US 109372

I'll search for this specific patent number and check for any current litigation.

US 109,372 — Verified Summary

Note on the number: US 109,372 is an 1870 patent. There is no modern "US 10,9372" or AIA-era application here — I interpreted the identifier literally as US Patent No. 109,372 (Google Patents lists it as US109372A, with a family/application variant US109372D).

Bibliographic data

Plain-language overview of the claims

The meter is a two-cylinder positive-displacement water meter in which two pistons in each cylinder are joined by a connecting-rod; the water pressure drives the pistons, and their reciprocation turns a crank that drives a rotary valve. The stated point of novelty is the shape of the slot in the connecting-rod plus the end-of-stroke stops — the slot is not straight-sided but is "wider in the middle than at the ends" (concave on each side of center). That geometry lets the crank-pin coast through its dead-center inside the widened part of the slot after the pistons have already been halted by the stops. The stated benefit: one piston can never be forced to move by the other while its valve is closed, giving cleaner valve timing.

Claim 1 (printed "l."): The piston-carrying connecting-rods having the described slots — formed with concavities to the right and left of the center — in combination with the stops o o, substantially as set forth. In plain terms: a piston/connecting-rod unit with a centrally-widened (cam) slot, claimed together with the stops that arrest the pistons at the end of the stroke. Both elements are necessary; the claim is a combination claim, not a claim to the slot alone.

Claim 2 (printed "2."): The piston-carrying connecting-rods having the described slots in combination with the crank m and crank-pin n, constructed and operating substantially as set forth. In plain terms: the same widened-slot connecting-rod, claimed this time together with the crank and the pin that rides in the slot.

Both claims are "substantially as set forth"-type combination claims, and neither is dependent on the other. There are no method claims and no dependent claims.

Litigation and prior art (relevant context)

• The only significant litigation over this invention is Union Water-Meter Co. v. Desper, 101 U.S. 332 (1880) (decided October Term 1879; Justice Bradley). It was brought on reissue RE 5,806, not on the original 109,372. The Court held reissue claim 4 (a five-element combination: rotary valve, two cylinders, double-acting pistons with a cam-slot in the rod, adjusting stops, and a crank-shaft) was not infringed by Desper's structure under his patent No. 144,747 (Nov. 18, 1873), because Desper's meter dispensed entirely with the crank-shaft, an essential element of the claimed combination. Syllabus point 5 in 101 U.S. (Reprint) summarizes it: RE 5,806 "are not infringed by letters-patent No. 144,747." Sources: https://tile.loc.gov/storage-services/service/ll/usrep/usrep101/usrep101332/usrep101332.pdf ; https://www.courtlistener.com/opinion/90131/water-meter-co-v-desper/
• The same decision notes the invention was not novel in all its parts: projecting stops and the crank-shaft-to-slot arrangement were said to appear in an Ericsson 1851 water-meter (slide valves rather than rotary), and the rotary valve plus cylinder/piston-rod/crank-shaft combination was the subject of an earlier Ball & Fitts patent of July 20, 1869 — which the 1870 specification itself acknowledges as the "similar" parent device.
• Google Patents lists as "similar documents" several contemporaneous fluid-meter patents, e.g. US13320A (Water-meter, 1855), US118392A, US119364A, US127126A, US144747A (Desper, 1873), US188214A (Water-meters, 1877), US337845A (Piston water-meter, 1886), and the reissue USRE5806E.

What I could not verify — please treat as uncertain

1. Filing date. No source I retrieved states an 1870 application filing date. Pre-1871 applications were not numbered in the modern serial-number fashion, so the "filing date" field is genuinely unavailable from these sources.
2. Assignee at issuance. The printed patent names only the inventors; whether there was an unrecorded assignment at issue is unknown. The Union Water-Meter Company relationship is documented only from the 1874 reissue onward.
3. 2026 FRCP/CAFC docket activity: I found none, and none is plausible. My web searches returned no Federal Circuit or district court docket for US 109,372 in 2026. This patent expired November 22, 1887 and has been in the public domain for roughly 139 years, so it cannot be asserted. I could not query PACER or the CAFC docket directly with the tools available — so this is "no evidence found," not a certified negative docket check.
4. OCR artifacts in the source text. The Google Patents OCR renders the slot letter inconsistently as "c" in the definitions list and "e" elsewhere; the correct element letter for the slot appears to be e (consistent with the reissue claim 4 language quoted in 101 U.S. at 334: "a single cam-slot, e"). The first claim is printed as "l." — that is a numeral 1, not a letter. I have not silently corrected these in the quoted claim text above.
5. Assignees listed on datamp.org are shown as "not known," and datamp notes the meter is "not known to have been produced" — but the existence of the Union Water-Meter Company and its 1870s price list ("PRICE LIST OF WATER METERS... MANUFACTURED BY THE UNION WATER METER CO., WORCESTER") suggests commercial manufacture did occur, at least of related Ball & Fitts meters.

Primary sources:

• https://patents.google.com/patent/US109372/en
• https://patentimages.storage.googleapis.com/4f/af/0e/a64aad84b50ab4/US109372.pdf
• https://www.datamp.org/patents/search/advance.php?pn=[109372](/patent/109372)&id=48670&set=1
• Water-Meter Co. v. Desper, 101 U.S. 332: https://tile.loc.gov/storage-services/service/ll/usrep/usrep101/usrep101332/usrep101332.pdf

I'll search for litigation records on this patent number.

Let me verify with more targeted searches, including the specific litigation databases.

Found a key case. Let me dig deeper into the district court record and any other suits.

Bottom line

Yes — one litigation is documented, but it is a 19th‑century case, and it was brought on the reissue of US 109,372, not on the original patent. There is no modern litigation (no district court, ITC, PTAB, or Federal Circuit activity) involving US 109,372, and there cannot be: the patent issued in 1870 and expired in 1887.

---

1. The known case

Claim 4 of the reissue as quoted in the opinion (the "instrumentalities" of the combination): the rotary valve g with ports; two cylinders b, b′; double‑acting pistons c, c′ each with rod d and a single cam‑slot e wider than the crank wrist n, permitting each crank wrist to pass its dead‑centre; adjusting stops o; and a crank‑shaft i transmitting motion from the pistons to the valve. The Supreme Court noted the patent "does not cover any of the separate parts of the meter, it being conceded that these were all known and used before the application," and identified the Ericsson 1851 water‑meter patent (and the earlier July 1869 Ball & Fitts patent referenced in the 109,372 specification) as prior art.

