The magnitude of the protein divergence strongly suggests that Globoderarostochiensis and G.pallida have experienced hardly any morphological evolution during a time period of millions of years (chapter II). These morphologically nearly indistinguishable potato cyst nematode species are discriminated from one another by 70 % of their proteins revealed by two dimensional gel electrophoresis (2-DGE), which definitively excludes a recent divergence as has been suggested by a number of authors. The observation that even two families of rodents share more protein spots revealed by 2-DGE than the two potato cyst nematode species implies that morphological evolution is not correlated with protein evolution. This fits the hypothesis that protein differences are accumulated at an approximate constant rate in all species and that organismal evolution proceeds at a variable rate.A small fraction of the proteins from the potato cyst nematodes appeared to be heat stable. In view of the large genetic distances between G.rostochiensis and G.pallida it is not surprising that also a considerable part of these thermostable proteins isolated from the two species are species specific (chapter III). The relative abundance and simple partial purification of these species specific proteins are suitable properties for developing a serological assay to differentiate the two species. In the near future it should be possible to use such a diagnostic test as an advisory tool. A reliable diagnosis of field populations offers possibilities to optimalize the control by means of resistance.Because of the high resolution and the large number of gene products resolved by 2-DGE, the number of proteins exhibiting intraspecific variation was in most cases ten-fold higher than revealed with other biochemical techniques used to study the genetic variability of G.rostochiensis and G.pallida . Another desirable trait of 2-DGE is that differences in allele frequencies can be estimated in a convenient way. The variant protein spots detected by the comparison of conspecific populations were divided in two classes: isoelectric point variants (IP-variants) and nonisolelectric point variants (NIPvariants) (chapter IV). The IP-variants have the appearance of protein differences caused by amino acid substitutions that change net charge. Corresponding IP-variants were assumed to be encoded by alleles at the same locus and the relative protein quantities were used as a measure for the allele frequencies (chapter V and VI). The NIP-variants are a remaining group for which no proper genetic interpretation is available and may include differences in regulatory genes and modifying genes. The IP-variants are the most reliable characters for estimating affinities between potato cyst nematode populations, because homologous proteins are compared. Relationships based on NIPvariants may also be informative, but should be interpreted with care, because the number of genes involved is unknown and NIP-variants are more difficult to distinguish from artefacts due to, e.g., experimental variations and differences in physiological or developmental stages.In current breeding programs the number of populations tested in estimating the effectiveness of certain genes for resistance is rather arbitrary and depends mainly on the screening capacity. In chapter V and VI we advanced a strategy for a more rational introduction of new genes for resistance by using genetic relationships revealed with (2- DGE), as a guidance for a representative survey. It is hypothesized that in the absence of selection by the relevant genes for resistance in Europe, variations in virulence and proteins among European populations are predominantly determined by the same processes: the genetic structures of the initial populations introduced from South America and the occurrence of random genetic drift and gene flow in Europe. Because these processes affect the whole gene pool of a population, genetic similarities revealed by 2-DGE are also reflected at virulence loci, including those not yet resolved by the current pathotype scheme. The feasibility of this approach is demonstrated by the observation that populations which are closely linked after constructing a similarity dendrogram often have the same pathotype designation. It is also apparent that the international pathotype scheme is incapable of reflecting the genetic diversity introduced into Europe.Several authors have emphasized the necessity of inferring phylogenies of extant nematode species in order to arrive at a more stable classification. A major constraint in reconstructing phylogenies of congeneric nematode species is the limited number of morphological features. The number of characters can in theory greatly be expanded by using biochemical techniques. In this thesis (Chapter VII) we introduced an effective method to infer phylogenies from 2-DGE protein patterns. Proteins of which the evolutionary rates are tuned to the required divergence dates were traced by taking advantage of the approximate constant rates at which proteins evolve in each lineage. This selection procedure resulted in 10 evolutionary conservative proteins which yielded valuable phylogenetic interpretations for the 7 Heterodera species studied. Starch gel electrophoresis, a standard technique for many organisms, is probably of little value in studying these ancient processes of speciation. Many of the enzymes sampled with starch gel electrophoresis evolve too fast and will reveal no shared evolutionary events between the more distantly related Heterodera species. Although 2-DGE has rarely been applied to phylogenetic problems, its general application should be evaluated. The wide application range with regard to divergence dates and the minute amounts of biological material required are desirable features in studying organisms which are difficult to investigate by other biochemical techniques.