WildNothos
THE NOTHOBRANCHIUS SITE
Collecting Nothobranchius – my special souvenir from Kenya 2007
By Béla Nagy
Author’s note: since this article was written, the populations of N. melanospilus from southern Kenya have been described as N. kwalensis by Costa in 2019.
From the end of the year 2006 until mid-January 2007 I spent some very pleasant days with my family in the coastal area of southeastern Kenya. As there are known habitats of the genus Nothobranchius in that area, I had planned to carry out a couple of days of field study and collecting while there.
Kenya is located in East Africa, the country straddling the equator. According to the findings of many archeologists, the beginning of the human evolution occurred in Kenya. The landscape of the country is characterized by the fault-lines of the Great Rift Valley, picturesque lakes, and imposing volcanoes. Many different forms of vegetation are present, from tropical rain forest, through savannah, to desert. Alongside the Indian Ocean lies the coastal plain which is about 310 miles (~500 km) long and between 12 and 50 miles (~20 and 80km) wide. Many rivers flow through the plain in a west-east direction, among them the two longest rivers in the country, the 440-mile-long (~708 km) Tana River and the 340 mile-long (~550 km) Sabaki River. The long white sandy beaches along the coast of Kenya makes this an ideal place for a holiday. The climatic conditions are characterized by periodic rainy seasons. Generally, in the coastal plain area the longer of two rainy seasons begins in March and ends in May, and the shorter rainy season begins in November and ends in mid-December. When the rainy season arrives, the rivers flow over the banks and inundate the floodplains. This periodic flooding is essential for the survival of the Nothobranchius species that occur there. The natural habitats of these fishes are the seasonal pools that dry out during the dry seasons.
Several species of Nothobranchius are known along the coastal plain of Kenya (Wildekamp, 1982, 1987, 2004; Wildekamp and Berkenkamp, 1979). D.J. Wilson collected the first identified Nothobranchius species in the vicinity of Mombasa in 1891. The specimens collected were sent to the British Museum in London, where the well-known ichthyologist, the late Boulenger identified them as Nothobranchius guentheri. We now know, however, that N. guentheri is present only on Zanzibar Island, and the species from Mombasa probably was N. elongatus (Seegers, 1985).
About 13 miles (~21 km) north of Mombasa, N. interruptus can be found in the swamp of Kikambala, in the drainage of the Mtomkuu River. The distribution of this species is somewhat restricted. West and northwest of Mombasa N. elongatus occurs in the drainage systems of the Manjewa, Kombeni and Rare rivers. The distribution of that species is wider than that of N. interruptus. In the northern part of the coastal plain of Kenya, in the vicinity of Malindi and northward from that city, habitats of N. jubbi are known in the drainage of the Sabaki River. In the drainage of the Tana River, N. jubbi, N patrizii, N. willerti and N. microlepis are found. These species occur sympatrically in different combinations in the same habitats. South of Mombasa, in the drainage of the Ramisi and Umbwa rivers, several localities of N. melanospilus and N. palmqvisti are known, often occurring sympatrically (Wildekamp 2004).
Southern area and vicinity of Mombasa
On my first collecting excursion I planned to visit the region south of Mombasa. From the hotel, which was situated north of Mombasa, I crossed the city and drove along the south coast. It was interesting to see that at 7.00 a.m. there were a lot of people on the road, most of them walking to their places of work in the city. Mombasa is situated on an island, connected to the mainland by two bridges, one to the north and the other to the east. However, to the south there is only a ferry crossing. According to the local people, it would be very expensive to build a bridge there, because the biggest ships have to reach the port through that channel. Three ferries were running across the channel and it did not take too long to reach the other side. First the cars were allowed to board the ferry, and then the remaining free space would be filled by bicycles and pedestrians. Interestingly, the transportation fee was only paid the cars. Accordingly, the ferry is always more than full with people. After crossing the city I drove along the southern part of the coast until I came to the Ramisi River. The tarmac road was in very good condition, although without any markings, and I was able to reach the river very quickly. Along this road there were numerous police control points, where the road was blocked by a 4-in (10 cm) high spiked barrier. Usually the cars had to slow down and sometimes the driver had to explain the purpose of the trip.
