The highest number on an assay certificate is often the least useful number in a new prospect. A small miner may have chosen the brightest copper fragment or the quartz chip with the most visible gold precisely because it looked exceptional. The laboratory can measure that piece accurately, but its result does not describe the rock beside it, the depth below it, or the amount that can be mined. “Proving a claim” is the patient work of replacing that vivid first impression with enough reliable evidence to answer a harder question: is there a valuable mineral deposit in place that can be extracted, processed, and sold under real costs and lawful conditions? A claim location and a high-grade specimen are beginnings, not conclusions.
The BLM discovery guidance sets a legal frame. A valid lode discovery needs mineral-bearing rock in place within the claim and enough value to justify prudent expenditure, with a reasonable prospect of profitable marketing. A placer has its own proof requirements. This is distinct from a company advertising a formal “proven reserve,” a technical term used in mineral reporting under defined standards. The USGS resource-and-reserve explanation illustrates that more sampling raises confidence and that only economically recoverable material belongs in reserves. A small miner should be careful with words as well as rocks: “mineral occurrence,” “prospect,” “estimated resource,” and “reserve” do not mean the same thing.
Begin with the question the sample is meant to answer
A sample is useful only in relation to a question. “Is copper present in this green crust?” can be answered with a selected chip, provided it is labeled as selected. “What is the average copper grade across this three-foot zone?” requires a different design, such as a representative sample across the measured interval where authorized and safe. “Does the zone continue for 200 feet?” cannot be answered by either single chip; it needs mapped exposures, repeated samples, or perhaps drilling. “Can the deposit make money?” adds mining loss, processing recovery, transport, permitting, reclamation, and a buyer. These questions grow in cost and consequence. The cheapest valid next test is the one aimed at the biggest remaining uncertainty.
The USGS’s introduction to mineral-deposit models distinguishes local observations from regional interpretations. That is a good model for sampling. First map what is actually exposed: rock types, fractures, bedding, alteration, vein widths, and weathered versus fresh material. Do not extrapolate a surface line across a covered hillside as a measured ore body. Draw a cross section with solid lines for observed contacts and dashed lines for inferred ones. Keep the original field notes. A later geologist may revise the interpretation, but they cannot recover the observations if a promotional sketch replaced them.
For a southwest New Mexico example, imagine a hypothetical narrow copper vein in granite, not a specific vacant claim. One selected green specimen assays high. The first uncertainty is whether the mineralized material occurs in place and how wide it truly is. A second is whether the primary rock below the oxidized surface contains similar or lower values. A third is whether the vein persists along strike and down dip. Only after those are tested does it make sense to discuss tonnage or a mine plan. The Burro Mountains mineral chapter shows why a surface copper color can be a small secondary feature or part of a much larger system; the color alone does not choose between them.
Selected, grab, channel, and bulk samples have different jobs
A selected sample deliberately includes the best-looking material. It can identify a mineral or demonstrate a possible high-grade component. It should never be described as representative grade. A grab sample is a small piece collected without a defined equal-volume method; it may characterize a location loosely but carries selection bias. A channel sample, taken across a measured, exposed interval with a consistent method, can estimate grade for that interval more fairly. A bulk sample tests a larger volume, often to understand processing or product quality. Each requires appropriate rights, authorization, safety planning, and a record of exact location and method. The label should say which type it is.
Representative sampling is not a magic word. A channel cut through only the colorful portion of a three-foot vein is not representative of the full mining width if a miner must also break the lower-grade margins. A sample from an easy-to-reach surface outcrop may not represent the weathered or fresh rock at depth. A bulk sample of hand-sorted gem pieces may establish what fine stones look like but not how many saleable stones the whole pocket yields. Sampling design follows the proposed extraction method. It needs to include material that will dilute the product and material that may be lost in mining, not just the mineral one hopes to sell.
For a placer, test volumes and locations matter. A pan can confirm a gold grain, but a single pan from a natural trap is an intentionally favorable point. A meaningful evaluation maps the gravel unit, samples at documented positions and depths, and measures how the payable material varies. Bedrock crevices, flood layers, and older terraces may have different grades. The Pinos Altos rockhounding chapter explains the lode-to-placer relationship; it cannot promise a continuous pay streak in a modern drainage. A placer estimate needs a defined area and thickness as well as grade.
An assay certificate is a chain of custody
Every sample needs a unique identifier tied to a map and notebook entry. Record the collector, date, exact context, material type, width or volume, and whether the sample was in place or float. Photograph before and after collection where allowed. Seal and label bags so contents cannot be swapped. Keep a split when the sample type and size permit, and document how the split was made. Choose an accredited or reputable laboratory and request the analytical method appropriate to the mineral and expected grade. A result reported in parts per million is not automatically equivalent to a saleable product. Mineral form and recovery matter.
