Nothobranchius rachovii Ahl, 1926

Last updated 15 April 2001

Photograph 1: Nothobranchius rachovii {Photograph Valdesalici}


rachovii :(Lat.) = Named after Arthur Rachow, a renown pre-world war German aquarium pioneer.

First Description

Ahl, E. - 1926 - Neue oder selten importierte Fische III - Blätter für Aquarien und Terrarienkunde 37(17): 346-348, 3 figs.

Terra Typica

Beira, Mozambique

Meristic Data

D= 14-16; A= 15-16; scales in mid-lateral line series = 25-27

Karyotype: n= 9

According to Ahl's first description of the species in 1926, both the body and the head heights are comprised 3 times in the body length. The snout is short and large, somewhat shorter than the eye diameter; the mouth is directed upwards, the lower jaw protruding. The few teeth in the first row are larger, stronger, conical in shape and curved towards the back. Beyond this row the other teeth are implanted in a wide row and in the most posterior row the teeth are not large. The eye is comprised 3 1/3 tot 3 1/2 times in the head and 1 1/2 times in the interorbital width. The distance between eye and upper-lip is very short, the latter forms a turn ending in a depression near the eye. The dorsal fin has 15 rays, placed at an equal distance between the posterior border of the eye and the caudal peduncle; the longest rays 2/3 to 3/4 of the head length. The anal fin has 15-16 rays, placed almost opposite the dorsal fin. Pectoral fins are about 3/5 of the head length, reaching beyond the beginning of the ventral fins. The ventral fins are hardly located nearer to the caudal peduncle than to the snout tip or at an equal distance from both these points. The caudal fin is rounded. The length of the caudal peduncle is similar to its height. There are 25 to 26 scales along the longitudinal row over the body length and, just in front of the ventral fins, there are 22 scales around the body.



This species is considered to belong to the sub-genus Zononothobranchius.
Populations known to the hobby include:
N rachovii "Beira 58" = Aquarium strain (Roloff's import of 1958)
N.rachovii KNP Black "Kruger National Park" (Vermaak's collection of 1984)
N.rachovii "Beira 91" (Rosenstock's import of March 1991)
N.rachovii "Beira 98" (commercial collection of 1998 from the Beira area [pers. comm. B. Watters]

N.rachovii "MOZ 99/1 - MOZ 99/2 - MOZ 99/3 - MOZ 99/7 - MOZ 99/9 - MOZ 99/10 - MOZ 99/11" (Wood, Ippel, Kearney, du Plooy & Riley collection of March-April 1999)

N.rubroreticulatus is defined as the type species of the subgenus Zononothobranchius, to which the following species most probably also belong: N.rachovii Ahl, 1926; N.kirki Jubb, 1969; N.furzeri Jubb, 1971; N.brieni Poll, 1938; N.polli Wildekamp, 1978; N. malaissei Wildekamp, 1978; N.symoensi Wildekamp, 1978; N.taeniopygus Hilgendorf, 1891; N. neumanni Hilgendorf, 1905; N.lourensi Wildekamp, 1977 and N.korthausae Meinken, 1973 (Radda, 1981:4).






Males can reach a maximum length of about 55-60 mm. Females remain considerably smaller, they never become larger than 45-50 mm.



Distribution & Habitat

The entire distribution area of N.rachovii is not yet exactly known but looking at the dispersion of presently known collecting sites, one can obtain a rough general idea. The type material originated from Beira, on the Indian Ocean Coast of Mozambique. Further to the west, the species is also known from temporary pools found within the Pungwe river drainage system, near villa Machado. The southern border of the distribution area lies within South Africa, where a darker sub-population can be found in the region of the Olifants and Limpopo rivers, within the Kruger National Park, along the borderline with Mozambique. From the Kruger National Park, this sub-specie has also been disseminated in adjacent pools in order to minimise the risk of total disappearance of the population in periods of extreme drought. The northern border of the distribution area is unknown, but lies probably somewhere in middle Mozambique, the Zambezi River possibly forming the northern border of the distribution area.

Known collecting sites of N.rachovii in Mozambique & South Africa.


