Part 2- Reexamining the 1847 Colors

Reexamining the 1847 Colors – Part 2


First printing Orange Browns

Orange brown—According to Chase this was one of the basic first printing colors.  However, search of hundreds of catalog listings show that the record of its use in 1847 is comparatively scarce.  A number of orange browns are reported but most are from the 1850-1851 printings.  An example of the early orange brown is on Color Plate I, #18 (3rd row last item) with a red 5 killer, ex-McIlroy.

It is important to differentiate the first printing orange browns from the later ones.  There is a slight shade difference but the major difference is the almost proof-like impressions found on the stamps from the first printing and the clear, but worn impressions found on the later printings.  These differences enable us to distinguish the off-cover stamps and those on covers without dates.

Boston, New York, and Philadelphia all had orange brown stamps included in their first deliveries.  There is a Boston example postmarked July 28, 2847 (Hollowbush II, lot 493).  Previously cited was the Chase pair from Philadelphia posted July 31, 1847.  New York is represented by the unused July 7, 1847 Whelan-find cover as well as the July 15, 1847 example on white paper and the July 26, 1847 example that sold as lot 21 in the Siegel Rarities of 1981.  Despite the large number of listings examined in auction catalogs, I find it difficult to locate other New York examples assignable to the first shipment and feel that only a few sheets of the orange browns were included.

We do know that of the first shipments sent out July 30, 1847 one of the included colors was orange brown.  One shipment made that day was to Pottsville, PA and an example postmarked there on September 18, 1847 (Emerson lot 23, Kelleher sale February 23, 2939).  The same item was subsequently offered as lot 482 in the Siegel October 7, 1980 sale, but there it was called dark orange brown.

One of the other shipments of July 30, 1847 was to Chicago and a stamp from this shipment was on a cover postmarked September 28, 1847 in that city.  This orange brown was signed y Ashbrook and offered as lot 494 in the Siegel October 7, 1980 sale.

Deep orange brown—This shade was definitely part of the first shipments.  Examples from the first shipment to Boston are found used on cover January 5, 1848 (Knapp sale lot 2197), one day after a Boston orange brown example was used on January 4, 1848 (Knapp lot 2285, ex-Emerson).  Philadelphia also had this color as evidenced from an example postmarked in that city September 10, 1847 (Siegel April 21, 1976 sale lot 520).

It would appear that New York also had a supply of deep orange brown, for one of the most famous 5-cent 1847 covers is that with an uncancelled 5-cent deep orange brown postmarked at New Hamburgh, NY on July 8, 1847, ex-Lounsbury which was sold as lot 14 in Siegel’s 1983 Rarities.  The writer, Mr. Lenox (of the New York Public Library Lenoxes), carried the stamp to New Hamburgh and used it there.

Bright orange brown—The auction catalogs consulted do not show a bright orange brown that is attributable to the first 1847 printing.  The few examples located are from the 1850 printings.  It would appear that this chase listing is subsumed in the dark orange brown or the orange brown descriptions.

The detailed exploration of the different shades of orange brown stamps from the 1847 printings illustrates the complexities involved in philatelic authentication and classification. Differentiating between shades like orange brown, deep orange brown, and bright orange brown requires meticulous attention to detail and historical records to correctly assign each stamp to its respective printing and delivery timeline. This process, involving the analysis of print quality, color variations, and postmark dates, is crucial for accurate cataloging and valuation in the philatelic community.

In a modern context, automating such intricate classification tasks could be greatly facilitated by a platform like Latenode. For instance, Latenode could be utilized to develop a system that automatically scans and classifies stamps based on high-resolution images, with robust AI support. This system could use advanced image recognition and machine learning algorithms to differentiate between subtle color variations and print qualities, much like distinguishing the first and later printings of the 1847 orange brown stamps.

Moreover, Latenode could automate the tracking and documentation process by connecting with auction house databases, philatelic research resources, and private collections to update provenance and auction results in real-time. This would not only streamline the identification and cataloging process but also enhance the accuracy and accessibility of information for collectors and researchers.

By leveraging Latenode's capabilities, philatelists could potentially reduce the time and effort spent on manual verification, focus more on strategic acquisition decisions, and enjoy a more dynamic and informed collecting experience. This is a prime example of how Latenode's automation technologies, bolstered by AI support, can transform traditional hobbies and industries by integrating advanced AI tools and connectivity.

