Yet Another Update to the Pettit-Mellowes Y-DNA Project

The story has been told before. Still, briefly...

Joseph Pettet was born in Pennsylvania in about 1815. When a young child, his family settled in Muskingum County, Ohio (1819). We know this and the general structure of the family from his older sister, Lovey Hearing, who provided the narrative for a 1905 article — at the age of 96.¹ Her father was Thomas Pettet of Pennsylvania, the son of Elias Pettet (-1794), also of Pennsylvania. Thomas's brother, Elijah, arrived at Zanesville with his family four years earlier. But researchers could find no definitive place on the family tree to place Joseph Pettet despite the fact he lived smack dab in the middle of this extended family. We spent decades tracking him through three states over several decades, collecting all the Pettet material in the relevant places. Other than census records, no reliable information about his birth in Pennsylvania was found as well as no information about his death (in Missouri). There is no precise reference to him among family records. It's possible he was the "one boy who was lame," or a deceased brother as inferred in Thomas Pettit, Jr's 1888 death notice.² Still, a great deal of circumstantial evidence points to a close relationship to the larger Pettet group in Muskingum, Morgan, and Perry Counties, Ohio. Beyond that, we could only speculate.

Nevertheless, proof that Joseph was indeed of these Pennsylvania-Ohio Pettets has been found. And wouldn't you know, it was the Y chromosome that came to the rescue. About a decade ago a well-documented descendant of Thomas Pettet's son William tested. And about two years ago a distant Pettet cousin of mine, also from my Joseph, came along and was found to be a perfect match to the first tester. The obvious conclusion, supported by several autosomal tests, is probably true: The one unidentified son of Thomas's was our Joseph.

But that's not the story I want to tell now. I want to describe the Pettit-Mellowes Y-DNA Project and revel in the progress we've made in just a year and a half. This story starts with Puritan immigrant Thomas Pettit, born in England about 1609, and his wife Christian, who might not have been a Mellowes as rumored for about a century. I'd be hard-pressed to guess how many living descendants there are, but suffice it to say that the family has been reproductively successful.

During the years I published The Pettit Correspondent I learned there were (and are) Pettits of multiple origins in North America. We now have genetic proof of it. Indeed, I've been in touch with four Big Y testers for which Y-DNA testing has shown no paternal connection to our Pettit-Mellowes clan. Even that much is an accomplishment for a researcher — to know with certainty to whom you are not related. But DNA, and the Y chromosome in particular, has the power to permanently dispel myths — ghosts that have been visiting us for no real reason at all. But its most practicable use is as a means to estimate the degrees of relationship between patrilineages. To do that, we need first understand the Y's inheritance pattern. With that knowledge, every reader can appreciate the significance of the following tree. Each tester, including my cousin, is genetically related to one another by varying degrees. The objective is to home in on a more precise relationships through both genetics and genealogy, which are inextricably linked. In short, genetics can answer the simple question, Is my ancestor your ancestor?

Haplogroup R1b-BY200368

Click image to enlarge

I'm reminded nearly every day that a number of researchers neither understand nor wish to understand genetic inheritance patterns nor their relationship to genealogy. Understandably, they merely want to know who their 6th great-grandfather was. I get it. But that, more often than not, is not found in traditional genealogical or historic records. But I now fully understand, despite my passion for teaching this stuff, the reader has to take a degree of responsibility if they wish to have any understand of what is, in fact, a very simple idea. Still, it incurs some subtleties.

Forget everything you might have learned about DNA. The Y chromosome is a different beast since it carries the male sex gene (the SRY gene). Therefore, it passes only from the father to his sons. It's really that simple. (If you don't want to know more, skip to What It Means).

Y-DNA Inheritance

Despite this unusual pattern, the process doesn't break the genetic inheritance rule that commands we get half of our genome from each parent. It's simply an emergent quality due to the presence of SRY gene on the father's Y chromosome.

Our chromosomes come in pairs, each half of which is a reshuffled half of each the parents' own genomes. The tricky part is the 23rd pair, the so-called sex chromosomes. Women have an XX pair — one of her mother's two Xes and the father's X — and men are XY, again the X is from one of the mother's. Since the Y chromosome comes from the father that's exactly what he got — the father's Y. Because of this, brothers have exactly the same Y. It's a toss up, however, as to which of the mother's two Xes each child gets.

