The Atlases of A.J. Johnson

(This is an updated version of an article published in the Portolan, No. 49, Winter 2000-01)

Background

Alvin J. Johnson entered into the business of publishing atlases in 1860.  Having previously been a book canvasser who sold maps and atlases for the well-known map and atlas publisher, J.H. Colton, Johnson was introduced to the profitability of atlases as a reference book.  Whether he originally planned to start his book publishing business with atlases or not is unknown.  It does appear that, when he came to New York City in 1857, he began to support Colton in the publishing of his atlases and maps.  In 1859, Colton’s General Atlas was published by the firm of Johnson & Browning, and the next year he published the first edition of his own atlas, titled, Johnson’s New Illustrated (Steel Plate) Family Atlas, With Descriptions Geographical, Statistical and Historical.  The early editions of the Family Atlas, as it is most commonly referred to, was comprised of maps developed by the Colton firm, and the title page announces that the publishers, Johnson & Browning, are “formerly (successors to J.H. Colton and Company…”  However, rather than successor, Johnson became a competitor of the Colton firm, and the Family Atlas became a competitor to the atlases of both Colton and S. Augustus Mitchell, the other prominent atlas publisher of the day.  (For a more comprehensive discussion of the relationship between Johnson and Colton go to the Alvin J. Johnson and his role in 19th Century Map Making in America page of this website).

 

Editions

The Family Atlas was produced in editions dated every year from 1860 to 1887, with two known exceptions, 1875 and 1876, years for which no atlases have been discovered.  These for the most part were not true editions, but the years printed on the title pages can be more accurately described as production dates for an individual atlas.  Rather than wait until a set time every year to start printing a new edition which includes all of the changes made over the last year, in Johnson’s atlases they began using a newly updated or designed map as soon as it was available.  More than likely, they started using the new map when they ran out of the old variation.  As a result, several atlases with the same date on the title page may have different states/variations of the same maps depending on the time of the year the atlas was assembled.  This was especially true in the early half of the 1860's, and less likely in the 1880's.  There were no changes at all between 1884 and 1887, the years between when Johnson died and the company closed.

There were only four times over the 27 years that the atlases were published in which all the maps changed at the same time; there were two other times when all but one map in one instance and three maps in another (and these changed either one atlas before or later).  For the most part the creation of a “psuedo-edition” was related to changes in pagination due to added maps, changes in the name of Johnson’s company and the related publisher attributions on the title page and maps, and changes in whether the back side of the maps were either blank or printed with gazetteer pages.  These same items, along with geographical changes, map title changes, and other incidental changes, are also used in determining the various states/variations of the individual maps.  A discussion of the most prominent factors used in determining pseudo-editions of the atlases and map states/variations follows.

Publisher Attribution Changes 

While A.J. Johnson was the major publisher of the Family Atlas throughout its existence, the name of his company changed a number of times.  Each atlas and each map within it included an attribution which identifies the name of the Johnson’s company at that time.  This was one of the most consistent changes found in the maps, and in each atlas, every map included an attribution of the publisher, “Published by...”, identical to that on the title page of the atlas.  While one might find an atlas in which the maps had two different versions of the border, on each map the attribution of the publisher were always the same. 

From 1860 to early 1863, the name of the company which published the atlases was Johnson & Browning.  Ross C. Browning was an agent who worked in Johnson’s book canvassing business in Cleveland prior to moving to New York.  Some of the 1860 and 1861 atlases have Richmond where Browning appears to have lived at least in 1860 as the city of publication.  During those same years atlases with identical maps were published with New York as the city of publication.  Atlases from all years following were published only in New York.  It is not clear if the first issues was published first in Richmond and then in New York, or in Richmond and New York at the same time.  Only a few atlases have been identified so far with the Richmond designation, indicating that it was a short-lived practice and a testimony to Browning’s role, most likely as a financier and sales representative, in the publishing of the atlases.

In 1863, Johnson and Ward became the name of the company publishing the Family Atlas.  Benjamin P. Ward, like Browning had been an agent for Johnson, who appears to have been brought in as a financial backer, when Browning left.  Even while receiving attribution as a joint publisher with Johnson, Ward never really was part of the company, and remained out in the field as a major agent in charge of selling the atlases by subscription in the west (Cleveland and Chicago).  The company remained Johnson and Ward until 1866, after which Johnson bought out Ward’s interest in the business.

