Several factors suggest that further Irish Y-chromosome analysis is warranted. First, its position at the European extreme in the cline of R1b3 frequency points to an unusual population history, perhaps with genetic signatures of past demographic events that are relatively undisturbed by migration. Second, as in many European societies, Y chromosomes and Irish surnames share (in the main) patrilineal transmission. These names are among the oldest cultural lineage markers in the world and emerged from an earlier tribal nomenclature that also emphasized patrilineal relationships. Hence, Y-chromosome comparisons between surnames may be informative with respect to older genealogical links that stretch back into Irish prehistory and mythology. Finally, the indigenous medieval Irish or “Gaelic” social order differed from that of much of Europe, survived until the 16th century, and was, arguably, highly patriarchal and pastoralist. These sociocultural features may have left a distinctive biological legacy.

We initially examined 796 Y chromosomes from all areas of Ireland. DNA samples were collected, along with written informed consent, via buccal swab from volunteers at a range of locations in Ireland (Hill et al. ²⁰⁰⁰; McEvoy et al. ²⁰⁰⁴). Each Y chromosome was typed hierarchically for 11 Y-linked binary polymorphisms that define common European haplogroups (Y Chromosome Consortium ²⁰⁰²; Jobling and Tyler-Smith ²⁰⁰³). The SNPs M269, SRY-2627, SRY-1532, M35, M9, 92R7, and TAT were genotyped using a PCR/RFLP method. M170, M26, and M89 were typed using Tetra-primer ARMS-PCR (Ye et al. ²⁰⁰¹). Insertion/deletion polymorphism 12f2 can be assayed by PCR alone. The most common haplogroup in Ireland, R1b3, roughly equivalent to haplogroup 1 in an earlier nomenclature (Hill et al. ²⁰⁰⁰), is defined by a T→C transition at the M269 SNP (Cruciani et al. ²⁰⁰²). To allow higher-resolution studies of Irish Y chromosome lineages, samples were also typed for 17 microsatellites—DYS19, DYS388, DYS390, DYS391, DYS392, DYS393, DYS434, DYS435, DYS436, DYS437, DYS438, DYS439, DYS389I, DYS389B (calculated by subtracting the DYS389I repeat score from that of DYS389II), DYS460, DYS461, and DYS462—in three PCR multiplex reactions, essentially as described elsewhere (Bosch et al. ²⁰⁰²). Of the samples, 221 had been typed elsewhere for DYS19, DYS390, DYS391, DYS392, DYS393, and DYS389I (Hill et al. ²⁰⁰⁰). The Y-chromosome haplotypes of these were extended to include all of the above markers. Full genotype data from this study is available in online-only tab-delimited text files 1 and 2 (which can be downloaded and opened into an Excel spreadsheet) and at B.M.’s Web site.

The vast majority (85.4%) of the Y chromosomes belonged to the R1b3 haplogroup. The greater resolution afforded by the 17-marker system disperses the AMH lineage into 94 different haplotypes, none of which predominates. However, it simultaneously uncovers a distinct alternative Irish modal haplotype (IMH), also within haplogroup R1b3, that, together with its one-mutational-step (IMH+1) neighbors, accounts for 8.2% of the island sample (IMH 17-marker haplotype with use of the loci as ordered above is 14-12-25-11-14-13-9-11-12-15-12-12-13-16-11-10-11). The IMH distribution is uneven, showing a distinct frequency peak in northwestern Ireland, where it accounts for 16.9% of Y chromosomes (21.5% when one-step derivatives are included) (fig. 1A). The inequitable distribution is further illustrated by comparison of two regional median-joining (MJ) networks of R1b3 Y chromosomes. These graphically display the phylogeny or interrelationship of different Y-chromosome haplotypes. In the northwest (fig. 1B), IMH+1 Y chromosomes form a predominant and phylogenetically coherent starlike cluster that is consistent with a single origin. This feature is absent from a corresponding network of Y chromosomes from southwestern Ireland (fig. 1C).

The distribution of the IMH lineage was also examined in a broader British Isles context, through use of Y-chromosome data for 1,525 individuals across 22 sampling points reported by Capelli et al. (²⁰⁰³). However, since the former were typed for only six microsatellites (DYS19, DYS388, DYS390, DYS391, DYS392, and DYS393), comparison was truncated to include only these loci. The 6-STR IMH is virtually absent from much of Britain but reaches frequencies of up to 7.3% (16.7% including likely one-step derivatives) in western and central Scottish locations. Scotland is known to have had substantial historical and prehistorical links with the northern part of Ireland; for example, the Celtic language still spoken in parts of western Scotland is the closest relative of Irish Gaelic.

The global occurrence of a 7-STR locus IMH (DYS19, DYS389I, DYS389II, DYS390, DYS391, DYS392, and DYS393) was next examined in 28,650 Y chromosomes from 249 geographically defined populations (but ignoring sample populations of <10 individuals) deposited in the Y Chromosome Haplotype Reference Database. The localization of the haplotype was evident with only 38 matches detected (a worldwide occurrence of just 0.13%), with an Irish population sample showing the highest frequency (4.8%). Interestingly, it was also observed in multiple North American population samples. For example, ~2% of European American New Yorkers carry the truncated IMH. Large-scale emigration to North America from Ireland is well recorded. Indeed, given historically high rates of Irish emigration to North America and other parts of the world, it seems likely that the number of descendents worldwide runs to perhaps 2–3 million males.

