Catharina van Malabar, SM/PROG - DNA: U2c or N21?

This follows from a discussion here: http://www.geni.com/discussions/152268?msg=1063884

And pertains to a Discussion about "Catharina van Paliacatta, SM/PROG - Confirmation by genetics?" http://www.geni.com/discussions/142201?msg=1063879

Beth Hawkins hasn't been on Geni for a year, and her profiles are being managed by her sister, Deborah Cecilia Hawkins

I can't see Charlette Louise Hoppe's DNA results on her profile. Am I being dumb? Do we have them somewhere?

I fail to see DNA for Charlette Louise Hoppe

This is Beth Hawkins' matriline as we have it on Geni at the mo:
Private >
Mary Cecilia Williams>
Elizabeth Anne Williams>
Anna Cecilia Williams>
Johanna Catharina Elizabeth Booyens, d10e20>
Catharina Elisabeth Oberholzer>
Geertruy Jacomina Viljoen>
Elizabeth Catharina van der Westhuizen>
Anna Elizabeth van Eck>
Elizabeth Bronkhorst>
Geertruyd Boshouwer, SM>
Adriaantje Gabrielsz, SM>
Catharina (Catrijn) van Malabar, SM/PROG

The immediate problem with Charlette's motherline is that the profile Terence Willie Fitzgerald Visser (aka Raldy / Joyce Phyllis Barnard) has a birth certificate that suggests she's male.
Hmmm - help?

Charlette Louise Hoppe matriline as we have it on Geni at the moment:
Beryl Adele Barnard>
Terence Willie Fitzgerald Visser (aka Raldy / Joyce Phyllis Barnard)>
Anna Johanna Visser (Fitzgerald)>
Maria Catharina Fitzgerald, e9f1>
Geertruida Susanna Grobler, d12e8>
Maria Catharina Klopper>
Maria Catharina Venter, b3c4d2e1>
Anna Catharina Venter>
Katharina Bezuidenhout>
Maria Catrina Bezuidenhout>
Catharina Pyl>
Cornelia Bastijaans aka Pijl, Pyl, Botha, SM>
Catharina (Catrijn) van Malabar, SM/PROG

Thank you Jan et al. This is stil way over my head so I would rather leave this up to the gurus.☺

Summary of points on the discussion that initially took place on Francoise's profile:

Private User: There are a few scenarios that may account for these mtDNA results. One is that somewhere along the maternal line, there was an adoption. Say a woman dies in childbirth, leaving her husband with a newborn along with several other children. The husband can`t care for the infant and work to support his other children, so he gives the newborn girl to someone else to raise. (Maybe a brother? a sister? a neighbor?) That newborn girl would then have a paper trail with her adoptive family and not her biological family, and that paper trail would look legit.

Roland Henry Baker, III: N is a huge group - one of the two main branches of L3 from Africa. L21 is generally from South East Asia.

I know very little about South African genealogy and history. What is the history of Cham, Malaysia and Indonesia immigrants to South Africa? Possibly also from Cambodia, etc. What about these peoples and their relationships to the mother country?

Would this admixture be expected in the ancestry of this immigrant?

If we understand these questions we might have some clues about how a female from South East Asia got into the mix.

If we get the full list of markets we can look it up on Ian Logan's site:

http://www.ianlogan.co.uk/mtDNA.htm

Haplogroup N21 - In ethnic Malays from Malaysia and Indonesia Hill, C.; Soares, P.; Mormina, M.; MacAulay, V.; Meehan, W.; Blackburn, J.; Clarke, D.; Raja, J. M.; Ismail, P.; Bulbeck, D.; Oppenheimer, S.; Richards, M. (2006). "Phylogeography and Ethnogenesis of Aboriginal Southeast Asians". Molecular Biology and Evolution 23 (12): 2480–91. doi:10.1093/molbev/msl124. PMID 16982817

Tracing the Austronesian Footprint in Mainland Southeast Asia: A Perspective from Mitochondrial DNA 2010
http://mbe.oxfordjournals.org/content/27/10/2417.full

A Mitochondrial Stratigraphy for Island Southeast Asia
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1876738/

...
Good question Jan! As stated I am ignorant about the genealogy and history of South Africa. So I was surprised when you stated that one of the descendants tested was from N21 which derives from Southeast Asia - see above.

