Introduction to the Breeding and
Maintenance
of the Genus Nothobranchius:
African Killifish

by Charles A. Nunziata

Every good fish breeder knows that there is no substitute for knowledge when it comes to the maintenance and propagating the animals in their charge. In fact, success is generally in direct proportion to the amount of study and effort one puts into the process. The “dedicated aquarist” is one that consistently provides a healthy environment and appropriate nutrition. The former reflects how fastidiously one adheres to a regime of water changes, water preparation where necessary, and disease prevention. As importantly, environmental needs in the form of space, plantings, and spawning site simulations are also necessary to both the overall health of the fish and the ability to propagate it. Nutritional needs and the ability to find methods of providing suitable alternatives are equally necessary if one is to succeed. The so-called difficult fish are only considered so because their needs may outstrip the breeders ability to provide for their environmental or nutritional needs either through ignorance of what those needs are, or by the lack of time or interest. This article proceeds from the notion that most “difficult” freshwater fish become much less so when a thorough knowledge of their needs are both known and are within the ability of the breeder to provide them. Given that the breeder has the time, resources and interest level to provide for the physical needs of their fish, then it is the understanding of the behavioral and environmental stimuli particular to the group that will determine breeding success.

Nowhere is this more important than in the maintenance and propagation of one of the most, if not the most beautiful group of freshwater fish in the world, the members of the killifish genus Nothobranchius. Considered “difficult” by those outside the killifish community and by some within it as well, many members of the genus Nothobranchius are fairly easy to both maintain and breed. This is due in large part to their extraordinary life cycle, their extremely rapid growth, early sexual maturation and fecundity. When other elements of behavior and embryo development are well understood, success will follow.

Nothobranchius rachovii (Click to Enlarge)

Nothobranchius Breeding and Maintenance:

The unusual life cycle of Nothobranchius is an adaptation to their environment. Nothobranchius inhabit temporary bodies of water. These include pools and wetland environments, which contain water for only part of the year. These repeatable wet/dry climates are common in Africa where Nothobranchius are found, and in South American where the distantly related genera Pterolebias, Cynolebias, Austrolebias and their kin have a similar life cycle.

The life cycle is a continuum of course, but to examine it in its simplest terms, we start with the newly hatched fry.

With the first post-drought rain, fry hatch from eggs buried in the substrate. Newly hatched fry grow rapidly and reach sexual maturity in 3 to 6 weeks. Spawning commences and continues throughout the lifetime of the fish. New eggs are deposited in the muddy substrate of the pool. The rains cease, the water eventually evaporates, and all the fish expire. The pool remains dry throughout the dry season. The new rains arrive, and the mature eggs hatch from the substrate, restarting the life cycle.

Nothobranchius are classified as “annuals” because they do not live more than one annual cycle in nature. And even under aquarium conditions their lifespan does not exceed 1 ½ years.

Nothobranchius melanospilus (Click to Enlarge)

By carefully studying this process, one can determine nutrition requirements, breeding methodologies and the manner that eggs should be handled. When the annual rains come and fill depressions to produce ponds, or flood lowland areas to produce wetlands, there is an enormous explosion of life in and around the flooded areas. Insects and animals are attracted to these new water sources, further adding to the ecological diversity that abounds. And it is into this vibrant environment that Nothobranchius hatch. Recognizing this, we can make several conclusions about the fish and its requirements.

The upturned mouth of the Nothobranchius implies that food can be taken from the surface. We can easily see how the temporary pond would attract flying insects, and float terrestrial insects at its surface. We can also assume that worms and small crustaceans will abound in the bottom. Because Nothobranchius grow so rapidly they must feed immediately on the life there present rather than on vegetation that will take some time to develop. So our Nothobranchius must be meat eaters, and research bares this out. In aquaria, live foods of all kinds are immediately taken, and the fry of all but a few species can take newly hatched brine shrimp. Frozen and freeze dried foods will also be taken after it is recognized as a food source. Fry are best started on newly hatched brine shrimp the first week. Frozen baby brine shrimp, microworms and other small living foods can be added until week 3 or 4. Finely chopped frozen foods can be introduced thereafter. Mosquito larvae and Daphnia are also excellent foods if available, and freeze dried mosquito larvae and bloodworms are dry foods that are usually taken as juveniles and adults. Feeding Nothobranchius is not a problem as they will grow and mature under virtually any combination of live or prepared foods.

