Gold mining is mostly a business of moving dirt. The pits are enormous, the haul trucks are enormous, and the metal hiding inside all that rock gets measured in grams per tonne rather than pounds. Two grams to the tonne is a working mine. One gram can still be worth digging if the pit is shallow and the mill is close enough.<\/p>\n
Ramelius Resources just reported an assay from Western Australia that came back at 996.<\/p>\n
The hit is 3.38 meters of drill core, roughly 11 feet, cut from 333 feet below the surface at the Never Never deposit on the Dalgaranga project, about 50 miles northwest of the company’s Mt Magnet processing plant in the Murchison. Australian Mining<\/a> reported the result on July 22 along with the rest of the company’s second-half drilling, and the same Never Never program produced 13 meters at 34.5 grams, 8.8 meters at 20.2 grams and 16.1 meters at 14 grams.<\/p>\n Those other three numbers would be the headline on almost any other day.<\/p>\n Gold gets priced by the troy ounce, which is 31.1 grams. So 996 grams to the tonne works out to a shade over 32 troy ounces of gold sitting in one metric ton of rock. With gold trading around $4,120 an ounce this week, that single ton carries roughly $132,000 of metal.<\/p>\n For contrast, look at what the same company plans to feed its own mill. Ramelius told the ASX in April that it wants a future Gilbey’s underground mine to displace the low-grade material already scheduled through the plant, specifically 1.9 million tonnes at 0.8 grams in FY29 and 1.8 million tonnes at 0.6 grams in FY30<\/a>.<\/p>\n A tonne of that 0.6-gram rock holds about $80 of gold at today’s price. The 996-gram intercept holds about 1,600 times as much, out of the same ground, at the same mine, drilled by the same company in the same financial year.<\/p>\n And even then, the spectacular tonne is 99.9 percent not gold. Just under 2.2 pounds of metal in a 2,205-pound ton of Murchison greenstone, which is the part that never quite lands when people picture a gold mine.<\/p>\n This was infill drilling, which is worth understanding because it changes what the number means. Exploration drilling goes looking for rock nobody has tested. Infill drilling goes back into ground the company has already modeled and puts holes between the existing ones, to confirm the grades and shapes it has been telling the market about.<\/p>\n Ramelius has good reason to be checking. Never Never carries a maiden probable ore reserve of 7.0 million tonnes at 7.3 grams per tonne for 1.6 million ounces, and the pre-feasibility study the company filed in October 2025 built an 11-year mine plan for 1.8 million ounces at an all-in sustaining cost of A$1,128 an ounce<\/a>, or about $790. That is a very low cost per ounce, and it only works if the grade is really there.<\/p>\n The mine is no longer theoretical. Ramelius hauled its first Never Never ore in January, 31,000 tonnes of development rock at 3.6 grams, and trucked it the 50 miles to Mt Magnet rather than building a separate plant. Higher-grade stockpiles were slated into the circuit during the June quarter, and the company closed FY26 at 192,182 ounces, inside its 185,000 to 205,000 ounce guidance for a sixth straight year.<\/p>\n Trucking ore between sites is its own quiet industry, and the fleets doing it have been getting stranger. Komatsu’s driverless haulers are now moving gold ore across Nevada with nobody in the cab<\/a>, on a fleet that has shifted billions of tons of rock.<\/p>\n Two things keep this honest. The first is that 996 grams describes 3.38 meters of core, not a mine, not a stope, and not a mining block. The reserve grade at Never Never is 7.3 grams. Nobody at Ramelius is suggesting otherwise.<\/p>\n The second is the nugget effect, which is the reason experienced geologists get careful rather than excited when a number like this lands. Where gold occurs as coarse visible particles instead of fine disseminated specks, a single assay can swing wildly depending on whether the drill happened to clip a cluster of grains. Hit one and the sample reads spectacular. Miss it by a few centimeters and the same vein reads barren. It cuts both ways, and it is why companies drill hundreds of holes instead of trusting one.<\/p>\n It is also worth knowing that Australia has produced better. The Swan Zone at Fosterville in Victoria went into reserve at 58.8 grams per tonne and was mined at around 23 grams<\/a>, and that was an entire zone rather than one interval. Infill drilling at Swan returned 6.4 meters at 598 grams and 3.6 meters at 625 grams<\/a>, over widths a couple of times what Ramelius just logged.