Sources: https://supreme.justia.com/cases/federal/us/101/332/ · http://www.worldlii.org/us/cases/federal/USSC/1879/190.html · https://tile.loc.gov/storage-services/service/ll/usrep/usrep101/usrep101332/usrep101332.pdf · https://case-law.vlex.com/vid/company-v-desper-water-885307848

---

2. Identifier caveats you should carry forward

• The asserted instrument was the reissue, not the original. Any litigation search on "109,372" alone will miss this case; the case is indexed under RE 5,806 / 101 U.S. 332. Google Patents' family listing for the 109,372 family does show a member USRE5806E, "Improvement in liquid-meters," dated 1874-03-24, which corroborates the reissue identity.
• Transcription conflict on the reissue number. The published Supreme Court opinion text reads "reissued letters No. 5806." At least one secondary aggregator (sooperkanoon.com) renders it "No. 5808" in its syllabus text. Per your instruction not to auto‑correct identifiers, I am flagging both as they appear rather than silently harmonizing them; the authoritative U.S. Reports text and the Google Patents family listing both point to 5806.
• The Google Patents "family has litigation" banner is unverified. The page carries a boilerplate Darts‑IP link (family=2178845) with the label "Family has litigation / First worldwide family litigation filed." I could not open or verify any concrete proceeding behind that flag; I treat it as consistent with, but not independent evidence of, the 1879 case.

---

3. Negative findings (what I did not find)

• No modern litigation. No district court complaints, no ITC § 337 investigations, no PTAB IPRs/PGRs, and no Federal Circuit appeals involving US 109,372. The Unified Patents litigation caselist, PTAB/PTActs records, and CAFC opinion searches returned nothing for this number.
• No other 19th‑century suits confirmed. I found only the one reported case. It is plausible Union Water-Meter Company brought other infringement bills in the 1870s–1880s that were never reported or are not digitally indexed, but I have no evidence of any, and I will not speculate about parties or docket numbers I cannot ground.
• Databases that cover modern litigation cannot help here. PACER begins in the late 1980s; Docket Navigator, Unified Patents, and Darts‑IP track modern dockets. A pre‑1900 equity case is reachable only through the reporters and Supreme Court records, which is how the case above was located.

---

4. Do not confuse this with similar numbers

Your search results were dominated by an entirely different patent — US 11,019,372 ("Scale Video Coding LLC v. Cisco Systems, Inc.," E.D. Tex. Nos. 2:23‑cv‑… and 4:25‑cv‑00095; PTAB IPR2024‑01371; Alice motion; Equitable IP). That is a 2021 patent, unrelated to Ball & Fitts. I have excluded it, and any litigation list for "109372" that cites Scale Video/Cisco is a wrong‑number match.

I'll verify the structured "no proceedings" signal against public sources before reporting.

Let me do one more targeted check for any PTAB/CAFC activity under the exact patent number, and to rule out confusion with the unrelated seven-digit patents indexed as "'372".

Proceedings overview

Zero AIA trial proceedings. The structured PTAB block (USPTO ODP) returns no IPR, PGR, or CBM for US 109372, and independent web searching on the exact patent number surfaces no petition, institution decision, Final Written Decision, or Federal Circuit appeal — the breakdown is therefore 0 active / 0 claims invalidated by PTAB / 0 claims sustained by PTAB / 0 settled / 0 institution denials. Defensive posture: "no PTAB history exists because none can meaningfully exist" — this is an 1870 Letters Patent (Phinehas Ball and Benaiah Fitts, "Improvement in Liquid-Meters," granted 1870-11-22) whose term ran out in 1887, and every modern assertion risk runs through its 1874 reissue, not the original. This is the opposite of the "the patent has survived two IPRs and is hardened" scenario; there is no PTAB record to either help or hurt you.

Important disambiguation before you rely on any search hit. The terms '372 patent and 109372 are heavily polluted by unrelated live patents in PTAB and district court dockets. None of the following involve US 109372:

• US 11,019,372 — Cisco Systems, Inc. v. Scale Video Coding LLC, IPR2024-01371 (claims 1-3, 5-8, 10-13, 15 challenged under § 103 over Shao-028 and Shao-028 + Shao-423). This is a video-coding patent with a Cisco/Scale Video district court case; it is a seven-digit patent number, not the 1870 six-digit patent.
• US 11,737,372 — Western Digital v. Godo Kaisha IP Bridge (magnetic tunnel junction / MgO oxygen-vacancy art).
• US 8,408,109 — Provisur Technologies v. Weber (food-slicing).
• The 2024 laser-weeding '372 patent in Carbon Autonomous Robotics Systems v. … (C.D. Cal. 2:24-cv-03012).

If a demand letter or a vendor's invalidity search points you at an "IPR on the '372 patent," confirm the full number and the issuing year before treating it as relevant. A six-digit 1870 patent number is not a substitute for a modern 11,xxx,xxx number.

Proceedings on file

None. There is no {PROCEEDING_NUMBER} — {Petitioner} v. {Patent Owner} entry to report, and I will not manufacture one. For completeness, an empty record means there is also no:

• Judge panel — N/A (no PTAB panel ever convened).
• Petition grounds — N/A.
• Institution decision — N/A.
• Final Written Decision — N/A; no claim of US 109372 has ever been canceled or sustained by the Board.
• Settlement / termination — N/A.
• Appeal — no CAFC appeal of a PTAB decision exists.

Historical adjudication of this patent (NOT a PTAB proceeding — flagged for context)

The only judicial adjudication of US 109372's subject matter I can verify is a 19th-century Supreme Court case. Treat it as historical context, not as an IPR substitute:

• Case: Union Water-Meter Co. v. Desper, 101 U.S. 332 (1879) — appeal from the Circuit Court for the District of Massachusetts. CourtListener: https://www.courtlistener.com/opinion/90131/water-meter-co-v-desper/
• Patent at issue: reissued Letters Patent No. 5806 (some sources render the number 5808; Google Patents lists USRE5806E, "Improvement in liquid-meters," 1874-03-24), issued 1874-03-24 to the Union Water-Meter Company as assignee of the original No. 109,372 (granted 1870-11-22 to Ball and Fitts).
• Asserted claim: claim 4 of the reissue — a combination claim to "a rotary valve, g, provided with suitable ports or openings … two cylinders, b and b' … the double-acting" pistons, piston rods, slotted connections, crankshaft with pinion and cranks, and "adjusting stops" in the piston-rods/slots.
• Disposition: non-infringement affirmed. The Court (Bradley, J.) held the defendants' meter (built under Desper's US 144,747, granted 1873-11-18) omitted a material element of the claimed combination and supplied no equivalent, applying the rule that "the claim of a combination is not infringed if any of the material parts of the combination are omitted," and that "the patentee makes them all material by the restricted form of his claim."
• Notable prior art discussed: the earlier Ball & Fitts patent granted 1869-07-20 (the patentees themselves conceded the rotary valve and the cylinders/piston-rods/crankshaft/rotary-valve combination were the subject of that earlier patent); and "a patent granted to Mr. Ericsson in 1851 for a water-meter having slide valves instead of a rotary valve." The opinion notes the only additional elements in the 1870 patent relative to the 1869 patent were the adjusting stops and the rectangular (enlarged-middle) slots in the piston-rods — which aligns with the two claims actually issued in US 109372 (claim 1, slotted connecting-rods with concavities + stops o o; claim 2, the slotted connecting-rods + crank m and crank-pin n).
• Defensive value today: effectively nil as an IPR matter, but as a claim-construction and scope point it is unusually strong for a defendant — the patentee's own recoverable scope was construed narrowly and confined to a five-element combination that a differently-arranged meter avoided. Separately, note the record's "Family has litigation" flag on the Google Patents page points to Darts-ip family ID 2178845 (https://patents.darts-ip.com/?family=2178845); that is the 19th-century litigation, not a modern campaign.

Strategic summary

Claim status of US 109372. No claim has been canceled, narrowed, or amended by the PTAB, because no AIA petition was ever filed. Google Patents labels the patent "Expired - Lifetime" with an anticipated expiration of 1887-11-22 (17 years from the 1870-11-22 grant) — consistent with the 1874 reissue litigation being the last enforcement activity. So, on the output taxonomy you asked for: claims 1 and 2 of US 109372 are UNTESTED by the PTAB and unavailable to enforce by operation of term expiry; the reissue Re. 5806 claim 4 was adjudicated (non-infringed) in 1879 and has likewise long since expired. A defendant facing a demand letter citing "US 109372" or "Re. 5806" has, in the ordinary course, a threshold problem for the plaintiff that has nothing to do with prior art: the asserted right cannot be infringed because it no longer exists. That is a Rule 11 / § 285 conversation, not an IPR.

Estoppel landscape. § 315(e)(2) estoppel is inapplicable — there is nothing to estop. Estoppel attaches to a petitioner (and its privies/real parties in interest) that was party to an instituted IPR that reached a final written decision; no IPR was instituted here, so no petitioner is barred, and no defendant inherits a narrowed claim set or an adverse FWD. Conversely, there is no FWD to borrow as a "the Board already killed this claim" argument. Every prior-art ground — including the Ericsson 1851 water-meter reference discussed in Water-Meter Co. v. Desper, the Ball & Fitts 1869-07-20 patent, and Desper's US 144,747 — remains "available" in the trivial sense that no tribunal has ruled on it, but § 102/§ 103 art is irrelevant to a patent that expired in 1887. Note also that because the patent predates the AIA, it is not a first-inventor-to-file patent, so PGR is unavailable by definition and CBM would fail the "financial product or service" (technological-invention) tests.

Pattern signals. No repeat petitioner (no petitioner at all). No PTAB appeals pursued by Ball/Fitts or the Union Water-Meter Company — the assignee's enforcement was a single 1879 Supreme Court appeal. There is no defensive aggregator (Unified Patents, RPX, etc.) in the chain, and no litigation-funding or IPR-financing signal. The absence of any PTAB activity over a century and a half of AIA-trial availability is not the "well-asserted patents eventually attract IPRs" signal you'd normally flag — it is the signal of a patent that expired in 1887 and is not a rational target for a post-grant challenge.

Recommended next steps

1. If you are a defendant and someone is asserting US 109372: do not spend money on an IPR. First, demand the plaintiff's chain of title and the asserted claim numbers, and confront them with the patent's face — granted 1870-11-22, expired 1887-11-22 (Google Patents: https://patents.google.com/patent/US109372/en). If the assertion is instead on the reissue, note the record's discrepancy (Re. 5806 per the opinions' bodies and Google's USRE5806E; 5808 appears in one secondary source's syllabus rendering) and that Union Water-Meter Co. v. Desper, 101 U.S. 332 (1879) held reissue claim 4's combination not infringed where a material element is omitted — https://www.courtlistener.com/opinion/90131/water-meter-co-v-desper/. Preserve a copy of the 1879 opinion; it is directly quotable on the narrow scope of the combination claim.
2. Confirm the number, not the nickname. Before acting on any "the '372 patent" reference, verify you are not looking at US 11,019,372 (Cisco v. Scale Video, IPR2024-01371), US 11,737,372 (Western Digital v. IP Bridge), or another modern '372. The empirical search results are dominated by those, and conflating them would produce a wildly wrong validity and estoppel picture.
3. No trial-stage calendar to track. There is no institution-decision deadline, no oral hearing date, and no statutory one-year FWD due date to monitor, because no trial was instituted. The only "next milestone" concept that applies is a real-world one: determine whether the assertion is an error, a records-search artifact, or a bad-faith demand.
4. Verify the negative yourself before filing anything. Re-run the check on PTAB E2E / PTAB Center (https://ptacts.uspto.gov/ptacts/) and the PTAB Decisions library (https://www.uspto.gov/patents/ptab/decisions) using the full six-digit number, and confirm the Office's own file history shows no AIA trial for US 109372. The absence of PTAB activity is itself the answer here — not a gap to be filled by inference.

I'll verify the assignment-record situation before drafting anything.

Now let me check whether pre-1980 assignment records are retrievable at all, and what became of the historical assignee.

Let me confirm the Smithsonian's institutional history and check whether NARA has digitized the pre-1957 assignment series.

Screening note before the chain

One threshold fact determines this entire analysis, so I'm stating it first.

The USPTO's Patent Assignment Search / Assignment Center — the primary source you specified — contains only assignments recorded back to August 1980 (USPTO Assignment Database Webinar deck, https://www.uspto.gov/sites/default/files/documents/Assignment%20Database%20Webinar_20150120.pdf; Finnegan Prosecution Pointer 119; Lexology, "Who owns what: finding patent assignment information," 2016-03-14). US 109,372 issued 1870-11-22 and its term ended 1887-11-22. It therefore cannot appear in that database as a matter of system coverage, not merely as a matter of the patent being obscure.