We need to see if this is a full sequence or a marker test or if it was just a "predicted" haplogroup based on a couple of markers in a panel like 23andMe does or NatGeo does. Because if it is just predicted the haplogroup call could be way off. And that's going to really through off the investigation. If we have a female from Southeast Asia as progenitor that could offer clues. But as I said N is a super-group which spans the whole globe and if the call is wrong we could be looking in the wrong places.

If they tested at FTDNA, the type of test and all the markers and name of the reference used would be a good start and I believe this is on the downloadable PDF certificate they give you. So it can be cut and pasted.

The full sequence is always nice but you would want to email that off to Mr Logan and that will take at least two weeks to process. Still it could be very interesting especially for the N21 sample.

...
Above I meant to type N21 is from Souteast Asia not L21. Very interesting!

...
N is also the mother haplogroup of U:

http://www.phylotree.org/tree/main.htm

...
On the other hand we have U2c which is common in India. Does the progenitor have a historical link to India?

...
So yes, we really want to know if these are based on a full sequence or a partial sequence or just a couple of markers. If both of these are based on a full sequence that we can start by looking that these geographical origins and compare this to the historical origins of the progenitor. Whichever matches the progenitors origins would probably satisfy the requirement of a null hypothesis. In other words if the progenitor is from Indonesia then N21 is probably the correct call. If the progenitor is from India then U2c is probably the correct call. If the ancestor is from Europe - that we are in deep water over our head :) But it would give us a null hypothesis to start with.

Erica Howton
Catharina van Malabar, SM/PROG. Birth: circa 1637 Malabar or Coromandel in India

Private User: Hmmm another piece of South African puzzle, is that a person that stepped on land here (progenitor), was attributed to the land from which he or she left port from, not land of birth. So we may have to assign probabilities there too...

Roland Henry Baker, III:
Thanks for the link!

I wouldn't put any probability on it.

What I mean by null hypothesis is we simply have no idea. So we make the best guess based on the fact that the boat came from India. We create a hypothesis that she was probably U2c. Then we try to kick the daylights out of that hypothesis and see if we can disprove it. It doesn't mean we believe the hypothesis or we don't believe it. Nor do we give it a probability. It just means we have something to test.

But before we do that do we actually have a source for that? Do we know for certain 100% that she came on a boat from India? Or did the boat from from nowhere? Because I know I have a lot of ancestors who "came on a boat from Devon, England" and when I go to actually find a source for it I find there is nothing... so it would be a good time right now to see if there is an actual piece of paper like a ship's passenger manifest that records that she was a passenger, etc. Otherwise we're just making stuff up. And as Ericka knows I like to delete data that's made up.

And we need to know if these are both full sequences and not just marker tests. Because U is a branch off of N. And if N or U is just a guess we could actually be looking at the same haplotype - no joke. I doubt that's the case but we don't want to spin our wheels if these data are not full sequences. If they are not full sequences now might be the time to upgrade them because things just got interesting! With FTDNA you don't need to retest or re-sample - you just pay and upgrade price which is a deal.

How much fun is this?

wonderful news! My brother is Gerrit Nieuwoudt du Plessis, so now I also know exactly where the Gerrit also comes from. Thank you for the good work. where do you do your DNA testing?
Blessings and Happy New year, Marietjie

Birth certificate shows her name as Terence Willie Fitzgerald Visser. Nicknamed Raldy by her family. Married c 1928. On her daughter Beryl Adele Barnard's birth certificate the name appears as Terence William Fitzgerald - maiden name Visser. She changed her first names by deed poll in 1931/32 to Joyce Phyllis - death notice TAB MHG 3126/49 dated 24 June 1949.

Yes, as said by Peter - The father made the mistakes on purpose (names and gender), but it was a she. All the dates check out and is sourced as well.

Jan - what is the story behind the father making the mistake on purpose?

I do not know.

