Berlese Funnel Experiment Kids Can Build Safely

Scoop a double handful of leaf litter from under a hedge. It looks like dead leaves. Put it in a Berlese funnel for a day and a half and somewhere between twenty and three hundred animals will walk out of it and fall into a jar: springtails, mites, tiny beetles, the occasional pseudoscorpion. This Berlese funnel experiment kids can build from a plastic bottle makes a community visible that is genuinely there and genuinely uncountable by eye. Two things make this version different from most: the lamp is chosen for safety rather than tradition, and the collecting jar holds damp paper instead of alcohol, so everything you catch goes home alive.

Key Takeaways

• You will build a funnel from a cut bottle, load it with leaf litter, dry it gently from above for 24 to 48 hours, and count what collects in the jar below.
• The concept is soil mesofauna: a dense invertebrate community lives in litter and topsoil, and it is extracted by exploiting the animals’ avoidance of light and dryness.
• Best for ages 8 to 15. Younger children can load and watch; sorting mites from springtails under a lens needs patience and a steady hand.
• The usual failure is litter that is too wet or packed too tight. Nothing moves down and you conclude, wrongly, that there was nothing there.
• The heat source is the one real hazard. Use an LED bulb, or an incandescent bulb of 25 W or less at 25 cm, on a stable lamp, never unattended overnight.

What You’ll Need

Item	How much	Where to get it	If you don’t have it
Clear plastic bottle, 2 litre	1, cut in half	Recycling	A kitchen funnel of 10 cm or wider
Mesh for the funnel neck	A 10 cm square, holes 3–6 mm	Plastic fruit netting, a scrap of window screen, a tea strainer	Crossed sticks or a loose tangle of twigs
Glass jar, 250–500 mL	1	Kitchen	A yoghurt pot
Kitchen paper	2 sheets, dampened	Kitchen	A scrap of damp cloth
LED desk or clamp lamp	1, with a 5–10 W LED bulb	Home	A 25 W or lower incandescent, at 25 cm, attended only
Leaf litter	Two double handfuls, 1 litre loose	Under a hedge, a woodland floor, a leaf pile	—
Hand lens, 10×, or phone macro	1	Toy box, hardware shop	A magnifying glass and a bright side light
White tile or paper, fine paintbrush, teaspoon	1 each	Kitchen and art box	—
Gloves and a bucket for the litter	1 each	Shed	Wash hands thoroughly instead

Time: build 20 min, collect 15 min, run 24–48 h, count and release 40 min Cost: free if you already have a lamp Mess level: do the loading outside

Do the Experiment

Step 1 — Build the funnel and the trap

Cut the bottle in half and keep the top as your funnel. Lay the mesh inside, resting at the shoulder where the bottle narrows, so litter sits on it and animals can fall through the holes. Stand the funnel neck-down in the mouth of the jar so it is held firmly and nothing can escape round the edge. Put two sheets of damp, not wet, kitchen paper in the bottom of the jar.

Step 2 — Collect litter without flattening it

Take the top 3 to 5 cm of leaf litter and the crumbly layer just below it from a shaded, damp spot. Lift it with a trowel into a bucket rather than squeezing it into a bag, because compacting the litter kills the animals and blocks the passages they use to move down. Note where you collected, the date and whether the litter was damp or dry.

Step 3 — Load loosely and position the lamp properly

Tip the litter into the funnel to a depth of about 8 cm without pressing it down. Then set the lamp so the bulb is at least 25 cm above the litter surface, angled down, on a base that cannot be knocked over, with the cable routed where nobody will trip on it. An LED bulb is the right choice: it gives the light without the heat, and the drying it causes is enough to drive the animals down. If you only have incandescent, use 25 W or less and treat the setup as something you do not leave.

Step 4 — Run it for 24 to 48 hours, with the lamp attended

Leave the funnel somewhere undisturbed and run the lamp only while somebody is at home and awake. Switch it off overnight; the extraction continues more slowly because the litter keeps drying, and this single rule removes the fire risk entirely. Check the jar after a few hours out of interest, and properly at 24 and 48 hours.

Step 5 — Count, identify to group, then release everything

Tip the jar contents onto a white tile and work through them with the hand lens and a damp paintbrush, which lifts tiny animals without crushing them. Count by group rather than species: springtails, mites, beetles, fly larvae, millipedes, centipedes, woodlice, spiders, ants, pseudoscorpions, worms, other. Photograph anything striking. Then carry the tile, the jar paper and the dried litter back to exactly where you collected them and tip it all out in the shade.

What you should see: Movement in the jar within the first two hours, usually springtails first because they are fast. By 24 hours a good litter sample from a damp shaded spot yields 30 to 200 animals, and mites plus springtails are nearly always the two most numerous groups by a wide margin. Larger animals like woodlice and millipedes often refuse to go down and will still be sitting in the litter at the end, which is normal and worth recording separately. A dry lawn-edge sample may give you five animals. The contrast between two habitats is the result, not the raw number from one.

