Z18 Panel Results and Nostalgia

As per usual, while I wait, I write. Lucky for me there are lots of great new things over at the R-Z18 and Subgroups project. One of these is the recent returns from the Family Tree DNA Z18 panel. As the news says, there are some SNPs that didn't get coverage and a few that had issues, but there are some that are pretty interesting as well.

Z18 Panel/Big Y Conglomerate

The Z18 group site has the panel results split out from the big Y results. I decided to do a DF95 mashup of the different results pages.

Once you figure out how you're all alike, the next thing you want to know is how you're different. Everyone who has tested big Y so far is in the 458.2 side of the house...or incommunicado.  So the Z18 panel has offered us a way to see some of the results from the normal 458 side of things (at a price that is pretty affordable in the Y testing world).

The problem with our 458.2 plus or minus groups is that 458.2 is not really an SNP. It acts somewhat like an SNP, but it's really a STR mutation..one that seems to stick. So my reference to 458.2 is not one you'll see in the big Y or Z18 panel results pages. At some point down the road, I expect it will become just one of many branches and categories a Cumberland cluster person will fit in and as STRs become a thing of the past, and SNP testing becomes cheaper, 458.2 will cease to be a category at all.

Anyway, on with it.

As per usual, if you want official information and good use of the scientific method, see the Z18 group or the U106 group.

Here is a horizontal table of SNPs across the top (along with DYS458.2) and testers on the left. I've got the Big Y testers, hopefully in position based on their SNP matches.  I have them filling in at the bottom with yellow names to differentiate them from the panel testers at the top. I also shaded DF95, which has been our defining SNP.

What you'll see here is that everyone is pretty much the same. White squares are basically, untested. The red ones are positive and the green ones are negative. You can see that Edwards there has an issue with Z369. I'm guessing that is a "false negative".

So that is the Cumberland world, generations and generations of common ancestors that inevitably come down to one single person that all of us are related to. Call him Mr, S4022...or further back Mr. DF95...or Mr Z370...etc. When contemplating the vast expanse of time boiling down over and over again to one man who is the father of all of us, it helps to remember that we're the descendants that lived to this point, and tested. That man had uncles and brothers, but they just might not have made it over the long haul or we haven't tested them yet.

Here is a second table where things start to break up some more. Again, white blocks were intended to be tested, but for one reason or another there are no results. I've thrown in a black vertical bar to separate the SNPs that are currently in the panel (on the left) from those in big Y (on the right). I'm using the Z18 project names for these SNPs but it's good to remember they have multiple names. For instance, ZP129 is also called A2277 at YSEQ and ZP121 is called Y15995 at the U106 project (I'm guessing named by Y-Full).

In this last one, we start to show our stripes.

ZP84 has some issues, it seems to be unstable in the big Y results, sometimes dividing people who are more closely related..seemingly at random.

ZP85 is down right exciting. ZP85 is so pervasive that you have to wonder if Corson's negative is a false negative. If it's a real negative result, then that would put the Corson(Jansen) men on an older branch than anyone else.

Here I threw in DYS458.2. It seemed to me that it's not tied to ZP85 because some men who are 458.2 negative have ZP85, so if 458.2 is really that stable, then it's younger than ZP85.

Mind the gap. There were several DF95 related SNPs that didn't get reported in this round. You can see that several of the Yellowish big Y testers have them and they are pretty common among them, but there just isn't that information for the panel people. It makes it hard to place them, since we would now be watching Corson to see how these came out for him and the other 458.2 negatives.

I shaded S8387. Again here it's very common in all the big Y testers (as is S8388), but it was only positive in Emery from the Z18 panel. So there is a clear divider there as both 458.2 negative and 458.2 positive men are S8387 negative. It comes after the 458.2 break up.

The panel ends, kind of at the last match point for Lund and Ovens who form the oldest branches of the Cumberland big Y men. If those S8387 results are stable and correct, then it seems like Olds, Edwards and Burghgraeve are now on  the oldest branch of 458.2 men though. They can, in turn, look back at Wright and Little. Everyone here would have to look back at Corson.

Beyond the big black bar is only Big Y land (for the time being) where we have a few more branches (explored and misused a bit more here on a broad scale and here in the nearer term) .

