references · updated 2026-08-27
The First Owens-Illinois Catalog (1930–1935)
confidence: single-source weakest: ev-0018
Analysis of the first Owens-Illinois catalog (1930-1935) and its coffee-jar pages bearing on the Crystalvac dual-sourcing question
The first catalogue issued by the Owens-Illinois Glass Company after its 1929 formation survives as a bound archival photocopy, scanned and published in 25 PDF parts totalling 536 catalogue pages. It was never officially dated; internal evidence brackets its assembly between 1930 and 1935 [ev-0018].
It matters to this expert for one reason. It is a primary record of what Owens-Illinois offered — including coffee jars — during exactly the years the Hoffmann-Hayman Crystalvac dual-sourcing question turns on, and that question is the hardest live case in telling Three Rivers ware from ware it is confused with.
What this expert actually holds
This is the part to read before citing anything below. The catalogue is not held here. What is held is three documents: the site’s presentation page, which carries the dating analysis, the provenance account, and the file-by-file structure [ev-0018]; a transcription of the coffee-jar pages in File #15 [ev-0016]; and a reading of the Closure Division in File #22 with the vacuum-cap pages transcribed [ev-0017].
The other twenty-three scan files are image-only, carry no text layer, and have never been read. The site’s own page says the linked PDF parts are a bounded collection that was not ingested [ev-0018]. Any claim about their contents would be a claim about a document nobody here has opened, and none is made.
The dating case
The site’s URL slug says 1933. The evidence points to a multi-year assembly instead [ev-0018].
Begin about 1930 or early 1931. A handwritten notation reads “Catalog finished Dec. 1930,” and the introductory plant roster still pictures the Hazelhurst, Pennsylvania plant, which was closed and demolished in March 1931 [ev-0018].
Content after 1933. A full Liquor Ware Division opens with a joking reference to the return of legal “Spiritus Frumenti” — Repeal came in December 1933 — and the catalogue shows Applied Color Label glassware, a process Owens-Illinois is not documented using commercially before about 1933 [ev-0018].
Content from the mid-1930s. Libbey “Safedge” tumblers including Donald Duck designs, and the character debuted in June 1934; the employee-club term “OnIzed,” introduced in 1934; and Libbey itself, absorbed by Owens-Illinois in 1935 [ev-0018].
The retired engineer Russ Hoenig’s summary — that “it looks like a work in progress… eventually assembled to what we have” — is the position this expert adopts [ev-0018]. A claim sourced to a specific page should not be dated more precisely than 1930–1935 without separate corroboration. That is the single most important practical consequence of the dating case, and it is why the catalogue cannot by itself date a jar.
Provenance
The photocopy was provided to the site by two retired Owens-Illinois engineers, Phil Perry and Russ Hoenig — the latter a co-author of the Owens-Illinois monographs in this field. A cover page signed by dozens of workers calls it “…the first Owens-Illinois Catalogue” [ev-0018].
This account reaches the expert from the site alone and is not independently confirmed.
Structure
The site’s page gives the division breakdown across the 25 files [ev-0018]:
| Files | Division or contents |
|---|---|
| 1–2 | Introductory: company history, glassmaking science, plant roster, indexes |
| 2–10 | Pharmaceutical and Proprietary Division |
| 10–11 | Liquor Ware Division |
| 11–13 | Prescription Ware Division |
| 13–18 | Food Container Division — packers, canning jars, coffee jars, condiments |
| 18–19 | Large Ware Division |
| 19 | Beverage Division |
| 20–21 | Milk Bottle Division |
| 21 | Package Division |
| 21–23 | Closure Division |
| 23–24 | Tumbler Division |
| 24–25 | Mold number index and Classified Index |
Page numbers in the catalogue are alphanumeric by division — L for Liquor, F for Food, Clo for Closure [ev-0018].
The coffee jars, and why they matter
File #15 contains a dedicated “Coffee & Pantry Jars” section — nine moulds across two pages, transcribed in full [ev-0016].
Jars C-1011 and C-1012 are square with waffle-grid embossing and a round label panel, which is the configuration closest to the Hoffmann-Hayman Crystalvac jars. C-1007 and C-1012 at 28 to 30 oz. sit in the capacity range a three-pound coffee jar implies, and the section’s sizes run from 6⅞ oz. to 46 oz. Every jar photographed carries the Owens-Illinois Diamond-Oval-I basemark [@ev-0016, pp. F56-F57].
That establishes documentarily that Owens-Illinois was actively cataloguing coffee jars, in Crystalvac-like styles, inside the window in which Hoffmann-Hayman launched Crystalvac [ev-0016].
The Closure Division completes the picture. “Knife-Opening Vacuum Caps,” lithographed “VACUUM PACKED — Keeps Contents Fresh,” were offered in the “G” GCA 450 thread — the same G-450 finish the coffee jars carry — with a sidebar photograph showing one fitted to a branded coffee jar; knife-opening non-vacuum variants are listed separately [@ev-0017, Clo-11, Clo-12]. Vacuum coffee packaging was a complete Owens-Illinois system, jar and cap, within the catalogue window.
What this does not show. It shows that Owens-Illinois offered jars of this kind. It does not show that any particular Crystalvac jar came from Owens-Illinois rather than Three Rivers, and the catalogue cannot date one, because its own pages date only to a five-year window.
How to cite it
Cite a specific page by its catalogue page number together with the file part, and date any claim drawn from it only to 1930–1935 unless something else narrows it [ev-0018]. For period style names and bottle typing, the site’s author regards this catalogue as the most comprehensive single record of the era’s dominant producer, which makes it useful comparative context even for non-Owens-Illinois ware [ev-0018].
What would improve this article
Reading the other twenty-three files. The two that were read produced the two most useful things in this article — the coffee-jar moulds and the matching vacuum caps — and they were selected because the Crystalvac question pointed at them. A vision-OCR pass over the Food Container and Closure divisions in full, and over the mould-number index, would let a mould number on a real jar be looked up rather than guessed at, which is the capability this article currently lacks.