Imagine a new genealogical index with more than six million entries! What if those entries, pointing to early American families, had very little duplication with the International Genealogical Index,™ Ancestral File,™ World Family Tree, PERiodical Source Index, or the U.S. Census Indexes? And what if over half of those entries had never appeared in an index before, and most entries cited lineage-linked (multi-generational) collections? Are you interested? Well, what red-blooded American researcher would not be?

It's all true-except for one tiny fact: the index is not new. In fact, parts of this index have been around for more than fifty years. Today it indexes more people than any other printed (book) index in the world of genealogy and family history.1 It is also the most comprehensive index and one of the easiest to use. Yet, most modern researchers have not heard of it, and those who have, seldom use it. It is truly one of genealogy's best kept secrets.

Historically, the index has been available in less than two hundred libraries. But that has changed. Fremont Rider's American Genealogical-Biographical Index was recently posted on the Internet at Ancestry.com's Web site and will be made available on CD-ROM later this year. In view of the tremendously increased access to this important but often overlooked tool, it is time we all became better acquainted with AGBI.

Defining AGBI
The American Genealogical-Biographical Index is actually the second edition of an index begun by Fremont Rider in 1936. It is often referred to in older literature as the Rider Index (after its creator), or in contemporary literature as AGBI. Rider, the librarian at Wesleyan University in Middletown, Connecticut, was an avid genealogist, and had a vision of an every-name index to hundreds of published genealogical works. He personally borrowed and collected hundreds of published family histories, some local histories, and published church and vital records, and began a subscription index. Volunteer indexers combed through nearly six hundred volumes to create the index. Between the years 1942 and 1948, The American Genealogical Index was published in forty-eight volumes.

However, every indexer knows that some sources were left out. Rider added about 220 more sources and began again, this time naming the work, American Genealogical-Biographical Index. He founded a small, private library, the Godfrey Memorial Library, in Middletown, to support the indexing and publishing. By carefully calculating the number of names on each page, and the pages in each volume, we now know that the final index has about 6.25 million names.

The index was published in large, library-quality volumes, with about 520 pages per volume. Beginning in 1952, the second edition was issued, with an average of about four volumes per year. That indexing project, now forty-seven years old, is approaching the final volume, which should be issued this year. As of January 1999, there are 197 volumes that index the surnames from A through Wicoff.

The Scope of AGBI
The most impressive aspect of AGBI is its scope. It is an every-name index to about 850 sources for eastern United States research. No other index has even attempted to accomplish what AGBI has succeeded in doing. But determining the exact number of sources indexed is problematic as a few titles were added throughout the indexing.

One authoritative list of sources indexed includes at least 841 titles. A vast majority of these titles are the approximately 723 published family histories, most of which pre-date 1920 and belong to the first great wave of printed genealogies. It appears that less than half of the index entries trace back to the published family histories. The rest of the names come from a wide variety of sources (see sidebar, p. 40).

It is only natural that a large majority of the sources indexed have a New England flavor. The index was created in that region, and those states were the most prolific at publishing genealogical material during the nineteenth century. However, there is important coverage of other eastern states as well and it is not limited to the 1790 census or the military lists. Family histories include families from North Carolina, South Carolina, and Maryland. Outside New England, Pennsylvania is probably the best represented state; however, titles pertaining to the first twelve colonies, Vermont, Tennessee, and Kentucky, as well as Mississippi, Missouri, and Alabama also appear on the list.

Accessing the Index
Well, with such content, why do so few family historians use AGBI? That situation can probably be traced to two factors. The first is the false perception that this is just a New England index. The second, and perhaps real problem, is that the index is not well distributed throughout genealogical libraries. As with most ongoing publication projects, the index has been produced to meet a subscription list. However, that list began in the early 1950s when there were far fewer genealogical libraries than today. They go out of print after the first several volumes have been published and distributed. So what library would then want to subscribe to an index for which the early volumes were not available? Recently the first one hundred volumes or so have been reprinted, but they are still expensive for many smaller libraries.

The current subscription list is only about 170, which likely includes a few private persons as well as research libraries. Thus, AGBI is not widely distributed. Only two LDS Family History Centers (both in California-Gridley and Sacramento) subscribe to AGBI, yet there are some two thousand centers in North America alone.

Fortunately, the accessibility of AGBI is changing. Previously, leadership at the Godfrey Memorial Library (after Fremont Rider's death in 1962) only saw their task as publishing the index. But new leadership, under director Nancy Doane, realizes the value of the index for researchers, and its importance to the operating budget of the library. Working with the Family History Library, Doane arranged to have the volumes microfilmed so Family History Centers could access the index. With microfilm masters, the Godfrey Memorial Library can now sell microfilm to institutions that have not previously subscribed. Doane also recognized the importance of making AGBI more widely available through electronic publishing. Last year the library contracted with Ancestry.com, Inc. to convert the more than one hundred thousand pages of typed text to an electronic database for commercial sale.

At present you should be able to locate print copies of AGBI in major research libraries, but only at an average of two or three libraries per state. In the state of Utah, for example, even with its genealogical orientation, only two complete copies of the index are known to be available (at the Family History Library in Salt Lake City, and the Lee Library at Brigham Young University). Family History Center patrons can now order microfilm copies of volumes 1 through 186 (Walton). On the Family History Library Catalog on CD-ROM, simply type in 49554. For the rest of us, see http://www.ancestry.com/agbi.htm on the Internet, as AGBI has recently been released on the Web site.

Format
The AGBI is easy to use, but like all large indexes, you need to understand the format to make an effective use of it. Each entry has up to six parts-the surname, first names, birth year, state of residence, biographical data, and citation.

