Who Cut the Molds

Ask who designed and cut the molds at Three Rivers and the collector literature answers with a sculptor. The record answers with a mould shop of six, named on a list one of them signed, and with one page of advertising copy that has been mistaken for a source ever since.

The two named craftsmen

David J. Jones, mould maker. The worker roster compiled for the 1973 marker application lists him plainly: “DAVID J JONES MOULD MAKER 1928”, with an address at 1107 E. Bowie, Beeville [@ev-0041, p. 7]. He was still alive when the roster was made — by then at the Beeville Machine Shop [@ev-0041, p. 54] — and he appears in the dedication photographs among the former employees [@ev-0041, pp. 21-22].

He is also the only person in the entire record who claims a specific Three Rivers design as his own, and he claims it by denying one. Against a Big Chief soda bottle displayed in Lee Troell’s collection, the compiler recorded: “David Jones said he designed a BIG CHIEF, but not the one on this bottle. The one he designed was much larger and was nearer the base of the bottle” [@ev-0041, p. 54]. That is a man distinguishing his own work from another’s at a glance, forty-odd years on. It is also the only first-person design attribution the vault holds from anyone who actually worked in the shop.

Carl Stendebach, mold designer. He has been in this wiki’s sources the whole time and was never counted. The Stendebach family history, written with family input for the Three Rivers centennial series, says it flatly: “Carl worked as a mold designer at the Three Rivers Glass Factory until it closed in 1937. He and Daisy then moved to Pennsylvania where he worked for a glass factory. Later, they moved to Palestine, where they lived the rest of their lives” [ev-0015].

Unlike Jones, Stendebach can be followed in federal records. His WWII draft card, registered 16 October 1940, puts him at Parkers Landing, Armstrong County, Pennsylvania, employed by the Wightman Bottle & Glass Co.; the 1950 census finds him in Palestine, Texas, aged 39, occupation “shop foreman,” industry “glass company,” working 62 hours in the prior week [ev-0021]. Those are records, not recollection, and they make him the best-documented glassworker in this expert’s holdings. What they do not document is the one thing the article most wants: they confirm his trade, not his job title at Three Rivers. That he was a mold designer there rests on the family account [ev-0015] and on the 1973 compiler placing his name on the mold-making page of the binder [@ev-0041, p. 29].

So the standing claim that Jones is “the only individual the record explicitly attaches to the mold shop” is wrong twice over. Stendebach is attached to it by a family account and is the one with primary federal records behind him; and four more mould men are named on a page of the same binder, as the next section sets out.

The roster David Jones signed

An earlier version of this article said the record named two craftsmen and no more, and listed “any period document naming a Three Rivers designer other than Jones and Stendebach” among the things this expert does not hold. That was wrong, and the document contradicting it is in the same binder the article was already citing, four pages past the passages it quoted.

On pages 53 and 54 of the marker research is a staff list headed “INFORMATION from Mr. David Jones, EMPLOYED in the THREE RIVERS GLASS FACTORY: April 1929 through September 1930. Returned to work there, 1932 through 1934,” and signed at its foot “David J Jones, Beeville Machine Shop, Beeville, Texas” [@ev-0041, pp. 53-54]. It is organized by department. Under MACHINE UPKEEP and DESIGNER it gives one name:

“C.B.Hoops” [@ev-0041, p. 53]

and under MOULD MAKERS, five:

“Ross Ayers — David Jones* — R.A.LaPell# — Theo. Williams# — Ballentine#”, with the note “The three marked (#) came to the Factory after it had been in operation for some time.” [@ev-0041, p. 53]

That is a mould shop of five, plus a designer, named by a man who worked in it, with a rough order of arrival attached. It is recollection given in 1973, not a payroll, and Jones is naming his own crew, which is both why it is worth having and why it cannot be checked against itself. But it is a period participant naming period participants, and it is the only such list this expert holds.

The same page names the rest of the plant’s staff, which no article here has used: C. R. Tipps as general manager; his secretary Emma Lee Bryant, later Mrs. Dick Epperson; factory superintendent O. M. Wilson; chemist Steve Coleman; glass maker and furnace supervisor Albert Bowman; construction engineer and draftsman H. O. Knape; bookkeeper Eddie Tyre; paymaster C. S. Steadman; secretary and treasurer R. A. Ellerman; machine operators Jim Woodruff, Frank Schotell, Whity Knight, Victor Johns, Otto Schotell, Luke Wagner and Harry Campbell; shift foremen A. N. Qualis, Sam Lungo, Joe Tetskus and Chester Peters; and truck drivers Guss Diez, “Fats” and “Hook” Hutchens, Herbert Simmons and Claude Casey [@ev-0041, pp. 53-54].

