The DENNISON Patrilineage 2 DNA Project

DENNISON DNA Patrilineage 2

The Recent Virginia Origins

The ancestors of most, if not all, of the American members of this DENNISON patrilineage appear to have settled in Pittsylvania County, Virginia, by the late 1700s, and migrated from there to a cluster of south central Kentucky counties in the early 1800s. Even though Pittsylvania lies just to the SE of the Blue Ridge mountains, and just outside the Valley of Virginia where the Scotch-Irish DENNISONs of Patrilineage 1 settled, the yDNA results shows that these two groups of Virginia DENNISONs are completely unrelated.

Now, as it turns out, although the PittsylvaniaCo DENNISONs appear to constitute a single extended family, they actually represent two different patrilineages: this DENNISON Patrilineage 2, but also DENNISON Patrilineage 6, for which I have yet to construct a project page. The evidence for this is complicated, and I have addressed it thoroughly in the set of reports for which links are provided under “Other Genealogical Resources & Evidence” in the upper left navigation panel of this page. Everything that I had earlier written about Patrilineage 2 in the paragraphs and sections below, however, still stands.

It’s highly unlikely, prima facie, that the PittsylvaniaCo DENNISONs were Scotch-Irish, for a number of reasons, beginning with the fact that most British DENNISONs are English, not Scottish, thus the vast majority of American DENNISONs have English, not Scotch-Irish roots. However, for reasons that I will develop below, I do not think that the ultimate ancestry of this patrilineage was English.

Second, Pittsylvania County was overwhelmingly settled, not by the Scotch-Irish whose initial settlements were beyond the mountains, but by tidewater Virginians slowly trickling west. This was true of Pittsylvania's parent county, Halifax, to the east, and of its child county, Henry, to the west. And the same east to west pattern applied to the next northern tier of counties, from Lunenburg, to Bedford, to Franklin County (where I happen to live). Furthermore, these early settlers of Virginia’s piedmont, and “south side”, continued west mostly in the wake of the Scotch-Irish, progressing to Tennessee and Kentucky a decade or two later.

Third, the early settlers of these cismontane counties, including these DENNISONs, do not, by and large, bear Scotch-Irish surnames, nor are their given names confined to the restrictive Scottish (and Scotch-Irish) set, and they don’t appear to observe the customary Scottish and Scotch-Irish onomastic (child-naming) pattern either (see my definitive paper on Scottish naming practices, including the Appendix showing the frequencies of the 20 most popular given names).

At least some of the Pittsylvania DENNISONs came from Maryland

DENNISON Surname Project member, Kathy-52 (of the DENNISON Patrilineage 6 Project) has found the key evidence that links at least her ancestor, Benjamin of BarrenCoKY and PittsylvaniaCoVA, back to Maryland—specifically, a pair of matching records just a couple of days apart for his marriage to Drucilla Pearce—the name “Drucilla”, of Benjamin’s wife, having followed Benjamin, first to PittsylvaniaCo, and then BarrenCoKY, where this couple ended up. The first of these records, dated 26Oct1779 is in PrinceGeorge'sCoMD, and the second, dated 28Oct is in MontgomeryCoMD. It looks as though Benjamin obtained his license in PrinceGeorge’sCo where he lived, but married in MontgomeryCo, probably where he went to church. Thus, he, and presumably other DENNISONs of his extended family, probably lived in PrinceGeorge’sCoMD, somewhere along its extensive border with MontgomeryCo. Also, Benjamin of PittsylvaniaCo’s oldest son was named Zachariah, and there was another marriage record for a Zachariah in PrinceGeorgeCoMD, in 1777, less than two years before Benjamin’s—quite likely this was for an older brother.

There are apparent problems with this evidence, in the form of the widely variant surname spellings in first these MD marriage records, and then in PittsylvaniaCoVA, but after my own now exhaustive research of the underlying primary records, I’ve dealt with this and the other issues that complicate the story of the PittsylvaniaCo DENNISONs, and have presented my conclusions and arguments in the set of linked reports under the head “Other Genealogical Resources and Evidence”, opposite.

