U.S DEPARTMENT OF COMMERCE
BUREAU OF THE CENSUS
Martha Farnsworth Riche, Director
Bryant Benton, Deputy Director
Paula J. Schnelder, Principal Associate Director for Programs
Arthur J. Norton, Chief
Compiled and edited by Richard L. Forstall
The editor acknowledges with thanks the contribution of many Census Bureau staff members in bringing this publication to completion. Special thanks are due to Clarissa Edwards, Joyce Williams, Nina Gressens, Gayle Hubbard, and Tracy Merriman for compilation and checking; Marie Pees, Elizabeth Newman, and Kathleen Morris for computer processing; Edwin Byerly and Donald Dahmann for production supervision; and Campbell Gibson, Ray Bancroft, and John Long for their editorial suggestions and their strong support for the project. The contribution of Edward Lehmann of the National Technical Information Service also is acknowledged with appreciation.
Corrections or suggestions for this publication are welcome. They should be directed to Chief, Population Distribution Branch, Population Division, Bureau of the Census, Washington, DC 20233.
Suggested Citation: Forstall, Richard L., Population of States and Counties of the United States: 1790 to 1990, U.S. Bureau of the Census, Washington, DC, 1996.
International Standard Book Number: ISBN 0-934213-48-8
Population of States and Counties of the United States: 1790 to 1990 is for sale by the National Technical Information Service, U.S. Department of Commerce, 5285 Port Royal Road, Springfield, VA 22161; (703) 487-4650. Price $35 plus handling, refer to order number PB96-119060.
Population of States and Counties of the United States: 1790 to 1990. Electronic Version is available as a text file that can be read by popular microcomputer spreadsheet programs. Price $35 plus handling, refer to order number PB96-500525.
Part I. Introduction and General Explanation iv Sources of Population Data Corrected Census Counts U.S. Total Population American Indian Population State History State Total Populations Census Coverage Census Dates Notes Giving Historical Information on Individual Counties Dates of First Census and of Last Significant Boundary Change Number of Counties and Other State Subdivisions ix County Codes ix Part II. Population of the United States and Each State: 1790-1990 1 Part III. Population of Counties, Earliest Census to 1990 7 Part IV. Historical Dates and Federal Information Processing Standard (FIPS) Codes 189Part I. Introduction and General Explanation
This publication provides the populations of States and their counties according to the twenty-one decennial U.S. censuses conducted from 1790 to 1990. This is the first time in almost 100 years that data for counties back to 1790 have appeared in a census publication.
The taking of a census every 10 years is mandated in Article 1, Section 2 of the United States Constitution, to ensure that the States will elect representatives to Congress in proportion to their population. The United States was one of the first countries in the world to conduct a regular census.
There are four parts to this publication. Part I provides an introduction and explanation of the information contained in the report. Part II is a single table giving the populations of States, for the censuses taken since they were established as Territories or States. So far as possible the States are shown in their present-day boundaries. If the State or its predecessor Territory had a different geographic extent at a census, a note in Part III explains the difference.
Part III consists of tables for each of the 50 States and the District of Columbia, with populations for counties or equivalent areas at each census. In these State tables, after the last column of populations a Notes column provides information on the origin of each county, other names it has had, etc. If this information is too extensive to fit here, the Notes column refers to a numbered footnote, which appears below the table for the State.
Following the population tables, Part IV gives for each State the date of the first census for each county, the date of the census since which the county has not had any significant alterations in boundaries, and the Federal Information Processing Standard (FIPS) code for the county.
Sources of Population Data
The populations in this publication essentially are those reported in each census, except that misprints and other errors made in the original census publications have been corrected. Such errors were fairly frequent in the early censuses, often resulting from mistakes made in adding up the various categories of population for a given county, or in totaling the counties of a State. Most such errors were corrected at the time of the 1870 census, when the earlier published tables were reviewed and their addition rechecked (Census Office, 9th Census, 1870, Vol. I, The Statistics of the Population... (Washington, 1872), pp. xliv-xlvii). Generally the State populations that now appear for 1790-1860 in official publications are those settled upon in 1870. Some exceptions and unusual situations are mentioned in footnotes.