A lead this expert did not run down, now run down: it holds. An earlier version of this section recorded the Knape-Coleman Company as a coincidence of surnames, said that nothing held connected it to Three Rivers, and called the check cheap. The check was cheap, and it had already been paid for. The connection is made under oath, in the same trial record this article cites elsewhere, by one of the two men.

Steve Coleman — a Texas A&M graduate of 1927, and the name the Jones roster gives as the plant’s chemist [@ev-0041, p. 53] — testified as president of the Knape-Coleman Glass Company [@ev-0059, part 04, before fol. 3430]. Asked what he had done before that, he answered: “I had been chief chemist for the Three Rivers Glass Company, at Three Rivers, Texas” [@ev-0059, part 04, fol. 3430]. Asked how he came to Knape-Coleman:

When the Three Rivers Glass Company went into receivership, they were closed down for a period of time and Mr. Knape and I became interested in a plant that the West Texas Utilities, during Mr. Insull’s expansion had built at Santa Anna, Texas. [@ev-0059, part 04, fol. 3430]

That plant had been moved down from Oklahoma, run three or four months and shut. In late 1932 and early 1933 the two men acquired its assets — held as the Texas Glass Company and owned substantially by the First National Bank of Dallas — and organized the Knape-Coleman Glass Company in June or July 1933 on $72,000 of capital, some ninety-eight percent of the stockholders from Austin and a Mrs. Ed R. Harrell a principal one. Knape and Coleman ran it themselves, with a superintendent named Hooper who had been chief engineer of the Liberty Glass Company at Sapulpa [@ev-0059, part 04, fol. 3431].

And Knape came out of this mould shop. Cross-examined about what his partner actually knew, Coleman was specific:

He operated a milk bottle machine, an LA machine, there for quite a period of time when he first went to work at Three Rivers. Later he went into the mold shop and was for some time mold shop foreman there. [@ev-0059, part 04, fol. 3452]

The roster’s “construction engineer and draftsman H. O. Knape” is the same man, and the roster records only where he finished. He ran a milk-bottle machine first and then ran the mould shop — which makes him the one person this expert can place in charge of the room this article is about. No other source held here, and no other person in any of them, is described as foreman of a Three Rivers mould shop; the other foremen on record are of shipping, cartons, packing, and the shifts [ev-0041].

What the lead was reaching for, it found. The conjecture was that two of this plant’s technical staff left and built a feeder of their own, which is the same thing Tremblay did at the same moment in the same state. That is what happened. Hartford-Empire’s files show the Knape-Coleman feeder held to be infringing, competing with the Liberty Glass Company, and the matter commercially serious enough that Hartford prepared a suit and consulted its Texas attorney Searcy about “the Texas courts and as to the procedure that would be followed in such an infringement suit” [@ev-0059, part 16, before fol. 59495]. The receivership of 1932 did not scatter this mould shop into obscurity. It sent the shop’s foreman and the plant’s chemist up to Santa Anna to build the thing again — and Hartford met them there with the same notice it had served on Three Rivers in 1929.

And Hartford’s licence register records how it ended. The same tabulation that dates Ball’s acquisition of Three Rivers carries, seven months earlier, the entry “4/11/36 - Kna[p]e-Cole - Licenses expired upon financial failure” [@ev-0059, part 18, line 68242]. The venture two of this mould shop’s men built ran from the summer of 1933 to April 1936 and failed for want of money, some seven months before Ball Brothers bought the plant they had left. It failed in the posture its founders had learned here: licensed by Hartford for some ware and held to be infringing on a feeder, which is precisely the position Three Rivers had occupied on milk bottles since 1929.

That entry is read out of a two-column table badly damaged by the scan, with the neighbouring column’s text interleaved into it; the words following “financial failure” are not legible. The date, the name and the fact of expiry are.

One page, read twice, counted as two

The famous “six full-time bottle designers” and the Borglum attribution do not come from two places. They come from one undated page of the company’s own promotional brochure, preserved in the marker binder under the heading “SHAPE AND DESIGN OF EVERY BOTTLE MUST FIRST BE FASHIONED IN STEEL”:

“Mould-making for various sizes and designs of bottles is a fine and skillful art. Moulds must be as exact as die-cutting tools. Many of the shapes are intricate. All of them have curves and angles requiring engineering science of utmost exactitude. Thousands of special trade-mark moulds are in our customer ‘files’. Six experts are kept constantly busy designing and making new ones. The famous sculptor, Gutson Borglum, has designed bottles for Three Rivers Glass Company.” [@ev-0041, p. 29]

Read what it actually says. The six experts are “kept constantly busy designing and making” the moulds — that is a description of the mould shop, one crew doing both crafts. It is not a design staff of six sitting apart from the mould makers. And Borglum “has designed bottles for” the company: a past commercial relationship, stated without a date, without a job, and without a bottle.

