James Alfred Casey Family

Family of James Alfred Casey and Annette (Tucker) Casey, ca. 1899. First Row (left to right): James Alfred Casey, Bonnie Casey, Annette (Tucker) Casey, Linda Casey, Lucinda Casey, Columbus Casey. Second Row (left to right): Louis Casey, Ella Casey, Perry Casey, Maude Casey
 



Analysis - South Carolina cluster

Analysis of the Casey South Carolina cluster – Haplogroup L226

by Robert Brooks Casey
Version 3
Date: July 29, 2011

What is new from previous analysis:

This analysis only covers one cluster for the Casey DNA Project and is a new format that I have used in the Brooks DNA Project. This cluster has strong ties to South Carolina from 1760 to 1800. This review includes two new submissions – 166618 and 172505. It also includes a new Ancesty.com submission 9BCZF (John Kersey, b. 1616, Oxford, England). Submission 9BCZF is the most significant submission of this cluster and partially breaks the brick wall as it is the first submission in this cluster that resides outside the United States and will dramatically alter the MRCA of this cluster. Also, two 12 marker submissions were analyzed for the first time. The MRCA for this cluster has been changed for one marker (from 460 = 13 to 460 = 12). The NPE section has been radicially improved and two new tables have been added - the rarity of DNA markers table and the Y-Search NPE table. Also, expanded analysis of deep ancestry has radically assisted with the analysis of this cluster.

Feedback, additions and corrections are encouraged

I am sure I may have made some mistakes in this analysis. The content and organization of each iteration makes it hard to keep a consistent format and flow. I welcome any input on this analysis (both positive and negative). Be gentle and kind on the negative - but also be firm. I need more direct feedback on questions, concerns, corrections, etc. but have received very little feedback to date.

Step 1) - Select cluster to analyze

Select one of the primary haplogroups and then select one of the larger clusters within this haplogroup. According to the current DNA summary spreadsheet, the South Carolina Cluster is the largest cluster and is the most closely related cluster found in the Casey DNA Project. This cluster has tested positive for the L226 haplogroup and is one of two clusters that tested positive for L226. Deep ancestry L226 is believed to have originated around 1,300 years ago. This cluster also includes my DNA submission as well (77349). It also includes one submission that is a very high probability Casey NPE (the Hanvey line) and one surname variation (the Kersey line).

Step 2) - Verify cluster is a genealogical cluster

Verify that all submissions in this cluster are closely related Casey submissions. Determine all closely related submissions from entire Casey Surname Study that exceed 50 % probability at 12 generations using FTDNA MCRA utility (my speculative personal definition of a closely related cluster).

Found 17 submissions that were closely related with at least 32 or more markers (one has 111 markers, twelve of these submissions have 67 markers, two have 37 markers and two Ancestry.com submissions. Below are the probabilities of relatedness from the FTDNA MRCA calculator at 12 generations (genealogically significant time frame). These comparisons were done at 25 markers, 37 markers and 67 markers when available. Expressed in mutations from the MRCA (56130) and percentage of relatedness at 12 generations to the MRCA (56130). Submission 56130 currently matches the "majority rules" MRCA (but must be upgraded soon to 111 markers to remain the true MRCA). However, the Kersey submission will obviously replace 56130 as the MRCA in the near future. However, the Kersey Ancestry.com submission only has 29 markers and can not be used as the MRCA until this submission is upgraded to 67 or 111 markers. This Kersey submission clearly has the DNA fingerprint of the South Carolina cluster with 393 = 12, 458 = 16 and 449 = 30. The Kersey submission is also genetically older than all other submissions as markers 460 = 11 and 464b = 13 are closer to the L226 MRCA. Until the Kersey submission is upgraded and other Kersey submissions are available, the Kersey submisisons will be kept separate even though they definitely belong to the South Carolina Casey cluster. This Kersey submission is a major first step in breaking the brick wall for this cluster:

FTDNA
Number
Description of Line

Mutation/
% Related
at 25 markers

Mutation/
% Related
at 37 markers
Mutation/
% Related
at 67 markers
56130
John (SC)
MRCA
MRCA
MRCA
99381
John/Levi/William
2 / 51
3 / 79
3 / 91
93773
John/Levi/Francis
1 / 78
1 / 97
1 / 99
45068
James H. (SC)/Willis
1 / 79
1 / 98
1 / 99
119586
Moses (SC)/Moses
0 / 94
0 / 100
0 / 100
166618
James G. (TN)
0 / 94
1 / 97
NA
172505
Randolph/Levi
0 / 78
3 / 72
3 / 88
58301
Henson/Arvle
0 / 94
0 / 100
0 / 100
40325
Henson/Jackson
0 / 94
2 / 91
NA
51924
Abner/Pleasant/Elsberry
0 / 94
1 / 97
1 / 99
54166
Abner/Pleasant/Pleasant
0 / 94
1 / 97
1 / 99
107623
Abner/Turner
0 / 94
2 / 89
2 / 96
9BCZF
John Kersey, b. 1616, England
1 / NA
2 / NA
NA
WN8TA
Unknown Casey
1 / NA
2 / NA
NA
29956
Hanvey NPE
0 / 94
1 / 97
1 / 99
56874
Jesse/Anthony
0 / 94
2 / 89
2 / 96
77349
Ambler/John
0 / 94
2 / 89
2 / 96

