Thompson Bevel Mysteries
On October 12, 1858 an unusual bevel was patented by Meriwether Jeff Thompson of St. Joseph, Missouri. That bevel has received little notice in tool-related books or publications other than brief cites in the March 1987 M-WTCA Gristmill, the September 1988 EAIA Chronicle, and the April 1997 CRAFTS Tool Shed. Attempts to learn more about it and the parties connected with it have uncovered some rather surprising facts, but have not produced answers to all of the original questions and have raised some new ones. I offer my results to date in the hope that some readers can help resolve the mysteries that remain. In the paragraphs that follow I will describe the differences between eight known variations of the bevel, provide some details about those connected to it, and discuss a few peculiarities of the patent, the instruction booklet, and the tool itself.
THE VARIATIONS
For discussion purposes I have designated the eight variations A1, A2, B1, B2, C1, C2, C3, and D in the general order of their production. The letter designations are for major variations, and the number designations are for less significant variations within the letter styles.
All Thompson bevels have a brass blade with a quarter circle segment on the outer end. Besides a protractor-like 0 to 90 degree angle scale around the outer arc of that segment, it has lines marked "3 SIDES," "4 SIDES," etc. up to "16 SIDES." The lines designate the proper miter angles needed to make the cuts for constructing frame assemblies (polygons) with anywhere from 3 to 16 sides. The front side of the body has formulas for determining some (but not all) of the lengths of each side of such frames relative to circle sizes so that they would either just fit inside of a circle or just fit around a circle (i.e., for inscribing or circumscribing a polygon to a circle). The formulas include multipliers expressed in both inches and thousandths of a foot. Related to this, there are distinctive markings of a triangle inside a circle and a triangle around a circle. The back side of the bevel's body has a graph-like scale for measuring thousandths of a foot. The combination of miter-angle markings and side-length formulas form the basis of Thompson's patent.
A STYLE: The primary distinctive characteristic of the two A styles is that angles are aligned on the side of the body. A small brass inset strip there gives a firmer alignment base than the wood side alone would. Another characteristic is the use of thumb-style locking nuts. In general, the assembly and marking of the A styles is noticeably cruder than the later B and C styles.
A1 STYLE: This variation is shown in Figure 1. Nothing is known about it beyond what can be seen in that picture, which came from a secondary file and cannot be traced back to its source. As the variation most closely resembling the patent drawing, it is assumed to be the first type produced. The similarities with the A2 example that can be seen in the picture below suggest that it is probably the same in most other respects.
Figure 1. A1 Style Thompson Bevel
A2 STYLE: This is shown on the right side of Figure 2. The most obvious difference from the Al style is that the part of the blade below the arc is slotted. This example is stamped "Wm. CONAWAY / PHILA" on the blade. It has additional lines of markings on the back side of the butt end of the body; one of those can be made out to be the patent date. Time and wear have obscured the others, but they appear to be (with some help from the imagination) something like "THOMPSON'S / PATENT" and "L.E. YORKE & CO. / PROPRIETORS." Even more imagination is needed to visualize some pivot end scratches as "CHAs. MILLER / SOLE AGENT." The body is birch or beech and is 11-1/4" in length. Some of the side number lines are on the back side of the blade.
Figure 2. C1 Style (left) A2 Style (right)
B STYLE: The primary distinctive characteristic of the two B styles is a rectangular window in which the angle is aligned. The use of this alignment window dictates all angle markings being on the front side of the blade rather than some being on the back as in the A styles. Note that the use of this window puts the pivot point of the blade at the axis of all angles on the blade's scale. This was not true of the A styles; on them, each angle had to align with an axis point that, being on the edge of the body, was slightly offset from the blade's pivot point and varied as the blade was rotated. Accordingly, the blades of the A styles must have been more difficult to lay out and mark than the B and C styles. Although the B2 style preceded the C styles, there is reason to question whether the B1 might not have succeeded them. All B and C styles are made of boxwood vs. the beech/birch used in the A styles and are a slightly shorter 10-5/8" long.
B1 STYLE: One example of the B1 style was sold as Lot #449 in the 1995 Brown Auction and is pictured in that catalog. Besides the rectangular window, the most obvious variation from the A styles is that it has a flat knurled nut. It has a ruled scale on the blade and that is the primary reason for thinking it might postdate the C styles; the C3 is the only other type having such a scale. On the butt end of the back side it is marked "L.E. YORK & CO. / PROPRIETORS." The pivot end back side of the body is marked "CHAs. A. MILLER / SOLE AGENT / PHILa." And, a marking of "THOMPSON'S PATENT OCT. 12, 1858" appears on one of the narrow edges of the body.