Between the Ramisi River and Lunga Lunga , at the border of Tanzania, numerous collections of N. melanospilus and N. palmqvisti have been made. Both species were collected by Holler in 1978, Willert and Forstner in 1983 (Willert, 1984, 1986a, 1986b) and 1985, Brown in 1989 (Brown, 1991a, 1991b), Fick in 1995 and 2001 (Fick, pers. comm.), and Watters, Cooper and Schmidt in 2001 (Watters, pers. comm.). In many cases these two species were found to be sympatric. When I crossed the Ramisi River, I started to look for suitable biotopes. There was water everywhere. The bridge over the river had been severely damaged by the flooding, and partly repaired only two days before I arrived there. As the whole region was under water, it was difficult to collect. I tried to collect Nothobranchius species in a number of potential habitats, but in most places I found only big frogs and tadpoles.
After the first unsuccessful attempts, I arrived at a pool situated in the shade of a big tree beside the road. I stepped into the pool and wet my net. After the former unsuccessful experiences, I could hardly believe what I saw: there was a female Nothobranchius in the net. As there were no markings on the body, it had to be a female N. palmqvisti. In the middle of the pool I caught only females. Where were the males? I tried with my net at the edge of the pool, where roots hung into the water, giving some protection to this part of the pool. Here I caught both males and females. Those parts of the habitat that seemed most hidden and protected yielded the largest and most colorful males. Directly under the big tree, among the roots, and in what was probably the most protected place, I caught a big male N. melanospilus. So, at my first confirmed locality I found two sympatric species. My next objective was to find some N. melanospilus females. Once I had collected enough specimens, I bagged them individually in clean water and packed them into a fish box.
After this first success I travelled toward the village of Lunga Lunga at the Tanzanian border. Just before I reached the village I turned northward. The first part of the dirt road to Kinango was in surprisingly good condition, despite the recent flooding. Along this road Fick had collected several times (1995, 1997, 2001, and 2005) from two localities (Fick, pers. comm.). The relatively good condition of this road did not last long because after several kilometers a river had destroyed the bridge. The water had flooded the road, making it impossible to cross, so I returned to Mombasa, toward some other locations north of the city.
There are two locations known from the vicinity of Bamburi. Arch Martin collected N. elongatus in the Conservation Park in August 1985 (Martin, pers. comm.). This population, which had been introduced into this habitat, did not survive more than a couple of generations because after some excavation work in the conservation park, the habitat became subjected to saltwater intrusion. The biotope was no longer suitable for the survival of the Nothobranchius species (Baer, pers. comm.).
Allen and Barbara Brown collected a Nothobranchius species in 1989, in a stream that had dried to form small pools, 5.2 km inland from the Bamburi Cement Factory (Brown, 1991a, 1991b). After they returned to the UK, the fish was identified by Ian Sainthouse as N. interruptus (Brown, 1991a, 1991b). This locality is associated with a small temporary river situated in the Nguuni Wildlife Sanctuary where, among other animals, giraffes and gazelles live. I obtained a special permit to collect there, but I only found cichlids and guppies in the pools of the temporary river.
I then looked for another locality near Bomani, where Karl Lung had collected N. interruptus in 1980. According to his directions, 5 miles (8 km) after the Mwtapa bridge one has to turn to the left and follow that road for 2½ miles (~4km) (Seegers, 1985). At that point I saw a heavily vegetated pool about 160 ft (~50 m) from the road. When I got out of the car to collect at this site the locals who were gathering nearby told me very emphatically to not go there, as a large snake inhabited the pond. I found it prudent to take their advice. So I moved on to the last collecting locality for the day.
Near the town Kikambala, about 13 miles (~21 km) north of Mombasa, there is an extensive swampy area approximately 325 feet (~100 m) across, on the west side of the main road to Malindi. This is a habitat of N. interruptus, from which several collections are known: Nagy in 1968, Holler in 1978, Willert and Forstner in 1983 and Willert and Langnickel in 1985 (Willert, 1984, 1986a, 1986b), Fick in 1985, 1995, 1998, 2001, and 2005 (Fick, pers. comm.), and Watters, Cooper and Schmidt in 2001 (Watters, pers. comm.). Walking around the swamp, I tried to collect at several places but was unsuccessful. I then stopped at the southern edge of the swampy area. In the shade of a big tree, I was able to capture a few male N. interruptus.
