US 110205

I'll verify this against live sources before summarizing.

Verified summary — US Patent 110,205 A

Identification (literal): The identifier 110205 maps, in USPTO/Google Patents records, to US 110,205 A, granted December 20, 1870. It is not to be read as a modern seven- or eight-digit patent number.

Abstract: None. The patent predates the statutory abstract requirement (35 U.S.C. § 112, added 1952). The operative summary is the single-sentence statement of invention: it "relates to certain improvements in the siphon pumps patented to me June 8, 1869, and consists in modifications of form and construction, whereby such pump is adapted to new uses, or to accomplish old results in a better manner."

Technical context: The patent is a continuation-in-substance of Coll's earlier No. 90,930 (June 8, 1869) — a steam siphon pump / water elevator having inner and outer pear-shaped hollow heads, a steam-injection pipe, and concentric inner and outer discharge pipes. US 110,205 modifies that structure by cutting off the blunt base of the inner pear-shaped head and cast-connecting it to the outer head with radial arms.

Independent claims — plain language

There are two claims, both in independent (non-dependent) form, as was typical of the "improvement" claim style of the era. Neither references the other by number.

Claim 1 — the "open-base inner head" claim (described by the 1882 court as "the principal and only material one"):
An inner head d that is open at its rear end, leaving an unobstructed annular opening around the steam-injection pipe so that solid matter entrained in the water (e.g., tan-bark) can pass through without clogging. That inner head is connected to the outer head a by radial arms a', substantially as described.

• Purpose stated in the specification: pumping or ejecting bilge-water and the water out of tan-vats, where bark and other solids would otherwise block the water passages.

Claim 2 — the nozzle-piece claim:
A nozzle-piece m that abuts against, and is used in combination with, the inner head d, substantially as described.

• Purpose stated in the specification: the interchangeable nozzle pieces let one pump serve different duty. Nozzle m (Fig. 1) abuts the forward end d' of the inner head to give a smooth internal joint and least flow resistance, and its throat is somewhat contracted before opening into a larger discharge pipe — intended for raising water to considerable height with comparatively little steam. A second nozzle n (Fig. 3) has a smaller throat and does not abut the inner pipe, leaving an annular gap so water also flows from the space between the two heads — intended for raising a larger body of water to a lesser height with the same amount of steam. Nozzle n "may be set loosely in the end of the main discharge-pipe, as shown, or otherwise attached at pleasure."

Caveat on the claim text: The OCR of the granted claim 2 is truncated in the record — it reads "…substantially as and" — evidently cutting off a closing phrase such as "substantially as and for the purposes set forth." I do not have an authoritative clean text of the full claim 2 tail.

Ancillary legal history (relevant, and worth flagging)

• Reissue RE 9,199 was issued to Coll May 18, 1880 on an application filed March 13, 1880 — i.e., roughly nine years after the original. The reissue had one claim to "A water ejector, provided with an inner head or shell…"
• In J. B. Sheriff & Son v. A. Fulton's Son & Co., 12 F. 136 (C.C.W.D. Pa. 1882) (Acheson, D.J.), the reissue was held invalid for unreasonable delay on the ground that the reissue enlarged the original claim after a nine-year delay. The court's decision turned on comparing the reissue against the original No. 110,205 specification and claim 1, and is a useful secondary source for the original's scope.
  • https://www.courtlistener.com/opinion/[8341150](/patent/8341150)/j-b-sheriff-son-v-a-fultons-son-co/
  • https://law.resource.org/pub/us/case/reporter/F/0012/0012.f.0136.html

CAFC 2026 docket check — important negative finding

I searched 2026 Federal Circuit dockets for the identifier 110205. No 2026 CAFC or PTAB proceeding involves US 110,205. The only 2026 "205" hit is US Patent No. 9,978,205 (Location Based Restrictions on Networked Gaming, AG 18, LLC), decided in DK Crown Holdings Inc. v. AG 18, LLC (Fed. Cir., May 6, 2026), involving claim 18 of that gaming patent:

• https://ipwatchdog.com/2026/05/06/cafc-affirms-ptab-ruling-draftkings-failed-prove-unpatentability-gaming-patent-claim/
• https://ipverse.greyb.com/litigation-news/cafc-affirms-ptab-ruling-that-draftkings-failed-to-prove-unpatentability-of-gaming-patent-claim

Per your strict rule I am reporting this result as found rather than discarding it, but I am not conflating it with your target: 9,978,205 is a different patent from 110,205. It is a 2017-era gaming patent, not the 1870 ejector. Taxpayer-funded CAFC appeals did not exist in 1870 (the Federal Circuit was created in 1982), and US 110,205 expired in 1887, so no live 2026 appellate docket for it is possible.

Uncertainty disclosures

1. Filing date of the original is not in the record — only the 1870-12-20 grant date.
2. No abstract exists — anything presented as one would be fabricated.
3. No assignee of record — ownership was Coll's personally.
4. Claim 2's text is OCR-truncated in the available sources.
5. Classifications shown (F04F 5/00, 5/44, 5/46 — jet pumps and nozzle arrangements) are retroactively applied by examiners/Google, not original 1870 classifications.
6. I did not verify whether related reissue records such as RE 8,904 E ("Improvement in ejectors," 1879-09-23) or RE 10,169 E ("Ejector," 1882-07-25) also trace to Coll; they appear in Google's "Similar Documents" list, which is a similarity listing and not a family relationship.

I'll search for litigation records involving US Patent 110205.

Let me run additional targeted searches to be thorough.

Litigation Search: US Patent No. 110,205

Patent identified (interpreted literally as US 110,205): "Improvement in Ejectors" / "Improvements in Siphon Pumps" — granted to Hugh Coll of Millvale Borough, Pennsylvania, December 20, 1870. Per the Google Patents record, it was "Anticipated expiration 1887-12-20," status "Expired – Lifetime."

Result: One (1) litigation identified — 19th century

I found one litigated matter that directly involves US Patent No. 110,205. It is an 1882 equity suit, and it was brought on the reissue of that patent (Reissue No. 9,199), with 110,205 named as the original patent.

