US 807449

I'll search for authoritative records on this patent number, then synthesize.

Let me verify the litigation angle more specifically before summarizing.

US Patent 807,449 — Verification Note

I searched for the exact number 807449 (and literal variants "807,449" / "8,074,449"). Results returned only the patent itself plus unrelated patents of the same era (US 1,055,027; US 1,185,647; US 837,920) and a German pharmaceutical document (DE 60305446 T2) that coincidentally contains the string "807449" as a mass-spectrometry value. No litigation, PTAB, or CAFC 2026 docket activity for US 807,449 was found. I state this as a negative finding from searches, not as a certified clearance search.

Two cautions on numbers:

• The Federal Circuit (est. 1982) post-dates this patent. US 807,449 was granted in 1905 and its term expired December 19, 1922 (per the Google Patents legal-status record), so modern CAFC docketing is essentially inapplicable.
• The famous "'449 patent" in i4i v. Microsoft is US 5,787,449 — a different patent, different art, different century. Do not conflate.

---

Bibliographic Data (US 807,449)

Abstract

US 807,449 has no formal abstract (abstracts were not used in 1905). Google Patents renders the specification's opening statement of invention as the "Definitions" text. In substance:

The invention relates to improvements in the type-actions of front-strike typewriting machines, which comprise pivoted type-bars and mechanism to swing them upward and backward to printing position on the front of the platen. The object is to provide highly efficient type-actions composed of simple, easily assembled parts.

Plain-Language Overview of the Independent Claims

A structural point first: all 61 claims in this patent are independent — none uses "as set forth in claim ___" dependency language. That drafting style was common in 1905. Rather than 61 separate recitations, the claims fall into a small number of recurring combinations, varied across four dimensions: (a) whether the intermediate element is a generic "lever" or a "sublever"; (b) whether the connection is "means" (generic) or a "pivotally connected link"; (c) pivot elevation arrangement; and (d) relative widths of the sublever system, key-levers and angular levers.

Group 1 — Generic "levers" (claims 1–4). A set of front-strike type-bars plus key-actuated devices, with a system of levers whose two arms are equal in length and whose ends travel on arcs whose chords are substantially horizontal. Claims 1–4 are the same idea with different combinations of generic "means" versus pivoted links at the type-bar end and/or the key end (1 = means/means; 2 = links-to-type-bars/means; 3 = means/links-to-keys; 4 = links/links).

Group 2 — Same idea expressed with "sublevers" and "key-levers" (claims 5–8). Claims 5–7 mirror claims 1–3 but name the key-actuated devices as "key-levers" and the levers as "sublevers." ⚠️ Claim 8 is not recoverable from the source text — its number is skipped and its text is missing in the OCR; claim 7 also ends mid-sentence ("...means connecting the ends"), suggesting OCR breakage in this region.

Group 3 — Uniform-arm systems (claims 9–12, 13). Adds the limitation that the arms are equal "to one another and to those of each other lever of the system." Again varying means-vs-links. ⚠️ Claim 13's opening is likewise garbled — it appears spliced onto the tail of claim 12. Claim 13's recoverable portion covers upper arms to type-bars / lower arms to key-actuated devices.

Group 4 — Sublevers mounted in front of the type-bar pivots (claims 14–17). The sublever system sits in front of a vertical plane cutting the type-bar set near their pivotal axes; upper arms connect to type-bars, lower arms to key-levers (15 uses links; 16–17 add that each lever's ends move side-to-side of a vertical plane through its own pivot).

Group 5 — Progressively raised pivots (claims 18–24). The recurring inventive hallmark: levers/sublevers whose pivotal axes lie at elevations increasing progressively from the middle to the sides of the system. Claims 18–24 cover this across generic levers and sublevers, with and without links, and with the "in front of the vertical plane" placement.

Group 6 — Interchangeability / reversibility (claims 25–28). Frames the manufacturing benefit the specification stresses: all levers of the system are interchangeable, all sublevers are interchangeable, each lever is "normally operative in a reversed position" (i.e., reversible end-for-end without loss of function), or both. The specification explains this springs from equal, symmetrical arms: the parts can be made with a single set of dies and assembled without care as to which end is up.

Group 7 — Straight, symmetrical levers (claims 29–30). Restricts the lever to a straight lever with symmetrical arms (29 with means/means; 30 with links).

Group 8 — Single type-bar claims (claims 31–37). Narrowed to one front-strike type-bar with: a lever/sublever in front of the plane cutting the type-bar near its pivot, upper and lower arms, a key-actuated device (or angular lever) behind the vertical plane containing the sublever's axis, and connecting links extending backward. Claim 36 covers devices forming operative connections between key-lever and type-bar; claim 37 covers "links connecting said levers with one another and with the type-bar."

Group 9 — Relative widths (claims 38–44). The spatial relationship the specification calls out explicitly: the sublever/lever system is equal in width to the set of type-bars, while the key-levers / key-actuated angular levers are wider than the sublever system (because the type-bar set is narrower than the keyboard).

Group 10 — Both sets with progressively raised pivots (claims 45–51). Combines Group 5's raised pivots with both the sublever system and the set of key-actuated angular levers having progressively increasing pivot elevations. Claim 45 additionally requires key-actuated means that move the sublevers' upper ends in substantially horizontal directions away from the vertical plane cutting the type-bars; claims 46–51 add the "arcs more nearly horizontal than vertical" language, parallel-arrangement limitations, and links.

Group 11 — Three-set combinations (claims 52–59). A system of parallel sublevers (width = type-bar set), a set of parallel key-levers (wider), and a third set of parallel levers/angular levers above the key-levers and behind the sublevers, with its width equal to the key-lever set, plus connections between each stage. Claims 56–59 add progressively increasing pivot elevations to both the sublever system and the third set; claim 59 is the most detailed, specifying parallel type-bar-to-sublever connections, key-lever-to-angular-lever connections of gradually increasing length from the center outward, and oblique links pivoted at their rear ends to the angular levers and front ends to the sublevers, "more and more obliquely arranged from the middle of the system toward the sides."

Group 12 — Narrow, single-bar and final combination claims (claims 60–61). Claim 60 is the single-type-bar counterpart with an angular lever and three links (upper sublever arm to type-bar; lower sublever arm to angular lever; downward link from angular lever to key-lever). Claim 61 recites key-levers wider than the type-bar set, sublevers with progressively increasing pivot elevations, ends moving on arcs "more nearly horizontal than vertical," the sublever system at the top being substantially equal in width to the type-bars, links to the type-bars, and means connecting the lower sublever arms to the key-levers.