Tables showing county populations back to 1790 appeared in the decennial census publications for 1830 and for 1870 through 1900. The 1910 census, however, showed county populations only back to 1890. In 1920 a table gave county populations back to 1850. The 1930 report, in separate State tables, carried the data back to 1890; 1940 and 1950 each gave only two previous censuses, and 1960 and 1970 each gave one. In 1980, a county table for each State gave populations from 1930 on. In 1990, besides a county table for each State with populations from 1940 on, a single U.S. table (1990 CPH-2-1, table 20) likewise presented 1940-1990 county populations.
In this publication, nearly all populations for counties from 1790 through 1900 agree with those published in the 1900 census (Census Office, 12th Census, 1900, Census reports..., Vol. I, pt. I (Washington, 1901), table 4). In a few cases, decennial numbers for counties and States have been changed from those shown in earlier publications on the basis of new information or to enhance consistency of treatment across all States. Generally such changes are mentioned in footnotes.
For a bibliography of all U.S. census publications through 1945, see U.S. Bureau of the Census, Catalog of United States Census Publications 1790-1945, by Henry J. Dubester (Washington, 1950). This was reprinted as part of Bureau of the Census Catalog of Publications 1790-1972 (Washington, June 1974), which also included publications of the 1950, 1960, and 1970 censuses. Publications of the 1980 and 1990 censuses are included in Census Catalog and Guide, issued annually with cumulations for 1980-84 in the 1984 catalog, and for 1990-94 in the 1994 catalog.
Corrected Census Counts
The misprints and computational errors referred to in the previous section should be distinguished from corrections made after the actual census counts were completed and tabulated. Prior to 1960, census reports only rarely made corrections of counts from earlier censuses. Beginning in the 1970s there was a great expansion in the use of census totals for distributing revenue to individual communities, and as a matter of policy the Census Bureau began to publish even very small corrections. These corrections made after census tabulations were completed are not reflected in this publication.
The majority of corrections were made after local information demonstrated that a municipal or county boundary had not been accurately reflected, so that some persons counted in the census had been credited to the wrong geographic area. Only a minority of corrections involved additions or subtractions of persons not originally included in State and national totals.
Users wishing to find the corrected numbers can do so by referring to the relevant census reports. The 1980 census publications, when citing data for 1970, reflect any corrections in the earlier counts, flagging them with an r symbol. The 1990 publications likewise reflect the same 1970 corrections, and also corrections for 1980; the 1980 corrections are flagged r. The States and counties with corrected numbers can be identified by the r symbol in, for example, 1980 Census of Population, Vol. I, Chap. A, Part 1, United States Summary (PC80-1-A1), table 17 (1970 corrections); and 1990 Census of Population and Housing, Population and Housing Unit Counts, United States (1990 CPH-2-1), table 30 (1980 corrections).
U.S. Total Population
The population totals for the United States in Part II represent the total resident population enumerated at each census in what is now the territory of the 50 States and the District of Columbia. The totals do not include Puerto Rico and other outlying areas.
Census publications have included the populations of Territories in the U.S. total if these were within what is now referred to as the conterminous United States (48 States and the DC). Alaska (covered in the census from 1880 on) and Hawaii (covered from 1900 on) generally were not included in the U.S. totals while they were Territories. However, beginning with the 1960 census, after Alaska and Hawaii were admitted as States in 1959, U.S. official totals have included Alaska from 1880 on and Hawaii from 1900 on.
The census totals for States generally have not included members of the armed forces and other U.S. population located outside the country when the census was taken. This additional population was reported as follows in 1830-1840 and in 1900-1990:
1830 5,318 1840 6,100 1900 91,219 1910 55,608 1920 117,238 1930 89,453 1940 118,933 1950 481,545 1960 1,374,421 1970 1,737,836 1980 995,546 1990 925,845This population category consistently has included U.S. military and naval personnel stationed abroad, or in home waters but not credited to any State; in some censuses it also has included such groups as U.S. merchant seamen, diplomatic personnel, U.S. citizens employed abroad by the government, and citizens abroad on private business.
