Archiver > GENEALOGY-DNA > 2010-02 > 1265771508

From: "Alister John Marsh" <>
Subject: Re: [DNA] TMRCA Estimate of a group of Marsh families>>
Date: Wed, 10 Feb 2010 16:11:48 +1300
References: <><>
In-Reply-To: <>


Many thanks for your analysis of my example. Perhaps I understand your
processes better through you working through this example.

Your assessment of there being two clusters is correct.

The age you estimate for the larger cluster (A to I) at 400 years ago is
possibly in the right ball park, although one or two of the haplotypes may
be from branches earlier. If anything, the cluster A to I may be a little
older than 400 years.

This smaller group of J and K is only two haplotypes, I suspect your
estimate of 100 years to branch common ancestor is considerably out.
However, because the haplotypes were the same, and only two of them, I think
there is not a large enough group to make predictions with statistical
confidence, which you in effect stated. You can't be criticized for
predictions on such limited information as 2 haplotypes.

The total group of A to K you estimate an age of 825 plus or minus 100
years, which would be between the years 1085 to 1285. If I was put on the
spot, and asked to make a single number prediction for the year of MRCA for
the total group, I would likely pick a number right in the middle of this
time slot. I however have been open to a wider range being "possible", and
I look on the extreme limits of probability as being between the years 500
to 1500. My probability range estimate is based on a mixture of facts,
speculation, and gut feeling. Also I suspect a number of back mutations and
parallel mutations which might be confusing your mutations count.

Following I will attempt to include the tree structure which I currently
speculate for these haplotypes... I hope the email programme does not mess
up the tree...

!-TYNDALL-John1637-------K USA
! Devon
!—TYNDALL?--!-TINDALE---Isaac1780--—- USA
! Northumberland??? USA
! perhaps before 1100????
-Common---!-------!-TEAGUE-----------------------------J USA
Ancestor ! Unknown origin, Ireland Speculated,
approx ! but de Tey/ de Tiege origin Essex 1100 pos???
1000? !
years ago!
! !--------James1812?------------------D Essex
! ! Essex
! ?
!-MARSH-!--!-----John1742?-------------------I Suffolk
Before ? Suffolk
1460?? !
perhaps !---!-Robert1707--------!---------B Australia
before ! West Wratting (WW)!---------C Sussex
1100??? ? !--------- Australia
!-John1787--!-William1806-!---A New Zealand
! (WW) ! !--- New Zealand
! !
? !-Thomas1814------ Australia
!-John1629--!-John1732-!------F Suffolk
! (WW) ! Linton !
! ! !-!-?--G Tasmania
! ! !
! ? !-?-— Australia
? !
! !---Charles1829---E USA
! Burwell
!-James1791-------------------H Cambridgeshire
Cambridge City

The haplotypes include Marsh families, a Teague family, and Tyndall/ Tindale
families. I have concluded that these families are related at some point by
a common ancestor who likely had the following marker scores which are off
R1b modal, and some are quite uncommon...

DYS439 =11
DYS458 =16
DYS456 =15
DYS607 =14
CDYa =35
DYS446 =12
DYS640 =12
DYS452 =29 (10)
DYS463 =25 (23)

The group is R1b1b2a1b, or R P312*.

On the tree I have indicated the name and date of birth of known ancestors
of different branches, and given ancestral branch locations, as well as
current locations for the families.

All of the Marshes so far have DYS511=9, which is a very uncommon marker
score in R1b, and seems to be a distinctive marker in the Marsh lines.
Teague is DYS511=10, Tindale/Tyndalls are unknown on that marker, but I
expect them to be 10.

Now, getting to your estimates Anatole...

MARSH FAMILY: I am reasonably confident that "most" of the Marshes (A to I),
particularly the ones tracing to near West Wratting in Cambridgeshire,
descend from a John Marsh born about 1560 at West Wratting, Cambridgeshire.
He had 2 sons who may have contributed to different branches. I can't
discount the possibility yet that all descend from this person, so to that
extent, your 400 year prediction is not too far out, and not able to be
disproved at this time.