We can also conclude from our study of the environment that Nothobranchius thrive in rapidly changing water conditions. Multiple heavy rain events, followed by a drying process and interspersed with animal use, make for an ever-changing chemistry that reaches extremes, especially during the drying out cycle. This is not to say that they will thrive in poor aquarium conditions, but rather that they do not require special water and will thrive under ordinary aquarium maintenance regimes. Weekly or bi-weekly water changes and avoidance of very rapid changes in water chemistry is all that is required to keep Nothobranchiushealthy and active. pH ranges from 6.8 to 8.2 are easily tolerated. And although wide ranges in water hardness are also tolerated, harder, rather than softer water is recommended as a disease deterrent.

Several diseases are less prevalent in water above 100 ppm, and one of these, Oodinium, is a disease to which Nothobranchius have little resistance. This dinoflagellate infection commonly known as “velvet disease”, is easily avoided through both the aforementioned water changes, and the addition of 1 tablespoon of coarse rock salt, or one teaspoon of fine non-iodized salt per 5 gallons. Appropriate amounts of salt should be added at each water change to maintain these levels.

The muddy or sandy bottom of a temporary pond or wetland tells us that Nothobranchius will require some kind of a pliable media in which to spawn. To be sure, Nothobranchius are such prodigious breeders that many species will attempt to spawn even on a bare take bottom. But the addition of a suitable substrate will result in a high degree of spawning activity, and allow the easy recovery of eggs. There are a few spawning medias in common use, but peat moss is by far the most popular choice. A good unadulterated finely chopped horticultural grade of sphagnum peat moss in loose form or compressed into a pellet, is recommended. Common garden peat typically found in nurseries is to be avoided because it often contains pesticides, fertilizers or other contaminants.

The best chance for early success by the beginner is the dedicated breeding set up. In this method, Peat moss is added to the existing tank, or to a bowl or tank specifically set up for breeding. Soak the peat thoroughly in a pail or other container until it sinks. Some breeders insist on boiling the peat moss and such a practice will allow it to sink more readily. Boiling is not necessary unless you are not sure whether the peat contains parasites.

Nothobranchius guntheri (Click to Enlarge)

The spawning container can be as small as a one-gallon bowl or up to a 2-½ gallon tank. Larger containers are not necessary. Fill the container with peat moss to a depth of 1 ½ “ and let settle. Add an air supply, with a slow to moderate stream of bubbles. No filters are required. Introduce the preconditioned breeders for a period 4 to 12 hours, after which they are to be removed and returned to their normal quarters. Net out the peat moss, squeeze out as much of the water as possible, and place the peat between a thick layer of newspaper for 24 hours. After that time, the peat will be damp, but neither wet nor dry. To test, squeeze a small amount of peat between the forefinger and thumb. This will dampen your finger without producing discernible water. Now place the peat in a plastic bag for storage, the importance of which is detailed below. Mark the bag with the name of the fish, and the date that you store the eggs.

You will note that we mentioned “preconditioned” breeders. Preconditioning refers to the practice of bringing fish to their optimal spawning condition before introducing them to the breeding setup. In Nothobranchius this involves isolating the males from the females and feeding both primarily live foods for a 2 to 3 day period. Once preconditioned in this manner, the fish will spawn readily, will produce the maximum number of eggs and will do so with minimal damage to each other.

Alternatively, breeding in a permanent tank is possible where a temporary dedicated spawning setup and preconditioning are not practical. Because of Nothobranchius are prodigious breeders, such arrangements produce good results even though they require more attention. Typically, a small container with the peat moss is inserted into the permanent take housing the males and females. Care must be taken to prevent dispersion of the peat moss during insertion. A minimum 3” deep plastic container is weighted with some gravel. Anything shallower will result in the fish throwing the peat out of the container during spawning. Cut a 1” to 1 ½” hole in the center of the cover. Fill the container with a 1” layer of peat, fill the container with water to the brim, snap on the cover and stretch a piece of plastic wrap over the top.