What’s Actually Going On

The funnel is an extractor, not a trap. It makes the top of the litter steadily less habitable while leaving a dark, damp refuge below, and the animals walk there themselves. Two behaviours do the work. Most soil invertebrates are negatively phototactic, moving away from light, which is adaptive for animals whose predators hunt by sight. And nearly all are acutely sensitive to humidity, because a springtail a millimetre long has a huge surface area relative to its volume and desiccates in minutes. As the lamp dries the upper litter the gradient steepens, the animals move down through the spaces between leaf fragments, and the mesh lets them fall into the jar.

This matters because it explains the method’s biases, which is more useful than the species list. Fast, small, soft-bodied animals that can walk down through litter are collected efficiently. Slow ones, large ones, ones that burrow sideways rather than down, ones already dormant, and anything that happens to be in an egg or a pupa are missed entirely. So a Berlese sample is not a census of the litter; it is a sample of the mobile mesofauna. Antonio Berlese’s original 1905 design used a water jacket for heat and Albert Tullgren’s later modification swapped in an electric bulb, which is the version almost everyone builds now. Both share the same bias.

What you are sampling is the engine room of decomposition. Penn State Extension’s soil summary puts one teaspoon of agricultural soil at a hundred million to a billion bacteria, hundreds of nematodes, up to a hundred small soil insects and five or more earthworms, with soil life weighing 2,000 to 30,000 pounds per acre. The animals you catch sit in the middle of that: they shred leaf fragments into smaller pieces, multiplying the surface area available to bacteria and fungi, and they eat each other. That shredding is why leaf litter disappears in months rather than years.

One honest limitation about the lamp. The traditional instruction is a 40 or 60 W incandescent bulb close above the funnel, and it does extract faster. It also puts a hot filament a few centimetres above a pile of dry leaves, overnight, in a house. That is a genuinely poor idea and the extra animals are not worth it. An LED at 25 cm extracts more slowly and reaches a similar total given an extra day.

How to Explain It to Your Kid

Ages 5–8: The leaves are a town

Tell them dead leaves are not just rubbish on the ground: they are a town full of animals too small to notice, and the lamp is like the sun getting too bright and too dry, so everybody walks downstairs to the cool basement. Let them load the funnel and check the jar. When the first springtail jumps, that is the whole lesson delivered in one second.

Ages 9–12: Counting by group, and per litre

Introduce mesofauna and the idea of standardising. Measure the litter by volume before loading, then report animals per litre so two samples can be compared. Have them build a tally by group and a bar chart. Then ask the better question: which habitat should hold more, and why? A damp hedge bottom usually beats a mown lawn by a factor of ten, and predicting that first makes the result feel earned.

Ages 13+: Extraction efficiency and what you missed

Give them the bias problem properly. Ask them to list the animals the method cannot catch, then design a check: hand-sort a measured sub-sample of the same litter under a lens and compare it with what the funnel delivered. The difference is extraction efficiency, and realising every method has one is a genuine step into field biology. Then add the comparison that makes it an investigation: three habitats, three replicates each, animals per litre, with litter moisture noted.

The question to ask: “We got two hundred animals from the hedge litter and six from the lawn. Does that mean the lawn has almost nothing living in it, or could it mean our funnel is bad at catching whatever lives in a lawn?”

Why Teaching This Early Pays Off

The transferable idea is sampling bias, and this is one of the cleanest demonstrations available in a kitchen. Every method prefers some things and misses others, and a result is only interpretable once you know which. Children meet this later in surveys, polls, medical screening and search results, almost always without being told. A funnel that catches springtails brilliantly and woodlice hardly at all teaches it physically.

The concept-specific payoff is a repaired mental model of soil. Most children, and plenty of adults, treat soil as inert brown stuff that plants happen to grow in. It is in fact the densest habitat most of us will ever stand on, and leaf litter is its surface layer. Once a child has counted two hundred animals out of a handful of leaves, decomposition, food webs, nutrient cycling and the reason gardeners care about soil all have somewhere to attach. The National Research Council’s framework names “analysing and interpreting data” and “constructing explanations” among its core practices, and animals-per-litre across habitats with a bias caveat is both.

A note on guidance rather than exploration. Kirschner, Sweller and Clark’s argument against minimal guidance applies squarely here: a child left to improvise this build with whatever lamp is nearest will produce either nothing or a hazard. The procedure matters, and so does the parent standing next to it.