Although it may seem a bit of a stretch that these DENNISONs of central Maryland, on the Chesapeake Bay opposite northern Virginia, should have leapt directly to PittsylvaniaCoVA on the border with North Carolina, Benjamin, one of the three PittsylvaniaCo patriarchs, his conjugal family, and scions of the other patriarchs, later made an even greater leap from PittsylvaniaCo to BarrenCo and other south central KY counties. The distance of this latter leap was even greater, while the due west direction was more typical of early American migrations, because in an era when virtually every family engaged in at least part time farming and pasturing, sticking to familiar climates and growing seasons was a prime desideratum.

The deviation from this rule that drove most of the early cismontane Virginians, the vast majority of whom (or their ancestors) were brought to the Chesapeaks and other tidewater counties as indentured servants, was due to the major obstacle of the mountains themselves, and the menace of the unknown wildnerness that lay beyond. When these early Virginia importees were able, subsequent to their term of service, to think of affording their own land (a readiness that was sometimes measured in generations), they were obliged to migrate to whatever counties of VA or NC then constituted the inner cismontane frontier and still offered desirable cheap and unoccupied land. And in the 1770s, PittsylvaniaCo (created in 1767 from HalifaxCo) was one of those counties.

That these DENNISONs should have been in Maryland in the first place nicely dovetails with the discovery, some years ago now, that the ancestry of the Pittsylvania patriarch James was ultimately Scandinavian, and that his immigrant ancestor was most likely an immigrant to the early (1638) New Sweden colony on the Delaware River, less than 100 miles north of the border between PrinceGeorge's and Montgomery Cos, MD—a subject that I shall take up next. And although Y-Chromosome DNA testing of descendants of the patriarchs James and Benjamin has shown that the former belonged to DENNISON Patrilineage 2, while Benjamin (and probably Richard) descend from DENNISON Patrilineage 6, I nonetheless think it likely that these two were related through the maternal line, as I've explained in a section of my Three Patriarchs paper.

The Ultimate Scandinavian Origins of DENNISON Patrilineage 2

As it turns out, the ultimate roots of DENNISON Patrilineage 2 lie in southern Scandinavia, with several lines converging on the area around Sweden’s principal modern port of Gothenburg. There is also a possibility that the line was originally Norwegian.

Lois-10, the only one of the Amercian Patrilineage 2 Project members who has even tested to 67 markers, has FTDNA-reported matches to two foreigners: Englishman Lee Daniels (member 11), who has traced his ancestry back to 18th century Buckinghamshire, and Simo Roman (member 33), who is a Finn with ancestors who trace back to 18th century Skaraborg County, Sweden, which was recently merged with adjacent Gothenburg. And Lee and Simo in turn have reported matches to Norwegian, Petter Vennemoe (member 35), and to a cluster of men surnamed LONGACRE who have solid descents from one Andrew Peterson Långåker born 1657 in Goteborg, Sweden, who apparently emigrated to the New Sweden Colony on the Delaware.

Petter has traced his Norwegian ancestry even farther back than the LONGACREs, to one Sivert [Olsson?] born say 1590 and a resident of the farm of Skarebøl, Akershus County, a large rural farming community that lies just to the west of Oslo, and to Sivert’s son, Svend, who took up residence in the adjacent farm of Bøen, also in Akershus. Akershus borders Sweden to the east, and lies about 200 miles from the port of Gothenburg overland, though the connection from Oslo to Gothenburg by sea would probably have been more convenient in the 16th and 17th centuries.

Scandinavian Surnames, Bynames, and Patronymics

The LONGACRE surname presumably derives from the Swedish Långåker (“long field” or phonetically “long acre”): at least Reaney & Wilson’s etymological Oxford Dictionary of English Surnames shows no English derivation for the surname. And Petter Vennemoe, the Norwegian member of this patriineage, has found that there was a Swedish farm near Gothenburg named Långåker. Modern Americans might be inclined to say, well, that’s interesting, but so what: certainly it’s possible that the New Sweden family could have hailed from there, but after all, most American and British surnames are derived from personal or occupational names, and only a small minority are locative in origin—derived from place names.

Scandinavian personal names and surnames, however, follow a radically different pattern from what Americans are used to. In the first place, virtually all Americans have had surnames from the time they stepped off the boat—permanent, hereditary, surnames that were passed down by the father to his children. And in England, from whence the largest contingent of American immigrants came, such surnames have been the rule since the 16th century and before. However, in Scandinavia, Russia, and a few other areas, permanent hereditary surnames didn’t become general until as late as 1900.