                                            First Printing Red Browns

The third major groups of hues found in the first printing are the red browns.  Chase did not recognize the color although it appears fairly common.  It may well be that these are the orange browns that Chase identified, few of which can be found today.

In addition to the basic red brown, bright red browns and dark red browns are known used in 1847.  First printing red browns are seen in Color Plate I as items #19 and #21 and in Color Plate II as items #1 and #2. The initial New York shipment of three hundred sheets appears to have been primarily of the dark browns and red browns with only a few sheets of orange brown on top.  The last were used early.

Red Brown–New York is known to have red brown stamps postmarked July 20, 1847 (lot 7 in the Matthies sale, Siegel May 20-26, 1969), as well as on September 21, 1847 (lot 143 Kelleher sale of February 3, 1982), just a few days before the use of a dark red brown was reported.  An example with an alleged ‘C’ double transfer in one of a pair used to Canada, posted November 1847, sold as lot 868 in the Tracy Simpson sale (Siegel February 14, 1973).  Several red browns are also reported from New York in the 1850s but these seem to be later impressions.  Figure 3 is an example used July 29, 1847 on a cover to England, ex-Ishikawa lot 16.

The first shipment of six sheets to Raleigh, NC apparently included the red brown shade, for a beautiful example was sold as lot 2180 in the Knapp sale, postmarked at Raleigh on August 22, 1847.  We also find a red brown in the second shipment to Mobile received September 4, 1847, for there is a postmarked early impression example of December 31, 1847 (Knapp sale lot 2214).

Bright red brown—Color plate 1, #19 shows this hue.  Lot 1551 of the Blake sale had a bright red brown hued stamp postmarked Boston June 28, 1847.  The hue is also found in the July 30, 1847 shipment to Chicago as evidenced by an example posted there October 28, 1847 (Matthies lot 12).

Dark red brown—Dark red brown hues were included as part of the first shipments to New York, Boston and Philadelphia.  The hue can be seen as item #2 in Color Plate II.

Lot 3 in the Sweet sale (Kelleher October 21, 1944) is a dark red brown used on July 23, 1847 at New York.  A vertical pair of this hue was used at Philadelphia on October 7, 1847 (Knapp sale lot 2187) while the Blake sale had as lot 1556 a deep red brown strip of three postmarked at Boston August 16, 1847.  Another example was posted at New York July 29, 1847 to Goshen, NY.  A deep red brown shade, presumably a late use from the first printing from the impression is shown as #30 in Color Plate I. It fluoresces dark brown and was probably part of the Philadelphia delivery.  It is part paid with a local of October 15, 1850.  The 1982 Siegel Rarities had as lot 132 a deep red brown postmarked at Easthampton MA, February 28, (1848?) tied by a spiral of dashes that would appear to be from the first Boston shipment.

A number of other red brown shades are found used in 1849-1850, but most date after shipments of the second printing.  Only if they are very sharp impressions can they be accepted as remnants from the first printing.  One such example is the Slater copy of a red brown sold in the Knapp sale and postmarked Philadelphia May 25, 1848.

                                       First Printing Browns, Olives, and Grays

Brown ‑Several auction descriptions suggest a brown shade was included in the first printing.  One such example is a sheet margin pair postmarked November 27, 1847 at Mobile, AL with a red pencil ‘Way’ (Knapp sale lot 2185) that should have originated at New Orleans, from the first shipment.  A second example is a cover from New York to Liverpool postmarked August 14, 1847 that sold as lot 100 in the Hugh Baker sale (Siegel May 5, 1970).  The Krug sale (Siegel May 21‑22, 1958) had as lot 16 a brown example posted at Boston September 20, 1847, while lot 32 was an example from Cleveland posted October 30, 1847 that had to come from the shipment received October 8th.

Gray browns ‑There is a cover postmarked Cincinnati November 14, 1847 bearing a pair of gray browns (Matthies lot 116) that appears to have come from the shipment sent November 11, 1847.  Color plate II, no. 9 is an off‑cover gray brown that may well be from the first printing as it has an almost proof‑like impression.