As always, the locals appeared immediately to watch my activities. By shouting “mamba, mamba” they informed me, that there was a crocodile in the swamp. Staying on the higher bank of the swamp, they had spotted the animal about 100 ft (~30 m) from me, lying only half in the water. They watched the crocodile, while I tried to catch some fish near the edge of the swamp. Obviously, collecting was not easy. I had not caught any female specimens but, because of the presence of the crocodile, I did not want to go into the swamp too far from the bank. I was standing there and did not know what to do.
One boy from among the spectators advised me to try a pool on the other side of the road. Beside the road the vegetation was dense and I did not see the pool. However, I saw a culvert under the road, and I thought there must be a connection with the swamp during the rainy season. The pool on the other side was not far from the road, but because of the vegetation I could only reach it by a detour of about 350 ft (~100 m).
While I was walking with the net, a very friendly man suddenly appeared in the front of me and welcomed me. He was happy that I was back to collect again. It was a funny situation, because I had never been there before, and our friend probably mistook me for one of the earlier collectors. When I reached the pool I found that it was close to drying out, as the water depth was no more than 4-8 inches (10-20 cm). Here I was able to collect a lot more of both male and female N. interruptus. I bagged them and took them back to the hotel.
Later, when I arrived back in Europe, I met Steffen Fick. He had collected at this last locality several times. At his last visit to the site he saw a young boy who had traces of a crocodile attack on his body. The boy had been attacked a few years earlier while swimming in the pool.
North of Malindi
During my second collecting excursion I visited the northern part of the coastal plain. Several collections of Nothobranchius are known from the area north of Malindi: Leakey in 1962, Whitehead in 1965, Holler in 1978, Lung in 1980, Willert and Forstner in 1983, and Willert and Langnickel in 1985 and 1989 (Willert, 1984, 1986a, 1986b), Hartig-Beeken in 1986, Seegers 1996, Fick in 2001 (Fick, pers. comm.), and Watters, Cooper and Schmidt in 2001 (Watters, pers. comm.).
Unfortunately, at the time of my visit the area was dry and all the potential habitats had been totally dried out. I tried to get further to the north, to the Tana River area but, about 34 mi (~55 km) from Malindi there was a military road block which I had already heard about. The area beyond this road block was dangerous, due to frequent attacks on travellers by bandits and the many Somali refugees who could also cause problems. After the necessary greetings at the road block, I made an agreement with the officer on duty that I would take an armed escort with me, and I was allowed to proceed.
After the road block, the area was still dry for about another 19 mi (~30 km). Then, suddenly, there was water everywhere. In many cases, the bodies of water alongside the road extended for several miles. At the road crossing to Garsen, we turned to the north toward Garissa. I tried to reach the type locality of N. willerti, which was approximately 0.6 mi (~1 km) from the crossing at Mnanzini (Willert, 1986). My guard told me that in order to travel beyond Garsen, officially, two armed guards are needed. The last robbery was two days before, when a car was attacked and the passengers killed. He agreed to proceed as far as Mnanzini, and he promised to protect me. The road beyond Garsen was still tarmac and, because of the very limited traffic, still in good condition. After about 12 mi (~20 km) the tarmac road ended abruptly and it became dirt again. From that point it was in a very poor condition and difficult to drive on. According to the news coming from this part of the country this was what I had expected. The guard told me that after heavy flooding during the last few months, the first car that could cross the area had arrived three days earlier from Garissa. After crossing several difficult sections of the road, we arrived at a road section we could not cross. The main road was in very bad condition and a short detour had been built some time ago. Unfortunately, neither the main road nor the detour was passable. According to the approximate GPS data, we were about 2 kilometers from the crossroad to Mnanzini. The guard knew the area very well and he confirmed this. As there was water everywhere I tried to collect at the point where we turned back and in several very promising biotopes on the way back, but I did not capture any Nothobranchius species. In a pool that had a substrate of cow dung I tried again and found only a Barbus species. A woman was washing clothes in this pool, in spite of the fact that it had a terrible odor. Every time we stopped, the armed guard immediately jumped out of the car to look around and secure our safety. It was fascinating to note that, despite the never-ending roads and the lack of villages, a lot of people moved around in the area. These were refugees from the many tent camps organized for the Somali exiles. These camps were also used recently for the refugees of the flooded areas. On our return to Malindi we arrived at the military block, where I left the armed guard and went back to the hotel.