Instead, individuals with the same given name were differentiated through bynames that applied solely to them, and expired when they did. The most common form, and system, of bynames was the patronymic—a temporary surname derived for a child from its father’s given name (thus Peter, the son of Anders, was known as Peter Andersson). Until 1900 or so, virtually all Scandinavians were known primarily by their given name and patronymic, but as rural populations, and commercial intercourse, grew from the early centuries of scattered settlement, the relatively small stock of Scandinavian given names, each with its derivative patronymic, proved insuffient to create for each person a more or less unique name and identity. This problem was endemic throughout much of Eurasia, and it was solved in various ways, but in Norway, and to a lesser extent in Sweden and Finland, most individuals acquired, at least implicitly, as an additional byname, the name of the farm (or farming district) that they resided within.

Thus, the full names of Petter’s earliest known ancestors were Sivert [Olsson? or Olsen?] of Skarebøl, and his son, Svend Sivertsen of Bøen, or just Svend Sivertsson Bøen. And when a great-great grandson of Svend finally removed to a different farm, he became known as Hans Bergersen, of Kongtorp.

And when Anders Petersson of the New Sweden colony (the son of Peter Andersson originally of Gothenburg, Sweden) chose LONGACRE to be the hereditary surname for his American line, it is highly likely that he would have used his family’s farm name, and that his father and other ancestors and collaterals should have had their roots in the large Swedish farm, Långåker, near Gothenburg, before they were lured to town to participate in the New Sweden adventure.

A Theory of Convergence for this Patrilineage to a Most Recent Common Ancestor

Besides finding a likely area of origin for the ancestors of his LONGACRE cousins, Petter has also proposed a theory to account for the convergence of his line with theirs, and with it, a probable MRCA for the patrilineage as a whole. His theory is predicated on the circumstance that the patronymic, Olsson, that he had independently projected for his probable earliest known ancestor, Sivert of Skarebøl (born say 1590), is the same as the patronymic of the LONGACRE’s earliest known ancestor, Anders Olsson of Gothenburg (born say 1585). And this in turn raises the possibility that these two were brothers and sons of the same father, named Ole.

There are, of course, serious objections to this theory. In the first place, Ole has always been one of the most popular Scandinavian given names, and there are innumerable Oles everywhere. And in the second, the putative sons of these men lived in two different countries, Norway and Sweden.

However, with respect to the latter objection we might observe that the Swedish LONGACRE family was evidently one highly prone to migrate away from their putative local landed area of origin—first to the Gothenburg, and then, almost at the earliest opportunity, to New Sweden, in the New World. Moreover, not long after that passage, perhaps, another descendant of this line who was ancestor to patrilineage member, Lee Daniels, emigrated to England. Given this family history, it is only a small stretch to suppose that the son of a farmer of Långåker, near Gothenburg, should have heard of better opportunities 150-200 miles to the north, just across the border into Norway, and resolved to seek them out. There is no direct evidence that any such thing happened, but it is a perfectly plausible scenario nonetheless.

With respect to the innumerable Oles, we need to avoid anachronism and adjust our thinking to a period when the population of Scandinavia was sparse. Whereas the Swedish population today is around 9.5 million, and the Norwegian about 5 million, interpolating from some early population estimates the population of Sweden about 1590 may have been roughly 700,000, and of Norway, 300,000. Meanwhile, given that the vast majority of modern Scandinavian surnames are derived from patronymics, we can work up a rough estimate of the relative frequency of the given name Ole in Sweden and Norway: about 9% of Norwegians were named Ole, and 5% of Swedes. The product of these estimates yields about 27,000 Oles in all of Norway, and 35,000 in all of Sweden.

We also need to take into account the distribution of this population. Today, a large percentage of Scandinavians live in the big cities and their suburbs (Stockholm and Gothenburg; Olso and Bergen) but things were vastly different in 1600. Although the roots of the city of Oslo/Christiana extend back to the 11th century, the port of Gothenburg wasn’t founded until 1621. The population of Oslo in 1349, just before the Black Death struck, has been estimated at 3,000, but historians have estimatated that that figure had been cut in half by 1400, and, judging by other areas, it would have taken a couple of centuries for it to recover. Meanwhile, Bergen became Norway’s principal port, as part of the Hanseatic league network.

Thus, Scandinavia around 1600 was almost entirely rural, with its population evidently widely scattered, and the denizens of the local farming districts specifically relavant to the Scandinavian ancestors of this patrilineage, are to be measured in the hundreds. Under these circumstances the number of Oles becomes manageable, the number of those with sons named Anders and Sivert even more so.