Pre-1957 assignment records are by regulation held by NARA, not USPTO: 37 C.F.R. § 1.12(a)(2) — "All records of assignments of patents recorded before May 1, 1957, are maintained by the National Archives and Records Administration (NARA)… Certified and uncertified copies of those assignment records are provided by NARA upon request" (https://www.customsmobile.com/regulations/title37_chapterI_part1_subpartA_subjgrp74_section1.12). MPEP 301.01 repeats this and adds that pre-1957 records cannot even be ordered by reel/frame from the Office. My second search timed out before I could confirm whether the relevant RG 241 assignment series has been digitized in NARA's catalog, so I will not assert that it has or hasn't.

Consequence: there is no reel/frame to cite anywhere in this report. No recorded-assignment metadata — no conveyance type, no recording date, no correspondent of record — exists for this patent in any source I could retrieve. Anything I wrote as "Reel NNNNNN/NNNN" would be fabricated. The chain below is reconstructed from a court of record instead.

---

Inventors

Pattern check — "inventors departing within 12 months of filing": not present, and structurally impossible. Both inventors were founders of the assignee company two years before this patent issued, and the specification itself states the meter is "similar in its operation to that patented by P. Ball and B. Fitts, July, 1869" — i.e., this is the second patent in a continuing Ball-&-Fitts program, not a lone inventor's asset. There was no employer/employee split to produce a departure. No fire-sale precursor exists here.

One caution on element naming that carries into any assignment analysis: the Google Patents OCR of the definitions list renders the connecting-rod slot as "c," while the body and the 1879 Supreme Court opinion both use e ("a single cam slot e," 101 U.S. at 334). The OCR is wrong; the slot is e. Likewise the first claim is printed "l." — a numeral 1, not a lowercase L.

---

Original assignee

None is named on the face of the issued patent. The printed specification names only "PHINEHAS BALL AND BENAIAH FITTS, OF WORCESTER, MASSACHUSETTS." No assignee line, no corporate designation. On the document as issued, the inventors were the owners.

The de facto original assignee, established by the Supreme Court, is the Union Water Meter Company, Worcester, Massachusetts — the company the inventors had founded in 1868. Water-Meter Co. v. Desper, 101 U.S. 332, 333 (1880): reissued letters No. 5806 were "a reissue of original letters patent No. 109,372, granted 22d November, 1870, to Phinehas Ball and Benaiah Fitts… the reissue being made to the complainant as assignee, on the 24th of March, 1874." The patent was thus assigned to Union Water Meter Company at some point on or before 1874-03-24.

Did the assignee ship a product embodying the claims? Yes, unambiguously. The 1870 specification is itself the product description for a reciprocating-piston meter, and the Smithsonian holds working Union Water Meter Company instruments:

• Rotary-piston meter, Worcester, MA, serial no. 155,010, ca. 1920 — "Union's example was deemed the best and was awarded a silver medal" at the Massachusetts Charitable Mechanics Association fair of 1874.
• Columbia turbine meter, serial 141,208, and a King disc meter (introduced 1905), both Union products.
• "By 1890, Union had made and sold some 31,000 meters. Most were of the rotary form, and most were used on hydraulic elevators and locomotive stand pipes."

Smithsonian institutional history: "The Union Water Meter Company was established in 1868, by Phinehas Ball, a hydraulic engineer who had worked on the design and construction of the Worcester Water Works, and a local mechanic named Benejah Fitts. Its initial aim was manufacturing the reciprocating meters designed by Ball and Fitts. The firm was reorganized in 1875, prospered for many years, and folded in the mid-1930s." (https://collections.si.edu/search/detail/edanmdm:nmah_1411265)

Primary line of business: water-meter manufacture — reciprocating-piston meters first (the 1870 patent line), then rotary-piston, disc, and turbine meters sold into house service, hydraulic elevators, and locomotive standpipe markets.

Current status: dissolved. The company folded in the mid-1930s — roughly half a century after US 109,372 itself expired. There is no successor assignee of record that I can evidence. I found no bankruptcy proceeding, no 10-K/8-K, and no acquisition record for the firm; the mid-1930s wind-down appears to have been an ordinary failure of a family-scale meter maker, not a distressed portfolio event. I did not find a successor-in-interest, and I will not guess at one.

---

Assignment timeline

The USPTO Assignment Center has no records for this patent. That is a coverage fact, not a gap in my search: the database begins August 1980 (USPTO Assignment Database Webinar, n.p.; Finnegan Prosecution Pointer 119; Lexology 2016-03-14), and pre-1957 records sit with NARA under 37 C.F.R. § 1.12(a)(2). No reel, frame, conveyance type, recording date, or correspondent of record exists to report.

Per your instruction I would stop here. Because the ownership question is nonetheless answerable from a different class of primary source, I'm supplying the remaining sections in compressed form, each labeled for what it actually is, rather than leaving them blank. Flag if you'd prefer the hard stop.

The only documented ownership events, neither of them a USPTO recordation:

• On or before 1874-03-24 — assignment from Ball and Fitts to Union Water Meter Company. Conveyance type not stated in the source; evidenced only by the reissue being "made to the complainant as assignee." No reel/frame. Assignor's motive: the assignee was their own company, founded 1868 to manufacture these meters — an internal founder-to-company transfer, not a sale.
• 1874-03-24 — Reissue RE 5,806 granted to Union Water Meter Company (a reissue is a new grant, not an assignment; it nonetheless re-titles the property in the company's name).

Commercially significant fact about the reissue: the asserted reissue claim 4 was a five-element combination — rotary valve g; two cylinders b and b'; double-acting pistons c, c' each carrying a rod d with a single cam slot e wider than the crank wrist; adjusting stops o; and a crankshaft i. Ball and Fitts broadened their position on reissue, and the broadening failed.

---

Timeline diagram

timeline
    title Ownership of US 109372
    1868 : Union Water Meter Company founded
    1870 : Patent issued to Ball and Fitts
    1874 : Assigned to Union Water Meter Company
         : Reissue RE 5806 granted
    1875 : Union reorganized
    1879 : Suit against Desper decided
         : Patent held not infringed
    1887 : Patent term expires
    1935 : Union folds mid 1930s

---

NPE / troll-pattern signals

Signals 1–8 are not present, with one qualification explained at the end. The reason is temporal, not evidentiary: every signal in your list presupposes a post-1990s assertion apparatus (reel/frame recordations, licensing-only LLCs, repeat correspondent attorneys, defensive aggregators). None of that infrastructure existed, and the patent expired 1887-11-22 — roughly 106 years before Intellectual Ventures was founded. A 2026 NPE reading is not merely unevidenced; it is chronologically foreclosed.

