The streetlight problem
Everybody looked in the flask, because the flask was where the light was.
Open almost any aquarium book. It will say that Nitrosomonas turns ammonia into nitrite and Nitrobacter turns nitrite into nitrate. Nearly everyone in the hobby has learned that sentence. The second half of it is wrong, and it has been known to be wrong since January 1998.2
Here is why everybody got it wrong, and the lesson goes far beyond fish tanks. Before molecular methods, the only way to study a bacterium was to grow it. Take a scrape of filter slime, put it in a flask of nutrient broth, wait, and study whatever appears.
But a flask is not a neutral place. It is rich in food, well stirred, warm, and free of competitors. Nitrobacter loves that. It is a fast grower that thrives when nitrite is plentiful, and in a flask it wins in days. Nitrospira hates it. Nitrospira is a slow grower adapted to a world where nitrite is scarce, which is exactly what a healthy mature filter is. For decades it barely showed up in laboratory cultures at all.
So the organism that grows in a culture is not the organism that dominates in a real filter. Culturing does not show you the community. It shows you the members of the community who happen to enjoy flasks. Nitrobacter went into every textbook and stayed there for fifty years.
What the 1998 study did. Four things. It read the DNA barcodes of everything living in biofilm from several freshwater aquaria, with no growing and no flasks. It watched the same filters again and again as a new tank matured. It counted, using probes designed to stick only to one target group. And it tested a commercial bacteria supplement of the day, one that contained Nitrobacter.2
Three papers, all free to read
- 1996
- Hovanec and DeLong. The survey. Freshwater and saltwater filters hold different bacteria1
- 1998
- Hovanec, Taylor, Blakis and DeLong. The famous one. Nitrospira, not Nitrobacter2
- 2001
- Burrell, Phalen and Hovanec. The other half. Which bacteria eat the ammonia3
- 3
- peer reviewed papers in total. Everything else is a poster, a patent or trade press1234
What the 1998 paper found2
- 01
- No Nitrobacter. Not a little. Nitrite oxidisers of that group were not detected in the samples
- 02
- Nitrospira-like RNA was nearly 5 percent of all RNA extracted from the biofilm2
- 03
- The barcode appeared exactly when the tank started converting nitrite to nitrate
- 04
- The aquaria dosed with the commercial Nitrobacter product grew no Nitrobacter. They grew Nitrospira, like the tanks that got nothing
Was he first?
Yes, for aquaria, and 1998 was the year the whole field turned over at once.
He did not discover Nitrospira. The best studied member of the genus was described and named in 1995 by a German group.5 What he did was show that Nitrospira and not Nitrobacter does the nitrite work in aquaria. On that narrow claim the record supports him: received 4 September 1997, accepted 27 October, published in the January 1998 issue2.
Then, within eight months, three other groups reached the same conclusion in a laboratory bioreactor, in municipal activated sludge, and in a fluidised bed reactor.6 Four teams, four kinds of water system, overlapping but independent methods, one conclusion in twelve months. That convergence makes the finding stronger, not weaker. It is about as solid as microbial ecology gets.
Disclosed at the time
He was already working for the aquarium industry when he did the research.
He joined Aquaria Inc., the parent of Marineland, as Chief Science Officer in 1990, and finished his doctorate in 1998.4 Those overlap. This is not hidden. The 1998 paper lists two addresses for him, the university department and the company, and gives the company as his contact address. The funding note says the work was supported in part by National Science Foundation grants to his supervisor and by assistance from Aquaria Inc. to him.2
So the conflict of interest was declared, in the paper, in 1998. That is the correct way to handle it. You can find the declaration in the paper.
1998 to now
Confirmed, and then repeatedly overtaken.
That is the normal life cycle of a good scientific finding, and it is not a criticism of him. Read it alongside the newer research below for a fuller picture.