Johnson ran the company himself from 1866 on, and starting in early 1866 the atlases and maps simply attributed as, “Published by A.J. Johnson.”  This was about the same time that Johnson’s relationship with J.H. Colton changed.  Prior to this time, the publisher attribution on the atlas title page had recognized the role of Colton in the development of the maps in the atlases, and both the Johnson and Browning and Johnson and Ward companies were both noted as being “successors to J.H. Colton”.  As Johnson became the sole publisher of the Family Atlas, he no longer gave Colton credit for a role in the development of the maps, which was appropriate because by that time most of the original Colton-derived maps had been replaced by maps drawn by Johnson’s company.

The company remained A.J. Johnson, Publisher, until 1879.  In that year, Johnson’s son was brought into the business and the attribution on the atlas title pages and the maps became “A.J. Johnson and Son.”  This lasted only a short time, and by 1881, the name had changed to “A.J. Johnson and Co.”, even though his son remained with the business and ran it after his father’s death.  The Johnson firm published under this name until it closed in 1887.

The publisher attribution is one of major items that change on the maps that aid in state/variation identification.  Although the name of the company changed five times, there are six versions of the publisher attribution.  This occurs because there were two versions of the Johnson and Ward attribution.  Starting in 1860, both the Johnson and Browning and the Johnson and Ward attributions were printed in plain block letters.  A more elaborate double-faced outlined font for the Johnson and Ward publisher attribution began being used in late 1862.  By early 1863, all but three of the maps had made this change, and those three started using the double-faced outlined font in 1864. This same double faced type was used for the publisher attribution through 1887.  The following chart demonstrates the history of publisher attributions throughout the publishing life of the Family Atlas.  There was a third variation that was used in the 1861 United States map from 1861-64 when the new 1864 United States map was published, which utilized the double-faced outlined letters.

Publisher Attribution	Font	Years Used
Johnson & Browning	Plain Block Letters	1860-1862
Johnson & Ward	Plain Block Letters	1862-1863
Johnson & Ward	Double-Face Outlined Letters	1862-1866
A.J. Johnson	Double-Face Outlined Letters	1866-1878
A.J. Johnson and Son	Double-Face Outlined Letters	1879-1880
A.J. Johnson and Co.	Double-Face Outlined Letters	1881-1887

Borders

Perhaps the most well known changes in the maps of the A.J. Johnson atlases are the map borders.  Over the years the atlases were published, there were four different borders that appear on the maps.  The borders were used in various years and can be used to help identify the various states/variations of the individual maps.  The first border was used starting in 1860.  The second border was first appeared in 1863.  During that year, 40 percent of the maps began to use the second border; the other 60 percent adopted the second border during 1864.  The second border was used through 1869. All of the maps began to use the third border in 1870.  The fourth border is a variant of the third border, yet distinctly different.  It was used on all maps from some time in 1883 through 1887, however, 12 maps also used the fourth border for one year extra prior year, 1880.

Map Title Changes

Several of the maps have a change in title one or more times, while the map itself remained basically the same. The reason this was done was to include new states of the union as they were established.  This occurred most often with the maps which included several central or western states, but it also happened in the east when West Virginia was established in 1863.  The following chart details the title changes and new states established for those maps in which title changes occurred on the same basic map.

Original Title (date)	Name(s)  Added	Year Added	Map State	Year Established
Nebraska & Kansas (60)	Dakota Colorado Idaho Montana	1861 1861 1863 1864	3 3 7 9	Dakota Territory 1861 Colorado Territory 1861 Idaho Territory 1863 Montana Territory 1864
Nebraska, Dakota, Idaho, &    Montana (65)	Wyoming	1869	4	Wyoming Territory 1868
Washington & Oregon (60)	Idaho	1863	8	Idaho Territory 1863
California, Territories of New     Mexico & Utah (60)	Arizona Colorado Nevada	1863 1863 1863	8 8 8	Arizona Territory 1863 Colorado Territory 1861 Nevada Territory 1861

Page Numbers and Reverse/Verso Sides

The page numbers and contents of the reverse side are second most consistent items that define “editions” of the Family Atlas.  Pagination of the maps changed frequently, usually due to the addition of a new map or subtraction of an old one.  When these changes took place at the beginning of the atlas, such as the addition of the United States map in 1861 or the deletion of the New England map in 1862, all of the maps changed page number.  When a map was added closer to the back of the atlas, only the portion of the maps after that in the atlas changed page number.