1. Shell-entity transfer — not present. No licensing-only LLC ever touched this patent. The one transfer was founder-to-own-company. Union Water Meter Company had products in commerce (31,000 meters by 1890; Smithsonian-held specimens).
2. Known asserter in the chain — not present. No assignee matches Acacia, Marathon, IV, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, or any Spangenberg entity. Checking Unified Patents or the RPX directories for an 1870 patent would return nothing, and I did not find any listing. The assignee is a defunct 19th-century meter manufacturer.
3. Repeat correspondent across the chain — not present / unascertainable. No correspondent of record exists to check, because no electronic recordation exists. The nearest analogue on the document is the witness block: "JOHN C. OTIS" and "ABRAHAM FITTS." John C. Otis (Worcester, MA) signed as a witnessing attorney on adjacent Ball-and-Fitts-era instruments; Abraham Fitts appears to be a family member. I want to be explicit: witnessing a deed is not the same as being correspondent of record, and I am not treating Otis as a repeat-player correspondent finding. This is an observation, not a signal.
4. Cascading transfers — not present. One transfer, in the direction of the inventors' own operating company.
5. Pre-litigation transfer — not present. The transfer preceded the first litigation by years (by 1874-03-24; the Desper suit was decided in the October Term 1879). More importantly, the transfer's purpose was manufacture, not assertion — the company existed to build the meter, and the Smithsonian record shows it built them for six decades. Note also the direction of the reissue: Union broadened claim scope in 1874 and then lost on the merits, which is the opposite of the well-crafted pre-suit continuation practice that signal 5 is designed to catch.
6. Bankruptcy fire-sale — not present. No Chapter 7/11, no § 363 sale. Union Water Meter Company simply folded in the mid-1930s — by which point US 109,372 had already been in the public domain for nearly fifty years, making the patent worthless as an asset in whatever wind-down occurred.
7. Privateering — not present. No operating company transferred to an NPE. Union sued Desper itself, on its own patents, in its own name, against a direct market competitor.
8. Defensive aggregator — not present. The chain ends at dissolution, not at RPX, AST, LOT, Unified, or OIN. Inverse-signal check: the patent was not neutralized by a defensive aggregator; it was neutralized by the passage of its own term in 1887.

The one qualification, and it cuts the other way. If you want the historical analogue, US 109,372 is a clean specimen of operating-company assertion: Union Water Meter Company made and sold the patented meters and sued a competitor, Henry A. Desper — who held his own competing patent, No. 144,747 (1873-11-18) — for infringement of the reissue. That is precisely the fact pattern your verdict rubric calls "Operating-company assertion," and the Court's finding that Desper's meter omitted the crankshaft — "The patent, as it stands, occupies very narrow ground. It requires the presence of every one of the elements specified" — is a 19th-century illustration of the same narrow-combination problem that produces modern § 112 and Alice losses.

---

Verdict

Insufficient data.

There are no recorded assignments to analyze: the Assignment Center database covers only August 1980 forward (https://assignment.uspto.gov/patent/index.html), and pre-1957 records are in NARA's custody under 37 C.F.R. § 1.12(a)(2), so this 1870 patent has no reel/frame, conveyance, recording date, or correspondent of record in any source I could retrieve — I will not invent them. The single documented transfer (Ball and Fitts → Union Water Meter Company, effective on or before 1874-03-24, evidenced by the reissue grant in Water-Meter Co. v. Desper, 101 U.S. 332) was a founder-to-own-company transfer to a manufacturer that shipped the patented meters, so no NPE signal is present, but the record is too thin for anything stronger than "insufficient data."

Verification link: https://assignment.uspto.gov/patent/index.html — search patent number 109372. Expect a null result; that null is the finding.

To actually complete the chain, the records must be ordered from NARA, Record Group 241 (Records of the Patent and Trademark Office), pre-1957 patent assignment records — not from USPTO. If you want, I can be pointed at that problem next: the useful targets are (a) the 1868 incorporation records of Union Water Meter Company with the Massachusetts Secretary of State, (b) whether the Ball & Fitts assignment was executed before or after the 1869 companion patent issued, and (c) the mid-1930s wind-down filings, which would establish whether any successor-in-interest exists.

I'll research this systematically — first confirming what the USPTO record shows for US 109372, then working through each cited/similar reference.

Now let me look for the two prior-art references actually identified in the record — the earlier Ball & Fitts patent and the Ericsson meter — and check whether Google Patents lists any formal citations.

I found the Ericsson reference (US 7,869) and the record confirms the earlier Ball & Fitts patent. Let me pin down the Ericsson date/number and try to identify the 1869 Ball & Fitts patent.

Bottom line first

US 109,372 has no patent-citation list at all — not on its face and not in any index. I searched for citations against the number and found none, because there is nothing to find: this is an 1870 grant, and the "References Cited" / "Cited By" apparatus on U.S. patents did not exist at that time. Anyone who hands you a § 102 anticipation chart for 109,372 "based on the citations" is working from something other than the patent.

What does exist is a thin, fully traceable set of prior art: one reference the inventors admitted in the specification itself, and one reference the courts supplied during the only litigation over this family. I lay both out below, together with the algorithmic "Similar Documents" list that Google Patents shows — which is not a citation list, and which is mostly post-dating.

---

1. What the USPTO-derived record for 109372 actually contains

Process caveat, stated plainly: I could not execute a native USPTO Patent Public Search / PatentCenter query with the tools available to me. Pre-1871 U.S. patents are image-only in USPTO full-text search, so a number search there returns a scanned document rather than a searchable citation record. The record I analyzed is the USPTO-sourced Google Patents entry (https://patents.google.com/patent/US109372/en), the scanned original (https://patentimages.storage.googleapis.com/4f/af/0e/a64aad84b50ab4/US109372.pdf), and datamp.org's USPTO-derived entry. No citation list appears in any of them.

Identifier discipline (per your instruction): a literal search on "109372" also surfaces US 11,019,372, US 2005/0255372 A1, US 8,394,551, CN 110352371, and similar strings. Those are unrelated modern documents of different length and are excluded from this analysis. Likewise USRE5806E is not prior art — it is the 1874 reissue of 109,372 itself.

---

2. The prior art that genuinely applies to US 109,372

Reference A — Ball & Fitts, patent of July 20, 1869 (same inventors)

Reference B — J. Ericsson, U.S. Patent No. 7,869, "Water Meter," patented January 1, 1851

---

3. The Google Patents "Similar Documents" list — not citations, and mostly not prior art

The 109372 page carries a 25-record list headed "Similar Documents." These are algorithmic similarity matches, not citations. I reproduce all 25 as the record shows them, with a prior-art status determination:

Tally: 9 of 25 pre-date the November 22, 1870 grant; 16 do not.