Now slowly insert the container to the bottom of the tank. Slowly remove the plastic wrap exposing the opening in the cover. This will prevent dispersing the peat accidentally throughout the tank when inserting the container. The Nothobranchius will find the hole in the container after a few hours and will spawn vigorously in the peat. The container cover will keep most of the peat in the container during the spawning process. Since the fish have not be preconditioned, leave the container in the tank for 3 or 4 days after which it is removed and the peat handled as noted above. Some peat will inevitably escape the container and this should be removed during regular maintenance.

The common perception is that the male is dominant, and initiates the breeding activity. The female is considered passive and accepts of the male's advances if she has eggs, while refusing to do so when she does not. The relationship among males is one of dominance and subservience in breeding opportunities and ultimate success. As we will see, these two issues are closely interrelated. Figure I shows a graphical representation of data (Haas, 1976a) regarding the relative reproductive activity of dominant and subordinate N. guentheri males with females that have not been isolated or preconditioned. Please note that in the first two hours the number of encounters are few and essentially the same for both dominant and subordinate males. Neither are pursuing females in any vigorous way. It is not until the third hour that a significant change in the number of encounters occurs. It has been shown that Nothobranchius female ovulation takes place some hours after daylight, or “lights on” in the fish room Post-daylight ovulation seems to explain the delay in activity if and only if it is the female that initiates the spawning activity. Her physical state when encountering a male seems to determine whether or not spawning will occur. Figure 1 shows how spawning related activity of dominant males increases dramatically between the third and sixth hours and then drops off almost as dramatically between the sixth and eighth hours, presumably because of fatigue and the need to feed.

Figure 1(Click To Enlarge)

The above data supports the recommendation that isolating and conditioning females prior to spawning will produce good results, but more importantly, it also indicates that almost all the eggs will be laid in the first few hours of contact. This is the major reason for the recommendation noted above for a dedicated spawning setup and preconditioning to assure breeding success.

We must also be mindful of what is happening within the male-female relationship in order to understand what characteristic we are purposefully or ignorantly propagating. And to address this, we must understand how mates are selected. Having established that the female initiates spawning, we need to know whether mate selection preferences play a role in this process. As it turns out, mate selection not only determines how the progeny look, but how large they get and how long they live.

The issue of mate availability in nature is not really applicable to Nothobranchius because males are not territorial and swim randomly throughout their environment (Haas, 1976b), making them continuously available to females. Females need only become aware of their presence, and upon selecting them will spawn immediately upon physical contact. One may ask whether this makes sense from an adaptive standpoint, and it does with respect to the females. Because the female does not need to seek out the males, she conserves energy that in turn permits the greater utilization of food resources for egg production.

The need for a simple, rapid, and efficient breeding process must be supported by simple and efficient methods of recognition and attraction. The implications flowing from the observation that females primarily initiate spawning activity dictate that any mechanism that enhances the circumstances by which females become aware of males will predominate. Therefore, display strategies, as well as color and size recognition, would appear essential to this idea. However, Nothobranchius males do not display, but immediately pursue, clasp and spawn with the receptive female. That leaves color pattern an/or color intensity as primary factors in the female’s response to males.

Fundulopanchax gardneri ... Another Type
Of African Killifish. (Click to Enlarge)

We are concerned here with two primary response mechanisms: homospecificity and color attraction. Since it is not recommended to keep more than one species of Nothobranchius together, homospecificity is not an issue. Color recognition is the mechanism that most affects mate selection in the artificial environment.

Females of virtually all Nothobranchius species are almost universally brown or grayish, with virtually transparent fins. In the murky waters of the temporary pools they are for all intents and purposes, invisible. Males on the other hand sport the dramatic colors and combinations that make them very visible in all but the poorest light. Although all the fins as well as the body of most male Nothobranchius are richly colored, it is the caudal fin that is the dominant feature. Large and brightly colored caudal fins are the flags of recognition in the Nothobranchius pond. There are three dominant caudal fin colors: Red and related, Blue and Yellow.