Update 2015-09-05

I was browsing out at the Z18 Project site and noted that a new result came in that has some bearing on ZP85. Schmidt came back ZP85 negative as well. Very interesting to see the layers coming out. I still have my fingers crossed that we'll get one of the Cumberland A people a big Y test so they can start seeing their own branches of our family tree.

Nostalgia

What made me smile about this group of men in the Z18 panel and Big Y conglomerate is that I see a few familiar names from my years writing this blog and speculating.

Back in 2010 I was scouring Y databases searching for matches and patterns. What makes us alike and what makes us different see?

So for fun here are some references to different matches I found and my attempts to place them on maps and in the context of my own weird results.

The Knowltons. Really the first big family of matches I could identify and my longest running research partners. Without their family tree work and testing, you wouldn't see the results we have today: http://thompsonhunt.blogspot.com/2010/10/16-slow-markers-map.html

Winne (murdered to Winnie...Sometimes Winner) and Corson/Jansen:  http://thompsonhunt.blogspot.com/2010/11/migrations-3-other-databases.html

And from 2012, Mr Emery who has helped me out immeasurably: http://thompsonhunt.blogspot.com/2012/08/z14-it-is.html

Unfortunately, I cannot show a link for Mr. Burghgraeve although, it seems like he is a node on one of my maps holding down the fort in Belgium.

Of course the Elmers who I totally disregarded in the early days at SMGF. Luckily Mr. Emery made me take another look at them: http://thompsonhunt.blogspot.com/2011/05/while-im-waitingyet-another-map.html

It's funny how it always boils down to the people who help you out or give you a new direction or a fresh look at an old direction. This list cannot possibly contain all the people who have been resources for me over the years, although I wish it did because I'm so curious about the Coens, the Edwardsons, the Hudsons and the Steiners...it goes on and on.

Partial Return from YSEQ

I thought I would mark the time a bit in a post about my experience with YSEQ so far.

I ordered six SNPs at $17 apiece on the 20th of June. My kit arrived at my house a few days before I sent it back out on the 27th of June. The return address for the kit was Berlin Germany so I expected it to take a bit to make it there. YSEQ changed my status to processing on July 9th. Four of my six SNP results were returned on the 15th of July.

That is lightening speed in the world of genetic Y chromosome testing. Typically, we would wait for the seasons to change before looking for results back.

You can see my post about what we expect from big Y and YSEQ here, just after I mailed off my kit.

Here is the chart of our testing expectations.

• I tested for one SNP that was shared by the Knowlton family (A2277).
• Four SNPs shared by our two Elmer testers (M1 and L2) (A2278, A2280, A2283, A2284) 
• One singleton SNP of L2 (A5920).

Here are my partial results:

1984	A2280	ChrY	15779806	15779806	A+
1984	A2283	ChrY	18553735	18553735	T+
1984	A2284	ChrY	21186403	21186403	A+
1984	A5920	ChrY	18578248	18578248	A-

My kit there is number 1984. The SNP names are assigned by YSEQ. The two number columns are the SNP locations on the Y.  The results are positive or negative for the variant.

I'm still processing  the Knowlton SNP A2277 and the Elmer SNP A2278. At this point,

I was negative for the A5920 singleton SNP for L2 in the chart. I was hoping for that smoking gun, but with only one singleton SNP from L2 to work with, my odds were not that good. M1 meanwhile has four SNPs to himself. The odds of identifying a branch from his singletons will be better.

Where does this leave me. Well, I have a hunch that I'll be positive for the Knowlton/Elmer SNP A2277. The Elmer SNP A2278 is a bit harder for me to call. I of course want it to match so I can, at a minimum, close the case file on me and Ed Elmer with a non-ambiguous win.

If I nab that Elmer SNP I can pin Ed Elmer down and bump my last known paternal ancestor to 1613..ish.

Whether or not I pick up that last Elmer SNP I will need to wait for the big Y results from our pillar R1 to see how we all fit with each other.

Depending on how things go with those results, I may run through a few more singletons for M1 and R1 to see where I stand. Should all of those comparisons leave me right where I am now, I'll probably then shoot for Y Prime (sticker shock at $750) and try to convince L2 to test some of MY singletons...moooohahahahaha! Before that though, I really need to get back and follow up on the U152 Thompsons. Their brick wall match may be a few hundred dollars away.

Common Ancestors and Speculations

While I'm waiting (which is very hard for me to do) I'm prone to fits of speculation. With another round of age estimates, I thought I would pick up where my last post, kind of, left off.

