Crossroad Gold Corp - An Open-File Forensic Look at the Pambula System, EL 9833
Every so often a piece of ground rewards the person who reads its file carefully rather than the person who simply walks it. The Pambula goldfield on the far south coast of New South Wales — held today under Exploration Licence 9833 by Rox-ex Pty Ltd, now a wholly-owned subsidiary of Crossroads Gold Corp (TSXV: CRG) — is one of those places.
Pambula is a low-sulphidation epithermal gold–silver system hosted in the bimodal Boyd Volcanic Complex, sitting at the southern end of the ~320 km-long Eden–Comerong–Yalwal Volcanic Zone, the same Late Devonian rift corridor that hosts Yalwal (~81,000 oz historical production) to the north. It is a proven gold producer, not a conceptual one: the recorded field production is 1.253 t of gold (roughly 40,264–40,908 oz) from about 40,908 t of ore between 1892 and 1941, which implies a recovered head grade in the order of 30 g/t Au. Grades of that order are rare in any Australian goldfield, and rarer still in one that has seen only 39 drillholes in its entire history.
What makes Pambula a compelling case study for open-file reinterpretation is the mismatch between how rich the system is and how lightly it has been tested. Recovery in the early mining era ran as low as 60% because the gold is exceptionally fine-grained and partly sulphide-locked — meaning the mined ore actually carried materially more gold than was ever recovered. The system is demonstrably silver-rich (electrum, naumannite, stephanite, proustite and a suite of Ag sulphosalts and selenides), yet silver was never systematically assayed. And no hole has ever tested the system below ~230 m vertical, despite fluid-inclusion evidence pointing to depth continuity beneath the preserved workings.
Crossroads' acquisition of the licence in 2026 places the entire historical goldfield — every known lode and every archived drillhole — under a single, modern, well-capitalised holder for the first time in a generation. That is a well-placed tenure position: the ground consolidates a coherent, high-grade epithermal system that prior eras only ever tested in fragments. The opportunity here is not a new discovery so much as the validation and modern re-reading of ground Crossroads already holds. This article walks through what the open file reveals when the eras are cross-referenced — and why, on the public record, the grade at Pambula is still open.
2. Exploration history — one era, one operator at a time
The Pambula file reads as a sequence of capable operators, each of whom did good work within the model, technology and commodity environment of their day, and each of whom left for reasons of corporate cycle rather than geology.
Historical mining era (1889–1941 · numerous small operators). Gold discovered 1889; production principally from the Faulkner mine (~60% of field output), Victory, Black and Berry's, Pambula Tunnel, Hidden Treasure, Morning Star and Diorite/Killaloe workings. Objective: direct mining of high-grade fissure lodes. Recovered grade ~30 g/t Au; recovery as low as 60% because the fine, sulphide-locked gold defeated the milling technology of the period. Mining wound down with the exhaustion of easily-treated oxide ore and the economics of the era, not the exhaustion of the system.
1982–84 · Arruba/Circular Quay Holdings → Getty Oil Development Co. Deposit model: epithermal Au across the district. Work: mapping and soils; assessed "potential for economic gold deposits high." Departure: relinquished on corporate grounds following Texaco's acquisition of Getty — a boardroom event a continent away, entirely unrelated to the ground.
1984–86 · Homestake Australia (with Circular Quay Holdings JV). Model: blind high-grade gold shoots below the boiling interface. Target: repetition of structurally-controlled shoots. Work: the field's first modern drilling — three rotary-percussion holes (P1–P3, 1984), two RC/diamond (P4–P5, 1985), and four diamond holes (P6–P9, 1986; ~972 m). This is the campaign that produced most of the field's best intercepts. Departure: consistent with major-company portfolio cycling of the late 1980s.
1986–87 · Gold Fields (RGC). Model: en-echelon mineralisation within NW-trending shears. Work: field mapping. Concluded mineralisation confined to thin N–S fractures — a reading appropriate to the structural paradigm of the day.
1987–91 · Millaroo Mines NL / Circular Quay Holdings JV. Model: geophysically-defined structural targets. Work: gradient-array IP, ground magnetics, 61 stream-sediment samples, and 23 shallow RC holes (PRC10–30, 1988) plus 5 further RC holes (PRC31–35, 1990) on magnetic anomalies. This era produced the isolated 31 g/t bonanza intercept in PRC32. Departure: the joint venture relinquished in 1992 on RGC's opinion that open-pittable (bulk-tonnage) potential was minimal — an assessment made explicitly against the bulk-tonnage paradigm then in vogue, when the gold price and the industry's appetite were both oriented toward large low-grade targets rather than narrow high-grade shoots.
c.1995 · Eddaglide Pty Ltd. CHIM–MPF geochemistry over the ground; limited campaign.
