Walter V. Hickeyarchives specialist with the National Archives and Records Administration–Northeast Region (Boston)writes in "A Gold Mine Of Naturalization Records In New England" (at NARA's Prologue, Fall 2004) about an unusual set of naturalization records held at the NARA branch in Waltham, Mass. These are "photostatic copies, called 'dexigraphs,' made in the 1930s, of naturalization proceedings in all courts—federal, state, county, and local—in five of the New England states (Maine, Massachusetts, New Hampshire, Rhode Island, and Vermont) between 1790 and September 26, 1906."
Early American naturalizations are among the most difficult public records to locate, largely because so many courts could claim jurisdiction. Some court records have never been microfilmed, and those which have are sometimes poorly indexed.
These NARA copies provide a one-stop index of New England naturalizations. The researcher, upon finding a relative in the dexigraphs, can then track down in the relevant court (or state archives) the immigrant's declaration of intention and original petition for citizenshipoften invaluable sources of information on his origins. It is not unusual for these records to include the place (and sometimes the date) of birth, place and date of arrival, and recent places of residence of the petitioner.1
Hickey's article casts fresh light on this important research tool, and gives information on obtaining dexigraph copies from the Waltham branch. Similar indexes are available for the courts of New York and Illinois.
1Before relying on these records, see Carmen J. Finley's "Original Naturalization Records: A Reliable Source for Birth Dates?" National Genealogical Society Quarterly 91:60 (Mar. 2003). The article cites a study conducted in Sonoma County, California, which found a 1.3% error rate in birth dates recorded in multiple filesdiscrepancies ranging from a few days to 23 years (the latter probably due to a clerical error). As this error rate is only for records in which a birth date is repeated, the actual error rate in reported birth dates is probably much higher. (Anecdotal evidence supports this conclusion.)