Case 1: J. B. Sheriff & Son v. A. Fulton's Son & Co.

Basis for the holding: The court walked through the lineage of Coll's patents, expressly reciting that "the original letters patent, No. 110,205, were issued to Coll December 20, 1870," and reproduced verbatim claim 1 of 110,205 ("The inner head, d, open at its rear end so as to leave an unobstructed opening around the steam-injection pipe for the passage of such pieces of solid matter as may be brought up by the water, such inner head, d, being connected to the outer head, a, by means of radial arms, a, all arranged substantially as described."). The court then found that the reissue specification added advantages "neither mentioned nor hinted at in the original specification (No. 110,205)," and that the nine-year delay in seeking reissue with an enlarged claim was fatal.

Sources: 12 F. 136 (law.resource.org PDF) · 12 F. 136 (HTML) · CourtListener, J. B. Sheriff & Son v. A. Fulton's Son & Co., 12 F. 136 (1882)

No modern / contemporary litigation found

I searched patent-litigation-focused resources (Unified Patents litigation caselist, CourtListener/RECAP, general docket databases) and found no modern cases involving US 110,205. That is expected: the patent's term expired December 20, 1887, so it has been unenforceable for roughly 139 years.

Negative-result caveat (avoiding a "similar number" false positive)

Every search hit other than the 1882 case was a different patent that merely shares the trailing digits "205." I explicitly flag these as not US 110,205:

• U.S. Patent 9,978,205 — DK Crown Holdings Inc. v. AG 18, LLC, Fed. Cir. No. 2024-2078 (IPR2022-01446) — networked gaming
• U.S. Patent 10,593,205 — Navog LLC v. TomTom Intl BV, No. 2:24-cv-00365 (E.D. Tex.)
• U.S. Patent 6,910,205 — Oracle America, Inc. v. Google Inc., No. 3:10-cv-03561 (N.D. Cal.)
• U.S. Patent 7,024,205 — Google, Inc. v. Unwired Planet, CBM2014-00005
• U.S. Patent 9,373,205 — Liberty Access Tech Licensing LLC v. Dormakaba Holding AG, No. 2:26-cv-00455 (E.D. Tex.)
• U.S. Patent 12,109,205 — Amicus/Aurobindo matter (D. Del.), per DrugPatentWatch

Confidence and gaps

• High confidence: US 110,205 issued Dec. 20, 1870 to Hugh Coll; it expired in 1887; and J. B. Sheriff & Son v. A. Fulton's Son & Co., 12 F. 136 (C.C.W.D. Pa. 1882), is a real, reported case in which 110,205 is the original patent underpinning the asserted reissue.
• Not verified: I did not retrieve the full opinion text, so I cannot confirm the precise final decree (e.g., bill dismissed vs. decree for defendants), the exact filing date, or whether any additional 1870s–1880s suits naming 110,205 or its reissue exist in Federal Cases or unreported circuit files. Those pre-1900 records are not comprehensively indexed in Unified Patents, PACER, or RECAP, which is why only the reported Federal Reporter case surfaced.
• I also cannot rule out unrecorded/licensing disputes that never reached a reported opinion.

Proceedings overview

Total AIA trial proceedings on US 110205: zero (0). Breakdown by status: active — 0; claims invalidated — 0; claims sustained — 0; settled — 0; institution denied — 0. The structured PTAB block drawn from the USPTO Open Data Portal returns no AIA trial proceedings for this patent as of the most recent ingest, and my web searches (PTAB decisions, Federal Circuit docket, CourtListener, general web) surfaced no IPR, PGR, or CBM naming US 110205A, US 110205D, or "Hugh Coll." Two independent searches for PTAB/CBM activity on this patent returned empty result sets.

The defensive posture is therefore neither "hardened by surviving IPRs" nor "claims canceled." It is moot: this is an 1870 patent that expired on 1887-12-20 (17 years from the 1870-12-20 grant, per Google Patents' legal-status entry "Expired - Lifetime / Anticipated expiration 1887-12-20"). The PTAB did not exist until the America Invents Act created it in 2012 — 125 years after this patent's term ran out. There is no AIA trial record to analyze, no FWD to quote, no estoppel to map, and no live claim scope for a demand letter to reach.

Because there is no proceeding to write up, the per-proceeding template below is populated with the closest historical analogue on the docket, clearly flagged as not a PTAB proceeding, plus the reasons an AIA trial is structurally unavailable. Do not read the subsection heading as an AIA trial number — there is none.

No AIA proceeding exists — structural reasons

• Type: N/A (no petition ever filed; none could have been filed during the patent's life).
• Filed: N/A.
• Status: N/A. The patent itself is "Expired - Lifetime" (verbatim USPTO/Google Patents legal status), with anticipated expiration 1887-12-20.
• Judge panel: N/A.
• Petition grounds: N/A.
• Institution decision: N/A.
• Final Written Decision: N/A.
• Settlement / termination: N/A.
• Appeal: N/A.
• Defensive value: The patent cannot be asserted. Its term expired 1887-12-20, so (a) there is no infringement cause of action to defend, (b) any IPR would be a purely academic exercise, and (c) § 315(e)(2) estoppel is irrelevant because no petitioner ever appeared. If a demand letter cites US 110205, the correct response is not an IPR — it is a laches/expiry letter and, if the sender persists, a § 285 fee posture.