Johnson and other atlas publishers of his era advertised the gazetteer part of their atlases with as much gusto as they did the maps.  The title page of the early editions proudly announced that the atlas included “...descriptions geographical, statistical, and historical, including the latest federal census, a geographical index and a chronological history of the Civil War in America, and the existing religious denominations in the world.  Text by Richard Swainson Fisher.”  It is interesting to note that Fisher also wrote the “Descriptions” for Colton’s atlases of that era.  When the atlases were first published they included a “Descriptions” sections on Physical Geography, and several sections on Descriptive Geography of various areas of the world, and a Geographical Index; an Appendix to the Geographical Index was added in 1864.  These were updated on occasion, and after 1867 were no longer attributed to Fisher.  Regardless of the fact that Johnson kept the maps up to date from year to year, the Geographical Index and the Appendix to the Geographical Index did not change from the 1860's through 1887.  In 1870, there were major changes to all of the “Descriptions” other than the Geographical Indexes; they were from that time on attributed to A.Guyot, with a section on religions by Rosewell D. Hitchcock.   

From 1860 to early 1863, these sections were in separate areas of the atlas from the maps, and the backs, or versos, of all of the maps were blank.  At some time during 1863, the “Descriptions” began to be printed on the reverse side of the maps.  This continued through 1872, after which the “Descriptions” were removed from the reverse of the maps and returned to separate sections of the atlas.  During the years when the “Descriptions” were on the reverse of the maps, they were occasionally moved around creating unique states/variations of the various maps. When these changes occurred, they affected every map in the atlas, defining a pseudo-edition of the atlas.  There were several sections that regularly appeared on the reverse side of the United States maps: a Geographical Index, the Appendix to the Geographical Index, History, Physical Geography, Animals.  Two other sections show up for two or three years: History of Mexico and History of West India.  Other sections appear on the reverse of the maps of other areas of the world.  The following chart demonstrates the contents of the reverse sides of the United States maps used from 1863-1872.

Reverse Side Section (Code)	Years Used
Physical Geography (PG)	1869-1872
Natural History: Interesting and Curious  Animals (AN)	1870-1872
Geographical Index (GI)	1863-1872
Appendix to Geographical Index (AGI)	1864-1872
Geography of North America(GN): Historical and Statistical View of North America(HN)	1863-1872
Historical and Statistical View of the United States(HU)	1863-1870
Historical and Statistical View of Mexico and Central America(HM)	1864-1868
Historical and Statistical View of the West Indian Islands or Columbian Archipelago(HW)	1867-1868

 

Geographical Detail Changes

An extremely important feature of the maps of the Family Atlas was the fact that the geographical detail was updated on a regular basis. As discussed in a later section of this article, it was important for sales promotion that the atlases be as up to date as possible.  As a result, the maps in Johnson’s atlases, as well as Colton and Mitchell’s, can be used to track the growth of the United States, including new towns and cities, new counties, new states and state boundaries, and new townships.  Most of the map states/variations detailed in this study have accompanying changes in the geographical detail, along with the more easily recognizable changes in the map borders, pagination, publisher attribution, and contents of the reverse sides.  There are even a few map states/variations for which the only defining quality is one or more geographical changes.

Counties.  The most easily recognizable geographical changes are the development of new counties.  Using Knox’s American Counties as a guide, every new county that was created in the United States between 1860 and 1882 was added to any map in which that county can be seen (whether or not it is in one of the titled states).  For the most part, new counties showed up on the maps within two years of their establishment. 

States.  New states tended to be added much more expeditiously.  West Virginia, for example, was introduced on the maps during the year of its division from Virginia.  It was obviously added in a hurry, and early states/variations of maps on which West Virginia was shown, often have the name “Virginia” placed so that it infringed on West Virginia’s territory.  The Virginia map was redrawn in 1864; in this new map the name “Virginia” is placed so that it no longer infringes on the state of West Virginia.  Other new states and territories were entered similarly as fast as they were established.  For the most part, territories were shown as if they were states.  Arizona was shown as part of New Mexico until 1863 when it formally became a territory.  Wyoming was shown several years before it formally became a territory. The following chart demonstrates the year that new territories and states appeared the atlases.

