Dear Robin, You need to identify the place and time of the family name. Before the 1750s, spelling was not really standardized. Later the tried to spell phonetically. Naturally there were those educated and not so educated, so you would have variants. Look for legal documents and such for the accepted spellings. Generally speaking, I try to standardize the spelling in family groups. However ... For example: Searles is a sound/spelling variant of Sales with a cockney slang. Yet Sales comes from the older Sailes. Thus in family grouping the first spelling in that group tying into the next and the next. Another example is: Rosencrans, Rosenkrans, Rosecrans, Rosekrans and some 26 variants all have the same ancestry. Yet each group is a distinct family group in time and location. Yet all have the common "Rose", thus searches are done by "ROSE*" with the "*" as a wild card. In your Bennett with variants of Bennet, Benet, etc, the wild card search with the last name "Ben*". I use various genealogical programs, but I like PAF (Personal Ancestry File) with the exception of PAF 3.0 [ :( ]. Paf 2.31 and the new PAF 4.0 use a Random Information Number or RIN. You can down load PAF 4.0 for free at: http://www.familysearch.org/OtherResources/Other_Resources.asp Once the RIN is assigned in the data base, it stays the same unless you renumber by deleteing and GEDCOMing the data to another data base. If you keep the data as is, the numbers are the same. This is a great help when trying to quickly find one William Carpenter in a data base of some 48,000 names. As I was writing this, I did a quick search and found 586 William Carpenter variants. I can filter the date of birth or location also, but it is easier to go with a known RIN number. The PAF 4.0 search filters are more advanced than version 2.31, but it takes longer to do a search. In my e-mail address is JRC RIN 001. Can you guess what number I am in the data base? I hope this answers your questions. John R. Carpenter La Mesa, CA Robin Carpenter wrote: > > Hi John: > Thanks much for the Bennett data. Please advise how your system deals > with variant spellings. E.g., does the info you sent include "Bennet" as > well, but adopt uniform spelling, or is that separate? > (I have similar issues with Carpenter marriages to Pierce vs Peirce; I've > never really settled my mind as to whether these are "deliberate" different > families, or just idiosyncratic variations (or even just typos!). > Also, can you explain your numerical coding system a bit? Maybe this > would be of general interest to our List subscribers...in that you are > obviously a major data source for all of us. > Robin