| Year | Finding | What it changed |
|---|---|---|
| 19982 | Nitrospira, not Nitrobacter, oxidises nitrite in freshwater aquaria | Still true. Confirmed independently within months |
| 20117 | Ammonia oxidising archaea were the more numerous ammonia oxidiser in 23 of 27 freshwater filters. In 12 of those, archaea supplied every detectable ammonia oxidising gene | Often it is not bacteria at all. And the less ammonia in the water, the more the archaea dominate, so a mature lightly stocked planted tank is more likely to be run by archaea than by the bacterium in your book |
| 20148 | Archaea outnumbered bacteria in every one of six freshwater tanks over four months. In the marine tank the reverse, by three to five orders of magnitude | The freshwater and saltwater split first spotted in 1996 is real and it is deep |
| 2015910 | Two teams described organisms that do complete nitrification alone, ammonia straight through to nitrate. Called comammox | The two organism model, standing since the nineteenth century, is wrong. And the joke is that the comammox organisms are themselves Nitrospira |
| 201711 | A stable nitrifying core of ammonia oxidising archaea plus comammox Nitrospira in a freshwater fish farm biofilter | First sighting of the new picture in a filter that looks like ours |
| 202412 | Across 38 freshwater aquarium biofilter samples, comammox Nitrospira genes were in all 38 and dominant in 30. Archaea present in 35, dominant in 7. Classic ammonia oxidising bacteria were rare | Comammox is the normal case in a home aquarium. In saltwater filters, no comammox was detected at all |
| 202513 | Three home aquaria watched from scratch for twelve weeks. Ammonia and nitrite undetectable by week 3 in two tanks and week 8 in the third. By week 12, archaeal and comammox genes confirmed in all three | How the community actually assembles, in a normal household tank |
Where that leaves the 1998 paper. Correct on its own question, and superseded as a complete picture. Nitrospira rather than Nitrobacter for nitrite in freshwater aquaria still stands, and is now understood to be even more central than he thought, because Nitrospira also turns out to run the ammonia step in many tanks. Two organisms, one per step, is wrong. Bacteria doing the ammonia step is often wrong in fresh water. Different communities in fresh and salt water is confirmed and strengthened.
He was right that the textbook name was wrong. He was right about which organism replaced it. He was, inevitably, not right about how complicated the answer would turn out to be. Nobody was.
Stated plainly
From laboratory research to aquarium products
Find something out, patent it, build a product. It is one of the most ordinary paths in applied science, and most of the useful things in a fish room got there that way. What matters is knowing which claims rest on refereed work and which rest on advertising.
| Date | Event |
|---|---|
| 1990 | Joins Aquaria Inc., parent of Marineland, as Chief Science Officer |
| Sep 1997 | The Nitrospira paper is submitted |
| Oct 1997 | Paper accepted |
| 22 Dec 1997 | Earliest priority date of the first nitrite oxidiser patent family |
| Jan 1998 | Paper published. Doctorate awarded the same year |
| 2001 | Ammonia oxidiser paper published. First three US patents granted |
| c. 2003 | Marineland launches a refrigerated live nitrifying bacteria product |
| 2007 | Leaves Aquaria Inc. and founds his own company |
The patent priority date is eight weeks after the paper was accepted and about three weeks before it appeared in print. That is entirely normal and legal practice, because publishing first destroys your own novelty. The commercial application was planned alongside the research. Thirteen granted patents exist, all assigned to the company or its successors rather than to him personally or to the university, and he is the named inventor.4
Verified
What the company says that checks out.
That there is a real Dr Tim with a doctorate from UC Santa Barbara in ecology, evolution and marine biology, and that his dissertation was on nitrifying bacteria in aquaria and Mono Lake. True. That he was Chief Science Officer of Aquaria Inc. for seventeen years. True. That his research led to patents. True, thirteen of them.4 That he was first to demonstrate Nitrospira as the active nitrite oxidiser in aquaria. True for aquaria, with the boundary drawn above.
Marketing, not science
Where it stops being checkable.
Any claim that a specific bottle will cycle a specific tank in a specific number of days. No independent peer reviewed trial of the product exists that we could find. Any claim of outperforming a competitor, because the comparison work is conference material. And any implication that the 1998 paper endorses the current products. It cannot. The paper predates them, and the only product it tested was somebody else's, which failed.
The biggest gap is the saltwater side, which rests on a 2004 conference poster and a patent4. There is no saltwater equivalent of the 1998 or 2001 papers. A poster is not checked by anybody.
The fair summary
A stronger evidence base than almost any competitor, and still twenty five years old.
The organism in his freshwater ammonia product is a real organism that his own refereed work identified as effective at starting nitrification in new tanks. That is more than most bottled bacteria can say. It is also three papers from 1996 to 2001 in a field that has been rewritten twice since, by the archaea in 2011 and by comammox in 2015. And no bottled product yet supplies the archaea or the comammox Nitrospira that current work says actually dominate mature freshwater filters.
The distinction the hobby collapses
Nitrifying bacteria do not eat waste. They cannot.