The 1870 census was widely recognized to have underenumerated the population of the Southern States. In the report of the 1890 census, the 1870 undercount was estimated at 1,260,078, and the correct U.S. total for 1870 as 39,818,449 (Census Office, 11th Census, 1890, Report on Population of the United States..., Part 1 (Washington, 1895), p. xii). This estimate was made by assuming that the White population of the Southern States as a group (comprising Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, South Carolina, Tennessee, Texas, Virginia, and West Virginia) increased by the same percentage from 1860 to 1870 and from 1870 to 1880, and that the Black population of these States likewise increased by the same percentage in the two decades. This approach was seen as justified, because after incorporating the results of this estimate the resulting U.S. growth rates for the two decades were quite similar to the observed growth rate for 1880-1890. However, the hypothesis of equal growth in 1860-70 and 1870-80 seems to make no allowance for above-average losses of population during the 1860s resulting from the Civil War. No estimates of the 1870 undercount were published for individual States, and the 1870 data in this report do not reflect any corrections.
American Indian Population
American Indians generally were not counted in the censuses prior to 1890, unless they were considered to be part of the broader non-Indian society, as evidenced by their paying local taxes and living in settled communities, often alongside non-Indians. The censuses from 1850 through 1880 included estimates of American Indians "retaining their tribal character," sometimes by State or Territory. In 1890, most Indians were enumerated, but were reported separately from the general population, usually by reservation rather than by county. A volume of the 1890 reports dealt exclusively with Indians and included a historical review of earlier estimates (Census Office, 11th Census, 1890, Report on Indians... [Vol. X] (Washington, 1894). Beginning with the 1910 census reports, the 1890 Indian population has been included in the appropriate State totals. In Part III of this publication, the tables report the 1890 Indian population by county when that is possible, and by reservation in Oklahoma and South Dakota; notes explain cases where Indians are included in State totals but not in any State subdivision.
The inclusion of most American Indians in the census beginning in 1890 naturally has some effect on comparisons for the relevant States and counties with censuses before that date. However, even if earlier censuses had enumerated all of the Indian population, much of it would not have been found at its 1890 location at the earlier dates, because of extensive forced removal, migration, and resettling as the settlement frontier advanced and Indian reservations were established.
Since 1790 many new Territories and States have been created, some States have been formed from others, and territory sometimes has been transferred between States. In Part III, just before the numbered notes for each State, two paragraphs appear giving a historical summary for the State. The first paragraph outlines the territorial evolution of the State, and the second paragraph describes the extent of its census coverage.
The territorial description extends from the time the Nation was established in the case of the 13 original States and the States subsequently formed from them, and from the time of acquisition by the United States for the other States. This historical summary is designed to indicate the geographic extent of each State (or Territory) at the time of each census in which it was reported. (The U.S. census of course did not cover any State or Territory prior to its acquisition by the United States.) The discussion of territorial evolution is based in large part on Boundaries of the United States and the Several States, by Franklin K. Van Zandt (Geological Survey Professional Paper 909; Washington, DC, U.S. Government Printing Office, 1976). In the notes on individual States, wording referring to "essentially the current boundaries" is intended to remind users that there may have been other very minor boundary changes besides those documented here.
State Total Populations
Throughout this report, to facilitate comparisons over time, each State is shown as closely as possible in its present-day boundaries, as far back as separate census data are available for counties now included in that State.
For example, West Virginia (formed in 1863-66 from counties originally part of Virginia) is shown in Parts II and III for every census back to 1790, and the former Virginia counties that now are part of West Virginia appear in the West Virginia table in Part III. The Virginia table does not include these counties, and the Virginia totals exclude them. However, the Virginia table does include the present Arlington County and the city of Alexandria, both of which were part of the District of Columbia from 1791 to 1846. Following the Virginia table, the State note for Virginia gives the total population at the early censuses for the State as then defined, including the West Virginia counties in 1790-1860 but excluding Arlington and Alexandria in 1800-1840.
Where areas have been transferred between Territories or States, revision of earlier census totals has been made if one or more whole counties was involved in the transfer, in which case such counties appear under the State of which they are now a part and are included in that State's totals. The State totals always are mutually exclusive; any given county is included in the tables for only one State. For inter-State transfers of territory smaller than a county, the population transferred usually is not available; for a few such transfers, the population affected is specified in the notes.
The U.S. census always has been taken on a resident (de jure) basis, so far as possible counting people at their usual place of residence, not where they happened to be at the time of the census. Since 1900 the census has covered essentially the whole territory of the United States; that is, no significant settlements have been omitted. In all censuses prior to 1900, there were some settled areas within the then territory of the United States that census enumerators did not reach. These areas outside the coverage of the census generally did not have large populations (other than American Indians, most of whom were intentionally omitted prior to 1890), but the gaps in coverage may have resulted in some significant understatement of the population totals of the newest and least populated Territories and States.