According to U. de V. Pienaar (1968), the original distribution of N.rachovii in South Africa was restricted to the Kruger National Park, more specifically to a shallow pool on the border with Mozambique. It was collected "very near the original collecting site of N.orthonotus, along the eastern boundary north of Nwanedzi. Several males and females were captured in a series of small pans in the sandveld around Pumbe picket, some 6 miles [10 km] north of Mtomene, and on the boundary with Mozambique". These pans form the only natural habitats in the Kruger National Park and in the entire Republic (of South Africa).

de V. Pienaar (1968) mentions that "the habitats of N.rachovii in the Kruger National Park are all rather shallow, natural pans with a sandy or pebbly substratum and almost choked with aquatic vegetation, particularly Cyperus spp., the water-loving grass Echinochloa holubii, Nymphaea spp. and other aquatic plants. The fishes appear to take refuge amongst the stems of water grasses and forage in the small patches of open water around the edges of the pan. It is probably here in the shallows that spawning activity also takes place and breeding seems to be later than in the case of N.orthonotus, i.e. during March and April. N.rachovii feeds like N.orthonotus, mainly on tiny aquatic animals such as Ostracods and other micro-crustaceans, as well as small insect larvae, particularly those of mosquitoes."

As the natural distribution area of N.rachovii in the Kruger National Park was limited, both in size and in extend, and because these few locations were lying in a drought-stricken area, it was decided to introduce the species into several similar habitats located in the northern part of the Park, between the Limpopo and the Luvuvhu rivers.

In the 1978 edition of "The Freshwater Fishes of the Kruger National Park", de V. Pienaar (1978) mentions that "on the 26th April 1976 a total of 30 N.rachovii were netted in the pan at Pumbe picket and ferried by helicopter to the northern Nyandu sandveld where 8 and 10 were released respectively in two small pans immediately east of the large Mathlakuza pan and a further 12 in Klawer pan. A resurvey in during March 1977 failed to establish the presence of any of these rare fishes in the new habitats but this may have been due to the large bodies of water and not necessarily to an unsuccessful colonization". He further mentions that "on 22nd February 1978 a further batch of 250 N.rachovii were captured in the original Pumbe habitat, transported by helicopter and released in the Nwambiya and Shirombe pans in the northern Nyandu sandveld. With this latest effort there is all hope that this rare and endagered species will be firmly established in at least one additional habitat in the Park and the Republic of South Africa" (de V. Pienaar, 1978).

Original and introduction sites within the Kruger National Park, Rep. South Africa {after U. de V. Pienaar, 1978}.


Bowmaker et al. (1978) mentions the occurrence of N.rachovii in the Limpopo River drainage basin, in the Pongolo and Incomati rivers and in the Buzi and Pungwe rivers.

According to Bruton et al. (1982), in South Africa (Kruger National Park) N.rachovii inhabits shallow pans with dense vegetation and a sandy or pebbly bottom. The eggs are capable of surviving habitat desiccation. In their natural habitat they feed on benthic invertebrates and mosquito larvae.

Langton (1994:17) mentions that Vermaak & S. Grant collected Nothobranchius orthonotus Pongola blue around 1984 in the Pongola area of northern Natal, South Africa, together with N.rachovii KNP black and N.orthonotus KNP red in the Kruger National Park on the border between South Africa and Mozambique.

During end of March beginning of April 1999, Trevor Wood, Johan Ippel, Pieter Kearney, Jan du Plooy and Peter Riley travelled to Mozambique and collected several colour forms of rachovii in the middle and lower Limpopo drainage system (including the Olifants River). The various location codes where rachovii was found during that trip include: MOZ 99/1, 99/2, 99/3, 99/7, 99/9, 99/10 and 99/11 (Wood, 2000:115-119).

Collecting sites of the 1999 Mozambique collecting trip [After Wood, 2000:116].


N.rachovii MOZ 99/2 {Photograph B. Morenski}

N.rachovii MOZ 99/10 {Photograph T. Wood}


An African mega-wildlife park

The governments of South Africa, Zimbabwe and Mozambique have signed an agreement on 12 November 2000 by which the largest African cross-border conservation park was created to promote bio-diversity and ecotourism. This mega Park would cover 35.000 km [14.000 miles] - an area larger than Portugal. The new Gaza-Krüger-Gona-Re-Zhou Transfontier Park will straddle the borders of the 3 countries. It will comprise the Krüger National Park in South Africa, the Gona-Re-Zhou National Park in Zimbabwe and an area in Mozambique known as Coutada 16. It is expected that part of the excess wildlife will move from South Africa to Mozambique where it has been decimated by wars, disasters and pouching [Herald Tribune 11-12 November, 2000: 2 & De Morgen, 13 nov.00:12].