Historical note — NOT a PTAB proceeding: J. B. Sheriff & Son v. A. Fulton's Son & Co., 12 F. 136 (C.C.W.D. Pa. 1882)

• Type: 19th-century equity infringement suit in the U.S. Circuit Court for the Western District of Pennsylvania (Judge Acheson). Included only because it is the sole adversarial record touching this patent family; it is not an AIA trial and carries no estoppel effect.
• Filed: Suit date not stated in the reported opinion; opinion reported 1882.
• Status: Decided (1882).
• Judge panel: Acheson, D.J. (single-judge circuit court — the PTAB did not exist).
• Asserted claims: The suit was brought on reissued letters patent No. 9,199, issued to Hugh Coll 1880-05-18 on an application filed 1880-03-13 — a reissue of the original No. 110,205 (issued 1870-12-20). The reissue carried a single claim: "A water ejector, provided with an inner head or shell, extending backward over the water-induction opening, and arranged with relation to the jet-pipe so as to protect the steam-jet from the direct impact of the inflowing current…" — i.e., the reissue reached subject matter the 1882 court found was not in the 1870 disclosure.
• Grounds / holding: The court held that the reissue claim was enlarged beyond the original, that the added advantages (the inner head acting as a "breakwater or dam" to reduce steam condensation) were "neither mentioned nor hinted at in the original specification (No. 110,205)," and — per the headnote — that "a delay of nine years is unreasonable on an application for a reissue wherein the original claim is enlarged." The reissue was therefore unenforceable against the defendants.
• Settlement / termination: N/A — judgment, not settlement.
• Appeal: No Federal Circuit appeal is possible (that court was created in 1982). No reported Supreme Court disposition surfaced in my search.
• Defensive value: Historical only. It confirms that Claim 1 of 110205 — the "inner head d, open at its rear end… connected to the outer head a, by means of radial arms" claim — was the "principal and only material" claim of the patent, and that the patentee's own attempt to broaden it via reissue failed on laches/enlargement grounds. Do not cite this to a court as IPR estoppel — it is a pre-AIA circuit-court decision about a different (reissue) claim set and has no § 315(e) effect.
• Source: CourtListener, J. B. Sheriff & Son v. A. Fulton's Son & Co., 12 F. 136 (1882); Federal Reporter Vol. 12, p. 136 (law.resource.org)

Strategic summary

Claim status: every claim expired, none canceled, none sustained, none tested. US 110205 has two claims as issued: (1) the inner head d, open at its rear end around the steam-injection pipe and connected to the outer head a by radial arms a′; and (2) the nozzle-piece m abutting against and in combination with the inner head d. Both expired with the patent term on 1887-12-20. There is no "surviving claim" set to work with, because there is no surviving patent. A parallel reissue, No. 9,199 (1880-05-18), was held invalid in 1882 for unreasonable delay in seeking an enlarged claim — but that is the reissue's problem, and it too is long expired.

Estoppel landscape: empty. IPR estoppel under 35 U.S.C. § 315(e)(2) attaches only to a petitioner (or its real party in interest or privy) that has instituted an IPR and reached a final written decision. No petition was filed, so there is no petitioner, no instituted ground, no "reasonably could have raised" set, and no estoppel bar on anyone. Nothing about 110205 constrains any defendant's prior-art arguments in any forum — trivially true, since the patent cannot be asserted in the first place.

Pattern signals: none. No repeat petitioner, no serial filings, no PO appellate history before the Federal Circuit (which did not exist during this patent's life), no defensive aggregator involvement — Unified Patents was founded in 2012, well over a century after expiry. The absence here is not the usual "this patent is uninteresting to challengers" signal; it is a categorical signal: AIA trials (IPR/PGR/CBM) were established by the Leahy-Smith America Invents Act, effective 2012-09-16, and this patent's enforceable life ended 1887-12-20.

Caveats on scope of my search. My searches for PTAB activity on this patent returned no results, and the structured ODP block is canonical and reports none. I did not find, and cannot rule out, an obscure or mis-indexed filing, but I have no evidence of one and no basis to assert one. I also did not independently verify the 1882 opinion's full disposition beyond the reported headnote and opinion text quoted above — treat the laches/invalidity characterization as the reported holding of that decision, not my independent legal conclusion.

Recommended next steps

• If you are a defendant receiving a demand citing US 110205: do not prepare an IPR. There is no AIA proceeding to leverage, and none is needed. Respond that US 110205A issued 1870-12-20 and its term expired 1887-12-20, so no claim of it can be infringed today; the same is true of reissue 9,199 (1880-05-18). If the sender cites a live patent number, ask them to identify it precisely — the similarity in this family (90,930 / 110,205 / RE 9,199) invites number confusion.
• Escalate if the demand persists: a demand on a patent that expired 139 years ago supports a Rule 11 / § 285 fee-shifting posture against the sender, and potentially state-law claims. Preserve the letter and any invoicing.
• If you were told there is an active PTAB trial: that is incorrect. There is no institution decision, no oral hearing, and no statutory one-year FWD deadline running, because no petition exists. Any docket reference you were given should be treated as fabricated or misattributed until a matching PTAB E2E or ODP record is produced.
• Verify independently in one step: search PTAB E2E / the ODP "Proceedings" tab for application 110205 and for patent 110205; expect an empty result set. Confirm the expired status in USPTO Patent Center (status "Expired - Lifetime," anticipated expiration 1887-12-20).
• Bottom line: no PTAB activity exists on US 110205, no claims have been canceled or sustained because none are alive, and the only historical adversarial ruling in the family — the 1882 Sheriff decision — concerns the reissue, not the original patent, and has no estoppel effect on anyone.

I'll check the assignment record sources before writing, since this chain's outcome depends entirely on what the Assignment Center actually holds.

The key procedural fact is already surfacing: USPTO's Assignment Search covers August 1980 to present, with pre-1980 assignments held at NARA. Let me confirm what that means for an 1870 patent and check for 19th-century ownership evidence.

Assignment & Ownership Analysis — US Patent 110,205 A ("Improvement in ejectors," Hugh Coll, issued 1870-12-20)

Threshold finding (governs everything below)

The USPTO Assignment Center contains no records for US 110,205, and structurally cannot contain any. Two independent reasons:

1. Database coverage. USPTO's own guidance states the Assignment Search database contains "all recorded Patent Assignment information from August 1980 to the present," and that "Patent Assignments recorded prior to 1980 are maintained at the National Archives and Records Administration." The codified version is 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)."
   • https://web.archive.org/web/20210623092848/https://www.uspto.gov/patents/maintain/patents-assignments-change-search-ownership
   • https://www.govinfo.gov/content/pkg/CFR-2024-title37-vol1/html/CFR-2024-title37-vol1.htm (37 C.F.R. § 1.12)
2. No possible post-1980 event. US 110,205 issued 1870-12-20 with an anticipated expiration of 1887-12-20 (17 years from grant, the then-governing term). There is no transaction of any kind that could be recorded against it in a 1980–present register.