B2 STYLE: Except for the rectangular window, the B2 (Figure 3) has more in common with the C styles than with the B1. It has a wing nut vs. the B1 style's knurled nut. While the patent marking is in the same place, it has a slightly different "THOMPSON PATENT / OCT. 12th, 1858" format (Figure 4). The Miller marking is the same (Figure 5). In place of the YORK & CO. marking, it is marked "MADE BY / L.C. STEPHENS & CO. / No. 3" (Figure 6). Most of the C styles have these same markings. Note that the Figure 3 bevel, as compared with those in Figures 1 and 2, has the nut mounted on the opposite side of the body. It is not certain whether this was the way it was made or just a reassembly reversal, but such a reversal would be virtually impossible with most of the other styles checked.
Figure 3. B2 Style Thompson Bevel
Figure 4. Patent Marking
Figure 5. Agent Marking
Figure 6. Maker Marking
C STYLE: The primary distinction of the three C styles is that the window is domed on one end vs. being rectangular. All C styles noted so far have wing nuts and most of them have the same markings as the B2 style with the exception of the number with the Stephens & Co. marking. Those are serial numbers rather than model numbers; as such, they provide a basis for tracking the chronology of the changes.
C1 STYLE: This is shown on the left side of Figure 2. The C1 style was made both before and after the C2 style. All known examples are marked with some or all of the three markings shown in Figures 4, 5, and 6. These marks are stamped quite lightly and easily worn away, so any absences of markings are apt to be due more to such wear (or masking grime) than their not having been marked at all. An example which is probably "Ser. No. 95" (but might possibly be 35) has a small "26" stamped a ways to the side of the patent marking; no such marking was noted on any other examples, and its significance is unknown.
Note that with no examples known numbered between 3 and 95 (or, possibly, between 35 and 123) there might have been other styles within that sequence; further, the serial number dividing lines between the different styles are uncertain. Although additional C1 styles are known, the only other one from which a serial number has been confirmed is No. 182.
C2 STYLE: The primary distinctive characteristic of the C2 style is brass plates on the butt end of the body that are as long as they are wide. Note that all other examples pictured have plates that only extend a short way up the bodies. This plate covers the area where the Stephens & Co. marking appears on earlier and later examples; there is no such marking on the C2. However, a serial number is marked above the Miller marking on the pivot end. The one example known is No. 123 (or possibly 128).
C3 STYLE: The C3 style has a one-foot ruled scale along the edge of the blade. The B1 is the only other style to have this, and it is the only difference between the C3 and the C1. Since the scale has not been seen on any examples with serial numbers below 219 (the highest serial number found to date), it is assumed to be a change made toward the end of the bevel's production vs. an option or such.
D STYLE: I have been told that there is one known example of a Thompson bevel that has an all brass body. I have not seen that one and have no other details about its form or markings.
THE PARTIES INVOLVED
Meriwether Jeff Thompson who is quite well known to Civil War buffs as the "Swamp Fox," was a C.S.A. Brigadier General who commanded up to 5000 men in the Mississippi River arena of the war. He was born January 22, 1826 in Harper's Ferry, Virginia (now West Virginia). Moving from Harper's Ferry to Liberty, Missouri, in 1846, he settled in St. Joseph in 1847. After initially working as a store clerk, he and a partner opened their own store in 1852. He later converted that business to a commissary traveling with a railroad surveying party. During those travels he learned enough about surveying to be put in charge of that party.
In 1857, he was elevated to head the construction of a 25-mile section of the Hannibal to St. Joseph rail line. That section was not completed until February 1859, but Thompson also served as the St. Joseph City Engineer in 1858 and a one-year term as mayor in 1859. He then engaged in various railroad and real estate activities until shortly after the April 12, 1861 bombardment of Ft. Sumter when he organized a battalion to fight for the C.S.A.
After surrendering on May 9, 1865, he worked briefly in Memphis, Tennessee, before moving to New Orleans, Louisiana. Initially a storekeeper there, the influence of some of his wartime associates led to his appointment as Chief Engineer of the Board of Public Works of the State of Louisiana. In 1876, taking a leave of absence due to ill health, he returned to St. Joseph where he died on September 5, 1876. No clues have been found as to how Thompson made a connection with a "Sole Agent" working in Philadelphia.
