A population approach to sampling the genetic resources of forest species is inherent in the prominence of the 'provenance' concept in forest management. It is known from the sampling theory for selectively neutral alleles that the main quantity of interest, the allelic richness or number of alleles per locus actually present in a sample, is expected to increase linearly with increasing population size. In contrast, the allelic richness of the sample increases in proportion to the logarithm of the sample size. This implies that the reward for extra sampling effort at a site follows the law of diminishing returns. An adequate sample is defined as one that includes with 95% certainty, at least one copy of an allele with arbitrary frequency 0.05 and this requires 59 unrelated gametes. Larger samples are needed when there are several alleles at the boundary frequency - the worst-case scenario of 20 equally frequent alleles requires a doubling of the size. Fruits are often the unit of sampling in trees species and we examine the effect of single versus multiple paternity of fruit on the number of fruit required. When paternity is unknown, a worst-case scenario is that all the fruit from one individual form a single full-sib array. It follows that fruit should be sampled from at least 15 trees to be sure of an adequate sample. Strategies for the sampling of many populations depend on the amount of genetic diversity in each population and the genetic divergence between them. The best strategy accords priority to so-called 'local' common alleles, or those that exceed 0.10 in frequency in only one region or population. Such alleles make up an appreciable fraction (10-20%) of the genetic resources of trees. This strategy is in line with the provenance concept and contrasts with samples of 172-plants-per-species, which is an approach recently advocated by Lawrence and colleagues.

The little available information on the domestication of forest tree species is mainly devoted to the genetic consequences of domestication rather than the process itself. Domestication of forest tree species is fragmented and follows the classical horizontal organization of management systems, thus making evaluation difficult. However, systematic monitoring of the various steps of the domestication process (selection-breeding methods and strategies, and seed and seedling production) has identified cases of unintentional directional selection where genetic diversity could be affected. Phenotypic selection has been proven effective in capturing most of the genetic variation existing in natural populations. The progress from breeding population to production populations did not substantially reduce genetic variability; however, the chance for rare, endemic allelic loss increased with a reduction in number of parental trees in seed orchards. Use of a breeding strategy that provides for sampling over several geographical locations and selection for various goals, as well as adaptation to different environments (i.e. Multiple Population Breeding System) was demonstrated to be effective. Generally, understanding the species biology and the role of biology on genetic diversity is of great importance. The need for change and/or the implementation of management practices is effective in maintaining the genetic variability.

Milnesium swansoni sp. nov. is a new species of Eutardigrada described from the tree canopy in eastern Kansas, USA. This species within the order Apochela, family Milnesiidae, genus Milnesium is distinguished by its smooth cuticle, narrow buccal tube, four peribuccal lamellae, primary claws without accessory points, and a secondary claw configuration of [3-3]-[3-3]. The buccal tube appears to be only half the width of the nominal species Milnesium tardigradum for animals of similar body length. The species adds to the available data for the phylum, and raises questions concerning species distribution.

No abstract

No abstract