N.rachovii MOZ 99/3 {Photograph B. Morenski}



The largest specimen available to Ahl (1926) had a total length of 48-mm. The discovery location was Beira in Portuguese East-Africa (presently Mozambique). Ahl described the species on the basis of three specimens, made available by A. Rachow, and one specimen provided by the Berlin Aquarium.

Nothobranchius rachovii is surely not a newcomer into the hobby. It has already been known since 1926, when it first was imported into Germany where it was described by Ahl on the basis of 4 specimens, three belonging to Rachow and one to the Berlin aquarium. In those early days, the species did not survive for long in the hobby as its reproduction was considered, even by experienced breeders, to be more of a gambling game: would something hatch or not? According to Seegers (1985), this pre-war population did not survive the First World War.

Two new imports, dating back to the fifties are however known: one by Hansen and the second by Roloff. Hendrik Hansen (1957) is mentioning the first post-war import. In an article published in "The Aquarium" (1957), he mentions to have caught N.rachovii together with N.kuhntae. The exact collecting site is however not known but would most probably lie some 100-150 km (about 50 miles up-country) westwards of Beira. Initially, he could not catch himself any Nothobranchius, but after a couple of days, a native woman brought him, in an entirely rusted can, a handsome male specimen of N.rachovii. She had collected it in a pool, full of trash and rusty cans, next to her hut. HANSEN hardly could believe it. The water was perhaps 18 inch (50-cm) deep and very turbid. In addition to N.rachovii and N.kuhntae he also caught a lungfish in this pool. Hansen further described that next to nearly every hut there was a pool for catching-up rainwater during the rainy season. In pools containing clear water, he could not find any Nothobranchius; only pools with very turbid water contained Nothobranchius. Over 1 000 fishes were collected, but because of long transportation times, only a handful fishes made it live to the USA. According to Scheel, one also failed in that time to establish an aquarium population out of this post-war import.

In mid-July of the year 1958 [usually this period is already too late in the year as most pools have dried-up], E. Roloff imported once again the species from Beira in Mozambique. This time he collected it about one mile (1.5-km) outside the city borders. Roloff discovered N.rachovii together with N.kuhntae along the road to the airport, in an almost dried-up road ditch. It still contained water in its deepest parts. The bottom was covered with humus and half-decayed vegetation whilst the edges were covered with flourishing reed. In many pools, the water level was so low that beautifully coloured N.rachovii males were awaiting dead, lying on their sides. During the rainy season, this entire lowland area forms a freshwater sea and the fish can spread everywhere. At collecting time, the water temperature was 20C (68F) in the morning and increased during the day to reach 25C (77F). The species N.rachovii and N.kuhntae lived mostly sympatric and each species did not appear to be restricted to its own pools.

In Beira, the coolest temperatures (average of 22.5C - 72F) can be found in July. During the hottest period (November-December) the average temperature reach 27.5C (82F). In general, pools start to dry-out in May or June. In Beira, the rainy season starts in November and by end-June-July, most pools where N.rachovii lives have dried-out (Roloff, 1962). In their natural habitat, the eggs of the coastal populations lie thus for approximately 5-6 months in/on wet to dry soils. The area around Beira, rich in swamps and marshes, generates a local microclimate by which the eggs, enfolded in nocturnal foggy clouds and early morning dew, are being protected from complete desiccation. Dr. Geisler analysed a water sample provided by Roloff and found a temporary hardness of 6.1DH and a total hardness of 8.3DH. The pH-value was 7.1. An explanation for the high salt content of 11 mg/l can be found in the geological soil structure, which around Beira is made of a salty "red clay". Dr Geisler (in Bertholdt, 1962:473) reported that at the beginning of the rainy season, when the eggs of bottom spawners hatch, the water is much softer: DH of 1.0. The collector Hansen (Bertholdt, 1962:476) reported that the water temperature of the pools in the neighbourhood of Beira fluctuated in the month of July, during the coldest period of the year, between 68 - 77F [20 - 25C]. During the hot season the water temperature could rise to over 90F [32C].