2006–08 · Ironbark Gold Limited. Model: Au, with a bulk-resource lens. Work: digitised the full 38-hole drill database, 3D modelling, core inspection; recommended five confirmation RC holes. Concluded the data were insufficient for a JORC resource and targeted a "small, relatively high-grade" body. Departure: the confirmation holes were never drilled — consistent with junior funding conditions through the 2008 financial crisis.
2007–11 · Oroya Mining Limited. Model: regional epithermal Au. Work: a 178-site BLEG + multi-element drainage survey that defined the Yowaka River "Anomaly 1" on the southern projection of Pambula stratigraphy. Departure: the survey lapsed with ~29 planned sites unsampled when the programme ceased — an incomplete regional dataset, not a negative one.
2017–20 · Sapphire Resources Pty Ltd. Model: Au, with an explicit deep-feeder and second-solution 3D reading. Work: negotiated FCNSW Forest Permit and LiDAR access, executed a native-title clearance, commissioned independent 3D voxel modelling of the full database (which resolved a previously-unrecognised east-dipping trend), and designed, budgeted and permitted-in-progress a 1,000 m + 300 m diamond programme. Departure: the drilling was never executed — a program that reached the starting line but not the drill pad.
2025–present · Rox-ex Pty Ltd → Crossroads Gold Corp (TSXV: CRG). EL 9833 granted 17 November 2025; Rox-ex acquired 100% by Crossroads, closing 28 May 2026. This is the latest and best-informed chapter: for the first time the whole goldfield sits with a single modern holder able to act on the accumulated open-file knowledge — including a fully-designed drill program and an archived diamond core library waiting to be re-read.
3. Why the grade is still open
Cross-reference the eras and a consistent picture emerges: each operator tested a slice of the system with the tools and model of their moment, and the slices never overlapped enough to close it out.
The depth axis is untested. No hole has ever gone below ~230 m vertical, yet fluid-inclusion temperatures (>300 °C) and salinities (12–17 wt% NaCl eq.) are high for the epithermal environment — the signature of a system whose higher-temperature roots remain below the drilled floor. The 1980s steep-fissure model simply never reached for them.
The geometry axis is under-sampled. Independent 3D voxel modelling of the same 1980s assay database (2018) resolves shallowly east-dipping high-grade trends. Almost every historical hole was drilled −50° to −65° toward ~270° — a geometry that would strike obliquely down an east-dipping body rather than cross it. If the 2018 model is right, historical drilling systematically under-sampled true widths.
The silver axis is essentially blank. 36 of 39 holes were assayed for gold only. In a system mineralogically defined by silver sulphosalts and selenides, the silver endowment of every headline intercept — including the 28.33 g/t Au metre in P7 and the 31 g/t pair in PRC32 — is simply unknown on the public record.
The pathfinder axis doesn't exist yet. No modern multi-element (As–Sb–Se–Te–Hg–Tl) vectoring has ever been run on the core, so the standard tools for reading "up-system vs down-system" have never been applied here.
Two intercepts were never even headlined. PRC13 ended in mineralisation at 39 m with its final three samples averaging 6 m at 4.97 g/t Au (last metre 7.0 g/t), and its twin PRC13A returned 24 m at 1.61 g/t — results that only surface when the raw digitised assays are recomputed, not from any historical summary.
The miss, precisely — (copy-paste blurb): Pambula was mined at roughly 30 g/t and then tested by just 39 shallow holes, none below 271 m, almost all assayed for gold alone and drilled to a steep-fissure model that a 2018 re-modelling of the very same data no longer fully supports. The deep boiling-zone target was designed, budgeted and permitted by an earlier operator — and then the funding cycle turned before the rig arrived. The blind spot was never the ground; it was the commodity prices, the assay budgets and the structural paradigms of the eras that drilled it. Every one of those constraints is an artefact of history, and every one is now resolvable at desktop-and-re-assay cost by the current holder.
4. Significant intercepts
Computed values are length-weighted, uncut downhole averages recomputed from the digitised 1 m/2 m assay records; "as reported" figures are the historical composites. True widths are not established for any intercept.