---

Uncertainty Disclosure

1. Claim count integrity: The patent is recorded as having 61 claims, and the text runs 1–61, but claims 8 and 13 are unrecoverable from the source text owing to an apparent scanning/OCR defect (claim 8 is entirely absent; claim 13's opening is spliced onto claim 12). I have not fabricated their content. If you need exact claim 8 and claim 13 language, retrieve the original document from USPTO PatentCenter or the Google Patents PDF image for US807449A.
2. Assignee/ownership history: Google Patents lists the original assignee as UNION TYPEWRITER CO and the inventor as Jacob Felbel. Google explicitly disclaims having performed any legal analysis of the assignee list; chain of title was not verified here. Note that Felbel filed related, later applications assigned by mesne assignments to Union Typewriter Company (e.g., US 1,055,027, US 1,185,647), indicating an ongoing relationship — but that is context, not proof of recordation for the '449 patent.
3. Litigation: No CAFC or district-court records for US 807,449 specifically surfaced. Historic cases involving Union Typewriter Company (e.g., Union Typewriter Co. v. L. C. Smith & Bros.) concern different patents (Daugherty's Nos. 457,258 and 470,990, per the opinion text) and should not be attributed to 807,449. I cannot rule out an obscure early-20th-century infringement action that is not indexed in the sources searched.
4. "Abstract": The abstract above is a paraphrase of the specification's object statement; no abstract of record exists for this 1905 patent.

Primary source: https://patents.google.com/patent/[US807449](/patent/US807449)/en

I'll search for litigation involving US patent 807449.

The initial searches returned only unrelated modern patents with similar numbers. Let me run more targeted searches to confirm.

Let me run a few final confirmatory searches.

Bottom line

I found no known litigation involving U.S. Patent No. 807,449. No district court case, Federal Circuit/CAFC appeal, ITC investigation, PTAB proceeding, or other adversarial proceeding naming this patent appeared in any source I searched.

I want to be explicit about a caveat: absence of records in modern litigation databases is expected here and is not proof that no suit was ever filed. See the limitations section below.

The patent at issue (as confirmed literally, per the record)

• Patent number: US 807,449 (kept literally as "807449"/"807,449"; not auto-corrected)
• Title: "Type-Writing Machine"
• Inventor: Jacob Felbel
• Original assignee: Union Typewriter Company (Jersey City, New Jersey)
• Application serial no.: 165,540, filed July 15, 1903
• Granted: December 19, 1905
• Term/expiration: Anticipated expiration December 19, 1922 (Google Patents legal-status entry: "Expired – Lifetime"), consistent with the then-applicable 17-year term measured from grant
• Subject matter: front-strike typewriter type-actions (type-bar segment, sublevers, angular levers, links)

Searches performed and why the hits are not this patent

I searched for the number itself and in litigation contexts (Unified Patents portal material, PTAB filings, district court opinions, Federal Reporter excerpts). Every "449"-flavored hit returned was a different patent whose shorthand also happens to be "'449." These are important to flag so they are not mistaken for US 807,449:

None of these involve US 807,449.

I also noted a 19th/early-20th-century opinion snippet referencing that a "key lever mechanism would infringe Felbel's patent." That reference is in a Patent Office interference/priority appeal and is dated c. 1900–1901 — before US 807,449 even issued (1905) — so it cannot refer to this patent. It likely refers to an earlier Felbel patent (e.g., the Felbel/Gabrielson ribbon-feed patent No. 703,339 of June 24, 1902, or No. 686,043). I would treat it as unrelated unless the underlying record is pulled directly.

Reasons litigation records would be absent

1. The patent expired December 19, 1922, meaning any enforcement suit would have been filed roughly 1905–1922.
2. PACER (the federal courts' electronic docket system) generally does not reach back that far — coverage is effectively from the late 1980s/1990s forward, and free litigation aggregators (Unified Patents portal, Docket Alarm, CourtListener/RECAP) are even more modern in scope. A 1905–1922 case file would not appear in these databases even if it existed.
3. CAFC did not exist until 1982, so any appeal from a pre-1922 infringement trial would have gone to a regional circuit court of appeals, not the Federal Circuit.

Limitations and suggested next steps

• I cannot affirmatively rule out historical litigation (e.g., a Union Typewriter Company enforcement suit against a competitor such as Underwood, Remington, or Smith Premier, or a Patent Office interference involving the Felbel type-action). I found no evidence of any, and I will not represent that none occurred.
• If you need a definitive historical answer, the productive sources would be: Federal Reporter / Federal Cases volumes (1905–1923), United States Circuit Court and District Court records in the relevant districts (New York/New Jersey would be most likely given the assignee's location), the Annual Reports of the Commissioner of Patents (for interferences/appeals), and the contemporaneous typewriter trade press (e.g., The Typewriter and Phonographic World, Business Equipment Topics). None of these is indexed in the litigation databases you named.

If you'd like, I can pursue those historical reporter sources specifically, or run a targeted search on Union Typewriter Company as plaintiff from that era.

Proceedings overview

Zero AIA trial proceedings on file. The USPTO Open Data Portal structured block returned no IPR, PGR, or CBM proceedings for US 807449, and targeted web searches for the patent number surfaced no petitions, institution decisions, Final Written Decisions, or Federal Circuit appeals involving this patent — so the breakdown is active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0. The defensive posture is not "the patent survived an IPR and is hardened" and it is also not "claims 1–5 were canceled" — it is that this patent cannot be asserted at all: US 807449 issued 1905-12-19, its anticipated expiration is logged as 1922-12-19, and its legal status is "Expired - Lifetime." Your bottom line is therefore a standing and enforceability problem, not a validity problem. If a demand letter cites US 807449, the letter is citing a patent that expired more than a century ago.

A caution that matters more than the PTAB question: verify the number in the demand letter digit-for-digit. US 807449 (seven digits, 1905, Jacob Felbel / Union Typewriter Co.) is not US 8,074,449 (eight-digit, 8-million series, a 2011-era patent). Several of the "'449" PTAB hits that come up in any search of this number belong to entirely different patents, listed below.

