"First in the Path of the Firemen"
The Fate of the 1890 Population Census, Part 2
![]() |
Confederates and veterans of earlier wars often show up on the special schedule for Union veterans. (NARA, Records of the Veterans Administration, RG 15) |
The Special Enumeration of Union Veterans and Widows
Often confused with the 1890 census, and more often overlooked or misjudged as useless, are nearly seventy-five thousand special 1890 schedules enumerating Union veterans and widows of Union veterans.(27) Nearly all of these schedules for the states of Alabama through Kansas and approximately half of those for Kentucky appear to have been destroyed before transfer of the remaining schedules to the National Archives in 1943. Nearly all, but fragments for some of these states were accessioned by the National Archives as bundle 198. Many reference sources state or speculate that the missing schedules were lost in the 1921 fire. The administrative record, however, does not support this conclusion.
The Pension Office requested the special enumeration to help Union veterans locate comrades to testify in pension claims and to determine the number of survivors and widows for pension legislation. Some congressmen also thought it scientifically useful to know the effect of various types of military service upon veterans' longevity.(28) To assist in the enumeration, the Pension Office prepared a list of veterans' names and addresses from their files and from available military records held by the War Department. The superintendent of the census planned to print in volumes the veterans information (name, rank, length of service, and post office address) compiled from the 1890 enumeration and place copies with libraries and veterans organizations so individuals could more easily locate their fellow veterans.(29)
Question 2 on the general population schedules inquired whether the subject had been "a soldier, sailor, or marine during the civil war (United States or Confederate) or widow of such person." Enumerators were instructed to write "Sol" for soldier, "Sail" for sailor, and "Ma" for marine, with "U.S." or "Conf." in parentheses, for example, Sol (U.S.) or Sail (Conf.). The letter "W" was added to these designations if the enumerated was a widow.(30) According to enumeration instructions, if the veteran or widow responded "yes" to Union service, the enumerator produced the veterans schedule, marked the family number from the general population schedule, and proceeded to ask additional service-related questions.
The upper half of each page on the veterans schedules lists name, rank, company, regiment or vessel, date of enlistment, date of discharge, and length of service. The lower half contains the post office address, any disability incurred in the service, and general remarks. The question on disability was included because many veterans claimed pensions, under an 1862 act, based on service-related disabilities.(31) The "General Remarks" column usually provides the most colorful, anecdotal, and meaningful information on the schedules.
Although the special enumeration was intended only for Union veterans of the Civil War and their widows, enumerators nevertheless often listed veterans and widows of earlier wars as well as Confedederate veterans.(32) Veterans of the War of 1812 are sometimes listed, and there are especially numerous entries for Mexican War veterans. Susan Arnold of Pennsylvania was listed, though her husband died in New Orleans coming home from the Seminole War (1828-1833).
John Yost is listed as serving in the French army under Maximilian. Several sources note that Confederates are inadvertently recorded in this enumeration; actual study of the records reveals that there are some Confederates listed for every extant state (excluding the fragments on bundle 198). Schedules consisting nearly entirely of Confederates are not altogether uncommon, especially in extant schedules of Southern states.(33) The Confederate names are sometimes crossed out or marked as errors (presumably by census supervisors), but the information is usually readable.