My mental questions are basically, what happens if Norway is considered a refuge where our Y waits for centuries before moving on. What if the Netherlands is such a refuge later on? What if (as the new SNPs suggest) Poland is not a source, but a destination for these people? What if England is another refuge for these Ys until they break out into the Americas?

With the earliest U106 positive human in Sweden in the early 2000s BC. R1b-Z18 being heavily Norse and DF95 being positioned by others after big Y testing as coming out of Norway (because Ireland seems unlikely). I'm feeling more comfortable with my past speculations on Scandinavia's involvement in pushing my Y around the North Sea and Baltic.

So here are some of the more recent Big Y related age estimates for a common ancestor shared by the U106 group and some labels to help sort them out a bit:

• All U106 testers so far: 2420 BC
• All Z18 testers so far: 2100 BC
• All DF95 testers so far: 219 BC (This is the most recent ancestor with the Blue Stars)
• Unamed 16524131: 538 AD (most recent ancestor for Blue Squares and Blue Diamonds)
• Elmer/Knowlton/Lunceford: 849 AD (most recent ancestor for these families)
• Elmer/Knowlton: 1030 AD 
• Elmer: 1614 AD

Here is that map of the DF95 big Y testers (rough positioning):

Norway has deep ties to Ireland and Scotland. The Norse in Ireland are well documented. I think that is one reason that Ireland is not seen as the origin for our group. There isn't a great record of the Irish invading Norway, but we definitely know the reverse is true. The high incidence of Scandinavians in brother clades to DF95 (like Z372 and it's child L257) also add weight to a sort of  Y "patient zero" in the Scandinavian world. 

It occurred to me today that we might think of some of these locations as reservoirs for a Y if not really an "origin". Considering our Y DNA like a virus, where does it hang out between outbreaks? It may have sprung up further down the continent for all we know, but it seems to have found a safe haven in Scandinavia for a number of centuries. 

Excuse the crude drawing, but this might explain what I'm thinking about these big Y results better than words:

Blue Stars 219 BC

So we have our reserve of Y DNA, from that reserve you could pull any of the DF95 Ys except for those that are localized outside the reserve. The Y testers from Norway and Ireland didn't stop in time, they continued to create SNPs down their respective lines. You could pull a Y out of the hat that was closer to either one, it just happens at this time that we haven't. 

So while I would expect to find people closer to our Ireland tester in Ireland, we could also find people closer to him than the Norway tester back in Norway. The two lines of men may have lived side by side for hundreds and hundreds of years.

Wait, they are farther back in time as a common ancestor and diverge from there, how could both exist in the same place to be picked from?

For that, we can think about the example of the oldest known Y in the world. Found in the descendants of an African slave in the US. His Y A00 is something like 300,000 years old. It's not that his people colonized the Americas 300,000 years ago. He was pulled fairly recently from Cameroon where some other inhabitants share his very old Y. Tellingly, not all the inhabitants of this part of Cameroon share his Y, just some of them. So we have an example of a very old Y living next to younger Ys in the same village and then randomly being selected for an overseas voyage (in this case, unfortunately into slavery).

What I'm scratching at is the idea that even though everyone shares a common ancestor at about 219 BC it doesn't necessarily mean they parted company in 219 BC. SNPs aren't necessarily created by travel. We can keep going back to the well and getting Ys that diverged from each other a long time ago but continue to exist in the same village.

We're all Blue Stars after all, the rest of us just happened to get more testers on our branches to test.

I gave them an "orange-ish line to denote their differences from the "maroon lines" people.

Maroon Lines 538 AD (16524131)

Initial thoughts.

The orange-ish line runs from Norway over to Ireland. I've put the maroon line people coming out of the water and rolling basically southwards out of the common reservoir of DF95.

These maroon lines represent the 4 or 5 (one is kind of troublesome) SNPs shared by both the blue squares and blue diamonds. Although I have multiple lines, they are really a single man that the Diamonds and Squares are related to in 538. Again, we could be staring at the mark of some migration to another part of the reserve or we could be looking at someone who lived side by side with the Blue Stars and had a minor difference that he passed down to the Blue Squares and Diamonds. It could be the mark of a 700 year layover in some part of the DF95 world. We can't be sure because we have a longer run of SNPs without a geographic pin for some divergence. 