Turbidity selectively absorbs more short-wavelength light than long-wavelength light. This shifts peak transmittance towards the longer (red) wavelengths, and because pure water has its greatest transmittance at 480 nm(blue-green), red, orange and yellow are most visible in turbid water.(Kinney, 1967). A bright red, orange, or yellow caudal enhances the visibility of males in a murky environment, while as noted above, plain females are virtually invisible to both predators and males. It appears that the survival strategy is for the female to notice the male, and to do so efficiently in the turbid water. Note that the other colors most visible in these cloudy environments, yellow and orange, are the most common caudal colors of Nothobranchius that have no red. These observations seem to further imply that those Nothobranchius whose primary caudal colors are blue or green should be found in clear water, but I have not seen a study confirming this implication.

Figure 2 confirms these observations. Here we see a preference profile highly dependent on the environment. The "X" axis is a turbidity measure from clear to turbid in percent visibility. The "Y" axis is a measure of response from none (0%) to always responsive (100 %). Responsiveness is defined in this study as one fish going towards and remaining near the other: female to male (F/M), female to female (F/F), male to male (M/M) and male to female (M/F).

In clear water, females spent three times as much time near males than near females, clearly confirming their preference for males. Likewise, males spent approximately twice as much time near females than males. These ratios do not change appreciably when visibility is reduced to 68%, but between 68% and 45% visibility, males lose identification with females, and at 45% visibility males spend about as much time near females as they do near males. in other words, they can’t see them well. Females, however, maintain strong preferences for males because, it is thought, color attraction begins to augment, and then supersede, homospecific attraction.

Figure 2 (Click To Enlarge)

Female Color Intensity Preference:
Male Target	Time/male (sec)
Intact Caudal	355
Bright Caudal	354
Pale Caudal	221
Amputated Caudal	219

Fig. 3.

The implications for the breeder is obvious, when given a choice, females will breed with the brightest colored males regardless of their other attributes. If there are more than one male in the breeder setup, the female will tend to spawn with the male that has the most colorful and largest caudal fin. Of a less obvious nature, side-by-side tank setups where the female can see the male in the next tank may divert her attention from her intended mate. In fact, activity in the neighboring tanks may divert attention in any case, and it is best to put some barriers between tanks to avoid these distractions.

So what is the downside of female selection of mates dependent on the size and intensity of color or the caudal? After all, we as hobbyist make a similar decision when we select the breeders, and normally assume that the males, which develop first, are the strongest, the most colorful, and the most prolific.

Studies, however, do not support this contention (Markofsky and Perlmutter, 1973.) and in fact point to a completely different conclusion. The data suggests that the earliest maturing males are the shortest lived and are smaller at the end of their lives than the slower maturing males. Therefore, if we select the largest most colorful males from the brood for breeding, we may be propagating shorter lived fish, and ones that will not ultimately reach their largest size.

This counterintuitive twist is supported by the data presented in Figure 4. At 240 days of age, the length of the short-lived males is 6% larger, and from the data one can assume a much larger disparity at younger ages. With breeding beginning in 3-5 weeks, one can see that these fast-growing males will dominate the spawning of available females. Yet these males grow at a much slower rate beyond 240 days, and virtually stop growing at 360 days. The slower-growing specimens are still growing at a substantial rate at 360 days of age, and have by that time attained a larger size. Furthermore, the short-lived fast-growing fish die at 420 days, whereas the slower growing specimens die at 480 days, a full 14% longer life span. Beyond 280 days, the slower growing specimens are clearly superior from a vitality standpoint and will no doubt be at least longer lasting breeders.

Figure 4 (Click To Enlarge)

To bolster this, Figure 5 shows the relationship between life span and the onset of maturity.(Markofsky and Perlmutter, 1972). The percentage of late maturing specimens alive as a function of time is much higher than for the early maturing specimens. This graph shows a substantial percentage of early deaths, nearly 20%, between 4 and 6 months of age among the early maturing specimens. Thereafter, there is a steady loss among this group, which is not mirrored, in the slow maturing group. At 14 months of age, most of the late maturing specimens are still alive, whereas less than 50% of the fast maturing specimens remain.

Figure 5 (Click To Enlarge)

The implications for the breeder are once again clear. Rapidly developing males should be removed and isolated from the rest of the spawn. This will allow the slower growing and slower maturing males to develop at a faster rate. This should be continued as additional males develop until all of the males have revealed themselves. Then and only then can the breeder evaluate which fish to breed. Most breeders select for color intensity, body shape and fin characteristics. Remember that the smaller fish are likely to reach the largest size, and if maximum size is a characteristic desired, select the late maturing fish.