Listings for widows can also provide telling insights to the veteran's service, her life or remarriage, even their relationship. Eliza Smith of Pennsylvania was simply listed as the "grass widow of a soldier." A Pennsylvania widow living at the Home for the Friendless claimed she knew nothing of her husband's fate but thought him dead. A Wyoming widow remembered no particulars, only that her husband wore a "blue coat." Enumerators were instructed to list the widow's name above the name of the deceased veteran and fill out the record of his service during the war but list her present post office. Remarried widows were listed in this manner with their new surname. Dependent mothers are also sometimes listed, as in the case of Pate Halberts of Ohio, who knew little English, but enough to tell the enumerator her son died in Andersonville.(34)
Enumerators often noted the battle or circumstances in which a death or disability had been incurred, such as "shot dead at Gettysburg, July 3rd 1864" or "lost right arm at Resaca." They also had the unenviable task of diagnosing the described ailments such as "harte disease," "indestan of stomic," and "thie woond." Men recounted the loss of eyes, ears, and appendages. They told of falling from and being trampled by horses, being crippled on trains "wrecked by rebels," and going insane from the "noise of war." Allan Hobbs of Salt Lake, Utah, claimed partial paralysis of his feet from freezing in Libbey Prison, and George Search of Baltimore claimed his constitution was broken after six months at Andersonville.(35) The perils of bad wartime medicine are evident as well. Many reported blood poisoning or crippling from an impure vaccination. One widow told the enumerator her husband died by eating too much morphine. Without a doubt, however, the most widespread permanent disabilities reported by the 1890 veterans were diarrhea (spelled in many creative ways) and piles.
The schedules may reveal anecdotal or unique information. They sometimes briefly chronicle an individual's military career, like that of William Martin of North Carolina, who rose from private to general. Josiah Dunbar's widow claimed her husband was one of the first, if not the first, to enlist in his county, and Bernard Todd remembered he had played in Custer's band at the Appomattox surrender. Ohioan James Stabus admitted he had been captured and paroled by the notorious raider John Hunt Morgan. Jackson Mitchell of Pennsylvania said he was born a slave and compelled at first to serve in the Confederate army. Others proudly noted their service in the U.S. Colored Troops, in specialized units, or as spies. Dennis Arnold of Allegany, Maryland, said he "would go again tomorrow." The schedules may even provide clues about enlistment under "secret or varied names." For example, Samuel Polite, Marcus Moultair, and August Gadson of Sheldonship County, South Carolina, all reported they had enlisted in the Union Army under "secret" names, which the enumerator listed according to instructions, with lawful name preceding the alias.(36) In some instances, the pension certificate number is provided. At least two Missourians were listed on the veterans schedule and overlooked in the general population census.(37)
A less noble side of some veterans is revealed, as well. Some individuals falsely claimed to be veterans, hoping to receive government pensions. "Deserter" is entered in the remarks column often enough, although it is often unclear by whom this information was provided. William Robertson of the Oklahoma Territory was found "sick on drink when visited." One North Carolina enumerator disgustedly reported on a case of pension fraud, noting: "Brown and Branvell were both deserters from the Confederate Army. Brown now draws a pension from 'Uncle Sam' under the plea that he has scurvy of the mouth."(38)
At the completion of the 1890 enumeration, the special schedules were returned with a preliminary count of 1,099,668 Union survivors and 163,176 widows. A large number of schedules were found to be incomplete, and many veterans had been overlooked. The Census Bureau sent thousands of letters and published inquiries in hundreds of newspapers hoping to acquire missing data. As appropriate, corrections and additions were made to the schedules. The initial work of examining, verifying, and classifying the information was suspended in June 1891, awaiting congressional appropriation for publication of the veterans' volumes.(39) During that same period, anticipating the publication, the bureau began transcribing information from the schedules onto a printed card for each surviving veteran or widow, later to be arranged by state and organization. No fewer than 304,607 cards were completed before this work was also halted. These cards do not seem to be extant, nor does there appear to be a final record of their disposition. Some cards may have been placed in individual service files.(40)
The veterans' publication seemed doomed. Adequate funding was not available, many considered other census work more pressing, and searches for information in the manuscript veterans schedules were cumbersome and costly. In 1893 Carroll Wright, then in charge of the census, argued that too much time had already passed to make any veterans' publication accurate; the general schedules provided an approximate number of Union veterans and widows. He recommended these special schedules be transferred to the Pension Office or the War Department, and in 1894 Congress authorized their transfer to the Commissioner of Pensions for use in the Pension Office and transferred them "shortly thereafter."(41) The schedules were arranged and stored in bundles, generally alphabetically by name of state or territory, and numbered sequentially. In 1930 legal custody of the schedules passed from the Pension Office to the newly formed Veterans Administration, where they remained until accessioned by the National Archives in 1943 as part of Record Group 15.(42) Clearly these schedules were maintained apart from the population schedules and used for different purposes in a different location. Moreover, no reporting from the fires of 1896 or 1921 mention these schedules among the damaged series. It seems nearly impossible they were involved in the Commerce Building fire in 1921.