In the notes on each State in Part III, the second paragraph specifies when census coverage included all of the State's present area, and summarizes the extent of coverage at the prior censuses.
The early censuses were begun in the summer but most of the recent ones have been conducted in the spring. In principle each census has had a reference date, and attempted to count all persons living in a given area on that date, although often the actual enumeration took place later. The reference dates for the censuses are as follows:
1790 August 2 1800 August 4 1810 August 6 1820 August 7 1830-1900 June 1 1910 April 15 1920 January 1 1930-1990 April 1Because the reference dates have varied somewhat, the periods between censuses have not always been exactly 10 years long.
Notes Giving Historical Information on Individual Counties
In Part III, the Notes column that follows the population columns in the State tables is included primarily to provide information on the geographic origin of the county, if it did not already exist at the date of the State's first census.
Following a *, the Notes specify the county or counties in whose territory this county was located at the census preceding its first appearance. This is not necessarily the same as the county(ies) from which this county was first formed, for example if a number of formations and boundary changes occurred within a decade.
The originating counties generally are listed in the approximate order of how much territory they contributed to the new county. In a few cases the order has been changed to sequence the originating counties by how much population they contributed, when that information is readily available, for example if a sizable town was involved. If an originating county's name appears following a semicolon, it means that it made only a very small contribution to the new county.
If a note is too long for the space available in the Notes column, it appears as a numbered note after the table.
This information on county boundary changes between censuses should be regarded as approximate. It is based largely on Map Guide to the U.S. Federal Censuses, 1790-1920, by William Thorndale and William Dollarhide (Baltimore, Genealogical Publishing Company, 1987). The authors kindly have granted permission for this use by the Bureau of the Census.
The emphasis in specifying originating counties is on population effects. Thus, if all of a county was formed from areas that were Indian lands or unorganized territory at the time of the preceding census, no originating area is specified because such areas would not have been enumerated in the preceding census anyway. If a county was formed mainly from Indian lands with a small portion from another county that was enumerated at the preceding census, only that county is mentioned as an originating area. Those concerned with tracing county territorial changes over time need a map source like the Thorndale and Dollarhide atlas and should not rely on these tables alone.
The term "reported" (for example, Jones was reported as part of Smith) is used to indicate what was presented in the publications of the relevant census. References to what was reported, census coverage, etc., apply only to decennial censuses from 1790 to 1990 (and the 1907 special census of Oklahoma). Additional population data for counties may be available in State or local census publications or from censuses taken by the States or predecessor colonies before 1790; these are outside the scope of this publication. For pre-1790 data, see Evarts B. Greene and Virginia D. Harrington, American Population Before the Federal Census of 1790 (New York: Columbia University Press, 1932; reprinted Baltimore: Genealogical Publishing Company, 1993). State censuses are listed in U.S. Bureau of the Census and Library of Congress, Census Library Project, State Censuses..., an annotated bibliography... prepared by Henry J. Dubester (Washington, 1948); and Ann S. Lainhart, State Census Records (Baltimore: Genealogical Publishing Company, 1992).
In the tables in Part III, if "---" appears for a census year, the county did not exist (or was not enumerated separately) at the date of that census. The tables always give the county information in terms of what the census showed. Counties shown in the census with no population (generally counties not yet organized or in process of formation) are shown with "---" unless an earlier census had reported a population for them, in which case they are shown with a population of "0" (zero).
Prior to 1840, the census publications regularly listed the few large towns of the time separately from their counties, and sometimes as though they were counties themselves, or independent of any county. In fact, this was true only of New York City, which corresponded to New York County; after 1854 Philadelphia also corresponded to Philadelphia County. In these tables, generally cities have not been shown separately from their counties unless they were actually independent of them (for some exceptions see the notes for the District of Columbia and Virginia).
Also appearing in the Notes column, preceded by "to", are the destinations of counties that have been abolished. Destinations are given on an analogous basis to origins -- they are the counties reported in the following census in which the territory of the abolished county was located at the time of that census. County names that were substantially different from the present name are mentioned in the Notes column, specifying the period during which the other name was used in the census. Minor differences (such as La Salle vs LaSalle or New-York vs New York) are mostly not mentioned. Where a county name has changed, references to the county for the period prior to the change generally use the earlier name with the present name in parentheses.