I therefore cannot produce reel/frame numbers, correspondents, or execution/recording dates for this patent. Producing any would be fabrication. Per your instructions, I state that plainly — and I note explicitly that I am nonetheless completing the downstream sections because your output spec requires a verdict, and "Insufficient data" is a specified, meaningful verdict rather than an omission.

Verification links (both will return zero rows for patent number 110205):

• https://assignmentcenter.uspto.gov/
• https://assignment.uspto.gov/patent/index.html

Where the real record, if any, would live: NARA's pre-1957 patent assignment records, searched by patent number, grant date, or assignor name (Hugh Coll), with certified copies issued by NARA under § 1.12(a)(2). I did not query NARA in this session and have no NARA results to report.

Inventors

• Sole inventor; no co-inventors appear on the face of the patent. Correspondence address, if any, is the standard "Witnesses: JOHN GLENN, G. H. CHRISTY" block — these are attesting witnesses to the executed specification, not assignees, employers, or correspondents of record.
• Employer: The specification is silent. Coll self-describes the invention as an improvement on "the siphon pumps patented to me June 8, 1869" (No. 90,930), which indicates independent inventorship / personal ownership, not an employee-invention assignment. Consistent with the previous section's finding of no assignee on the face of the patent.
• Unusual-pattern check — inventor-departure signal: Not applicable / not assessable. The "all inventors departing the original assignee within 12 months" heuristic presupposes an assignee-employer. There is none of record, and no 1870 employment records were located. Millvale Borough (now Millvale, Allegheny County) was an industrial river town; I found no evidence Coll worked for a named pump manufacturer, and I will not speculate.

Note on the "Similar Documents" list: Google Patents lists US 104,986 ("Improvement in ejectors," 1870-07-05) and US 110,289 ("Improvement in injectors," 1870-12-20) adjacent to this patent. These are similarity-list results, not family members and not evidence of a common assignee. I could not verify their inventors in this session and make no claim about them.

Original assignee

None. The patent issued to Hugh Coll personally; no assignee is named on the face of the grant, and my prior section reached the same conclusion independently. The record is therefore an inventor-owned patent, not an operating-company-owned one.

• Entity named on the issued patent: N/A — individual ownership.
• Did they ship a product embodying the claims? Unclear for the original patent. The strongest available circumstantial evidence is the 1882 infringement action J. B. Sheriff & Son v. A. Fulton's Son & Co., 12 F. 136 (C.C.W.D. Pa. 1882), which was litigated on reissue RE 9,199, not on US 110,205, and which presupposes competing commercial ejectors in the Western District of Pennsylvania market. That indicates the invention had commercial embodiments by 1882. It does not establish that Coll himself manufactured or sold them.
• Primary line of business: Underdetermined for Coll. The patent's stated uses — bilge-water ejection and "the water out of tan-vats" — point to the Allegheny County tanning and river/marine trades as the intended customer base, but that is a reading of the specification's stated purpose, not a finding about Coll's business.
• Current status: N/A (natural person, deceased long ago; no successor entity of record). The patent itself is "Expired – Lifetime," anticipated expiration 1887-12-20.

Flagged inference (not a finding): The 1882 case caption lists J. B. Sheriff & Son as complainant on the reissue. A complainant in a patent infringement action must ordinarily hold title or an exclusive license, which suggests Coll's rights had passed to Sheriff & Son by 1880–1882. I retrieved only the opinion's excerpted text, which does not recite an assignment, and the previous section already cautioned that RE 9,199 and RE 10,169 appear only in Google's similarity list. I therefore record this as an unverified inference, not a recorded transfer.

Assignment timeline

No recorded assignments exist in the USPTO Assignment Center for US 110,205. The chain of title is blank from the USPTO's 1980-onward register, and any 1870–1887 conveyances would be NARA-held pre-1957 records that I did not retrieve.

Per the format above: there are no reel/frame entries to report, and I am not going to invent placeholder reel numbers.

Timeline diagram

Rendered from the only dated ownership-relevant facts actually on the record. It is a docket/status timeline, not an assignment chain — there are no assignment events to plot.

timeline
    title Ownership status of US 110205
    1869 : Coll granted earlier siphon pump US 90930
    1870 : US 110205 issued to Hugh Coll
         : No assignee named on the grant
    1880 : Reissue RE 9199 applied for by Coll
         : Reissue granted to Coll
    1882 : Infringement suit on the reissue
         : Sheriff and Son named complainant
    1887 : Patent term expires
    1980 : USPTO assignment register begins
         : Register contains no record for this patent

NPE / troll-pattern signals

Every signal is scored against the specific evidentiary standard you set. Because there are zero recorded transfers, seven of the eight are structurally impossible to substantiate. Marking them "not present" would be a false negative — the honest label for a signal that cannot be populated is unclear / not assessable.

Cross-check on the previous sections: my prior section's findings — no assignee of record, no abstract, filing date absent, claim 2 OCR-truncated, reissue RE 9,199 invalidated in 1882 for unreasonable delay — are all consistent with the above. No contradiction surfaced. The 1882 invalidity holding is on the reissue, which is a separate legal instrument; it does not disturb the original US 110,205, which by then had expired or was within two years of expiring.

Verdict

Insufficient data

Justification. There is no assignment chain to evaluate: the USPTO Assignment Center's register opens in August 1980 (37 C.F.R. § 1.12(a)(2); USPTO: "Patent Assignments recorded prior to 1980 are maintained at the National Archives and Records Administration"), while US 110,205 issued 1870-12-20 and its anticipated expiration is recorded as 1887-12-20. No reel/frame, assignor, assignee, or correspondent of record exists to cite, and every NPE signal in the framework is either not present or chronologically impossible for a patent that expired 43 years before the modern assignment register and 138 years before this analysis. The one ownership hint in the file — J. B. Sheriff & Son as complainant in the 1882 reissue suit — is an inference from a case caption on a different instrument, not a recorded transfer, and I decline to elevate it to a finding. Resolving this properly requires a NARA pre-1957 assignment search by patent number and by assignor name "Hugh Coll," which was outside this session's reach.