Presumed distribution area of N.rachovii in Mozambique & South Africa.


According to Scheel (1971) and Seegers (1985) the present aquarium-strain of N.rachovii still originates from Roloff's first post-war imported fishes. For those of us who likes codes, this ancient population could thus be defined as "Beira 1958". This year, she could also have reached her 40 years in the hobby, which easily could correspond to over 60 rachovii generations. After all those years, the ancient "Beira 58" aquarium strain still looks the same as newly caught wild specimens of the "Beira 91" and "Beira 98" imports. This tends to indicate that over the years hobbyists have done a pretty good job in preserving the original colours and forms. There is indeed little to improve on this beautiful species.

The N.rachovii Kruger National Park (KNP black) population presently in the hobby originates from fish and eggs collected originally by Vermaak in 1984. There have been some suspicions that in recent years few hobbyists may have crossbred the KNP population with the N.rachovii "Beira 91" and/or the ancient aquarium population but this has not been properly documented and is probably insignificant as only a handful of dedicated hobbyists maintain this rare species. According to Watters (1998), the N.rachovii KNP black population is only known from the Kruger National Park. He further mentions that Nothobranchius occurring in South Africa (N.rachovii and N.orthonotus) are on the List of Protected Species and as such are not allowed to be caught, kept alive and bread in tanks, let alone exported without a valid permit. According to Watters (1998), it appears very unlikely that special permits be granted to re-collect this fish from the wild in the Kruger National Park, therefore every effort should be made to maintain it in the hobby. In addition, its natural habitat in the KNP is in a precarious condition as its lies within a drought stricken area.

In 1991, Rosenstock collected, on March 24th, some specimens of rachovii and kuhntae, close to the Dom Carlos Hostel, located where the road from the airport [about 4 km from the airport] enters the town of Beira. The swampy area consisted of a low-lying pan protected from the coastline by sand dunes. It contained numerous rice fields. The water temperature reached 31C (88F). Originally, he only could find N.kuhntae. N.rachovii was found along the first earthen path, on the left side coming from the Dom Carlos Hostel and past an ancient sawmill and a couple of huts on the right side. Where the rice fields started, the water was shallow, 5 to 15-cm.

In the past, it was generally accepted that the fish had to be kept in very soft and light acidic water. Analysis of water originating from the natural habitat disclosed that the pH-value was around 7, thus neutral water, with a total hardness of 7.0 GDH.


Ahl (1926) who based his first description on alcohol preserved specimens gave following description of N.rachovii: "the colour of males is brownish grey; on the side of the head there are a few diffuse carmine red stripes or worm-like drawings; over the entire body there are some 12 thin carmine red crossbars, running obliquely from the posterior back towards the front belly. Dorsal and anal fins are bluish with large red sports placed in rows or in red stripes; the caudal fin is, in its frontal portion, bluish with large red spots, which can be organised into crossbars, and whitish in its distal part, with a somewhat darker border. The ventral fins are transparent, the pectorals darker at their tip". Apparently, besides from presenting some incorrect and incomplete descriptive characteristics of the species and also omitting a description of the female, the first description does not fit the present mode of describing Nothobranchius species. It is however, to date, the only valid description of the species.

In live males, the blue-green iridescent base colour over the flanks extends into the dorsal and anal fins and also halfway through into the caudal fin. Depending on the light reflection, this reflecting blue can appear light or deep. The scales are edged with orange-red resulting -especially in the distal part of the body - into the appearance of some transversal orange-red bands over the body. The small external edge of the caudal fin is brilliant black, which contrast heavily with the inner orange band. The red to red-brown and even black dots, which are present on the dorsal, anal and caudal fins vary in number, size and coloration.

The South-African population originating from the Kruger National Park differentiates itself from the Beira population in so far that the colour is far less brilliant and the spots on dorsal, anal and caudal fins are smaller but also more numerous. This latter form is much less known as aquarium population (Wildekamp, 1979).