When on thinks about this thought, it does make sense, ever from an ecological standpoint. Going back to our pond, the largest, fastest and most colorful males will breed first, but they will also be the first victims of predators. Their removal allows the maturation of the remaining males, assuring a continuous supply of maturing males until he pond dries out.

To summarize:
Females, not males, initiate the spawning process.
In a dedicated spawning setup:
Isolate and condition females at least 24 hours prior to spawning.
Remove breeders after 12 hours.
The female will spawn with the most colorful male near her.
Minimize distractions from neighboring tanks.
Select breeders for color, pattern, and shape.

Nothobranchius Eggs

We know from our study of the environment that the eggs of Nothobranchius can both persist and develop in a semi-dry, waterless state. Research shows that there are three points in the embryonic process of the Nothobranchius embryo, designated as “Diapause I, II, and III” where development virtually ceases.(Leibel, 1977) Diapause I occurs immediately after fertilization and initial cell division. This stage may persist for a few weeks to many months, or even years. Diapause II takes place at an intermediate stage after major body structures are developed, and the heart is beating. Diapause III occurs at the end of the developmental cycle and immediately before hatching. It may persist for weeks, but once at Diapause III, the clock is ticking since it is at this stage that the life functions are fully operating and if hatching does not occur within a reasonable time, the embryo will consume the available nutrition and die within the egg.

These pauses in the embryonic developmental cycle should be allowed to occur in order to obtain healthy, viable fry. Therefore, aside from some extraordinary processes, which are beyond the scope of this article, the eggs must be in a semi-dry state during incubation. Actually, it is not too difficult to provide the environment that allows the natural Diapause events to occur. The eggs simply need to be kept damp but not wet, warm but not overly so, and free of contamination for a period of time that mimics the period during which they naturally subsist in the dried out pond or wetland. The storage of eggs in peat moss discussed above mimics this process, and allows the eggs to go through their normal pattern of Diapause and development in a fashion that is similar enough to natural conditions.

Different species of Nothobranchius have different periods of development. These range from a few months to several months to first hatch. The term “first hatch” itself implies there can be more than one hatch event. In fact, after hatching and removal of the fry from the peat storage media, the peat can be redried and stored for another few months. Second and even third wettings may produce fry. This is so because the eggs do not all develop at the same time or at the same rate.

Nothobranchius foreshi (Click to Enlarge)

In the chaotic environment of the temporary African pool, we need to consider the fact that not all rain events are substantial enough to keep the pool filled for months at a time. In most cases, but not always, once the rains start they more or less continue unabated throughout the rainy season. However, an isolated rainfall followed by a dry period will fill the pool, the fry then ready to hatch will hatch, and the pool will dry again, perhaps before maturation and reproduction occurs. Under these conditions, if all the eggs were to develop at the same rate, then the population would die out after the isolated rain event. Likewise, in periods of drought, no rain will fall for months or even years. Once again, if all the eggs were to develop at the same rate, then the population would die out when the eggs reached Diapause III during the drought. In one of the most striking examples of the persistence of life on our planet, Nothobranchius eggs mature at vastly different rates in response to the uncertainty of the environment. Within a given clutch, some eggs will remain at Diapause I for months or years, and likewise some at Diapause II and III. It is only those at Diapause III that have a discrete time within which to hatch, but in the other two states, the embryo may persist for enormous periods of time, bridging both erratic rain patterns or drought. By employing this delayed developmental strategy, some proportion of the eggs will be ready to hatch any time the rains come. This is a stunning and miraculous process. Aside from the spectacular beauty of the Nothobranchius group, participating and observing this life process provides an extra dimension to the joy of fish keeping not provided by other groups of fish.