Proceedings on US 807449

None. There is no proceeding to report at claim-level granularity, no judge panel, no institution decision, no FWD, no settlement, and no CAFC appeal. I will not invent a proceeding number to fill this section.

Two structural reasons this is unsurprising, and one is dispositive:

• Timing. IPR and PGR became available for petitions filed on or after 2012-09-16; CBM review ran from 2012-09-16 until its 2020-09-16 sunset. US 807449 expired 1922-12-19 — roughly ninety years before the PTAB's first trial. There is no window in which a petitioner could have filed.
• No enforceable claim scope. The patent's term ended in 1922, so there is nothing to cancel and no infringement exposure to eliminate. An IPR on a long-expired patent is possible in principle but has no rational petitioner; the absence of activity here is a function of the patent's age, not of litigants' inattention. Compare the well-asserted modern patents in this space, which do attract serial petitions — the search hits for other "'449" patents show exactly that pattern (three parallel IPRs against one patent owner, discretionary-denial fights under General Plastic).
• No reexamination either. My searches surfaced no ex parte reexamination on this patent. I state that as "nothing found," not as a verified negative — the ODP block covers AIA trials, not reexam.

False-positive watchlist — do not import these into an 807449 analysis

If your clearance search or a vendor's docket report returns "'449" PTAB activity, it is almost certainly one of these:

All four appear in PTAB E2E: https://ptacts.uspto.gov/

Strategic summary

Claim status on US 807449. Every claim — the specification ends with 61 numbered claims, from claim 1 ("a set of front-strike pivoted type-bars, a set of key-actuated devices, a system of levers each having arms of equal length…") through claim 61 — is untested in any AIA trial, but that is the wrong frame. The correct frame is that all 61 claims are expired, not canceled, not sustained, and not vulnerable; simply unenforceable because the statutory term ran out on 1922-12-19 (seventeen years from the 1905-12-19 grant, under the term then in force). There is no "surviving claim set" to work around and no narrowing that helps or hurts you. The claim numbering is also a trap for anyone quote-mining: claim 13 and claim 18 in the printed patent are garbled in the OCR of the original, and old-style claims like these will not map onto modern infringement theories (e.g., the "key-actuated devices" of claim 1).

Estoppel landscape. There is none, and it cuts in no one's favor. § 315(e)(2) estoppel applies only to a petitioner that obtained an FWD or that settled after institution; with no proceeding, no petitioner, no privy, and no real party in interest is estopped as to any ground. Freedom-of-action analysis is correspondingly simple: you do not need § 102/§ 103 art, an invalidity contention, or a § 112 theory. A § 282 defense of expiration/unenforceability, or simply a statute-of-limitations and standing inquiry into whoever is sending the letter, ends the analysis. Note the practical consequence — because no FWD exists, there is also no PTAB record to point the court to for a § 315(e)(2) or collateral-estoppel shortcut, so any defense must be made on the face of the patent and the term calculation.

Pattern signals. No serial petitioner, no repeat patent owner litigation campaign, no defensive aggregator (no Unified Patents or RPX proceeding), no CAFC appeal. The patent's own family tells the story: Felbel assigned to Union Typewriter Company (Jersey City), and his later work in this same line was assigned to Remington Typewriter Co. (see, e.g., US 1,185,647, US 1,185,299, and US 1,392,408, all listing Jacob Felbel). This is a turn-of-the-century typewriter-mechanism portfolio, not the raw material of a modern assertion campaign. The only modern-art antecedent worth noting is that old front-strike type-bar geometry does resurface as prior art against later patents — but it would be cited as a printed publication, not as an 807449 PTAB record.

Recommended next steps

• If you are a defendant: do not spend budget on an IPR. The patent expired 1922-12-19 (Google Patents records the anticipated expiration at https://patents.google.com/patent/US807449/en). Go straight to a Rule 12 motion or a letter replying that the asserted patent is expired and unenforceable, and demand the serial number, issue date, and a certified copy of the assignment chain. Ask for the exact patent number in writing — if opposing counsel meant 8,074,449 or 6,895,449, the PTAB analysis above changes entirely and you should re-run this exercise against the correct number.
• If you were expecting a hardened patent: the absence of PTAB activity here is emphatically not the "well-asserted patents eventually attract IPRs" signal. It is the opposite: a 1905 typewriting-machine patent that expired in 1922 has never been the subject of an AIA petition because there is nothing to petition against.
• If a proceeding does surface: ODP may lag recent filings. Re-run the search against PTAB E2E (https://ptacts.uspto.gov/) and the PTAB Decisions page before relying on this memo. I found none, but I cannot rule out a filing made after the ODP ingest date.

Confidence note: I have high confidence in the "no AIA trial activity" conclusion for US 807449, based on both the ODP structured block and independent searches. I have not verified the reexamination history (no AIA-style structured data covers it) or any historical pre-1922 litigation, which would predate any electronic docket and would be a separate archival exercise.

I'll search for the assignment record before writing anything.

Ownership-Chain Reconstruction — US 807,449 ("Type-writing machine")

Method note up front, because it governs everything below. I could not execute a live query inside the Assignment Center itself with the tools available to me — the tool is a general web search, and assignmentcenter.uspto.gov is a JavaScript application whose result sets are not exposed as crawlable pages. What follows is therefore a negative search result, not a confirmed null query. I also hit a hard constraint that predates any of this: the USPTO's own training material states the Patent Assignment Search "contains all recorded Patent Assignment information back to August 1980" (USPTO, What are Assignments?, webinar deck, 2015-01-20, p.2). A 1903-executed assignment, if recorded, sits on pre-1980 microfilm reel/frame — outside the searchable window. I did not fabricate a reel/frame to fill the gap.

---

Inventors

Employer at time of filing. Not stated on the face of the patent and not determinable from any source I retrieved. The contextual evidence points at the Union Typewriter Company orbit: Felbel's later, related applications were assigned to Union Typewriter Company (US 1,055,027, "assignor, by mesne assignments, to Union Typewriter Company") and to its successor Remington Typewriter Company (US 1,185,647, US 1,185,299, US 1,392,408). His 1938 New York Times obituary describes him as "an authority on patents and typewriter inventor... who formerly was with the Underwood Elliott Fisher Company and with the Remington Typewriter Company and its predecessors" (NYT, 1938-12-14, p.25). I treat the employer determination as inferred, not evidenced.