Maria Magdalena Pienaar (du Plessis) the best place to do mtDNA testing is FTDNA. here is there product page:

https://www.familytreedna.com/products.aspx

A mtDNA marker test starts at $69 and a full sequence is $199. If you take the cheaper test they will keep your sample and you can just pay the different to upgrade to a full sequence later. They ship internationally. I suggest the full sequence if you can afford it. They have sales a few times a year like on America's mother day and father's day and you can buy a full sequence for $169. Sales are often posted on Facebook DNA groups.

Sharon thanks for posting the summery to get this discussion started.

We need to get the exact information on which test these two descendants took. Was it a full sequence from FTDNA? Or was it a marker test from FTDNA? Or was it just part of a NatGeo or 23andMe general genome test? We should also get a full list of their markers versus the RSRS and Cambridge standards. If they are on FTDNA that information will be posted with their results. They could just take a screen shot.

We also need someone how is familiar with the genealogy of this person to see if there is an actual passenger manifest showing where Catharina van Malabar came from. N21 is from Malaysia and Indonesia and U2c is from India.

The rest of the steps involved from a genetic point of view will be to narrow the range of uncertainly on each lineage by confirming each lineage and squeezing the number of generations in question.

Then we might want to see if any other 4th or 5th cousins who are matrilineal descendants want to take an mtDNA test who are on the same two pedigrees above. If we can work our way down each pedigree a few generations we should be able either confirm the results are the same or show that there is a discrepancy and limit that discrepancy to a finite number of generations on one pedigree.

If both women have done autosomal testing it would be helpful to see if they have matches to their 4th cousins via the lineal descent above. If they do have 4th cousin matches and better yet if they can triangulate them then that would support their lineage at least that far back and narrow the number of questionable generations as well.

This is basically a "process of elimination."

Roland - I dont think your suggested approach of autosomal testing would work in this case (and likely in any RSA case) - these two individuals are not only 'paper trail' related on the maternal lines, but also through more than one other line and much closer than the maternal line...

Hmmm... I think the 'check' relationships is the best, have done the Malaysian line, but need someone to provide sources for the Indian one - there are a few that are not sourced.... maybe aim should be to be 'reluctant' in using DNA if we don't have sourced profiles for the entire line?

Here are the respective 'closest' lines:

Catharina van Malabar, SM/PROG 
→
Cornelia Pyl (Claasen), b3 SM 
her daughter
→
Catharina Botha/Brits (Pyl), b3 
her daughter
→
Magdalena Gouws (Brits), b7c3 
her daughter
→
Anna Gouws, b3c4d3 
her daughter
→
Marthinus Gerhardus Booyens, b6c2d9 
her son
→
Marthinus Gerhardus Johannes Gert Johannes Booysen 
his son
→
Anna Cecilia Williams (Booysen) 
his daughter
→
Elizabeth Ann Williams 
her daughter
→
Mary Cecilia Williams 
her daughter
→
<private> Hawkins (Sellars) 
her daughter
→
Elizabeth Anne ten Have (Hawkins) 
her daughter

AND

Catharina van Malabar, SM/PROG 
→
Cornelia Pyl (Claasen), b3 SM 
her daughter
→
Catharina Botha/Brits (Pyl), b3 
her daughter
→
Martha Elizabeth Botha, b2c8 
her daughter
→
Jan Andries Swanepoel, b3c3 
her son
→
Pieter Andries Swanepoel, b3c3d2 
his son
→
Jan Andries Swanepoel, b3c3d2e3 
his son
→
Roelof Johannes Willem Swanepoel, b3c3d2e3f12 
his son
→
Maryna Maria Francina Petronella Barnard (Swanepoel) 
his daughter
→
Marthinus Christophil Barnard 
her son
→
Beryl Adele Barnard 
his daughter
→
Charlette Louise Hoppe 
her daughter

I am still wondering why Noelle did not answer the question of why it might be important that the lines are clear?

Thanks for laying that out, Roland.

Catharina van Malabar would have been brought to SA as a slave. Both Malaysia and India are options, and her birthplace, as opposed to the place she was sold on from, could be different, and typically just not recorded.
I've been mooting a DNA mapping of the East African slavetrade for a while now.
I wonder if that's on the cards anywhere?

Jan, before we proceed I think we should wait for input about why Terence Willie Fitzgerald Visser (aka Raldy / Joyce Phyllis Barnard)'s father registered them as a male, and why the birth certificate was never changed if that was wrong.