State	Date Territory Estab.	Date Statehood Estab.	Date Appearing in Specific Maps
West Virginia		1863	United States; Pennsylvania & Virginia/both in 1863
Colorado	2/1861	1876	Nebraska & Kansas; California/both in 1861
Nevada	3/1861	1864	California/1861
Dakota	3/1861	1889	Nebraska & Kansas/1861;Minnesota & Dakota/1860
Arizona	2/1863	1912	United States/shown wrong in 1862, correct in 1863; California/shown wrong in 1860, correct in 1863
Idaho	3/1863	1890	United States; Nebraska&Kansas; California; Washington & Oregon/all in 1863
Montana	5/1864	1889	United States; Nebraska & Kansas;Washington & Oregon/all in 1864
Wyoming	7/1868	1890	United States/1866; Nebraska & Kansas/1864

Railroads.  The growth of the railroads in the United States is equally well represented on the Colton/Johnson and Johnson maps.  Railroads are depicted in three ways in the Family Atlas.  There is no legend, so it is not clear exactly what the different markings mean, however, one can guess.  Railroads marked with what are called “railroad markings” (solid lines with cross hatching) in the Johnson U.S. Map Project, and abbreviated as “RR(m)”, appear to be established railroads.  Those marked with solid lines, abbreviated “RR(s)” in this project, appear to be either proposed railroads or those under construction, or both, as a RR(s) often becomes a RR(m) in later states/variations of a map.  At times, a RR(s) will slowly evolve into a RR(m) in stages over several years.  On a rare occasion, some railroads are depicted with a dashed line, abbreviated as RR(d).  It appears as if the RR(d) is proposed railroad, and they are just as likely to disappear in later editions than they are to become a RR(s) or RR(m).  When research on the development of the railroads done by the Johnson firm discovered that a railroad had been placed on a map in error, or had been abandoned, they were removed from the maps.  An example of this is found on the North and South Carolina map of 1860, where in 1863, on state/variation 8 of the map, a railroad present on early map states/variations, from Greensboro, North Carolina to Danville, Virginia, disappears. 

For the most part, railroad changes on the maps are not the sole defining feature of map state/variation and they have not been comprehensively described for all states/variations of all maps, and they are not all listed in the Map Identifier data sets.  (Researchers who want more information on the railroad changes should read the section on Archival Data in this website.)

Cities.  Similar to the railroads, during the era of the Family Atlas there was a significant growth in the number of cities and towns.  As maps changed, they represented this new population growth.  There are no instances, however, where a new city or town was the sole change defining a new map state/variation.  Unlike Colton, who included numerous city maps in his atlases, Johnson used only a few.  Starting with Washington, DC in 1860, New York City was added in 1862, with Boston and Philadelphia not being added until 1870 and 1873 respectively.  These city maps demonstrate the growth in these cities, including new neighborhoods, roads, trains, buildings, parks, and ferries. 

Picture Changes 

During the years that Johnson used Colton derived maps in the Family Atlas, Johnson also often added pictures as decorations for the maps.  Almost all of these pictures can be found on large Colton wall maps, and it is clear that Johnson’s relationship with Colton included the use of these pictures.  Johnson added these pictures to some of the maps that were identical to those maps used by Colton in his own atlases, but Colton published them without the pictures. These pictures become important in the Johnson maps, not only because they make the maps pleasingly decorative, but also because a particular picture would at times appear and disappear from the map, and/or its placement might have changed.  Sometimes the printing of the picture title even changed.  For the most part, the changes regarding the pictures follow no pattern, and often make little sense.  The changes are very useful, however, in determining the state/variation of a map.

One example comes from the 1860 map of Delaware and Maryland.  On the first state/variation of the map, there is a picture of the U.S. Capitol building under the title.  Later in 1860, on state/variation 2, four other pictures were added: “Harpers Ferry” over the title, and “General Post Office,” “Treasury Building,” and “Patent Office” all across the top of the map.  In 1862, the Capitol and Harpers Ferry pictures disappear in map state/variation 4, being replaced by an inset map of the District of Columbia.  By late 1864, map state/variation 11, all of the pictures are gone.  In another example, there is a picture of the Washington Monument on the 1860 map of the District of Columbia.  When it first appears, in 1860, map state/variation 1, there is shading around the spire of the monument.  In 1862, state/variation 2, the shading disappears, only to show up again in 1864, state/variation 6, and disappear the next year, 1865, state/variation 7, and return again in 1867, state/variation 9.