Verification: https://assignmentcenter.uspto.gov/ (search patent number 110205 — expect zero results), https://assignment.uspto.gov/patent/index.html, and for the historical record, NARA pre-1957 patent assignment holdings.

Note on scope discipline: this patent is a useful control case for the framework. US 110,205 has no NPE exposure not because it lacked an assertion history, but because it was asserted — if at all — in an 1882 Pennsylvania circuit court on a separate reissue, in an economy with no LLC shells, no correspondent-of-record conventions, and no possibility of a 2026 appellate docket. A "no signals present" result here should not be read as a clean bill of health for the family; RE 9,199 is the instrument where the assertion behavior actually occurred, and it was struck down for nine-year reissue delay.

I'll verify the citation record for this patent against live sources before answering.

Let me check whether the USPTO/Google record for this patent contains any cited references, and verify the earlier Coll patent.

The record shows no citation table so far. Let me verify the specific pre-1870 documents that appear in Google's similarity list, since those are the only realistic §102 candidates.

I hit the tool-call limit mid-verification, so I'm reporting below exactly what I confirmed and where the gaps are. Two of my searches returned results before the cutoff; two others did not complete.

Prior art analysis — US Patent 110,205 A

Bottom line

There is no citation record for this patent. The Google Patents/UPSTO record for US 110,205 A contains no "Patent Citations" table and no "Cited By" table — the only reference-like content is a "Similar Documents" list and a set of Google-extracted "Prior art keywords" (water, head, improvement, nozzle, ejectors). Under your strict rule, I am not reclassifying algorithmically-similar documents as citations. They are not the same thing, and treating them as citations for §102 purposes would be fabrication.

The only patent the specification itself references is Coll's own earlier US 90,930 (June 8, 1869) — and it is a self-reference to the parent invention, not third-party art. It is nevertheless the single most relevant reference for §102/§103-style analysis, and I analyze it below.

1. Why the citation record is empty (and why that is not a search failure)

Two structural reasons, both grounded in the record itself:

(a) Front-page "References Cited" is a 20th-century feature. The granted 1870 printing (reproduced in the Google Patents scan) has no front-page citation block. What the record shows instead is a "Prior art keywords" block (water, head, improvement, nozzle, ejectors) — these are Google's machine-extracted topical terms from the claim and specification text, not references. The word "ejectors" there is simply the title of record.

(b) The blank "Application number" field is a second symptom. As the earlier section of this analysis flagged, the application number field is empty and no filing date appears. A prosecution file with no recoverable application number has no recoverable examiner's citation record either.

Evidence that these old patents are cited as art in the other direction, but not here: US 57,035 A (1866) appears in the "Patent Citations (10)" table of US 4,274,812 A (Elliott Co., filed 1921-12-07). That confirms the document is citation-eligible in modern databases, and simultaneously confirms that no comparable citation table exists on US 110,205 A itself.

Net finding: zero prior-art citations of record on US 110,205 A. Any element-by-element §102 mapping I produce is therefore analyst-constructed, not record-constructed — and I label it as such.

2. Reference of record in the specification: US 90,930 (Coll, June 8, 1869)

Disclosure, as recited by the 1882 court: outer and inner pear-shaped hollow heads; each head tapered at one end into a discharge pipe (inner pipe much smaller); steam-injection pipe screwed/ cast onto the rear end of the outer head and extending into the inner head; the inner head "at its rear end, is screwed on or otherwise fastened to or cast on the injection pipe"; "back of the discharging end of the steam-injection pipe inlets or openings are made in the inner head"; the claim is "exclusively for the combination of the inner and outer pear-shaped heads with the injection pipe."

§102 assessment:

• Claim 1 — NOT anticipated, but narrowly avoided. Claim 1 requires the inner head d to be open at its rear end, "so as to leave an unobstructed opening around the steam-injection pipe for the passage of such pieces of solid matter," and to be connected to the outer head a by radial arms a'. US 90,930 discloses the opposite geometry at the rear: the inner head is fastened to the injection pipe, with discrete inlets in the head wall behind the steam nozzle. A closed, pipe-mounted rear end plus side inlets is not an unobstructed rear opening, and 90,930 has no radial-arm connection to the outer head. So 90,930 does not anticipate claim 1. It is the anticipatory-art benchmark that claim 1's two positive limitations (open rear end; radial arms) were written to clear.
• Claim 2 — Weak anticipation candidate, on the "inner head" sub-element only. Claim 2 covers "the nozzle-piece m, abutting against and in combination with the inner head d." US 90,930 discloses the inner head/tube into which a discharge pipe leads; if 90,930's inner discharge pipe were shown to seat against the inner head with a smooth internal joint, that sub-combination is arguably old. I cannot confirm this — I have the court's summary, not the 90,930 drawing or text. Treat as a hypothesis, not a finding.
• Note this is also the self-collision problem: under the 1836 Act's "patented or described" branch, a patent to the same inventor is a fragile basis for §102; claim 1's validity in 1870 rested on the structural departure, not on novelty over 90,930 as third-party art.

Also relevant (not prior art): Reissue RE 9,199 (issued May 18, 1880; application filed March 13, 1880) is the same disclosure as 110,205 plus added advantages. A reissue of the same patent cannot be separate prior art against it, and the 1882 court held that reissue invalid for unreasonable delay because it enlarged the original claim after nine years, adding advantages "neither mentioned nor hinted at in the original specification (No. 110,205)."

3. Candidate art by grant date (analyst-identified, from Google's similarity list)

I sorted Google's "Similar Documents" list — which is a similarity listing, not a citation list — by publication date relative to US 110,205's December 20, 1870 grant, because only pre-grant publications can function as §102 art.

