Round eight: New Jersey Mirror index (lane A2), 2026-09-12

Working archive: Burlington County Library System index to the New Jersey Mirror (1818–early March 1947), queried via https://netscape.bcls.lib.nj.us/newspapers/newjerseymirror/njmirror.phtml?keyword=<TERM>&type=any&limit=200&sort=dateasc&fields[]=category&fields[]=headline&fields[]=page&fields[]=summary with a browser user-agent. type=any appears to OR the keyword’s words across category/headline/summary; there is no working date-range parameter, so limit=200 with sort=dateasc truncates common-surname searches before reaching a target year (confirmed with “Clevenger,” 272 total hits, cut off in the 1930s before an 1857 record could surface at all with limit=1000).

Assigned leads: none confirmed in this index

George Lungin, frozen to death Dec. 21, 1831, reported Jan. 4, 1832. Searched “Lungin,” “Lungen,” “frozen to death,” “Martha Furnace,” “Tuckerton,” “exposure,” “abandoned house,” “frozen,” “frozen Martha,” “exposure Tuckerton,” “Lungin Tuckerton.” The index does hold January 1832 issues (a Jan. 5, 1832 marriage notice appears under “Tuckerton”), but no death/exposure item for this date or name turned up under any spelling tried. Not written up — could not source a citation this session.
Thomas Hyland, 65, Atsion carpet weaver, suicide April 19, 1903, Stone Bridge/East Fruitland. Searched “Hyland” (16 hits, all 1892–1938 marriages/deaths unrelated to this case, though one 1892 obituary confirms a Thomas Hyland living at Atsion by 1892), “Stone Bridge” (23 hits, none near 1903), “East Fruitland,” “Fruitland,” “carpet weaver” (all zero results). Not written up.
Jonathan Godfrey struck by John Recchia, Chatsworth, Oct. 1928. Searched “Recchia,” “Ricchia,” “Ricci,” “Godfry” (zero results each); “Godfrey” (58 hits, 1890s–1940, none in 1928; confirms a real Jonathan Godfrey bought the Chatsworth Club House in 1908, consistent with the forum’s Chatsworth-area Godfrey, but the 1928 assault item itself is not indexed); “Chatsworth assault” / “assault Chatsworth” (irrelevant 1938 court roundup). Not written up.
Atwood Tomlin’s 1932 shot-size conviction (Game Warden Roy R. Carson). Searched “Atwood Tomlin” (zero), “Tomlin” (164 hits, not narrowed further given zero hits on the specific angle), “Roy Carson” (3 hits, all unrelated marriages), “shot size” (1 hit, an unrelated 1921 Riverton murder). Not written up.

The Mirror index simply does not appear to hold the specific issues these four leads depend on (they may have run in a different Burlington-area paper, or in an issue outside this library’s holdings/indexing).

My-choice entries (search.json checked by subject before writing; none of these three subjects appear there)

George Lemmon, arson/self-death near Retreat, April 10, 1920 — NEW entry `lemmon-george-arson-suicide-retreat-1920`

Found via keyword “lost in the woods” (an incidental hit) and confirmed via “perished in the woods.” New Jersey Mirror index, April 14, 1920, p.4, col.5, headline “Burned to Death in a House He Fired.” Summary: George Lemmon, 60, living with brother James Lemmon and family near Retreat (Woodland Township), set fire to the outside of the house while James and his wife were off fighting an unrelated woods fire, went inside, and locked the door; two children in the house witnessed the fire being set; the elder, age 7, ran for help; the house and barn were destroyed and Lemmon’s remains found in the ruins. Wrote this up as a plain index-summary case: no coroner’s report, no obituary, and no independent confirmation of Lemmon’s identity or age was located this session, so the entry is explicit about relying on a single index summary. “Retreat” location pin borrowed from the site’s existing retreat-blueberry-flats-women entry coordinates, used as an area center only.

Isaiah Fouche, found dead near Mary Ann Forge, April 1873 — NEW entry `fouche-isaiah-mary-ann-forge-1873`

Found via keyword “found dead in the woods.” New Jersey Mirror index, April 10, 1873, p.3, col.1, headline “SUDDEN DEATH.” Summary: Isaiah Fouche, about 50, living in a cabin at Bear Hole Branch “near Mary Ann Ford’s” [Forge] in Pemberton Township, left home in good health to chop wood, did not return, and was found by his daughter lying dead beside a tree he had felled. Inquest by Benjamin Taylor, Esq.; verdict, death from heart disease. A natural death, written plainly with no embellishment. Location pinned to the site’s existing ghost-of-mary-ann-forge coordinates as an area center, since Bear Hole Branch itself is not otherwise located in this record.