And the number turns out to be right. Jones’ roster gives five mould makers and one man on “MACHINE UPKEEP and DESIGNER” [@ev-0041, p. 53] — six men doing exactly what the brochure says six men were doing, listed by one of them. That is a real corroboration and it should be stated plainly: the “six experts” sentence is advertising copy that happens to be accurate, and the article’s predecessor was right to distrust it and wrong about what it was describing. What the roster does not corroborate is Smith, who converted the same six into “six full-time bottle designers” on a payroll of over 150 and hung a sculptor on the end of them.

Michael David Smith’s 1989 history renders the same passage this way:

“the Three Rivers Glass Company had over 150 employees on its payroll, including six full-time bottle designers. It was during this period that Belgian sculptor Gutzon Borglum, who was residing in nearby Beeville at the time, joined the design staff of the Three Rivers plant. Borglum had become famous for the creation of Mount Rushmore, and his addition to the company payroll added an element of prestige to the entire operation. His design influence almost certainly left its mark on all Three Rivers glass produced from that time on.” [@ev-0039; @ev-0038]

Every disputed particular in the collector literature enters here, and none of it is in the brochure. The brochure does not say Borglum was Belgian; it does not place him in Beeville; it does not put him on the design staff or the payroll; it spells him Gutson. Smith converts a crew that designs and cuts into six pure designers, converts a past commission into employment, and then — in a sentence that gives the game away — extends the influence to “all Three Rivers glass produced from that time on” while conceding the word “almost certainly.”

This is not a second witness. It is the same page with additions.

Kevin Mackey reaches the same conclusion independently, working from the 1921-22 trade press rather than from the brochure: “Smith, it seems, took much of this information from the written record Charles Tips provided and was read for the dedication of the marker. Tips only stated the workers were ‘brought down from the north’” [ev-0040]. Mackey is making the point about a different embellishment — the claim that plant manager Warrick personally brought the northern glassmen, which Tips never said [@ev-0041, p. 16] — but the finding is the same one, and it establishes the habit rather than the single instance.

The consequence for this wiki is direct: this article’s predecessor separated “designers” from “mold makers” as distinct crafts and reasoned at length about which one Three Rivers staffed. The primary document it was reasoning about says they were the same six men.

What survives about Borglum

Stripped back to the brochure, the claim is narrow: at some point before an undated advertising leaflet, a famous sculptor designed bottles for this company. That is a company advertising its own prestige, which is the weakest possible class of witness for exactly this kind of assertion, and this expert holds no second source for it of any kind.

Against it:

• The Borglum archives are silent. The TSHA Handbook biography, which covers his Texas period in detail — the 1925 move for the Trail Drivers monument, the Menger Hotel, the Brackenridge Park studio, the failed Corpus Christi waterfront plan — mentions no glass and no Three Rivers [ev-0012]. The Gutzon Borglum Texas Papers at Texas A&M–Corpus Christi, the one collection assembled specifically around his Texas work, surfaces nothing in its finding aid [ev-0010].
• Two local negatives, neither of them from a Borglum source. The 1933 Texas Almanac entry for Live Oak County describes Three Rivers as carrying “a large glass manufacturing industry” and says nothing of Borglum, though Borglum appears elsewhere in the same volume on unrelated business [ev-0035]. A 1972 Beeville Bee-Picayune piece on a Three Rivers bottle collection shown at the county fair — precisely the venue where the prestige story would be told if it were current locally — does not mention him [ev-0036]. Neither is decisive; together they show the claim was not circulating in the two places most likely to carry it.
• The Beeville detail belongs to the other Borglum. Beeville’s own 2016 Texas Historical Marker honours Lincoln Borglum — Gutzon’s son and Rushmore collaborator, who married a Beeville descendant, settled there, and served on the city council [ev-0007]. Smith’s “residing in nearby Beeville” fits the son, whose Bee County life began after the factory era, and not the father, whose Texas base was San Antonio [ev-0012].

What does not weigh against it: the chronology is possible. Borglum wintered in San Antonio through these years, roughly sixty miles away, and took commercial commissions. No held source rules the commission out. It simply rests on the company’s own advertising, and no Three Rivers bottle has ever been attributed to him by anyone, anywhere in this expert’s holdings.