Unusual patterns — two, and neither is the classic one.

1. Not present: inventor departure. There is no twelve-month-departure signal. Felbel's filing relationship with Union/Remington spans at least 1903–1919 across multiple patents, so this is a sustained in-house inventor relationship, not a portfolio pre-fire-sale tell.
2. Present: inventor is also a patent practitioner. Felbel appears as counsel — not as a party — in Howell v. Hess, a Court of Appeals of the District of Columbia interference appeal over a type-bar dust shield: "Mr. William W. Bodge, Mr. Daniel A. Carpenter, and Mr. Jacob Felbel for the appellee" (CourtListener). This matters for the correspondent analysis below: on his own later filings, the attorney of record and the inventor may be the same natural person.
3. Name-variant hazard. The Times headline renders the surname "FELVEL" while the body text and the patent read "FELBEL." Any assignor/assignee index search must be run both ways.

Related but non-dispositive datum (do not import into this patent's chain). Howell v. Hess records that in December 1900 Underwood's solicitors concluded that "the key lever mechanism would infringe Felbel's patent." That predates US 807,449's July 1903 filing by two and a half years, so it cannot refer to this patent. It is evidence of Felbel's earlier patent estate in front-strike type-actions — most plausibly US 703,339 (filed 1900-08-10) or a sibling — and it is why I would not treat a search hit for "Felbel patent" litigation as automatically belonging to 807,449.

---

Original assignee

Union Typewriter Company, of Jersey City, New Jersey, a corporation of New Jersey — named on the face of the grant.

Shipped a product embodying the claims? Yes, on the record's own terms, though I did not verify it machine-by-machine. The specification is written as a production machine's type-action (segment-mounted type-bars, sublevers, angular levers, a shiftable segment). The Museum of Science and Technology "Leonardo da Vinci" catalogs Felbel, Jacob as "progettista" (designer) of the Monarch typewriter, activity dates 1887–1923 (Lombardia Beni Culturali, record ST120-00043). The Monarch was Union's own front-strike visible-writing answer to the Underwood, and the Monarch Typewriter Company was founded by Union in 1904 — the year after this application was filed. Note that US 1,185,647 (Felbel, assigned to Remington) repeatedly specifies construction "as usual in the Monarch machine" and describes the "Monarch machine," reinforcing the Felbel–Monarch design link.

Primary line of business. A manufacturing combination/trust — not a licensing entity. Formed 1893 out of Remington Standard plus Smith Premier, the American Writing Machine Company (Caligraph/New Century), Yost, and Densmore; president Clarence W. Seamans. Contemporaneous reporting in the Salt Lake Herald (via Utah Digital Newspapers) calls it "the typewriter trust," controlling Remington, Smith-Premier, Monarch, New Century, Yost and Densmore, and states it was "organized under the laws of New Jersey in 1893" (newspapers.lib.utah.edu). Later patents in the family (US 1,055,027, 1913) describe it approvingly as a "corporation of New Jersey."

Current status — dissolved / absorbed. Union was reorganized in 1913 and renamed Remington Typewriter Company, with Seamans continuing as president; the Monarch brand's factory was shut down and production moved to Remington's plant in 1914–15 (willdavis.org/Union.html). I flag that this source is a hobbyist history, not a corporate record. The subsequent chain — Remington Typewriter → Remington Rand (1927) → Sperry Rand (1955) → later divestitures — is general corporate history. I am not confident about the later disposition of the typewriter business and I am not asserting any of it as a recorded patent transfer.

⚠️ Contradiction flagged. Google Patents lists the "Current Assignee" as UNION TYPEWRITER Co and the "Original Assignee" as UNION TYPEWRITER CO (https://patents.google.com/patent/US807449/en). Google expressly disclaims having performed any legal analysis of the assignee list. Since the named entity was reorganized away in 1913 and its affiliates' assets passed through Remington, "Union Typewriter Co" as of 2026 is a face-of-the-patent artifact, not a live ownership determination. Treat Google's assignee field as bibliographic, not operative.

---

Assignment timeline

No assignment records for US 807,449 were retrieved. I searched the patent number, the inventor's name in both spellings (FELBEL / FELVEL), and the assignee name against indexed sources, and found no reel/frame entry, no conveyance type, no recorded date, and no correspondent of record. The reasons are structural, not evidentiary gaps in my search:

• Searchable window. The USPTO states assignment-data coverage "back to August 1980" (source above). A 1903/1905 transaction, if recorded at all, lives in the pre-1980 bound-volume/microfilm series and is not surfaced by a name or patent-number query.
• No post-1922 transfer is even possible in principle. The term expired 1922-12-19. Every assignment regime, correspondent convention, and NPE entity the task asks me to compare against post-dates that date by decades or a century.
• The only ownership fact I can document is on the face of the patent itself.

The one documented transfer, from the grant, not the Assignment Center:

• Executed on or before 1905-12-19 / recording date not retrievable — Reel not retrievable / not available in the searchable series
  • Conveyance: Assignment (direct, pre-issuance; inferable from publication-on-the-face of the assignee)
  • Assignor: Jacob Felbel, of the borough of Manhattan, City of New York, New York
  • Assignee: Union Typewriter Company, of Jersey City, New Jersey, a corporation of New Jersey
  • Correspondent: Not stated on the printed patent text available to me. Note the practitioner overlap above — for a self-filing inventor-agent, the correspondent and the assignor may be the same person.
  • Context: Inventor-to-employer/company assignment, standard for the era. Not a fire-sale, not a securitization.

Evidentiary corroboration that this was a direct (not mesne) assignment. The face of US 807,449 reads "JACOB FELBEL... ASSIGNOR TO UNION TYPEWRITER COMPANY." By contrast, Felbel's later US 1,055,027 reads "JACOB FELBEL... ASSIGNOR, BY MESNE ASSIGNMENTS, TO UNION TYPEWRITER COMPANY" (US 1,055,027 PDF). "By mesne assignments" is patent-drafting shorthand for a broken chain that had to be papered through intermediate assignees before reaching Union. US 807,449 is on the direct-assignment side of that line. That is a real, if modest, chain-of-title finding: there is no evidence of any intermediate holder between Felbel and Union for this patent.

Subsequent corporate history (not recorded assignments; listed so it is not mistaken for one).