3a. Published BEFORE December 20, 1870 — the only possible §102 candidates

§102 mapping — I am deliberately declining to state anticipation. §102 requires element-by-element disclosure of the claim. I have verified only the existence, title, date, and (for one) inventor of these documents. I have read none of their specifications or drawings. Asserting that US 57,035 A or US 64,539 A "anticipates claim 1" would require me to have confirmed (i) an inner head open at its rear around the steam-injection pipe, and (ii) radial-arm connection to an outer head — the exact two elements that distinguish claim 1 over US 90,930. I have confirmed neither. My honest position: US 104,986 A and US 64,539 A are the highest-value targets for follow-up, on title and date alone.

3b. Published AFTER December 20, 1870 — expressly NOT prior art

These appear in Google's similarity list but cannot anticipate US 110,205, because they post-date the grant: US 121,376 A (1871), US 119,852 A (1871), US 116,104 A (1871), US 129,090 A (1872), RE 8,904 E (1879), US 246,084 A (1881), US 265,246 A (1882), RE 10,169 E (1882), US 401,753 A (1889), US 531,111 A (1894), US 611,454 A (1898), US 646,640 A (1900), US 749,875 A (1904). I flag these explicitly so they are not mistaken for the citation record.

Two of these are Coll's own lineage and worth naming: RE 8,904 E ("Improvement in ejectors," 1879-09-23) and RE 10,169 E ("Ejector," 1882-07-25). As the earlier section noted, I have not verified that either traces to Coll — Google's similarity list is not a family relationship, and I will not assert one.

4. Two contradictions / data hazards you must know about

(a) Identifier collision on the bare string US110205. A patent-aggregator page (patenthub.cn) lists "US110205" as the application number of US 4,857,379 A — Sheetlike structure of fibers, especially as a reinforcement for plastics components, filed 1987-10-19, granted 1989-08-15, inventors Margret Plöntges and Herbert Cox. That is an application-serial collision, not the same document: US 4,857,379 A is a 1980s textile patent. Under your literal-identifier rule I am reporting it rather than discarding it, but it is not the 1870 ejector patent and it contributes nothing to the prior art of US 110,205 A. This is a live false-positive risk for anyone searching the bare string 110205.

(b) The task's legal frame is anachronistic. US 110,205 A was examined under the Patent Act of 1836 as amended (and the Act of July 8, 1870), not the modern 35 U.S.C. § 102 (1952/2011 versions). In 1870 there were no IPRs, no "prior art date" concept in the modern sense, no printed References Cited, and no statutory abstract. Google's "Prior art date: 1870-12-20" is simply the publication date, not a determined critical date. I am answering in the modern § 102 vocabulary you requested, but the mapping is an analogy. I also cannot state the applicable grace period with confidence and will not guess at it.

5. Confidence and gaps

• High confidence: US 110,205 A carries no patent citations of record; the only patent referenced in its specification is Coll's own No. 90,930; the December 20, 1870 grant and December 20, 1887 (17-year) expiration are as recorded; US 90,930 does not disclose claim 1's open rear end plus radial arms.
• Moderate confidence: US 57,035 A (1866-08-07, Charles Barnes, "Improvement in steam-jet pumps") is a genuine pre-grant steam-jet-pump patent and is citable art (it is cited in US 4,274,812 A).
• Not verified — do not rely on: the specifications or drawings of US 57,035 A, US 64,539 A, US 57,057 A, US 54,615 A, US 104,986 A, or US 103,494 A; the actual content of the 90,930 drawing; whether RE 8,904 E or RE 10,169 E belong to Coll; and whether any additional 1870s–1880s suits name 110,205 beyond the reported 1882 case. My two verification searches (US 64,539 A, and an attempt to surface a citations table on the Google Patents page for US 110,205 A) were cut off before returning results.
• Unverified and deliberately not asserted: any statement that a specific one of these references "anticipates" claim 1 or claim 2.

Recommended next steps for a complete answer: pull the full OCR of US 104,986 A, US 64,539 A, US 57,035 A, US 57,057 A and US 54,615 A from Google Patents/USPTO Patent Public Search, and check their Fig. 1 geometries specifically for (i) a rear-open inner head surrounding a steam-injection pipe and (ii) radial-arm mounting to an outer head.

I'll ground this analysis in the actual "Prior Art" listings on the US 110,205 page, then verify what those references disclose. Let me pull them.

Section: Obviousness Analysis under 35 U.S.C. § 103 — US Patent 110,205 A

Scope note: I was limited in further searching, so this analysis rests on (a) the full text of US 110,205 as supplied, (b) the "Prior Art" material actually present on the Google Patents page — the "Prior art keywords" field and the "Similar Documents" listing — and (c) the reference data I was able to verify before the search budget was exhausted. Where I could not read a reference's interior, I say so and frame the combination conditionally rather than asserting what it discloses.

1. Threshold framework problems (must be flagged before any § 103 reasoning)

1. § 103 postdates the patent by 82 years. US 110,205 issued December 20, 1870. Section 103 was enacted July 19, 1952. In 1870–1887 the governing test was the judge-made "invention" requirement of Hotchkiss v. Greenwood, 52 U.S. (11 How.) 248 (1851), later hardened by the "flash of creative genius" language of Cuno Eng'g Corp. v. Automatic Devices Corp., 314 U.S. 84 (1941). The Supreme Court expressly repudiated Cuno in Graham v. John Deere Co., 383 U.S. 1 (1966), and § 103 codified that repudiation. A modern § 103 analysis is therefore materially more permissive of combination than the standard actually applied to this patent — i.e., applying § 103 today makes invalidation easier, not harder. Any conclusion of obviousness below must be read with that directional bias in mind.

2. The analysis is legally moot. The Google record shows "Anticipated expiration 1887-12-20," status "Expired – Lifetime." The § 103 question here is academic/hypothetical: there is no enforceable term to invalidate.