According to U. de V. Pienaar (1968), "the males" of the Kruger National Park population "are particularly brilliantly coloured and display a bluish-green iridescence over the dorsal surface and a shimmering greenish-gold lateral surface. The iris is golden and the throat and chest suffused with crimson. There is usually a series of irregular slanting crimson bars over the lateral surface and the pectoral fins are tipped with white. In breeding males the body becomes very dark blue, almost black. The caudal fin of mature males has a very distinctive sub-terminal, broad, bright red crescent bordered distally by a marginal deep blue band. The dorsal and anal fins also have blue marginal zones and are spotted with irregular dark blue or indigo blotches."

The N.rachovii MOZ 99 populations [MOZ 99/1, 99/2, 99/3, 99/7, 99/9, 99/10 and 99/11.] present relatively little variation in spite of the fact that they were collected over a fairly wide area. One major noticeable difference is to be found in the caudal submarginal band that in some populations is red-orange whereas in other populations it is either absent or only slightly developed or extends over only 1/2 of the caudal fin or can even be of a pale blue colour. As with many other species that display a wide distribution area, a progressive change in colour pattern from one end of the range to the other can be observed. This is because populations from one end of the range may not easily crossbreed with populations at the other end, behaving therefore as if the populations living on the edges could, at times, be even considered as different species.

The females are pale brownish-yellow, without any particular markings. The belly is silverish above which a faint light blue reflection can be seen. Fins are all colourless and transparent. According to U. de V. Pienaar (1968), "the females" of the Kruger National Park population "are drab in contrast and a more uniform olive or yellowish-brown with a slight greenish iridescence."

Maintenance and Breeding

Those of us who want to maintain this utmost handsome fish must definitely not be disappointed if it passes away after only 5 to 6 months. In seasonal fishes, all major biological functions unwind at a much faster paste. They have to utilise their short lives, which in nature last for about 5 to 6.5 months, to secure that enough offspring's are produced. They have to grow and produce eggs as long as there is water in the pools/swamps, the ditches and/or rice fields at their disposal.

Maintenance can sometimes cause problems; particularly the sensibility to velvet (Oödinium) can be important, but this is also valid for most other Nothobranchius species. One can partially prevent this by adding a teaspoon of kitchen salt per 10 litres tank water.

The reproduction of the species can also at times cause trouble. Not that they don't produce enough eggs, healthy females can have a large egg production, but the hatching percentage can be at times very variable.

Also, males can be at times aggressive towards their kin's and in that case it is better to separate them or transpose them into larger tanks before casualties are allowed to occur. After having spent some time in a tank, each male establishes his own territory, which he eagerly defends against intruders. Relatively speaking, N.rachovii is not as aggressive as many other Nothobranchius but they certainly can be aggressive to the point where females can suffer fatal damage. Wild specimens generally tend to be more aggressive than aquarium-raised specimens of the same species. In general, the best way to deal with aggression is to raise and breed them in groups (e.g. 5-6 males and 10-12 females) and do not at any time separate them. Taking one male out of a group and placing him back say a week later will assuredly create trouble. Raise them together as fry and as they grow and mature remove only those that aren't needed or that aren't wanted until all that are left in the tank corresponds to the chosen breeding group.

In a group the aggression is spread around although the less dominant males and smaller females will be picked on and the displaying amongst males will draw their attention away from the females to some extent. Another advantage of group spawning is a larger gene pool but of course, this does also mean that larger tanks are needed (25-40 litres or 10-15 gallon tanks).

Roloff (1962) collected eggs laid during the transport from Beira to Germany and left them into water to look what would happen if they did not undergo a dry period. On 08.IX.1958 [after only 5-6 weeks] the first young hatched. However, all but a few were belly-sliders.

According to Scheel (1959), the eggs are not quite circular and they range from 0.95x0.85 mm to 1.05x0.80 mm.

This species prefers a species-specific aquarium, with dark bottom cover and richly furnished with immersed wood chunks. This dark to shadow-rich environment definitely enhances colour contrasts. In a community aquarium, and especially when introduced whilst too young, the fish will stay put and will starve to death letting the other inhabitants get away with the food. The preferably neutral water will have to be changed regularly as the metabolism of these fishes is quite high; weekly partial water changes of about 10% are considered appropriate.