It is in my view probable that the West Wratting Marshes further back
descend from a John "March" perhaps born about 1460, who lived at West
Wratting. There was a Mareys/ Marsh family who owned land at West Wratting
up until about 1360 at least. There was a Simon de Merc/ Marsh born about
1100 who owned land at "Wratting", and he would have to at least be a
"possible" ancestor of the West Wratting Marsh clan. Although it is possible
that the surname may have passed via a daughter heiress at an early
generation, so the current Marshes may not be "direct male line" descendants
of the de Mercs. A DNA clue suggests it is at least possible that the Y-DNA
line may not be continuous back to 1100, but that is a matter to resolve
which is "work in progress".

Haplotype I's line first appears in Essex near London in 1812, and has
DYS449=29, matching Teagues and Tyndalls (all of the other Marshes are 28),
and I am open to the possibility that I is from an early branch of Marshes,
and that perhaps DYS449=29 was ancestral, and the Cambridgeshire/ Suffolk
Marsh branches had a branch ancestor who mutated to 28.

The Simon de Merc born about 1100 who owned land at Wratting, is suspected
of coming from an Essex de Merc/ de Merk/ Marsh land owning family. A land
owning Marsh family which lived in about 1300 near where I's family turns
up, may be another branch of the early Essex de Merc/ de Merk family, which
was a Norman de Marcq family originally which came to England in 1066.

If Simon de Merc was the ancestor of A to H Marshes which spread out around
West Wratting, then it is possibly a Kinsman of Simon de Merc who was the
ancestor of I. That would make the common ancestor of all the Marshes born
about 1100 or before. However, it can't be discounted that "I" many descend
from a West Wratting Marsh who went to London in the 1700s.

TEAGUE FAMILY: A prominent early Essex Marsh family had it's main estate at
"Marks Tey". "Mark" derives from the surname de Merc/ de Merk/ de Marcq.
"Tey" derives from an early word for an enclosure, which originally was
called Tiege or Tege. Several de Merks living at Marks Tey appear to have
used the surname "de Merk de Tey", and in France/ Belgium where the Essex de
Merk/ de Marcq family originated prior to 1066, and seemed to keep estates
for a while, there was a family called "de Marcq de Tiege".

Not only that, but when Andrew de Merk inherited the Marks Tey Estate about
1280, at an advanced age, he changed his name to de Tey. It is unclear if
some of his sons also changed their name to de Tey. But Andrew's son who
inherited the Marks Tey estate, had an only daughter heiress, who by
marriage passed the estate to another "de Tey" family, reputed to be
descended from a Simon de Tey born at Marks Tey about 1100. Could Simon de
Merc/ de Merk born about 1100 who owned land at Wratting possibly be from
the Marks Tey estate, and could he be the same person as Simon de Tey born
about 1100 at Marks Tey?

It is tempting to wonder if the Teague who matches the Marshes Y-DNA could
have descended from a person surnamed de Tey/ de Tege/ de Tiege, somehow
connected to the Marks Tey estate in Essex, where de Merk and de Tey
families were confusingly intertwined from about 1100 to 1360 in particular.
This is clearly just a speculated possibility, but if it were true, the
Marshes and Teagues might share a common ancestor some time between about
1050 and 1300. Your speculated TMRCA Anatole was 1085 to 1285, so your
prediction would not discount that possibility.

TYNDALL/ TINDALE FAMILIES: Haplotype K traces to a Rev John Marsh born about
1637 from Devonshire England. Traditionally, most Tyndall families are
thought to descend from a de Tyndall (Tyne-dale) family in Northumberland in
about 1100. Although the family has a Norman sounding name, the Tyndall
family descended on female lines from Scottish Royal lines, and the original
Tyndall Y-DNA is thought to possibly descend from Northumbrian Anglo-Saxon
Royal lines.

Through marriages of prominent Tyndall heiresses, there were at least 2
other Northumberland Y-DNA lines who adopted the surname Tyndall. But the
"original" Tyndall Y-DNA line sent a branch to Northamptonshire and Norfolk
in 1220, which was prominent in those areas for many hundreds of years. The
estates of this Tyndall family in that area were near to where some branches
of the de Merk family had estates.

