Definitely the spread of mtDNA H in Europe is older than Neolithic.
First we had the reports on abundant mtDNA H in Upper Paleolithic Morocco and Epipaleolithic Portugal, based only on HVS-I, what is admittedly confusing, and never properly published either one. All this was used by the gurus of Neolithic replacement to deny them any validity at all.
However, a few weeks ago we got notice of a study that found mtDNA H in two caves of Cantabria, dated to the Magdalenian period. That alone should have put the Neolithic replacement theory to rest but, of course, faith is blind and irrational, and some seem to be grasping for the proverbial straws... even if themselves do not know which are or where can those imaginary straws be found.
Reality checks sometimes do hurt, I know.
For whatever doubts that may still persist, today I have got notice (h/t Jean) of yet another paper that found mtDNA H in pre-Neolithic europeans. In this case in proto-Basques from Western Gipuzkoa and, less clearly, proto-Greeks from Argolid.
Marie Lacan, La Néolithisation du bassin méditerranéen: Apports de l'ADN ancien. Doctoral thesis (Biologie, Santé, Biotechnologies (BSB)), 2011. Freely accessible as PDF (in French language only).
Dr. Lacan (I really hope she earned her gallons with this quite impressive paper) first ponders in a quite balanced manner on the issues surrounding the Neolithic colonization of Europe on light of modern and ancient DNA. Then, she goes straight to solve the problem by properly testing several individuals from several Epipaleolithic caves from Southern Europe.
That direct and ambitious approach pays off:
The ages of the conclusive results are:
- Linazeta cave, dated to 6230-6100 cal. BCE (more than 1000 years before any Neolithic), Deba, Gipuzkoa (more info in this paper) - H(xH1, H3)
- Franchthi cave, Koilada, Argolis (dates from final Mesolithic to final Neolithic) - H(xH1,H3).
The dates of Franchthi are inconclusive but in the case of Linazeta it is most clear. Specially if we add to it the recent findings in Cantabria (two individuals from two different Magdalenian sites) and the previous, less conclusive, findings in Portugal and Morocco.
The conclusion can't be more clear: in SW Europe at least, mitochondrial haplogroup H existed since at least Magdalenian times.
Also the authors tested two individuals from Santimamiñe cave, a site in Biscay with deep roots dating to the Gravettian culture and showing continuity of use until the Iron Age. This site is of particular importance to me personally because my paternal family comes from just a few kilometers away from there. These are also the oldest DNA sequences from Biscay, however they are already from the Neolithic period (c. 4000 BCE):
This, together with the Paternabidea findings, would confirm that Basques appear to have matrilineal continuity since at least the Neolithic, however this conclusion loses some relevance on light that this continuity now appears to have even deeper roots, into the Epipaleolithic at the very least, in line with the classical archaeological interpretations.
Many details have yet to be unraveled but we can certainly conclude that the coffin of the Neolithic replacement theory has all the nails in it and is more than ready for formal burial.
Another site dealt with in this ambitious paper is the Dolmen of Pierre Fitte, near Paris, dated to c. 2750 BCE (Seine-Oise-Marne culture, a local variant of both Danubian and Megalithic influences), which shows haplogroups K and K1 (HVS-I only) and Y-DNA I2a1 (two individuals).
The paper also discusses in some depth the previously researched necropolis of Treilles and L'Avellaner in the Mediterranean side of the region and quite clearly of likely trans-Mediterranean origins, a component of Neolithic arrival that nowadays is less important even if it has indeed persisted as minority element.