← The land

Story · 1859–1941 · v2.1

Kopjes Kraal: a farm that lost a quarter of itself, and a road nobody uses

Crown land surveyed in 1859 at 6,325 morgen and resurveyed four years later at 4,982 - about 1,343 morgen fewer, unexplained. Granted in 1871 to Gerolm Marincowitz of Trieste, whose surname reaches the Karoo from the Adriatic and not from the Netherlands. A stock route across it was ended by a line in a provincial gazette in 1926, and half of what was left became Kromrivier in 1934.

Crown land, surveyed in 1859, granted in 1862, resurveyed in 1863 and granted again in 1871 to a man born in Trieste. Between the first survey and the second the farm lost about 1,343 morgen, a quarter of itself, and nothing in the record explains why.

Two surveys, four years apart, and a quarter of the ground missing

In November 1859 J.A. Thwaits and J.W. du T. Melvill measured it as Crown Land in the Field Cornetcy of Traka: 6,325 morgen 330 square roods, bounded north by Perseken Hoogte, east by Minies Kraal and Oude Muure, south by Klip Fontein, west by Rondable. They marked an outspan on it. It was granted on 11 September 1862 under P.A.Q.1-17.

In November 1863 L. Marquard measured the same ground again and made it 4,982 morgen 370 square roods.

Same farm, same four boundaries named in the same order, 1,343 morgen fewer. That is not a rounding difference or a conversion error; it is roughly the size of a small farm. No document here explains it, and this page does not guess. not yet verified

Thwaits is not a stranger here. He came back twenty-four years later to resurvey Middelwater for the Oosthuizens, which is how one surveyor's hand runs through two of these farms a generation apart.

The man it was granted to

The 1863 resurvey carried its own title, and on 17 May 1871 the grondbrief P.A.Q.1-28 was issued ten gunste van GEROLUM MARINCOWITZ.

Gerolm Marincowitz was born in Trieste on 11 February 1805 and reached this country in 1838. He married Hester Carolina Adriana Hitge at Uitenhage on 5 June 1844 and they had ten children. He died at Vrolykheid on 13 May 1890.

The Marincowitz name reaches the Karoo from the Habsburg Adriatic, not from the Netherlands. Marincovich is the older spelling. Almost every other surname here comes down through the Cape from the Dutch, the German states or the Huguenot arrivals, and this one does not. It comes from a port on the Adriatic, and it is here because one man crossed and stayed.

GEROLUM on the deed is worth keeping as written. The registers spell this family Gerolm, Ger'olm, Gerholm and Gerolum, and the document's spelling is kept rather than tidied, because spelling drift is evidence.

The trek that was cancelled

Across the 1863 diagram runs a trek: a stock route, the old right to move animals over somebody's ground.

On 15 December 1926 the Surveyor-General's F.F. Elliott wrote on the sheet that the trek from x to y, indicated by Surveyor J.A. Loubser, had been cancelled under section 254 of Ordinance 13 of 1917, by notice in the Provincial Gazette of 5 March 1926, Notice 71, page 359.

A right of way older than the title, ended by a line in a gazette and a note in an official's hand on a sixty-three-year-old diagram. The Kromrivier and the Traka River both run through this farm on that same sheet, the third of this district's records to put the name Traka on surveyed ground.

Then it is cut in half

In 1934 Kromrivier was cut off as Portion 1: 2,327.9295 morgen, survey record E.714/1933, diagram 3281/33, deed 1934-73-3876. Nearly half of what was left after 1863 went in one operation, and the rest became the Remainder.

Seven years later, on 30 October 1941, Gerolm Samuel Marincowitz was buried on the farm: the grantee's grandson through his son Nicolaas James, carrying his name. His is the only stone recorded here in the Kopjes Kraal burial ground, and it has not been read. The dates come from eGGSA's caption, which is somebody else's transcription of a photograph.

The name, and the trap it sets

There is a Koppieskraal in this district's war history: the farm where Gideon Scheepers was captured, ill and abandoned by his own commando, on 11 October 1901.

It is not this farm. Ours is Kopjes Kraal No. 123, out towards Klaarstroom at 22.71° east; the capture farm lies west of Prince Albert near Bosluiskloof, about 102 km away. The names are identical and the ground is not, which is the oldest trap in this district and worth writing down every time it is avoided. That finding belongs to the war story and is repeated here because anyone who meets both names in one week will join them without help.

Built from four Surveyor-General sheets, their endorsements and the farm's own chain of title. The 1,343 morgen the farm lost between 1859 and 1863 is an open question, not a rounding: it is stated here as unexplained rather than smoothed over.

What this story is about

Sources

Everything this story rests on, with a link to the original wherever one exists. Where a citation is incomplete it says so, an incomplete citation is a research gap, not a formatting problem. If your work is here and the credit is wrong or thin, please tell us and it will be fixed.