The men behind the names

The roster gives names and a department and stops there. Beyond Jones and Stendebach nothing else in this expert’s holdings says anything further about any of them, and it is worth being exact about how little that is.

Tips’ own 1973 account of the hand-blown years says only: “Some skilled glass blowers and a few mould makers were brought down from the north and a few local men were employed” [@ev-0041, p. 16]. No names, no count, no origin more precise than “the north.” H. S. Warrick — rendered H. L. Warrick in some of the same trade press — was the Cleveland glass man who answered Tips’ advertisement and managed the plant from 1922, having lost his own factory to a fire the year before [@ev-0038; @ev-0040]; he was replaced by John Finkbiner when the machines came online in early 1925 [ev-0042]. He is the obvious channel for northern hiring, but the claim that he brought the workers himself is Smith’s, not the record’s [ev-0040].

Albert F. Tremblay, hired from Monterrey, built the company’s forming machinery, not its moulds. His U.S. Patent 1,648,792 for a “Bottle-forming machine” — filed 26 June 1926, granted 8 November 1927 — resolves the “Harold Trembly” of Smith and the BRG chapter to Albert F. [@ev-0051; @ev-0042], and the binder holds the $6,000 bill of sale behind it [@ev-0041, p. 46]. Nothing credits him with cutting or designing a mould. His machine shop was nonetheless the condition for in-house mould work: interchangeable blank and finishing moulds are the machine’s consumable, and a plant running them needs someone on site to maintain and recut them [ev-0051].

The physical evidence of these anonymous hands is the marks themselves. The Bottle Research Group, unable to date the maker’s-mark variants, concludes that they “may only have been changed at the whims of the mold makers” [ev-0042]. That sentence is the closest thing the record has to a portrait of the shop: a group of men with enough authority over the tooling to change the company’s own mark without leaving a reason.

That authority has one documented consequence, and it is not in the collector literature at all. Twice the shop’s output put the company outside the licence it held. On 1 June 1931 Hartford-Empire wrote to Tips about pressed packers tumblers appearing in the trade from “a licensee new in the pressed business” — a class of ware his Field of Permitted Ware did not cover, and one whose existing licensees were “naturally quite jealous” of their rights in it [@ev-0059, part 16, fol. 60862]. In August 1934, with the plant in receivership, an internal Hartford memorandum recorded that Three Rivers had made brilliantines on Single Feeder No. 295 and checked them as toilet ware — the correct classification, “but they have not the right in their Field of Permitted Ware to make ‘Toilets’” [@ev-0059, part 16, fols. 60789-60791].

Neither document names the man who cut the mould, and neither is evidence about design in the sense the Borglum story means. What they establish is narrower and better attested than anything the brochure says: somebody in this shop was cutting moulds for ware the company had no right to make, twice, five years apart and on both sides of the receivership, and the patent holder found out from the trade rather than from the plant.

What this expert does not hold

• A payroll. Jones’ roster [@ev-0041, pp. 53-54] is the closest thing held, and it is a former employee’s recollection written up in 1973, not a company record. It gives no dates of service for anyone but Jones, no wages, and no spelling authority — “Ballentine” has no given name and “Whity Knight” and “Theo. Williams” are as the page has them. An earlier version of this article said no period document named a Three Rivers designer beyond Jones and Stendebach; that was a failure to read four pages further into a source the article was already citing.
• Any Borglum-side document of any kind — no contract, no correspondence, no drawing, no payment. The claim exists in one sentence of the company’s own advertising and in what has been repeated from it.
• The company brochure’s date. The page is undated in the binder; its companion pages give a furnace capacity of 125 tons and forty-five tons of glass drawn daily [@ev-0041, p. 27], which would place it well after the 1926 expansion, but this expert has not fixed it.
• The trade literature on mould-making as a craft. A description of the period trade — the division between designers, mould machinists, and chipper-engravers, and the hours an ornate pattern took to cut — was compiled into this article’s predecessor from a 1990 Wheaton Arts publication. That source is not admitted: deferred on the question of whether it touches any named Three Rivers person at all, it was refused when the question was answered — it names none, and no named hand of this shop connects to the trade it documents (record/refusals/in-0016). Everything resting on it has been withdrawn rather than carried forward uncited. The craft distinction it supplied also turns out to be the wrong lens for this shop, whose own brochure describes one crew doing both jobs.
• The 1930 census enumeration for Three Rivers, whose occupation column would name the mould shop outright, and the local newspaper run. The 1940 enumeration has been searched for Stendebach and does not contain him [ev-0021]; the 1930 has not been worked at all.