Hole | Operator (year) | Intercept | Source report | Why it matters |
PRC32 | Millaroo/Circular Quay (1990) | 2 m at 31 g/t Au from 26 m | GS1990/262 | Isolated bonanza hit 470 m ESE of the main cluster, on a magnetic target — and no follow-up hole was ever drilled before relinquishment. |
P7 | Homestake (1986) | 5 m at 6.62 g/t from 99 m, incl. 1 m at 28.33 g/t | GS1988/075 | Deepest significant hit; the 28.3 g/t metre was independently confirmed by GSNSW sampling. Mineralisation persists to 235 m — the system does not die downward. |
P1 | Homestake (1984) | 5 m at 7.60 g/t from 20 m | GS1984/319 | First modern hole into the Pilot Fissure; anchors the near-surface high-grade shoot. |
PRC13 | Millaroo (1988) | 6 m at 4.97 g/t from 33 m (final 2 m at 7.0 g/t) | GS1990/262 | Hole terminated in mineralisation at 39 m with grade rising downhole — a shoot left open by the drill string, not closed off. |
PRC13A | Millaroo (1988) | 24 m at 1.61 g/t from 31 m, incl. 8 m at 2.35 g/t & 2 m at 5.0 g/t | GS1990/262 | The re-drill of PRC13 shows real width around the high-grade core; only emerges on recomputation of raw assays. |
P4 | Homestake (1985) | 22 m at 2.13 g/t from 27 m | GS1985/026 | Broad Pilot Fissure interval (reported 2.20 g/t at 1 g/t cut-off); demonstrates width, not just fissure. |
P6 | Homestake (1985) | 23 m at 1.10 g/t from 64 m, incl. 9 m at 1.77 g/t | GS1985/026 | Down-dip of P4; confirms the shoot continues below the near-surface intercepts. |
5. Geological setting
Pambula sits on the far south coast of NSW, ~5.5 km south-west of Pambula township, within Nullica State Forest, at the southern end of the Eden–Comerong–Yalwal Volcanic Zone — a narrow, north-trending Late Devonian rift corridor compared in the literature to the Yellowstone–Snake River Plain and to active intracontinental rifts like the Taupo Volcanic Zone. The host is the bimodal Boyd Volcanic Complex: coherent rhyolite lavas and dykes, ignimbrites and tuffs, basaltic lavas and intrusives, and intercalated fluvial-lacustrine sediments, all metamorphosed to prehnite–pumpellyite grade. The package was gently folded and faulted during the Early Carboniferous Kanimblan event, producing the north-trending Bald Hills Anticline on whose crest Pambula Central sits.
Mineralisation is a low-sulphidation epithermal Au–Ag system: disseminated, stockwork and high-grade vein styles with chalcedonic (crustiform, colloform) quartz, bladed-carbonate pseudomorphs and breccia fill. The ore bodies characteristically "lack any form of distinctive reef" — limits are gradational and defined by assay, which is exactly why a well-designed modern drill-and-assay program is the tool the ground has always needed. The rich shoots occupy three parallel, north-striking structures — the Pilot, Hidden Treasure and Diorite fissures — interpreted as basement-related extensional structures, all within ~2 km of the ESE-trending Pipeclay Creek lineament. The mineralogy is distinctly silver-rich (electrum the principal gold mineral, with naumannite, stephanite, proustite, acanthite and other Ag sulphosalts and selenides), with Au:Ag ratios reaching 1:1 in the lodes.
6. Top 3 targets
Target 1 — Deep boiling-zone / feeder beneath Pambula Central (conceptual). All historical work tests only the upper ~230 m of a system whose fluid chemistry points to higher-temperature gold styles below the boiling interface, with Homestake's own 1985 memo framing the play as "repetition of blind high grade gold shoots" below that interface. Testing it — via the ~1,000 m diamond hole already designed and budgeted by the previous operator in 2018 — would resolve whether Pambula has depth continuity and a feeder root, the single question that most governs the system's ultimate scale. This aligns directly with a drill program Crossroads already inherits fully permitted-in-progress; it is the highest-value single hole on the project.
Target 2 — East-dipping mineralised trend (data-supported, geometry unverified). The 2018 voxel model defines shallowly east-dipping high-grade zones inconsistent with the steep-fissure orientation of all 39 historical holes. A single purpose-oriented diamond hole (the 2018 program's ~300 m companion) would test whether the historical database has been systematically under-reading true widths. Testing this complements a modern re-interpretation strategy: it re-reads existing data rather than chasing new ground, precisely the kind of leverage a consolidated single-holder position enables.
Target 3 — PRC13/PRC13A shallow high-grade zone (data-supported, drill-ready). This zone sits under less than 50 m of cover, is open down-dip and down-plunge, has grade increasing at the PRC13 termination, and has never been tested directly beneath. It offers near-term, low-cost, shallow drill validation of a high-grade shoot — a fast, de-risking complement to the deeper conceptual targets, and the kind of quick result that underwrites a longer program.
Not a JORC-compliant statement; historical results reported from open-file sources and not verified by a Competent Person.
All data sourced from publicly available NSW DIGS open-file records and public ASX disclosures. CAADIA Metals has no interest in, and no commercial relationship with, Crossroads Gold Corp; the subject tenure is wholly held by Crossroads Gold Corp (via Rox-ex Pty Ltd).
This article is general information only and not investment advice or a recommendation regarding Crossroads Gold Corp securities.
Full technical synthesis, intercept overview, drill plan and source register available on request.