County names that were used only during an intercensal period (such as 1872-1878) are not specified, or referred to in statements about county origins.
Dates of First Census and of Last Significant Boundary Change
These dates appear in Part IV. The first date is that of the first census in which the county or State appears with a separate population; it also can be derived from the population tables in Parts II and III. The second date is that of the census since which the State or county has had no significant territorial change, that is, a boundary change large enough to have a significant effect on the county's population as of the preceding census. However, changes that involved a sizable area (generally more than 25 square miles) are treated as significant even if they involved little population. Typical of changes treated as not significant are adjustments of a boundary to follow a survey line, replacement of an irregular line along a drainage divide with straight-line segments, exchanges of small uninhabited islands in a boundary river, and changes involving only water area.
For States, the date since which there has been no significant territorial change refers to the area represented by the State population total given in the table -- the total of the counties now included in the State. For example, the date given for both Virginia and West Virginia is 1860, because the population of the present territory of each State can be determined for 1860 by adding up the counties that formed West Virginia in 1863-66 and subtracting them from the 1860 total for Virginia as then defined. Prior to 1860 this cannot be done exactly, because the present Virginia-West Virginia State boundary did not correspond to county boundaries in 1850 or earlier. Thus, for these two States comparisons of the 1850 or earlier populations with populations for 1860 or later are affected by significant geographic changes.
For some States, the date since when there has been no significant change is for a census prior to complete census coverage of the State. This date has been selected on the basis that the areas not enumerated had very few if any non-Indian residents, so that even if that census had covered the State completely, the total population would not have differed significantly.
The evidence checked for these determinations consists of the Thorndale and Dollarhide atlas, all notes about county changes since 1920 reported in decennial census publications, and the 1976 U.S. Geological Survey publication already cited.
All statements about county origins and first census dates refer to the inclusion and boundaries of the county in decennial census publications, and may not correspond to the county's legal history, since, for example, new counties sometimes were not separately reported in the next census, and some legal changes in county boundaries were not yet recognized when the census was conducted.
In a few cases a census population is available for a county that was not created until after that census, and is given in the notes.
Changes in counties since the 1990 census are not reflected.
Number of Counties and Other State Subdivisions
The counts given below are the totals of counties and other subdivisions (parishes, independent cities, Indian reservations, etc.) reported in each census and shown with a population in the tables in Part III. They include a few cases where the population given in the table is 0 (zero). The count for 1860 includes 1 each for Colorado and South Dakota (Dakota Territory), which were reported as single units in that census rather than by counties.
1790 292 1860 2,080 1930 3,110 1800 419 1870 2,295 1940 3,108 1810 574 1880 2,570 1950 3,111 1820 759 1890 2,813 1960 3,133 1830 988 1900 2,862 1970 3,142 1840 1,279 1910 2,962 1980 3,137 1850 1,623 1920 3,076 1990 3,141The formation of new counties in the United States proceeded at a rapid rate from 1790 until 1860. The 1850s witnessed an increase of 457 in the number of areas shown in the tables, the largest of any decade; however, the largest relative increase (more than 40 percent) took place in the 1790s.
After 1860 the number of counties continued to grow, but at a somewhat slower rate. Since the 1920s, changes in the number of State subdivisions reported in the census have been due largely to the creation of new independent cities in Virginia and to increases (or decreases) in the number of subdivisions reported for Alaska. Only a few actual counties have been created or abolished since 1930.
In all, the State tables include 3,279 counties and other subdivisions, comprising the 3,141 entities reported in 1990 and 138 reported only in one or more prior censuses.
The Federal Information Processing Standards (FIPS) code issued by the National Institute of Standards and Technology, U.S. Department of Commerce, is given in Part IV for each county that currently exists. Each county has a 3-digit code within its State, so the full county identifier has 5 digits, including the 2-digit FIPS State code.
Codes also are given for counties that no longer exist. The code system now adopted by FIPS originated in the private sector about 1930, and counties abolished between 1930 and 1990 have the code in Part IV that was originally assigned to them when they were in existence. Counties abolished before 1930 have the preceding county code with a decimal added, to place them in the proper alphabetical position. Decimals have been used because it is often impossible to assign additional codes in the FIPS system while retaining both alphabetic sequence and a 5-digit numeric format; for example, in Nevada a former county alphabetically precedes current county 001.