Annie O’Donnell, drowned near New Lisbon, March 1909 — NEW entry `odonnell-annie-new-lisbon-1909`

Found via keyword “cranberry bog drowned.” New Jersey Mirror index, March 24, 1909, p.3, col.1, “LOCAL BRIEFS” item. Summary: Mrs. Annie O’Donnell, wife of Patrick O’Donnell, living in a house on H. M. Black’s cranberry bogs near New Lisbon, drowned in a creek there on March 22, 1909; the index item says the working theory was that a strong wind blew her into the water while she was drawing a bucket from the bank, and Coroner Seeds ruled it accidental with no suspicion of suicide raised in the item. Location pinned to the site’s existing New Lisbon locality coordinates as an area center, since the bog and creek are not otherwise located in this record.

Rejected leads (considered, not written up)

1936 Chatsworth–Warren Grove–West Creek forest fire (five killed, incl. Ranger Ira Morey) — fully confirmed in the index (New Jersey Mirror, May 27, 1936, p.1, “Five Killed, Many Injured in Greatest Forest Fires…”), but search.json already carries this exact subject (“The 1936 forest fire: five deaths, conflicting descriptions”), so not duplicated.
Mark Anthony Foulks parricide at Hanover Furnace, 1861 — fully confirmed in three index records (Mirror, July 4, 1861 “HOMICIDE”; Sept. 26, 1861 trial-calendar notice; Oct. 3, 1861 “TRIAL OF FOULKS, THE PARRICIDE,” verdict of second-degree murder, seven-year sentence), but another lane had already written entry/foulks-joseph-hanover-furnace-1861.html (found on disk, untracked, timestamped earlier the same session) before I got to this lead. Not duplicated.
Margaret “Peggy” Clevenger, burned in her Old Shore Road cabin, Dec. 1857 — the forum gives exact Mirror dates (Dec. 10 and Dec. 17, 1857) for this, but the surname “Clevenger” returns 272 hits with no working date filter, and none of “Peggy Clevenger,” “Margaret Clevenger,” “Dothey Clevenger,” “Blake Clevenger,” “Clevenger cabin,” “Clevenger burned,” “Mount Misery Cedar Bridge,” “hogs poisoned,” “old woman opium,” “Joseph W. Cox Hanover,” “Joel Mount Juliustown” surfaced the 1857 item. Given the corpus clearly holds 1857-era material (other searches return 1850s hits routinely), this looks like a genuine gap in what this keyword search can retrieve, not proof the record doesn’t exist. Left unwritten rather than guessed at.
Wesley Warner’s 1893 trial for murdering Lizzie Peak (Mount Holly), the 1906 Florence Allinson murder (Moorestown), the 1863 Keturah Ann Brooks trial, the 1872 “A Dead Body Identified” (Whitesville, Ocean Co.) and 1887 Isaac Shockley case (near Berlin, Camden Co.) all turned up in general “found/lost in the woods” searches but were set aside as being outside the Pine Barrens proper (western Burlington County suburbs, or too vaguely located) rather than stretched to fit.
Roy R. Carson (game warden) and Atwood Tomlin’s other biographical details (shanghaied to England, 1947 death in a picker’s shack) were not pursued further once the specific 1932 conviction could not be sourced.

Notes on the search tool

type=any behaves as an OR across the words given; there is no AND operator found (type=all returned zero results on a two-word query rather than intersecting).
No date-range parameter was found; limit can be raised past 200 (tested to 1000) but a search still returns only its true hit count sorted as requested, so a common surname with hits spread over more than a century can still bury or exclude a specific year within the returned page — there is no reliable way to jump to a decade for a generic surname search.
Distinctive multi-word phrases (a headline fragment, a rare surname combination) are far more effective than a single common surname.

Validation (root fact-check)

Adversarial re-check, September 12, 2026. All three index searches were re-run independently against netscape.bcls.lib.nj.us with the required fields[] parameters, searching by surname rather than by the discovery keyword.

All three index rows exist exactly as cited — date, page and column all match:

Lemmon → 28 results; hit 12, accession 20647, category “Obituary”, headline “Burned to Death in a House He Fired”, April 14, 1920, p.4 c.5.
Fouche → 2 results; hit 1, accession 60872, category “OBITUARIES”, headline “LOCAL AFFAIRS. / SUDDEN DEATH.”, April 10, 1873, p.3 c.1.
O'Donnell → 50 results; hit 19, accession 75241, category “OBITUARIES”, headline “LOCAL BRIEFS.”, March 24, 1909, p.3 c.1.

Source links in all three entries were repointed from the incidental discovery keywords to these surname queries, which reproduce the record reliably, and each now states the hit count, accession number and what was and was not inspected.