West of Mombasa
For the third collecting excursion I travelled to the Mariakani area. From that area several collections of N. elongatus were known: Leakey in 1962, Fanara in 1973, Walpole, Greenal, Hemmingway and Wiggens in 1974 (Walpole, 1975a, 1975b), Holler in 1978, Willert and Forstner in 1983 and Willert and Langnickel in 1985 (Willert, 1984, 1986a, 1986b), Brown in 1989 (Brown and Brown, 1991a, 1991b), Sadler in 1996 (Sadler, pers. comm.), and Watters, Cooper and Schmidt in 2001 and Watters, Cooper and Bishopp in 2003 (Watters, pers. comm.). Mariakani is not far from Mombasa, but the condition of the road was terrible. This is the first part of the Mombasa–Nairobi main road. The road suffers from the countless trucks carrying goods from the port of Mombasa to Nairobi, and also to Uganda and Rwanda. On this road the average speed was about 12 mph (~20 km/h), so this short distance took two hours.
At Mariakani I turned to the south at the sign of the Mwawumbo Secondary School onto the C107 main road in the direction of Kinango. From here I drove about 2 mi (~3.5 km). On the right side of the road there was a fairly large swampy area. Watters, Cooper and Schmidt had collected at this locality in 2001, and they had found N. elongatus (Watters, pers. comm.). Here I did not catch anything. On the other side of the road I saw some water, about 165 ft (~50 m) from the road. Apparently, the whole area had been under water during the rainy season, but at the time of my visit there were only small pools remaining. At the first attempt I captured a female of N. elongatus. I could rapidly collect a suitable number of specimens, mainly at the edge of the pool. The edges were overgrown with grass and there was no true aquatic vegetation either floating or submerged. There were large, fully grown fish in very good condition, although not very colorful. After I had bagged the fish, I returned to Mariakani.
From Mariakani I went in the direction of Kaioleni, on the C107 road. I found a swampy area totally overgrown with sedge. I was able to just reach the edge of the swamp and cleared a small part of the pool by pushing aside the vegetation with my foot. There, within a minute, I found an abundance of N. elongatus. Taking into consideration the limited extent of the cleared area, there must have been thousands of specimens in the pool. I bagged a few pairs individually and the rest were returned to the pool. The fish of this population were young, very colorful and vigorous. Still at the same locality, there was a culvert pool on the other side of the road. I decided to try this pool and again caught males of N. elongatus. I was simply curious to see if there were any fish in the culvert pool so I did not keep any of them. Traveling a further 2.7 miles (~4.3 km) on that road I arrived at the Mwatsuma River. Here, Walpole, Greenal, Hemingway and Wiggens had found a Nothobranchius species in isolated pools formed in the river bed (Walpole, 1975a, 1975b). This time there was still a lot of water in the river. I made an effort to fish but only found the ubiquitous Barbus kerstenii.
In the area near Mariakani, I visited one more locality which I was especially interested in. From about 1988 until 1990 Sabine Baer, Rehabilitation Manager of the Bamburi Lafarge Eco Systems had, on numerous occasions, found Nothobranchius species between Mariakani and Samburu. The interesting part of the story is that aside from the relatively common, red tailed N. elongatus, she was able to capture a second species, N. microlepis. Sabine is still working in Kenya, cultivating the vegetation in the area destroyed by the Bamburi Cement Factory. I contacted Sabine before my arrival in Kenya and we met at the hotel. She showed me photos of the locality and the fish. This increased my curiosity. She told me that she had visited the locality about 20 times while carrying out a field study on lungfish. She often found N. elongatus there, but the other species, N. microlepis was found only three or four times.