There are some captive environments that can substanially affect developmental rates. It is well recognized that storage temperature has a profound effect on the rate and quality of egg development in Nothobranchius. For example, most fish hobbyists, and killie hobbyists in particular, know that the higher the temperature the faster the egg development. To the Nothobranchius breeder wishing to accelerate the "dry" time, this knowledge was exemplified by the practice of storing eggs on shelves near the ceiling, near a heat source, or in temperature-controlled storage containers such as "warm boxes" and more elaborate incubators. But for the new Nothobranchius breeder, we do not recommend altering the developmental cycle, but rather have the patience to wait out the natural processes. Temperature manipulation of the developmental process should be left to the experienced breeder.

Consider one caveat that applies to everyone, however. There is strong evidence that eggs left in the aquarium or spawning setup will not develop at the same rate as those in peat storage. (Levels, 1986) It has been shown that Both Diapause I and II can be induced when eggs are kept in the presence of adult females because of what are thought to be pheromone secretions. This finding is backed by hobbyist’s experience that eggs left in tanks with spawning adults show very little development even several weeks after being spawned. This is the reason why we recommend using the date the eggs are packed as the base date from which to calculate the wetting time.

There are an enormous number of Nothobranchius species to select from, but those most suitable to the beginning breeder are the members of the “Palmqvisti Group”, which all belong to the subgenus adiniops. Within this group, we recommend: foerschi, guentheri, korthausae and palmqvisti. All of these, except N. korthausae have red caudal fins, with korthausae sporting a striped yellow caudal. From the zoonothobranchius subgenera we recommend the spectacular N. rackovii with its orange and black banded caudal, and from the Nothobranchius subgenera, N. melanospilus, largest among this group, with a bright red caudal and caudal peduncle. All are relatively easy to breed, are beautifully colored as shown in the accompanying photographs, and thrive under aquarium conditions. The recommended incubation time for these species, at room temperature, is:

Species	Storage in Weeks
N. foerschi	12 to 16
N. guentheri	12 to 16
N. korthausae	12 to 16
N. melanospilus	12 to 16
N. palmqvisti	12 to 16
N. rachovii	20 to 28

ish Association and some online services.

Although Nothobranchius are not the easiest tropical fish to keep, they do represent one of the most rewarding, unusual and interesting challenges to the hobbyist. Try a Nothobranchius species or two, you will not be disappointed.

References:

Haas, R. (1969). Ethology and sexual selection in the annual fish, Nothobranchius guentheri.. Ph.D. thesis, Univ. Cal., Los Angeles.

Haas, R. (1976a) . Behavioral biology of the annual killifish, Nothobranchius guentheri. Copeia (1):80-89.

Haas, R. (1976b) . Sexual selection in Nothobranchius guentheri. (Pisces: Cyprinodontidae). Evolution, 30(3):614-622.

Kinney, J. A. S., S. M. Luria, and D. O. Weitzman. (1967). Visibility of colors underwater. J. Opt. Soc. Am., 57(6)::802-809.

Leibel, W. S. (1977). Killifish annualism: Old world annuals. J. Am. Killifish Assoc., 10(2):19-26.

Leibel, W. S. (1982). Review of recent observations on the phenomenon of annual egg diapause. J. Am. Killifish Assoc., 15(3):83-100.

Levels, P. J., L. J. H. van Tits, and J. M. Denunce. (1986). The effect of the presence of adult fishes, gonad homogenates, and embryo homogenates on the dispersion-reaggregation phase during early embryonic development of the annual fish Nothobranchius korthausae. J. Exp. Zool. 240:259-264.

Markofsky, Jules and A. Perlmutter, (1972). Age at sexual maturity and its relationship to longevity in the male annual Cyprinodon fish, Nothobranchius guentheri. Exp. Gerontol., 7:131-135.

Markofsky, Jules and A. Perlmutter, (1973). Growth differences in subgroups of varying longevities in a laboratory population of the male annual Cyprinodon fish, Nothobranchius guentheri. Exp. Gerontol., 8:65-73.

Nunziata, C. A., (1989) Water incubating Nothobranchius eggs. J. Am. Killisish Assoc., 22(1):3-22.

Terceira, A. (1974). Killifish, their care and breeding. Pisces Publishing Co. Rhode Island. 143 p.

All photographs courtesy of Anthony Terceira, Scituate, RI.
Copyright, Anthony Terceira, all rights reserved.

Click Here to go to TOP OF PAGE or click here to go to TABLE OF CONTENTS