3. The only court to test this patent's family did not use obviousness. In J. B. Sheriff & Son v. A. Fulton's Son & Co., 12 F. 136 (C.C.W.D. Pa. 1882), the court invalidated Reissue No. 9,199 for unreasonable delay in enlarging the claim. Per the prior sections' record, the court noted the reissue added advantages "neither mentioned nor hinted at in the original specification (No. 110,205)." That finding cuts against nonobviousness: the advantage the patentee later relied on was not in the original disclosure, so it cannot be used to rescue the original claims. But it also means the original claims of 110,205 were never adjudicated invalid for want of invention — that negative fact must be preserved.

4. Claim 2's text is OCR-truncated ("…substantially as and"). The analysis of claim 2 assumes the conventional contemporaneous tail ("…substantially as and for the purposes set forth").

5. Date discrepancy. The task header states "Current Date: April 26, 2026"; the system context states 2026-09-12. I note the inconsistency rather than resolving it; it does not affect the analysis.

2. Screening the "Prior Art" material on the page

The page's "Prior art keywords" — water, head, improvement, nozzle, ejectors — are auto-extracted text tokens, not an examiner's art field. The "Similar Documents" list is keyword/title-similarity driven, not a citation or family relation. Screening it against the Dec. 20, 1870 date is mandatory, and doing so kills most of the list:

Contradiction flagged (live-search vs. listing). The listing presents US 103,494 A among documents similar to 110,205, with the garbage OCR title "Jthographefl." Live retrieval shows 103,494 is Charles A. Peavey's "Feed-Water Connecting-Pipe for Locomotives" (granted May 24, 1870) — a rubber-sleeved flexible hose joint between a locomotive boiler and its tender tank, classified in F16L (pipe joints), claimed as "the flexible joint E… constructed and operating substantially as and for the purposes set forth." It has no ejector, nozzle, head, steam-jet, or solids-passage teaching. It must be excluded from the § 103 combination. I am reporting the contradiction rather than deferring to the bare listing. The same screening logic excludes US 101,619 (horse-collars) as a keyword artifact.

Consequence: the usable pre-1870 art in this record is thin and, on the record available, is largely identified by title only. US 57,035, 54,615, 57,057, 64,539, and 104,986 are all plausibly pertinent by title, but I have not read their interiors. The combinations below are therefore stated as conditional: each requires verification that the named reference actually discloses the element attributed to it.

3. The person having ordinary skill in the art (PHOSITA), c. 1870

A mechanic or engineer in the steam-injector / siphon-pump trade: no formal degree expected; familiarity with Giffard-type injectors (1858), steam-jet water elevators, foundry casting practice (the spec itself says the head is "connected, in casting"), and the elementary hydraulics of throat area vs. jet velocity vs. head. The retroactively applied classifications F04F 5/00 (jet pumps), F04F 5/44 (component parts), and F04F 5/46 (arrangements of nozzles) correctly mark the art as crowded and mature: nozzle arrangement is itself a recognized sub-art (5/46), which bears directly on claim 2.

4. Claim 1 — Graham analysis

Claim 1, as reconstructed from the record: an inner head d, open at its rear end, leaving an unobstructed opening around the steam-injection pipe for the passage of solid matter entrained in the water, the inner head being connected to the outer head a by radial arms, substantially as described.

Why a PHOSITA would have made the change (motivation, all KSR-recognized):

1. The specification supplies its own motivation. It states the purpose outright — pumping bilge-water and tan-vat liquor where "tan-bark is necessarily carried up with the water" and "the holes or passages for the water are apt to become clogged, unless ample room be left for the flow of both." That is a stated problem in a known field, and the solution is "leave ample room."
2. Simple substitution / change of degree, not function. Cutting away a constricting closed base to enlarge a flow area is exactly the "simple substitution of one known element for another to obtain predictable results" rationale. The head still defines a converging flow path toward d'; the invention changes how much of the base is closed, not what the part does.
3. Radial-arm (spider) support of a coaxial member inside a flow duct was notoriously old — the ordinary way to hold a coaxial element in a conduit while leaving an annular path. Nothing in the claim specifies arm number, section, or spacing beyond "in any desired number," which the specification itself treats as a free design choice. That is the KSR "finite number of identified, predictable solutions" situation.
4. Both modifications are consequences of one another. Once the base is removed, a spider is the natural way to keep the head centered; conversely, once one wants a spider to minimally obstruct flow, removing the base follows. The two differences are a single design move, which is why they were made in a single casting.
5. No new result is claimed. The asserted benefit (passing solids) is inherent in the geometry — an open annulus sized to pass bark inherently passes bark. There is no showing of an unexpected property, critical dimension, or result not predictable from the structure.

Combination 1A (strongest, by admission): Coll 90,930 (pear-shaped inner head, closed base, concentric heads, steam injection pipe) + the ordinary mechanical skill evidenced by the cast-spider/radial-arm mount + the express problem statement in the field (clogging in debris-laden pumping) → claim 1.

Combination 1B (independent-art version, avoiding the "own patent as art" objection): US 57,035 (steam-jet pump, 1866) or US 64,539 / US 104,986 (water ejectors, 1867/1870) in view of US 54,615 (siphon/force-pump pressure-heads, 1866) for the concentric-head housing, plus radial-arm mounting as known structure. Conditional on verification that these references disclose a closed-base inner shell and/or axial support arms.

5. Claim 2 — Graham analysis

Claim 2: nozzle-piece m abutting against and in combination with the inner head d.

The specification supplies the whole technical basis itself:

• Nozzle m (Fig. 1): abuts d' "so as to give a smooth joint inside and least resistance to the flow of water"; throat "contracted somewhat, from which it opens into a larger pipe" → raising water to considerable height with comparatively little steam.
• Nozzle n (Fig. 3): smaller throat, does not abut — an annular gap remains, so water also flows from between the heads → larger body of water to less height, same steam.
• The nozzles are expressly "interchangeable, and may be used with the same pump."

That paragraph is, functionally, an admission of the principle of operation and a teaching of the variability of the claimed element. Nonobviousness therefore has to come from something other than the described throat/head relationships.