The process is the same for the remaining 22 pairs. The results are simply more subtle than sex determination. We get one of our father's two number 1 chromosomes, each of which is a reconstructed half of each of his parents, and one of the mother's two number 1 chromosomes. This continues from one generation to the next, halving an ancestor's DNA with each iteration. Our brains can handle only so much recursion making it difficult to visualize. But the basic rule is simple enough and results in fewer and fewer bits of DNA from any one ancestor. Exponentiate it to any degree you wish. The base rule never changes, except in the case of rare mutations, such as a third copy of chromosome 21. That results in Down Syndrome. The point being is that the sex chromosomes behave no differently except that the sex-determining gene always comes from the male. Either it's there or it's not, the Y being nearly an exact clone almost forever.

23 Pairs of Chromosomes



Which are Ma's and which are Pa's?

The results of the XX or XY pairing is unusual, but the SRY gene could just have easily originated on the number 10 chromosome. There's nothing special here. The process is found in all duo sex species. (The New Mexico whiptail lizard is entirely female and clones from one generation to the next — and lacks genetic diversity.) Because of this duo sex system, we have the straight male-to-male inheritance pattern. An interesting sidebar to that is that the X is genetically robust, while the Y tends to shrink due to the lack of recombination. Rather than gaining new parts through recombination, genes break (as genes do), lose functionality, and become largely meaningless. This results in few genes on the Y, and they tend primarily to support the SRY gene. This so-called "junk" DNA regularly gets sloughed off and the Y shrinks in size and is, therefore, the smallest of all chromosomes.

The Y chromosome between paternal cousins, even in the 6th degree, might vary by only 0.0000001% between one another. Because of this cloning of the Y, our male Pettits have the same Y and they look virtually identical to Thomas Pettit's Y of 400 years earlier. His body doesn't need to be exhumed to prove this due to that simple fact. Every living, biological male-lineage descendant has those same markers, as will any living descendant of Thomas's paternal brothers, cousins, uncles, etc. We have proof of it in our SNP tree. The Y for kit #764323 looks exactly like that for #261925. There's only one genetic path by which that could have happened, and that path has been confirmed by hundreds of thousands studies, exhumed body or not.

Who Gets the Y?

Quite frankly, even that much doesn't need to be understood to appreciate what the project has accomplished to date. One merely needs to play the same game quantum physicists have to play: Accept the fact that it works: only men have a Y and every son inherits it. Just go with it.

The Pettit-Mellowes tree at top reads just like any genealogy tree of human families, the Darwinian descent of fish, the branching of languages, or the XX-XY tree above. Each node on a family tree represents a couple. That same basic pattern iterates over and over. It's a binary tree to the extent that it traces back through only one ancestral path — the father, not the mother lineage. By the same token, only one of any number of possible lineages are illustrated in the linguistic tree. We know, for example, that several languages influenced modern English. Still, a linguistic branch is distinguished from a neighboring branch once a sufficient number of words and syntax changed to the point that it has its own characteristics. These words and syntax variances are analogous to the genetic differences among the different fish species, all of which descended from a long-extinct proto-fish, if you will. (You might enjoy the book and documentary, "Your Inner Fish" by Neil Shubin.) The branches in our DNA tree represent the rare occasions when a single genetic letter on the Y chromosome mutates from one value to another of the four letters (A, C, T, or G). For example, the marker that presently defines the Pettit-Mellowes Project, BY200368, mutated from a G to a C at position 8202198. In all these examples, the top represents the origin, the bottom the latest or current generation. All the testers have that marker for the very reason of the nature of Y genetic inheritance, explained above.

A new branch on the tree occurs each time a tiny genetic letter mutates (one out of 57 million possibilities on the Y). Because of this inheritance pattern its passed to all subsequent generations, just like an unmanned boat being swept along a river. Like the boat, the SNP mutation continues until it reaches its terminus. And, just like the boat, it has a starting point — the man in which the mutation occurred.

Here's another made up tree to illustrate the point. The ancestor has a G at 8202198. (Thomas, or most likely a predecessor of his, acquired a C at that position, which passed to all his male-lineage descendants.) In our pretend tree, the G continues to pass down. But one branch mutates from the G to an A. Testing the descendants of these lineages, we find three distinct markers, G, C and A. We can't distinguish between the various G branches, but the differences between the A and C lineages are telling and recognizable.