Any 1913 reorganization transfer of Union's patent estate would have been effectuated by corporate merger/reorganization rather than a discrete recorded assignment, and by 1913 this patent had nine years of term left — so a bulk transfer of record is not implausible, but I found none and I will not assert one.

---

Timeline diagram

Per the task's stop condition I am short-circuiting the analytical sections after this point; the diagram records only dated facts I can actually verify, and I have deliberately kept the corporate-reorganization entry out of the chart because it is not a recorded assignment.

timeline
    title Ownership of US 807449
    1903 : Application filed by Jacob Felbel
         : Serial 165540 filed Jul 15
    1905 : Patent granted Dec 19
         : Face names Union Typewriter Co
    1922 : Patent term expired Dec 19
    2026 : No USPTO assignment records found

Annotations: the 1905 entry is a face-of-the-patent assignment, not a reel/frame entry. The 2026 entry is a negative search result, not a confirmed null query inside the Assignment Center.

---

NPE / troll-pattern signals

Rendered for completeness — the verdict framework explicitly contemplates a "no records" outcome — but every call below is constrained by the 1922 expiration, which makes the comparison to modern NPE lists structurally inert.

---

Verdict

Insufficient data — no Assignment Center records, and only the original assignment is documented (and that one only from the face of the grant, not from a reel/frame entry).

Justification. The chain I can actually evidence is exactly one link: Jacob Felbel → Union Typewriter Company, executed on or before the 1905-12-19 grant, corroborated by the face of the patent and by the absence of "by mesne assignments" language that appears in his later US 1,055,027. No reel/frame, execution date, recording date, or correspondent is retrievable, because the USPTO's searchable assignment window opens in August 1980 and this transaction is a pre-1980 record if it was recorded at all. Every subsequent event in the corporate line — the 1913 Union → Remington Typewriter reorganization, Remington Rand (1927), Sperry Rand (1955) — is corporate history I cannot tie to a recorded transfer, and the patent itself expired 1922-12-19, which forecloses any modern NPE signal by chronology alone. There is no evidence of a shell transfer, a known asserter, a cascading LLC chain, a pre-suit transfer, a bankruptcy sale, privateering, or defensive aggregation.

Verify: USPTO Patent Assignment Center and assignment.uspto.gov — search by patent number 807449, and separately by assignor FELBEL, JACOB / FELVEL, JACOB. If the Assignment Center returns nothing on the number, request the pre-1980 bound-volume record or a certified copy of the assignment abstract directly from the USPTO Assignment Recordation Branch; that is the only route to a reel/frame for this patent.

Confidence. High confidence that no searchable assignment record exists for US 807,449 in the post-1980 electronic series. Low confidence, and I want this stated clearly, on whether the pre-1980 paper record contains a recorded Felbel → Union assignment or any 1913 reorganization filing — I did not obtain that record, and the negative result above should not be read as certifying that the paper chain is empty.

I'll research the citation record for US 807449 across USPTO and patent databases.

Let me verify the citation record directly from the patent's own front page and Google Patents citation data.

US 807,449 — Citation Record and Prior-Art Analysis

1. Identification (verified against the record)

The USPTO-derived record at https://patents.google.com/patent/US807449/en matches the identifier literally: US 807,449 A, "Type-writing machine," Jacob Felbel (Manhattan, N.Y.), assignee Union Typewriter Co. (Jersey City, N.J.), application US 165,540, filed 1903-07-15, granted 1905-12-19, anticipated expiration 1922-12-19, CPC B41J 7/02. That identification is the starting premise and is not repeated here.

A targeted PatentCenter/USPTO-database search for the string "807449" returned no direct USPTO-hosted hit — the only exact-string match was an unrelated German document (DE 60305446 T2) in which "807449" appears as a mass-spectrometry value. The authoritative text used below is therefore the digitized printed patent (the OCR rendering on Google Patents), not a live PatentCenter bibliographic record. I flag that so the sourcing is not overstated.

Distractor note (one line, per prior sections): do not read any of the "'449" hits below as this patent — US 8,074,449, US 6,895,449, US 5,787,449, and US 9,999,449 are different patents.

---

2. The decisive finding on "patent citations for 807449": there are none of record

I could not enumerate citations for US 807,449 because the patent has no citation record to enumerate. This is a substantive finding, not a search failure. Three independent reasons:

1. The printed 1905 grant contains no "References Cited" list. The examiner-citation front-page convention on US patents did not exist in 1905; patents of this era printed only the drawing sheet(s), specification, and claims. The full text of US 807,449 as fetched contains the bibliographic block, "Prior art keywords," classification, the specification, and 61 claims — and no references-cited section whatsoever.
2. The Google Patents record shows no backward-citation ("Patent Citations") table and no forward-citation ("Cited By") table. What it does show is a "Similar Documents" table — an algorithmic CPC/text-similarity listing, not a citation record. Treating "Similar Documents" as citations is the most likely error in this analysis, and I am expressly not doing it.
3. The citation record for this era lives in the file wrapper, not the patent. For a 1903 application, examiner's references, the applicant's Information Disclosure-equivalent, and any interference papers are in the application file. Those 1903–1905 file wrappers survive (wholly or partly) in NARA Record Group 241 and, where digitized, in USPTO PatentCenter. That is where citations for US 807,449, if any, must be retrieved. I did not retrieve them, and I will not fabricate a list to fill this section.

Bottom line for the citation question: zero patent citations of record. Any list of "citations for 807449" produced by a vendor or an AI that shows a neat table of US patents cited on the face of this patent should be treated as fabricated or as a mis-mapped "Similar Documents" table.

---

3. Most relevant prior art — independently identified

Because there is no citation list to report, the useful work is to identify the art that would have been cited, and the art that bears on the claims today for historical/validity purposes.

3.1 The governing law — important framing correction

US 807,449 was filed 1903-07-15, so it was prosecuted and granted under R.S. § 4886 (the 1870 Act as amended), not under 35 U.S.C. § 102, which was enacted in 1952. The pre-AIA § 102 categories map onto the old statute in substance, but two old-statute rules materially change the analysis and are routinely overlooked:

• There was no § 102(e)-style "secret prior art" in 1903. A third party's US patent issued after the applicant's filing date was not prior art against the application. Only references granted or published before the applicant's invention date (for § 102(a)-analog), or more than one year before the filing date — i.e., before 1902-07-15 (for § 102(b)-analog), could be used.
• Consequently, most of the documents in the "Similar Documents" table are legally incapable of anticipating US 807,449 despite being topically near-identical, because they post-date the filing.