Three points you should not skip:

1. These are not citations. Google Patents' "Similar Documents" is a text-similarity output. It carries no legal weight and was not considered by any examiner — there was no examination record of this kind in 1870.
2. Items 7, 9, 11, 18 have partially destroyed titles ("Self and w," "Henry," "Samuel w," "William a") — OCR failures, not real titles. I did not retrieve their disclosures, so I cannot state what they teach and I will not guess. They are flagged as date-qualified § 102 candidates requiring document-level review.
3. US 144,747 (item 1) is the single most important record here — and it is not prior art. It post-dates 109,372 by three years; it is the patent under which the accused meters were built. Google's similarity algorithm surfaces it next to 109,372 because the two meters are structurally close, not because it anticipates.

---

4. Claim-by-claim § 102 analysis

Legal-framework caveat (do not skip): 35 U.S.C. § 102 as such did not exist in 1870 — Title 35 was enacted in 1952. The governing provision was § 24 of the Patent Act of 1870, which barred a patent where the invention "had been patented or described in any printed publication" before the applicant's invention (the antecedent of § 102(a)) and contained no one-year § 102(b)-style grace period. I map to § 102 nomenclature only as a convenience. Anticipation requires every claimed element in a single reference.

Claim 1 — connecting-rod with centrally-widened (concave) slot, in combination with stops o o

• Ball & Fitts 1869 supplies the connecting-rod, pistons, cylinders, rotary valve and crank-shaft, but not the concave slot and not the adjusting stops (101 U.S. at 334 expressly identifies those two as "the only additional elements in the present patent"). Alone: no anticipation (§ 102), but a strong § 103-type combination case.
• Ericsson US 7,869 supplies both the stops and the piston-rod-to-crank lost-motion connection — its claim 2 is nearly verbatim the claimed combination. The only possible point of distinction is that Ericsson's opening is described as an elongated hole / yoke slot of plain form, not "wider in the middle than at the ends." That geometric recital is the thin reed on which claim 1 survives. Potentially anticipatory under § 102(a), with that one limitation in genuine dispute.

Claim 2 — connecting-rod with the widened slot, in combination with crank m and crank-pin n

• Ericsson US 7,869 claim 1 discloses two pistons connected to two cranks of a crank shaft so each rests at stroke end while the other drives — i.e., crank + crank-pin + lost-motion coupling. Potentially anticipatory under § 102(a) save for the concave slot geometry and the rotary-valve context.
• Note that claim 2 does not recite the stops. It is therefore the broader and more exposed of the two claims; and the litigation record shows the whole inventive contribution was conceded to be combination-only — "it being conceded that these [separate parts] were all known and used before the application."

Combination-claim vulnerability carried forward from the prior sections: 101 U.S. 332 held claim 4 of the reissue not infringed because the accused structure omitted the "adjusting stops" element, while also holding that a court may not declare any element of a combination claim immaterial. That doctrine cuts both ways here: if stops are an essential element of claim 1, then a slot-only reference cannot anticipate claim 1 — but by the same token, claim 1's scope is narrow.

---

5. What I could not verify, and what I am not asserting

1. The number of the July 1869 Ball & Fitts patent — unknown. Neither the specification nor 101 U.S. 332 states it. I did not locate it and have not fabricated one. It is the single most useful missing document for a complete § 102 file.
2. No "References Cited" or "Cited By" data exists for 109372, and I found none. Stated as a positive finding, not an absence of effort.
3. Contents of the nine pre-1870 "Similar Documents" (items 2, 5, 7, 8, 9, 11, 17, 18, 19) were not retrieved. I verified dates and titles only. Titles like "Self and w" and "Henry" are OCR truncations of inventor names, so those four are effectively unidentified documents.
4. The defendants' pleaded prior art in the 1870s suit is unavailable. 101 U.S. at 333 records that Desper's answer specified "various older patents in which, as they allege, it was described, and divers persons who had known and used it" — but the reported opinion does not list them. There may therefore be additional pre-1870 references in the original 1870s record that are not digitally indexed. I have no evidence of their identity and will not speculate.
5. The reissue-number transcription conflict noted in the earlier sections persists and I have not harmonized it: the Supreme Court text reads "No. 5806"; at least one aggregator renders a syllabus as "No. 5808." Google Patents' family listing shows USRE5806E, which supports 5806.
6. On the OC's of the original patent: the slot letter is rendered "c" in the definitions list and "e" elsewhere; the correct letter is e, per the reissue claim 4 quoted at 101 U.S. 334 ("a single cam-slot, e"). The first claim is printed "l." — that is numeral 1.

Sources: https://patents.google.com/patent/US109372/en · https://patentimages.storage.googleapis.com/4f/af/0e/a64aad84b50ab4/US109372.pdf · https://patents.google.com/patent/US7869 · https://patentimages.storage.googleapis.com/4e/2d/7a/45d0e89b3d6b40/US7869.pdf · Water-Meter Co. v. Desper, 101 U.S. 332 (1879): https://tile.loc.gov/storage-services/service/ll/usrep/usrep101/usrep101332/usrep101332.pdf, https://www.law.cornell.edu/supremecourt/text/101/332, https://www.courtlistener.com/opinion/90131/water-meter-co-v-desper/ · The Artizan, vol. 9 (1851) (independent reprint of the Ericsson claims) · https://www.datamp.org/patents/search/advance.php?pn=109372&id=48670&set=1

I'll ground the obviousness analysis in the actual prior-art references listed on the patent page, plus the two references the Supreme Court identified. Let me pull what I can on the key ones.

Good — I located the Ericsson reference. Let me pin down the July 1869 Ball & Fitts patent and check the remaining pre-1870 references.

Obviousness Analysis — US 109,372 under 35 U.S.C. § 103

0. Three framing caveats before the analysis

(a) § 103 is anachronistic to this patent. 35 U.S.C. § 103 was created by the Patent Act of 1952. It did not exist on November 22, 1870. A § 103 analysis of this claim is therefore retrospective and hypothetical — an exercise in applying modern obviousness doctrine to an old claim, not a statement about what any 1870 examiner should have done (and the original examiner apparently allowed the claims without citing Ericsson). Pre-1952 law tested "invention" (the Cuno Engineering "flash of genius" standard, 314 U.S. 84 (1941)), which § 103 replaced with the "PHOSITA" inquiry. This is analysis, not legal advice.

(b) The prior-art window. The effective date is 1870, so only art predating the invention/application counts. Two notable consequences for the Similar Documents list on the patent page:

A careless analyst would cite Desper's US144,747A against this patent. It cannot be done — it issued three years after 109,372. USRE5806E is the patent's own reissue, not art.