  1. Surveyor-General diagram, 1927/1860, November 1859 Surveyor-General / Deeds e-portalSG Dgm 1927/1860
  2. Surveyor-General diagram, 1035/1871 (pt 1), November 1863 Surveyor-General / Deeds e-portalSG Dgm 1035/1871 pt1
  3. Surveyor-General diagram, 1035/1871 (pt 2), November 1863 Surveyor-General / Deeds e-portalSG Dgm 1035/1871 pt2
  4. Diagram verso, 1035/1871 (verso) Surveyor-General / Deeds e-portalSG Dgm 1035/1871 (verso)
  5. Kopjes Kraal No.123 diagram (1859 survey) Surveyor-General / Deeds e-portalSG Dgm 1927/1860
  6. Kopjes Kraal No.123 diagram, part 1 (1863 survey) Surveyor-General / Deeds e-portalSG Dgm 1035/1871 pt1
  7. Kopjes Kraal No.123 diagram, part 2 (description) Surveyor-General / Deeds e-portalSG Dgm 1035/1871 pt2
  8. Kopjes Kraal No.123 - reverse: deduction table Surveyor-General / Deeds e-portalSG Dgm 1035/1871 (verso)

Research notes

[8 Aug 2026] Written as the second of the farms that had a record and no history: 8 dated events, 4 Surveyor-General sheets and no page. THE 1,343 MORGEN IS LEFT AS AN OPEN QUESTION - two surveys four years apart name the same four neighbours in the same order and differ by a quarter of the farm, and no document here explains it. The Scheepers Koppieskraal was adjudicated in STO-0032 as a DIFFERENT farm 102 km away; that finding is repeated on this page because anyone meeting both names will join them without help.

[12 Aug 2026] VOICE REWRITE to STORY-VOICE.md, both languages. en 0/3/0 -> 0/0/0, af 5/4/0 -> 0/0/0 (narration/bold-sentence/punchline). Five Afrikaans narrations, all the same move - 'hierdie projek weet nie waarom nie', 'hierdie rekord se plase', 'wat hierdie projek aanteken'. The two open questions are UNCHANGED and still stated as open: the 1,343 morgen keeps its [TO VERIFY], the eGGSA caption is still named as somebody else's transcription of a photograph, and the stone is still recorded as unread.

THE MAN IN THE GRAVE IS THE GRANTEE'S GRANDSON, and the page did not say so. It named Gerolm Marincowitz PER-000020, granted the farm in 1871, and Gerolm Samuel MarincowitzM PER-000056, buried on it in 1941, and left a reader to guess whether that was one man, a son, or a coincidence - on a page whose closing section is about exactly that kind of trap. parentage.csv already held the answer: PER-000056 -> father PER-000021 Nicolaas James -> father PER-000020. Now stated in both languages with Nicolaas James linked.

DRIFT FLAG CHECKED, both records harmless. PER-000056 gained a note of 11 Aug 2026 saying he MAY be the sitter in the double portrait MED-000022, on initials and age, marked [TO VERIFY] - an open lead, not a contradiction, and too thin for the page. He also gained a baptism at Prince Albert on 18 Sep 1870 (SRC-358), which corroborates his BIRTH; the page cites eGGSA only for the 1941 burial, so its caveat stands as written. PAR-000029 Oude Muure gained a note of 10 Aug 2026 warning that a SECOND Oude Muur exists in the Calvinia division - the page names Oude Muure only as an 1859 boundary of farm 121 in Prince Albert, which is the right one, so nothing to correct.

Bold came off two links and off the Habsburg sentence; the Marincowitz-from-Trieste point reads as prose now. Closing note stops naming ownership rows and names the chain of title. verify_af passing.

[12 Aug 2026] BOLD PASS, both languages, en 25 runs -> 1 and af 24 -> 1. 0 page and said it still had lots of bold words and phrases, on a file the voice check had scored 0/3/0 and closed. He was right and the check was blind: rule 4 counts bold carrying a whole SENTENCE, and this story had one of those in English and none in Afrikaans while running 11.5% of its words bolded against the S035 benchmark's 1.1%. A density check and an en/af asymmetry check now exist in checks.py so no file closes on that gap again. WHAT WAS BOLD AND IS NOT: every date, extent, deed and survey number, the field cornetcy, the outspan, the grantee's name in caps, Kromrivier, Portion 1, Koppieskraal, the 102 km. STORY-VOICE.md names those exactly - not every name, date, farm, extent or deed number qualifies simply because it could appear in an index, and the linked ones already carry weight. The benchmark S035 bolds nothing in its narrative at all. WHAT IS KEPT: '1,343 morgen fewer' / '1 343 morg minder', the discrepancy the whole page is about. AND ONE RULE APPLIED RATHER THAN RELAXED: 'trek' was bold and is now ITALIC, because it is a Dutch/Afrikaans term and that is what italics are for here. 'It is not this farm.' lost its bold and reads as it did in Afrikaans, which never had it - a whole sentence is never bolded for emphasis. NO WORD, FACT, HEDGE, [TO VERIFY] OR LINK CHANGED: the edit removed emphasis markers and added two. verify_af 1/1.

The research behind this page, kept separate from it: dated working notes, open questions, and what is not yet settled.