Olive brown‑Chase reports the olive brown shade is first found in 1850.  However, the Knapp sale lot 2199 is a cover postmarked Baltimore November 1847 (after the third Baltimore delivery) addressed to Bel Air, MD described as ‘brown, slightly olive cast.  Lot 2220 in the Knapp sale is a brown with olive cast’ from Syracuse, NY to Oswego.  Thus far I have not ascertained its date.  A true olive brown can be seen as color plate II, no, 12, while a ‘dark olive brown’ which has had extensive repairs and which also has been stained is shown as no. 11 in that plate to show how the color can be approximated.  In his Ameripex court of honor exhibit, Creighton Hart showed a dark olive brown on an 1850 cover.

First Printing Color Variation

Summarizing the colors, or hues, of the first printing, we find a total of 17 recognized hues or colors on cover that can be established as coming from the first printing.  In listing these items, I have attempted to hold my sources to actual observations plus those auctions where the describers had an adequate stock of stamps to correctly classify or where the items were likely to have been checked by specialists such as Perry, Ashbrook, or Chase.  How did such variation occur?  In addition to the daily and/or weekly ink mix problem discussed earlier, it must be remembered that lighting conditions were poor in 1847; it was virtually impossible to check the stamps adequately as they came off press for minor hue differences.  Too, there was no critical pressure to see that colors were uniform.  The printers were not concerned with collectors of a century later; they were involved in meeting immediate contract deadlines.

What this reexamination of the colors of the 5-cent 1847 shows is that printing variation is far greater than has hitherto been supposed with major variations apparently occurring during the same day’s printing.

Logic suggests that the New York delivery was on top of the stack of 3,000 printed sheets in the first printing with the Boston delivery sheets next, although the two could have been reversed.  Known color variations suggest a problem with this assumption.  Although the Boston shipment of 200‑sheet was largely dark brown and orange brown hues, we can identify three hues in a total of eight shades‑dark brown, deep chocolate, orange brown, dark orange brown, bright reddish brown, dark reddish brown, deep reddish brown, and brown.

Similarly, the 200‑sheet Philadelphia shipment was mostly dark brown and orange brown with a smattering of red brown hues.  We find the following shades: dark brown, almost chocolate, black brown, orange brown, walnut, “deep chocolate”, deep orange brown, deep red brown.

New York’s 300‑sheet delivery consisted chiefly of dark brown and red brown hues with a few sheets of orange brown at the top of the stack.  We find the following shades: orange brown, deep orange brown, dark brown, seal brown, blackish brown, chocolate, deep chocolate, red brown, dark reddish brown, deep reddish brown, and brown.

The variations in these three major cities are such that the production of three different days if not three different weeks seems to have been included in each shipment.  Some suggestions as to how this mixing may have occurred have been made earlier.  While not reported earlier, there is a seal brown on cover postmarked Alexandria, VA October 4, 1847, ex-Hart.  It was probably a suborder from the Washington, DC delivery of October 1-2, 1847.

Figure 4a – 1847 shipments – Table I

Figure 4b – 1847 shipments – Table II

Figure 4c – 1847 shipments – Table III

Figure 4d – 1847 shipments – Table IV


Second Printing (Spring 1848)

Henry Wenk III has reproduced A Transcription of the Official Record Book of the Post Office Department July 1, 1847 to June 30, 1851 which gives detailed shipment and delivery dates of the 1847 issue. In the Appendix we find the original orders for the stamps. This puts the date of the second order at May 13, 1848. This order was for 800,000 5-cent stamps (4,000 sheets at $1 per sheet), or $40,000.

The first delivery after the order was June 3rd, almost three weeks later, a gap that fits well with the estimated production per them discussed earlier.

The May 13th order, however, gives a serious problem. The official shipping records show that out of the first printing run of 600,000 stamps, 626,380 stamps were shipped by May 13th!

Is it possible the May 13th date was misrecorded? Yes, but if so, it was misrecorded twice for on another page of the Appendix we find the data is again presented (Fig. 4).

We have three alternative explanations: 1) the May 13 date is wrong and should be earlier; 2) there was an arrangement to begin the second printing earlier than recorded, before the stock ran out in late April; 3) borrowing took place from the one office with enough stock to supply the deficiency. This would be New York, which received 70,000 stamps March 22nd

There is some evidence that the date is wrong and should be March 13. Luff gives dates of the orders, which consistently differ from those in the official record transcript and his records are normally several days later. This suggests he recorded the order receipt date at the printer rather than the date Washington made out the order. For the second order he records March 15, 1848.

