Quite recently, a scientific paper from Denmark, published in Microbial Ecology, discussed the results of monitoring of the biofilm function on solid substrate, collected in two Danish streams. Both streams receive large amounts of effluent from a sewage treatment plant and run-off from agricultural areas, rich in nitrate and organic matter. These biofilm layers are similar to those in uncleaned aquariums, but they build up to 1 mm in thickness.

 

I found this information fascinating, for I have been trying to obtain data on the biomechanics of biofilms for years but even our sewage treatment laboratories have kept out of this field. Sewage aeration tanks, where digestion of organic matter occurs, depend on very high levels of solids present, where the bacteria can populate the surfaces of the whirling particles. If these particles are larger than the bacteria, then the process becomes more effective the more the dirt in the sewage (source: University Ph.D. thesis, U.S.A.). Aquarium water of any quality is crystal clear by comparison.

 

The Danish scientists actually measured the microprofiles of oxygenand nitrate by use of a microsensor. The object of the research was tomeasure the depthwise distribution of denitrification and oxygen respiration. Within the 1 mm of film thickness, penetration of oxygen and nitrification occurred to a depth of 0.33 mm; then followed a thin (0.03mm) layer of mixed nitrification-denitrification, and finally, the denitrification took over to a depth of 0.88 mm, where diffused nitrate was zero.

 

Denitrification always appeared when oxygen was depleted and the capacity of the bacterial community to perform rapid shifts from oxygen respiration to denitrification seemed to be the principal adaptation to fluctuations in oxic/anoxic conditions in the biofilm. It was also noted that the bacterium Thiosphaera pantotropha, abundant in some water purification systems, performs oxygen respiration and denitrification at oxygen concentrations near to air saturation levels.

 

The required supply of carbon was shown by these studies to be derived partly from algal cells and new research into biofilms of tricklefilters indicates that layers may switch from nitrification to denitrification, according to the light/dark regime.

 

Practical application of the denitrification process in the aquarium hobby faces the same problem as do the water treatment facilities. Actually, the similar levels of nitrate (100-200 ppm) are relatively low for a rapid treatment but the use of thick layers of bacteria is really a long-term process. French scientists have produced a simple battery of vertical pipes and they thereby achieved almost zero levels of nitrate. I am searching for more details, particularly the materials and grading used in their system, as a clue for similar filter construction.

 

At present, the solution lies in the installation of a separate tank (or chamber), which follows the trickle-filter oxidation (nitrification), since the denitrification stage must not receive free (dissolved) oxygen. In effect, oxygen must be consumed before or at least in the early part of the denitrification tank and this means an airtight lid and siphon inlet type of arrangement. The bacteria established in the denitrification stage also require a source of carbon, but methanol, with its combustible and explosive nature, is not the answer. Carbon dioxide is more convenient but unfortunately, its higher oxygen content reduces the efficacy of the process.

 

Water velocities through the denitrification stage should be slow and a bypass is recommended for regulation of the flow. The infill media should have even more area than in trickle-filters, because there is no airwater mixture to need larger passages. Properly graded sand, spent activated carbon, lava rock or expanded rock are suitable, as offering large surface areas for the growth of bacteria.

 

An example of an aerobic/anaerobic aquarium unit is shown in TFH Nov/88, together with a detailed description of its practical aspects. Certainly, the measured results are impressive.

 

The operation of a complex nitrification/denitrification filter is relatively reliable, except when power failures affect the circulation pump. A circulation stoppage of no more than 20 min is permissible, after which bacterial death will lead to an environmental disaster. A time-control switch, not allowing automatic re-start of the system, is probably the best solution, but unfortunately, cleaning and restart of a unit to re-establish new bacteria, is a slow process.

 

The design of the whole unit also requires provision for stepwise cleaning of the infill media in any part of the filter, to keep areas of established bacterial growth above the minimum working levels.

 

The main operational parts of a nitdfication/denitrification system are:-

 

  Prefilter (strainer. foam or floss mat) to remove large particles. A combination of useful gadgets, such as surface skimmers, etc., certainly improves performance. The prefilter area should be cleaned frequently, the more often the better.

 

  Nenitrification stage (trickle-filter, trickle trays, sprayed layers of foam or filter-wool). Cleaning is required to maintain an unobstructed flow. Excessive growth of bacterial film is culled to 1/3 of the maximal volume at any time.

 

  Denitrification stage (anaerobic chambers). Cleaning is due whenever clogging obstructs flow. Temporary nitrate buildup after cleaning does not threaten life in the system. Bacterial colonisation of the cleaned elements is rapid, from the trickle-filters.

 

  Settling well needed for settlement of shed layers of biofilm and mineralised matter derived from bacterial activity.

 

  Recirculation pump to provide the needed flow through the system.

 

The above-described system provides the ultimate answer to closing the biological cycle. For many hobbyists it may possibly amount to an ‘overkill' but for serious ones and breeders, it is the only way to avoid massive wastage of water. In view of predicted future shortages of water resources, this is likely to become a very important consideration.

 

Rainbowfish in the wild are foragers: they feed on Algae, higher plants, small crustaceans, aquatic insects, terrestrial insects, tadpoles and the occasional small fish. Algae are the one item of food that appears to be consumed the most. When a rainbowfish is taken from its creek, river, billabong or lake, the first thing to he noticed is that it excretes a long, dark green string of fibrous faeces. The fish has usually a streamlined shape and 1 think they get that way from having to work so hard to survive in their hostile environment.

