by Don Pusch, May 1999
In the late afternoon of Monday, January 10, 1921, just after the federal workday had ended, a small fire broke out in the basement of the Commerce Building in Washington, DC. Starting in a carpentry shop at the west end, the fire remained unnoticed until it had reached well into the adjacent hallway and had begun moving toward the southeast end of the building. In that area, outside the protection of the building’s fire-proof vault, were stacked the original population schedules of the 1890 U.S. census. Conditions for their destruction could not have been better. Lacking space in the vault, Commerce Department officials had placed the 1890 records on rows of pine shelves—more than one and a third linear miles, according to one estimate—separated by 20-inch aisles. The 30-year-old volumes, dry and well ventilated, were quickly engulfed.[1]
By the time the first fire company arrived at the Commerce Building, heavy smoke was pouring out of the low basement windows, making it impossible to determine the exact source of the fire’s ignition. In response, firefighters directed hoses indiscriminately into the entire basement area, either through broken windows or through holes that were hurriedly chopped through the first floor. This eventually flooded, to a depth of several feet, most of the basement area. The fire took three hours to control and involved five Washington-area fire companies.[2]
Assessments made immediately after the fire showed that the 1890 population schedules had been heavily damaged, either as a direct result of the fire or from the large quantity of water used in fighting it.[3] Attempts were made in the days following to dry out and salvage remnants of the census; however, those efforts were largely unsuccessful. Although many of the salvaged volumes were retained, a large number of these were considered to be of little value and were disposed of in 1935. Record fragments from ten states and the District of Columbia—less than 1 percent of the 1890 population schedules—are all that currently remains of an enumeration that contained an estimated 62 million entries.[4]
What is left of the population schedules today is contained in National Archives microfilm publication M407, Eleventh Census of the United States, 1890 (3 rolls). At Clayton Library, this film is located in the second-floor microprint area, cabinet 10, drawer 3. Population schedules (or fragments of schedules) for the following geographical areas are imaged on these rolls:[5]
Alabama: Perry County (Perryville Beat 11 and Severe Beat 8).
District of Columbia: 13th, 14th, 15th, Corcoran, Q, R, Riggs, and S Streets, and Johnson Avenue.
Georgia: Muscogee County (Columbus).
Illinois: McDonough County (Mound Township).
Minnesota: Wright County (Rockford).
New Jersey: Hudson County (Jersey City).
New York: Westchester County (Eastchester), Suffolk County (Brookhaven Township).
North Carolina: Gaston County (South Point and River Bend Townships) and Cleveland County (Township #2).
Ohio: Hamilton County (Cincinnati) and Clinton County (Wayne Township).
South Dakota: Union County (Jefferson Township).
Texas: Ellis County (J. P. #6, Mountain Peak, and Ovilla Precinct), Hood County (Precinct #5), Rusk County (Precinct #6 and J. P. #7), Trinity County (Trinity Town and Precinct #2), and Kaufman County (Kaufman).
Personal information collected in the 1890 population schedules was similar to that collected in 1880; however, readability was improved through use of the family schedule, which placed the information on each family member in a column format and provided space at the top of the form for the family’s full address. Several questions were added for the first time in 1890, including three related to residency and citizenship: number of years in U.S., whether or not naturalized, and whether or not naturalization papers had been submitted. This was also the first U.S. census to inquire about the status of veterans, the significance of which will be explained shortly.
If you are lucky enough to have an ancestor listed in the surviving population schedules of the 1890 census, finding that ancestor is relatively simple. A card index to the 6,160 names on the surviving schedules is imaged in National Archives microfilm publication M496, Index to the Eleventh Census of the United States, 1890 (2 rolls). Roll 1 lists individuals with surnames beginning with A through J; roll 2, those with surnames beginning with K through Z.[6] At Clayton, this film can be found in the same drawer as microfilm publication M407. Destruction of the 1890 population schedules was, indeed, a tremendous loss for genealogists. However, at the same time that enumerators were completing the population schedules, they were also collecting data for a special “veterans schedule” that had been recently authorized by Congress to support Civil War pension administration. Fortunately, most of these schedules survived. In accordance with an act approved by Congress on April 21, 1894, the veterans schedules had been transferred to the Commissioner of Pensions and were in a separate repository at the time of the fire in the Commerce Building.[7]
Impetus for the collection of information on Civil War veterans and widows stemmed from problems experienced by the U.S. Pension Office in processing veterans’ claims. By 1890, more than 250,000 Civil War veterans’ claims had been rejected for lack of corroborating evidence, evidence that, in many cases, could only be provided by other veterans who had known the claimant.[8] It was hoped that, by collecting and publishing the names and addresses of surviving Civil War veterans, contacts could be made and corroborating testimony secured. To implement the collection of this data, Congress included appropriate authorization in an act passed March 1, 1889, which provided for a “special enumeration of survivors of the war of the rebellion.”