<\/p>\n So 996 is a genuinely exceptional intercept rather than a record. Which is roughly how the company itself framed it, tucked into a results release alongside a dozen other holes.<\/p>\n Ramelius does not have a long history at this deposit. Never Never came with Spartan Resources, the ASX-listed owner of Dalgaranga, in a deal worth about A$2.4 billion that closed in July 2025. Ramelius took over operations at the site the following month.<\/p>\n The logic was straightforward. Ramelius had a processing hub at Mt Magnet with spare capacity and a mine plan increasingly reliant on low-grade material. Spartan had a very high-grade deposit 50 miles up the road and no plant worth feeding it into. Buying the grade was cheaper than finding it.<\/p>\n Australian ground with genuinely scarce material in it has been repricing hard for a couple of years now, whether that is gold or the critical minerals nobody thought about until China restricted them. A 168-year-old antimony mine in New South Wales drew a rejected $470 million takeover bid<\/a> on much the same logic.<\/p>\n The 996-gram hole came out of a program that got expensive. Ramelius turned 165,000 meters of drill core in the second half of FY26 alone, more than 100 miles of hole, and finished the year having spent A$103.9 million on exploration and resource definition against guidance of A$80 to A$100 million.<\/p>\n Dalgaranga took 7,350 meters of infill drilling at Never Never and another 12,075 meters targeting the Four Pillars and West Winds areas of Gilbey’s underground, where holes returned 1.6 meters at 130 grams and 3.7 meters at 38.2 grams. An updated resource estimate is underway there to support a pre-feasibility study on a second underground mine.<\/p>\n Mt Magnet did better than most sites would consider possible. The Galaxy underground mine returned 20.2 meters at 40.7 grams from 189.1 meters, including 11 meters at 71.7 grams. Franks Tower returned 14.76 meters at 54.5 grams, with half a meter of that running 740. Over at the Cue project, Lena produced 19.7 meters at 5.73 grams from 445.4 meters, deep enough that Ramelius now plans a scoping study on it.<\/p>\n Peter Ruzicka, the company’s executive general manager of exploration, tied the results directly to the money spent. “The application of expanded exploration budget in FY26 resulted in successful identification and development of multiple high-grade targets,” he said in the release, pointing at Galaxy and Gilbey’s as the areas most likely to displace low-grade material already sitting in the mine plan.<\/p>\n The board has approved A$90 to A$110 million for FY27, covering roughly 240,000 meters, or close to 150 miles of drilling. Most of that goes back into the Murchison, a part of Western Australia remote enough that mines out there increasingly generate their own electricity, like the WA gold operation that ran 155 hours on solar, wind and a battery with every engine switched off<\/a>.<\/p>\n Ramelius is due to publish an updated group mineral resource and ore reserve statement in late August, and that document decides considerably more than one core run does. It is where the infill drilling either firms up the Never Never grades the mine plan is built on, or quietly does not.<\/p>\n Reconciliation will settle the rest. Mills have a way of reporting what a deposit is really worth, and Mt Magnet has been blending Never Never ore into its feed since the March quarter. If the blocks come back close to the model, Ramelius has an 11-year, 1.8-million-ounce mine running at A$1,128 an ounce while gold sits above $4,100.<\/p>\n If they come back light, 996 grams per tonne will have been a very good day at the assay lab and nothing more. Either way, it is the kind of number that gets printed out and taped to the wall of the core shed, which is probably where it belongs.<\/p>\n","protected":false},"excerpt":{"rendered":" Gold mining is mostly a business of moving dirt. The pits are enormous, the haul trucks are enormous, and the … <\/p>\nAlmost a kilo of gold per tonne, and it’s still 99.9 percent rock<\/h2>\n
The hole was drilled to check homework, not to find anything<\/h2>\n
One assay is not an orebody, and Australia has already seen richer<\/h2>\n
Ramelius paid A$2.4 billion for the company that owned this hole<\/h2>\n
A$103.9 million of drilling bought a lot more than one good number<\/h2>\n
The number that actually settles this arrives in August<\/h2>\n