The extant schedules are available for part of Kentucky through Wyoming, Lincoln Post #3 in Washington, D.C., and selected U.S. vessels and navy yards. The schedules are generally arranged by state and county and thereunder generally by town or post office address. The bundle containing schedules for Oklahoma and Indian Territories are arranged by enumeration districts. Although veterans schedules from the states of Alabama through Kentucky (part) are not known to be extant, bundle 198 on roll 118, "Washington, DC, and Miscellaneous," also contains some schedules for California (Alcatraz), Connecticut (Fort Trumbull, Hartford County Hospital, and U.S. Naval Station), Delaware (Delaware State Hospital for the Insane), Florida (Fort Barrancas and St. Francis Barracks), Idaho (Boise Barracks and Fort Sherman), Illinois (Cook County and Henderson County), Indiana (Warrick County and White County), and Kansas (Barton County). All of the accessioned schedules have been microfilmed and are available as National Archives Microfilm Publication M123 (118 rolls).(43)
There is no comprehensive index to the 1890 special enumeration, but indexes to some states or specific areas have been prepared by various publishing companies and private groups. These special enumerations are well worth examination. Although it may be time-consuming to wade through an unindexed county, the information rewards can be priceless and uncommon. Few series in the National Archives rival this one for anecdotal information and local color.
![]() |
The Washington Herald and other newspapers reported the events of January 10, 1921, and decried the loss of valuable records. |
Of course, there is no real substitute for the lost 1890 or any other comprehensive federal census. Records relating to elections, tax or criminal legislation, impending statehood, war, economic crisis, vital statistics reporting, and other local events may provide alternative information sources. There are some state and territorial censuses available for the years near 1890. For example, the federal government assisted the states and territories of Colorado, Florida, Nebraska, New Mexico, and the Dakotas in an 1885 census. There is an 1890 territorial census for some areas in Oklahoma.(44) The 1890 poll lists or "Great Registers" for selected counties in Arizona and California are extant and available at the respective state archives. The Arkansas Genealogical Society has sponsored a statewide program to reconstruct the missing 1890 federal census using tax and other local records. Ann Lainhart's State Census Records (Genealogical Publishing Co., Inc., 1992) includes state-by-state listings of census resources, including some census and other alternatives for the 1890 federal census.(45) Researchers are encouraged to contact state and local repositories to inquire about alternative resources and verify records arrangement, availability, and content.
The loss of the 1890 schedules and absence of part of the
special veterans enumeration are especially painful information
losses for which there is no real balm. However, all of the
federal censuses (pre-1920) might have been destroyed in that
1921 fire, especially if it had consumed the entire Commerce
Building. It is a wonder now, as it was to the secretary of
commerce at the time of the fire, that such a large number of
records were saved.(46) Most researchers in federal records are
frustrated at some point by gaps in records, lack of indexes and
description, poor quality images, or unknown records provenance.
More than 150 years passed between the signing of the Declaration
of Independence and the establishment of a U.S. National
Archives, however, and the nation paid a high price for this
delay. Critical records succumbed to war, fire, flood, theft,
moves, agency reorganization, administrative error, improper
filming, ignorance, apathy, and the ravages of time. It is really
quite remarkable that so many valuable records are extant and
available for research. The tragedy of the 1890 census remains a
constant reminder of the necessity for a vigorous National
Archives and unrelenting vigilance about the historical record.
The Fate of the 1890 Population Census, Part 1
The Fate of the 1890 Population Census, Part 3 (Notes)
Articles published in Prologue do not necessarily represent the views of NARA or of any other agency of the United States Government. |