At line #2 of the 1890 population schedule, enumerators checked whether or not the person listed was a “soldier, sailor, or marine during the civil war (U.S. or Conf.), or widow of such person.” For each person identified as a Union veteran or widow, an entry was then made on the special schedule. Although census takers were instructed to record only Union veterans and widows, these instructions seem to have been ignored in many cases. If one examines carefully the special schedules for some—mostly southern—states, the names of Confederate veterans will often be found. On the veterans schedule for Opelousas, Louisiana, for example, there are more Confederate veterans listed than Union! In general, one will find lines drawn through these Confederate entries and the annotation “Conf.” written at the schedule margin. For the most part, the line-throughs do not significantly detract from the legibility of the entries. If you are searching for either a Union or a Confederate veteran (or widow) who survived to 1890, you should definitely check this special veterans schedule.
The information contained in the veterans schedules is often substantial. The 12 columns of data provide name, rank, company and regiment or navel vessel, dates of enlistment and discharge, length of service, and current post office address. Columns 11 and 12 are reserved, respectively, for information about injuries sustained during the war and for any remarks that the census taker wanted to include. We find, for example, John W. Witson, then living in Kingfisher, Oklahoma, “who had both feet frozen off while serving in Kansas state militia.” James M. Curd of Huntsville, Texas, was “crippled in knee joint from horse falling on it at the battle of Nashville.” James Blith of Washington, D.C., was captured on the gunboat Fanny in October 1861 and held for seven months until paroled. Jules Larquet, Jr., a corporal in the Louisiana Artillery (CSA) was shot in the left arm and taken prisoner at Franklin, Louisiana. The remarks column was sometimes used to explain the lack of other data. For example, we find that Albon E. Bently, whose home was in Hammond, Louisiana, was “absent from state...family knows nothing of his record.” George Smith of Hamilton, Louisiana, “has been in the Louisiana penitentiary.”
The originals of the 1890 veterans schedules are part of National Archives Record Group No. 15, Records of the Veterans Administration. They are imaged in microfilm publication M123 on 118 rolls. At Clayton Library, the film is stored in cabinet 59, drawers 10 and 11. Although veterans schedules were completed for each state in the 1890 census, not all of the 1890 veterans schedules still exist. Sometime between their transfer to the U.S. Pension Office in 1894 and their subsequent relocation to the National Archives in 1943, most of the schedules for the states of Alabama through Kansas were lost. Those states and counties whose veterans schedules do still exist are listed in the microfilm catalog.[9] On the microfilm itself, the schedules are arranged by bundle number, with one roll of film containing the images of one bundle of schedules. The quantity of records varies with the state, ranging from a single bundle for states like New Mexico and Nevada up to 15 bundles for the state of Ohio. There is also a separate bundle for “U.S. Vessels and Navy Yards,” on which will be found the names of many individuals who remained in, or returned to, the U.S. Navy following the Civil War.
Finding individuals in the 1890 veterans schedules may take a little time unless you have access to a state index. Clayton Library owns microfiche indexes for the states of Kentucky, Louisiana, and Texas. These are located in cabinet 68, drawer 1. For these indexes, the compiler has included all names that appear in the schedules, including those of lined-through Confederate veterans. The index gives the veteran’s (or widow’s) name, state, county (or equivalent), city (or other local jurisdiction), the supervisor’s district number, and the enumerator’s district number. With this information, it is fairly easy to find the veteran (or widow) in the veterans schedules. Other indexes have been published in book form, and at least two of these—for Missouri [10] and Tennessee [11] —are available at Clayton. Even if you have no index to guide you, the fact that the schedules are arranged by state and then by county makes the task of directly searching the schedules fairly easy, provided that you know the approximate location of the person’s 1890 residence.
Although destruction of most of the 1890 census left a tremendous gap in the genealogical record, researchers should not be too quick to dismiss this census as a possible source. The remaining 1890 population schedules do contain valuable data, and the Civil War veterans schedules contain an abundance of information not found in any other federal census.