That single point eliminates the majority of the apparent "prior art" surface.

3.2 Candidate references, with statutory treatment

3.3 The most likely true closest art — and it is not in any database section

The single most probative prior art for a patent of this type is ordinarily the interfering application or patent of a rival, and there is a documented signal that one existed. Prior work in this analysis flagged a Patent Office interference/priority appeal of roughly 1900–1901 containing the statement that a "key lever mechanism would infringe Felbel's patent." That proceeding pre-dates the 1905 grant, so it concerns an earlier Felbel patent — but it establishes that Felbel's type-action claims were contested in interference practice before 807449 issued. Interference papers are the one place where the actual anticipatory reference is named. Retrieving that interference record is the highest-value single action available on this patent, because (a) it identifies the rival's structure, and (b) it tells you which of the 61 claims the examiner and the rival considered the real invention — which explains why the claim set is drafted as 61 narrow, structural variations on one lever geometry.

3.4 Non-patent art (must not be ignored for a 1903 filing)

For a 1903 application the practical § 102 risks were public use and known-or-used-by-others, not printed patents:

• Production front-strike machines of the 1898–1903 period — Underwood's front-strike machines, Smith Premier, and Remington — as physical devices, catalogues, and trade descriptions. Any of these showing a key-lever → sublever → link → type-bar train with the sublever on a fixed support would bear directly on claims 1–12 and 31–37.
• Contemporaneous typewriter trade press — The Typewriter and Phonographic World, Business Equipment Topics — for dated disclosures of the Union Typewriter Co. machine itself. Note the specific hazard: the patent is assigned to the manufacturer, so the assignee's own pre-filing sale or public use of the machine is a § 102(b)-analog bar risk, not merely third-party art.

---

4. Claim-level anticipation assessment (screening opinion only)

Overall assessment: I found no reference that anticipates any claim of US 807,449. The realistic historical risks were (i) an interference-based loss of particular claims, (ii) the two-year statutory bar if Felbel's own pre-1901-07-15 patent or the assignee's own public use is documented, and (iii) a "lack of invention" (§ 103-analog) attack on the groups whose elements appear separately in the art (38–44, 45–59). Claim 8 and the opening of claim 13 remain unrecoverable from the OCR and cannot be assessed.

---

5. Limitations — stated explicitly

1. No citations to report. US 807,449 has no references-cited section in the printed grant and no citation tables in the digitized record. The "Similar Documents" table is algorithmic similarity, not citation data. I did not obtain the file wrapper.
2. Dates marked "not captured" must be verified before any of this is relied on. For US 687,500 and US 686,043 the exact issue dates are outcome-determinative under the two-year bar, and I did not verify them. I have not guessed them.
3. I did not retrieve the disclosures of the pre-1903 in-class documents (US 576,329; 626,915; 643,786; 657,927; 668,713; 679,675; 686,487; 697,250; 700,687). They are the correct retrieval targets; mapping them to claims without the texts would be fabrication.
4. The governing statute is R.S. § 4886, not 35 U.S.C. § 102. § 102 as a codified section is a 1952 enactment; the substantive categories map backwards, but the absence of a § 102(e) analog in 1903 is a real and often-missed limitation on what can be cited.
5. The interference record (c. 1900–1901, referencing infringement of "Felbel's patent") remains the best lead on the true closest art and has not been pulled.
6. Claim 8 is entirely missing from the OCR and claim 13's opening is spliced onto claim 12, consistent with the defect already flagged in this analysis.

Next steps, in priority order: (1) pull the US 807,449 file wrapper from USPTO PatentCenter / NARA RG 241 for examiner's references and interference papers; (2) verify issue dates for US 687,500 and US 686,043; (3) retrieve the disclosures of the eight pre-1902 in-class documents; (4) search Federal Reporter/Federal Cases 1905–1923 for Union Typewriter Company enforcement actions and for the c. 1900–1901 interference opinion; (5) search the typewriter trade press 1898–1903 for the assignee's own machine disclosures.

I'll research the prior-art documents listed on the patent page before building the § 103 analysis.

Good — I now have a key lead: US 657,927 (Felbel, Sept. 18, 1900) appears both in the patent's Similar Documents list and in the 1900 Census of Manufactures list of notable front-strike patents. Let me pull the remaining references.

Obviousness Analysis — US 807,449 (Jacob Felbel / Union Typewriter Co.) under 35 U.S.C. § 103

Scope note. The task directs me to the "Prior Art" section of the US 807449 page — i.e., the Similar Documents list at https://patents.google.com/patent/US807449/en. One structural caution governs everything that follows: that list is similarity-ranked, not date-filtered. It is not an examiner's citation list, and it contains no rejection, no art unit annotation, and no indication of which documents the Office actually applied. Fifteen of the twenty-four entries post-date the '449 filing date and are therefore chronologically incapable of being prior art. I triage them below rather than treating the list as a ready-made § 103 attack.

I also flag at the outset that both previously generated sections establish the patent expired 1922-12-19. This analysis is therefore retrospective — it is the analysis an examiner, a defendant, or a later patentee citing 807449 as art would have to perform, not a live invalidity posture.

---

1. Governing law

US 807449 was filed 1903-07-15 with no priority claim (Google Patents logs the priority date as identical to the filing date). The pre-AIA version of § 103 therefore governs, applied through Graham v. John Deere Co., 383 U.S. 1 (1966), as refined by KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007).

The critical date for statutory bars under pre-AIA § 102(b) is 1902-07-15. Two doctrinal points do real work here:

1. Pre-AIA § 102(b) contains no "own work" exception. A patent granted to an inventor more than one year before that inventor's U.S. filing date is a statutory bar against that inventor's own later application. This matters acutely because the '449 list contains US 657,927 — Felbel, granted 1900-09-18, more than two years before the '449 filing.
2. Pre-AIA § 102(a) requires "by others," so Felbel's own pre-1902 patents are not § 102(a) art — but they are § 102(b) art, and § 102(b) art can be combined for § 103 purposes.

