How do I measure the specific gravity of a mineral at home?

Weigh the specimen dry, weigh it again while it hangs fully under water, and divide the dry weight by the weight it lost. The answer is its specific gravity (SG): how many times heavier it is than the same volume of water [1]. Quartz comes out near 2.65, galena near 7.5 and gold near 19 [1][2]. You need a scale that reads to 0.1 gram or better, a cup of water, thread and a calculator. This page walks through the measurement, a worked example, the errors that throw the number off, and a table of values to compare against.

What the number means

Specific gravity is the ratio of a mineral's weight to the weight of an equal volume of water [1][2]. Because a gram of water takes up almost exactly one cubic centimeter, SG is in practice the same number as density in grams per cubic centimeter [2]. The Maine Geological Survey's classroom activity puts it this way: specific gravity is "nearly identical" to density [2].

Collectors sense SG before they ever measure it. Pick up a piece of barite and a piece of quartz of the same size and the barite feels wrong in the hand, too heavy. That impression is called heft, and the measurement turns it into a number you can look up.

The two sources this page uses specify slightly different water temperatures for the reference (23 °C in one, about 20 °C in the other) [1][2]. At room temperature the difference is far smaller than the error any household scale introduces, so ordinary tap water at room temperature is fine.

What you need

• A digital scale reading to 0.1 g, or better 0.01 g. Small jewelry and postal scales work for small specimens.
• A cup or beaker of water deep enough to cover the specimen without it touching the bottom or the sides.
• Fine thread or thin fishing line, or a small wire cradle, to hang the specimen.
• A pencil and a calculator, or the specific gravity calculator.

Pick a specimen that is as close to a single mineral as you can find. The Queensborough Community College gem course notes advise choosing "the purest specimen possible," because matrix, inclusions and other minerals shift the result [1].

Measuring it step by step

There are two ways to get the weight lost in water. Both give the same number.

Method 1: hang the specimen and weigh it under water

This is the method in the Queensborough course notes [1].

1. Weigh the dry specimen. Write it down as A (weight in air).
2. Tie the thread around it, or set it in the cradle.
3. Hang it from the scale (some scales have a hook underneath; otherwise rig a bridge over the pan) so the specimen sits completely under water and touches nothing but the thread. Write down the reading as W (weight in water).
4. Subtract: A minus W is the weight lost, which equals the weight of water the specimen pushed aside.
5. Divide: SG = A ÷ (A minus W).

Method 2: put the cup on the scale

This version needs no hook. It works on the same principle: the water pushes up on the specimen with a force equal to the weight of water it displaces, and the scale under the cup reads that force.

1. Weigh the dry specimen: A.
2. Put the cup of water on the scale and press tare so it reads zero.
3. Hold the specimen by its thread and lower it until it is fully under water, touching neither the bottom nor the sides. Hold it still.
4. The reading is the weight of displaced water, the same as A minus W in method 1. Call it D.
5. Divide: SG = A ÷ D.

A third option: measure the volume directly

The Maine Geological Survey activity uses a graduated cylinder instead of a second weighing: drop the weighed sample into a cylinder filled to a set level, read how much the water rises in milliliters, and divide the weight in grams by that volume [2]. Their example is a 42 g sample that raises the water 11.6 ml, giving 42 ÷ 11.6, about 3.62 [2]. This is easy with a classroom cylinder, but reading a cylinder to a tenth of a milliliter is hard, so the weighing methods are usually more precise at home.

A worked example

You have a colorless, glassy, six-sided crystal and you want to know whether it is quartz or something heavier.

A result of 2.65 matches quartz [2]. If the same crystal had come out at 3.5 or higher, quartz would be ruled out and you would look at topaz, garnet or other heavier species.

The Queensborough notes give a gold example: a 50 g specimen weighs 47.368 g in water, loses 2.632 g, and has an SG of about 19 [1]. A dense mineral loses very little weight in water, which is exactly why small errors in that little number matter so much.

What throws the number off

To see how scale resolution matters, take a 10 g piece of quartz. It should lose about 3.8 g in water. A scale that reads only to 0.1 g could show 3.7 or 3.9, and the SG would come out anywhere from about 2.56 to 2.70. On a 100 g piece the same rounding barely moves the answer. When the result matters, weigh twice and average.

Minerals you should not put in water

A few minerals dissolve, swell or crack in water, among them halite and other salts, some borates and some clays. Do not run this test on them. Check minerals that must not get wet before you dunk anything you have not identified, and dry every specimen fully before it goes back in its box with its label.

Typical values to compare against

These values come from the Maine Geological Survey, the Queensborough Community College course notes and the Mineralogical Society of America's identification key. Ranges show the spread each source gives.

Cinnabar is a mercury mineral and galena a lead mineral. Handle them as toxic minerals, wash your hands, and do not leave the water where anyone will drink it.

Where SG settles an identification, and where it does not

Specific gravity is at its best separating minerals that look alike but differ greatly in weight: barite (4.5) from calcite (2.71) or celestite from calcite, and gold (about 19) from pyrite and other brassy minerals [1][4]. It is weaker between species with close values, such as quartz (2.65) and calcite (2.71), where hardness and the way the mineral breaks do the job faster; those pairs are worked through on common lookalikes. On a mixed specimen, the result tells you little about any one mineral in it.

Running the numbers with the calculator

If you would rather not divide by hand, the specific gravity calculator takes your weight in air and weight in water and lists common minerals in that range. Put its answer together with hardness, streak and locality on identifying a mineral, and if the tests still disagree, getting a specimen tested explains what a lab can tell you and what it costs.

References

1. Queensborough Community College (CUNY), Roland Scal, Gems and Semiprecious Stones course notes, "Specific Gravity" (undated, read October 2026). https://www.qcc.cuny.edu/biologicalSciences/faculty/rscal/gems/specgrav.html
2. Maine Geological Survey, "Activity 19: The Density of Minerals," developed by Duane Leavitt (undated, read October 2026). https://www.maine.gov/dacf/mgs/education/lessons/act19.pdf
3. Mineralogical Society of America, The Mineral Identification Key, Table IIB-1: Streak not colored, cleavage prominent, hardness less than 2½ (undated, read October 2026). https://msaweb.org/msa/Collectors_Corner/id/mineral_id_keytiib1.htm
4. Mineralogical Society of America, The Mineral Identification Key, Table IIC-1: Streak not colored, cleavage prominent, hardness 2½ to 3 (undated, read October 2026). https://msaweb.org/msa/Collectors_Corner/id/mineral_id_keytiic1.htm
5. Mineralogical Society of America, The Mineral Identification Key, Table IIA: Minerals with a nonmetallic luster, definitely colored streak, and hardness 1 to 6 (undated, read October 2026). https://msaweb.org/msa/Collectors_Corner/id/mineral_id_keytiia.htm