Finally, although besides these circumstantial considerations, no direct evidence has yet been found linking the New Sweden LONGACREs with Långåker farm, nor any evidence that Petter’s earliest known ancestor, Sivert, was a close relative of the Swedish LONGACRE family, given the extremely tight DNA relationship between Petter and LONGACRE descendants today (Genetic Distance of 1 at 67 markers), it is appropriate that we theorize aggressively to find a possible convergence of their lines in a common ancestor.

In fact, FTDNA’s TIP calculator (adjusted for no known common ancestors for at least 8 generations) says that there's a 38% chance that Petter’s earliest known ancestor Sivert [Olsson?] was also the father of the LONGACRE’s earliest known ancestor, Anders Olsson, and a 76% chance that their lines converge within two generations back from Anders, and since patronymic bynames typically change every generation, with the given name, the most likely hypothesis to account for this data is that these two were sons of the same Ole. What is needed is some intensive research into both the specific records and the history of Långåker and the Gothenburg area of Sweden, and for the Norwegian farms Skarebøl and Bøen to the east of Oslo, near the Swedish border, for the period 1525-1675. Out of this it may be possible to construct a scenario to explain how these two putative brothers ended up in these two different, though not particularly remote, places. I would not foreclose the possibility either, that this patrilineage was originally Norwegian, though I suspect that these distinctions of nationality and ethnicity that we moderns are prone to make are probably somewhat anachronistic as well.

Thumbnail Ancestries of DENNISON Patrilineage 2

What is known of these ancestries may be summarized thus:

(1) Simo Roman has traced his surname patriline back to one Sven Larsson ROMAN, born in Skaraborg County (near the port of Gothenburg) about 1811, and rather remarkably, he has been able to push his patronymic ancestry back another three generations before the adoption of an hereditary surname in his line to one Anders Larsson, born say 1695, though it isn’t known where. However, Anders’s son, Olof Andersson, is known to have died in Söne, Skaraborg County, Sweden, not far from Gothenburg, Sweden’s most important port, so that marks one anchor point for this patrilineage.

The proliferation of surnames here is explained by the fact that prior to the 1800s, the surnames of virtually all Swedes were formed from the (single) given name of their father—the suffix “-son” being used for sons, “-dotter” for daughters. This was essentially the same patronymic system used by the Russians, and for that matter, Welsh, and in all of these areas permanent hereditary suranames didn’t become the rule until well into the 1800s.

(2) The two LONGACREs (who aren’t yet members of the project) descend from one Peter Andersson, an immigrant to the American colony of New Sweden, on the Delaware River, who was himself the son of Anders Olsson of Gothenburg, Sweden, born say 1585. This provides the earliest known anchor point for this patrilineage.

The first colonists arrived from Gothenburg late March 1537/8, under the guidance of the Dutch adventurer, Peter Minuit. The LONGACRE ancestor, Peter Andersson, is thought to have emigrated in 1639 (though I have not myself, seen the evidence for this), and to have shuttled back and forth several times before settling finally in Kingsmessing, near the confluence of the Delaware and the Schuylkill Rivers (now a district of SW Philadelphia). Peter’s son, Anders Petersson, was the first to adopt the permanent hereditary surname LONGACRE (Swedish Långåker), which begins to be associated with his name in the records in 1693. The colonists of New Sweden were among the very first Swedes to have adopted permanent hereditary surnames, as part of their assimilation into English culture.

Significantly, both Simo and his LONGACRE cousins have traced their ancestry back to the vicinity of Gothenburg (Swedish Göteborg), Sweden’s major port. Simo’s earliest ancestors resided in the nearby county of Skaraborg (see this map of the Swedish counties). The specific location of the LONGACRE ancestors is unknown, though it places the line in the Gothenburg area as early as the mid-1600s.

(3) Lee DANIELS, who is English, has traced his DANIELS line back to Thomas DANIEL[S] born say 1762, probably in Buckinghamshire, near London, Britain’s most important port. Considering the mutational pattern of Lee’s DNA, it’s likely that his line branched off the main trunk of the patrilineage tree very early, some time in the early 1600s, and a near ancestor of Thomas, perhaps his father, probably emigrated to London from Gothenborg some time between 1650-1750.