Sabine kindly offered to show me the locality. On the road from Mariakani we crossed a bridge at Maji ya Chumvi. This name means salt water in Kiswahili, due to the somewhat higher salt content of the local water. Some kilometers after this bridge we turned south along a dirt road. The name of this area was Matope which means swamp. This area is very dry for most of the year but during the rainy season it turns into a big swamp. On that road we had to drive about 2.5 mi (~4 km). This was not an easy drive, because after a short disance there was a lot of water covering the road and a big Jeep was stuck in the middle of it. We could not pass the road where the Jeep stood, but we were able to drive into the savannah and bypass the difficult place. The 4-wheel-drive was very useful. Soon we arrived at a crossroad where we turned to the left and drove about 0.2 mi (~300 m) along that road until we crossed a small river. Sabine immediately showed me the dried out river bed behind the bush where she had previously collected N. elongatus and N. microlepis. In the deeper part of the river bed I found only Barbus species. In shallow pools formed in the river bed I again caught some N. elongatus. I walked in both directions in the river bed but I did not catch any other species. Near the river two concrete well rings were placed, with some water inside. I put my net inside and caught N. elongatus in there. After searching in this area for a while, I preserved some specimens and left the locality. Watters, Cooper and Bishopp had visited this locality in 2001 and 2003 and had caught N. elongatus in the same river (Bishopp, 2004; Watters, pers. comm.). On the way back to the main road, after several hundred meters, we stopped at a culvert along the road where I again collected N. elongatus. While I was looking for the fish, Sabine collected some plants she would use in her work.
When I arrived back at the hotel some hours later I put the fish bags on the table and observed that the males were still trying to fight with each other even though they were in separate bags. After a day of collecting I usually put a trio to spawn in some small plastic tanks with some peat I had brought with me. This also gave me the security of having some eggs in the event that the fish did not survive the journey home. The fish were bagged individually and daily water changes were carried out. All the fish arrived safely in Europe and after a few days of acclimatization most of them started to spawn in the aquariums.
Localities
1 – N. melanospilus Ramisi KEN 07-1 and N. palmqvisti Ramisi KEN 07-1
A temporary pool on south side of the A14 road, about 4.3 mi (~7km) from the bridge over the Ramisi River. The pool measured about 10 x 16 ft (~3 x 5 m), without any aquatic vegetation and only some decaying leaves on the bottom. The water was muddy and about 12-20 in (~30-50 cm) deep. The temperature of the water was about 81°F (27.2 °C), the pH was 6.5, and the water contained 100 ppm total dissolved solids. Date of collection: January 2, 2007.
2 – N. interruptus Kikambala KEN 07-2
An extensive swampy area on the west side of the Mombasa–Malindi main road and a small pool on the opposite side of the road, near Kikambala. The latter pool measured about 33 ft (~10 m) in diameter. The water was muddy and only 4-8 in (~10-20 cm) deep. The temperature of the water was 88.9°F (31.6 °C), the pH was 7.04, and it contained 640 ppm total dissolved solids. Date of collection: January 2, 2007.
3 – N. elongatus Mariakani KEN 07-3
A muddy depression on the east side of the C107 road, about 2.1 mi (3.5 km) south from where it crosses the main road in Mariakani. The pool was 33 ft (10 m) long and 6.5-10 ft (2-3 m) wide. The water was muddy and about 3 ft (~1 m) deep in the middle. The water temperature was 85°F (29.5°C), the pH was 7.82, and it contained 1140 ppm total dissolved solids. Date of collection: January 6, 2007.
4 – N. elongatus Mwatsuma KEN 07-4
An overgrown swampy area and culvert pools on both sides of the C107 road between Mariakani and Kaioleni, about 2.7 mi (4.3 km) before the Mwatsuma River (and in the vicinity of the village Mwatsuma). The water was muddy and about 4 in (10 cm) deep at the edge of the swampy area. The temperature of the water was about 85°F (29.4 °C), the pH was 6.64, and it contained 200 ppm total dissolved solids. Date of collection: January 6, 2007.