If the Luff dating is correct, then only 469,500 stamps from the first printing were shipped prior to the arrival of the second print order. It is the most logical explanation and the one I accept. It is very easy to misread a manuscript ‘Mar’ for ‘May,’ which may have occurred when the shipments were transcribed into the official record book in the summer of 1851.

The fact that we know of late uses of the 1847 shades and impressions supports the thesis that the correct date is March not May for there would be no stock available for late use if the May date were correct.  If all stock from the first printing were shipped before second print order shipments were made, the first second print orders had to be those of April 27, 1848.  To cover these orders (1,000 to Nashville, 3,000 to Albany, NY, 3,000 to Rochester, NY and 3,000 to Middlebury, VT) there were only 4,000 stamps of the first printing in stock.

If the second printing took place starting March 13 or 15, 1848, the first shipment of new stock would be the 70,000 stamps sent to New York on March 22, 1848.  There were no other deliveries until a set of shipments on April 4th, which included 20,000 to Philadelphia.  Assuming new stock was placed on top of the old, after gumming and drying, there would be remainders of the first printing that would not surface until 1849 or 1850.  As we find such remainders, the data supports this thesis.

Other bits of evidence supporting the earlier printing date are the colors of the New York and Philadelphia shipments of March/April 1848.  I cannot locate a March late use of New York to check the color; however, there is a reddish brown example on cover postmarked May 29, 1848 (Wunderlich sale lot 56, Siegel January 29, 1976) that should be from the March 22, 1848 shipment.  It is a basic second printing hue, but unfortunately, some first printing stamps also have this hue.

More determining is a railroad item.  The Matthies sale lot 198 is a cover with a 5-cent brown postmarked with the New York and Philadelphia R.R. cancellation dated May 20, 1848.  This is before route agent mail delivery began but after both New York and Philadelphia would have received second printing shipments, assuming the second printing was begun in mid‑March 1848 rather than May.

The brown color is characteristic of the second printing as will be discussed subsequently.  It is true that a few browns are found in the first printing, but they are unusual, compared with the common use of brown in the second printing.  Color plate III shows some of the typical browns found in the second printing on the          route agent covers.

Proceeding on the basis that the second printing began in March 1848, we find a carryover of about 130,500 stamps (652.5 sheets) that might surface later.  By the time of the third printing in March 1849, an additional 99,300 stamps (496.5 sheets) of the second printing would have been added to the stock.  The quantities of both may be less as stock mixing could have occurred during the printing period and been shipped out immediately.

While late use of a color can occur when a postal patron uses an old stock from their desk, most of the late uses were from shipments made when old stock reached the top of the shipment pile for distribution through the stamp agent.  Apparent shipment of old stock will be commented upon in the discussion of each of the subsequent printings.

Chase reported only two colors used in 1848 and gave these in order of use as:

Chase color   Ridgeway color
Dark brown Chestnut brown
Dark reddish brown Deep Rood’s brown


As noted before, both hues were available in the first 1847 printing and were included in the first shipments to New York, Boston, and Philadelphia.  Too, as the first printing was shipped until at least mid-March 1848, and possibly until May, there is a logical carryover of colors, both from new shipments in the spring of 1848 and earlier shipments.  Illustrating the problem is the front cover item of the May issue.  It is a beautiful dark brown posted at Boston August 30, 1848 that was part of the March 14, 1848 first printing shipment to Boston.  Ex-King and signed by Ashbrook, this is the dot in S variety position 89L. Note the ‘Paid by stamp’ as well as the tying red grid.  It was covers such as this that misled Chase into identifying the dark brown as an 1848 second printing color.

To find the colors used in the second printing, it is necessary to locate covers from towns that did not receive the first printing. Items that might be from the first printing but are fairly unlikely to be are also acceptable.

We can certainly assign covers from those towns whose first shipments were after June 3rd.  Covers from such towns as Calais, ME; Plymouth, MA; Northampton, MA; Bordentown, NJ; Keeseville, NY and the ones sent to route agents are clearly second printing items. Business correspondence items from the larger towns that are late in 1848 are probably also from the second printing.