At some point the two groups diverge genetically and they eventually diverge geographically. Here is my initial map of what that might look like. Later though when looking at the Blue Diamonds more closely, I have some second thoughts and maybe some alternate timelines. 

Blue Diamonds

So my basic Idea of the Blue Diamonds is that they also represent the Scandinavian influence in both the Netherlands and Poland. My thought is both those places saw significant Scandinavian input over the centuries with the Netherlands suffering right along with France with the Viking invasions in the 700s and 800s and Poland saw norse settlement, trade and raiding like  "Jomsborg". 

The same men who bring you the "Great Heathen Army in 800s England are also invading Dorstad in the Netherlands and attacking future Belgium as well as France.

I've spent a lot of time speculating about Norse input into Poland. They have the proximity and the history. With input from both the Swedes and Danes at different times. It helps to remember that the Rus in Russia were Scandinavians. 

It isn't hard for me to imagine waves of Norse influence pulling from that common reserve over and over. So my map for the Blue Diamonds would be like the one above for the Maroon Lines. 

An Alternate Timeline?

I wondered if I hadn't really given the Blue Diamonds as much thought as they deserve. I am evaluating them based on my thoughts about the Lunsfords, Knowltons and Elmers which I've asked for age estimates on, but I didn't have estimates for the blue diamonds. 

So, they diverge from the Lunsford/Knowlton/Elmer man, but where do they go from there? 

If you look at the Elmers (regardless of the Lunsfords and Knowltons) as an example, they have about 8 SNPs different from the common ancestor of the "16524131" group. So one of the Blue Squares is about 8 SNPs away from the last ancestor of the Blue Squares and Diamonds.  

I realize that you can't just count SNPs like this, but I wanted to think about it a different way to see if there is a more meaningful explanation.

I decided I should look to see how many SNPs were shared by the two blue diamond testers to see how far they were from the same 16524131 ancestor.

Looking at them with new eyes, it seems they share about 8 SNPs with each other.  Making them roughly as distant as any two Elmers (if that is the way it works out). The Elmers are aged for a common ancestor at about 1614.  

If the SNPs stack up the same 8 and 8, then maybe the two people from Netherlands and Poland are roughly about as close to each other as the two Elmers? Even if they had a common ancestor in the 1500s that would be fairly close to the present.

I decided rather than pursuing a connection directly with Denmark or Sweden to look at relations between Poland and the Netherlands in the range of the 1600s. It turns out that the Netherlands and Poland have a pretty extensive past as well and a good relationship as nations today. Apparently the Dutch had a major influence on Poland and Dutch settlers moved to Poland in the 1600s. 

Keeping in mind that I am deep in speculation territory now without any real estimates..I wondered if my last map should look more like this:

Maybe this map would better show how closely related the Netherlands tester and the Poland tester are. if we were to estimate them like the Elmers then it seems reasonable that they would branch off around 1600 too. Including myself here for a moment (dangerous because my tests aren't back). It would roughly be like considering the Elmers and Thompsons migrations here in North America with our common ancestor in 1614 only it takes place on the European continent.  

After a Rough Estimate

The age Estimate for this group came back as 1099 AD with a range of 558 to 1573 (Thanks U106 group volunteers!). So roughly 1100 AD. Not exactly as close as a simple count of SNPs would make them. I could not find much in the time period that would help me place them. Specifically, I was looking for connections between the Netherlands and Poland that might mark this geographic split. 

My idea for an alternate timeline (above) could still fit here if they hung out in the area of the Netherlands for several centuries as two lines of anciently related men and then one person from those lines happened to move on to Poland. Certainly we see something similar to this in the Blue Square group of men and the Americas.

Also my map could still fit if we consider the Viking input into the Netherlands and then the transformation of some of those vikings into regular Holy Roman Empire/Frankish/Dutch citizens. Eventually, like in Normandy, some of the vikings were just accepted as part of the empire. 

Perhaps they moved into the Netherlands with the great heathen army and set up shop like Rorik becoming part of the people of the Netherlands with time, then moved on later to Poland either with the Franks or as part of that later colonizing Netherlands to Poland effort?

As always, I think we need more layers of men that we can pin to more locations.

Blue Squares 849 AD (Lunceford - Elmer - Knowlton)

I'm making the supposition that these men are all British. The Elmers and Knowltons have some paper work that makes that reasonable. We found two Luncefords that match STRs, the big Y tester has an idea that they are British, but no good tie back to England. The matching Lunsford STR tester shows Scotland as an origin. I chucked the Lunceford square in England for ease and because there are places called Lunsford in Sussex.  

