tentatively questioned.

Radular myoglobin and protein variation within and among some littorinid species (Mollusca : Gastropoda)

The radular muscles of several littorinid species, including Littorina littorea, L. saxatilis, L. obtusata, L. striata and Melarhaphe neritoides, contain myoglobin (Mb). Here we report on the presence of radular Mb in eight other littorinids: L. compressa, L. arcana, L. fabalis, Nodilittorina punctata, N trochoides, N. radiata, Littoraria undulata and Littoraria cingulifera. Using native Polyacrylamide gel electrophoresis (PAGE) and isoelectric focusing (IEF) we compared the Mb and soluble protein (SP) profiles of these species. This suggested that: (1) L. saxatilis and L. arcana may have specific Mb/SP profiles, (2) Littoraria spp., Nodilittorina spp. and L. striata share similar Mb patterns, (3) Mb is remarkably diverse in the genus Littorina, (4) L. littorea shows intraspecific Mb/SP variation, (5) L. saxatilis does not show geographic Mb/SP differences, and (6) IEF uncovers substantial hidden Mb/SP heterogeneity not shown by PAGE (particularly for Melarhaphe neritoides). Hence, littorinid Mb/SP may be a useful taxonomic marker whose ecophysiological significance deserves further study, even if its genetic basis remains unclear.

Software dependent phenetic relationships of non-nodulating African species of **Acacia**

The phenetic analysis of non-nodulatingAcacia species by Harrier et al. (1997) was repeated to illustrate how different computer programs may generate alternative UPGMA trees for the very same data, even in the absence of data input order effects (ties). For example, all Harrier et al.s UPGMA dendrograms produced by software from the Scottish Agricultural Statistics Service differed from those obtained by the packages NTSYS and MVSP87. Particularly, the positions ofA. albida, A. rovumae, andA. pentagona, as well as the relationships betweenDiacanthae andTriacanthae were affected by this phenomenon. Hence, whenever clustering techniques are used, care should be taken to consider possible software-dependent caveats and artefacts. Nevertheless, all programs provided clusterings that largely coincided with the subgeneric and sectional groupings proposed by Vassal (1972) although the positions of some species varied depending on whether morphological or molecular data were considered (e.g.A. albida andA. rovumae).

Dependence of the Ligation Efficiency of Large DNA Fragments Isolated from Agarose Gels on the Purification Method

The isolation of large DNA fragments (> 5 kb) from agarose gels can be problematic. The DNA yield is often low, and the purification may be insufficient for subsequent reactions such as sequencing or ligation. Here we have compared a number of commonly used methods and commercial kits for DNA recovery. Large DNA fragments (12 and 14 kb, respectively) were isolated from agarose gel and purified, the recovery yield was tested, and a well-defined amount of purified DNA was used in a standard ligation reaction. We observed dramatic differences in the efficiency of this ligation reaction depending on which recovery method had been used. However, the respective ligation efficiencies of the different methods were not related to their recovery yields.