Why claim 2 is obvious:

1. Variable / interchangeable throat area to trade head against volume is elementary injector practice. A smaller throat raises jet velocity and head; a larger throat moves more water at lower head. Giffard-type injectors and steam-jet pumps of the 1860s were built around this relationship. US 57,035 ("steam-jet pumps," 1866) and US 104,986 ("ejectors," July 1870) are the natural loci of that teaching.
2. Butt-jointing a nozzle to a conduit to present a flush bore and reduce shock/turbulence loss is a routine mechanical expedient. Motivation is stated in the specification itself ("least resistance to the flow"). "Make the joint smooth" is a design incentive, not invention.
3. The only structural element added over claim 1 is the location of the abutment. If a nozzle abutting a conduit in a steam-jet/ejector was known (near-certain), claim 2 is a combination of old elements producing no new or unexpected result — the classic pre-1952 combination rejection of Great Atlantic & Pacific Tea Co. v. Supermarket Equipment Corp., 340 U.S. 147 (1950) (skepticism toward combination patents), later channeled into § 103 and then chilled by KSR. Under present law, KSR subsumes it.
4. Claim 2 is effectively dependent in substance on claim 1's structure. A rejection of claim 1 over the art, combined with "smooth joint + interchangeable nozzle throat" as known design choices, yields claim 2.

Combination 2A: Claim-1 combination + a steam-jet/ejector reference showing an abutted, flush-bore replaceable nozzle (candidate: US 57,035) + the admitted interchangeability and throat relation in Coll's own specification → claim 2.

6. Summary of KSR rationales and where they attach

7. Secondary considerations (Graham prong 4)

• Commercial success: no evidence in the record. Not shown.
• Long-felt but unmet need: arguable — ejector clogging on debris-laden water is a real problem, and the patentee states it. But there is no record evidence that others tried and failed, and the need was addressed by a geometric enlargement that any mechanic could have made; a bare statement of purpose in the patent is not the "long-felt need" evidence Graham contemplates.
• Failure of others / copying: the only infringement signal is the 1882 suit. Copying is a recognized secondary consideration, but (a) I could not retrieve the full decree in Sheriff, so I cannot confirm an infringement finding, and (b) the bill was founded on the reissue, which was held invalid for delay — so the case supplies no adjudicated finding that the original claims were valid and infringed.
• Unexpected results: none asserted in the original specification. Critically, the Sheriff court found the reissue added advantages "neither mentioned nor hinted at in the original specification." Advantages absent from the original disclosure cannot support the original claims' nonobviousness — a party cannot retroactively supply the "unexpected result."
• Licensing / industry acquiescence: not in the record.

Note also the nine-year reissue delay (application filed March 13, 1880 for a patent granted Dec. 20, 1870). Delay of that length is not itself an obviousness factor, but it is inconsistent with a narrative of an immediate, recognized technological breakthrough.

8. Verdict

Claim 1 would very likely be held invalid under modern § 103 — assessed at moderate-to-high confidence (~75–80%) — primarily on Coll's admitted prior patent of June 8, 1869 (No. 90,930) in view of the ordinary mechanical skill of radial-arm/spider mounting, with the clogging problem stated on the face of the patent. Because the change is a truncation of the patentee's own known pear shape, produces a result inherent in the geometry, and involves no dimension or structure beyond a free design choice ("any desired number" of arms), this is close to a textbook case on the "predictable results" and "obvious to try" lines of KSR.

Claim 2 would probably be held invalid — ~70% confidence — as a combination of (i) the claim-1 structure, (ii) an abutted flush-bore nozzle in a steam-jet pump/ejector, and (iii) the patentee's own admitted teaching that nozzle pieces are interchangeable and that throat size trades head against volume. Conditional on verification of the nozzle-abutment teaching in US 57,035 or US 104,986.

Two crucial caveats on those numbers:

• Under the 1870s "invention" standard actually applicable at the time (rather than § 103), the outcome would be substantially closer, and the anti-combination climate of Atlantic Works v. Brady (1883) and A&P (1950) cuts both ways. The percentage figures above are § 103 figures and should not be read as historical ones.
• No court ever held claims 1 or 2 of US 110,205 invalid for obviousness. The only court to pass on this family (Sheriff, 12 F. 136) invalidated the reissue, on delay, not on obviousness — and the original claims, notably, were not the basis of the holding.

9. Verification gaps I am not papering over

1. I did not read the interiors of US 57,035, 54,615, 57,057, 64,539, or 104,986. I verified only US 57,035's existence, inventor (Charles Barnes), and date (Aug. 7, 1866), via the Chicago Public Library patent index; and US 103,494's full text. Every element attributed to those references is by title inference and must be verified before the combinations in §§ 4–5 are relied on.
2. I could not verify US 90,930's specification directly. Its disclosure is taken from the 110,205 specification's own reference to it and from the recitation in Sheriff. That is a strong basis (it is an admission), but it is second-hand.
3. US 110,289 ("Improvement in injectors") bears the identical grant date, 1870-12-20. Whether it is available as art depends on the original's filing date — which, per the prior summary, is not in the record (the Google "Application number" field is blank). Same-day documents also raise the 1836/1839-era question of the applicant's invention date, equally absent from the record. I flag this rather than assume.
4. The "Similar Documents" list is not a prior-art citation list. As shown by US 103,494 (a locomotive feed-pipe) and US 101,619 (horse-collars), it is keyword-similarity noise. Any § 103 analysis built naively on that list — rather than on the pre-Dec.-20-1870 subset and the admitted 1869 patent — would be unsound.
5. The full text of the 1882 opinion and its final decree were not retrieved, so the secondary-considerations inference from that case is provisional.

Sources: https://patents.google.com/patent/[US110205](/patent/US110205)/en · https://patents.google.com/patent/[US103494](/patent/US103494) · https://apps.chpl.org/services/invent/patent.asp?patent=1509 (US 57,035, Charles Barnes, 8/7/1866) · https://law.resource.org/pub/us/case/reporter/F/0012/0012.f.0136.html and https://www.courtlistener.com/opinion/[8341150](/patent/8341150)/j-b-sheriff-son-v-a-fultons-son-co/ (Sheriff, 12 F. 136 (C.C.W.D. Pa. 1882)).