Sand, Silt and Clay: The Three Soil Particles
Every mineral soil is a mix of three particle sizes, and the mix sets how a soil drains, holds water and stores nutrients.[1] Below: the size limits from boulders down to clay, how each particle behaves, how labs measure the mix, and how US topsoils mix them.
What’s the mix at your address?
Enter a US address, city or ZIP to see the sand, silt and clay USDA maps for the topsoil there, plotted on the texture triangle.
US addresses only: street address, city and state, or ZIP.
Medians of each fraction, measured separately.
How big are sand, silt and clay?
USDA sorts the mineral grains smaller than 2 mm into three groups, the separates. From largest to smallest they run sand, silt, clay, and the largest sand grain is a thousand times wider than the largest clay particle.[2] Each one feels different between your fingers, and in a jar of water each settles at its own pace.[3]
| Particle | Size range | How it feels | Settles in a jar test |
|---|---|---|---|
| Sand | 2.0–0.05 mm (2000–50 µm) | Gritty | In about a minute |
| Silt | 0.05–0.002 mm (50–2 µm) | Smooth, like flour | In about 2 hours |
| Clay | Under 0.002 mm (under 2 µm) | Sticky | Only when the water clears, usually 1 to 3 days |
Those settling times are the basis of the jar test, which you can run at home.
Soil particles from largest to smallest
Above sand, USDA sorts the pieces 2 mm or larger, the rock fragments, by diameter: the 3-inch (76 mm) limit separates gravel from cobbles, 10 inches (250 mm) separates cobbles from stones, and 24 inches (600 mm) separates stones from boulders. With the three separates below them, that makes a ladder from boulders down to clay.[2]
| Particle | Diameter | Counts toward texture? |
|---|---|---|
| Boulders | more than 24 in (600 mm) | No, a rock fragment |
| Stones | 10 to 24 in (250 to 600 mm) | No, a rock fragment |
| Cobbles | 3 to 10 in (76 to 250 mm) | No, a rock fragment |
| Gravel | 2 mm to 3 in (76 mm) | No, a rock fragment |
| Sand | 2.0 to 0.05 mm | Yes |
| Silt | 0.05 to 0.002 mm | Yes |
| Clay | under 0.002 mm | Yes |
Where rock fragments are common, USDA adds a modifier such as gravelly to the texture name; the loam page gives the thresholds.[2]
Particles and aggregates
Texture counts single particles. USDA defines soil texture as the weight proportion of the separates among particles less than 2 mm across, as a laboratory particle-size analysis measures them; pieces 2 mm or larger are rock fragments.[2] Aggregates are something else: groups of soil particles that bind to each other more strongly than to the particles around them.[4]
That’s why the jar test starts with hard shaking: it breaks the aggregates apart into single mineral particles, so each can settle by its own size.[3] How readily particles cling together into aggregates is the soil’s structure.[5]
How each one behaves
- Sandy soils drain quickly and don’t hold water and nutrients well: sand grains are the largest particles and leave large pores.[5]
- Silt particles are microscopic and feel smooth; silt makes smaller pores and holds water better than sand.[5]
- Clay soils hold a great deal of water and nutrients but drain slowly, can be short of air, and get very hard when dry.[5] “Sand and silt don’t store nutrients; they’re just rocks,” says Oregon State soils instructor James Cassidy.[1]
In USDA’s data, silt loam, silt and silty clay loam topsoils tie for the most plant-available water, a median of 2.4 inches per foot (200 mm per meter); sand holds the least, 0.84 inches per foot (70 mm per meter).[6][7] Silt loam also covers 24.2% of alluvial soil acres.[6]
How labs measure sand, silt and clay
The hydrometer method
One lab method splits a sample into sand, silt and clay by how fast each settles in water, a difference in settling rates described by Stokes’ law. Ohio State’s soil fertility lab runs it with a hydrometer, which measures the density of the soil particles still in suspension at set times. Its protocol, revised in July 2020, runs like this:[8]
- Add 25.0 mL of sodium hexametaphosphate dispersing solution to each beaker of soil.
- Read the hydrometer at 40 seconds: that reading gives the percent silt plus clay.
- Read it again at 2 hours: that gives the percent clay. Each reading is corrected against a blank and divided by the sample weight.
- Subtract percent clay from percent silt plus clay to get silt, then subtract percent silt plus clay from 100 to get sand.
Those reading times belong to that lab’s protocol, not a universal standard.[8]
The soil jar test is the home version: the same settling by size, measured with a ruler on the side of a jar.[3]
Sieve analysis and the grain-size curve
Engineers size soil grains by sieving and mechanical analysis and plot the results as a grain-size curve. The curve’s shape shows the type of gradation: on semi-log paper, poorly graded soils plot as steep curves and well-graded soils as smooth ones.[9]
Two numbers describe the curve. The coefficient of uniformity (Cu) measures the size range: Cu = D60 ÷ D10, where D60 is the size that 60 percent of the sample is finer than and D10 the size 10 percent is finer than. The coefficient of curvature (Cc) describes the curve’s shape: Cc = D30² ÷ (D60 × D10).[9]
The fines that pass the No. 200 sieve, 0.074 mm in USDA’s engineering handbook, are classed as silt or clay by plasticity, which the Atterberg limit tests measure, rather than by grain size. The handbook cites ASTM D6913 as the standard test method for particle-size distribution by sieve analysis.[9]
The two systems also draw the sand line in different places: the Unified Soil Classification System counts sand down to the No. 200 sieve, while USDA’s sand stops at 0.05 mm. The soil types page covers the engineering groups.[9]
How US topsoils mix them
Across 3,386 detailed USDA soil surveys, the sand, silt and clay of 1.9 billion classified topsoil acres fall mostly in the loam and silt-loam middle of the texture triangle.[6]
| Fraction | 10th | 25th | Median | 75th | 90th |
|---|---|---|---|---|---|
| sand | 10% | 23% | 38.6% | 65% | 78% |
| silt | 15.8% | 24% | 40% | 60% | 69% |
| clay | 5% | 8.5% | 15% | 22.5% | 31% |
Acre-weighted percentiles of each fraction, measured separately across classified topsoils.