They are chemoautotrophs. Read the word in two halves. Chemo means they take their energy from a chemical reaction: they burn ammonia to make nitrite, or nitrite to make nitrate, using oxygen. Autotroph means they build their bodies out of carbon dioxide dissolved in the water, exactly as a plant does. So a nitrifier needs three things: ammonia or nitrite, oxygen, and carbon dioxide. It needs no food in the ordinary sense. Put one in a bowl of fish flakes and nothing happens. It cannot digest them.
That is why a bottle of nitrifying bacteria will never clear a dirty substrate, and it is the single most important fact in this topic.
Waste bacteria are the opposite in every respect. They are heterotrophs, and hetero means other: they take both their energy and their body carbon from organic matter that something else made. Leftover food, faeces, dead leaves, dying roots, the brown mulm in the gravel. They eat it, grow on it, and release the leftovers, and one of the leftovers is ammonium.
So the two groups are a supply chain, not a competition. The heterotrophs make the ammonium. The nitrifiers eat it. They compete only for oxygen and space, and that competition only becomes serious when there is a lot of organic matter around.
| Nitrifiers | Heterotrophs, the waste and sludge bacteria | |
|---|---|---|
| Energy source | Oxidising ammonia or nitrite | Oxidising organic matter |
| Carbon source | Dissolved CO2 and bicarbonate | The same organic matter |
| What they consume | Ammonia, nitrite, oxygen, CO2, alkalinity | Proteins, fats, starches, cellulose, oxygen |
| What they produce | Nitrite, nitrate, acid, very little new cell material | CO2, ammonium, phosphate, a great deal of new cell material |
| Growth speed | Slow. Days to weeks to establish | Fast. Hours |
| Biomass yield | about 0.15 g of cells per g of nitrogen oxidised14 | about 0.63 to 0.67 g of cells per g of substrate consumed15 |
| Maximum growth rate | well under 1 per day for nitrite oxidisers14 | 3 to 6 per day at 20 C15 |
| Where they live | Thin, deep, slow turnover biofilm on filter media, substrate grains, glass, plant leaves | Everywhere organic matter is. Mulm, detritus, the outer skin of every biofilm, the water column |
| Diversity | A handful of narrow specialist groups | Thousands of species, most uncharacterised |
| Do they clear sludge? | No. Physically impossible for them | Yes. This is literally their job |
Two numbers to remember. A nitrifier turns a gram of nitrogen into roughly a seventh of a gram of new bacteria1426. A heterotroph turns most of a gram of food into new bacteria15. That single asymmetry explains almost everything downstream: why nitrifiers take weeks to build a population and heterotrophs take a day, why a filter clogs with slime and not with nitrifiers, and why nitrifiers are always the population you lose first.
And the step the hobby diagram leaves out. The usual picture starts at ammonia. It never says where the ammonia in a piece of rotting food comes from. Most of the nitrogen entering a tank does not arrive as ammonia at all. It arrives as protein, in fish food, and leaves the fish partly as ammonia across the gills and partly as undigested protein in faeces. Heterotrophs release enzymes that cut that protein into peptides and amino acids, take those up, use the carbon for energy, and dump the surplus nitrogen out of the cell as ammonium, because fish food carries far more nitrogen than their own cells need. That ammonium is precisely and only what a nitrifier can eat. The heterotroph is not competing for the nitrifier's food. It is manufacturing it.
Where they genuinely do compete
Position in the biofilm, and it is measured.
Two nitrifying biofilm reactors were given acetate, an easily eaten carbon source. Nitrification fell. Imaging showed why: a heterotrophic layer formed on top of the nitrifying biofilm and cut off its oxygen supply.16 The nitrifiers were not eaten, not poisoned and not starved of ammonium. They were buried.
An aquaculture biofilter study put numbers on it. Feeding 2 mg/L of total ammonia nitrogen, raising the carbon to nitrogen ratio to 0.5 cut ammonia removal by 30 percent. At higher ratios nitrification efficiency fell by 50 percent while the number of attached bacteria doubled.17 Twice as many bacteria on the media, half as much nitrification. That is an overfed tank in one sentence.
Folklore
"My cycle crashed because I overfed and the ammonia spike killed the bacteria."
Almost certainly not. Ammonia at those levels does not kill nitrifiers; they eat it. What happens is the sequence above. The extra food is carbon, the carbon feeds heterotrophs, the heterotrophs multiply within hours and blanket the media, and the nitrifiers lose their oxygen. Nitrification falls by 30 to 50 percent17, ammonia and nitrite rise, and the hobbyist concludes the bacteria died.