(4) Strictly speaking, member Terrance Nolan (D-45) probably falls just outside this DENNISON Patrilineage 2. Although he comes up as “Possibly Related” to member Simo Roman on the 37-marker GD chart, and improves to “Related” on the 67-marker chart, also picking up a “Probably Related” to Lee Daniels (D-11) of England, his haplotype diverges significantly when expanded to 111 markers. When I compare Terrance’s 111-marker result to that of Norwegian member Petter Vennemoe (D-35), and also to the one LONGENECKER who has extended to 111 markers, Terrance comes short, with a GD=13 relationship (“Unrelated”) to LONGENECKER, and he’s only GD=15 to Petter.

However, I’ve accepted Terrance as a member because his Irish ancestry may shed some light on the deep ancestry of this patrilineage. A GD of 13 projects to a MRCPA with the Scandanavian-rooted ancestors of DENNISON Patrilineage 2 born about the 13th century, give or take a couple of centuries. That puts this MRCPA on the near cusp of the Viking Diaspora, and it seems unlikely that Terrance’s ancestor would have migrated to England during the 14th century—the century of the Black Death—during and after which most Englishmen acquired surnames—English surnames. NOLAN is a distinctively Irish surname that seems to have originated in County Carlow in the SE of Ireland and spread out from there in all directions, but that doesn’t tell us much because so many Irishmen only acquired surnames centuries later. If we can get some other project members to extend to 111 markers, the relationship of Terrance’s line to that of the DENNISON Pat2 line proper, may become clearer.

(5) All the other DENNISON Patrilineage 2 members appear to trace their ancestry back to Pittsylvania County, in south-central Virginia (or possibly to its parent county of Halifax) by the late 1700s, although for only one member, Julie-08, can we now make a specific connection to one of these PittsylvaniaCo families: my recent compilaton and analysis of all the DENNISON entries in the annual tax records of PittsylvaniaCo from 1782-1812, makes it highly probable that Julie’s previous earliest known ancestor, William, who married Jenny Barron in PittsylvaniaCo, was the son of James DENNISON/DONALDSON of PittsylvaniaCo.

Meanwhile, new DENNISON Surname Project member, Kathy Warren Baker (Kathy-52), has traced her ancestry credibly back to Benjamin DENNISON of PittsylvaniaCo, and from there back to Maryland. And since all the indications but one, suggest that James, Benjamin, and a third Pittsylvania DENNISON paterfamilias, Richard, were brothers, James presumably came from Maryland too. I have throughly developed the strong circumstantial evidence in favor of this presumption in my new analytical report on the DENNISONs of PittsylvaniaCo, a link to which, along with other supporting evidence will be found under “Other Genealogical Resources & Evidence” in the top left navigation panel of this page.

However, that one contraindication that James and Benjamin were brothers (or at least full brothers) is fatal: their two credible descendants, Julie-08 and Kathy-52, belong to different DENNISON patrilineages, and they have no common ancestor for at least 2000 years! Clearly an NPE (Non-Paternity Event) has occurred in one of their DENNISON ancestries, but we can pretty well dismiss the possibility that this happened downstream of PittsylvaniaCo, because both have at least two patrilineage cousins who can trace back to different progeny of PittsyvaniaCo principals—though none can as yet actually connect to their senior PittsylvaniaCo ancestor. The resolution to this paradox is that James and Benjamin were probably half brothers, while Richard was presumably a full brother of one of them, most likely Benjamin, considering Richard’s close proximity to Benjamin in the county tax records that I’ve just compiled and analyzed. However, we need to test a patrilineal descendant of Richard to know this for sure.

Finally, given that DENNISON Patrilineage 2 is Scandinavian in its ultimate origin with patrilineage cousins named LONGACRE, and the fact that the area of Maryland where the PittsylvaniaCo DENNISONs came from is only about 100 miles SW of the 1638 New Sweden settlement on the Delaware River, the only important Swedish emigration to America before the 19th century, suggests the possibility that they descended from Peter Andersson, the New Sweden colonist and the immigrant ancestor of the LONGACREs, or a close male relative of his who assumed the surname DENNISON and whose descendants drifted down a ways to Maryland. I’ve considered this specific hypothesis in detail in the next section analyzing the DENNISON Patrilineage 2 DNA.