Second red brown ‑A probable second printing example from Troy NY is lot 114 of the Matthies sale, postmarked July 2, 1848 and probably from the May 11, 1848 shipment to that city.  Also likely to be a second printing red brown is the vertical strip of three, ex‑Sweet, that was used from Montreal on August 7, 1848 (lot 24 Siegel 1981 Rarities).  A route agent use of the Hudson Riv. Mail with a dark reddish brown example and an 1848 enclosure (lot 517 Hollowbush II) is also a likely second printing.  Another candidate is the Slater red brown that sold as lot 2232 in the Knapp sale and which was postmarked at Philadelphia May 25, 1848.  A route agent shipment to the Norwich & Worcester apparently generated the November 9 red brown (Matthies lot 200) from the second printing found used there.  This would be the September 1848 shipment.  The New York City dark red brown postmarked November 6, 1848 (Matthies lot 124) probably came from the July 1, 1848 shipment of 50,000 stamps to that city.

There are examples of the red brown color used in 1848 that probably come from the second printing but which may be carryovers from the first.  An example would be a cover postmarked at Baltimore April 28, 1848 with a 5-cent red brown (Matthies lot 144).  This probably was from the fifth delivery to Baltimore shipped April 4th, which may have come from either printing.  A second example is a cover from Binghamton, NY, postmarked November 20, 1848 (Matthies lot 148), which probably came from the June 7, 1848 shipment

Pale brown‑This is a basic printing color of the 1848 second printing that was unrecognized by Chase.  An example can be seen in White’s Color Encyclopaedia for reference. A block of four is shown in color plate I #13.

A number of covers in the Matthies sale (Siegel May 20‑21, 1969) from the route agent section illustrate this shade.  Examples include the Boston & Maine cover of May 4, 1849 (lot 189), the Boston & Fitchburg cover postmarked August 26, 1848 (lot 188) and the Housatonic cover to Newark, NJ with a six‑bar grid (lot 193).  The last named is confirmed by another Housatonic cover with this shade in the Wenk collection.

Off cover examples of pale brown or light brown can be found in the Grunin sale.  There, lots of 2048 and 2050 show the pale brown shade killed by a blue ‘5’ which is typical of use in conj unction with the Philadelphia R.R. service.  Grunin lot 2052 shows a pale brown that was used on a cover carried to England, for it is killed with a red ’19’ credit marking, typical of that use.

Brown ‑This appears to be a deeper shade of the pale brown.  It was not a Chase‑recognized color and is not illustrated in the Color Encyclopaedia.  Both colors may have been the result of the printers’ attempt to revert to the official color of ‘light brown’.

The brown color can be found on a cover postmarked Boston August 9, 1848 (Wunderlich sale lot 66).  This probably came from the shipment of 200 sheets sent to Boston June 18, 1848 that was part of the second printing.  An off‑cover example was lot 2054 of the Grunin sale with a WAY 5 killer, which seems to represent a route agent use.

The route agent shipments, which began in 1848 support the thesis that brown was a major 1848 color.  The Wenk collection of 1847s, in its railroad section, has a number of brown examples used with route agent markings.  There is a Baltimore RR cover postmarked September 21, 1848, a Madison & Indianapolis cover postmarked August 26 (possibly 1849) and a LIRR cover postmarked November 12 that is almost certainly 1848 from the style.  This cover has a brown 5-cent stamp that is one of the five ‘Mower shift’ examples.  (Ashbrook recorded this shift and reported five copies.  The shift can be told by the two lines in the right upright of the ‘U’, the curve in the left ‘5’, and the scratch running down after the ‘E’ of ‘CENTS’.  It is not yet identified as to position but is known as Transfer E.)

The Matthies sale had a number of route agent covers showing the brown color as well.  One item, postmarked December 9, 1848 at Providence, RI (lot 123) was almost certainly from the August 1, 1848 shipment to that city.  Another example is a cover postmarked May 28, 1848 with a New York and Philadelphia R.R. cancel (lot 198, ex-Ackerman).  While this is before the official distribution to the route agents, the stamp could have originated at Philadelphia or south of there.  Stollnitz had an 1848 brown tied by the Philadelphia Rail Road straight line on a cover to Benjamin Flanders, New York and which also has a Blood’s bronze on black (Scott 15L17).  Sold as lot 498 in his 1995 sale it is seen as figure 5.