* Surname. This is not repeated for each entry, rather the surname appears at the top of each column (two columns per page). Compound surnames are filed as one name, so Van Der Bilt is filed with Vanderbilt. Names beginning with O' and D' are filed as if the apostrophes were not there, while M' is rewritten as Mc and filed as such. Mac is not interfiled with Mc. Both are filed under their own spelling, thus Mac before Mb, while Mc comes after Mb. Umlauts, tildes, and other foreign accents are ignored in copying and filing the surname.

* First names. These names lead each entry, and include full names, with middle names or initials, as given in the source. Abbreviations are written out if the source is clearly using an abbreviation (usually indicated by use of a period in the source, i.e. Dan. for Daniel). Married women are indexed under their maiden and married surnames.

* Birth year. This key identifier is used when the source suggests a birth date. This provides better identification than any indexes in the sources themselves. The exact year is used if given. If only a decade can be determined, a question mark is used for the final year, i.e. 168?. Heads of household in the 1790 census are given the birth decade of 175?.

* State(s) of residence. This information appears after a double dash (--) to further identify the subject. Two letter abbreviations are used for states, but they don't always match current U.S. postal abbreviations. The most confusing abbreviation is Ms for Massachusetts (not Mississippi, which is Mp). Well-known countries use three letter abbreviations (Eng, Ger, Can). Other countries are spelled in full.

* Biographical data. Following the location, a very brief biographical statement is sometimes given, if obvious from the source. Generally this is the person's occupation and/or highest military rank or political office.

* Citation. Following a second set of double dashes (--), is the reference to the source indexed. Each title is abbreviated, usually with two words, such as Currier Gen, which represents the full title of the work, in this case Genealogy of Richard Currier of Salisbury, Massachusetts. There is a Key Index in the front of volumes 1, 10, 34 and 54, with a supplement in volume 70. This provides complete bibliographic information, including the number of pages. After the short title is a colon (:) followed by the page reference. The entries for the Boston Evening Transcript are abbreviated as Transcript: followed by the date published and the query number. With these references, it is important to use both the date and number, as the numbers were reused after a few years.

Locating Citations
Of course, an index doesn't do much good if you can't locate the materials it cites. This, perhaps, is the only drawback to AGBI: accessing the sources indexed. However, it is not as difficult as some believe. For starters, it only indexes published sources, so most of them are likely to be found at many different libraries.

Indeed, there are two libraries that are virtually guaranteed to have all the sources indexed. The Godfrey Memorial Library, in Middletown, Connecticut, where the index was created, and the Family History Library. The Godfrey Memorial Library borrowed some of the volumes it indexed in the early years. However, they also operated a microcard (before microfiche) sales operation for some time and likely acquired indexed sources not previously at the library. Of course, some of the sources may have been lost in the past fifty years, but that would be a very rare situation. While most of us cannot travel to Connecticut just to look up a citation, you can contact the library for their assistance. Be aware that this is a small, private library with a very small staff. As a private library, they need to charge for their services, but the fees are quite nominal.

Most Ancestry Magazine readers have heard about the tremendous collection of the Family History Library also. It is correct to assume that they have most of the indexed sources at that library as well. According to the library's own finding aid for AGBI, only fifty-six titles are not at the library. In other words, 93.3% of the titles are at the Family History Library representing about 98% of the entries or more, since the sources the library does not have are generally the older, briefer family histories. Also, virtually all of the sources they have are available on microfilm through the branch family history centers.

Of course, if you do your research at any of the other major research libraries around the country, they will likely have a vast majority of the sources as well. For a handy list of major U.S. genealogical libraries, see Appendix B in Printed Sources: A Guide to Published Genealogical Sources (Salt Lake City: Ancestry, 1998).

AGBI Availability
Now, some of AGBI's sources are already available electronically. More than a year ago Ancestry.com posted the entire 1790 U.S. census on their Web site, while Savage's Genealogical Dictionary of New England was on the first genealogical CD-ROM issued by Automated Archives more than five years ago. Currently it is available from Br¿derbund software. Most of the published marriages for the early colonies are also available at http://www.ancestry.com/ancestry/

DBList.asp?GROUPID=10>. Where practical, the CD-ROM and Internet version of AGBI will point to easily accessible copies of the cited sources.

In his preface, Fremont Rider said that the American Genealogical-Biographical Index "should be the first tool used by every genealogical researcher in seeking an answer for every question." Of course, that was before the advent of the International Genealogical Index,(tm) Ancestral File,(tm) and the Family Finder Index.(tm) However, the addition of these newer files, most of which are also based on early published works, does not mean that AGBI should be ignored. Rider's subsequent statement still rings true almost fifty years later: "In the long run, this use of it will save [the researcher] time and trouble."

AGBI is a more thorough, more comprehensive index to early-published literature than any other index in genealogy. It is the only every-name index to cover almost 850 sources. With AGBI you can find a specific individual in an obscure, otherwise unindexed book, whose surname has nothing to do with the main topic of the book. Now, isn't that something we have all been waiting for?

Kory Meyerink, AG, MLS, editor and primary author of Ancestry's Printed Sources: A Guide to Published Genealogical Records, is a professional researcher, former publications coordinator for the Family History Library, and past president of the Utah Genealogical Association.

Endnotes
1. AGBI's only rival, the Biography Genealogy Master Index, has about eleven million entries, but with its duplication rate, only identifies four million persons. Also, most of those entries pertain to twentieth century persons, and the index is not in one alphabetical sequence.