By texture class, silt loam covers the most classified US topsoil, 33.0%, then loam (20.8%) and sandy loam (20.4%); all twelve are on the soil types page.[6]
Highest median sand
| State | Median |
|---|---|
| Florida | 95% |
| Michigan | 71% |
| South Carolina | 70% |
| Georgia | 68% |
| Delaware | 66% |
Highest median silt
| State | Median |
|---|---|
| Illinois | 70% |
| Alaska | 65% |
| Missouri | 65% |
| Iowa | 61% |
| Indiana | 59% |
Highest median clay
| State | Median |
|---|---|
| Hawaii | 31% |
| Iowa | 28% |
| Kansas | 23% |
| North Dakota | 23% |
| South Dakota | 23% |
States with at least 500 classified components.[6]
Download the data: national percentiles (CSV) and per-state medians (CSV).
Find your own soil’s mix
A jar test gives you the percentages, the ribbon test gives you a class by feel, and the texture triangle turns three numbers into one of 12 texture classes. New to the chart? See how to read it, or print a worksheet.[3][2]
Which texture class is my mix?
Here are a few common mixes, classified with the same USDA triangle the texture tool uses. Loam isn’t a particle size: it’s a texture class, a combination of sand-, silt- and clay-sized particles.[3]
| Sand | Silt | Clay | USDA texture class |
|---|---|---|---|
| 40% | 40% | 20% | Loam |
| 20% | 60% | 20% | Silt loam |
| 65% | 25% | 10% | Sandy loam |
| 85% | 10% | 5% | Loamy sand |
| 10% | 83% | 7% | Silt |
| 30% | 35% | 35% | Clay loam |
| 20% | 30% | 50% | Clay |
Frequently asked questions
- What's the difference between sand, silt and clay?
- Their size, and how that size feels. Sand is 2.0 to 0.05 mm and feels gritty, silt is 0.05 to 0.002 mm and feels smooth like flour, and clay is under 0.002 mm and feels sticky.[2][3]
- Which soil particle is the smallest?
- Clay, at under 0.002 mm across.[2]
- What are the soil particles from largest to smallest?
- Boulders, stones, cobbles and gravel, the rock fragments, then sand, silt and clay. Texture counts only those last three, the particles smaller than 2 mm.[2]
- How big is silt?
- 0.05 to 0.002 mm: smaller than sand, bigger than clay.[2]
- Is gravel part of soil texture?
- No. Gravel runs from 2 to 76 mm (3 in), and texture counts only particles smaller than 2 mm; gravel and anything larger are rock fragments.[2]
- What is the hydrometer method?
- A lab method that separates sand, silt and clay by their settling rates, as described by Stokes' law. Ohio State's soil fertility lab reads a hydrometer in a dispersed soil suspension at 40 seconds (silt plus clay) and at 2 hours (clay), then works out silt and sand by subtraction.[8]
- What is sieve analysis of soil?
- Sizing a soil's grains by sieving and mechanical analysis, with the results plotted as a grain-size curve whose shape shows how well graded the soil is. USDA's engineering handbook cites ASTM D6913 as the standard test method.[9]
- What size are loam particles?
- Loam has no particle size of its own. It's a texture class: a soil with a combination of sand-, silt- and clay-sized particles.[3]
- Which soil holds the most water?
- In USDA's data, silt loam, silt and silty clay loam topsoils tie for the most plant-available water, a median of 2.4 inches per foot (200 mm per meter); sand holds the least, 0.84 inches per foot (70 mm per meter).[6][7]
- What mix is the typical US topsoil?
- About 39% sand, 40% silt and 15% clay, taking the median of each fraction separately.[6]
How we calculate this
We read each soil component’s sand, silt and clay from its first horizon below any organic surface layer, the same surface rule as our texture data. We drop miscellaneous areas, organic soils and any component without a sand/silt/clay split, then weight every figure by each component’s share of its map unit across 3,386 detailed USDA soil surveys.[6]
The percentiles, the dominant-fraction shares, the heat grid and the per-state medians are all acre-weighted. Texture-class shares and water per foot come from our texture-class profiles of the same surveys, which leave organic soils (Histosols) unclassified. Data generated 2026-10-07.
Sources
- Oregon State University Extension, Understanding soil texture is key to better gardening (K. Pokorny, quoting J. Cassidy; 2025)
- USDA NRCS, Soil Survey Manual (Agriculture Handbook 18), Ch. 3 “Examination and Description of Soil Profiles” (Soil Science Division Staff; eds. Ditzler, Scheffe, Monger; 2017)
- Colorado State University Extension, Estimating Soil Texture (GardenNotes #214)
- USDA NRCS, Soil Quality Indicators: Aggregate Stability (June 2008)
- Colorado State University, Landscape for Life, How to determine soil type
- USDA NRCS, Soil Survey Geographic Database (SSURGO), via Soil Data Access
- USDA NRCS, National Soil Survey Handbook, Part 618, section 618.6: Available water capacity (amended August 2024)
- Ohio State University Soil Fertility Lab, Procedure for Soil Particle Size (Texture): 2-Hour Hydrometer Method (rev. July 2020)
- USDA NRCS, National Engineering Handbook, Part 631, Ch. 3 “Engineering Classification of Earth Materials” (210-631-H, 2nd ed., 2022)
Marko Visic · MPharm, BSc Physics · Last reviewed November 13, 2026