(c) The critical limitation. Stripped of boilerplate, both claims reduce to one structural feature: a connecting-rod slot "wider in the middle than at the ends," i.e., concave to the right and left of center. Everything else — two cylinders, opposed double pistons on a common rod, a crank and crank-pin, a rotary ported valve, stops — is, on the patentees' own admission, old.

---

1. The field and the PHOSITA

Field: positive-displacement liquid meters and, more broadly, reciprocating machines using a crank-pin-in-slot (lost-motion) coupling. All of the candidate references are in this field or the directly adjacent one (steam-engine valve gear), which makes them analogous art without any "reasonably pertinent" stretch. KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398, 402 (2007).

PHINEHAS BALL had himself patented a water meter (US 73,430, Jan. 21, 1868), and the specification of 109,372 names a prior Ball & Fitts meter of July 1869 — so the artisans here were literally the same two men.

PHOSITA circa 1870: a machinist/mechanical engineer with shop training, experienced in valve gear and pump/meter mechanics, able to lay out a cam slot, size a crank-pin for lost motion, and file an adjusting stop. Not a person requiring anything beyond routine mechanical skill.

---

2. The single most important reference: Ericsson US 7,869 (1851)

The Supreme Court in Water-Meter Co. v. Desper, 101 U.S. 332, 334 (1879) expressly held that in the patentee's own meter, "Projecting stops for arresting the movement of the pistons, and much of the mechanical arrangement between the crank-shaft and the slots in the piston- rods, used for giving the proper motion to the crank-shaft, are to be found described in a patent granted to Mr. Ericsson in 1851 for a water-meter having slide valves instead of a rotary valve."

That reference is US 7,869 (J. Ericsson), retrieved at https://patents.google.com/patent/US7869 and https://patentimages.storage.googleapis.com/4e/2d/7a/45d0e89b3d6b40/US7869.pdf. Its claims read:

1. Connecting the two pistons with the two cranks of a crank shaft … so that at the end of each stroke of either of the pistons it shall remain at rest while the crank shaft is being impelled by the other piston, so that the valves shall be shifted while the piston is at rest …
2. … determining the range of motion of the pistons by means of stops connected with the cylinders and the pistons, in combination with the connection of the pistons with the crank or cranks, by means of a joint having sufficient play to permit the pistons alternately to remain at rest while the crank shaft continues to rotate …

And its description discloses "the horizontal slot in the yoke to form the connection with the crank pin," anticipates the alternative of "a hole … so elongated as at the end of the up and the down stroke to permit the crank shaft to have an independent motion," and notes that rotary motion is taken off the crank shaft to a registering apparatus via an eccentric.

This is claim 2 of 109,372 with one feature missing. Compare:

Claim 1 fares no better structurally: its two elements are the slotted rod and stops o o — and Ericsson's claim 2 is precisely stops + a slotted/jointed crank connection, claimed as a combination. Ericsson's stops are fixed rather than adjustable, but the Supreme Court expressly noted that Desper's accused meter used "Ericsson's stops projecting inside for arresting the movement of the pistons, though these stops are fixed and not adjustable" — i.e., adjustability was treated as an immaterial commercial variant, not an inventive departure.

---

3. The self-admitted prior art: the Ball & Fitts July 20, 1869 patent

The specification itself states the meter "is similar in its operation to that patented by P. Ball and B. Fitts, July, 1869, the principal difference being in the slots in the connecting-rods." The Court characterized that earlier patent as covering "the rotary valve, and the combination of the cylinders, piston-rods, crank-shaft, and rotary valve," and summed up: "The only additional elements in the present patent are the adjusting stops and the rectangular position of the slots in the piston-rods." (101 U.S. at 335.)

Under Riverwood Int'l Corp. v. R.A. Jones & Co., 324 F.3d 1346 (Fed. Cir. 2002) (en banc), an applicant's own specification statements describing prior art are usable as prior-art admissions. Here the admission is explicit, and it is the patentees' own words. It converts the case into: the entire meter except the slot contour and the stops was already in a patent by the same inventors.

⚠️ I could not verify the patent number of the July 1869 Ball & Fitts patent from any source I retrieved. Both the specification ("July, 1869") and the Court ("20th of July, 1869") agree on the date; the number remains unidentified. I am flagging that rather than guessing.

---

4. The pre-1870 valve-gear and pump art in the Similar Documents list

These supply the technique of a varying-profile slot/cam for controlling when a follower is positively driven versus free:

• US 87,448A, "Improvement in steam-engine valve-gearing" (1869-03-02)
• US 83,903A, "Improvement in valves for steam-engines" (1868-11-10)
• US 70,841A, "Improvement in steam-engines" (1867-11-12)
• US 69,792A (1867-10-15); US 68,748A (1867-09-10); US 64,166A (1867-04-23)
• US 53,373A, "Improvement in oscillating pumps" (1866-03-20)
• US 13,320A, "Water-meter" (1855-07-24)

I retrieved titles and dates only for most of these; I did not retrieve their texts, and I will not assert what any of them disclose. They are cited here only as evidence that (i) varying-profile slot-and-follower geometry and (ii) lost-motion/positive-drive transitions were standard shop knowledge in the 1850s–60s. A separate caveat: my search for US 13,320A returned US 13,198 (Curtis & Hoard, "Water-Meter," July 3, 1855) instead. Per the no-auto-correction rule I am not equating 13,320 with 13,198; the content of US 13,320A remains unverified.

---

5. Claim-by-claim obviousness

Claim 1 — slotted connecting-rod (concavities R/L of center) + stops o o

Primary combination: Ericsson US 7,869 in view of Ball & Fitts, July 20, 1869.

• Ericsson supplies every element except the slot's central widening: stops, slotted rod, crank-pin, and the very functional result claimed (piston halts, crank then passes through).
• The '69 Ball & Fitts patent supplies the meter environment (cylinders, opposed pistons, crank-shaft, rotary valve) that the 109,372 specification concedes.
• The only remaining gap is the shape of the slot.

This is obvious as a matter of routine design. Once the slot is understood as a lost-motion coupling — which Ericsson already teaches, by requiring "a joint having sufficient play" — the slot profile is the single variable controlling (i) how much lost motion exists, and (ii) whether the piston is positively engaged at the ends of the stroke. Enlarging the slot's midsection while tapering it toward its ends is the standard way to obtain a varying velocity ratio from a slot-and-pin drive: wide in the middle = free run-through at dead center; narrow at the ends = positive, cushioned re-engagement. Ericsson's own text identifies the design goal; achieving it by contouring the slot is "the predictable use of prior art elements according to their established functions." KSR, 550 U.S. at 417.