 

Tropical areas have two main seasons a wet and a dry. During the wet season, food is plentiful and this is when most reproduction occurs. The proteinaceous foods (crustaceans, insects, tadpoles, fish) are abundant. However, rainbowfish collected during the wet season still excrete the green fibrous faeces, after capture. When put into a well planted tank, the first thing they do (especially the larger ones) is consume your plants. Thus the inevitable conclusion is:-

 

 

What happens if we feed a rainbowfish on an all-meat diet? I don't know exactly but I have heard from a person who keeps beautiful and well proportioned rainbowfish in Queensland, that experiments are being conducted in relation to their diets. Rainbowfish that have been fed high protein diets are being dissected and they prove to have deposits of fat around their internal organs. Any animal that has too much fat in its system is stressed and the first lesson that we learn as aquarists is that stressed fish are more likely to be affected by disease organisms, which are present in all aquariums.

 

I therefore believe that the answer to poorly proportioned rainbowfish, ulcer disease and most of their other problems is in their diet, as well as in good clean water and uncrowded conditions.

 

If you don't wish to take my word but are having troubles with deformed or diseased rainbowfish, then try the following experiment, which is easy to conduct. Take two equal-sized aquariums, set them up in exactly the same way and breed your favourite rainbowfish, keeping the same number of fry in each tank. Label one tank 'Vegetarian' and the other 'Normal' and keep a notebook to record differences between them. Feed the fish in the vegetarian tank with vegetarian flake food and frozen food mixture containing at least 50% vegetable matter but provide those in the normal tank with whatever foods you would normally use to obtain maximum growth. The growth rates in the two aquariums may well be different: in the vegetarian tank the fish may develop more slowly.

 

A suggestion for a frozen food containing plenty of vegetable matter would be cooked zucchini, boiled spinach or peas, or cooked pumpkin for one half of the mixture, the other half being prawns, cooked or frozen, and fish fillets. Stay away from beef heart and other land-animal products.

 

Insects form another important food for rainbowfish and freeze-dried or frozen mosquito larvae are usually available from the aquarium shop. Fruitflies are easy to culture and form a good alternative live food. Starter cultures of vestigial-wing fruitflies are available from Southern Biological Supplies in Melbourne and a recipe for a suitable culture medium is given in the book 'Australian Native Fishes for Aquariums', which should be in our library.

 

Rainbowfish will eat almost anything put in front of them, so it is up to us to make sure that they get the right foods to keep them in good health.

 

My property in this region forms part of an elevated ridge, for the most part covered in rainforest and attaining a maximum height of 550m, some 100 m above the surrounding (and largely settled) plain. The average yearly rainfall is about 1800 mm (72"), i.e. nearly three times as much as we get in Canberra. Numerous gullies carry temporary creeks during the wet summer season but these peter out as the rains decline and dry out completely during the winter, when the periodic light showers are not enough to provide the needed run-off to sustain them. Not surprisingly, fish are absent from these temporary creeks. However, at the base of the ridge, both on my side and the opposite one, on a friend's property, are spring-fed perennial creeks that I was for a while tempted to regard as permanent, and these present a very different picture.

 

To my delight, some years ago and soon after acquiring a toe-hold in the district, I detected thriving populations of the Northern Trout Gudgeon (Mogurnda mogurnda) in both creeks and of the Eastern Rainbow Fish (Melanotaenia splendida splendida), plentiful on my friend's place but with smaller numbers also on mine. Both species readily responded to a scattering of bread crumbs but the Gudgeon, being a bottom dweller rather than a free swimmer, seemed to prefer to feed on small pieces of cheese that immediately sank beneath the surface.

 

My irregular monitoring showed both species to have been present for several consecutive years until the drought commenced in mid-1992. Little by little, the flows decreased until, in August, they ceased altogether, at least upon the surface, and the creeks were reduced to a few isolated and rather murky-looking pools. By September the pools had also gone and the creek-beds appeared to be completely dry.

 

At this stage domestic water-bores were showing an ever descending water table and the rainforest trees responded with an unusually heavy leaf-fall. What, I wondered, had become of my treasured fishes?

 

The dry spell continued with little relief until December, when more than 500 mm (20") of rain suddenly renewed all systems and my creeks started to flow once more. I approached them with bated breath but the water was so cloudy that it was impossible to tell whether the fish had survived or not and I had to curb my curiosity until the end of the 1993 summer wet.

 

Now (August 1993), the creeks appear to be back to normal and I'm glad to report that thriving populations of the Gudgeon (including some quite large specimens) are still present, although the Rainbowfish have disappeared. The former, being bottom dwellers, presumably managed to survive in the few remaining damp spots under the creek beds but the free swimming Rainbowfish probably perished when the last pools dried out.

 

Although this drought was one of the worst on record, similar ones must have occurred in ages past and my creek has no doubt dried out many times before. Perhaps the Gudgeon has always been present, in active or passive form, but it seems likely that the populations of Rainbows come and go. Presumably they are replenished after such disasters by upstream migration from the safer havens of the larger river systems of the surrounding plain, for which the local creeks are merely feeders. I certainly hope so and I shall be keeping a keen lookout for the return of these beautiful fish.