In 1858, Charles A. Miller worked at 415 Commerce Street as a "Hardware Commission Agent." A Philadelphia directory which lists nine individuals/ companies in that category shows eight of them working in the 400 and 500 blocks of Commerce Street, the apparent hub of that type activity. A later directory lists only Miller as a "Merchant" at 505 Commerce Street; that location might imply that he was still actually working as an agent. By 1872 he was a partner of a George Zinn, working as Miller and Zinn at 1401 Walnut Street. What they did is not specified. The specific services provided by Miller, and his links to the other parties involved remain elusive.
The instruction booklet, as well as the B1 (and probably the A2) style tool, bears the name "L.E. Yorke & Co." (but note the "e" is not in the mark on the B1 style); it contains the equivalent of a copyright notation dated 1858 that was registered in the Philadelphia District Court. That fact plus Miller's working in Philadelphia led to an assumption that this was a Philadelphia company. However, no record of them can be found in any Philadelphia directories or other sources of data about Philadelphia businesses.
A check of St. Joseph, Missouri, (Thompson's home town) records was also unsuccessful. The possibility of Yorke & Co. being only a publisher/ printer would seem to be denied by their name on the bevels. The use of the term "Proprietor" would be unusual for a maker, so it is suspected that they were a hardware wholesaler or such. Major questions remain regarding where this company worked, what type business they were "Proprietors" of, and what their relationship to Miller was.
A William Conaway is known to have been working as a sawmaker in Philadelphia in 1858, and he was initially assumed to have made the A2 style bevel. While that still seems probable, another possibility has surfaced. The data on Conaway's activities as a sawmaker are somewhat confused. One source links him to Johnson & Conaway before 1857, when Disston acquired that business. Another source shows him having owned a Union Saw & Tool Co. in the early 1850s. An 1861 source shows an Eagle Saw & Tool Mfg. Co. as his successor, but that name does not appear in subsequent years, and he continues to be listed as a sawmaker.
Directories consistently spell his name Conway vs. Conaway and list his work address at 402 Cherry Street (1855-60) or 404 Cherry Street (until 1866). His home address in 1858 was 415 Cherry Street, but in 1859 that was shown (spelled Conway) as 1416 Ridge Road. That same 1859 Directory also lists a William Conaway for "hardware" at 1410 Ridge Rd. Was that the same man? Whether so or not, might the marking have been made as a dealer vs. as a maker? Finding additional examples of A style bevels that are or are not marked by Conaway should resolve this question.
Lorenzo Case Stephens was born in 1809 and died in 1871. He was working as L.C. Stephens & Co. in New Hartford, Connecticut, in 1858. By 1864 when the company moved to Riverton, Connecticut, it was being run by Deloss B. Stephens, L.C.'s son. Deloss continued to run the company until 1901 when it combined with H. Chapin's Son & Co. to form the Chapin-Stephens Co. L.C. Stephens himself obtained the January 12, 1858 patent for the well-known No. 36 combination rule/level/bevel/etc. tool which was still being made by Stanley (who bought out Chapin-Stephens) into the 1940s.
It seems likely that Stephens started making the Thompson bevels in 1859 or 1860 at the latest. There is no evidence that it was ever sold by Stephens as one of their own products. Miller and/or L.E. Yorke & Co. probably handled marketing. The changes of styles suggest that Stephens made the bevels over some period of time, but it seems doubtful that was more than a few years. It can be imagined that Stephens produced them in relatively small groups, with changes being made at some of the times that a new group was ordered. If Conaway did not make the A styles, Stephens may have.
Who made the B1 style (and when) is quite debatable. Whoever was ordering the bevels might have obtained the C2 group from some other maker. Regardless, Stephens & Co. appears to have made most of the Thompson bevels ever produced.
THE PATENT
There are some discrepancies within the patent. The text of the patent cites only miter-angle lines for 3, 4, 5, 6, 8, and 10 sided assemblies. The patent drawing, however, clearly shows side-length formulas for those and 12, 14, and 16 sides. The formulas for 7, 9, 11, 13, and 15 sides were never put on any of the tools, and those are the lines put on the back side of the arc on the A styles. Whether the decision to provide them was made· at the same time the other three were added, or even later, is not known. The lack of a brass alignment piece for these back side lines adds to the suspicion they were afterthoughts.