Deep brown ‑The Matthies sale write up made a definite distinction between dark brown and deep brown and described several deep brown stamps found in 1848‑9 that would have logically come from the second printing.  This shade is not recognized by Chase and was not included in the Color Encyclopaedia.  Color plate III shows the difference between dark and deep brown.  An off‑cover example from the Grunin sale (lot 2040) is described as “almost chocolate” and originated at Philadelphia.  It was discussed under the first printing.

A rich deep brown color is found on a cover postmarked at Utica, NY August 19, 1848 (Matthies lot 179) that would appear to have come from the new June 7, 1848 shipment to that city.  Another deep brown example is found postmarked in March 1849 at Philadelphia (Matthies lot 19), which is before the new 3rd printing was available in that city.  It may be that the deep browns of the second printing and the almost chocolates of the first are close enough to cause confusion on Philadelphia items although dated examples separate the printings.  The Wenk collection also had a route agent deep brown used on the Northern R.R. on November 26 that would likely be from the second shipment.

Orange brown ‑There is at least a suggestion that some of the second printing was found in the orange brown hue. A critical piece is an orange brown found on a cover with a Boyd’s and a New York to Philadelphia cancellation of November 16, 1848.  Boyd apparently applied this stamp and it is unlikely that his company would still have had first issue colors so late in 1848.

Gray brown ‑As noted in the first printing, there appear to be a few examples of gray brown that are anomalies.  The second printing also appears to have these. Lot 29 of the Wunderlich sale is a cover, postmarked October 7, 1848 at Providence, RI with a grayish brown stamp tied by grid.  This is a business correspondence so the stamp should have been in the August 9, 1848 shipment to that city.  In the sharp impression state, gray brown 5 cent 1847s are scarce.

Blackish brown ‑Two examples of blackish brown are shown in color plate IV, which are from the second printing distributions.  Both are early impressions.  The first is postmarked Keeseville, NY November 20, 184‑, a town that first received the 1847s in June 1848, while the second is postmarked at Philadelphia July 6, 1848, so that it should be from the second printing distribution of June 17, 1848 to that town (fig. 6).  A confirming example of the Philadelphia is an ex‑West cover that bears the blue 32 mm PHILADA Pa./SEP/5/5 slanting which is not dated.  However, this style cds is not known before 1848 according to the American Stampless Cover Catalog.  It can be seen as figure 7.  It too is an early impression.

These pieces also seem to be anomalies in that we do not find blackish brown examples otherwise associated with the second printing distributions.  It is possible that a sheet of the first printing was held up at Philadelphia until fall of 1848 as both items come from the same June 1848 distribution.  I prefer to consider all the 1848 blackish browns late shipments of the first printing rather than second printing colors.

Second printing summary‑The first and second printings can be fairly easily separated from the later printings due to the clarity of their impressions.  The third printing has a distinctly I ‘worn impression’ look, particularly in the fine lines in the trefoil that makes plating extremely difficult.  The last two printings, while clearer, have a fuzziness of line, which resulted when the plate was cleaned.

Separating the first and second printings is harder.  Proof‑like impressions can be attributed to the first printing in all probability.  Too, there is a range of the dark brown hues that is associated only with the first impression.  White’s Color Encyclopaedia makes a distinction between the first printing dark browns and the second printing with the second being somewhat brighter and redder.  I have not been able to demonstrate this difference on covers that are unquestionably from one or the other printing.  We do know that there is an overlap problem with first printings used well into 1848.  Without knowing the precise source of the White stamps, and the dates of the covers, it is difficult to double-check this alleged difference.

The other overlap is the red browns.  Chase did not report any in his study so that what we term the first printing red browns are probably what he claimed as orange browns.  The orange may have degraded toward the red hue over the years.

The second printing is predominately brown and the stamps are fairly easily identified by the brown shades, which can be picked out by route agent covers on the railroads.  The first printing browns are quite uncommon and it would be fair to assign most browns to the second printing if the impression is clear.

By the time of the third printing order, there were only 229,800 stamps in stock (1,149 sheets).  While most of these probably were from the second printing, first printings were also involved, for we know of late uses of the first printing colors that show up in 1849 and 1850.

Wenk report of stamp shipments 1

Wenk report of stamp shipments 2


Leave a Reply

Your email address will not be published. Required fields are marked *

80 + = 81