Implications of the DNA Evidence for DENNISON Patrilineage 2:

Additional yDNA Testing is Needed

First, although I’ve tried below to milk what I can out of the yDNA evidence so far, it would be highly desirable to extend to 111 markers the haplotypes of the rest of the current members of DENNISON Patrilineage 2, particularly the American members, in order to look for possible Closer Cousin Cluster (CCC) relationships among them that could be useful in focusing and guiding further research to connect them back to PittsylvaniaCoVA. Extending would also provide a better read on the closeness of relationship between the American DENNISONs and the LONGACREs and other likely descendants of the New Sweden settlement.

As a separate matter, it would also be desirable to try to track down descendants of Richard DENNISON of PittsylvaniaCo and test them to see whether they too belong to Patrilineage 2, or to Kathy-52’s DENNISON Patrilineage 6. Knowing that would help in looking for the NPE (Non-Paternity Event) that must have occurred, probably in Maryland, in the siring of one of the three PittsylvaniaCo DENNISON brothers: James, Benjamin, and Richard.

The Genetic Distance Evidence

To provide an overview of the closeness of these patrilineage interrelationships I have updated the DENNISON Patrilineage 2 67-marker GD matrix for the five 67-marker haplotypes that have surfaced for this patrilineage, as well as the less accurate 37-marker matrix, into which I’ve incorporated all the additional 37-marker LONGACRE haplotypes. The 67-marker matrix shows these DENNISON, DANIELS, LONGACRE, and ROMAN haplotypes are unequivocally “Related ” (in FTDNA’s parlance), and the 37-marker matrix shows that they have many tested LONGACRE cousins, a couple of whom might profit from extending to 67 markers. As a matter of fact, it would be desirable at some point for all the members of this project to extend their haplotypes further to 111 markers, both to look for Closer

In light of these expanded GD matrices, I have adopted Simo ROMAN’s haplotype as the new RPH for DENNISON Patrilineage 2. Simo’s largest GD at 67 markers with any of the others (which is 6), projects to a MRCA (Most Recent Common Ancestor) for all 5 of the patrilineage members who have tested to 67 markers to a man who was born say 1500, while the maximum GD between any of these 67-marker haplotypes is 7, which projects back to a common ancestor born say 1425. However, there is so much slop in these estimates that one needs to make a mental note to append to any such estimates: “give or take a century or two”. Extending all these haplotypes to 111 markers should tighten these loose estimates significantly, bring the variance down to less than a century either way. Now how do these TMRCA estimates, predicated entirely on the mutational probabilities, square with the known genealogy?

The GDs of 6 and 7 separating Englishman Lee DANIELs from his Scandinavian cousins are in line with the idea that his earliest known ancestor, Thomas DANIEL (born say 1762) was a descendant of a Swede who emigrated between 1600-1750 from Sweden’s most important port, Gothenburg, to England’s most important port, London.

Likewise, the GD 6 and 7 matches between Lois, a DENNISON, and her ROMAN and LONGACRE cousins from the area of Gothenburg, Sweden, are consistent with the hypothesis that she and her PittsylvaniaCoVA cousins descend from a fairly close patrilineal relative of Peter Andersson, the ancestor of the LONGACREs, who emigrated from the Swedish port of Gothenburg to New Sweden on the Delaware about 1639. However, it is just as likely, considering her GD of 7 from Englishman Lee DANIELS, that her line, and that of the other Virginia DENNISONs, is a bit more closely related to the English DANIELS line and that they came, or were transported, to Virginia as indentured servants, like the 80% or so of Virginia immigrants during the colonial period—more on this below.

What seems much harder to swallow is the very low GD between Simo ROMAN, a Finn with exclusively Finnish and Swedish ancestry, and his two American LONGACRE cousins, whose direct ties with their country of origin presumably ended with the death of the immigrant Peter in 1679 about 8 generations back, and whose only known son first assumed the surname LONGACRE. Meanwhile, Simo has traced his patrilineal ancestor 7 generations back to a man named Anders Larsson, born say 1695, so supposing, say, that Simo’s earliest known ancestor was a great-grandson of Peter’s father, the expected (or most likely) number of mutations would be 3.6, and the probability that there had been only one mutation would be only 10%.