The practical consequence is different too. If they had died you would need to cycle again. Because they were smothered, removing the organic load usually restores the rate within days. Vacuum the mulm, rinse the filter in tank water, stop feeding for two days.
Folklore
"A bacterial bloom means the tank is cycling."
A white or milky cloud in the water is a bloom of free living heterotrophs, not nitrifiers. Nitrifiers are attached organisms in far smaller numbers and would never make water visibly cloudy. A bloom means one thing: a sudden supply of dissolved organic carbon. A new tank, new soil, a dead snail, a big feed. It is a heterotroph event and it says nothing directly about whether nitrification is established.
How outnumbered
At least 6 in 7 of the organisms in your filter are not ammonia oxidisers, and on the filter sponges it is closer to 24 in 25.26
A 2025 study followed three ordinary home aquariums for twelve weeks. By week 12, ammonia oxidisers made up 1.6, 3.1 and 3.8 percent of the community on the biofilter sponges. On the beads the range was 0.6 to 14.1 percent.13 Everything else was overwhelmingly heterotrophic, and the most common genera were waste processing organisms.
The same paper notes that heterotrophs help nitrifiers establish, by forming the initial biofilm that nitrifiers then colonise.1318 Heterotrophs build the house. Nitrifiers move in. Only when the heterotrophs keep building without limit does the house get buried.
Products
An enzyme is a molecule. It cannot multiply, and dissolving sludge is not the same as exporting it.
Many products sold as sludge removers or waste digesters are not organisms at all. Some are enzyme blends, some are enzymes plus bacteria, and some are bacteria with an enzyme sounding name. The three behave differently, so it is worth being precise.
| Enzymes in a bottle | Live bacteria in a bottle | |
|---|---|---|
| Alive? | No | Yes, or spores that become alive |
| Can it multiply? | No. You get exactly what you poured in | Yes, if conditions suit it |
| How long does it last? | Short. It is protein, so other bacteria eat it | As long as it has food |
| What does it do? | Breaks solid organics into dissolved organics | Breaks solids down and then consumes the products |
| Does it remove anything? | No. Hydrolysis turns solid waste into dissolved waste. The nitrogen and phosphorus are still in the tank | Also no in the strict sense, but the atoms end up in cells that can be filtered or skimmed out |
| Risk | Low | Oxygen demand if overdosed |
The decisive row
A bacterium that secretes protease is a self replenishing protease factory.
A bottle of protease is a fixed number of molecules that will be degraded within days. If both work, the bacterium is strictly better value, because it makes more of itself and more enzyme. That is why the serious products in this category are bacterial rather than enzymatic.
The row nobody mentions
An enzyme that makes detritus invisible has not removed it.
It has turned it into dissolved organic carbon and fed the water column heterotrophs, which will consume oxygen and release the nitrogen and phosphorus back as ammonium and phosphate. If your problem was that the substrate looks dirty, that is a solution. If your problem was that nitrate and phosphate are too high, it is not.
Verdict
Mechanism plausible, and untested in an aquarium.
Enzymatic hydrolysis of sludge is a genuine, active, well funded research field in wastewater engineering, so the biochemistry works.19 But that work doses enzymes into concentrated sludge, in a closed reactor, at controlled temperature and pH, to feed a downstream digester. An aquarium is dilute, open, cool, and has nowhere for the products to go.
A literature search for enzyme dosing into aquarium or pond water returns essentially nothing usable. The substantial aquaculture enzyme literature that does exist is about feed additives, added to fish food to improve digestion, not added to water to clean it.20 It is not that studies were done and found nothing. It is that the studies do not appear to exist, and anyone claiming either that these products work or that they do not is going beyond the published evidence.
Naming trap
A product name containing "zyme" does not mean the product contains enzymes.
One well known product with an enzyme flavoured name is described by its own manufacturer as containing over 300 million live bacteria per teaspoonful. Another is described as containing live freshwater specific strains of nitrifying bacteria, and names no genus or species at all.25 Read the ingredient description, not the brand.
Verdict
The soil carbon dioxide claim is partly true, and the true part is real science stretched past where the evidence goes.
