The unambiguous identification of closely related species is useful for many practical purposes in forest tree species. For example, international laws require timber identification and the control of the origin of forest reproductive material. In this paper, we present a mitochondrial DNA marker which can be used to differentiate among groups of fir species (Abies spp.). Eight Mediterranean and one North American fir species (used as reference) were analysed at the fourth intron of the NAD subunit 5 gene. A total of six different haplotypes was identified, one in the American Abies concolor, the other five in Mediterranean species. Two different haplotypes were found each in the widespread A. alba and in A. cephalonica, one haplotype being shared among the two species. A single species specific haplotype was found in the near-eastern A. cilicica. The two southwestern species A. pinsapo and A. numidica shared one haplotype. The fifth haplotype was shared by all remaining eastern Mediterranean firs, A. cephalonica, A. bornmuelleriana, A. equi-troiani, and A. nordmanniana. Differences in haplotype sequences were mainly due to large insertions/deletions. Agarose gel electrophoresis thus provides a fast, cheap and reliable diagnosis method for species or species group identification.

Size homoplasy of microsatellite loci can be detected by single‐strand conformation polymorphism analysis (SSCP) or sequence analysis. The present note demonstrates pitfalls of sequencing microsatellite loci. A complex chloroplast microsatellite locus was polymerase chain reaction (PCR) amplified from 33 Abies alba individuals, and the products cloned and sequenced. After 2–7 clones per individual had been sequenced and aligned, differences in repeat numbers occurred among clones of the same individual. It was concluded that the DNA polymerase slippage during PCR most probably caused this variation. A solution specific to the analysed complex locus was found by splitting it into two new loci.

Summary 1We investigated spatial genetic patterns in the clonal herbaceous plant species Galium odoratum L., which has previously been shown to be significantly associated with ancient woodlands. 2 A maternally inherited chloroplast DNA marker was developed for G. odoratum . Seven chloroplast haplotypes were detected in the study area and mapped at different spatial scales. Their spatial patterns suggest that both local and non-local gene flow occurs via seeds. 3 Fingerprints from DIG-labelled AFLP technology were validated as markers for G. odoratum clone identification. A single primer-enzyme combination was sufficient to distinguish between clones. A typical patch of G. odoratum is composed of numerous genets, mainly occurring in separate clusters of ramets, which may be several meters in diameter. 4 Pairwise similarities, calculated from the AFLP banding patterns, were used for spatial autocorrelation analysis. The distinct spatial genetic structure at the patch level may be due to the genetic similarity of neighbouring genets or to clonal growth. 5 The spatial patterns obtained using the two DNA markers, suggest that a mixture of local and non-local gene flow via both seeds and pollen is occurring. Repeated seedling recruitment (RSR) would be consistent with the life history of G. odoratum in ancient woodlands with long-term spatial continuity.

The applicability of DNA markers to purely maternal tissues has been scarcely addressed in trees. We have focused on non-parenchymatic maternal tissues of the fruits and seeds of pedunculate oak (Quercus robur L.) and silver fir (Abies alba Mill.) and investigated whether they can be used for a direct molecular identification of the mother trees. Total DNA with sufficient quantity and quality was extracted from single woody pericarps of acorns as well as from single dry wings of silver fir seeds. The DNA was analysed by PCR at highly polymorphic microsatellite loci. A comparison of the multi-locus genotypes from pericarps and wings with those of the respective mother trees revealed absolute identity. Thus, mother trees could be identified by genotyping their fruits or seeds. The results demonstrate the applicability of DNA fingerprinting to woody and/or dry seed tissues without the destruction of embryos and endosperm or a significant contamination. Progress is now expected in dispersal biology as well as in forensics and forest management.

Clones of Norway spruce (Picea abies L.) were grown for several years on an altitudinal gradient (1750 m, 1150 m and 800 m above sea level) to study the effects of environmental × genetic interactions on growth and foliar metabolites (protein, pigments, antioxidants). Clones at the tree line showed 4.3-fold lower growth rates and contained 60% less chlorophyll (per gram of dry matter) than those at valley level. The extent of growth reduction was clone-dependent. The mortality of the clones was low and not altitude-dependent. At valley level, but not at high altitude, needles of mature spruce trees showed lower pigment and protein concentrations than clones. In general, antioxidative systems in needles of the mature trees and young clones did not increase with increasing altitude. Needles of all trees at high altitude showed higher concentrations of dehydroascorbate than at lower altitudes, indicating higher oxidative stress. In one clone, previously identified as sensitive to acute ozone doses, this increase was significantly higher and the growth reduction was stronger than in the other genotypes. This clone also displayed a significant reduction in glutathione reductase activity at high altitude. These results suggest that induction of antioxidative systems is apparently not a general prerequisite to cope with altitude in clones whose mother plants originated from higher altitudes (about 650-1100 m above sea level, Hercycnic-Carpathian distribution area), but that the genetic constitution for maintenance of high antioxidative protection is important for stress compensation at the tree line.