Quality control begins before the laboratory. The USGS primer on geochemical QA/QC emphasizes that contamination and analytical artifacts can begin in study design and field handling. A dirty tool reused after a rich sample can contaminate the next. A bag labeled after several similar bags are filled can be misassigned. Duplicate samples help reveal variability and preparation differences; blanks can reveal contamination; reference materials help check accuracy. A small miner may not have a corporate laboratory budget, but simple chain-of-custody discipline prevents a surprisingly large number of false conclusions. If an anomalous result would justify expensive work, verify it independently.
A lab report should be read with its detection limits and preparation notes. “Below detection” does not mean zero; a value above a method’s reliable range may need reanalysis by a different technique. A composite can hide high and low intervals, while an isolated rich chip can exaggerate a deposit’s grade. Certain minerals can cause difficult analytical or processing behavior. For gemstones, a chemical assay may be less relevant than a documented yield of cuttable material from representative rock. Choose measurements that match the product one intends to sell.
Why the average is harder than the best piece
Consider a purely illustrative vein divided into four equal-width intervals. One visually selected interval returns a high copper assay, while the other three return much lower results. If all four intervals would have to be mined together, the attractive number is only one-quarter of the width-weighted story. If the high interval is also much narrower than the rest, its contribution is smaller still. The exact arithmetic is simple; the difficult part is knowing which intervals exist, which would be mined, and whether the samples truly represent them. A miner who reports only the best number can be truthful about that sample and misleading about the deposit at the same time.
The converse can happen. A broad, modest-grade zone may be more useful than a narrow spectacular seam if it is continuous, can be accessed safely, and can be processed cheaply. Large copper mines illustrate the principle at industrial scale, but a small operation faces it too. Width, waste dilution, and recovery determine what reaches a buyer. A channel sample that includes barren margins may feel disappointing next to a cabinet piece, yet it is precisely the sample that can improve an honest mining estimate. The goal is not to make every number high. It is to make the numbers answer the correct question.
This is why a sample register should include every location selected before results arrived. If only successful samples remain in the story, the deposit will look more uniform than it is. Plot the lows and highs on the same map. Look for a geologic explanation: does grade rise near a fault intersection, change with host rock, or end at a dike? A pattern can guide the next lawful test. Randomly discarding low values because they are inconvenient destroys the very information exploration is meant to obtain.
Map continuity before multiplying numbers
Grade times volume is tempting arithmetic, but volume is a geological interpretation. Measure the exposed length and width of each mineralized interval and the orientation of the body. Note where the exposure ends because the mineral ends and where it ends because soil or talus covers it. Look for faults that offset or truncate the zone. A vein can swell into a rich pocket and pinch to nothing within a short distance. A skarn body may be irregular around a limestone contact. A placer pay streak may follow an older buried channel rather than today’s surface stream. Each geometry needs a different model.
Do not treat “shown on two outcrops” as continuous ore between them without supporting evidence. A map can show an inferred connection with a dashed line and state the distance not observed. Drilling or other subsurface work may eventually test it, but such activity needs surface authorization and can be expensive. The BLM surface-management guidance explains that work beyond casual use can require a notice or plan; New Mexico has its own exploration permit categories. A claimant should design a lawful exploration program before placing a drill, not treat the claim notice as a drilling permit.
Historical production is useful context and a dangerous multiplier. If a neighboring mine shipped ore for decades, it establishes that a deposit existed there under its own geology and economics. It does not extend grade through the untested hill into a new claim. Conversely, a historic mine may have left lower-grade material that becomes interesting under a different price or process, but that possibility has to be tested under present conditions. The Small-Time Prospecting series separates district-level history from individual mine evidence. A prospect estimate must do the same.
Recovery changes the value of a ton
An assay measures metal or mineral in the sample; a plant sells what it can recover into a marketable product. Gold locked in fine sulfides may require a different process from free gold grains in gravel. Copper oxide and copper sulfide material can demand different treatment. Turquoise may be plentiful by weight but too fractured for high-value cutting. Fluorite with impurities may not meet a buyer’s specification without sorting or processing. The intended sale product must be named before one can sensibly compare grade with cost. A small operation that cannot access affordable processing may not be economic even with a genuine mineralized body.
Metallurgical or lapidary tests should use material that represents what would actually be mined. A carefully selected attractive stone can prove a cutting method works on that stone; it cannot establish average recovery from the whole pocket. A tabletop gravity test can show that a small amount of gold reports to a concentrate; it may not predict recovery at scale, water needs, or disposal of tailings. A buyer may discount small, irregular shipments, charge for impurities, or require a minimum quantity. The marketability component of the BLM discovery standard forces the miner to ask these questions in the real world.
Water is a good example of a cost that can be invisible in the assay. Panning a few pounds of gravel in a creek does not establish a lawful or sufficient water supply for washing tons of material. Dry processing may save water but change recovery and dust control. A small flotation or leaching concept can require technical skills, permits, reagents, and residue management that overwhelm a tiny ore body. Even hand sorting generates waste and transport decisions. The mine plan should state the proposed processing route in ordinary words and identify every input and residual material. If the route is not yet known, the economic model is not complete.