The claim under test is this. Bacteria in the soil produce CO2, plant roots absorb it and use it to build tissue, and that is an alternative to taking carbon in through the leaves. It circulates in the dirted tank community and it is used to justify not injecting, so it deserves a verdict rather than a survey.
| Question | Verdict | Confidence |
|---|---|---|
| Do soil bacteria produce CO2 in a submerged aquarium substrate? | True | High |
| Does CO2 build up in the pore water above water column levels? | True, and by a large factor | High |
| Can roots take up dissolved CO2 from the substrate at all? | True for free CO2, and in some plants it is the main carbon route. Almost certainly not for bicarbonate | High |
| How much does it contribute for a typical aquarium plant? | Mostly overstated. Large for rosette plants with big roots in very soft carbon poor water. Small to negligible for stem plants, and for any tank with elevated water CO2 | Medium-high |
| Is there a better explanation for why dirted tanks grow plants well? | Yes, largely. Nutrition, iron and cation exchange explain more of the observation than root CO2 does | Medium |
Parts one and two are not in doubt. Oxygen in a flooded soil is consumed within hours to a day, and microbes then work through nitrate, manganese, iron, sulfate and finally CO2 itself.21 Every step in that sequence still releases CO2. Even full methanogenesis is not a CO2 sink on balance: complete methanogenic breakdown of cellulose yields equal amounts of CO2 and methane.21 So the half worry that an anoxic substrate makes methane instead of CO2 is wrong. It makes both, roughly one for one.
And because the gas is produced inside a water filled porous solid, it cannot escape quickly, so it builds up. Measured pore water free CO2 in lake sediments runs commonly 10 to 250 times the air equilibrium value23, and in flooded paddy soils it is far higher again22. The instinct that there is a lot of carbon dioxide down there is correct.
Part four is where the claim breaks, and one study settles it. Five species were grown in split chambers, leaves in one compartment with almost no carbon and roots in another with CO2 varied. Four of the five were tissue cultured plants from a commercial aquarium plant nursery23, which makes it about as close to an aquarium experiment as the literature gets.
| Species | Water at air equilibrium, about 0.7 mg/L | about 4.4 mg/L | about 22 mg/L | about 44 mg/L |
|---|---|---|---|---|
| Lobelia dortmanna, an isoetid | 98 % | 86 % | 58 % | 46 % |
| Lilaeopsis macloviana | 86 % | 50 % | 23 % | 20 % |
| Vallisneria americana | 75 % | 34 % | 15 % | 14 % |
| Ludwigia repens, a stem plant | 28 % | 6 % | 3 % | 3 % |
| Hydrocotyle verticillata | Could not use root CO2 at all, at any concentration | |||
Read the rows, then read the columns. Across the rows, plant shape decides everything. The best predictor was the ratio of shoot area to root area.23 A plant with roughly as much root surface as leaf surface can supply its leaves from below. A plant with fifty times more leaf than root cannot.
Down the columns, the contribution collapses as the water gets better. Even the isoetid falls from 98 percent to 46. The stem plant falls from 28 to 3.23 This is the fact the hobby claim gets backwards. Root CO2 uptake is not an addition to leaf uptake. It is a substitution for it. The sediment route saturates at a low ceiling and the leaf route does not, so the root fraction is large precisely and only when the water is carbon poor.
So saying "my dirted tank gets carbon from the soil, therefore I do not need to inject" inverts the logic. The soil route matters because the water has no carbon in it. It is a symptom of the reason to add CO2, not a replacement for it.
And there is a second ceiling. In every species tested, including the isoetid, the maximum photosynthesis achievable on root carbon was below the maximum achievable on leaf carbon. For Lobelia it was 69 percent of the leaf ceiling. For Vallisneria, 17 percent. For the stem plant, 3 percent.23 No amount of carbon in the soil closes that gap, because the limit is the plant's plumbing rather than the supply.
What it means for what you keep
Stem plants: effectively a myth. Epiphytes: zero.
Ludwigia, Rotala, Hygrophila, Bacopa, Limnophila, Myriophyllum. Large leaf area, modest roots, a long transport distance to the growing tip. The measured case got 3 percent of maximum photosynthesis from roots.23
Anubias, Bucephalandra, Microsorum, mosses. No sediment contact at all, so the mechanism cannot apply.
Rosette plants with heavy roots, Vallisneria, Echinodorus, Cryptocoryne, Helanthium, Sagittaria, are where there is a real and sometimes large contribution, in soft carbon poor water only. Carpeting plants are morphologically the closest thing in the hobby to an isoetid, which makes them the most likely beneficiaries, but nobody has measured them. That is a hypothesis, not a finding.
The ecological cross-check
Sediment carbon users lose ground where the water has carbon.
A survey of fifteen submersed plant communities in acidic lakes measured free CO2 in both the water column and the pore water. The relative importance of sediment CO2 using species was significantly negatively correlated with free CO2 in the water column.24
Root CO2 use is a low carbon specialism with a growth rate cost. It is the trait of a plant that has given up on fast growth, which is the opposite of what a hobbyist quoting the claim is hoping for.