Bertholdt (1962:473) reported that he preferably bred N.rachovii in water with a hardness of 6.0 DH and a pH of 6.2-6.5. He boiled the peat moss for three to five minutes, leaving sufficient acid matter in the peat. "If the peat moss is boiled out completely, practically all the acid is released, and the moss loses its capacity to kill off the harmful bacteria which destroy the eggs of the bottom spawners". Bertholdt (1962:476) reported on his first breeding experience with the species: Initially, he intended to wait three months before wetting the peat moss, stored in plexiglass receptacles. Because of lack of time a full year passed. He was afraid that the long waiting time had killed the embryos but was surprised to find that the five dishes of peat moss produced 35, 61, 75, 80 and 87 healthy fry, with no single belly slider. This happened in spite of the fact that (or most probably because of) the eggs had been stored in the dishes in a room where the temperature dropped for six weeks to 59F (15C). In addition the peat moss with the eggs had been dried out to such a degree that he feared the eggs had been killed.

These days, the aquarium population of N.rachovii can be maintained and reproduced in ordinary tap water with a pH of and around 7 and a total hardness between 11 to 18GDH. During the day, the maintenance temperature would preferably be maintained in the vicinity of 22-24C (72-74F) and could be allowed to drop at night to around 19C (66F). After a week of spawning, the peatmoss containing the eggs is removed, slightly dried and poured into a plastic bag which is then sealed, keeping at least 3/4 of the bag volume filled with clean air. The 1-mm eggs are transparent but not very difficult to localise in wet peatmoss. The moisture content of the peatmoss containing the eggs should not be allowed to fluctuate too much during incubation. The sealed and labelled plastic bags are kept in a dark place were the temperature will be maintained around 22-24C (72-74F). The incubation time is very dependent on the storage temperature. Good hatches will usually be obtained if incubation temperatures are kept between 20 and 26C {65 to 80F}.

All further steps are now a question of patience. The incubation time of N.rachovii eggs lasts between 5 to 7 months and requires several hatching phases before one can ascertain that all eggs contained in the peat have hatched. Treat (1995), and we can confirm the good results of this methodology, pours for the first time water onto the peatmoss after only 4-5 months incubation. He then only gets a limited number of fry but hardly any belly-sliders, provided soon after these first young have hatched and been placed into a container with water of similar quality, the remaining peatmoss with unhatched eggs is removed from the water. Him (1995) also mentions an incubation period of only 4 to 5 months at a conservation temperature of 24 to 27C {77 to 80F}.

Hatching the fry happens by pouring either aged or freshwater onto the peatmoss containing the eggs. The hatching water should however be chilly, having a temperature between 16-18C (61-64F). The addition of salt {1 teaspoon of salt per 5-10 litre water} will help combat velvet disease and increase the longevity of the baby brine shrimp that one will have given as food, but this is not absolutely necessary. Treat (1995) even adds 1 to 2 drops of Liquid Fry for egg-layers at the moment of peatmoss immersion. This ensures a faster development of infusoria for the fry. At hatching time, the water layer, which comes to cover the peatmoss can easily have between 8-10 cm depth. With such a water layer, I always have had good hatching results after between 1 to 24 hours immersion. This large water volume renders faster water changes superfluous and dissolves to a greater extent any salt and waste products. In addition, at the bottom where the eggs await hatching, the necessary oxygen shortage by which the developed fry break through the eggshell is created at a much faster paste; not entirely developed embryos remain however quite in their resting stage. A light aeration of the water can happen but is not absolutely necessary. Him (1995) also indicates that the newly hatched N.rachovii fry are small and should first be fed with infusoria and/or micro-worms {this is indeed the case if the first supplied freshly hatched brine shrimp nauplii are too large to be eaten by the fry}.

Rearing of the fry usually does not cause any further problems. The small fry grow fast and after nearly 2 months the first fish become sexually mature. Fry mortality rate is high if one does not change water on a very regular basis. Him (1995) mentions that obtained sex ratios tend to favour females. My experience however tends to indicate that the sex ratio is in favour of males when the first months of egg incubation happen at around 2O-24C (68-74F) {during winter months} whilst it favours males when the first months of incubation take place at higher temperatures {26-28C - 79-82F), e.g. during summer months}.

In order to obtain a good egg production, one will have to separate sexes during growth and wait until the fish have grown stronger and larger before starting to breed. If this important aspect is systematically neglected, after some generations, the females will grow much smaller than males and remain behind in strength. Seasonal fishes, because of their high metabolism, require a strong, adequate and regular food supply in order to keep their energy demanding egg production at full capacity. If food is in short supply, the fish will soon perish. Treat (1995) also favours the separation of genders at the moment the males become sexual active and display their first colours. Smaller females and later males stand then better chances to grow to full size.