On the North and South Carolina map of 1860, the size of the font of part or all of the picture’s title changes in size.  On the Missouri and Kansas map of 1860 there are three pictures, two of which keep switching position with each other, between 1860 and 1865.  It almost seems as though, when a map was going through a printing, the individual who assembled the plate had to decide each time where each picture went on each map; and, sometimes it was done differently than others.

Other Incidental Changes 

Before Johnson began to produce his own maps, the Family Atlas used maps previously drawn and used by Colton.  Several of the Johnson atlas maps were created by the folio size printing of part of a larger wall map.  To these maps, Johnson added borders, and sub-borders.  The change in the borders is discussed above, but there were also changes in the sub-borders where the longitude and latitude of the area depicted on the map was printed.  The font and placement of the statements “Longitude from Washington” or “Longitude from Greenwich” would at time change on various maps.  Indeed, at times the longitude or latitude values would inexplicably change, only to be corrected again later.  One example of this can be found in the Arkansas, Mississippi and Louisiana map of 1860, where in 1862, state/variation 5, the longitude from Washington, which had been “12-17”, is mistakenly printed as “94-99”.  It remained this way until state/variation 7 of the map, in 1863, when it reverted back to “12-17”.

On occasion, there were blemishes that occurred in the map plate, which took some time to be fixed.  One example of this is in the Delaware and Maryland map of 1860, where, in 1862, map state/variation 4, a blemish shows up in the word “State Capital” in the map legend.  The blemish, over the letters “it” in the word capital, remained until map state/variation 6, in 1863, where it was repaired with what appears to be hand drawn letters “it”.  In the next map state/variation, 7, in the same year, the repair was completed with the “it” now being in the original font.  Many such changes can be used to determine states/variations of the maps, and are noted in the Map Identifier and Gallery and its data sets on this website.

 

A Short Note About Printing

The title page of the Family Atlas prominently states that it is a “Steel Plate” atlas.  This was used to demonstrate the high quality of the product and to enhance sales.  However, Ristow suggests, and the wisdom of others observing the maps concurs, that the maps used in the atlases were actually lithographs.  He points out in his book, American Maps and Mapmakers, that the title page of one copy of the Johnson Family Atlas, at the Library of Congress has the following notation on the contents page, “The maps are transferred and Printed by D. McLellan & Bros. 26 Spruce Street, New York.”  Ristow further indicates that McLellan and his brothers were lithographers.  The plates are felt to have originally been drawn on steel plates, and later transferred to stones for the actual printing of the atlases. (Ristow, W.W., American Maps and Mapmakers, Detroit: Wayne State University Press, 1985, p.325)

Selling Atlases By Subscription

The Johnson firm sold almost all of its atlases and books by subscription through door to door canvassers.  Johnson alludes in one of his letters that only on a rare occasion would an atlas be sold through the publishing office itself.  Other major atlas publishers of the times, Colton and Mitchell, also sold their atlases by this method.  (For a more comprehensive discussion of the selling of atlases by subscription go to the Alvin J. Johnson and his role in 19th Century Map Making in America page of this website).  The importance of this practice in terms of the map states/variations in Johnson atlases is that the updating of maps on a yearly basis was a sales promotion strategy as well as an academic exercise.  Every “gentleman” was urged to have the best and most up-to-date atlas in his home; and, the atlases needed to be changed often enough for the salesmen to convince the “gentleman” that the acquisition of the newest edition was necessary.  This appeared to be a successful sales technique, and, although Johnson had a practice of placing an advertisement coupon in some of his atlases offering to sell the buyer updated versions of any of the maps for a small price (and extra spines were included in the atlases for the purpose of inserting these maps), it is this author’s experience that this rarely if ever happened.  A supposition can be made that Johnson’s salesmen urged prior customers to buy new atlases rather than use the update option.

Conclusion

Alvin J. Johnson was not the most famous of American atlas publishers of the 19^th Century, in fact in most cartography texts he is merely an afterthought.  However, his atlases were extremely popular, as evidenced by their current availability relative to those of his competitors, and his success as a salesman and publisher helped establish the atlas as vital family reference book.  The fact Johnson most likely played a role in financially saving the failing Colton firm is probably as an important, if not greater, than his contribution to cartography.