So why do dirted tanks work?
Nutrition, not carbon.
A dirted tank is a well fed tank with a slow release nutrient bank, available iron, cation exchange, ammonium form nitrogen, and a stable microbial community. It is usually also a low light tank, which is what lets it succeed without injection: demand is matched to supply by turning demand down rather than turning supply up. The soil does produce CO2, and a fraction of that does reach roots that can use it. For rosette plants in soft water that fraction can be real. It is not the reason the tank works.
Not a fabrication, a promotion
What Walstad actually wrote about it.
Sediment CO2 uptake is the fourth of five carbon acquisition adaptations in her carbon chapter, about one page of a 225 page book, and she does not name it in her own summary of that chapter27. Her published literature index cites this same split chamber study and records its qualifiers correctly2823: substantial in carbon poor habitats, correlated with root porosity, small size and low water CO2. The full account is on the Walstad page.
On Tuesday morning
None of this changes what you do. It changes how much you trust a label.
The reliable way to start a tank is still fishless cycling. Dose ammonia to about 2 mg/L, keep the filter running and warm, test daily, and wait until the tank converts that dose to nitrate within 24 hours with no nitrite left. Then do a large water change and stock.
A bottle can help you start. It does not decide who lives in your filter long term. The 2001 work shows a good starter culture speeds up ammonia oxidation3. The 2011 and 2024 work shows the mature community is made of organisms that are not in any bottle712. Both are true.
The organisms in your filter are slow growers. Nitrospira and the ammonia oxidising archaea are adapted to scarcity, not to feasts.147 That is why cycling takes weeks rather than days, and why a filter you have scrubbed clean recovers slowly.
Never wash filter media in tap water. You are killing a slow growing population that will take weeks to rebuild. Rinse gently in a bucket of tank water.
Do not assume marine advice transfers to freshwater, or the reverse. The two communities are genuinely different and have been known to be different since 19961. In saltwater biofilters, no comammox has been detected at all12.
Low ammonia is not just safe, it selects a better crew. Archaea, which prefer low ammonia, dominate lightly stocked filters.7 A calm, planted, moderately stocked tank builds a different and arguably more robust community than an overstocked one.
Be sceptical of any label that names an organism. In 1998 the labels said Nitrobacter, and the Nitrobacter was not there.2 Labels are not tested by anybody.
Media boxes on this site are 5 to 2 and this banner is 16 to 5, both inside the range measured on grand prize contest photographs29.
The lesson is older and larger than a fish tank.
The 1996 paper ends with a sentence worth remembering. If nitrification is this poorly understood in a simple glass box, the authors wrote, then it is likely that nitrification in natural environments is even more complex and less well characterised with regard to the micro-organisms responsible.1 In other words: if we are this wrong about a fish tank, consider how wrong we are about a lake.
Thirty years later the aquarium filter is still one of the better studied model systems for this question, and the answer has changed three times. That is not a failure of the science. It is what a working field looks like from inside.
- Hovanec, T.A. and DeLong, E.F. (1996). Comparative analysis of nitrifying bacteria associated with freshwater and marine aquaria. Applied and Environmental Microbiology 62(8): 2888 to 2896. Free on PubMed Central, PMC168074. DOI 10.1128/aem.62.8.2888-2896.1996
- Hovanec, T.A., Taylor, L.T., Blakis, A. and DeLong, E.F. (1998). Nitrospira-like bacteria associated with nitrite oxidation in freshwater aquaria. Applied and Environmental Microbiology 64(1): 258 to 264. Free on PubMed Central, PMC124703. Received 4 September 1997, accepted 27 October 1997. Two affiliations are printed on it: the university department, and Aquaria Inc., Moorpark, California, which is also his contact address. This is the paper. DOI 10.1128/aem.64.1.258-264.1998
- Burrell, P.C., Phalen, C.M. and Hovanec, T.A. (2001). Identification of bacteria responsible for ammonia oxidation in freshwater aquaria. Applied and Environmental Microbiology 67(12): 5791 to 5800. Free on PubMed Central, PMC93373. All three authors at the company laboratory, no university affiliation. DOI 10.1128/aem.67.12.5791-5800.2001