An analogously predominant Y-chromosome lineage is found in Central Asian populations at a frequency of ~8% (Zerjal et al. ²⁰⁰³). The level of diversity assessed using 15 microsatellite markers supported an origin ~1,000 year ago. This, together with details of its geography, suggested that the lineage’s rapid rise to predominance occurred through an association with the male-line descendents of Genghis Khan and the dynasties he founded in the region. Several observations led us to propose that the high frequency of the IMH in these data may be the result of a similar linkage to an ancient, enduring dynasty.

Gaelic society placed great emphasis on family relationships organized around a strongly patrilineal system (derbhfine) in which land and title could be handed down to successors chosen from within a kin group of male-lineage relatives. This wider inheritance cohort resulted in a decreased likelihood of dissociation of lineage from power (O’Croinin ¹⁹⁹⁵). Also, whereas medieval Ireland was Christian, earlier marriage customs persisted and allowed divorce and concubinage. One feature of these customs was that illegitimate sons were claimed and had rights protected by law (Jaski ²⁰⁰⁰). As in other polygynous societies, the siring of offspring was related to power and prestige (Betzig ¹⁹⁹⁵). For example, Lord Turlough O’Donnell (d. 1423) had 18 sons with 10 different women and counted 59 grandsons in the male line (Connolly ²⁰⁰²).

Turlough and other O’Donnells were members of the most powerful and remarkably durable royal lineage in medieval Gaelic Ireland, the Uí Néill, literally translated as “descendents of Niall.” Gaelic genealogies were important records used to validate claims to prestige and power and linked most ruling families in the northern part of Ireland, prior to the Elizabethan conquest, to the Uí Néill, who claimed high-kingship of Ireland from the 7th to the 11th century ad (Pender ¹⁹⁵¹). The ultimate origin of this dynasty is attributed to the conquering sons of the eponymous and possibly mythological 5th-century warlord, Niall of the Nine Hostages. The historical region under Uí Néill power coincides with the peak in the frequency of the IMH.

In line with the early Irish emphasis on ancestry, it is not surprising that Ireland has one of the largest surviving bodies of early genealogical records. These records range from the solidly historical to the dubiously mythical. Furthermore, since the main purpose of the records was to validate claims to power and property, they were often altered or forged to accord with prevailing political circumstances. Nonetheless, they do present the opportunity to directly test the circumstantial geographic association of the IMH lineage and the Uí Néill dynasty. Modern Irish surnames began in the 10th century ad but often arose from the pre-existing dynastic/population groupings. The Y chromosomes of 59 men possessing names with a purported common origin within the Uí Néill genealogies were collected and genotyped for 17 Y-chromosome STRs. The phylogenetic relationships between these Y-chromosome haplotypes, shown in figure 2, were reconstructed using MJ networks (Bandelt et al. ¹⁹⁹⁹) with the software NETWORK, version 4.1 (Fluxus Engineering). To simplify networks, a reduced median (Bandelt et al. ¹⁹⁹⁵) algorithm was initially applied to the data, followed by analysis with the MJ method. A large coherent cluster centered on the IMH is predominant. The time to the most-recent common ancestor (TMRCA) of this lineage was estimated with the ρ statistic (Morral et al. ¹⁹⁹⁴) in NETWORK, with use of a mutation rate of 1 per 2,131 years for a 17-marker haplotype (Zhivotovsky et al. ²⁰⁰⁴). At 1,730 (SD 670) years ago, it is at least consistent with an early-medieval time frame. A similar TMRCA analysis of the IMH lineage in the general northwestern population sample (shown in fig. 1B) is also roughly consistent with this time frame (1,010 years ago [SD 390). Furthermore, the presence of the IMH across several surnames, which are up to 1,000 years old, certainly suggests an origin predating their adoption. However, it may be that the rise in frequency did not begin with this TMRCA but rather was associated with a group of descendents sometime later. The residual diversity in the Uí Néill sample is probably the cumulative consequence of nonpaternity events and the induction into the clan structure of unrelated males. If any of the Y chromosomes associated with these events were similar to the IMH, because of local homogeneity and recurrent mutation, then they will not be distinguishable here.

Figure 3 shows the distribution of mutational divergence from the IMH in haplogroup R1b3 samples from Ireland as a whole, the Uí Néill–surname population, and the northwestern region where they originated. The smooth distribution in Ireland generally is an obvious contrast to the low-divergence bias (zero and one-step mutational distance) seen in the Uí Néill sample. The disrupted distribution in the general northwestern population illustrates the marked impact of this single recent ancestor on the Y-chromosome structure of the entire region.

The significance of IMH prominence in the Uí Néill cohort was next tested using a Mann-Whitney test, with the number of differences from the IMH for each haplotype (excluding outlying non-R1b3 chromosomes) used as a metric. The Uí Néill sample (n=57) showed a significantly higher affinity with the IMH (P<.001) than with a general R1b3 northwestern Ireland geographic population (n=166). Importantly, a genealogically based control (n=54) composed of R1b3 Y chromosomes from men with surnames originating in the northwest but not within our cohort of Uí Néill derivatives was also found to be significantly different (P=.006). Thus, it seems unlikely that the IMH rose to a high frequency because of a general feature of northwestern Irish demographic history. The rate of increase in the frequency of the Uí Néill lineage is ~1.21 per generation, under the assumption of an evenly spread rise from ad ~500 to the present (some 50 generations, under a 30-year paternal generation interval [Tremblay and Vézina ²⁰⁰⁰]). This rate is somewhat lower than that associated with the patrilineal legacy of Genghis Khan (Zerjal et al. ²⁰⁰³) but still represents a greater increase than that conferred on loci by many known episodes of natural selection, including the sickle cell trait (Li ¹⁹⁷⁵) and lactose persistence (Bersaglieri et al. ²⁰⁰⁴).