We're extremely lucky with these big Y results to have some more layers in the Blue Square area. The estimate for a common ancestor in 849 AD puts that ancestor in the middle of a pretty crazy time in English history. It's just before the Great Heathen Army runs rampant in England for 14 years. It's at about the same time as the attacks on France and the Netherlands.

This nice map, I'm stealing from Wikipedia shows some of the movements over those years. 

I'm wondering if this layer in our DNA is a product of that movement of Scandinavians. I would also be painfully curious to see if there was a similar layer for our continental testers if we could break up their 8 SNP run the way the Isles people are broken up.

I'm marking this pinpoint on the map as when we have our earliest big Y SNP in England, but really, as much as I favor this dispersal method, we don't have all the data. Several Cumberland cluster A people are also in England, Ireland, Germany and the Netherlands. They would have traveled right along with the big Y testers, but they are not being represented in big Y yet. What would their SNPs and origins say?

In light of that, and the dates for our big Y Blue Diamonds. Maybe it would be helpful to consider another alternate timeline with a slightly different map.

Alternate Timeline 

In this map I'm trying to illustrate would could happen if we make a couple of movements into a couple of refuges. I left our friends from Norway and Ireland as such. Pulling from the reserve of Norway for their DNA, most likely with the Vikings.

It's impossible to talk about the Blue Squares at this point without the Blue Diamonds. So you see them in purple.

I'm also imagining a migration to Denmark with a layover of several centuries and then through Denmark to the home turf of the Angles and Saxons. So 219 BC, Norway who directly contributes to Ireland well down the road in the 800s and 900s AD. Norway sends some of it's group south into the soup of Germanic peoples that is Denmark, Frisia, the Saxon Coast..etc. Around 538 AD one group of Angles or Saxons or Frisians...maybe even Jutes (squares) has lobbed itself at Britain. While another group sticks to the local program (diamonds).

In the 800s the Blue Squares and Blue Diamonds are defending their respective territories from the Vikings (who they are, ironically or not, distantly related to). In the 1100s the Blue Squares are still knocking about Southeast England as Saxons in a newly Norman world, while the Blue Diamonds have managed a move across country from each other, likely as part of the Frankish world pushing into Slavic Poland through the various Polish/German wars.

I think this also fits fairly well with the dates and SNPs we have. Really only by defining more layers with our testers will we be able to sort this early stuff out any further. There is still some debate about when and whether and what size the Germanic migration to Britain was. Poland's history with the Rus and with Scandinavians in general is...well harder for me to find easily with some google searches. Pulling this story out from the data will continue to be fun for some time to come.

In England

I have more layers to work with, also a pretty good vested interest in the outcomes here so I've saved it to talk about separately. 

Whether the product of viking incursions or Germanic migrations. I always end up here in southeast England.

Of course, in this post I'm assuming the age ranges and estimates are on the money. We could find out some surprises in the future that would revise our speculation, but for now I'll move along like we're on the right trail.

I've been talking about the Knowlton family for a long long time, but the Luncefords are new for me. They are a happy surprise of big Y. Nicely the Lunceford's give us a nice pin in space and time. Their proximity to the Knowlton and Elmer traditional homelands makes them much easier to explain. Although we're waiting for more layers from other families to come in, here we have several hundred years of habitation in England.

There are a lot of Lunsford related place names in Southeast England, I don't know where the original ford (probably over a river) was that was associated with Lun or Lund or "a" lund. It's one of those transient place names. It ties your surname to a place, but maybe the place is just unknown now.

So without really knowing exactly where to put this pin, here is Lunsfords Cross in the general neighborhood I'm thinking of:

Why don't I know where to put the pin? The Lunsfords in the U.S. are a lot like the Knowltons or the Elmers. The records tying them to England are not rock solid. Most people place them in England, but it's the "where" that gets lost in their moves to the colonial Americas.

I think you could fairly safely assign us to Southeast England at some time in proximity to 845 AD. Either arriving with or defending against the Danes.

We should be clear that it's unlikely this common ancestor would have a surname at all. As far as I know that convention was brought to us through the Normans.