- Hovanec, T.A. Curriculum Vitae, nine pages, dated 16 October 2020, published on his company's site. A self published primary source, treated with the normal caution, but the only complete list of his output. Everything in it we could check independently, the three papers, the patent numbers and the university affiliation, checked out. PDF
- Ehrich, S., Behrens, D., Lebedeva, E., Ludwig, W. and Bock, E. (1995). Nitrospira moscoviensis sp. nov. Archives of Microbiology 164(1): 16 to 23. Paywalled. The paper that named the organism, three years before it was found in fish tanks. DOI 10.1007/BF02568729
- The 1998 convergence. Burrell, Keller and Blackall, a nitrite oxidising bioreactor, May 1998, DOI 10.1128/aem.64.5.1878-1883.1998. Juretschko and others, municipal activated sludge, August 1998, DOI 10.1128/aem.64.8.3042-3051.1998. Schramm and others, a fluidised bed reactor, September 1998, DOI 10.1128/aem.64.9.3480-3485.1998. All three free on PubMed Central
- Sauder, L.A. and others (2011). Aquarium nitrification revisited: Thaumarchaeota are the dominant ammonia oxidizers in freshwater aquarium biofilters. PLoS ONE 6(8): e23281. Open access. The archaea in 23 of 27 filters. DOI 10.1371/journal.pone.0023281
- Bagchi, S. and others (2014). Temporal and spatial stability of ammonia-oxidizing archaea and bacteria in aquarium biofilters. PLoS ONE 9(12): e113515. Open access. Six freshwater tanks over four months, and the freshwater against marine reversal. DOI 10.1371/journal.pone.0113515
- van Kessel, M.A.H.J. and others (2015). Complete nitrification by a single microorganism. Nature 528: 555 to 559. Free on PubMed Central; the journal version is paywalled. DOI 10.1038/nature16459
- Daims, H. and others (2015). Complete nitrification by Nitrospira bacteria. Nature 528: 504 to 509. The companion paper, same issue, independent group. Free on PubMed Central. DOI 10.1038/nature16461
- Bartelme, R.P., McLellan, S.L. and Newton, R.J. (2017). Freshwater recirculating aquaculture system operations drive biofilter bacterial community shifts around a stable nitrifying consortium. Frontiers in Microbiology 8: 101. Open access. DOI 10.3389/fmicb.2017.00101
- McKnight, M.M. and Neufeld, J.D. (2024). Comammox Nitrospira among dominant ammonia oxidizers within aquarium biofilter microbial communities. Applied and Environmental Microbiology 90(7): e0010424. Open access. The most current picture of what actually lives in a home filter: 38 samples, comammox in all 38 and dominant in 30. DOI 10.1128/aem.00104-24
- McKnight, M.M., Szabolcs, N., Graham, A. and Neufeld, J.D. (2025). Microbial community succession of home aquarium biofilters associated with early establishment of comammox Nitrospira. ISME Communications 5(1): ycaf212. Open access. The twelve week succession study, and the 1.6, 3.1 and 3.8 percent figures. DOI 10.1093/ismeco/ycaf212
- Blackburne, R., Vadivelu, V.M., Yuan, Z. and Keller, J. (2007). Kinetic characterisation of an enriched Nitrospira culture with comparison to Nitrobacter. Water Research 41(14): 3033 to 3042. Paywalled; abstract free. The 0.15 g per g yield and the 0.54 mg/L oxygen half saturation constant. DOI 10.1016/j.watres.2007.01.043
- Liwarska-Bizukojc, E. and others (2011). Calibration of a complex activated sludge model for a full scale wastewater treatment plant. Bioprocess and Biosystems Engineering 34(6): 659 to 670. Open access. The heterotroph yield of 0.625 to 0.666 and the growth rate range of 3 to 6 per day. DOI 10.1007/s00449-011-0515-1
- Nogueira, R., Melo, L.F., Purkhold, U., Wuertz, S. and Wagner, M. (2002). Nitrifying and heterotrophic population dynamics in biofilm reactors. Water Research 36(2): 469 to 481. Paywalled; abstract free. The heterotrophic layer forming on top and cutting off the oxygen supply. DOI 10.1016/s0043-1354(01)00229-9
- Michaud, L., Blancheton, J.P., Bruni, V. and Piedrahita, R. (2006). Effect of particulate organic carbon on heterotrophic bacterial populations and nitrification efficiency in biological filters. Aquacultural Engineering 34(3): 224 to 233. Author version free at Ifremer. The 30 percent and 50 percent nitrification losses, and the doubling of attached bacteria. DOI 10.1016/j.aquaeng.2005.07.005