Alternative single-reference route: Ericsson's specification, read in light of the elongated-hole embodiment and the "sufficient play" language, arguably discloses a slot enlarged beyond pin diameter to permit dead-center passage. If so, claim 1 differs only by the degree and distribution of that enlargement — a result-effective variable optimized by routine experimentation. In re Boesch, 617 F.2d 272 (CCPA 1980); In re Applied Materials, 692 F.3d 1289 (Fed. Cir. 2012).

Claim 2 — slotted connecting-rod + crank m + crank-pin n

This claim is the weakest of the two. It recites no cylinder, no valve, no meter. On its face it covers any connecting rod with a centrally-widened slot receiving any crank-pin. Ericsson's claims 1 and 2 recite exactly "connecting the two pistons with the two cranks of a crank shaft … by means of a joint having sufficient play," implemented by "the horizontal slot in the yoke … connection with the crank pin."

• Primary combination: Ericsson US 7,869 alone, or in view of any of US 87,448A / US 83,903A / US 70,841A (varying-profile slot-and-follower valve gear).
• Motivation: same field (positive-displacement liquid meters), same mechanism (crank-pin in slot), same stated objective — let the crank pass dead center after the piston is arrested, so the valve can shift while the piston is at rest, which Ericsson expressly ties to accurate measurement: "By reason of the shifting of the valves while the piston is in a state of rest the quantity of water introduced and expelled from the cylinder at each stroke must be accurately measured."

Note the irony: the Supreme Court observed that "the use of a crank in converting reciprocating into rotary motion is an old device … applied to the steam-engine a century ago," and that "Ball and Fitts had no claim to it, but only to the particular method and device by which they employed it." (101 U.S. at 336.) Claim 2, as drafted, claims far more than "the particular method and device."

The reissue's claim 4 (for context)

Claim 4 = rotary valve + two cylinders + double-acting pistons with cam-slot + adjusting stops + crank-shaft. The Court itself identified its provenance: elements 1, 2, 3 and 5 came from the Ball & Fitts July 1869 patent, and "the only additional elements in the present patent are the adjusting stops and the rectangular position of the slots in the piston-rods." A five-element combination whose first four and last are drawn from two references by the same inventors — one of which is admitted in the spec — is a textbook § 103 combination.

---

6. Articulated motivations to combine (KSR rationales)

1. Express admission by the patentees. The specification bridges Ericsson's field and the '69 Ball & Fitts patent in a single sentence. There is no need to infer a teaching, suggestion, or motivation; it is on the face of the patent.
2. Same field / same problem. All references address accurate volumetric measurement by arresting the piston and shifting the valve during repose. KSR at 402.
3. Known technique applied to a known device. Contouring a slot to vary the driven/free phases of a follower is old in valve gear (US 87,448A, US 83,903A, US 70,841A — pre-1870). Applying it to Ericsson's crank-slot is a predictable application, not an unpredictable one.
4. Finite, identified, predictable solutions. KSR at 417 — enlarging a slot's middle versus tapering its ends is a two-way design choice.
5. Optimization of a result-effective variable. The slot's profile relative to crank-pin diameter controls the water quantity per stroke and the dead-center transition — exactly the variables the reissue claim 4 itself frames as adjustable.
6. "Obvious to try." The inventors' own genesis story is empirical: shape the slot until the piston "reaches the end of the stroke and stops a little before the crank passes the center." That is iterative tuning within a known, bounded design space. KSR at 421.
7. Design pressure. Contemporary sources in my search record chronic meter inaccuracy complaints — the Cochituate Water Board's 1853 report ("it did not measure accurately") and the 1854 New York Croton report — establishing a market incentive to refine stroke termination and valve timing.

---

7. Secondary considerations (and why they likely fail)

• Long-felt need for accurate water meters is documented (1853–54 Boston/New York reports; the Union Water-Meter Co. price list showing commercial sale). But any long-felt need was to water meters generally, and the Ball & Fitts '69 patent already addressed it.
• Commercial success: the Union Water-Meter Company's price list (manufactured in Worcester) evidences sales. Nexus is the problem. The success is attributable to the meter as a whole — an assembly the patentee concedes was old — not to the slot's curvature. Without evidence that the concave slot drove purchasing decisions, there is no presumption of nexus. In re GPAC, 57 F.3d 1573 (Fed. Cir. 1995).
• Copying/industry praise: none found. Notably, the Desper defendant designed around the claims entirely by deleting the crank-shaft — the opposite of copying.

---

8. Conclusion

Caveat on the bottom line: Ericsson's slot appears parallel-sided but oversized, whereas the 109,372 slot is tapered/waisted. Ericsson therefore anticipates only if its disclosure is read to encompass a non-uniform slot; absent that reading, the case is § 103 (obvious design choice), not § 102. And because both claims end with "substantially as set forth," a narrow construction tying them to the illustrated curvature would make the motivation argument (rather than anticipation) the decisive one — which is still typically sufficient under KSR.

Finally, note the posture: no court has ever adjudicated the validity of US 109,372. Water-Meter Co. v. Desper was decided on non-infringement only — the Court never reached the defendants' pleaded anticipations (the July 1869 Ball & Fitts patent, "various older patents," and prior public use). The validity question is therefore genuinely open on the historical record, which is precisely why the § 103 analysis above is grounded in the patentees' own admissions rather than in any holding.

---

9. New uncertainties introduced by this section

1. US 7,869's exact issue date is not confirmed in the sources I retrieved; the page at https://patents.google.com/patent/US7869 is attributed to "J. ERICSSON" and the 1851 date comes from Justice Bradley's characterization in 101 U.S. at 334. The patent number range is consistent with 1851, but I flag this as inference.
2. The July 20, 1869 Ball & Fitts patent number is unknown to me. It is the single most probative reference for motivation, and I could not retrieve it.
3. US 13,320A's content is unverified; my search returned US 13,198 (Curtis & Hoard, 1855) instead. I did not conflate them.
4. The pre-1870 valve-gear references (US 87,448A, 83,903A, 70,841A, 69,792A, 68,748A, 64,166A) are cited by title/date only. I have not read them and do not assert what slot geometries they show.
5. Consistent with the earlier sections, the application filing date remains unknown, which prevents a clean pre-AIA § 102(b) one-year-bar calculation for the July 1869 reference. It is § 102(a)-type art (before the invention) on any reading, and § 102(b) art if the application was filed after July 1870.