Follow the SNPs

The G-to-C mutation that Thomas Pettit had likely occurred before his birth in c1609. The seven other mutations at that level of the tree didn't occur all at once. It might have taken 800 years for them to have been acquired by Thomas's paternal ancestors. This means we have no idea as to what order they emerged. BY200368 was chosen by the lab as the lead for completely arbitrary reasons. Another fact we merely must accept.

Now, let's take a look at the bare bones SNP tree for the project. We can easily follow each person's SNP path going up the tree; for example, the last kit listed, #261925. Both he and his testing companion have FT95921 as their terminal SNP. They also have Y42738, a SNP we now know belongs further up the tree by virtue of the fact that a somewhat distantly related family of Pettits also has it. And all the Pettit-Mellowes has BY200368. It's a simple idea. The more people sharing a particular SNP, the older the SNP, and the more genetic material you share with another, the closer you're related. Really, this is as straight forward as a Tom, Dick, Harry family tree. You could replace the SNP names with human names and all would become clear.

The SNP path for #261925 is BY200368 > Y42738 > FT95921 > #261925. We know this because each SNP was found in his sample, just like all the debris collected in a river ends up in the bay. But those of you who are still having trouble reading these tables, I've created a hybrid genealogical-genetic tree. Seeing this alongside the genealogy itself might help.

Four genetic branches of the family are clearly represented on our Pettit-Mellowes DNA tree. The first, BY200368, defines the group. All testers have it and all testers are (or presumed to be) descendants of Thomas and Christian.

The first of these belongs in the Nathaniel Pettit lineage. More precisely, the defining BY198412 appears to have emerged at the birth of George Pettit in 1719. We know this because only George's descendants have it. And that's proved by the fact that a descendant of George's brother, John, lacks it.

The two F7174 testers are descendants of Samuel Pettit Jr (1790-1874) and Sarah Gallagher. We don't know yet when F7174 first appeared in the lineage. Finding a collateral branch to test will tell us more. And we don't yet know from which of the second-generation brothers he descended. I suspect Joseph. But please note that the project has funds to help test a proven descendant of Joseph's.

The third lineage brings us to the best developed branch of the tree, Y42738. Two distinct subclades have been found, FT360712 and FT95921. As I understand it, the father of the Elijah Pettit who married Tabitha Brasheers hasn't been adequately proven. It's certainly possible he was a grandson of Joshua's, listed on his left. This is only one of several pressing mysteries, but the lineage is rich in SNPs. These might well point the research in the right direction. In any case, it's suspected that Joshua was the son of another Joshua (in fact, it seems nearly certain). However, the paper trail is simply not there. We need a tester from the Increase Pettit (1726-1795) lineage, the probably brother of Joshua II. As I understand it, his descent from Thomas and Christian is well documented as,

Was Increase indeed a brother of the Joshua? Most of us would guess yes. But genetics just might be the only way left to prove it. Did Increase have both Y42738 and FT360712, or only Y42738? (Inquisitive Pettits want to know!) No matter the scenario that presents itself, test results will provide more data with which to make a better judgment. To have a verified descendant Increase's step forward would be as important as having a Joseph tester (son of Thomas I, above). Therefore, some or all of the present funds could be applied to a new Increase tester rather than (or in addition to) a Joseph tester. Someone please accept one of these offers.

Further, the whole of the "Joshua" lineage is rich in personal SNP mutations / personal variants or PVs, shown in light blue on our SNP tree. Each would have come into the respective lineages via any number of the testers' ancestors. This branch has great potential to being fleshed out. Please, please ask me questions about that.

We next have the FT95921 subclade of Y42738. Two branches are known, a Fox family and a descendant of William Pettit (c1779-) and Rebecca Clayton. We have no idea who the families' Most Recent Common Ancestor (MRCA) was but we do have the happy circumstance of a defining SNP, FT95921. And that means they shared a common Pettit ancestor in the not-too-distant past. Beyond that, this group has a dearth of SNP mutations, the Fox tester having no additional markers. By finding others positive for FT95921, we can begin to piece together the puzzle and perhaps close in on their MRCA. We have only two SNPs involved, Y42738 and FT95921, each of which can be tested at YSEQ.net for $18. Again, I would welcome any questions about this.