Although most authors mention that the addition of kitchen salt to the water [1 teaspoon per 5 to 10 litre water] is advisable as a preventive measure against the occurrence of velvet, I experienced that the best way to be permanently released from velvet, even if it is already active in the tanks, is to secure a sufficiently important water movement in the aquarium {this is achieved with a simple power-head pump}. This water movement should of course not be too strong as Nothobranchius are in general poor swimmers. With an adequate maintenance and a water temperature of 22-24C (72-74F), the fish can easily live up to one year; Roloff (1962) has kept rachovii that reached 15 months of age.

Four main set-ups can be used for reproduction: (i) keep a trio [2 females and 1 male] in permanent spawning, although it is advised to use more females {up to five} per individual male; (ii) separate the genders and bring them together every couple of days; (iii) keep all females together in a large tank with a few males and change these regularly or (iv) raise a group (with more females than males) together in a larger tank and never change anything to the numbers present. Treat (1995) doubled the collection of eggs by bringing the fish together for only 50 minutes after a one-day separation. He came to this strategy by assuming that females, because of their high metabolism demands, were also very devoted eggs-robbers.

As a spawning substrate one should foresee a box wherein a 10 mm thick layer of peatmoss can be introduced. This device helps, to limit the use of peatmoss to a strict minimum, to concentrate the eggs and to avoid contamination of the eggs with waste from the tank. Treat (1995) uses for this species always freshly boiled and properly rinsed peatmoss because according to his experience, the egg revenue is lower when using recycled peatmoss. One can leave the fish to spawn for between 3-7 days, after which the peatmoss is removed, gently pressed to reach the known tobacco consistency [peatmoss containing eggs should not be allowed to dry out too much] and put in a plastic bag for further incubation.

At immersion time, after 5-6 months incubation at 22-24C (72-74F), the peatmoss cum eggs is first poured into a container after which cooler {18-20C or 64-68F} water is poured over it. The floating material is well rubbed between the fingers to allow the eggs to sink. The peat which still floats after above operation is removed manually whilst the sunken part containing the eggs is poured into a 2 to 5 litre tank and flooded with a 4-6cm deep water layer. Hatching can sometimes take up to 24 hours before the first free-swimming fry appear. The remaining peatmoss containing the not yet completely developed eggs is again stored away for an additional 6 weeks period. After which time a new batch of fry will surely hatch. According to Schrey (1976), consecutive hatching attempts of fry can, expressed as a percentage, present following values for the hatching rates: 100 : 60 : 30 : 15: 0, whereby the first hatching attempt is measured at 100%. As the newly hatched fry are very small, infusoria-promoting products, such as 1-2 oak leaves, can be added to the hatching water one week before it is used for hatching. Wildekamp (1979) mentions that the incubation period for the eggs can vary between 6 and 9 months, depending on several external factors.

The N.rachovii Kruger National Park (KNP black) population is a tough one and it takes at least 7 months or more to develop. According to Watters (1998), the eggs of the KNP population can even take up to around 3 years to develop. It is thus always a good idea to rewet the peat a month after the first wetting in order to obtain more fry.

The young grow rather rapidly and one can regularly raise the water level during consecutive water changes. After 5-6 weeks, one can identify and separate the sexes and two weeks later they reach sexual maturity. The fry is particularly active and always in motion looking for food.

Treat (1995) mentions that the occurrence of mortality rates over and above 50%, even up to 100%, constitute no exception during the rearing of the fry. Also, the older the fish, the more sensitive they become to radical water changes and drastic pH variations. An entire water replacement can easily be accomplished with young fish whilst with older specimens, the changes have to be much more partial and also more gradual.

Because of the very high metabolism in N.rachovii it is absolutely necessary to provide on a daily basis strong and sufficient food, as otherwise the fishes will be wasting away very rapidly.

According to Chan (1976) the incubation period should definitely not be lower than 6 months; it is only after this period that good hatching results can be expected. The incubation temperature can, according to him, be best maintained between 18-25C {65-75F} as the eggs develop well within this temperature range.