The table above represents frequencies found in ancient samples compared to modern Ukrainian ones. The dataset for the ancient sample is a combined pooling of both the Yamnaya culture and the succeeding Catacomb culture from the same area. Note the SNP rs12912832 [OCA2] in the HERC2 gene which predicts blue eyes is present in 65% of modern Ukrainians, but was only present in 16% of the ancient samples. Also note that the ancient samples have much lower frequencies in all three SNPs which predict light pigmentation including SNPs in the TYR gene which are associated with blondism. So while blue eyes and blondism were not completely unseen among the Proto Indo-European speakers...those phenotypical traits were indeed in the small minority.
SOURCE : Wilde et al., Direct evidence for positive selection of skin, hair, and eye pigmentation in Europeans during the last 5,000 y PNAS, Published online before print on March 10, 2014, DO:I10.1073/pnas.1316513111
Another article released a year later basically revealed the same thing, although this study focused the pigmentation of many ancient European populations in comparison to modern ones including pre Neolithic Western Hunter Gatherers and early Neolithic Farmers in Europe. But the study did also include a sample set from the Yamnaya culture which is identified by most linguists as the speakers of the Proto Indo-European language on the eve of their great expansion. Only 11% of the Yamnaya in this sample from the Samar Oblast region carry alleles for light eyes, as they are noticeably darker pigmented in skin and eyes than contemporary Southern Europeans. See data below.
In graph A you'll see a timeline which consists of ancient samples all the way up to modern CEU (which represents White Americans from Utah who are mostly of Northwestern European origin). In graph B you'll see ancient samples represented as well, and you'll also see that the Yamnaya are represented in graph B. Graph C represents modern European populations. Notice the red bar which measures the frequency of light eye color pigmentation alles is extremely low in Yamnaya only scoring 11%, lower than modern Southern Europeans such as the Spanish and Italians (the Italian sample is from Tuscany). This means the Yamnaya, who were the likely speakers of late stage or 'classical' Proto Indo-European were around 90% dark eyed. You'll also notice that the blue and green bars represent the presence of light skin color alles, and their frequency in Yamnaya is lower than modern Southern Europeans as well. The blue bar in Yamnaya is equal to modern Europeans, but the green bar is lower. The data is clearly pointing to the idea that most of the presence of blue eyes in Europe derives from pre Neolithic and pre Indo-European speaking Western Hunter Gatherers, who are uniformly blue eyed. The data also points that natural selection has been favoring an increase in light skin over the past 5,000 in most areas of Europe, peaking in contemporary Northern Europe.
SOURCE : Mathieson, Iain, et al. "Eight thousand years of natural selection in Europe." bioRxiv (2015): 016477.
Also of note, another 2015 major aDNA study was released just several months later. Interestingly enough, it found the same extremely high prevalence for brown eyes in the people of the Yamnaya culture. These aDNA samples were taken from the Kalmykia & Rostov Oblast regions, which are roughly 1,000 kilometers Southwest of the other Yamnaya group sampled in the earlier Maithieson et al study above, which were taken from the Samara Oblast region. However, the study and new sample locations yielded the same results.
"For rs12913832, a major determinant of blue versus brown eyes in humans, our results indicate the presence of blue eyes already in Mesolithic hunter-gatherers as previously described. We find it at intermediate frequency in Bronze Age Europeans, but it is notably absent from the Pontic-Caspian steppe populations, suggesting a high prevalence of brown eyes in these individuals."
Pretty much puts the nail in the coffin for the old idea of the Proto Indo-Europeans being a highly blonde and blue eyed group or phenotypically indistinguishable from present day Northern Europeans. Well, at least for those that support the Kurgan/steppe hypothesis for PIE origins. Of course, there is still the competing Armenian Highland hypothesis and (the now dying) Anatolian hypothesis, but there's no suspicion of predominately light pigmented populations in those prehistoric areas.
SOURCE : Allentoft et al., Bronze Age population dynamics, selection, and the formation of Eurasian genetic structure, Nature 522, 167–172 (11 June 2015) doi:10.1038/nature14507
Those curious of what the geographic distribution of early Indo-European linguistic expansion looks like according to the Kurgan hypothesis, here are some great maps I found from a fellow blogger.
All credit goes to the guy who runs this blog for making these maps.
So what does this mean? It means the long lasting and now hopelessly outdated conjectures from Nordicists about Proto Indo-European phenotype have been falsified. To be fair, we must not forget though that the Yamnaya had greater "genetic height" [Not actual height which is based on a complex combination of genetics and environmental/dietary upbringing] than modern Spaniards & Central Europeans, and perhaps one of the reasons modern Northern Europeans are on average the tallest in Europe is because they have the most Yamnaya like ancestry.
But back to the point, let us not forget that these conjectures about pigmentation, and the idea of predominately light pigmented proto Indo-Europeans [such as John V. Day's hypothesis] were based on little more than select literature references of personalities (both real and mythological) from ancient Indo-European cultures that existed thousands of years after the Proto Indo-Europeans existed, and in most cases many hundreds to thousands of miles away from the Proto Indo-European urheimat. Of course, that's not even getting into the problem of the relativeness of color terminology nor vagueness of the descriptions. The point is none of these later distant and distinct Indo-European populations in which ancient literature mentions of light pigmented individuals spoke the original or even a pure Indo-European language, they merely spoke a language which shared an ultimate linguistic ancestor with Proto Indo-European. In the future more data from ancient DNA I'm sure will show these populations to not be fully genetically continuous to the early Bronze age Proto Indo-European steppe populations (or even the "elites"), though they definitely may have significant Bronze age steppe admixture like most modern Indo-European speaking populations in Europe do.
Keep in mind, just because the Proto Indo-Europeans were overwhelming dark eyed and rather dark skinned and haired by Eurocentric standards, does not mean all later particular Indo-European speaking cultures or populations were. I don't want anyone to fall into that trap. We of course know that modern Indo-European speaking populations vary greatly in phenotype and pigmentation. We also know that even by the late bronze age, that many later Indo-European peoples were evolving not only their own unique languages, but also their own unique genetic identities away from the Proto Indo-European identity as well. So with this in mind, there's no reason to believe they weren't developing their own unique phenotypes distinct from the original Indo-European phenotype either. Many later derived Indo-European cultures like the Proto Indo-Iranians or the Proto Germanics were probably predominately light pigmented.
The beauty of science strikes again and throws us another surprise. Any contemporary seeker of truth is lucky to live in this era of scientific inquiry when humanity has thrown off dogmatic and religious assertions in search for genuine knowledge. I know I am.