First of all let us consider where the date of 1762 may have come from for the marriage of John Abercrombie and Ann Kellet.From what I have found it appears to included in Ancestry's US and International marriage database that so many use in their trees as a source.Just reading the source information should send out a huge BEWARE!It is not based on actually marriage records, but on Yates Publishing. The following is a quote from Ancestry. "This unique collection of records was extracted from a variety of sources including family group sheets and electronic databases. Originally, the information was derived from an array of materials including pedigree charts, family history articles, querie." Family group sheets, pedigree charts etc that were submitted by individuals to Yates over the years.Strickly opinion in most cases.So the year of 1762 or what you state ca1764 is not necessarily correct. It could have been 1760 for all we know.
Now that I have that out of the way we really do not know the exact date of marriage of John and Ann.So Colville could definitely be their son as well as James b ca 1765.As for the Christie Abercrombie.Maybe John and Ann did have a daughter by this name, but once again there is no proof only theories.We could go around and around all day on who may have been the child of certain parents.If they are not named in a will or any other legal document then it is no more than a guess.As for naming patterns for the children of John and Ann.Colvill may have been a good friend, a neighbor etc and then they named their next son James.OK?When it comes to when certain males married once again we only have what we have since I have yet to learn where any early marriage records that would aid in this puzzle may exist.That seems to be the bottom line.I have been told the records from the period we need no longer exist.They were destroyed or lost at some point.I just wish people would realize that databases such as the US and International Marriage database as well as Ancestry's Family Collection databases are secondary sources and should only be used as a tool and not a fact.