Q. Last summer I built a 3000-gallon liner pond in my backyard. I constructed a very large biological filter (I used large gravel for the media) and put three koi — each about 12 inches — into the pond in September.

My problem is ammonia. A week or two after I put the fish in the pond ammonia readings went to 0.25 parts per million (ppm) and they have not gone back down. I called a local pond "expert" who told me that the problem was that I had too few fish! Wow, was I shocked. He explained that because I had such a small fish load there was not enough ammonia to get my large biological filter working. So the ammonia built up. He said the best thing to do would be to buy about 12 more fish — they would produce enough ammonia to feed the filter and then everything would be fine.

Having read your column in the past I thought I was doing the right thing by having only a few fish. But, now I am afraid I won't be able keep my filter running properly. Please help.

A. I wish I could say this was the first time I had heard this nonsense, but it is not. Believe it or not, there really are some folks out there who fervently believe that too small a fish load prevents the biological filter from detoxifying ammonia, which in turn leads to higher ammonia levels in a pond than if there were more fish. This is "voodoo fishkeeping" at its finest.

Let's consider this. A biofilter is nothing more than a convenient place for nitrifying bacteria to hang out. The only special design feature of a biological filter is that it tries to cram as much surface area for bacteria to attach to into as small a space as possible while still allowing water to pass through freely.

Controlled laboratory experiments do indeed show that, given the room, the density of nitrifying bacteria is directly related to the concentration of ammonia in the water — all else being equal. So, higher concentrations mean more bacteria. The converse is also true: lower concentrations result in reduced bacterial colonization.

But, these bacteria are always in water. They live on pond walls, on plants and even float freely in the water. If there is any ammonia in the water, these bacteria are available and in numbers commensurate with the ammonia load. If there was not enough ammonia in the water to stimulate bacterial nitrification, the concentrations would be so low they would be irrelevant to the fish and unmeasurable by your test kit.

So let's keep our facts straight: The amount of ammonia produced by fish does not determine the capacity or effectiveness of a biological filter. It is the steady-state (long-term) operational capacity of a biological filter that determines the maximum ammonia load in a recirculating water fishpond.

Plunking more fish into the pond is a guaranteed way to kill off your fish by ammonia poisoning. The reason your filter does not appear to be working has nothing to do with the fish load. It is not working yet because you set it up as the weather was turning colder (nighttime temperatures in the 50s and below), which means it will take about 60 or more days to finally operate at stable capacity. Poor water conditions, such as high acidity (low pH — around 6), high organic content, high chlorine content and poor oxygen levels will also inhibit rapid and effective establishment of nitrifying bacteria colonies.

Also, check that the water flow through the filter is not channeling along a narrow path that prevents circulation through the gravel. Channeling creates a situation that is similar to having no biological filter at all.

Frankly, if you add some plants to your pond and keep the fish load right where it is you will not need a biological filter at all. The nitrifying bacteria on the pond walls and the plants will take care of everything. In fact, I'll bet that you will never get measurable ammonia readings again.

Keep in mind that the goal of pondkeeping is to cultivate fish, not nitrifying bacteria. Beware of pond experts who tell you that you need more fish to make your biological filter work properly — in the next breath they will offer to sell you some of their fish!

Follow the Clues
Q. I have a 55-gallon-long tank that houses four goldfish: one 5-year-old comet, one 2-year-old black moor, one 1-year-old fantail and one 2½-year-old orange oranda. My problem is with the orange oranda.

The fish started swimming upside down and seemed to be out of control. Constipation? Someone suggested placing one to two grains of Epsom Salts into its mouth behind the teeth to help clean out its digestive system — which I tried. It seemed to make the fish better, but it is still not acting normally.

I then tried a product called Quick Cure for two days as directed — leaving the carbon in the filter and raising the water temperature to 83 degrees Fahrenheit. Nothing. Now I am feeding tetracycline-medicated food for five days.

This is not the first, or even the second, time my orandas have gotten sick and died. I have tried every suggestion and read many books and magazines (including AFI) for advice. My fish is still alive, but not well. Enclosed is a tank maintenance schedule.

A. The writer enclosed a very useful detailed summary of her fishkeeping practices. It was really helpful to have all the facts outlined concisely. Among the things she reported: a 30-percent weekly water change, good mechanical and chemical filtration, good aeration, pH between 6.8 and 7.2, no ammonia or nitrite, water conditioning with salt and Novaqua, and feeding 3 tablespoons of soaked pellets once per day.

So what is the problem? Our first clue is that the oranda has trouble maintaining its normal balance. A second clue is that there are no external wounds or other signs of external parasites or bacterial infection. Our third clue is that it has happened before to other orandas she had kept. The fourth clue is that it has not happened to other goldfish varieties. Our fifth clue is that water conditions are fine. And the sixth clue is the feeding schedule.

Now, the first and second clues tell us that an internal problem is preventing the fish from orienting itself properly. This is commonly labeled as swimbladder disease, but whether or not the swimbladder is actually involved and whether or not it is diseased is not clear.

The selective propagation of bizarre body-type goldfish varieties does lead to misshapen or inadequate swimbladders. And some breeding stock may be more prone to this problem than others. So, if the writer keeps getting her orandas from the same supplier (clue number three limits the problem to orandas), then perhaps it is a genetic defect. The answer would be to get orandas from a different source.

The sixth clue raises the possibility that the problem may be feeding-related. Goldfish do not have true stomachs. Rather, they have very long intestine-like structures that wind around the inside of their bodies. A gas buildup inside the digestive tract can cause a goldfish to flip over for an hour or more after feeding.

Alternatively, the direct feeding of dry foods — pellets or flakes — can cause a similar effect as the food absorbs moisture in the fish's gut and expands. Because the writer soaks the pellets this does not appear to be likely.

But, feeding 3 tablespoons of food per day in one feeding could be the root of the problem. That is a lot of food for just four fish. I have a 300-gallon indoor pool with the same number and variety of goldfish and I feed them 1 tablespoon of pellets per day. My goldfish are big, healthy and six years old on average, so they are not starving.

Systematic overfeeding can cause chronic digestive problems, leaving a marginally balanced goldfish perpetually upside down. A marginally balanced goldfish is one that is already teetering on the brink of imbalance because of its body shape and swimbladder structure and placement.

I suggest you cut back your feeding to 2/3 to 1 tablespoon per day. I also suggest you divide the feeding into at least two, or perhaps three, meals per day. You might also supplement the feedings with some cooked and shelled peas every few days.

I also suggest you eliminate the salt in the water. It is not doing anything useful for the fish and it makes it harder for those that already have flotation problems to maintain their balance. (Yes, it's true. Using salt as a permanent additive to a goldfish tank has no practical effects. It does have therapeutic value when used as a short-term treatment, which is a different matter.)

Notice that there was no indication of infection or disease. After all, why would just orandas get sick? Were there hemorrhage marks on the body or fins? And the fish did not stop eating. Notice that the writer tried at least two different fish drugs, which had no effect.