Having multiple shared common ancestors due to cousins interbreeding on a regular basis is called endogamy and it's not a problem. You just have to use the proper technique with autosomal testing. We have the same issues with Cajun and Jewish populations in the united states. The idea is you prove which Nth grandparent the person inherited a DNA segment from that is shared with each Mth cousin. You start by testing closer cousins to prove closer Nth grandparents and then you work your way to more distant cousins to prove the relationship to more distant grandparents using those exact same segments. Phasing the results with the test of one parent is ideal but not ncessary. You can use and aunt, half-sibling or first cousin to narrow down which segments are from which parents. Then you can use 2nd cousins to sort those same segments to each grandparents, third cousins to sort to each great grandparent. So you eventually work your way up the tree. And if we find surprises than we really learn something. For example if we find a 2nd cousin has no match than we learn we have a problem with the genealogy. So we pay attention to surprises. This takes work but it is done successfully in American Jewish and Cajun populations all the time.

If there is no source for the origin of this person we have to just ignore that as a factor.

We still need to know which tests we taken.

Just want to check then... using that deck of cards example... we start with one unique deck and shuffle it with other decks for a few times. And then somewhere in between we shuffle half of the unique deck into the remainder half of the same initial unique deck going through another route.

We compare the resulting deck with another deck that also have been duplicated to the same extent (lets say 50% per shuffle), but in a different way. From expectations theory we actually expect them to be more related just due to 2 or more duplications, than what the decks should be if they were unrelated.

Don't we, when this happens, 'autosomal DNA match' close relations when they are in fact very far?

Agree that we need the full test for absolute certainty that it is not a predicted value, would it be OK if we post both sets in this discussion or where should it go?

Sharon, I will also try and find more info on the male birth record. But if you look further, you would see that the same person was listed as the mother and subsequently changed names.

In any case, seems to support the argument for more samples... still wonder if someone would source the U2c line and confirm it on this discussion.

I have been told that Catharina van Malabar came from Batavia, today known as Jakarta, and that it is in Indonesia.....

Anna Johanna Visser *Fitzgerald filled in and signed her husband's death notice - Johannes Petrus Visser. Only the 4 male children were included on this document. There were no girls included! So the mother seems to have been agreeable to the idea of relegating the girls. I have had contact with the descendants of one of the boys in the family and I mention some of the names of the girls that have come to light. I know of a descendant, also on Geni, of the daughter Hester Visser; there was a sister Olga mentioned in court papers at time of father's death - both sisters to Terence Willie Fitzgerald *Visser. Latter named for her grandfather. The Fitzgeralds all seem to know of the little girl with the boy's name who was nicknamed Raldy. There was also a boy William Terence Visser on the death notice and his daughter is also on Geni. I can only try and get them to do the test??

OK - The results are also out - in fact the test was the full mtDNA of FTDNA. N21:

Extra Mutations
C146T C152T T236C 315.1C 522.1A 522.2C C3340T T4592C A7804G G8839A G15043A G15217A A16182G C16189T

Missing Mutations
C195T

Link to N21 DNA

http://www.geni.com/documents/view?doc_id=6000000039389894821&amp;

Perfect - that's really helpful Jan!

In future though if people could get both the RSRS and rCRS tabs that would be even better. Because there is no quick way to compare the two. But this is good enough.

I'm glad it is a full sequence! Excellent!

So we know for sure this is N21 from Indonesia with fits perfectly with the Batavia origin story.

So now we need the details of the U2c test and also someone needs to double check the pedigree of the person who tested U2c. It is too early to tell but there may be an error in their pedigree since U2c is from India.

Hi, I have persuaded a second cousin - our grandmothers were sisters, plus a second cousin once removed - our great grandmothers were sisters to do the full sequence mtDNA test. So will let you know when results are out and thereby prove that my Granny is a girl, despite her birth certificate stating she is male. Have lodged my N21 results with GenBank.

My second cousin, once removed, has received her mtDNA results - full sequence. She is N21. So now there are 2 of us. Any progress with U2c1 and her timeline?

For me, the following unsourced profile in the U2c1 line is not making sense (higher up in that line, everything is "working":

Anna Cecilia Williams