Corrections made

Retreat is in Southampton Township, not Woodland Township. The Lemmon entry, the items record and the map pin all placed Retreat in Woodland Township “near Chatsworth.” Nominatim puts the Burlington County hamlet of Retreat at 39.9165, -74.7127 in Southampton Township, which matches the coordinate the pin was already using (borrowed from retreat-blueberry-flats-women); Chatsworth is roughly 17 km away in a different township. The index record names no township at all — it says only “near Retreat” — so the township attribution was an unsupported addition on top of a wrong one. Corrected throughout; the “near Chatsworth” claim is cut.

Arson and suicide characterisations hedged. The Lemmon page’s title, H1, lede and meta description stated as established fact that he set the fire. The index summary does say so, but it is an indexer’s précis of one newspaper report; the record carries no inquest, no verdict, no police conclusion and no arson charge, and the only witnesses named are two small children, at third hand. The page now states the sequence as what the report says and adds an explicit paragraph that arson/suicide is an interpretation of that one account. The slug retains “arson-suicide” (unchanged to avoid breaking links), but no rendered text asserts it.

Dateline attributions corrected in all three. Each entry claimed the article’s own dateline gave the exact date. It does not. The Lemmon and Fouche summaries print only “Saturday afternoon” and “Tuesday of last week”; those dates (April 10, 1920 and April 1, 1873) are arithmetic from the issue dates, and the pages now say so. The O’Donnell record does carry a date, but as the index’s parenthetical gloss (“presumably a reference to March 22, 1909”), not as the newspaper’s text; the entry now attributes it to the index. Weekdays were checked: April 14, 1920 was a Wednesday (so Saturday = April 10); April 10, 1873 was a Thursday (so Tuesday of last week = April 1); March 24, 1909 was a Wednesday (so Monday = March 22). All three resolutions are sound.

Unsupported detail cut. Fouche: “Benjamin Taylor, Esq., a local justice acting as coroner’s-inquest officer, as was routine practice for sudden rural deaths at the time” — the record gives only his name and that he held the inquest; the office, and the claim about routine practice, were invented. Cut. O’Donnell: “isolated company housing scattered through the Pine Barrens’ bog country” — the record says nothing about who owned or provided the house, or about any other families. Cut and replaced with a statement of what the item does and does not document.

Verdict language tightened. O’Donnell’s lede and meta said the death was “ruled an accidental drowning.” The record says Coroner Seeds believed it accidental and issued a burial certificate to that effect — a coroner’s disposition, with no inquest, jury or post-mortem mentioned. Reworded. Fouche’s inquest verdict, by contrast, is genuinely given in the record and stands as written.

Quoting reduced. Passages that quoted the indexer’s summary while framing it as the newspaper’s own wording have been paraphrased or relabelled. Where quotation marks remain they are short and explicitly attributed to the index summary.

Checks that passed

No duplication. search.json (374 entries) holds no entry for Lemmon, Fouche or Annie O’Donnell. The nearby New Lisbon and Retreat entries (new-lisbon-diabetic-missing-1954, new-lisbon-anderson-williams-1975, retreat-blueberry-flats-women) are unrelated subjects.
Factbox. All three carry “Research checked: September 12, 2026”.
Titles, meta descriptions and ledes now match the body text and each other.
Mary Ann Forge coordinate, flagged for scrutiny: it returns nothing in GNIS or Nominatim and cannot be independently verified. It traces only to session-provided NJPB community map data, backed loosely by the 1849 Otley & Whiteford map discussed in ghost-of-mary-ann-forge. It is therefore used here as an approximate landmark center only, never as a site pin. One point in its favour: reverse-geocoding 39.942894, -74.579638 returns present-day Pemberton Township, matching the 1873 record’s “in Pemberton Township” — so the existing pin’s “Woodland Township” label looks like the error, not the coordinate. Bear Hole Branch could not be located at all.

Map pins

All three were already area, correctly — no “near X” phrase was pinned as a site. Radii were widened to be honest about how little the records give: Retreat 1500 → 3000 m, Mary Ann Forge 1500 → 3000 m, New Lisbon 3000 → 4000 m. Names were changed from the specific unlocated feature (“Bear Hole Branch…”, “H. M. Black’s cranberry bogs…”) to the locality actually being pinned, so the pin label no longer implies a precision the coordinate lacks. Notes now say explicitly that no pin marks a dwelling: the Lemmon house and the O’Donnell house were both private homes and are deliberately not located.

Recommendation

All three entries are keepable as single-index-record notes. None should be promoted beyond that without the microfilm. The Lemmon entry is the one to watch: it is the only one making a claim about intent, and it rests entirely on one indexer’s précis of one report.