Berlese Funnel Experiment Kids Can Run: What Parents Should Know

Best age	8 to 15. Below 8, loading and watching works but the counting and the lens work do not hold attention
Supervision	An adult positions and switches the lamp, cuts the bottle, and runs the release. The lamp is never operated by a child
Real hazards	Fire. An incandescent or halogen bulb close above dry leaves is a genuine ignition risk. Use an LED, or an incandescent of 25 W or less at a minimum of 25 cm, on a stable base, with the cable clear of walkways, and never leave the lamp on unattended or overnight. Also: sharp edges on the cut bottle, which an adult should trim and tape; soil bacteria and animal faeces in litter, so gloves and thorough handwashing; and ticks in leaf litter in many regions, so check limbs afterwards. A few soil animals bite or pinch if handled, which is why you use a brush
Don’t	Do not use alcohol, antifreeze, soapy water or any preservative in the collecting jar; damp paper keeps the animals alive for release. Do not use a halogen work light, a heat lamp or a hair dryer as the heat source. Do not take litter from a protected site, a neighbour’s garden, or anywhere a dog uses. Do not keep the animals: release everything within 48 hours, where it came from

When It Doesn’t Work

• Almost nothing comes down. Usually litter that is too wet, so no drying gradient forms, or packed so tightly that there are no passages. Collect from a damp but not sodden spot, tip it in loosely, and give it a full 48 hours.
• Everything is dead in the jar. The jar paper dried out or the sample overheated. Re-dampen the paper, raise the lamp, and switch to an LED. A collecting jar that dries out defeats the point of a release-based version.
• You cannot tell a mite from a springtail. Springtails jump and most have a visible furcula tucked under the abdomen; mites have eight legs and walk steadily. At 10× that is enough. Use an “unsure” column rather than guessing.
• Only big animals, sitting in the litter, refusing to move. Woodlice, millipedes and snails frequently just stay put. Record them as a separate hand-sorted count, which is a legitimate method and not a failure.
• The second sample gives a wildly different number. Litter fauna is patchy at the scale of centimetres. Three replicates per habitat is the minimum for a defensible comparison, and reporting a range beats a mean from one sample.

Take It Further

• Three habitats, three replicates. Hedge bottom, mown lawn edge, and a compost heap surface, one litre each, same funnel, same duration. Animals per litre by habitat is a clean comparative result and a strong science-fair entry.
• Depth as the variable. Same spot, three samples: surface leaves, the partly broken-down layer below, and the top centimetre of soil. The group composition shifts markedly with depth, and that shift is a food-web story rather than just a count.
• Photograph and submit. Clear macro photos of litter invertebrates are genuinely useful to iNaturalist, where an observation reaches Research Grade with an accurate date, coordinates, a photo, a wild organism, and more than two thirds of identifiers agreeing at species level. Pair it with the compost jar experiment, which shows the microbial half of the same process, and with the pill bug choice chamber, which tests the humidity preference this funnel exploits.

What to Watch For Over the Next 3 Months

• Week 4: They turn over a log or a flowerpot and look underneath on purpose. Expecting animals in the dark damp places is the model working.
• Month 2: The sampling-bias idea surfaces somewhere unrelated, often as scepticism about a count or a survey. Encourage it, and ask what the method would have missed.
• Month 3: The follow-up worth chasing is a seasonal repeat. The same hedge sampled in spring and autumn gives different numbers and different groups, and that is a real finding. It links naturally to burying five things for sixty days, since these are the animals doing some of that work.

Frequently Asked Questions

Do we have to use a hot bulb?

No, and you should not. An LED bulb provides the light the animals avoid, and the gentle drying from above is what actually drives them down. It is slower, so give it 48 hours instead of 24. A hot incandescent bulb above dry leaves overnight is the one genuinely dangerous version of this activity.

Should we put alcohol in the collecting jar?

Research versions do, because it kills and preserves the specimens for identification. For a family project, damp kitchen paper keeps the animals alive so they can be counted and released, which is both kinder and a better lesson. You lose some identification precision and that is a fair trade.

Is leaf litter safe for children to handle?

With gloves and thorough handwashing, generally yes. Avoid anywhere dogs or cats use, do not let a child put hands near their face, and check for ticks afterwards if you are in a region that has them. A few soil animals can pinch, so move them with a damp brush rather than fingers.

Why do we have to release them?

Because you collected a functioning community from a working habitat, and it belongs there. Returning the litter and the animals within 48 hours, to the exact spot, costs you nothing and keeps the project honest. It also models the norm that real field biologists work to.

---

About the author

Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.

---

Sources

1. Penn State Extension. “Soil Quality Information.” https://extension.psu.edu/soil-quality-information
2. U.S. Environmental Protection Agency. “Types of Composting and Understanding the Process.” https://www.epa.gov/sustainable-management-food/types-composting-and-understanding-process
3. iNaturalist. “What is the Data Quality Assessment and how do I use it?” iNaturalist Help. https://help.inaturalist.org/en/support/solutions/articles/151000169936-what-is-the-data-quality-assessment-and-how-do-i-use-it-
4. Kirschner, P. A., Sweller, J., & Clark, R. E. (2006). “Why Minimal Guidance During Instruction Does Not Work.” Educational Psychologist, 41(2), 75-86. https://doi.org/10.1207/s15326985ep4102_1
5. Chi, M. T. H., & Wylie, R. (2014). “The ICAP Framework: Linking Cognitive Engagement to Active Learning Outcomes.” Educational Psychologist, 49(4), 219-243. https://doi.org/10.1080/00461520.2014.965823
6. National Research Council. (2012). A Framework for K-12 Science Education: Practices, Crosscutting Concepts, and Core Ideas. National Academies Press. https://doi.org/10.17226/13165