5 – N. elongatus Matope KEN 07-5
Several pools formed in a temporary river bed, between Mariakani and Samburu, 3.7 mi (6 km) after the bridge at Maji ya Chumvi and 2.5 mi (4 km) south of the main road. The water was muddy, and from a few inches to ~3 ft (1 m) deep, depending on the shape of the river bed. The temperature of the water was 84.5°F (29.2 °C), the pH was 7.26, and it contained 140 ppm total dissolved solids. Date of collection: January 6, 2007.
6 – N. elongatus Matope KEN 07-6
A pool at a culvert beside the dirt road to Matope, about 3.5 km south from the Mariakani-Samburu main road. The pool was about 3 ft (~1m) wide, the water was muddy and about 20 in (~50 cm) deep. The temperature of the water was 86.7°F (30.4 °C), the pH was 7.68, and it contained 220 ppm total dissolved solids. Date of collection: January 6, 2007.
Acknowledgements
Firstly, I would like to thank my family for their patience while I was engaged in collecting, maintaining, and breeding the fish. I want to express my special thanks to Sabine Baer (Bamburi, Kenya) for her very valuable help in the field, Susanne Schories and Manfred Schartl (Würzburg, Germany) for the official documents, Steffen Fick (Augsburg, Germany) for his helpful information about the localities, and Dr. Brian Watters (Nanaimo, Canada) for valuable information about the localities and for reviewing this manuscript.
References
Bishopp, B. 2004. Collecting in Tanzania, 2003. Part I. The Nekton. Saskatoon Aquarium Society, Saskatoon, Canada. 2004 (Jan.): 7-8.
Brown, A. and B. Brown. 1991a. Kenyan Killies. Aquarist and Pondkeeper, (1): 22-23.
Brown, A. and B. Brown. 1991b. Kenyan Killies. Aquarist and Pondkeeper, (2): 70-71.
Seegers, L. 1985. Prachtgrundkärpflinge. Die Gattung Nothobranchius: Systematik, Vorkommen, Pflege und Zucht. Deutsche Killifisch Gemeinschaft Journal, Supplementheft 1: 48 pp.
Walpole, M. 1975a. A Visit to East Africa in Search of the Genus Nothobranchius, Part 1. Killi-News. British Killifish Association, (124): 1-6.
Walpole, M. 1975b. A Visit to East Africa in Search of the Genus Nothobranchius, Parts 2. Killi-News. British Killifish Association, (125): 1-3.
Wildekamp, R.H. 1982. Die Nothobranchius Arten des Küstengebietes Kenias. Die Aquarien und Terrarien Zeitschrift (DATZ), 35 (9): 333–339.
Wildekamp, R.H. 1987. Notes sur le poisson annuel Nothobranchius jubbi Wildekamp and Berkenkamp, 1979 (Cyprinodontiformes: Nothobranchiinae) du nord-est du Kenya et de Somalie du sud. Revue française d’Aquariologie, 13 (4): 99–106.
Wildekamp, R.H. 2004. A World of Killies. Atlas of the oviparous cyprinodontiform fishes of the world. Volume IV. American Killifish Association, Mishawaka, Indiana: 368pp.
Wildekamp, R.H. and H.O. Berkenkamp. 1979. Untersuchungen zur Identität von Nothobranchius neumanni (Hilgendorf, 1905) aus Tansania, mit der Beschreibung einer neuen Art und einer Unterart aus dem Küstentiefland Kenias. Deutsche Killifisch Gemeinschaft Journal, 11 (5): 65–75.
Willert, M. 1984. Kenia 1983. Deutsche Killifisch Gemeinschaft Journal, 16 (7): 97-103.
Willert, M. 1986a. Neues aus Kenia I. Deutsche Killifisch Gemeinschaft Journal, 18 (4): 47-53.
Willert, M. 1986b. Neues aus Kenia II. Deutsche Killifisch Gemeinschaft Journal, 18 (5): 74–76.