Somewhat similarly, the text of the patent includes an example of using the tool to lay out a 10-sided column, but the portion of the patent drawing that relates shows only a 5-sided column being laid out. The patent describes its length formulas as replacing a standard procedure of stepping dividers around a circle and changing their spacing until the number of steps (equaling the number of sides) produces an exact return to the starting point. That is a more exact method than Thompson's formulas, and probably is easier for any carpenter not fairly proficient in mathematics.
THE BOOKLET
This booklet was probably written by Thompson. Its registry in 1858 suggests that it was written, possibly even printed, before the patent was issued. As originally written, the booklet describes an A1 style with the angles aligned on the side. My copy of the booklet includes a pasted in notice of an improvement that moves that alignment to the "Slot Hole". Whether a revised version was ever printed is unknown. The booklet covers all frames from 3 to 16 sides and includes the side-length formulas missing from the bevel itself (but does not repeat those on the bevel). It cites only angles for making 6, 8, 10, or 12-sided columns.
When the bevel is set to those sided frames, the arced segment of the blade blocks the intersection of the blade and body, so as to prevent laying the tool on a frame's angle to check it (see Figure 3). The booklet describes a process for reversing that angle to the other side to get the arc out of the way. The booklet also includes a comment relative to Thompson's use of thousandths of a foot. It describes such decimal measuring as "fast beginning to be a necessity" and predicts that very little more time would pass before inches would be "heard of no more."
THE TOOL
The bevel's design limits flexibility that would make it more satisfactory for its purposes. Explaining this gets a bit technical, but I'll try to make it as painless as possible. Every angle has a supplement — the difference between it and 180 degrees (e.g., the supplement of a 45- degree angle is 135 degrees). Either such angle can be expressed, different positions on the stock allow either to be used, and bevel users know which is appropriate. Accordingly, the fact that a Thompson bevel set to a marked 45-degree angle actually provides a 135 degree blade-to-body angle is not, in itself, a problem. (NOTE: The angles for measuring solid frames, which are all greater than the 90 degree range of the arc, are expressed as their supplements and have an exactly opposite meaning than the miter-angle markings in that the blade-to-body angle is the one to be used.)
Note, however, that as can be seen in Figure 3, the arc completely eliminates any inside blade-to-body angle on one side of the body, partially blocks a complete marking of such an angle on the other side, and the part of the blade below the arc might be too short to fully mark some frame miter angles. Thus, four possible positioning options that exist with a "T" bevel are reduced to only one that is fully satisfactory.
SUMMARY
There are reasons to doubt that more than about 250 of these bevels were ever made, or that many more than 30 can currently be found in tool collections. With a brass blade and distinctive form, it is an attractive and desirable collectible. That plus its limited availability makes it fairly pricey. The B1 style that sold in the 1995 Brown auction brought $4,200. However, the A2 style sold at the 1997 Brown auction for only $270; it was not identified as a Thompson in the catalog and its differences from "standard" Thompsons may have confused some potential bidders. Most C styles known to have been sold recently were in the $2,000 to $4,000 price range.
I would very much appreciate hearing from anyone who owns any style of one of these bevels. Both details about any characteristics or markings that differ from those described here, and confirmations of like examples (with serial numbers when applicable) are desired. To date, the C1 is the only style that I have seen more than one example of; more like examples would help establish the serial number ranges and quantities of the different styles. I would also like to know anything more about any of the parties involved (especially L.E. Yorke & Co.). I will prepare a follow-up article if/when enough additional or contradictory data is obtained to warrant your further attention.
For their contributions to this article, I would like to thank the following: David Arms; Phil Cannon; Charles Flynn; Chuck Granick; Don Johnstone; Herb Kean; Rick Kerns; Chuck Morgan; Sam Pickens; Terry Thackery; Don & Anne Wing; Historical Society of Pennsylvania; St. Joseph, Missouri, Historical Society; and St. Joseph, Missouri, River Bluffs Library.
Ed: The Tool Shed thanks Bob for allowing us to publish his research on the Thompson bevel. He's well versed on measuring devices as evidenced by his article in EAIA's June 1998 Chronicle on "George Miller's Carpenter Gauge." He also does the CRAFTS auction reviews for us.