The Tyndall who translated the Bible, was perhaps broadly in the same era as
the Rev John Tyndall born 1637. The Bible translator is said to have been
from the Northampton/ Norfolk de Tyndall Family, which traces back to the
Northumberland de Tyndalls. It is not inconceivable that Rev John Tyndall
was not native to Devon, but from Northamptonshire, or perhaps directly from
Northumberland. Religious people in that time move about a bit, sometimes
out of necessity to keep their heads on.

If Tyndalls and Marshes have very similar Y-DNA, what opportunities would
have presented for a connection? The only Tyndall family known in the same
back yard as the Marsh/ de Merc family, was the de Tyndall family
originating in Northumberland which moved to near the de Merc family in

Possibilities are that after 1200 a Tyndall married a Marsh heiress, and
changed his name to Marsh, or a Marsh married a Tyndall heiress, and changed
his name to Tyndall. If that occurred, my guess would be that it occurred
between 1220 and about 1460 if a Tyndall changed his name to Marsh, or
between about 1220 and 1600 if a Marsh changed his name to Tyndall.

However, I wonder if the Marsh Tyndall link may have been back as far as
1500 years ago, in some common Anglo-Saxon or Saxon link. The Marshes came
from an area in East Anglia which was settled by Anglo-Saxons 1500 years
ago. But even if the Marshes trace back earlier to the de Merks/ de Marcqs
from the Calais/ Flanders area, that area was thought to have been settled
to a large degree by Saxons at earlier times. So a "Saxon" origin could not
be discounted at a very early date. If the link is not after 1220, then
perhaps we need to think in terms of about the year 500 AD for the link?

could establish if the Tyndall Y-DNA which matches me comes from the de
Tyndalls of Northumberland in about 1100. If anyone knows Tyndalls which
might be from that line, please encourage them to contact me at

It would also help if I could find out what the de Tey family from marks Tey
had for a Y-DNA profile. The de Tey/ de Tay/ de Tea family from Essex is a
bit rare these days, but I would welcome any contacts which might link me to


You estimate Marshes A to I having a common ancestor about 1610. I estimate
"most" to have a common ancestor about 1560, but for the moment the
possibility is that Marsh D has a common ancestor with the rest of the
Marshes possibly as far back as 1100.

The Tyndall/ Teague common ancestor you estimated within 100 years. It
seems definitely not within the past 200 years, and if my speculations were
correct, not within the past 700 years. If anecdotal evidence from the
families concerned were correct, the connection is not in the past 400

Your estimate for the Marsh/ Teague/ Tyndall common ancestor between 1085 to
1285 is quite plausible. I however feel that I can't yet exclude the
possibility that it is somewhere in the range from the year 500 to the year

Many thanks to you for spending your time to look at this for me.


-----Original Message-----
[mailto:] On Behalf Of

Sent: Wednesday, February 10, 2010 1:54 AM
Subject: Re: [DNA] Variance Assessment of R:U106 DYS425Null Cluster

>From: "Alister John Marsh" < >
>I am not a mathematician, so you may be able to correct me, but would not
the history of this group have a bearing on the expected number of
mutations?  What I mean is, suppose B, D, E, F, G, H were brothers (who have

identical haplotypes), and the other two A and C of unknown relationship to
the 5 brothers?  Would that in effect mean that comparing haplotypes A, B,
C, would be a more meaningful way to assess age since common ancestor using
your formula for this group?

Dear John,

You cannot group haplotypes without a certain and justified reason.
"Suppose" and "what if" do not work here. When you toss a coin you cannot
say - "what if a coin always lands on a tail, and heads are just
aberrations? So we strike them out".

The criteria in haplotype handling which I have described earlier include
haplotypes grouping, dissection of them onto branches, however, based on
certain repetitive features in haplotypes, not single (random) mutations as
in haplotypes A, B and C which you have mentioned. A haplotype tree dissects
datasets into branches, however, sometime it can be done just visually (it
is not preferred, but sometimes possible, in rather obvious cases).     

Having said this, I will move to your specific example.

>I have an example for you to consider.  Can you make anything of the 11
haplotypes of 37 markers below? (Haplotypes are shown below - AK)

>I don't know when the common ancestor was for this group, but I know dates
it must have been before, and can speculate dates it was probably before.