In fact, from the little I have learned of the LONGACRE ancestry, the perfect match between the two LONGACRE descendants at 67 markers looks suspicious too. These two are supposed to be descended from two different sons of the immigrant Peter’s son, Anders Petersson LONGACRE (1657 - abt Nov1718), and if we guess that it is 9 generations back to Anders, the expected number of mutations is 4 between these two descendants, not 0, for which the probability is less than 2%.

This may all be just a case of long odds results turning up now and then, but the conjunction of two long odds results having a composite probability of just 1 in 500 (.1 x .02 = .002) leads me to consider alternative hypotheses for the respective genealogies here. There seems little room to question whether Simo’s patrilineal ancestry is purely Scandinavian, and there was precious little back migration to Europe for most American immigrants, so that leads one to wonder whether the father of his first ROMAN ancestor may not have been also a father or close collateral of Swedish emigrants to America in the early 1800s, who for whatever reason adopted the LONGACRE surname. These hypothetical later immigrants may, in fact, have been distant cousins of the first LONGACRE of New Sweden, who maintained their extended family ties over the nearly two centuries for which they were separated. The probability of this is increased if it is true that the original New Sweden immigrant, Peter, made several trips back and forth across the Atlantic before finally settling in America for good.

That would be one way of accounting for this anomaly, but perhaps a better way is to back up and look at all seven of the LONGACRE haplotypes at the 37-marker level. We find in the first place a mutation in common for all of them, CDYa=36, that differentiates them from the haplotype of Simo and all the others. Second, the set of seven LONGACRE haplotypes as a whole looks like a typical set of haplotypes descended from a mid to late 17th century American immigrant. The two 67-marker haplotypes that I’ve been presuming all along belong to the two LONGACREs who descend from different grandsons of the immigrant Peter are still anomalous in being identical, but it’s always possible that my presumption is wrong, and even if not, much stranger things than this have happened.

The Fine-Grained Mutational Pattern Evidence

A fine-grained comparison of the specific marker values across the patrilineage indicates that DANIELS and DENNISON share at least one marker value that none of the LONGACREs have: they are DYS447=24, while the LONGACREs are DYS447=25. There are two possible interpretations of this: either DYS447=24 was the original value, and the mutation to 25 occurred after the hiving off of DENNISON and DANIELS, but before the emigration of the LONGACRE ancestor, and was thus inherited by all the tested LONGACREs, or the original value of DYS447 was 25, in which case the mutation to 24, which was inherited by both DENNISON and DANIELs, occurred during the 1-3 generations that their lines ran together before their final separation, probably due to the emigration of one or the other out of Sweden.

Without having a detailed grasp of the LONGACRE genealogy, and being able to evaluate the evidence behind it, it is difficult to choose between these alternatives (or even to be quite sure about the claimed LONGACRE descent from one Andrew Peterson Långåker born about 1657, which the DNA suggests), however, I do note from statistics on DYS447 at the Sorenson Foundation that DYS447=25 is the prevalent value across all haplotypes (3x as common as 24), and that it is also the prevalent value in the LONGENECKER and other R1b projects that I’ve looked at, so I will guess that it is the DYS447=24 value that is the mutation here. In which case it is more likely than not that the DENNISON and the DANIELs line ran together for several generation after diverging together from the main trunk.

Not too many generations, though, because the DENNISON line saw a mutation of the RPH (Simo ROMAN & LONGACRE) marker value of DYS449=31, first to 32 (for the Lois-10 haplotype), then another mutation in the same direction to DYS449=33 (for the other three DENNISON haplotypes), while in the meantime, the DANIELS haplotype at some point mutated in the other direction (losing instead of gaining a repeat) to DYS449=30.

Since it appears that Englishman Lee Daniels’s ancestor emigrated from Sweden some time during the period of Swedish expansion (1600-1750), presumably sans surname, one would expect that the DENNISON line was most likely brought to England by the same ancestor or by a close collateral, who happened to adopt the different surname DENNISON—a surname that I note fits with the preferred Scandinavian patronymic pattern. Of course other scenarios are possible, but they all have to accomodate the fact that the fine-grained DNA evidence suggests that the line that was to become English DANIELS and that which was to become DENNISON ran together for just a few generations during this approximate period. And as I’ve noted above, extending the marker panels of these descendants to 111 markers would probably tighten up the very broad estimates that we are stuck with here.

Finally, I note that the assumption of a Swedish-English origin for the DENNISON line fits best with their emergence in Southside Virginia about the mid-1700s, just when the descendants of original tidewater servants were settling on this inner Virginia frontier, and the given names found in this Virginia family are also strongly suggestive of specifically English origins.