---

2. Triage of the "Prior Art" section by date

The practitioner's takeaway: the top of the list that Google ranks as "most similar" — US 788,763 (1905) — is later than the patent in suit and is worthless as § 103 art. The usable set is the eleven pre-1902-07-15 references. I was able to retrieve substantive text for US 556,460, US 576,329, US 626,915, US 643,786, and partial text for US 687,500 (which is not on the list and which I flag separately below). I could not retrieve text for US 657,927, 668,713, 679,675, 686,487, 697,250, or 700,687 — a verification gap addressed in § 8.

---

3. Level of ordinary skill in the art (PHOSITA)

Circa 1903: a mechanical designer with several years' experience in typewriter development, familiar with the front-strike machines then on the market (Remington, Underwood, Monarch, Densmore, Smith Premier, Daugherty), competent to design multi-stage lever-and-link trains, and attentive to the two commercial pressures of the period — visible writing (front strike) and manufacturing economy through interchangeable, die-formed parts. No witness would need extraordinary skill to swap one lever arrangement for another in this art; the machine was a mature assemblage of bars, links, and bell-cranks by 1903.

---

4. The state of the art, corroborated

The strongest available corroboration of the § 103 landscape comes from outside the patent list. The 1900 U.S. Census of Manufactures (Vol. X, p. 818) surveys typewriter patents and states that "in late years inventors have been unusually diligent in perfecting what is generally known as 'front-strike' machines," naming as examples Daugherty 553,158 (1896), Wagner 559,345 (1896) and 633,672 (1899), and Felbel 657,927 (1900) — https://www2.census.gov/library/publications/decennial/1900/volume-10/volume-10-p6.pdf.

This is significant in three ways:

• It is a federal government printed publication, usable itself as art.
• It establishes that the "front-strike type-action" problem was a recognized, crowded field by 1900 — the KSR "design incentive" and "known problem" factors apply forcefully.
• It places Felbel's own 657,927 in the same recognized family as Daugherty and Wagner. Whatever 657,927 discloses, it is same-inventor, same-field, § 102(b) art.

---

5. Disclosure summaries of the usable references

US 556,460 — Daugherty, "Type-writing machine," 1896-03-17 (https://patents.google.com/patent/[US556460](/patent/US556460)).

From the retrieved text, this reference discloses, in one machine:

• A curved end piece C carrying type-bar-supporting plates or hangers G on curved rods H and I, with the pivotal points of the type-bars "arranged in steps, one above the other" and "also arranged in horizontal steps and in series or rows."
• "Parallel bars which form the support for the levers which are placed between the key-levers and the type-bars" (Fig. 11), with "a modified form of support or pivotal rod for the said levers which are between the type-bars and the key-levers" (Fig. 12) — i.e., a mounted system of intermediate levers interposed between the keyboard and the type-bars.
• Bell-crank-form levers P connected to the finger-keys.
• A case-shift mechanism (transverse shaft 3, inward arm 1, arms 5 with operating pieces 4 at opposite sides of the keyboard) that raises the curved end piece C and elevates the type-bars — the same segment-shift function as the '449's shift-levers 45 / lifting-levers 46 / links 47.

This is, element for element, the architecture of '449 claim 1: front-strike pivoted type-bars + key-actuated devices + an intermediate lever system + connections at each end.

US 576,329, 1897-02-02 (https://patents.google.com/patent/[US576329A](/patent/US576329A)/en). Discloses a series of connecting-rods, a series of type-bar-operating "sub levers," sublevers of different lengths arranged over a universal bar, and fulcrum-posts 14 disposed upon a plate or support 15 so that a stated desideratum is obtained — i.e., the geometry of the sublever system is deliberately varied across the system to produce a uniform result. The reference also recites the aim of light, uniform key touch.

US 626,915 — Little, "Type-writer," 1899-06-13 (https://patents.google.com/patent/[US626915](/patent/US626915)). Franklin-class machine in which "the type-levers are disposed in the arc or segment of a cylinder." The stated object is "to provide an arrangement by which the keys may be located close together, occupying the same relative positions as in the square keyboard of the Remington," achieved by re-arranging the key-lever pivots. This is directly relevant to the '449's width-relationship limitations.

US 643,786, 1900-02-20 (https://patents.google.com/patent/[US643786A](/patent/US643786A)/en). Type-levers in horizontal alignment guided by stationary and spring-loaded guide arms g, h toward a single central printing point, with key-levers pivoted mid-length on a rod and connected by rods to angular projections on the type-levers. Relevant as a secondary teaching that disparate lever positions across a set are converged on a common printing point by guide geometry — the general principle behind the '449's guided, progressively-elevated lever system. It is the weakest of the usable references.

US 657,927 — Felbel, 1900-09-18. Same inventor. Listed on the face of the '449 page and identified by the 1900 Census as a notable front-strike patent. Text not retrieved. Its legal significance is disproportionate to what I can say about its content: as a § 102(b) statutory bar by the same inventor, it is the single most dangerous document in the list, and if its disclosure overlaps the '449 type-action it converts much of the § 103 case into a § 102(b) case.

US 687,500 — not on the Similar Documents list, surfaced in my searches (https://patents.google.com/patent/[US687500A](/patent/US687500A)/en). Discloses type-bars connected by rearwardly-extending links to the upper ends of upright bell-cranks, the lower portions of which are pivoted on horizontal key-levers; a comb 29 guiding key-lever treads; and — critically for the '449's width claims — the recitation that "the width of the key-lever system at the junction of the sublever system therewith is substantially equal to the width of the type-bar system or to the system of links, the bell-cranks extending vertically from the key-levers to the links," while the keyboard itself is wider. The drafting vocabulary ("sublevers," "antifriction-rollers," "width of the key-lever system at the junction of the sublever system") is closely allied to the '449 and to Felbel's later US 1,055,027. I could not confirm its inventor, assignee, or filing date; if it is pre-1902-07-15 art, it is highly material to claims 38–44, 52–59, and 61. Treat as an investigative lead, not a verified reference.

Felbel's later patents as evidence of the art's recognized objectives. US 1,055,027 (Felbel, filed 1911-07-13, assigned by mesne assignments to Union Typewriter Co.) states the objects as providing "uniform leverage and a uniform dip of the keys throughout the system" and "elimination of all, or substantially all, side strain or pull in each action throughout the system" (https://patentimages.storage.googleapis.com/2d/cb/cc/9e1724531c88ef/US1055027.pdf). Those are precisely the results the '449 attributes to its equal-arm, progressively-elevated sublever system ("only slight lateral or twisting strains are developed"). A reference post-dating the '449 is not § 103 art, but it is admissible evidence of what the art was trying to achieve — i.e., evidence that the '449's asserted advantage was a recognized goal, not an unexpected discovery.

