Below is Cavalli-Sforza’s elaboration on Occidental/European genetics with the emboldened and underlined parts being my stresses. As a preface, here are my comments:
  • Central Europeans are evidently the most homogeneous 'mega element' within the race.
  • Lapps are the most genetically alien “European Racial” element (Eurasian).
  • Celts are rather like Icelanders in being genetically ‘unique’ in meta/mega ethnic terms.
  • The Germanic ethnicities (including the French) are in the genetic core/center of the European Racial tree.
  • Norsemen & Swedes are indeed genetically distinct from other Germanic folks.
  • Frenchmen are essentically, in genetic terms, Germanic, with the caution of high heterogenity.
  • Dutchmen, Danes, and Englishmen are relatively speaking the same super-folk, genetically. I’ve been thinking this all my life, so I am not surprised by these findings. Danes are not as genetically close to Norsemen and Swedes as many have have long thought for centuries.
  • Austrians, Swiss, & Germans (I suppose Belgians too, but I find this baffling) are of similar close relations as the above… relatively speaking the same super-folk, genetically. Never argue with ‘ Pan-deutschers! ’ LOL
  • Longstanding objections over linguistics being meaningless in genetic groupings of European ethnicities are wrong-headed.
  • The Germanic ‘meta/mega ethnicity’ is the most COHESIVE/TIGHT-KNIT group among all European Racials and also, IMO, among all other human ‘meta/mega ethnicities.'
  • The Northern & Central European ethnicites OVERLAP each other, thus indicating in my mind ‘close’ genetic relations.
  • Greeks, Basques, and Finns are genetically unique ethnicities within the race.
<5.5. THE GENETIC PICTURE [in Europe]

Europe is the continent with the most genetic information but is also the least easily interpretable on the basis of trees. It owes the first distinction to the high density of universities and medical laboratories per square kilometer and to the great interest in historical problems that also makes Europe the most investigated continent from an archaeological point of view. The second distinction is due to a variety of factors. European populations are classified by countries, which are often heterogeneous, being composed of subgroups on which there is only occasionally enough information for separating them in a clear fashion. Many of these countries have been unified politically or linguistically for a sufficiently long time that internal migration has blurred some of the original differentiation. There is a need, however, to study in detail the internal genetic heterogeneity of these countries. We have done this work so far only in a limited way, but it is rewarding, and we will see examples for Italy, France, and Spain. Moreover, the central part of Europe is fairly homogeneous genetically and has little tendency to form a tree. To some extent, this may be due to the Neolithic diffusion, which has ironed out many initial differences in the central region, leaving a fairly smooth gradient.

A tree analysis (fig. 5.5.1), eliminating several less well tested populations, was made for 26 populations with 26.4% missing data. The average number of genes is 88. The tree shows a number of outliers, a few small clusters, and a larger cluster of central Europeans.

The extreme outliers are Lapps, followed by Sardinians; both populations are well known genetically. In a bootstrap analysis for testing the robustness of the tree, Lapps are outliers in 76% of the bootstraps, being replaced as first outliers by Sardinia 18% of the times; Sardinia is the next outlier in 63% of the bootstraps in which Lapps are first.

There is then a group of five less extreme outliers, in the following order: Greeks, Yugoslavs, Basques, Icelanders, and Finns. The rest of the populations form a series of small groups, all of which comprise geographically close neighbors or related populations, in the form of a linear tree. These groups, in order from the most peripheral in the tree to the most central ones, are:
  1. Celtic (Scots and Irish)
  2. Eastern Europeans (Russians, Hungarians, Poles)
  3. Southwestern Europeans (Spaniards, Portuguese, and Italians)
  4. Czechoslovaks, who do not join other Slav speakers, but are approximately intermediate between them and the central subcluster of the Germanic group below, to whom they are geographically close
  5. Northwestern Scandinavians (Norwegians and Swedes)
  6. French, who are related to the Germanic group below
  7. Germanic populations, including two subclusters, northern and central: the first subcluster is made up of Dutch, Danish, and English people; the second of Austrians, Swiss, Germans, and Belgians
It is clear that there is a basic linguistic association within these subclusters, with exceptions worth examining. The first small cluster includes the two Celtic populations in the sample. The second includes two of the four Slavic-speaking populations; the third (Czechoslovak) is isolated in the tree, probably because of its intermediate geographic position (it is located between Slavic- and German-speaking people and was part of the Austrian empire for some time). The fourth, Yugoslavian, is an extremely heterogeneous country, historically and genetically, and is not included in the cluster, being an outlier. Hungarians are linguistically intruders in this group, but that is discussed more thoroughly in section 5.6. Swedes and Norwegians are associated both geographically and linguistically: of the other three countries in Scandinavia, two (Iceland and Finland) are outliers for good reasons, also discussed in section 5.6, and the third (Denmark) belongs genetically and geographically to the northern Germanic populations. France is rather heterogeneous genetically and is discussed separately. The most cohesive group consists of people speaking Germanic languages; of these, Belgium and Switzerland are divided linguistically, with Flemish and German speakers being the majority in the two countries, respectively.

This tree most probably has no historical interest in the present case, but other approaches can help disentangle some of the complex historical and prehistorical events that shaped the history of Europe and they will be examined in the next sections. The PC map (fig. 5.5.2) based on an average of 88 ± 0.1 genes accounts for 50% of the original genetic variation and confirms the external position of the outliers Lapps and Sardinians. Other major outliers like Greeks, Basques, and Finns are on the borders. There are two relatively clear clusters, one of northern and one of central Europeans overlap with central Europeans, on the opposite side from the northern ones. Apart from outliers, the PC map reflects to some extent geography, as is often the case.

Table 5.5.1 gives fst genetic distances of Europeans. An interesting new method of analysis was introduced by Barbujani et al. (1989) and applied to human variation in Europe (Barbujani and Sokal 1990). It formalizes a proposal by Womble (1951) of identifying genetic boundaries as regions of rapid genetic change. The quantities that are used for this analysis are the local slopes of gene-frequency surfaces. The authors visualized boundaries as "lines separating two different regions, each one displaying comparatively little variation"; they recognized that no definition permits the unambiguous distinction of boundaries from gradients or dines ("wide areas of gradual biological change," in the authors' words). It seems a boundary is ideally a narrow dine with a recognizably high local slope, but it might also be a local change in the slope of a dine, and it is difficult to give numerical thresholds that might help definitions. It is unlikely that a sharp boundary in a continuously inhabited region is due to change in natural selection across the boundary; it is more likely that it signals a local decrease in genetic exchange. Boundary detection is greatly facilitated by using a multivariate approach and looking for boundaries that are visible with many independent genes, which is what the authors have done in practice. Local decrease could be due to one or more of a great variety of barriers limiting exchange between two regions. The first and most natural are geographic barriers: the sea, mountains, and major rivers….>

Citation: L. Luca Cavalli-Sforza, Paolo Menozzi, Alberto Piazza; The History and Geography of Human Genes (Princeton, New Jersey: Princeton University Press, 1994), pp 268-9.