How the History of the DENNISON Branch of the Patrilineage
    Might Intersect with Swedish History

If the DENNISONs of Pittsylvania County, in southern Virginia, were originally from Scandinavia, we need an explanation for how they ended up in the backcountry of colonial Virginia, and I believe that there is one ready to hand. First, though, we must consider a bit of Swedish history.

By and large, the emigree Swedes preferred the cold climates they were used to, so they migrated principally to Canada or the northern tier of the US (particularly Minnesota and the Dakotas), mostly in the mid to late 1800s. However, Sweden had expanded its sphere of influence about the same period that the Dutch were feeling frisky and establishing trading colonies in the New World, and the Swedish monarchs sponsored several of their own resettlement projects. But the only one that contributed significantly to American colonial history, was the colony of New Sweden (1638-1655), headed by Peter Minuit (also a director of the Dutch Colony of New Amsterdam), that was planted on the Delaware River before the North American colonial picture had quite gelled (here is another take on New Sweden).

This expedition was organized and launched from the Gothenburg area, and one presumes would have initially been confined to that population, but later waves of settlement were comprised largely of the Forest Finns—mostly ethnic Finns, but also poor Swedes whom the Swedish Crown had originally encouraged to settle in the northern waste areas of Finland, where they practiced slash and burn agriculture, and proliferated beyond the ability of this poor land to support them. It is claimed by LONGACRE descendants, however, that their ancestor, Peter Andersson, came over as early as 1639, so it is unlikely that he was part of the Forest Finn diaspora.

Inevitably, like the Dutch colony of New Amsterdam up the Atlantic coast from New Sweden, this Swedish project was elbowed aside by the much larger tide of English settlers, although the original Swedish settlers, who numbered about 600, were offered compensation by William Penn in the form of a 10,000 acre tract in Pennsylvania up the Schuylkill River called the Manatawny Creek settlement, and it appears that most of the colonists relocated there. Others, however, fanned out south into the Chesapeake tidewater region and merged with the population of tidewater Virginians who gradually pushed the line of settlement out into the Virginia and Carolina piedmont area on the doorsteps of the Blue Ridge Mountains.

It would seem that this history of Swedish colonization dovetails almost perfectly with the likely history of the DENNISON line in America. In particular this article on the breakup of New Sweden provides examples of Swedes filtering down to the Southside Virginia county of Bedford (created 1754 from LunenburgCo) which had a long border with Pittsylvania (where the Virginia DENNISONs first emerge) when that county was created in 1767 from Halifax County. Of course, it is also possible that these Scandinavian DENNISONs of Southside Virginia diffused initially west into the Pennsylvania hinterland, and came to Virginia later, taking the North-South roads paralleling the Blue Ridge Mountains down to BedfordCo and parts south.

Research Implications of the Theories Presented Here

Unfortunately, I am unable to comment on the prospects for further research in Scandinavia because that lies well outside my area of expertise. Nor have I heretofore had occasion to research the colony of New Sweden on the Delaware. However, it is a point in favor of the work that both Simo ROMAN has done, and that the LONGACREs have relied on that their respective patrilines appear to converge on the Gothenburg area in southwestern Sweden, while the DNA tells us that they must converge somewhere.

However, the colonial records of New Sweden, and of Pennsylvania and Virginia, are accessible to us American researchers, in the usual libraries, archives, online resources, and most of all, LDS films. From my correspondence so far with one LONGACRE descendant, I gather that most or all of what is known about the first two generations of their line comes from a pair of books by Peter Stebbins Craig: The 1693 census of the Swedes on the Delaware (1993), and 1671 census of the Delaware (1999). And he has been good enough to send me photocopies of the relevant pages on his ancestors, from which I have culled the material above. Craig cites his sources, both primary and secondary, though most of these resources are at present out of reach due to the COVID-19 closures. One of these sources, though, is the New Sweden Journal, and among the historical journals cited that I recognize is the Pennsylvania Magazine of History and Biography 2:226, so these secondary sources would be a couple of starting points for locating the ultimate, primary sources regarding the American immigrant generation of DENNISON Patrilineage 2.

Lee Daniels, meanwhile, needs to look for records and other material on immigration from Sweden during the period 1650-1750.

DENNISON Patrilineage 2 Descendancies