---

6. The § 103 attack, by claim group

The claim groups from the prior section are used as given; claims 8 and 13 remain unrecoverable and are not assessed.

---

7. Motivation to combine — stated expressly

A § 103 rejection must articulate why. Four motivations are available and independently sufficient:

1. Same field, same problem, same solution space. All the references are type-actions for front-strike typewriters. The 1900 Census confirms this was a single recognized class of machine being actively perfected by multiple inventors including Felbel himself.
2. Explicit design incentive from the patent's own specification. The '449 states that equal, symmetrical arms allow the parts to be "formed with a single set of dies," are "interchangeable," and are reversible. That is the applicant's own stated reason for the structure — squarely a manufacturing-economy rationale.
3. Known technique applied to a known structure. Stepping or progressively elevating pivots across a set of levers was known (556,460). Applying it to the intermediate lever set and to the key-actuated lever set of the same machine, for the same purpose, is the predictable use of a known technique to improve a similar device.
4. The asserted advantage was a recognized goal of the art. The '449's stated benefit — that "only slight lateral or twisting strains are developed" — matches the objects Felbel himself later recited in US 1,055,027: "uniform leverage and a uniform dip of the keys throughout the system" and "elimination of all, or substantially all, side strain or pull in each action throughout the system." When the goal is known and the mechanism is a predictable rearrangement of known parts, KSR forecloses patentability.

---

8. Where the § 103 case is strong, and where it is weak

Strongest: claims 18–24, 45–51, 56–59, 61 as to the progressive-elevation limitation — because US 556,460 expressly recites pivots "arranged in steps, one above the other" and "in horizontal steps and in series or rows." Comparison of these claims against 556,460 should be done line-by-line from the full text rather than from the abstract snippet I retrieved.

Weakest / most resolvable only by claim construction: the "arcs whose chords are substantially horizontal" and "more nearly horizontal than vertical" limitations (claims 1–30, 46, 61). These are functional-geometric characterizations of a lever's end-travel. A defendant would argue they are met by any sublever whose ends swing predominantly horizontally; a patentee (in the counterfactual world where this mattered) would argue they impose a specific angular relationship not disclosed. The width-relationship claims (38–44) are the ones most vulnerable to a genuine evidentiary gap, since I have not confirmed that any pre-1902 reference recites the sublever-system-width-equals-type-bar-set-width relationship in express terms — US 687,500 appears to, and verifying it is the single highest-value next search.

Counterarguments a patentee would raise:

• No express TSM in the references for combining 556,460 with 576,329 or 626,915. Post-KSR this is not dispositive, but a pre-KSR adjudication (which is what a 1905–1922 court would have applied) would have required it, and the record contains no "references suggest combining" language.
• Secondary considerations. The '449 was assigned to Union Typewriter Company, an aggregator of typewriter manufacturers; commercial success of machines embodying the type-action is theoretically probative — but there is no nexus evidence in the record, and the specification offers no comparative test data.
• § 112 problems cut against the patentee, not for it. Sixty-one independent claims with "means connecting" language and purely functional geometry would today be assailable under § 112(f)/(b) (and pre-1952, as indefinite and over-claimed). That is a validity problem, not a § 103 defense.

---

9. Verification gaps — be explicit

1. US 657,927 (Felbel, 1900-09-18) is unread. It is same-inventor, § 102(b) art listed on the face of the '449 page and identified as a front-strike patent by the 1900 Census. This should be read first; it could convert substantial portions of the analysis above from § 103 to § 102(b).
2. US 668,713; 679,675; 686,487; 697,250; 700,687 are unread. Only their titles and dates were confirmed from the list. Any could be closer art than 556,460.
3. US 687,500 is unverified as to inventor, assignee, and filing date and is not on the "Prior Art" section at all. Its width-relationship disclosure, if pre-1902-07-15, is the best art I have seen for claims 38–44 and 61.
4. US 556,460's full text was not reviewed. My disclosure summary rests on the Google Patents excerpt, which is substantive but partial; the stepped-pivot language in particular should be read in context in the printed patent.
5. Filing dates for the post-1903 publications (US 739,622; 788,763; 799,121; 887,976; 911,644; 972,940; 984,654; 990,169; 1,028,214; 1,030,503; 1,099,213; 1,099,270; 1,181,972; 1,564,329; 2,184,735) were not verified, so pre-AIA § 102(e) "secret prior art" status is indeterminate for them. Most are so late that they cannot qualify; that is a reasoned conclusion, not a verified one.
6. No on-sale or public-use inquiry was made. Given the Union Typewriter Company's commercial activity after 1901, a pre-1902-07-15 public use or sale of a machine embodying this type-action would be an independent § 102(b) bar. Not investigated.

---

10. Bottom line

On the record available, the strongest § 103 combination is US 556,460 (Daugherty, 1896) as the primary reference, alone or in combination with US 576,329 (1897) for the sublever system and US 626,915 (Little, 1899) for the wide-keyboard/narrow-type-set problem. Daugherty alone appears to disclose the full architecture of claim 1 — front-strike pivoted type-bars, key-actuated devices, an intermediate lever system mounted between them, and a shifting segment — and, importantly, expressly discloses pivots "arranged in steps, one above the other" and "in horizontal steps and in series or rows," which is the geometry the '449 treats as its distinguishing feature across claims 18–24, 45–51, 56–59, and 61. The motivations to combine are grounded in the references themselves (a common, crowded field and a common problem), in the '449 specification's own admission that equal, symmetrical, reversible arms exist to permit a single set of dies and careless assembly, and in the art's expressly recognized objectives of uniform leverage and elimination of side strain.

The principal risk to this conclusion is not legal, it is evidentiary: I have read only part of the relevant art. Before relying on this analysis, read US 657,927 (same inventor, § 102(b)), then US 556,460 in full, then verify US 687,500. If 657,927 discloses the same type-action, this stops being a § 103 question and becomes a § 102(b) one — and, given the patent expired 1922-12-19, either answer is academic as to enforcement.