- Tsuneda, S. and others (2001). Enhancement of nitrifying biofilm formation using selected extracellular polymeric substances produced by heterotrophic bacteria. Water Science and Technology 43(6): 197 to 204. Paywalled. Heterotrophs building the house. DOI 10.2166/wst.2001.0374
- Song, G. and others (2024) and Zou, X. and others (2024). Enzyme enhanced acidogenic fermentation and lysozyme coupled protease pretreatment of waste activated sludge. Water Research 249: 120889 and Bioresource Technology 414: 131626. Both paywalled. The wastewater engineering evidence that the enzyme biochemistry itself works. DOI 10.1016/j.watres.2023.120889
- Liang, Q. and others (2022). Application of enzymes as a feed additive in aquaculture. Marine Life Science and Technology 4(2): 208 to 221. Open access. Cited to show that the substantial aquaculture enzyme literature is about feed, not about dosing water. DOI 10.1007/s42995-022-00128-z
- Conrad, R. (2020). Methane production in soil environments. Microorganisms 8(6): 881. Open access. The flooded soil redox sequence, and the statement that methanogenic breakdown of cellulose yields equal amounts of CO2 and methane. DOI 10.3390/microorganisms8060881
- Greenway, H., Armstrong, W. and Colmer, T.D. (2006). Conditions leading to high CO2 in waterlogged and flooded soils. Annals of Botany 98(1): 9 to 32. Open access. The measured paddy soil production rates and partial pressures. DOI 10.1093/aob/mcl076
- Winkel, A. and Borum, J. (2009). Use of sediment CO2 by submersed rooted plants. Annals of Botany 103(7): 1015 to 1023. Open access. The central source for the soil CO2 verdict: both tables above, the shoot to root area ratios, and the saturating root concentrations. DOI 10.1093/aob/mcp036
- Titus, J.E. and Pagano, A.M. (2017). Fifteen submersed macrophyte communities in acidic Adirondack lakes, with free CO2 measured in both the water column and the pore water. The source for the negative correlation between sediment CO2 users and water column carbon. Cited through the primary literature record
- Product pages, quoted as manufacturer claims and nothing more. Read for their own wording on what is in the bottle. DrTim's Aquatics One and Only and Waste-Away; Seachem Stability; API Stress Zyme; Fritz FritzZyme 7
- Our own arithmetic. Not a published source. Two restatements on this page are ours. One, the nitrifier yield of about 0.15 g of cells per g of nitrogen oxidised from source 14, restated as roughly a seventh of a gram. Two, the "at least 6 in 7" headline, which is our rounding of the full range of ammonia oxidiser fractions in source 13. That study measured 0.6 to 14.1 percent on the bead media, so at the worst case 85.9 percent of the community are not ammonia oxidisers, which is 6 in 7. On the biofilter sponges it measured 1.6, 3.1 and 3.8 percent, so 96.2 to 98.4 percent are not, which is about 24 in 25. The headline quotes the weaker of the two so that it holds across every sample the study took. calculated, repeatable, unverified by a third party
- Walstad, Diana L. Ecology of the Planted Aquarium, fourth edition, April 2023, Echinodorus Publishing, 225 pages, ISBN 979-8-9875741-0-2. The 23 page preview is open access and carries the complete table of contents and the whole of Chapter I, which is what the chapter structure and page count above rest on. We have not read the carbon chapter itself. Preview PDF preview open access, full book paid
- Walstad, Diana. Aquatic Ecology Subject Index. 2023. Open access, 175 pages, roughly 1,490 author-year citations arranged by topic. Her own record of the primary literature, including source 23 with its qualifiers attached. PDF open access
- Aspect ratios of grand prize aquascape photographs, measured. Intrinsic
widthandheightattributes read off the shipped<img>tags of the IAPLC grand prize archive, twelve photographs: 3.27, 2.83, 1.90, 2.20, 2.57, 2.84, 3.13, 3.21, 2.39, 2.50, 2.84 and 3.44 to 1. Recorded in our own reading of Source page iaplc.com/j/grand_prize_works measured, repeatable against the live page
Continue
Where to go next.
Cycling microbes and ammonia doses
Meet the filter's microbes and understand why more ammonia is not the goal.
Treating tap water
Why a conditioner protects the microbial filter as well as the fish.
Fifteen mechanisms, rated
What a water change does once the filter is running, rated mechanism by mechanism.
Chemistry
The ammonia equilibrium, which is the single most useful piece of chemistry a fishkeeper can carry.
Electrics and power cuts
How long the filter you just cycled survives a power cut, and what to do first.