Knowlton-Elmer 1030 AD

The idea has been that the Knowltons are our closest Y cousins for various reasons, mostly based on their STRs.  Big Y testing and age analysis, point to that being the truth. At some point there is a "Knowlton-Elmer" man who is born around 1030 AD. Most likely, somewhere in Southeast England in pre-Norman times.

The Knowltons have ties to Kent, the Elmers to Norfolk and Suffolk, but I wonder if they first lived in Dorset. Knowlton is a place name and there is an abandoned village in Dorset named Knowlton. 

The Knowltons have some speculation of their own about Dorset. So I'm not breaking any new ground here. Their genealogy places them in the Americas as the descendants of either of two sons of one man, but their Y DNA shows that each branch was not related to the other within any meaningful time frame. They are in totally different haplogroups. So they have a mystery in the colonies.

One goal in big Y testing was to see if the Knowltons were really just a lost branch of the Elmer family that was adopted to the Knowltons. We're still waiting on definitive results from a final Elmer tester to try and sort that out. In the meantime we have this age estimate that puts the common ancestor back in England.

So Knowlton in Dorset is on their mind, but why is it on my mind?

It's because of this Domesday book entry: http://opendomesday.org/place/SU0210/knowlton/

Aelmer (or Athelmaer) is the previous lord in Knowlton in 1066 before it's passed to Ansgar the Breton.

Image blatantly lifted from the opendomesday site. 

Knowlton is mentioned again in a second entry with a nearby town changing hands from King Edward the Confessor to King William the Conqueror. 

Again, Aelmer here is not a surname. It's a given name. Knowlton is not a surname, it's a place name. It is nicely one time the two names come together around the time of  1030 AD. 

I'm not suggesting that the Knowltons and Elmers are directly related to this Aelmer, but that they might be associated with him by others. That they might be assigned a surname later based on this association or by an association with the town they lived in. One group becoming Knowltons and one becoming Aylmers based on the convenience of the people taking the tally. At this point, where the names might diverge, it could be just that arbitrary. 

The shared SNPs with the Knowltons will be very important as we try to bridge the gap to England. We may find Knowltons still in England that are more closely related to the Elmers or we may find Elmers that are more closely related to the Knowltons. We're examining only two branches of a single family at this point. They may have lived with their newly minted surnames, knocking about England for these hundreds of years before one person from each group moved to the new world.  

Maybe that is the best example of a reserve of DNA. We can speculate that the Lunsford-Knowlton-Elmers lived in England in the 800s. About 300 years later in the 1100s AD, the men related to that man are still there. Then some 600 years later, at least three men make their way to the Americas. Their DNA can only hint at the 800 or 900 years of shared history. 

With each of these DNA strands, we're not seeing the whole story. We're not even seeing the whole story of the Y. Just bits and pieces. That's why this is speculation. I think our resolution is getting tighter though.

Elmer 1614 AD

This age estimate comes without trees. We have one Elmer pillar related to Edward 2 and one Elmer whose tree is stuck. The results we'll see this fall, should help sort both the Elmers and the Knowltons in some definitive way. 

Without other big Y Knowltons to compare to, the Elmers are representing a nearer history on their own. 

The man we expect all these Elmers to be related to is Edward Elmer born around 1610..ish who was a founder of Hartford. It was very interesting to get 1614 as a date since that is right near the estimated time of Edward's birth. It's also nice that this estimate is independent of any research we've done. The group doing the estimates was not specifically made aware of the expected relationship between these men. 

Now they are of course, so when our big Y results from the line of Samuel come in we may see an adjustment in this estimate. For today, the estimate for a most recent common ancestor is 1614 AD.

This age estimate represents the Elmer history in the Americas. 

No one is really sure where he's from in England. He's linked to Bishop Aylmer in London. He's linked to the Sheriff of Norfolk and Suffolk. He's sometimes connected to fish mongers in London. At some point he is a member of a church in Essex and leaves from there to Cambridge Mass. 

I think we're at the point of searching British census records and looking for living Elmers, Elmores and Aylmers to test. Maybe a starting point would be in Dorset. Maybe in Salisbury about 19 miles from Knowlton. 

Here is an 1881 surname map for Elmer in England. Salisbury (teal..ish in south central England) is not as popular as points further east, but it is much closer to the cluster of 1881 Knowltons (second image, just to the south east). Images from Public Profiler. 

Elmers in 1881

Knowltons in 1881

Food for further speculation and maybe some recruiting. Back to waiting.