The buyer is equally real. A jeweler, smelter, processor, or mineral dealer may buy products under very different terms. A turquoise pocket’s value may depend on cuttable color and stability; a copper ore buyer may require a minimum grade and penalize contaminants; a fluorspar buyer may demand a chemical specification. A price quoted for a finished retail specimen cannot be applied to unprocessed rock at the mine. Seek actual offers or published terms suited to the expected product, and record delivery, testing, and payment conditions. “Marketable” means there is a plausible route from the ground to a paying buyer after costs, not that a similar-looking item appeared online at a high asking price.
There is no shame in discovering that a mineral occurrence is scientifically interesting and commercially poor. The geologic record remains valuable. A turquoise fragment, quartz vein, or mineralized skarn can teach how the district formed. A claim’s purpose, however, is to develop a valuable deposit. The difference between a collectible specimen and a mine is not an insult to the specimen; it is an honest account of scale, continuity, and recoverability.
Calculate a first-pass project without pretending precision
A simple preliminary model lists tonnes or tons one can support from measured geometry, a grade range based on representative samples, likely recovery, product price or buyer terms, and all major cost categories. Those costs include access, excavation, labor, fuel, water, crushing or sorting, transport, assay and supervision, permits, reclamation, insurance, and administration. If any of those is unknown, mark it as unknown or a scenario. Do not use a single exact “net profit” figure built from guesses. A range with clear assumptions can guide the next test; a false exact number can attract bad decisions.
Consider a hypothetical small vein exposed for 30 feet and averaging two feet wide at the surface. That surface view does not establish depth. Multiplying 30 by two by an imagined depth and a handpicked assay produces a number with no geological confidence. A better decision tree asks what depth or continuation could be tested lawfully at a cost the project can afford. If even an optimistic, clearly labeled scenario cannot cover transport and processing, the next exploration dollar may be better spent elsewhere. If the economics look promising only at the highest selected grade, representative sampling is the immediate priority. The model should tell you where ignorance matters most.
Price changes can reverse a marginal conclusion, but the miner cannot control the market. The historic Boston Hill manganese story illustrates how an industrial ore became valuable under particular demand and transport conditions. The mineral did not appear only when the market needed it. A modern prospect can be real and uneconomic today; holding it indefinitely still incurs claim and stewardship costs. A credible plan distinguishes present marketability from a speculative future price. It should not call a deposit a reserve because of a hoped-for boom.
What “proved” should mean in a small claim file
The file should let another qualified person reconstruct the conclusion. It needs a land-status and claim record, geologic map, measured exposures, sample locations and methods, laboratory certificates, quality-control checks, deposit-geometry interpretation, processing evidence, product market, cost assumptions, and the limits of each estimate. Keep selected high-grade specimens in the record, clearly labeled as selected. Include barren and low-grade results too. A claim with mixed evidence may still warrant more exploration, but the decision should be transparent. If the project seeks financing or a sale, outside technical and legal review becomes more important, not less.
The BLM discovery rule is a legal standard that can be examined and contested; an ordinary filing does not automatically adjudicate it. Formal resource and reserve statements have their own technical reporting conventions. For a family-scale prospect, the plain-language conclusion may be more useful: “Copper mineralization is exposed over a measured interval, three independent samples show variable grade, continuity below the exposure is unknown, and processing and access costs have not been established.” That is a real result. It tells the owner what to do next and what not to promise.
When the evidence strengthens, the next question is how to work without harming land or people. The operations chapter follows a small project from permitted exploration through a mine plan, worker safety, water, waste, reclamation, and closure. The transition is substantial. A good assay is a reason to design that work carefully, not a reason to start digging tomorrow.
Let a weak result change the plan
A proof program should include decisions that are uncomfortable to make. Set a minimum grade or product quality, a minimum demonstrated width, a plausible recovery rate, and a maximum cost per unit before the next sampling campaign. Then compare the new results with those thresholds. A run of lower-grade samples may show that the original rich piece was a narrow pocket. Poor recovery may turn an impressive head assay into unsaleable material. A deposit may also be genuine but too small or too remote for this operator. Such findings are not failed science; they are the reason to test.
Keep uncertainty visible. Show the difference between observed material and inferred extensions. Use low, middle, and high price and recovery scenarios, with the assumptions written beside them. Ask an independent geologist or processor to review the evidence before borrowing heavily against it. If the project needs every optimistic assumption to be true, it has not been proved for a small miner’s budget. The best decision may be to preserve the data, reclaim test disturbance, and move on to a better target.
Source notes
- BLM, Discovery states the prudent-person and marketability tests for valuable mineral deposits.
- USGS, Mineral Deposit Models and USGS mineral resource/reserve explanation support the distinction among occurrence, geologic confidence, and economic reserve.
- USGS, Quality Assurance and Quality Control of Geochemical Data explains why study design, sample handling, and analytical checks all matter.
- BLM, Surface Management and New Mexico Mining Act Reclamation Program govern disturbing exploration beyond casual use.