Cheek/Chick DNA Project

Summary of Results

As of September 1, 2007, we have we have received Y-DNA test results for 40 participants.  We have now identified 2 distinct family groups with the Cheek/Chick surname.  We are calling these Group 1 and Group 2.

Approximately 80% of our participants so far fall into either Group 1 or Group 2.  The remaining 20% of the participants have unique (non-matching) DNA results.  As the study progresses, we hope that some of our "uniques" will ultimately find a "match" that will reveal a new, as-yet-unidentified Cheek/Chick family group.

The participants in Group 1 are descendants of various Cheek and Chick families in Virginia, North Carolina, South Carolina, Georgia, Alabama, Mississippi, Texas, and other Southern states.  Their common ancestor is believed to be a man named John Cheek, Sr., who settled in Old Rappahannock (Essex) County, Virginia, during the mid-1600's.  Most immigrants to Virginia during this time period were from England, and the DNA supports the assumption that John Cheek was English.  The members of Group 1 have certain Y-DNA markers that put them into a DNA "super-family" known as "Haplogroup R1b1" which is typically found in men from the Atlantic coast of western Europe.  In the British Isles, Haplogroup R1b1 is associated with the Celtic populations of southern England, Wales, Scotland, and Ireland.  We think that John Cheek (or his paternal ancestors) probably came from southern England, possibly Devon, Dorset, Somerset, or Hampshire, where the Cheek and Chick surnames seem to have originated.  (For more information about the Cheek/Chick surname, click here.)

Two other participants in our study (Participants #18 and #34) can trace their ancestry to a family named Cheek/Chick who lived in Somerset, England, in the 1500's.  Participant #18's family migrated to Australia.  Participant #34's family came to the United States from England in the mid-19th century.  These participants may be related to Group 1, but more distantly (probably several hundred years).  This supports the theory that the family of John Cheek of Virginia originally came from southern England, possibly Somerset.

Small variations in the individual DNA profiles can be used, in some cases, to distinguish separate branches of the family.  For example, we have found a consistent difference on the genetic markers called "389-1" and "389-2" which seem to distinguish:

1. The descendants of John Sr.'s (probable) son Richard Cheek who died in Beaufort Co., NC, in 1745, and
2. The descendants of John Sr.'s (probable) grandson Robert Cheek who died in Orange Co., NC, around 1817-1823.

These two branches are closely related, but a single mutation occurred in the line of Robert "of Orange," which we now find (for the most part) only in his descendants.

Most members of Group 1, including the two Somerset participants, have an interesting genetic quirk: the value of the marker called "DYS# 464d" is "20."  This result is extremely unusual.  In fact, scientists had never previously reported a value of 20 on DYS# 464d.  It was always 19 or less.  The FTDNA lab had to run most participants' tests several times to make sure that the unexpected value was not an error.  After many delays and great angst (some of our early participants waited 4 or 5 months to get their results), the FTDNA scientists finally concluded that our quirk really does exist.  It's turned out to be very useful for our study.

The participants in Group 2 are definitely related to each other within the last 200-300 years.  The DNA results validate this family's "traditional" genealogical research, which traces their ancestry to a man named William Cheek who immigrated to Virginia from London, England, in 1754.  But is Group 2 related to Group 1?  When we compare the most common Group 2 profile to the most common Group 1 profile, the probability of a common male ancestor within 20 generations (about 500 years) is only 10%.  However, the probability of a common male ancestor within 40 generations (about 1,000) years is 59%—a little better than 50-50 odds.  So it's possible that the two groups are descended from a man named Cheek or Chick who lived in the England during the Middle Ages (750-1,000 years ago), around the time that surnames first developed.  On the other hand, our common ancestor could have lived more than 1,000 years ago, before surnames were invented in England, in which case the common surname is purely coincidental.

The unique participants do not match anyone else in the study so far.  A unique participant may be the first representative of a new family group which we will have to call Group 3, once we find additional matches.  Alternatively, a unique participant may have a "non-paternity event" in his direct male line such as an adoption, step-child name change, or illegitimate birth.  In such a case, it is often helpful to test more descendants from that particular family line.  According to the FTDNA company, the chance of a DNA mismatch due to a "non-paternity event" is about 2%-5% per generation.  That works out to 18%-40% in 10 generations.  So, if you test 20 men whose (presumed) common male ancestor lived 10 generations or roughly 250 years ago, you will probably have between 3 and 8 mismatches.

If you don't match the other descendants of your presumed ancestor, scour your family tree for clues such as remarriages with step-children, divorces and separations, and children who seem to be either too old or too young considering their mother's age and date of marriage.  You may be able to identify where the "break" occurred.  For example, one classic scenario is a child born out of wedlock who is raised by his maternal grandparents.  He will probably be the youngest child in the family and possibly separated from his next-oldest "sibling" by several years.  His "mother" (grandmother) may have been in her 40's or even 50's when he was born.  He will be identified as a son on census records because of the social stigma of illegitimacy.  Another common situation is a woman who marries after being divorced or abandoned by a first husband, or having a child out of wedlock, and the child takes the step-father's surname.  County divorce records, bastardy bonds and records of "base-born" children who were "bound out" can be rich sources of genealogical information.  However, you may have to search original records or microfilm at the county court, state archives or at an LDS Family History Center.  In most places, only a small percentage of the available court records have been indexed and published.

For those disappointed in their DNA results, don't despair.  If you go back 10 generations, assuming no cousin marriages, you have 1,024 great(x8)-grandparents.  That's 1,024 different lines of descent in 10 generations, and the number doubles with each additional generation.  Y-DNA testing provides information about only one line of descent—the exclusively male ("patrilineal") line.  Although Y-DNA gives us a unique insight into our male lineage, there are still plenty of other branches of the family tree to explore.  And of course, you could still be related to one of the main groups through a female line, such as a child born out of wedlock who was raised by his grandparents.