Finally, we have the very first, and rather nebulous, branch on the tree designated as BY200368* but not on the list just above for the reason that it's not a clearly defined haplogroup. The splat is a conventional notation. It simply means the branch is a member of BY200368 but has no distinguishing markers below it. Still, it's likely the two testers are also descended from Nathaniel. We do know with certainty that neither of the lines are descended from George, otherwise they'd have the BY198412 marker. Nor do they have the FT51308 SNP that emerged somewhere along the Pettit/Wood lineage. And that illustrates the concept of proof through negation, or the process of elimination. (I.e., I don't descend from George, nor from anyone listed to the right of him in our tree.) The Jonathan tester does, however, have two personal variants (Y163386 and MF44765). PVs are simply SNPs found in the tester's sample not yet matched with another tester. In other words, we have the potential in finding two new branches in his tree, one of which might (we would hope) define the whole of Jonathan lineage. But the Elias branch, that which I belong to through the probable brothership of my Joseph Pettet to William Pettet (1805-1885), has no PVs. Genetically, Elias's branch is the most SNP-naked of all Pettits tested to date. Still much thanks to my distant cousin for testing! His results represent a giant leap for our lineage.

My personal Pettet story (at top) provides an important lesson. Although Joseph Pettet's autosomes show up in my cells, the Y lineage was broken by his daughter's marriage to Greenbury Cooley. No Pettet Y for me. But not all is lost. As demonstrated with my cousin, I have Pettet cousins who are eligible for testing. The story is recorded in their Y chromosome, if minimally so in this case. But the deeper story is still present and the Pettet SNP archive has continued to the present date and, so far as I know, to succeeding generations. Lucky for me, lucky for all non-Y Pettets. In fact, I'm doing the same thing with my Duncans, my Johnsons, Woods, Aikens, Fisks, Struthers, and Harts. And more are yet to come once I've rattled enough cages! And that's another lesson. The Pettits here represent only a single example of the much bigger story, which doesn't change much. In other words, I could plug in any name and its corresponding SNPs and come with an article for Cooley, Strother, etc. Once you get this story, you've gained an understanding about this essential tool for unlocking all your ancestors' lineages.

One further bit before bidding adieu. If you are unable to donate your spit to the cause, or that of your brother, cousin, etc., please consider donating to our fund. It's sitting at a reasonably substantial amount right now (thank you!) but will disappear quickly. But for those of you who can't see benefit in testing, I would say that if you are not a Pettit-Mellowes descendant, there is no benefit. But those who are and are "eligible" testers by virtue of having the Pettit Y, belong in one of two camps. One, if you have a known lineage back to Thomas, your test will provide genetic data that will define your specific lineage, as above. Others can piggy back on those results, as I did with my cousin. Two, if the lineage back to Thomas is unknown but the connection is strongly suspected, then there's a good chance of receiving results that will dovetail with a known and tested lineage, markers magnanimously provided from men in Camp One.

In other words, if you know your lineage, your test will provide data for those who don't know it. A good example is the Fox tester who now knows he's most closely related (paternally) to William Pettit (c1779-). In that way, the more testers we have, the greater the number of connections we make. Simply, known lineages supply defining markers. Those same makers will be found among the unknown lineage and the genealogy tree gets bigger as the genetic tree grows. Just take a look at the first tree I produced for the group. The tester on the left is our first Pettit Big Y. (Note the presence of the BY200368 SNP.) The column to the right of him includes distantly-related surnames (none of real genealogical value to the Pettits). And this is what the tree looked at with five Big Y testers resulting in a total of one haplogroup and two subclades.

That's a far cry from the tree at top — thirteen testers and six haplogroups and subclades. And we're just starting. (And remember, funds are available to the right tester.) Genealogy-speaking, Y-DNA is a hugely successful tool. It has already revolutionized genealogy. Let's take this bus into town.

1. Michael Cooley, ed., "The Oldest Woman in the County: Mrs. Hearing is Ninety-Six Years of Age," The Pettit Correspondent, page 1.

2. More detail is found on my webpage for Joseph.