There is little doubt that all submissions in the above table are very closely related. Due to the extremely high degree of relatedness, it is believed that the MRCA of this cluster is very recent and is estimated to be around 300 years ago. With the oldest proven ancestor being born in 1736 (Randolph Casey), there is currently very little gap between the oldest proven ancestor (born 1736) and oldest common genetic ancestor (born 1700). However, the John Kersey, b. 1616, England submission has now replaced Randolph Casey as the oldest proven ancestor of this cluster and it appears that all other South Carolina submissions are a genetic branch of the Kersey submission.

Below is a list of 12 marker submissions that have a possibility of belonging to the South Carolina cluster but have insufficient markers tested to verify with any certainty. These submissions require upgrades to more markers to definitely be assigned to any Casey cluster. Additionally, there is little or no traditional documentation to connect these lines to the South Carolina cluster:

FTDNA
Number
Description of Line

Mutation/
% Related
at 12 markers

Mutation/
% Related
at 25 markers
56130
John (SC)
MRCA
MRCA
N65778
Unknown Casey - Average odds
1 / 36
NA
129039
Unknown Casey - Low odds
2 / 11
NA

It is very doubtful that two additional 12 marker submissions not listed above are part of the South Carolina cluster. Both additional submissions have 393 = 13 which does not match the DNA fingerprint of the South Carolina cluster which always has 393 = 12.

The table below shows the next most closest related Casey MRCA to the South Carolina cluster. This table clearly shows that the Munster, Ireland MRCA is not closely related to South Carolina cluster in the genealogical time frame of 300 years (12 generations). The researchers of the South Carolina cluster should not expect the Munster, Ireland submissions to be closely related enough to warrant traditional research in connecting these lines. This table also indirectly eliminates the Elisha Casey (SC) line as a candidate that can be genealogically related to the South Carolina cluster:

FTDNA
Number
Description of Line

Mutation/
% Related
at 25 markers

Mutation/
% Related
at 37 markers
Mutation/
% Related
at 67 markers
56130
John (SC) - MRCA South Carolina
MRCA
MRCA
MRCA
53484
Dennis (Kerry) - MRCA Munster
4 / 11
7 / 8.0
9 / 8.0

Note 1: Obviously there are no other members of the Casey Surname Study that are closely related to this cluster in a genealogical time frame (300 years). This analysis confirms that the cluster defined as the South Carolina cluster is a valid cluster and that everyone in this cluster is probably related in a genealogically significant time frame (possibility of sharing a common ancestor around 300 years ago). The inclusion or exclusion of any 12 marker submission is very problematic.

Note 2: The probability of being related to the MRCA only varies from 88 % to 100 % at 67 markers. 67 markers should always be considered more statistically accurate due to a larger sample size (in this case – more markers). As you increase the sample size – the accuracy of these percentages increases.

Note 3: The admin usage of the FTDNA MRCA utility is limited to analysis of only submissions that have tested with FTDNA and have officially joined the FTDNA Casey Surname Project. Therefore, the Ancestry.com submissions can not be analyzed by the FTDNA MRCA utility.

Note 4: The two Ancestry.com submissions are missing several markers found in FTDNA 37 marker submissions and have few common markers for markers 38 through 111. It is unfortunate that neither company does not offer a more competitive upgrade option to make available the missing markers that are available in the Ancestry.com submissions. FTDNA now offers a 111 marker test that includes all markers tested by Ancestry.com. These inconsistencies in offerings create some significant drawbacks in testing submissions from different testing companies.

Step 3) - Determine possible NPE connections

The Hanvey submission is very strong candidate as a Casey NPE. This strong NPE event has not been thoroughly researched to date. Very few submissions will qualify for NPE connections but the DNA marker values for the South Carolina are very uncommon. The Kersey submission is another form of a NPE but is really a surname variation. Both submissions clearly have the South Carolina DNA fingerprint which is very unique.

This step attempts to determine if any possible non-Casey submissions are "Non Paternity Events" (NPE) submissions that should be added to this cluster. First, check that all non-Casey members that are part of Casey FTNDA Surname Study to see if any belong to this cluster. Currently, except for one submission (Hanvey), all closely related submissions in the FTDNA Casey Surname Project have the Casey surname. Next, look at the Y-Search database to validate non-Casey near matches that could be NPEs. Must rule out overlapping haplotypes (common marker values). This is where unrelated lines with common marker values overlap due to having only 37 or 67 markers being analyzed.

If few non-Casey surnames appear to be within 6 mutations at 32 or more markers, then it is much more likely that those submissions may have Casey DNA but not a Casey surname. If most submissions do not have the Casey surname, then overlapping haplotypes probably exist and there would be fewer possible NPEs. If many non-Casey surnames exist, we would need to depend primarily on traditional genealogical research to prove these submissions are NPEs. Also, upgrading to additional markers could further separate other submissions that may be overlapping haplotypes.

Note: As a sponsor of my DNA submission, I can customize my profile to include matches that do not have the Casey surname. I highly recommend that all participants change their default profile to allow non-Casey submissions to be included in any search.

At 32 markers, there were 29 submissions found in Y-Search with 32 or more markers that had six or less mutations. The Ancestry.com 42/43 marker test shares 32 common markers with the FTDNA 37 marker test. Unfortunately, this misses the Kersey submission which only tested for 29 markers some time ago. It was missed in previous Y-Search searches and was actually found in the L226 database of submissions. Ten of these submissions were Casey submissions and one Hanvey submission is believed to be a Casey NPE.

Number
Mutations

Number non-Casey Surnames

Number Submissions

Percent NPE  
0
0
1
0
Casey
1
1
5
20
Casey (1 to 4), Hanvey
2
1
7
14
Casey (5 to 7)
3
1
10
10
Casey (8 to 10)
4
1
10
10
None Found
5
3
15
20
Bryant, O'Brien (4)
6
16
29
55
Brooks, Brown, Callahan, Forbes, Hart (2), Hogan, Kennedy, Murphy, Oakley, O'Donaghue, O'Haren, Upton, Wilson

The above table has very believable NPE rates between 10 and 20 percent up to five mutations but jumps to 55 percent at six mutations. The Kersey submission will eventually replace the current MRCA and will have a major impact on the above table as it will change at least two markers in the MRCA (460 and 464b). From the above table, it is very clear that the Hanvey submission is very closely related to the South Carolina cluster. It appears to be an excellent NPE connection. After exchanging information with the Hanvey sponsor, any doubts about this NPE connection were elminated due to the following additional information:

1) This Hanvey line does not genetically match several other Hanvey submissions that are supposed to be related to this Hanvey line.

2) This Hanvey line lived in the same western counties of South Carolina where all the South Carolina submissions resided.

3) Traditional documentation implied that this Hanvey line had economic ties with the Casey's that could not be explained.

Many other non-Casey submissions in the above table have South Carolina ties and ties to Ireland. The Hart line show ties to South Carolina. Eight other submissions show their origin as Ireland. Four others show their origins in Munster, Ireland: Clare - O'Brien (2), Limerick - Hart and Tipperary - Kennedy. Most L226 submissions (which the South Carolina cluster tested positive for) shows very strong ties to Munster, Ireland as well. These close ties to Munster, Ireland increases the odds for some of these non-Casey submissions having NPE connections.

The South Carolina Casey cluster appears to have less evidence of overlapping haplotypes (common DNA values that are not really related) for submissions that are more closely genetically related (Hanvey). There were four O'Brien submissions at five mutations which indicates that this O'Brien line and this Casey cluster could have NPE connections. However, this is not a very strong connection at five mutations and these O'Brien submissions do not share the South Carolina DNA fingerprint. Below is a summary of the most rare DNA values for the MRCA of the South Carolina Casey cluster. There are many useful facts included in this analysis:

1) The delta weight is an indication of how rare the DNA fingerprint within all R1b submissions. Having a total of 733 is very high and suggests a rare DNA fingerprint.

2) The L226 Off Modal, shows 288 which indicates that the South Carolina cluster also has very unique DNA fingerprint in the L226 cluster as well.

3) 393 = 12, 460 = 12, 534 = 14, 481 = 23 and 458 = 16 is a very unique DNA fingerprint for the South Carolina cluster.

Rare marker values for this cluster include:

Marker Number Rarest Marker Values % of R1b R1b Weight R1b Delta Weight L226 Off Modal
464a 13 2 98 78 MRCA
464b 14 2 98 69 MRCA
459a 8 3 97 92 MRCA
393 12 4 96 87 87
460 12 6 94 68 68
557 15 9 91 62 MRCA
464c 15 11 89 37 MRCA
534 14 13 87 38 38
481 23 17 83 53 53
459b 9 17 83 64 MRCA
458 16 18 82 32 32
439 11 22 78 39 MRCA
CDYa 37 29 71 1 1
CDYb 38 31 69 0 MRCA
449 30 32 68 9 9
456 15 36 64 4 MRCA
576 18 42 58 0 0
Total       733 288

At first glance, the O'Brien line appeared to be a reasonable NPE candidate (having four submissions at only five mutations). Unfortunately, all these submissions do not match the DNA fingerprint of the South Carolina cluster. O'Brien (one submission) has 393 = 12, Murphy and Wilson have 460 = 12, Hart (two submissions) and Callahan have 458 = 16, Oakley has 534 = 14 and Brooks and Oakley have 481 = 23. With the exception of the Hanvey submission, none of the non-Casey submissions within six mutations closely match the South Carolina DNA fingerprint. Therefore, it is pretty lows odds that these non-Casey submissions would share any NPE connection in the genealogical time frame (around 300 years), Future analysis will compare matches to the Kersey submission which could reveal more good NPE candidates based on a more accurate MRCA.

Step 4) - Determine impact of deep ancestry information

Determine if there are any other distantly related clusters that could be related to the South Carolina Casey cluster in a genetically significant time frame (after our ancestors first started using surnames approximately 24 generations ago - around 600 years ago). This is extremely important in determining the haplotype of the MRCA of each cluster. Even though there is almost no chance of solving any genealogical connections (200 to 300 years), any possibly related clusters in the last 600 years could result in a MRCA that lies somewhere between the two clusters. This is extremely important in determining the MRCA of each cluster. In order to qualify, submissions must exceed 50 % at 24 generations (personal definition of closely related).

Comparing the submissions included in the South Carolina cluster only, all show 100 % relatedness at 67 markers (except one at 99 %). This clearly shows that these submissions share a common ancestor at 24 generations (or around 600 years when our ancestors first started using surnames):

FTDNA
Number
Description of Line

Mutation/
% Related
at 12 markers

Mutation/
% Related
at 37 markers
Mutation/
% Related
at 67 markers
51924
Abner/Pleasant/Elsberry
MRCA
MRCA
MRCA
54166
Abner/Pleasant/Pleasant
0 / 91
0 / 100
0 / 100
107623
Abner/Turner
0 / 91
1 / 100
1 / 100
29956
Hanvey NPE
0 / 91
0 / 100
0 / 100
56874
Jesse/Anthony
0 / 91
1 / 100
1 / 100
77349
Ambler/John
0 / 91
1 / 100
1 / 100
56130
John (SC)
0 / 91
1 / 100
1 / 100
99381
John/Levi/William
2 / 37
4 / 99
4 / 100
93773
John/Levi/Francis
1 / 68
2 / 100
2 / 100
45068
James H. (SC)/Willis
0 / 91
2 / 100
2 / 100
119586
Moses (SC)/Moses
0 / 91
1 / 100
1 / 100
58301
Henson/Arvle
0 / 91
1 / 100
1 / 100
40325
Henson/Jackson
0 / 91
1 / 100
NA
166618
James G. (TN)
0 / 91
2 / 100
NA
172505
Randolph/Levi
0 / 91
4 / 96
4 / 99
9BCZF
John Kersey, b. 1616, England
1 / NA
2 / NA
NA
WN8TA
Unknown Casey
1 / NA
2 / NA
NA

Comparing the submissions of the Munster cluster to the South Carolina cluster at 24 generations, the degree of relatedness supports that these two genealogical clusters could share a common male ancestor when our Caseys first started using the surname of Casey (this needs updating to new South Carolina MRCA (460 13 to 12):

FTDNA
Number
Description of Line

Mutation/
% Related
at 12 markers

Mutation/
% Related
at 37 markers
Mutation/
% Related
at 67 markers
51924
Abner/Pleasant/Ellsberry
MRCA
MRCA
MRCA
34073
Michael (Limerick)
1 / 70
8 / 63
12 / 58
53484
Dennis (Kerry)
1 / 70
10 / 63
15 / 71
56031
Daniel (Clare)
1 / 70
13 / 48
16 / 71
131349
Elisha (SC)
1 / 70
12 / 28
17 / 40
42BNV
Martin (Limerick)
3 / NA
11 / NA
NA
6FW87
Daniel (Cork)
2 / NA
10 / NA
NA

At 67 markers, the average probability of relatedness is 60 % and qualifies as a closely related genetic cluster. As a closely related genetic cluster, both clusters could greatly influence the determination of the MRCA of each cluster. We now have five MRCA haplotypes to determine the relationships of Casey and Kersey lines under the L226 deep ancestry:

1) The MRCA of all L226+ submissions.

2) The MRCA for all combined three Casey & Kersey clusters.

3) The MRCA of the Munster, Ireland cluster.

4) The MRCA of the combined Kersey / South Carolina cluster

5) The MRCA of the South Carolina cluster.

We also have four strong branches between these MRCAs:

1) 534 (15 to 14) from L226 MRCA to all three combined Casey/Kersey MRCA.

2) 481 (22 to 23) from all three combined Casey/Kersey MRCA to the Munster, Ireland MRCA.

3) 393 (13 to 12), 458 (17 to 16) and 449 (29 to 30) from all three combined Casey / Kersey MRCA to Kersey / Casey (SC) MRCA.

4) 481 (22 to 23), 464b (13 to 14) and 460 (11 to 12) from Kersey / Casey (SC) MRCA to South Carolina MRCA.

These connectons can be best been seen visually:

DNA Descendancy Chart - L226 Casey cluster

Step 5) - Determine the MRCA for this cluster

Determine the MRCA of this cluster based on all submissions in this cluster. The discovery of the Kersey submission greatly expands our knowledge of the South Carolina cluster but greatly complicates the analysis as well. Until more markers are available and more Kersey submissions are available, the MRCA will only include the South Carolina Casey submissions. There are definitely enough markers in the Kersey submission to determine that this line is closely related to the South Carolina lines. However, there are not enough markers for the Kersey submission and there needs to be more Kersey submissions to properly analyze any Kersey submissions and their impact on the South Carolina cluster.

This cluster is very easy to determine the most likely haplotype of the MRCA using the "majority rules" methodology. The exception is marker 460 where around half of the submissions have one marker value and the other half have a second marker value. Since the South Carolina cluster descends from the L226 MRCA, it is very likely that 460 = 12 would be the closest marker value to the L226 MRCA. Additionally, the Kersey submission also has 460 = 11 which also implies 460 = 12 would be the most likely MRCA value for the South Carolina cluster. For a summary of the marker values, see the DNA Results summary of the South Carolina cluster:

DNA Results of South Carolina Cluster

Note 1) It is unfortunate that Ancestry.com submissions are missing many markers that FTDNA submissions have at 37 markers. The Ancestry.com submissions are missing the fast moving markers CDYa and CDYb. These two markers play a significant role in the South Carolina cluster (as it does with many clusters). If submission WN8TRA is upgraded to FTDNA's 37 markers, the possible (CDYb = 37) could tie this submission to the Ambler Casey and Jesse E. Casey lines. If WN8TRA is upgraded to FTDNA's 37 markers, (CDYa = 36) could tie this line to Randolph Casey and James G. Casey lines. With introduction of the Kersey submission from Ancestry.com, this submission is missing the key markers 481, 534 and CDYa which are very unique to the South Carolina DNA fingerprint.

Note 2) It is unfortunate that FTDNA markers and Ancestry.com markers are not more closely aligned and that economical upgrades are not available for missing markers. It is unfortunate that neither company offers economical upgrades to each other offerings.

Note 3) With thirteen submissions (out of sixteen) now having 67 markers, it is unfortunate that no additional mutations were discovered to separate the South Carolina lines further. On one hand, all these markers being the same increased the probability of relatedness between all lines in this cluster. On the other hand, having no additional mutations did not reveal any additional possible branches to further separate these lines. Also, the degree of relatedness significantly increased between the South Carolina cluster and Munster, Ireland cluster. This also suggests that the South Carolina Casey cluster probably may have ties to southwestern Ireland vs. northern Ireland as speculated in early manuscripts.

Step 6) - Determine Branches within cluster

This step is even dependent on traditional genealogical information to be reviewed in combination with the DNA evidence. Review of traditional genealogical research allows several mutations to be eliminated from any DNA descendancy chart as recent mutations that are not genealogically significant. Traditional documentation can also take a weak genetic connection and make the connection stronger based in traditional documentation. The analysis of this step is even more dependent on the accuracy of traditional research. If speculative information is entered as fact, the DNA analysis from this point on could be incorrect.

Based on traditional research, there are three submissions for the Abner Casey (TN) line, two submissions for the John Casey (MO) line and two submissions for the Henson Casey (TN) line. There are ten different South Carolina Casey lines included in this cluster and an estimated 40 or 50 unconnected South Carolina lines that exist. This suggests that we only have 20 to 25 % of the estimated lines tested to date. The line of Ancestry.com Casey submission is unknown at this point in time. Also, very little is known genetically about the Kersey branch as well and many more markers are needed for the Kersey submission and many more Kersey submissions are needed as well. Several of the South Carolina submissions have well proven connections to South Carolina and all other lines have proven connections to Tennessee (with speculated connections to South Carolina). I believe that DNA has proven that all submissions in this cluster probably have ties to South Carolina as we have all already speculated. DNA documentation now provides strong documentation that almost all lines with South Carolina ties are closely related. However, the Elisha Casey (SC) line appears to be one exception and belongs to the Munster, Ireland cluster. The Kersey submission is a quantum leap in breaking the South Carolina Casey brick wall. For the first time, we now have ties outside of South Carolina (England) and another surname variation (Kersey) to begin to research. Here is a summary traditional genealogy on the submissions in the South Carolina cluster:

Family Trees - South Carolina Cluster

Based on a combination of both DNA submissions and traditional genealogical documentation, the following chart is the best estimation of the possible branches associated with the South Carolina cluster. This "DNA Descendancy Chart" should be considered very speculative and will change over time as more information becomes available (this chart needs updating):

DNA Descendancy Chart - South Carolina Cluster

Note 1) The marker 460 is the most significant marker in this cluster as around half of the genealogical lines descend from one son of this most common recent ancestor (MRCA) and the other half descend from all other sons of this most common recent ancestor (MRCA). This implies that this ancestor did not have many sons and/or there were many fewer male descendants of the other sons. It is now fairly certain which marker value of 460 represents the DNA marker value of the MRCA of this cluster based on deep ancestry research. The "majority rules" methodology fails to identify the MRCA marker value for marker 460 since there is a pretty even split of submissions with each value. The Kersey submission, the Munster, Ireland MRCA haplotype, the L226 MRCA haplotype and marker value distribution table for 460, all imply that 460 = 12 is the obvious choice for the MRCA value of marker 460.

Note 2) There are also three strong candidates for genetic second level branches off of the 460 first level branch. The John Casey (MO) line has two submissions where both have marker value 391 = 10. Both genetic information and traditional documentation establish that mutation 391 (11 to 10) is probably a unique DNA fingerprint for this line. If other new submissions that are not connected to John Casey (MO) have marker value 391 = 10, then both lines would share a common ancestor that other lines do not share. The Ambler Casey and Jesse E. Casey line both share marker value CDYb = 37. CDYb is a very fast moving marker, so this second level branch is less strong than the John Casey (MO) branch. However, traditional speculative documentation implies that Ambler Casey and Jesse E. Casey were brothers and DNA evidence supports this connection. More submissions of additional sons of Ambler Casey and Jesse E. Casey are needed to verify this branch. CDYa = 36 is another second level branch that implies that James G. Casey and Randolph Casey share a common male ancestor. This is supported by traditional genealogical research that shows James G. Casey residing in common counties in Tennessee and later Illinois.

Note 3) Several other mutations that appear to be unique to each submission. These mutations could represent possible future branches in the DNA descendancy chart. However, these mutations could have occurred in the earliest two or three generations or could have occurred in much more recent generations. Further submissions of other sons or grandsons of these branches could push these mutations down the DNA descendancy chart or reveal that these mutations are earlier and truly represent early branches of the DNA descendancy chart. The James H. Casey line has a unique mutation 437 (15 to 16). If another son or grandson of James Casey were tested, this would prove that this mutation is recent (if the second submission came back with 437 = 15) or genealogically significant (if the next submission came back with 437 = 16). The Ancestry.com Casey submission (WN8TRA) has 389-1 (12 to 13) and represents another possible future second level branch. Unfortunately, no traditional information is known about WN8TRA making any analysis problematic.

Note 4) There are also several other sub-branches that are probably not as genealogically significant. The mutations of 385b=15, 447=24, 607=16 and CDYa=37 represents a sub-branch where genealogical information already has well proven connections. The mutation 385b=15 is a third level branch the John Casey (MO) line that is a fingerprint for John's grandson, William Casey (or any of his male descendants). This sub-branch has less significance because this mutation only separates the sons of Willis Casey and does not solve any unknown genealogical questions. The mutation of 607=16 is another mutation that is unlikely to be genealogically significant. This mutation distinguishes Jackson Casey (or one his male descendants) from Arvle Casey – both are known brothers. This mutation does not solve any genealogical questions.

Note 5) There are also other sub-branches that appear not to be genealogically significant if only considering the South Carolina Casey cluster by itself. However, if you look at other related MRCAs (Kersey submission, Munster, Ireland MRCA and the L226+ MRCA, there are two mutations that could be low-probability wild cards. The mutation of 460=11 appears to another fingerprint of the descendants of William Casey, grandson of John Casey (MO). The most likely scenario is that this represents a backwards mutation to original value of the Kersey and L226+ marker value. Since the 460 marker represents such a significant marker for this cluster, additional testing of another son or grandson of John Casey (MO) is warranted. Another sub-branch that could be much more important is CDYa=36. This appears to be a fingerprint for the Turner Casey line and separates Turner Casey (or any of his male descendants) from his brother, Pleasant Casey. However, this mutation would have to another backwards mutation to the original L226+ marker value.

Note 6) The above descendancy chart is the current highest probability scenario and could change over time as additional markers become available, as additional submissions are tested and as more traditional genealogical information becomes available. It should also be remembered that these mutations are assigned to the individual where the mutation first appeared. Over time, many of these mutations will be concluded to be recent mutations that will be pushed down the descendancy chart.

Note 7) The submissions 166618 and 172505 appear to share a common ancestor. DNA does not prove exact connections as most believe – but does imply common ancestral connections. Since Randolph Casey was born around 1738 and James G. Casey was born in 1805, it is very possible that James G. Casey could be a direct descendant of Randolph Casey due to the differences of their birth dates. However, DNA only implies that they share a common ancestor and can not be used as proof that James G. Casey is a descendant of Randolph Casey. With additional submissions for these lines, other branches could be determined and more common ancestors could be established.

Step 7) - Conclusions that can be reached with current submissions:

1) With the discovery of the Kersey submission, this is the first DNA submission that is closely related to the South Carolina cluster that does not originate from South Carolina. We now have genetic proof that a Kersey line from England since 1616 is the most closely related Casey / Kersey line and we should now research this Kersey line and other Kersey lines for connections. This Kersey submissions shares four L226+ off modal values and could share two more mutations with additional testing. DNA fingerprint matches (L226+ off modal mutations) are much more important the random genetic distance between submissions. This Kersey submission represents the most significant genetic discovery for the South Carolina cluster to date.

2) Even with the current submissions, there appear to be two very strong major branches established for the South Carolina cluster and two other strong branches. There are at least two other possible genetic based sub-branches as well. Additional submissions, additional markers for existing submissions and better traditional documentation will better verify these branches and add other branches – allowing DNA to connect the approximately 40 different Casey lines that originate from South Carolina.

3) For the many Casey submissions with ties to South Carolina (or Tennessee), only the Elisha Casey (SC) line appears to be the only known exception. Most researchers of South Carolina Casey lines have speculated that these lines were closely related and DNA evidence confirms this speculation (except for the Elisha Casey line).

4) The South Carolina cluster is a very unique cluster from a genetic point of view. This cluster includes almost exclusively submissions with the Casey surname and only one non-Casey surname which is believed to be a Casey NPE. At 37/32 markers, Y-Search shows more non-Casey submissions at five and six mutations from the MRCA of the South Carolina cluster. This cluster shows little evidence of overlapping haplotypes (DNA submissions with common marker values that overlap) or genealogically significant NPEs (except for the Hanvey submission). Because the South Carolina Casey cluster has pretty unique DNA values, any submission that is closely related genetically probably is closely genealogically as well – even for any future non-Casey surnamed submissions that have less than five mutations.

5) Any Casey researcher that believes that they are connected to any Casey line in South Carolina should be encouraged to submit 37 or 67 markers to verify this connection. These DNA submissions would almost always support that connection (encouraging additional research) or occasionally would be found not be part of this cluster (eliminating a lot of unproductive research). This is by far the best usage of funds in getting more lines tied together. We also now have five distinct branches and two possible branches discovered, therefore, not only would determine that your line belongs to the South Carolina cluster but you would determine which branch of the cluster that you belong to (narrowing down your research even more).

6) The South Carolina cluster and the Munster, Ireland cluster are probably related in the distant time frame (500 to 700 years ago when our ancestors first started using surnames). This means that both clusters probably share a common male Casey ancestor 500 to 700 years ago. Although these connections will probably never be made in the genealogically significant time frame, this connection is very important to both clusters. Most Casey submissions in the Munster, Ireland cluster have strong ties in the counties in the southwest part of Ireland. This is probably the same area where the South Carolina Casey lines originate as well. This disputes the traditional speculation that the South Carolina Casey lines originate in County Tyrone, Ireland. There is no doubt that the South Carolina cluster belongs to the L226+ haplogroup cluster which also have strong ties to southwestern Ireland. Of course, our Casey could have migrated to northern Ireland and then later to South Carolina. As known with the Kersey submission, we now know that the Kersey branch originated in southwestern Ireland and moved to England by 1600 where they changed their name to Kersey. The genetic DNA descendancy chart reveals that part of this Kersey line later migrated to South Carolina and changed their name back to Casey (or some variation of this scenario).

7) Both the Ambler Casey line and Jesse E. Casey line have CDYb = 37 which supports the speculative traditional connection that these two men were brothers. Henson Casey was speculated to be the son of Ambler Casey and DNA tends to discount this connection at this time. Ambler Casey and Jesse E. Casey both have CDYb = 37 but two submissions from Henson Casey do not have this marker value. The marker value of 391 = 10 appears to be a fingerprint for the John Casey (MO) line. Any future submission with that marker value could be closely related to the John Casey (MO) line. The Hanvey submission speculates that the NPE occurred between 1830s and 1860s when few Casey lines remained in South Carolina. The Hanvey line appears to be very closely related to the Abner Casey (TN) line. Researchers of the Abner Casey (TN) line should get submissions of these late SC lines and look at traditional methods to connect these lines. Because the John Casey (MO) line and the Ambler Casey line both resided in McMinn County, TN, there was speculation that Ambler Casey (TN) and John Casey (MO) could be brothers. Both of these lines married into the Ellison family of Roane County, Tennessee as well. DNA marker 460 tends to discount this connection since Ambler Casey has 460 = 13 and John Casey (MO) has 460 = 12.

Possible Action Items for cluster:

1) Definitely need more traditional information concerning the Casey submissions in this cluster. Please look at the menu, "Family Histories" and the sublink, "South Carolina" and send more detail concerning your lines. Please look at the the "Family Histories" menu and the sublink "How to Submit" for a complete details of how to summarize your family history. For the Ancestry.com submission WN8TRA, there is no traditional genealogical documentation other than the Casey surname, therefore, DNA analysis for this submission is extremely limited without the traditional documentation. The extremely important Kersey submission also has limited traditional documention as well.

2) Only eleven lines out of forty lines are currently covered in the Casey Surname Study. The Levi Casey line have no representation even though this line is well-researched. The highest priority remains getting as broad as coverage all known lines as possible. Everyone wants to find out more about their lines by submitting more submissions of their lines – but you must test the other unrelated lines to see which ones appear to be more closely related and to establish more genetic branches.

3) Researchers of this Casey cluster should seek out currently non-related Casey lines that they feel could be connected to their line and get their DNA submissions. These submissions will mostly prove unrelated if outside of South Carolina or Tennesse - but this will eliminate possible related lines that are no longer worthy of attempting to connect to this cluster. However, should you discover a related line that nobody knows how is connected, this could really provide some breakthroughs on this cluster. We have Casey lines with ties to Virginia that belong to the South Carolina cluster. This implies that these connections may not really exist or that we just not tested any these related Virginia lines to date. We have just scratched the surface of understanding the connection of the Kersey line with the South Carolina lines - we should all immediately focus on this connection to break the Casey ancestry brick wall.

4) Since there are many mutations between the MRCA haplotype of the South Carolina cluster and most of the submissions within this cluster. It always very beneficial to obtain a second and sometimes third son (or grandson) of the oldest proven ancestor of each submission. These additional submissions would assist in the determination of whether the existing mutations are from the older generations or more recent mutations nearer the donor (and are not genealogically significant mutations). The Ambler Casey and Jesse E. Casey lines need additional submissions to verify that CDYb = 37 represents a genealogically significant mutation or a recent mutation. The Henson Casey line needs a third son of Henson to be tested as the original two sons produced unique results (one submission with 460 = 12 and one submission with 460 = 13). As it currently stands, assigning this line to either branch is very speculative. Another submission of the James Casey (SC) line is required to verify that 447 = 16 is a fingerprint for this line or just a recent mutation. Another submission of a proven descendant of different son or grandson of Randolph Casey is needed to verify the possible branch for this line.

5) The Munster, Ireland cluster appears to be genetically related to the South Carolina cluster. Both clusters probably share a common Casey male ancestor around 500 to 700 years ago. Researchers of the South Carolina cluster should encourage additional submissions in the Munster, Ireland cluster as well as research these Irish Casey lines. Now that it is known that the South Carolina cluster probably has ties to southwestern Ireland, any Irish research should be focused on these southwestern counties of Ireland and the lines found in the Munster, Ireland cluster. The Munster, Ireland cluster is very important to the South Carolina cluster in determining MRCA haplotypes that are shared between these two clusters.

6) Since the Munster, Ireland cluster and South Carolina cluster appear to be related in the 500 to 700 time frame, the Kersey submission is now known and the L226+ MRCA is now known, additional testing of certain South Carolina lines may help verify these connections and increase the accuracy of the MRCA haplotypes. Another son of John Casey (MO) needs to be tested to verify the MRCA of the 460 marker. If another son (or grandson) has 460 = 11, this would further strengthen the tie to the Munster Ireland cluster and could be grounds for altering the MRCA haplotype for the entire South Carolina cluster. The submission of Abner Casey (TN) / Turner Casey shows CDYa = 36 and could represent MRCA haplotype for this marker. Testing yet another son of Abner Casey could come back as CDYa = 36 and establish that this marker value is the MRCA of this line as well as the entire SC cluster. If the third son comes back as CDYa = 37, this would further establish that the Turner Casey line is the source of CDYa = 36. These are lower odds scenarios - but could prove to be wild card mutations that could have a profound impact on the analysis of this cluster.

7) There is no doubt that you should compile documentation on all Casey lines found in this cluster and attempt to find connections via traditional research. I believe that future family history publications will not be limited to only family lines that can be connected via traditional research. These future publications should also include lines that are closely related by DNA evidence as well. Hopefully, these publications will not attempt to speculate too much on how these lines are connected as speculation in publications unfortunately turn in facts in future publications that use less strict standards for proof.