>I don't know the ancestral haplotype. What do you think the ancestral
haplotype was? How far back do you estimate to the common ancestor using
your system, and what confidence interval range would you give for that
estimate?  I have been puzzling over this group for years, and it would be
interesting to see how your estimates compare to my current thinking on the

Of course I "can make anything" on that series. Thank you for it. The series
is a bit more complicated compared to those I have shown earlier. It
represents two subsets, or branches, or "mini-lineages" meaning that the
principal lineage in the same, however, there was an offshot recently, just
a few generations before present, and it created a fork. Hence, the dataset
shows two "local lineages", each one with its common ancestor, which
deviates from each other by four mutations in the 37 marker haplotypes.

It can be pictured as an iceberg with two tips, one small and one taller,
however, deep underwater, it has its "common ancestor" of the both tips.
That is why there is no a clearly visible base (ancestral) haplotype in your
series. It is a superposition of the two series.

The age of one "tip of the iceberg" is about 400 years. It is of a larger
subset, haplotypes "A" through "I" in your designation.  The age of the
smaller subset, and only two haplotypes from it are shown in your series, is
"formally" 50+/-50 years, and I tend to place it at 100 years before
present. It is probably some older, when more haplotypes are included.
Judging (quantitatively) from 4 mutations between them, their common
ancestor lived about 825 years ago, in the 12th century, give or take a

Well, your call?      

Anatole Klyosov



For this group, I have many more haplotypes, but only the above complete to
37 markers.  The criteria I used for selecting this group, was that they
were all I had which were complete to 37 markers, so I have not tried to
make some tricky selection to mislead you.  Some of the above have been
tested to 130+ markers, and a number to 67 or 76 or more markers.

I have some speculative ideas about when the common ancestor might have
based in part on genealogical data which might imply opportunities for a
common ancestor to have lived.

I have an advantage over you in that I do have more marker
information, and some paper trail and context information.  But I would be
interesting to see how you evaluate this group based on the limited
information I have given you above.

A 13 24 14 11 12 14 12 12 11 13 13 29 16 9 10 11 11 25 15 19 28 14 16 17 17
 11 11 19 23 15 14 18 17 35 38 12
B 13 24 14 10 12 14 12 12 11 13 13 29 16 9 10 11 11 25 15 19 28 14 16 16 17
 11 11 19 23 15 14 18 17 35 39 12
C 13 24 14 11 12 14 12 12 11 13 13 29 16 9 10 11 11 25 15 19 28 14 16 16 17
 11 11 19 23 15 14 18 17 35 39 12
D 13 24 14 11 12 14 12 12 11 13 13 29 16 9 10 11 11 25 15 19 29 14 16 16 17
 11 11 19 23 15 14 14 17 35 39 12
E 13 24 14 11 12 14 12 12 11 13 13 29 16 9 10 11 11 25 15 19 28 14 17 17 17
 11 11 19 23 15 14 19 17 35 39 12
F 13 24 14 11 12 14 12 12 11 13 13 29 16 9 10 11 11 25 15 19 28 14 16 17 17
 11 11 19 23 15 14 19 17 35 39 12
G 13 24 14 11 12 14 12 12 11 13 13 29 16 9 10 11 11 25 15 19 28 14 16 17 17
 11 11 19 23 15 14 18 17 34 39 12
H 13 24 14 11 12 14 12 12 11 13 13 29 17 9 10 11 11 25 15 19 28 14 16 17 17
 11 11 19 23 15 14 18 17 35 39 12
I 13 24 14 11 12 14 12 12 11 13 13 29 16 9 10 11 11 25 15 19 28 14 16 17 17
 11 11 19 23 15 14 18 17 36 39 12
J 13 24 14 11 12 14 12 12 11 13 13 29 16 9 10 11 11 25 15 19 29 14 15 17 17
 11 11 19 23 15 14 18 17 35 37 12
K 13 24 14 11 12 14 12 12 11 13 13 29 16 9 10 11 11 26 15 19 29 14 15 17 17
 11 11 19 23 15 14 18 17 35 37 12

To unsubscribe from the list, please send an email to
with the word 'unsubscribe' without the
quotes in the subject and the body of the message

This thread: