Kushangaza at Halembe in Lake Tanganyika, and 
Observations on Brichard’s Tropheus

Kushangaza – A colourational anomaly
All true Tropheus species of Lake Tanganyika in Tanzania have been observed to include individuals that possess a colourational anomaly, a chromatic mutation that usually comprises a mixture of mostly black, yellow and orange colours, which is sometimes in combination with the normal colouration of the species, or variant. An individual of the tribe Tropheini with such an anomaly is locally and commonly known as a Kushangaza, meaning ‘surprisingly’ in Swahili. The term was coined in 1990 and applied to the first Tropheus individual discovered at Udachi (Q7) which was surprisingly golden and, was therefore named “Golden Kushangaza” {see picture here; see video clip here}. In southern Tanzania, Tropheus moorii “Red rainbow” is no exception to variants known to occur in the form of Kushangaza {see picture here}.

Fig. 1. A Kushangaza form of Tropheus sp. "Crescentic" just south of Halembe. Mutant coloured individuals are rare in this locality. The rocky shore south of Halambe village harbours the southernmost population of this species. The name "Crescentic" derives from Pierre Brichard’s observations of Tropheus in the late seventies and in the eighties.
Fig. 2 (left). Females and sub-adults of Tropheus sp. "Crescentic" may show a semi-striped colour pattern as this individual, photographed in the shallows south of Halembe village. Males are usually solid greenish brown. Fig. 3 (right). Villagers' canoes in the small creek at the centre of Halembe village. 

Tropheus brichardi
At Halembe (L5), about 35 kilometres (km) south-southeast of Cape Kabogo (J4) in Tanzania, a Kushangaza variety of the local and common Tropheus species is quite conspicuous, but also extremely rare. Individuals appear with exclusively black and yellow colours in a rather irregular combination (Fig. 1). The normal colouration of this Tropheus species comprises an all greenish brown, or dark-brown body, with a small round part of the chest being yellow, or yellowish. This Tropheus species is commonly identified as Tropheus brichardi. However, this is unlikely a correct identification. What may be called the real T. brichardi exists only at Nyanza (E3) in Burundi (the type locality) (Nelissen and Thys van den Audenaerde 1975) and, by all accounts, along the shores southwards from Nyanza (or Nyanza Lac sensu Brichard 1989) to the rocky habitat just north of the estuary of the Malagarasi River (I5) in Tanzania (Brichard 1989: 154; pers. obs.), as well as along the shores just north of Nyanza Lac (Brichard 1989: 168; 183, map). The colouration of the variants found from the border of Burundi and Tanzania to the Malagarasi River varies only slightly; earlier, these were together referred to as the ‘striped variant’ (Staeck 1985: 114). However, T. brichardi seems to be absent from the rocky habitat along the shores of Gombe Streams National Park (G3), north of Kigoma (H3) (Seehausen 1993: 170). Further on, the Tropheus variant found at Rumonge (D3), 25 km north of Nyanza Lac and across a sandy and swampy geographical barrier, belongs to another species (Brichard 1989: 162, 190).

Fig. 4. Tropheus brichardi at Masaka Point, Bulombora (I4). Southern populations of Tropheus brichardi seem to be the most colourful. The Tropheus brichardi from Tanzania (Kigoma variant) was discovered by Wolfgang Staeck in 1974 and first exported by Klaus Grom in 1975.
Fig. 5. The steep coastline between Bulombora and Moyobozi villages exhibits a beautiful reddish colour.
Fig. 6 (left). Just north of Moyobozi villgage (I4) Tropheus are rare due to suboptimal habitat. In the picture is a male individual in very shallow water. Fig. 7 (right). The southernmost population of Tropheus brichardi is found just south of Moyobozi village in Tanzania. The habitat is rich in sediment due to the Malagarasi River estuary, located just 20 km further south. Picture captured in 3 metres depth.

Tropheus sp. “Crescentic”
The Tropheus species at Halembe is here referred to as Tropheus sp. “Crescentic”. It was probably first identified as a distinct species by Brichard (1989: 158, table; 166, 169f; 180, table). He reported on one of its features being a “higher, longer, and crescentic tail” (1989: 166); thereby, its provisional name: Tropheus sp. “Crescentic”. However, Brichard did not apply such a descriptive name. He applied a rather contradictive name when referring this identified line (or lineage = descendants from an ancestral species) to the “Unidentified line”, and added a “line of unclear taxonomic status”. He did, however, list three rather distinct races (variants) of this line (here: species): the ‘Ubwari race’ in which the “tail is long and crescentic” (1989: 170); the ‘Yungu race’, whose “tail is crescentic and very long and high” (1989: 170) and of which the geographical distribution of 130 km covers the coast between Kasimia Bay (F2) (south of Ubwari) and Toa Bay (Mtoa Bay (K3); the area of Kavala Islands (K3)) (1989: 170); and the ‘Kavimvira race’ (Brichard 1989: 159, fig. second at the top, left; 169f). Brichard’s documentation of this species only concerned the geographical variants he found along the Congolese coast from north of the Kavala Islands and northwards via Yungu (H2) to the Ubwari Peninsula (E2), and at even further northern localities, such as Kavimvira (A2) (also known as Kamvivira, Poll 1952; Kullander and Roberts 2012: 372) in the extreme north (1989: 170). But the similar variants of this species present along the Tanzanian coast opposite the southern part of the mentioned Congolese coast are near identical, with no obvious specific differences, and therefore highly likely conspecific. However, Brichard did not recognise this, and he failed to realise that all of the greenish or yellowish dark-brown Tropheus variants found from south of Malagarasi River, and all the way southwards to Halembe, also belong to his ‘unidentified line’, i.e. T. sp. “Crescentic”, and not to T. brichardi (see 1989: 168–169).

The probably most distinguishing character of T. sp. “Crescentic” is unsurprisingly the crescentic tail fin with conspicuous pointed tips (Fig. 8). Even though it is not as crescentic as the one found in T. polli (Fig. 27) and possibly T. annectens, it is clearly distinct from the tail fin found in all other Tropheus species superficially resembling T. brichardi. It seems that T. sp. “Crescentic” also has a different body morphology, being slightly less deep-bodied posteriorly, and possibly having a slightly different head profile, as well as the head potentially being a bit larger, possessing a different colour of the inner eye (at least the Tanzanian variants), plus the tail fin being not only crescentic but also larger than many other Tropheus species, particularly T. brichardi. Furthermore, the vertical lateral stripes are seemingly thinner than those possessed by T. brichardi, and instead are reminiscent of those found in T. moorii. More or different characters appear to separate it from the other species similar to T. brichardi, e.g. a shorter body length. However, these observations are only provisional, pending an analysis.

Fig. 8. Along Kungwe Bay Forest Reserve between Segunga and Halembe live populations of Tropheus sp. "Crescentic" exhibiting a rather bland colour pattern of yellowish to greenish brown. The picture shows an adult individual at Kungwe Point.
Fig. 9 (left). A Kushangaza coloured individual of Tropheus sp. "Crescentic" in shallow water at Kungwe Point. The individual is showing only slight abnormal colouration. Fig 10 (right). Individuals with black markings are found in every Tropheus population. The picture shows an individual of Tropheus sp. "Crescentic" at Kungwe Point.

Tropheus sp. “Lukuga”
Moreover, regarding the field observations of Brichard (1989), Tropheus sp. “Crescentic” is partly sympatric with another undescribed Tropheus species along the Congolese coast. This is Tropheus sp. “Lukuga”, a name which also derives from the identifications of Brichard, who referred to it as the “Lukuga (Kalemie) line”, or the “Kalemie line (or Lukuga line)” (1989: 152, table; 158, table; 163, 166, 180). The name derives more specifically from its geographical distribution in the Congo laying ‘across’ the Lukuga River (L1), with roughly half of its distribution north of the river, and the other half south of the river (Brichard 1989). Its distribution in the Congo covers about 200 km (Brichard 1989: 166). According to Brichard (1989: 180, table), there are six distinct variants of Tropheus sp. “Lukuga” along the Congolese coast, with the northernmost located north of the Kabogo River (J2), and the southernmost at Mtoto (Q4). The latter variant was referred to as T. sp. “Lukuga” in Karlsson and Karlsson (2012). Both variants of T. sp. “Lukuga” found at Mtoto and Cape Tembwe were frequently collected and exported by us in the nineties, with the names of T. brichardi “Yellow – Mtoto” and T. brichardi “Lemon chin – Cape Tembwe”, respectively (Persson 1997: 23, fig. bottom, right). About five years later we also exported the variant from Kabimba, north of the Kavala Islands, known as “Canary cheek”. The colouration of T. sp. “Lukuga” at Mtoto comprises a greyish dark-brown body, anteriorly fading into a slightly yellowish head. The colouration of the variant found at Cape Tembwe is similar to the preceding, but with the yellow colouration more concentrated on the chin and cheeks.

Fig 11. Tropheus sp. "Lukuga" at Karilani Island. This variant is known as "Goldfin". The provisional specific name "Lukuga" derives from Pierre Brichard’s observations of the genus in the late seventies and in the eighties. 
Fig. 12. Bulu Point and Karilani Island. In the far distance is Pasagulu Point. The latter is located within the bounderies of Mahale Mountains National Park.
Fig. 13. Tropheus sp. "Lukuga" at Kalela Island (M4) (also known as Magambo Island). Adult induviduals of this species typically have a solid greyish brown body colouration and bright yellow pectoral fins.

Tropheus sp. “Lukuga” is also found in Tanzania, and the most commonly known Tanzanian variant is probably the one from Karilani Island (M4), also known as “Goldfin” or “Karilani goldfin” (Fig. 11), and was in the past collected and exported by us as T. brichardi “Goldfin – Karilani Island” (Zadenius 1991; 1992a; Johansson 1994). In Tanzania, T. sp. “Lukuga” has been found between Magambo Point (M4) and Lufungu Bay (N4) (pers. obs.). Its Tanzanian geographical distribution covers about 45 km. The Karilani Island variant was sampled by us for scientific studies in 2007 and deposited at the Swedish Museum of Natural History as Tropheus sp. “Lukuga – Karilani Island”.

Tropheus sp. “Lukuga” was probably first collected by L. Stappers during his expedition in 1911–1913. Boulenger, who worked on Stappers’ collection of fishes from Lake Tanganyika, reported on one single specimen which had been collected at Tumpa (presently Kitumba (L2), most likely); this specimen had six anal-fin spines and “is close to T. annectens, and even more so to T. moorii” (Boulenger 1920: 48). At Kitumba in the Congo there are two different species of Tropheus: T. sp. “Lukuga”, which possesses six anal-fin spines, and a species which has only four spines, possibly T. annectens (sensu Brichard 1989: 149f; pers. obs.; but see also Konings 2013a). Obviously, Boulenger’s specimen was a T. sp. “Lukuga”. However, it was referred to as T. moorii, and Boulenger (1920: 48) wondered whether T. moorii and T. annectens (the only two taxonomically recognised species by that time) were not really one and the same species.

Tropheus sp. “Lukuga” shows much resemblance to T. brichardi, and is not as easily distinguishable from the latter as T. sp. “Crescentic”. Minor, but important, differences are, for example, the colouration of the finnage (mostly pectoral fins (e.g. yellow vs. uncoloured, in adult individuals), but also pelvic and anal fins) and of the eyes. Dissimilarities in the presence of the vertical stripes also seem to exist. Furthermore, there is a slight morphological difference in the tail fin. The tail fin is rather truncated and near straight with almost rounded lobes in T. sp. “Lukuga”, in comparison to being indented with slightly pointed and sharper tips in T. brichardi. However, these observations are also only provisional, pending a thorough analysis.

Fig. 14. Tropheus sp. "Lukuga" at Lufungu Bay. A river by the same name is located at the innermost part of the bay. In the bay, crocodiles are common.
Fig. 15. Nkwasi Point (N4) is located on the central coast of Mahale Mountains National Park. During high tide the point is detached from the mainland and becomes an island. The rocky point is located just over 2 km north of Lufungu Bay, the southernmost locality of Tropheus sp. "Lukuga" (pers. obs.).

Behaviour and juvenile appearance
Although the differences in aggression between the various Tropheus species are rather subtle, they are observable nonetheless. Based on our experiences of wild individuals, in both their natural habitats and in aquaria, we would consider T. sp. “Crescentic” to be the most aggressive Tropheus species, closely followed by T. brichardi. T. sp. “Lukuga” is probably one of the least aggressive. The latter is frequently observed in so-called ‘foraging schools’, where thousands of individuals are foraging epilithic algae together, commonly within the feeding territories of other algae-eating species (pers. obs.) {see a video sequence of a ‘foraging school’ of Tropheus polli here}.

Regarding the juvenile colouration, there is a certain similarity in all of the three species (T. brichardi, T. sp. “Crescentic”, and T. sp. “Lukuga”) (Fig. 16, 17, 18). They all express a more or less reddish/yellowish colouration, which may suggest a close phylogenetic relationship (phylogenetic = relating to the evolutionary history of a set of organisms). However, whether such a relationship exists or not, the juvenile colouration is possibly an analogy, a character which may have evolved repeatedly likewise due to similar ecological constraints. The reddish/yellowish colouration may thus not necessarily have been derived from within a group of Tropheus which now possesses this juvenile colouration. There are species, such as T. polli, and especially of variants found along the shores of the Mahale Mountains (N4 {see a video sequence of Mahale Mountains here}, that not only possess a subtle reddish/yellowish sub-adult colouration, but also exhibit a more or less red dorsal fin in its juvenile state (Fig. 28). Moreover, there are additional Tropheus species in the south-eastern part of the lake that also possess a reddish/yellowish, or only partially reddish/yellowish, juvenile colouration, but which, in their adult state, appear phenotypically only distantly related to any of the three species (T. brichardi, T. sp. “Crescentic” or T. sp. “Lukuga”). The reddish/yellowish colouration within the genus Tropheus, whether fully or only partially evolved, may therefore be regarded as ancestral, and not derived. In addition, the change from juvenile to adult colouration seems to be correlated to the reproductive stage in the life history of the species (Karlsson and Karlsson 2014b).

Fig. 16 (top). Juvenile, size 4 cm of Tropheus sp. "Lukuga" at Karilani Island ("Goldfin"). Fig. 17 (bottom left). Juvenile, size 3 cm of Tropheus brichardi at Moyobozi. Fig 18 (bottom right). Juvenile, size 3 cm of Tropheus sp. "Crescentic" at Segunga north ("Yellow stain"). Unravelling the evolutionary mysteries of Tropheus may expose a phylogenetic network of populations.

Merely variations of Tropheus moorii
The population of T. brichardi at Nyanza Lac was probably first sampled already in 1937 by A. Lestrade, when four individuals were collected (Poll 1946: 266), and then described in more detail by Matthes (1962: 48ff; 50, table III). Both authors (Poll and Matthes) considered this variant to be a geographical colour variant of T. moorii, a name convention which was partially followed by Staeck (1974, 1975, 1985), who discovered and described the similar variant from Kigoma as a striped variant of T. moorii. In our travel log from Kigoma we also referred to it as T. moorii (Karlsson and Lundblad 1989), which was collected and exported to Sweden by us in September 1988. The Kigoma variant was first collected and exported by Klaus Grom in 1975, and shipped to his company Tagis-Aquarium in Sprendlingen (near Frankfurt), Germany; in his initial efforts of collecting and exporting Tanganyika cichlids from Kigoma, Klaus Grom was accompanied by Wolfgang Staeck and Trevor Davies (son of Malawi fish exporters Peter and Henny Davies) (Staeck 1975; Staeck, W. 2015, pers. comm., 17 Feb.). Later on, Berglund (1976: 3) summarised the differences of T. brichardi from T. moorii as a “wider mouth, longer head, shorter pelvic fins, and a narrower interorbital width, plus the complete differences in live colouration”. However, some authors on Tropheus felt that the original description of T. brichardi was not very convincing; and because of the interpretation that the “morphometric and anatomical values of T. moorii may vary”, the description of T. brichardi was questioned, and considered possibly invalid (Staeck 1977: 224ff; 1985: 109). T. brichardi was argued also elsewhere as being a geographical colour variant of T. moorii (Scheuermann 1975, 1976a, b; Hedfeld 1978). More information about T. brichardi, especially regarding its behaviour, was presented (Nelissen 1977, 1978), which likely resulted in the recognition of T. brichardi being more widely accepted. Based on the prevailing and available information, Zadenius (1981) suggested that authors on Tropheus should accept T. brichardi as an established species with a valid name, pending further studies. Today, when several DNA-based studies on Tropheus have been conducted, T. brichardi seems in all probability a genuine species distinct from both the taxonomically recognised T. moorii as well as several undescribed species (see, for example, Baric et al. 2003; Sturmbauer et al. 2005).

Early recognition
Some of the very first variants of T. sp. “Crescentic” were discovered by Staeck in 1974 along the rocky coast between the Malagarasi River {see a video sequence of Malagarasi River hereand Lugufu River (J5), and were referred to as geographical colour variants of T. moorii; they were together known as the ‘Green variant’ or ‘Goldmoorii’ (Scheuermann 1975, 1976b: 402; Staeck 1985: 114). These variants from just south of Malagarasi River (e.g. Karago (I4) and Maswa (J4)) are also known as ‘Green Wimple Moorii’ (see, for example, Herrmann 1987: 168; Konings 1988: 35; 2013b: 84; Schneidewind 2005: 108), a name which we also applied in our early exports in the late eighties (T. brichardi “Green wimple – Cape Kabogo”). ‘Wimpel’ is the German word for ‘pennant’, which refers to the pennant-shaped (wimpelförmige) tail fin in adult individuals (Schneidewind 2005). Konings reported on the differences between the ‘Green wimple’ and T. brichardi being the behaviour and the preferences of the type of habitat (1988: 35), and referred to two variants of the former as Tropheus spec. aff. brichardi II malagarasi and T. sp. aff. brichardi II kabogo, respectively (Konings 1988: 35, 38), a name and naming convention which was followed by others, e.g. Zadenius (1992a, b) and Burnel (1993), but not, however, by Konings (1998, 2013b). Although Konings also included a few more non-conspecific variants into the tentative species name Tropheus spec. aff. brichardi II (e.g. T. sp. “Kaiser/Ikola” and T. sp. “Doubleblot/Kirschfleck”) (1988: 28, fig. bottom; 38), this may be the earliest distinct name applied to this species (T. sp. “Crescentic”), when not considering the variants of it (T. sp. “Crescentic”) to be merely geographical variants of either T. moorii or T. brichardi.

Fig 19. Kiti Point just north of Karago with the Malagarasi River delta in the far distance. This area is heavily affected by turbid water from sediment discharge by the river all year round. The most affected area is the rocksy shore just north of the village (Kiti Point), but also the water south of the village is affected by poor visibility. Therefore the domestic variant of Tropheus sp. "Crescentic" (known as "Yellow band") is seldom collected for the ornamental fish trade. Instead the similar variant from further south (Maswa) is collected quite regularly and often sold as "Yellow band".
Fig. 20. The most northern variant of Tropheus sp. "Crescentic" in Tanzania ("Yellow band") is also the most conspicuous, with an almost 'perfect' yellow band running down the side. Discovered by Wolfgang Staeck in 1974. The picture was captured in shallow water at Kiti Point, Karago in the end of December 2007 (rainy season). The visability underwater was less than 1 metre.
Fig 21 (left). Tropheus sp. "Crescentic" in murky water on the southern side of Karago village. Depth 1 metre. Due to the nearness to the Malagarasi River estuary, during December to April the water visibility is extremely poor in this locality, often less than one metre. Fig. 22 (right). Tropheus sp. "Crescentic" at the northern side of Kibwe Bay. Depth 3 metres. The yellow band is quite speckled with a broad solid mark on the dorsum.

Geographical colour variants of Tropheus sp. “Crescentic” in the northern part of the Congo
The geographical distribution of T. sp. “Crescentic” in the northern part of the Congo seems rather unclear. Even though several observations of Tropheus have been made in the area, the reports concerning T. sp. “Crescentic” appear to be missing or contradictory. Marlier (1959) reported on three new geographical colour variants from the northern Congo, and one from Burundi. Then, many ecological reports followed, such as on the territorial behaviour (Kawanabe 1981), the social behaviour (Kohda 1991), the mating behaviour (Yanagisawa and Nishida 1991), and the vertical distribution (depth range) (Kohda and Yanagisawa 1992). These reports dealt with two Tropheus variants: one found at Luhanga (B2), about 10 km south of Uvira (A2), and another at “Bemba (=Pemba)” (B2), about 25 km south of Uvira, in northern Congo. However, all of the variants, including Marlier’s, pertain to another species, as of the time of the respective report, incorrectly identified as T. moorii.

Brichard (1989: 162, 170) discusses what he believes to be a “melting pot of varieties with very changing patterns” regarding the variants of his “black tail” line (also referred to as T. moorii moorii) and his “unidentified line”, distributed along the torrent estuaries from Makobola (B2) via Luhanga and Uvira to Kavimvira, in the northern part of the Congo. At both Uvira and Kavimvira, about 50% of the individuals of Brichard’s “unidentified line” (i.e. T. sp. “Crescentic”) are documented as possessing five anal-fin spines, whereas 50% are possessing six anal-fin spines. It is not unlikely that these so-called “hybrid populations” and the “melting pot of varieties” may in fact consist of the two mentioned species, coinhabiting several localities, presumably being very similar and cryptic in colouration, yet subtly distinct, and not necessarily hybridised and ‘melted’ into one. These are the “black tail” line and T. sp. “Crescentic”.

Furthermore, Konings (2012: 18) reports on the Luhanga variant, referring to it as T. sp. “Black” (the “black tail” line, or T. moorii moorii sensu Brichard 1989), and describing it as exhibiting “an olive-green to bronze color” (Konings 2012). This variant is possibly the same as the one referred to as the T. moorii bronze variety in a caption by Brichard (1989: 153, fig. top, left). However, judging by the appearance of the head profile, the decreasing body height posteriorly, and the apparently forked caudal fin of the fish in the picture, it rather seems to correspond to T. sp. “Crescentic”. In addition, the very same picture of T. moorii, the bronze variety, already appeared in Brichard (1978: 316, fig. top); the colours of this picture are differently adjusted, and the fish shows less of the bronze colours, and more of the olive-green colours. The colours of the fish in the latter picture may be more realistic and correct, and seem to agree with the olive-green variant of T. sp. “Crescentic” at Kavimvira, exhibiting a similar colouration (Brichard 1989: 159, fig. second at the top, left).

Schreyen (2001: 9) reports on two sympatric Tropheus species at Muzimu (D2), located close to the northern tip of the Ubwari Peninsula, with one of these two being “a black variant with a narrow yellow band” and the other one “with a small yellow triangle high on the dorsum” (Schreyen 2001), the latter of which being reminiscent of several variants found along the shores of Burundi, such as the variant found at Rumonge, having “a faint beige-brown triangular patch below the dorsal base” (Brichard 1989: 190; Staeck 1985: 113; pers. obs.: sampled by us in 1994), but which here likely corresponds to T. sp. “Crescentic”.
Moreover, Konings (2011: 10) confirms the presence of the variant at Muzimu, which has “a yellow patch on the upper flank”. In addition, Konings (2011) reports on two sympatric species of Tropheus found in the upper rocky habitat, slightly north of the southernmost tip of the Ubwari Peninsula, one of which he refers to as “a variant of Tropheus brichardi, also known as the Green Moorii”. However, this is more likely a variant of T. sp. “Crescentic”.

Geographical colour variants of Tropheus sp. “Crescentic” in Tanzania
In Tanzania, Tropheus sp. “Crescentic” is found from Karago (just south of the Malagarasi River) to the rocky shores just south of the village of Halembe, a distance of almost 70 km. From this part of the lake, several geographical colour variants of T. sp. “Crescentic” are known. From Karago comes a beautiful and distinct variant, labelled “Yellow band” by us in past exports. Individuals have a ‘perfect’ vertical yellow band on the body (Fig. 20), which is almost identical to the variant of T. duboisi at the same location. However, both T. sp. “Crescentic” and T. duboisi are rarely collected due to the near-zero visibility throughout the year, which in turn is due to the proximity of the outlet of the Malagarasi River, emptying its turbid water into the lake. The rocky shore just south of Kibwe Bay (J4) and Kirando village (J4) is known as Maswa (Fig. 23). In this area lives another beautiful variant of T. sp. “Crescentic” which is slightly similar to the preceding. However, the Maswa variant lacks the perfect band, and is instead adorned with a goldish yellow, or silvery yellow, large spot laterally, which sometimes may extend down to its belly (Fig. 24). Further south, passing the village of Lubengela (J4), the Kirandoo Point (J4), the Kabora Point (J4) by the outlet of Msehezi River (K5), a greenish brown variant (Fig. 26) appears at Cape Kabogo, which lacks any trace of yellow. Between the villages of Mkuyu (J4) and Segunga (K4), a yellow-stained variant is found, which has a yellow pectoral patch extending anteriorly and ventrally onto its chest (Fig. 29). Close to Mkuyu we have found several individuals of T. sp. “Crescentic” having the yellow pectoral patch extending even onto its gill cover (Fig. 30). From Segunga southwards to Halembe, along the shores of Kungwe Bay Forest Reserve, greenish to yellowish brown variants of T. sp. “Crescentic” are found (Fig. 34, 35), with a colouration including minor irregularities of yellow spots or marks, plus normally a small yellow spot on its belly. In the area from Segunga to Halembe a handful of different Kushangaza variants have been observed, with some exhibiting only minor chromatic mutations, such as yellow lips and parts of finnage (Fig. 9), while others have a more extended colourational anomaly and are draped in Kushangaza colours (Fig. 1). The yellow-stained Segunga variant was sampled by us in 2007 for scientific studies and deposited at the Swedish Museum of Natural History as Tropheus sp. “Crescentic – Segunga” (Fig. 38).

Fig. 23. Adjacent to Kirando village are the whitish rocks of Maswa. In crevices on the barren vertical wall of rocks grow plenty of Aloe plants.  On top of the cliffs a cassava field is spreading out. The picture was captured in December 2007. For a historical comparison see picture of the same view in Konings (1988: 25; 1998: 9) captured 20 years earlier.
Fig. 24. The variant of Tropheus sp. "Crescentic" at Maswa is regularly collected for the aquarium trade and sold as "Yellow band". The picture was captured in the shallow water just off the whitish rocks. Mitochondrial sequences have tentatively identified several distinct lineages being largely in agreement with morphologically and ethologically identified Tropheus species, such as T. brichardi, T. sp. "Crescentic" and T. sp. "Lukuga" (Baric et al. 2003, Sturmbauer et al. 2005).

Early uncertainties
It may be pointed out that Brichard (1989), at an early stage, probably did not fully recognise T. sp. “Crescentic” and T. sp. “Lukuga” as being two different species, or ‘lines’. He introduced them as one and the same by the name of “Green northern” or “Kalemie” line, and reported this line as existing between Mtoto and Ubwari (1989: 154). Moreover, he also stated this line to be “concurrent at each end of its range with other lines of Tropheus” (Brichard 1989: 154), alluding to T. annectens, or “Moba” line (sensu Brichard 1989: 149f, 177; 180, table) in the south at Mtoto, and to his ‘black tail’ line in the north at Ubwari (1989: 150). He initially also seems to have treated these two species (T. sp. “Crescentic” and T. sp. “Lukuga”) as one and the same when he reported on a species at Mtoto having a “yellow snout [T. sp. “Lukuga”] and normally notched tail [T. sp. “Crescentic”]” (Brichard 1989: 156), stating further that this Tropheus “has 6 anal spines, a normally forked caudal [T. sp. “Crescentic”], and a color pattern featuring a plainly colored body with a yellow snout [T. sp. “Lukuga”]” (1989: 177); however, it should be mentioned that Brichard did consider T. sp. “Lukuga” to have something of a forked caudal fin when stating: “caudal normally forked and not exceptionally long” (1989: 163). In our opinion, T. sp. “Lukuga” has one of the least forked caudal fins in the genus, and instead has more of a truncate and straight-cut caudal fin — only slightly emarginate. Nevertheless, elsewhere in the same chapter (chapter 6), Brichard seems to have changed his view on this line (the “Green northern”/“Kalemie” line). Firstly, in the table (1989: 152), two separate lines with names similar to the initial name are documented: the “Kalemie” line and the “Northern green” line, the latter of which additionally likely containing two different lines (species), indicated as “hybrids”. Secondly, the initial line (the “Green northern”/“Kalemie” line) is further treated more firmly as two different lines, i.e. the “Unidentified line” (T. sp. “Crescentic”) and the “Lukuga (Kalemie) line” (T. sp. “Lukuga”) (1989: 158, table; 163, 169; 180, table), which are also stated as being sympatric “from the islands of Toa Bay [Mtoa Bay; the area of the Kavala Islands] toward the north” (1989: 166), a statement which also agrees with the map showing the geographical distributions of the Tropheus species (1989: 182). The alternative names plus the redefinition of names may seem a bit confusing.

Fig. 25. The beautiful scenery of Cape Kabogo. In the lush forest of the Kabogo area, Japanese scientists studied chimpanzees in 1961-1962. A few years later the project shifted to Mahale Mountains (gazetted as national park in 1985). The project is today still ongoing.
Fig. 26. In the water of Cape Kabogo lives one of the least colourful variants of Tropheus sp. "Crescentic". The  species was first identified by Brichard (1989: 169f) and referred to as the ‘unidentified line’ and should not be confused with another of Brichard’s lineages, the so-called ‘crescent tail line’, which corresponds to T. polli (Brichard 1989: 154).
Fig. 27. Tropheus polli occurs in several different geographical colour variants. The picture shows a spawning pair of the colourful yellowish green variant at Katumba Point (M4). Geographically distributed in an area with the largest specific variation of Tropheus (i.e. the number of species, whether allopatric or sympatric), T. polli (and the sister species T. annectens) may represent (one of) the oldest member(s) of the genus, and possibly being ancestral to some of the other members. The number of anal-fin spines possessed by T. polli (sensu Axelrod 1977) and T. annectens further indicates a link between Tropheus and the less specialised Simochromis, which was what Boulenger (1900) had in mind when he chose the name ‘annectens’ (from the Latin ‘annectere’, meaning ‘to attach’ or ‘to connect’). Superficially, some populations of T. polli and S. diagramma share the greenish body colouration, as well as, and more remarkably, the red small dots on the body (e.g. T. polli, Isonga, Tanzania), the latter of which is also found in ‘Ctenochromishorei, another trophein (Tropheus-related) ancestor representative. Interpreted from molecular analyses, the two genera Tropheus and Simochromis are closely related (Sturmbauer et al. 1997).

Schneidewind (1993) speculates on the origin of Tropheus being in the northern part of the lake because of the supposed existence of more diversified species there. However, what is known to date is that the central part includes the same level of diversification, plus a higher number of species, including T. duboisi, which has been observed south of Lubulungu River (N4) (Konings 2013b; pers. obs.). T. duboisi is possibly (one of) the oldest species of the genus. However, it has been reported that T. duboisi is not a Tropheus, that it constitutes a parallel lineage to Tropheus, and that it is more closely related to Simochromis (Sturmbauer 2003: 27; Sturmbauer et al. 2003: 55, 61). T. duboisi has a jaw and teeth morphology very similar to S. diagramma, and very different from all other Tropheus species (pers. obs.). The origin of the true Tropheus is not unlikely somewhere in the central part of the lake.

Perhaps a species which would be generically known as Tropheus did evolve in the central part of the lake, and from there spread independently both northwards and southwards. Corresponding to this idea, there are neighbouring Tropheus species, in both the north central and south central parts of the lake, that are very similar in appearance to both T. polli and T. annectens.
Fig. 28. Tropheus polli juvenile at Pasagulu Point (M4), Mahale Mountains National Park. Size 2 cm. The individual displays redish markings on the body and a distinct red margin on the dorsal fin. The ancestral state in juvenile colouration of Tropheus is possibly dark brown, or dark reddish brown, with reddish finnage.

Morphological characters of Tropheus
Allopatric speciation often produces ecologically equivalent sister species. Such sister species may sometimes be very cryptic and difficult to identify. The earlier method of regarding all non-sympatric sister species as conspecifics is misleading. It is often claimed that Lake Tanganyika harbours far fewer cichlid species than Lake Malawi and Lake Victoria. But this is partially due to the acceptance of allopatric taxa in the latter two lakes being non-conspecific. Pending more research work in Lake Tanganyika and on its cichlid fishes, it is not clear whether this lake really harbours that many fewer cichlid species (Turner et al. 2001: 802).

More or less, any set of stable (non-variable) characters may work in order to diagnose species, including genetic characters. Two common diagnosable characters in fishes like the ‘modern haplochromines’, including the Tropheus, are the colouration and the number of hard and soft rays in the various fins. Tropheus are known to show a considerable degree of inter-specific variation in the number of anal-fin spines, but a rather low degree of intra-specific variation; therefore, all morphological studies on Tropheus mainly refer to this meristic character (Nelissen 1979). However, Poll (1956: 266f) argued for the number of both hard and soft anal-fin rays to vary in Tropheus, something which was disagreed upon by Nelissen (1979: 30), who considered this meristic character to be quite invariable within species of Tropheus. Another example of a meristic character is the number of caudal-fin rays, which usually shows low intraspecific variation (Fricke 1983).

The Tropheus study by Snoeks et al. (1994) focused on two meristic characters: the number of anal-fin spines and dorsal-fin spines. Their result showed that there is a great inter-populational variation in the number of anal-fin spines and dorsal-fin spines, and that the geographical distribution of these variations shows a distinctive pattern, which can be distinguished as discrete groupings, i.e. possibly species (Snoeks et al. 1994: 366). Brichard, who also considered the number of anal-fin spines in individual species of Tropheus to be rather constant (1989: 150, 152, 156), classified all of the Tropheus variants based on this character, in combination with the colouration. In the Great Lakes of Africa there are likely to be numerous sister species complexes in which live colouration is the most distinctive specific character, and often the only reliable diagnostic character (Marsh 1983: 9).

The colouration is a character that appears to evolve quickly, “sometimes completely out of step with genetic changes” (Stiassny and Meyer 1999a, b). The colouration of Tropheus can vary greatly, both within and between species. In some cases, genetically closely related Tropheus populations have rather different colourations (see pictures for comparison of geographical variants of T. sp. “Crescentic”). Therefore, contrary to the colouration being an important diagnostic tool, obviously not all geographical colour variants pertain to a distinct individual species, as species may be colour-polymorphic. Despite the pronounced colourational variation, Tropheus has otherwise remained largely unchanged morphologically for probably more than one million years (Meyer 1993: 284), or about one and a half million years (Sturmbauer 2003: 27). However, it is now possible with more advanced technology to detect even minor structural and shape-wise differences among Tropheus populations (Herler et al. 2010; Lechner 2010; Kerschbaumer et al. 2014). 

Fig. 29. Tropheus sp. "Crescentic" north of Segunga. This variant is distinguished by a yellow patch at the base of the pectoral fin and commonly known as "Yellow stain". Discovered by Horst Walter Dieckhoff in 1987 and labelled as originating from Isonga (Konings 1988: 124, fig. left, third from top), this variant is found from north of Segunga Bay to just south of Mkuyu village. It was rediscovered by our team in 1992 and exported as T. brichardi "Yellow stain". The population was sampled for scientific studies by us in 2007. The specimens were deposited at the Swedish Museum of Natural History by the name Tropheus sp. "Crescentic – Segunga".
Fig. 30 (top left). In December 2007 we found several individuals of Tropheus sp. "Crescentic" having the yellow pectoral patch extending onto the gill cover. This individual was photographed just south of Mkuyu village {see video clip of a similar individual from south of Mkuyu here}. Fig 31 (top right). Crystal clear water at the rocky shore just north of Segunga. Here lives the "Yellow stain" variant of Tropheus sp. "Crescentic". In the distance is Mkuyu Point. Fig 32 (centre left). On the northern side of Segunga Bay, Tropheus sp. "Crescentic" has a smaller yellow patch than individuals living further to the north. Fig. 33 (centre right). South of Segunga, induviduals of Tropheus sp. "Crescentic" with a bright yellow spot in the dorsal fin are encountered. Fig 34 (bottom left). Tropheus sp. "Crescentic" at Lusungura Point (K4). This variant is likely identical to the variant found at Kungwe Point 20 km to the north. Fig 35 (bottom right). Tropheus sp. "Crescentic" at Kangwena (K4), north of Halembe. In this locality induviduals with a yellow anal fin were observed (commonly females and sub-adults).

Stenotopy and allopatry
Tropheus are poor dispersers, and both males and females tend to defend feeding and breeding territories for several years, which imply a restricted gene flow. In addition, brood sizes are small. These two factors promote diversification and allopatric speciation, which seems to be the most common mode of speciation in Tropheus. Lake Tanganyika has close to 100 affluent tributaries (Kullander and Roberts 2012: 372), which effectively prevent many allopatric Tropheus populations from exchanging genes, and ultimately strengthen their non-conspecific relationship. In a recent molecular analysis, high genetic diversity and population differentiation were detected in a Tropheus species (T. moorii) from the southern part of Lake Tanganyika, whereas much lower variation and structure were found in O. ventralis, a moderately stenotopic rock-dwelling cichlid also found in the southern part of the lake (Sefc et al. 2007). Most of the Tropheus variants are distributed allopatrically, but phylogeographic data indicate that populations underwent recurrent cycles of population fragmentation and secondary contact in the course of lake-level fluctuations (Egger et al. 2008). Their evolutionary journey which led to the present-day geographical distribution is imprinted and preserved in their DNA. This phylogenetic network of evolutionary paths has been investigated recently (e.g. see Baric et al. 2003; Sturmbauer et al. 2005; Egger et al. 2007; Koblmüller et al. 2010, 2011).

Fig 36. Along the 60 km coastline from Segunga southwards to Kalela Island there is almost nowhere to find shelter from a raging lake. The local wooden boats commonly used on Lake Tanganyika are slow movers with a speed of about 5–10 knots (≈9–18 km/h), taking hours to reach the destination (obviously depending on the size of the outboard motor). Finding shelter in such small creeks as the one at Halembe village is the only way in which to avoid the waves when caught up in bad weather. The picture shows members of African Diving staff at Halembe securing the anchor line after having manoeuvred the boat (15 metres in length) by hand past the surf and parked it in the small creek.

DNA and Tropheus
Molecular phylogenetic analyses have in recent years resolved some aspects of the phylogeny of African cichlids by exploring various markers, such as allozymes, the control region of mitochondrial DNA (mtDNA), the genes for cytochrome b and ND2 in the mitochondrial genome, and noncoding regions in the nuclear genome (see list of authors in Takahashi et al. 2001: 2057). But even when some relationships are largely supported in analyses using single-locus markers, such as genes in mtDNA, the possibility of misinterpretation of phylogeny remains. In order to improve the reliability of inference, the use of multiple genetic markers is recommended (Maddison and Knowles 2006). This was followed in the phylogenetic analysis and species description of Neolamprologus timidus, where fragments of three mitochondrial genes and two nuclear genes were used (Kullander et al. 2014: 303). For an overview of DNA-related research on Lake Tanganyika cichlids, see Rüber (1998).

Molecular phylogenetic analyses of Tropheus have been published several times since the first by Sturmbauer and Meyer (1992), summarised in French by Fontaine (1993). These analyses have primarily relied on sequences of gene fragments of mtDNA, which, by the way, is widely recognised as an important tool for resolving relationships among closely related species (Kocher et al. 1995; Clabaut et al. 2005). In addition, mitochondrial DNA evolves faster than nuclear DNA, and phylogenies based on mtDNA (particularly of the quickly evolving control region) can provide the finer resolution necessary for examining the evolutionary relationships among young species (Meyer 1993). Moreover, short sequences from a fragment of a mitochondrial gene contain phylogenetic information extending from the intraspecific level to the intergeneric level (Kocher et al. 1989; Clabaut et al. 2005). Furthermore, sequences of mtDNA can provide independent tests in taxonomic work.

Regarding the initial mtDNA-based analyses on Tropheus, the revealing results were interpreted as not being in agreement with the prevailing Tropheus species taxonomy (e.g. Sturmbauer and Meyer 1992; Sturmbauer et al. 1997). As a possible consequence thereof, Sturmbauer and Meyer (1992) suggested Tropheus taxonomy being reviewed with molecular data. Several more molecular phylogenetic studies on Tropheus followed, with the immediate being Baric et al. (2003) and Sturmbauer et al. (2005). The main difference between these two and the initial from 1992 is the greater number of individuals (samples) and populations (sample sites). A higher number of samples and sample sites may likely result in a more accurate inference of a species (or population) phylogeny (Maddison and Knowles 2006). Some of the further work (Sturmbauer et al. 2005) was also aimed at reconstructing the origin and spreading of the different Tropheus lineages, based on the mtDNA phylogeny in previous work.

Fig. 37 (top left). Tropheus sp. "Crescentic" from Karago ("Yellow band"), NRM 61565. Fig. 38 (top right). Tropheus sp. "Crescentic" from north of Segunga Bay ("Yellow stain"), NRM 58001. Fig. 39 (centre left). Tropheus sp. "Crescentic" from Halembe. Fig. 40 (centre right). Tropheus sp. "Crescentic" from south of Segunga Bay. Note the yellow spot in dorsal fin. Fig. 41 (bottom left). Tropheus sp. "Lukuga" from Karilani Island ("Goldfin"), NRM 57969. Fig 42 (bottom right). Tropheus brichardi from Kitwe Point, Katonga (H3), NRM 61567. Molecular analyses of the quickly evolving portion of the mitochondrial genome may resolve the phylogenetic relationship within Tropheus.

Incomplete lineage sorting in the diversification of Tropheus
The results of molecular phylogenetic analyses are primarily genealogies of genes, and even if a considerable phylogeny of species may be interpreted from them, it has long been known that the genealogical genomic history of several species need not be identical to the history of the species themselves, i.e. gene trees are not necessarily in agreement with species trees (e.g. Pamilo and Nei 1988; Doyle 1992; Rosenberg 2002; Avise 2004: 143). The impression of the several DNA-based Tropheus studies (e.g. Sturmbauer and Meyer 1992; Baric et al. 2003; Sturmbauer et al. 2005) was that they essentially claimed to represent species (or population) phylogeny. At least this is how it was interpreted by some authors on Tropheus. However, several processes can explain the disagreement between a gene tree and a species tree, with the two most commonly suggested perhaps being (1) hybridisation and (2) lineage sorting (coalescent theory) (Doyle 1992; Maddison 1997; Rosenberg and Nordborg 2002). Although the two processes have conflicting effects on species diversification, either by inhibiting or by promoting divergence, they are often difficult to separate in species that have experienced multiple periods of isolation and expansion (Maddison and Knowles 2006).

Lineage sorting is a genetic process (concerning the dividing (splitting) of populations) by which alleles are inherited and lost over time. This process can be traced on the basis of molecular evidence. ‘Coalescence’ is a term used for the point (looking back in time) at which two alleles converged on a single ancestral copy; the term ‘deep coalescence’ refers to coalescence of alleles occurring significantly earlier than the divergence of the species containing those alleles (Leliaert et al. 2014). Deep coalescence may arise from the random loss of gene lineages by genetic drift of several gene lineages that coexisted in the past. Incomplete lineage sorting is the maintenance of genetic variation within a metapopulation* lineage from one speciation event to the next, resulting in deep coalescence and gene tree–species tree disagreement (Leliaert et al. 2014).

* A metapopulation is a set of populations connected by gene flow over a relatively short time interval.

Alternatively expressed: a complete lineage sorting is the merging (backwards in time) of the genealogy of multiple gene copies into their common ancestor within the duration of the species, or specific diversification event; an incomplete lineage sorting is the failure of these gene copies to merge (backwards in time) within the duration of the species — the gene lineages merge in an ancestral species.

Incomplete lineage sorting can give rise to gene tree–species tree disagreement for recently diverged species as well as for older divergences when the time between successive speciation is short. Deep coalescence in the past leads to gene tree–species tree disagreement in the present (Rosenberg and Nordborg 2002; Maddison and Knowles 2006).

Lineages that have diverged completely may also share genetic polymorphism in parts of the genome due to secondary admixture (hybridisation). Incomplete lineage sorting and admixture produce similar genetic patterns, and the sharing of alleles in two lineages may thus be the result of either process, or both (Maddison 1997). Moreover, both incomplete lineage sorting and admixture (hybridisation) can cause serious difficulties for phylogenetic inference (Maddison and Knowles 2006). The fact that various displaced, unnaturally occurring Tropheus species (most commonly T. sp. “Kaiser/Ikola” and T. sp. “Doubleblot/Kirschfleck”) have hybridised with local and naturally occurring species in several localities of the lake may very likely seriously complicate the construction of a phylogeny based on molecular analyses; see Karlsson and Karlsson (2013: 3; 2014a).

Fig. 43. Map over Lake Tanganyika showing distribution range of Tropheus brichardi, T. sp. "Crescentic" and T. sp. "Lukuga". For easy navigation, localities are accompanied by coordinate designations when first appearing in the text (corresponding to numbers and letters in the map margins).

Tropheus – A symbol of unfinished speciation
Incomplete lineage sorting occurs frequently in large populations (many interbreeding individuals; extra gene lineages occurring) with few generations before splitting (incomplete sorting of gene variants). This especially includes species with several ‘temporarily divided’ populations, such as some of the Tropheus species, of which some groups of populations are possibly being mixed up in a never-ending diversification or speciation event, with the result of unfinished speciation.
Based on the historically fluctuating nature of Lake Tanganyika, the specific and ‘subspecific’ geographical distribution of species and populations of Tropheus are predicted to have expanded and contracted, as well as increased and decreased in individual numbers, in response to these fluctuational events {see a video sequence of the lake crossing dispersal animated here}. The ancestral polymorphic gene copies (alleles) of Tropheus, after repeated population fragmentation and secondary admixtures, with too little time and with too large specific populations for a complete lineage sorting to take place, may end up widely distributed within the genus. Instead of Tropheus being the symbol of speciation, as coined by Marlier, it may better be viewed as the symbol of unfinished speciation.

A genetic reflection of Tropheus sp. “Crescentic” and T. sp. “Lukuga”
The molecular analysis and the phylogenetic inference of Baric et al. (2003) produced subsets of Tropheus which are largely in agreement with several morphologically identified subsets (species). The analysis was based on 365 specimens from 55 localities, covering a large part of the shoreline of Lake Tanganyika. Among these specimens were 243 types of mitochondrial patterns (haplotypes: a set of specific alleles) found, of which 11 series of such closely related haplotypes were detected. One of these series was the so-called ‘Lineage A1’, an mtDNA lineage inferred from samples of Tropheus collected within the borders of the geographical distribution of T. sp. “Crescentic” in Tanzania (Baric et al. 2003: 58, fig. 2; 61: fig. 4a–b); collecting sites were Kiti Point (I4), Kibwe Bay and Kabwe (the latter is also known as Halembe). No samples of T. sp. “Crescentic” from the Congo were included in the analysis; consequently, the mtDNA lineage A1 was not found there. Another series of closely related haplotypes was the so-called ‘Lineage A2’, which primarily included samples of Tropheus collected along the geographical distribution of T. sp. “Lukuga”, e.g. Cape Tembwe, Kavala Islands, Karilani Island/Bulu Point area (M4); a sample from the latter location was referred to as T. “yellow” (61: fig. 4a). This mtDNA lineage, however, also included haplotypes from specimens collected in the south, from populations not conspecific with T. sp. “Lukuga”. The genetic similarity is not unlikely due to incomplete lineage sorting of ancestral polymorphisms during successive turns of speciation (see e.g. Pamilo and Nei 1988), i.e. the two species (or specific populations of them) share an ancient set of polymorphic gene copies*. They are not necessarily the closest related species within the genus. Altogether, the molecular analysis reflects the existence of both T. sp. “Crescentic” and T. sp. “Lukuga”.

* Ancestral polymorphism is the genetic variation that originated prior to a specific diversification event. The presence of ancestral polymorphism in closely related species (shared ancestral polymorphism) may result in gene tree–species tree disagreement (Maddison and Knowles 2006; Leliaert et al. 2014).

The mtDNA phylogenies of Tropheus as presented in Baric et al. (2003), and also partially in Sturmbauer et al. (1997), have been interpreted by Schupke (2003), who incorporated these with his personal field and aquarium observations. In Schupke (2003) each mtDNA lineage (or a part thereof; or a combination of such) is treated as a distinct lineage of supposedly specifically related populations, which are designated as local colour forms, or variants. Included are lineages of populations which largely or partially correspond to the identifications of T. sp. “Crescentic” and T. sp. “Lukuga”, which are being referred to as ‘Lineage 5’ and ‘Lineage 3’, respectively (Schupke 2003: 46ff, 59ff). However, there is a misconception regarding the composition: T. sp. “Crescentic” (lineage 5) does not occur north of Malagarasi River, as is stated (2003: 61, fig.). Therefore, the three reported populations north of Malagarasi River (Rumonge, Nyanza, and Kigoma) do not exist; the remaining 11 populations exist and correspond to T. sp. “Crescentic”. Regarding T. sp. “Lukuga” (lineage 3, partially), this species does, in our opinion, not occur anywhere along the east coast north of the Magambo area in Tanzania. Schupke’s ‘Lineage 3’ (2003: 46ff) seems to be a mixed composition of the two species T. brichardi (partially) and T. sp. “Lukuga”.

Fig. 44. One of the authors and members of African Diving staff during an 8 week survey trip along the northern Tanzanian shores of Lake Tanganyika in December 2007 and January 2008. During a fish-survey diving-trip lots of equipment is brought along, together with several months of food supply.

The neighbour is usually the next-of-kin
Even though DNA analyses have shown that genes of distantly and remotely distributed Tropheus variants from different parts of the lake may be closely related (e.g. Baric et al. 2003; Egger et al. 2007), and thus pertain to a genealogy of genes, the conclusion of Brichard (1989: 150) is likely in agreement with species taxonomy, and may be conveniently followed, when stating races, variants and species from a certain part of the lake to “be more closely related to each other than to races [variants and species] living at the other end of the lake”. Therefore, Tropheus species from the central and the southern parts of the lake are not necessarily closely related to T. brichardi, even though they may possibly be slightly or grossly similar in appearance. The name Tropheus brichardi should not be applied to these more or less similar, yet different, species; rather, it should be restricted to the populations found in the Nyanza Lac – Malagarasi vicinity.

The total results of the molecular phylogenetic analyses that have been carried out on Tropheus so far have resulted in quite a bit of knowledge in the evolutionary history of the genus, but rather little has been added to the taxonomic knowledge. A better understanding of the relationship between the various Tropheus species seems to have failed to occur (comp. Staeck 2008, 2009). Furthermore, it is not only the gene tree vs. species tree that complicates things. It is also the concept of what constitutes a species, since species as an evolutionary entity and species as a taxonomic unit are two (among several ideas) that seem to compete for the status of being the ultimate ‘species’, even though they are seemingly two different things and will never be consolidated. In many people’s view, evolutionary, incrementally changing entities are populations, and the pattern of variation interpreted in these populations, based on any type and any number of characters, and, as static as they may be, are taxonomic species, i.e. ‘species’.

Considering the evolutionary events of repeated population fragmentation and secondary admixture that most Tropheus variants seem to have experienced, a strict DNA-based phylogeny may contradict useful and judicious species taxonomy. Is it needed to disentangle the possibly reticulate phylogeny of the Tropheus genome in order to taxonomically identify the present species?

Fig. 45. Thunderstorm approaching Maswa. If not quickly finding safety in a sheltered bay, you may soon be in the middle of a raging inferno with heavy winds and downpour.

Suggestion on how to cite this blog article;

Karlsson, M. and Karlsson, M. (2015) Kushangaza at Halembe in Lake Tanganyika, and observations on Brichard’s Tropheus. African Diving Blog. Available from: http://blog.africandivingltd.com/2015/03/kushangaza-at-halembe-in-lake.html (accessed [day] [month] [year])


Avise, J. C. (2004) Molecular markers, natural history, and evolution. Second edition. Sinauer Associates, Inc. Sunderland, MA, 684pp.

Axelrod, G. S. (1977) A new species of Tropheus (Pisces: Cichlidae) from Lake Tanganyika. The J.L.B Smith Institute of Ichthyology Special Publication (17): 1-12.

Baric, S., Salzburger, W. and Sturmbauer, C. (2003) Phylogeography and Evolution of the Tanganyikan Cichlid Genus Tropheus Based upon Mitochondrial DNA Sequences. Journal of Molecular Evolution, 56 (1): 54–68; doi:10.1007/s00239-002-2380-7.

Berglund, G. (1976) Två nya Tanganyikaarter. Ciklidbladet, 9 (4): 3. Nordiska Ciklidsällskapet.

Boulenger, G. A. (1898) Report on the collection of fishes made by Mr. J. E. S. Moore in Lake Tanganyika during his expedition, 1895-96. Transactions of the Zoological Society of London, 15 (1, 1): 1-30.

Boulenger, G. A. (1900) Matériaux pour la faune du Congo. Poissons nouveaux du Congo. Sixième Partie. Mormyres, Characins, Cyprins, Silures, Acanthoptérygiens, Dipneustes. Annales du Musee du Congo (Series Zoology), 1 (6): 129-164. 

Boulenger, G. A. (1920) Poissons de la Mission Stappers, 1911-1913, pour l’exploration hydrographique et biologique des lacs Tanganika et Moéro. Revue Zoologique Africaine, 8 (1): 1-57. 

Brichard, P. (1978) Fishes of Lake Tanganyika. T. F. H. Publications Inc., Ltd., Hong Kong, 448 pp.

Brichard, P. (1989) Pierre Brichard's book of cichlids and all the other fishes of Lake Tanganyika. T.F.H. Publications, Neptune City, NJ, 544 pp.

Burnel, P. (1993) Les Tropheus, aperçu géographique. Revue française des Cichlidophiles (127): 7-28. [March 1993]

Clabaut, C., Salzburger, W. and Meyer, A. (2005) Comparative phylogenetic analyses of the adaptive radiation of Lake Tanganyika cichlid fish: Nuclear sequences are less homoplasious but also less informative than mitochondrial DNA. Journal of Molecular Evolution, 61 (5): 666–681. 

Cohen, A. S., Lezzar, K.-E., Tiercelin, J.-J. and Soreghan, M. (1997) New palaeogeographic and lake-level reconstructions of Lake Tanganyika: implications for tectonic, climatic and biological evolution in a rift lake. Basin Research, 9 (2): 107-132.

Egger, B., Koblmüller, S., Sturmbauer, C. and Sefc, K. M. (2007) Nuclear and mitochondrial data reveal different evolutionary processes in the Lake Tanganyika cichlid genus Tropheus. BMC Evolutionary Biology, 7 (137): 14 pp; doi:10.1186/1471-2148-7-137.

Egger, B., Obermüller, B., Eigner, E., Sturmbauer, C. and Sefc, K. M. (2008) Assortative mating preferences between colour morphs of the endemic Lake Tanganyika cichlid genus Tropheus. Hydrobiologia, 615 (1): 37–48.

Fontaine, M. (1993) Divergence génétique, spéciation et “conservatisme” morphologique dans la lignée Tropheus. Revue française des Cichlidophiles (127): 29-36.

Fricke, R. (1983) A method for counting caudal fin rays of actinopterygian fishes. Braunschweiger Naturkundliche Schriften, 1 (4): 729-733. Braunschweig.

Hedfeld, M. (1978) Die Gattung Tropheus: Nur lokale Farbformen oder mehrere selbständige Arten? DCG-Informationen, 9 (6): 101-105. Deutsche Cichliden-Gesellschaft e. V.

Herler, J., Kerschbaumer, M., Mitteroecker, P., Postl, L. and Sturmbauer, C. (2010) Sexual dimorphism and population divergence in the Lake Tanganyika cichlid fish genus Tropheus. Frontiers in Zoology, 7 (4): 10 pp.

Herrmann, H.-J. (1987) Die Buntbarsche der Alten Welt - Tanganjikasee. Reimar Hobbing, Essen, West Germany, 239 pp.

Johansson, J.-Å. (1994) Importnytt från Åleds Akvarium AB. Ciklidbladet, 27 (5): 38-40. Nordiska Ciklidsällskapet.

Karlsson, M. and Karlsson, M. (2012) Tropheus moorii “Murago”. Ciklidbladet, 45 (2): 6-15. Nordiska Ciklidsällskapet.

Karlsson, M. and Karlsson, M. (2013) Lepidiolamprologus kamambae - A new predatory cichlid from Lake Tanganyika. African Diving Lakesite articles, 1 (1): 1-24; no. 1. First edition; printed. MMK, Sweden.

Karlsson, M. and Karlsson, M. (2014a) Displaced Tropheus in Lake Tanganyika. African Diving Blog. Available from: http://blog.africandivingltd.com/2014/05/displaced-tropheus-lake-tanganyika_31.html (accessed 27 Feb. 2015)

Karlsson, M. and Karlsson, M. (2014b) Variabilichromis moorii - A species of populations being phenotypically uniform but genotypically distinct. African Diving Blog. Available from: http://blog.africandivingltd.com/2014/11/variabilichromis-moorii-species-of.html (accessed 27 Feb. 2015)

Karlsson, M. and Lundblad, J. (1989) En tur till Tanganyikasjön. Ciklidbladet, 22 (1): 14-16, cover. Nordiska Ciklidsällskapet.

Kawanabe, H. (1981) Territorial behaviour of Tropheus moorei (Osteichthyes: Cichlidae) with a preliminary consideration on the territorial forms in animals. African Study Monographs, 1: 101-108.

Kerschbaumer, M., Mitteroecker, P. and Sturmbauer, C. (2014) Evolution of body shape in sympatric versus non-sympatric Tropheus populations of Lake Tanganyika. Heredity, 112 (2): 89–98.

Koblmüller, S., Egger, B., Sturmbauer, C. and Sefc, K. (2010) Rapid radiation, ancient incomplete lineage sorting and ancient hybridization in the endemic Lake Tanganyika cichlid tribe Tropheini. Molecular Phylogenetics and Evolution, 55 (1): 318-334.

Koblmüller, S., Salzburger, W., Obermüller, B., Eigner, E., Sturmbauer, C. and Sefc, K. M. (2011) Separated by sand, fused by dropping water: habitat barriers and fluctuating water levels steer the evolution of rock-dwelling cichlid populations in Lake Tanganyika. Molecular Ecology, 20 (11): 2272-2290.

Koblmüller, S., Sefc, K. M. and Sturmbauer, C. (2008) The Lake Tanganyika cichlid species assemblage: recent advances in molecular phylogenetics. Hydrobiologia, 615 (1): 5–20.

Kocher, T. D., Conroy, J. A., McKaye, K. R., Stauffer, J. R., Jr. and Lockwood, S. F. (1995) Evolution of NADH Dehydrogense Subunit 2 in East African cichlid fish. Molecular Phylogenetics and Evolution, 4 (4): 420–432.

Kocher, T. D., Thomas, W. K., Meyer, A., Edwards, S. V., Pääbo, S., Villablanca, F. X. and Wilson, A. C. (1989) Dynamics of mitochondrial DNA evolution in animals: amplification and sequencing with conserved primers. Proceedings of The National Academy of Sciences of the United States of America (PNAS), 86 (16): 6196-6200.

Kohda, M. (1991) Intra- and interspecific social organization among three herbivorous cichlid fishes in Lake Tanganyika. Japanese Journal of Ichthyology, 38 (2): 147-163.

Kohda, M. and Yanagisawa, Y. (1992) Vertical distributions of two herbivorous cichlid fishes of the genus Tropheus in Lake Tanganyika, Africa. Ecology of Freshwater Fish, 1 (2): 99–103.

Konings, A. (1988) Tanganyika cichlids. Verduijn Cichlids and Lake Fish Movies, Zevenhuizen and Herten, 272 pp.

Konings, A. (1998) Tanganyika cichlids in their natural habitat. Cichlid Press, El Paso, TX, 272 pp.

Konings, A. (2011) Burundi and northern Congo expedition 2011 - Part 1. Cichlid News, 20 (4): 6-10. Aquatic Promotions Division, Miami, FL.

Konings, A. (2012) Burundi and northern Congo expedition 2011 - Part 2. Cichlid News, 21 (1): 14-20. Aquatic Promotions Division, Miami, FL.

Konings, A. (2013a) New information on Tropheus annectens Boulenger 1900. Cichlid News, 22 (1): 12-16. Aquatic Promotions Division, FL.

Konings, A. (2013b) Tropheus in their natural habitat. Cichlid Press, El Paso, TX, 160 pp.

Kullander, S. O. and Roberts, T. R. (2012) Out of Lake Tanganyika: endemic lake fishes inhabit rapids of the Lukuga River. Ichthyological Exploration of Freshwaters, 22 (4): 355-376. http://www.pfeil-verlag.de/04biol/pdf/ief22_4_08.pdf

Kullander, S. O., Norén, M., Karlsson, M. and Karlsson, M. (2014) Description of Neolamprologus timidus, new species, and review of N. furcifer from Lake Tanganyika (Teleostei: Cichlidae). Ichthyological Explorations of Freshwaters, 24 (4): 301-328. http://www.pfeil-verlag.de/04biol/pdf/ief24_4_03.pdf

Lechner, W. (2010) Tropheus moorii & Tropehus polli - Populationen unterscheiden sich nicht nur an der Färbung. DATZ, 63 (5): 21. Eugen Ulmer-Verlag, Stuttgart.

Leliaert, F., Verbruggen, H., Vanormelingen, P., Steen, F., Lopez-Bautista, J. M., Zuccarello, G. C. and De Clerck, O. (2014) DNA-based species delimitation in algae. European Journal of Phycology, 49 (2): 179-196.

Maddison, W. P. (1997) Gene trees in species trees. Systematic Biology, 46 (3): 523-536; doi: 10.1093/sysbio/46.3.523

Maddison, W. P. and Knowles, L. (2006) Inferring phylogeny despite incomplete lineage sorting. Systematic Biololy, 55 (1): 21-30.

Marlier, G. (1959) Observations sur la biologie littorale du Lac Tanganyika. Revue de Zoologie et de Botanique Africaines, 59 (1-2): 164-183.

Marsh, A. C. (1983) A taxonomic study of the fish genus Petrotilapia (Pisces: Cichlidae) from Lake Malawi. Ichthyological Bulletin of the J. L. B. Smith Institute of Ichthyology (48): 1-14.

Matthes, H. (1962) Poissons nouveaux ou intéressants du lac Tanganika et du Ruanda. Annales, Musée Royal de l'Afrique Centrale, Tervuren, Sér. in-8°, Sciences Zoologiques (111): 27-88.

Meyer, A. (1993) Phylogenetic relationship and evolutionary processes in East African cichlid fishes. Trends in Ecology and Evolution, 8 (8): 279-284.

Nelissen, M. H. J. (1977) Über den taxonomischen Status von Tropheus brichardi Nelissen & Thys van den Audenaerde, 1975. Translated from English to German by Werner Schmettkamp. DCG-Informationen, 8 (2): 29-31. Deutsche Cichliden-Gesellschaft e. V.

Nelissen, M. H. J. (1978) Contribution to the ethology of Tropheus brichardi Nelissen and Thys 1975 (Pisces: Cichlidae). Acta Zoologica et Pathologica Antverpiensia, 70: 141-147.

Nelissen, M. H. J. (1979) A taxonomic revision of the genera Simochromis, Pseudosimochromis and Tropheus (Pisces: Cichlidae). Annales du Musée royal de l'Afrique Centrale, Série in-8°, Sciences Zoologiques (229): 1-54. Tervuren.

Nelissen, M. H. J. and Thys van den Audenaerde, D. F. E. (1975) Description of Tropheus brichardi sp. nov. from Lake Tanganyika (Pisces, Cichlidae). Revue Zoologique Africaine, 89 (4): 974-980.

Pamilo, P. and Nei, M. (1988) Relationships between gene trees and species trees. Molecular Biology and Evolution, 5 (5): 568–583.

Persson, R. (1997) De överträffar våra vildaste drömmar: Nyheter från African Diving. Ciklidbladet, 30 (4): 22-23. Nordiska Ciklidsällskapet.

Poll, M. (1946) Révision de la faune ichthyologique du Lac Tanganika. Annales du Musée du Congo Belge. Série C (Zoologie) (sér. 1), 4 (3): 141-364.

Poll, M. (1952) Poissons de rivières de la région des lacs Tanganika et Kivu recuellis par G. Marlier. Revue de Zoologie et de Botanique Africaines, 46  (3-4): 221-236.

Poll, M. (1956) Poissons Cichlidae. Exploration Hydrobiologique du Lac Tanganika (1946-1947). Résultats scientifiques, 3 (5B): 1-619.

Rüber, L. (1998) Herkunft der Tanganjikasee-Cichliden. Pp. 10-13 in: Stawikowski, R. (ed.) Tanganjikasee, DATZ-Sonderheft, 1998. Eugen Ulmer-Verlag, Stuttgart.

Rosenberg, N. A. (2002) The probability of topological concordance of gene trees and species trees. Theoretical Population Biology, 61(2): 225-247.

Rosenberg, N. A. and Nordborg, M. (2002) Genealogical trees, coalescent theory and the analysis of genetic polymorphisms. Nature Reviews Genetics, 3 (5): 380-390.

Scheuermann, H. (1975) Reich an Varianten ist Tropheus moorii. Das Aquarium, 9: 530-535; no. 78.

Scheuermann, H. (1976a) Der taxonomische Status von Tropheus moorii (Boulenger, 1898), T. annectens (Boulenger, 1900) und T. brichardi (Nelissen & Thys van den Audenaerde, 1975). DCG-Informationen, 7 (12): 226-230. Deutsche Cichliden-Gesellschaft e. V.

Scheuermann, H. (1976b) Die Varianten von Tropheus moorii. Aquarien Terrarien, 23 (12): 400-405. Urania-Verlag, Leipzig.

Schupke, P. (2003) Aqualog. African Cichlids II, Tanganyika I, Tropheus. (Cichlids of Lake Tanganyika, Part 1, The species of the genus Tropheus). Translation from German to English by Mary Bailey. Original title: Buntbarsche des Tanganjikasees, Teil 1: Die Arten der Gattung Tropheus. Verlag A. C. S. GmbH, Rodgau, Germany, 190 pp.

Schneidewind, F. (1993) Auf der Suche nach dem Ursprung - Phylogenetische Betrachtungen zur gattung Tropheus. DCG-Informationen, 24 (7): 141-151. Deutsche Cichliden-Gesellschaft e. V.

Schneidewind, F. (2005) Miniatlas. Tropheus. Bede - Verlag, Ruhmannsfelden, 336 pp.

Schreyen, M. (2001) Explorations on Lake Tanganyika: Safari to Ubwari. Cichlid News, 10 (3): 6-11. Aquatic Promotions Division, Miami, FL.

Seehausen, O. (1993) Beobachtungen an Vertretern der Gattungsgruppe Tropheini. DATZ, 46 (3): 168-172. Eugen Ulmer-Verlag, Stuttgart / Urania-Verlag, Leipzig.

Sefc, K. M., Baric, S., Salzburger, W. and Sturmbauer, C. (2007) Species-specific population structure in rock-specialized sympatric cichlid species in Lake Tanganyika, East Africa. Journal of Molecular Evolution, 64 (1): 33-49. (Epub 09 Dec 2006)

Snoeks, J., Rüber, L. and Verheyen, E. (1994) The Tanganyika problem: comments on the taxonomy and distribution patterns of its cichlid fauna. Pp. 355-372 in: Martens, K., Goddeeris, B. and Coulter. G. (eds.) Speciation in ancient lakes. Archiv für Hydrobiologie - Advances in Limnology, 44.

Staeck, W. (1974) Eine neue geographische Rasse des Brabantbuntbarsches. Aquarien Magazin, 8 (12): 504–509. Kosmos-Verlag, Stuttgart.

Staeck, W. (1975) Als Duboisi-Fänger in Kigoma - Der Weißpunkt-Brabantbuntbarsche und sein Biotop. Aquarien Magazin, 9 (11): 470-477. Kosmos-Verlag, Stuttgart.

Staeck, W. (1977) Cichliden: Verbreitung, Verhalten, Arten. Band 2. Engelbert Pfriem Verlag, Wuppertal-Elberfeld, Germany, 296 pp.

Staeck, W. (1985) Cichliden: Tanganjika-See. Engelbert Pfriem Verlag, Wuppertal, Germany, 124 pp.

Staeck, W. (2008) Half a century of experience with Tropheus species: A summary, part 1. Tropical Fish Hobbyist, 57 (4): 78-83; no. 633. [December 2008]

Staeck, W. (2009) Half a century of experience with Tropheus species: A summary, part 2. Tropical Fish Hobbyist, 57 (5): 82-86; no. 634. [January 2009]

Staeck, W. and Linke, H. (1994) African Cichlids II: Cichlids from eastern Africa. 1st completely revised edition. Tetra-Press, Melle, Germany, 199 pp.

Stiassny, M. L. J. and Meyer, A. (1999a) Cichlids of the Rift Lakes. Scientific American, 280 (2): 64–69.

Stiassny, M. L. J. and Meyer, A. (1999b) Buntbarsche - Meister der Anpassung. Spektrum der Wissenschaft, 22 (6): 36-43.

Sturmbauer, C. (2003) The evolution of the genus Tropheus. Pp. 26-27 in: Schupke, P. (ed.) Aqualog. African Cichlids II, Tanganyika I, Tropheus. (Cichlids of Lake Tanganyika, Part 1, The species of the genus Tropheus). Translation from German to English by Mary Bailey. Original title: Buntbarsche des Tanganjikasees, Teil 1: Die Arten der Gattung Tropheus. Verlag A. C. S. GmbH, Rodgau, Germany, 190 pp.

Sturmbauer, C. and Meyer, A. (1992) Genetic divergence, speciation and morphological stasis in a lineage of African cichlid fishes. Nature, 358 (6387): 578-581.

Sturmbauer, C. and Meyer, A. (1993) Mitochondrial phylogeny of the endemic mouthbrooding lineages of cichlid fishes from Lake Tanganyika in eastern Africa. Journal of Molecular and Biological Evolution, 10 (4):751-768.

Sturmbauer, C., Hainz, U., Baric, S., Verheyen, E. and Salzburger, S. (2003) Evolution of the tribe Tropheini from Lake Tanganyika: synchronized explosive speciation producing multiple evolutionary parallelism. Hydrobiologia, 500 (1-3): 51-64.

Sturmbauer, C., Koblmüller, S., Sefc, K. M. and Duftner, N. (2005) Phylogeographic history of the genus Tropheus, a lineage of rock-dwelling cichlid fishes endemic to Lake Tanganyika. Hydrobiologia, 542 (1-3): 335–366; doi:10.1007/s10750-004-4664-y

Sturmbauer, C., Verheyen, E., Rüber, L. and Meyer, A. (1997) Phylogeographic patterns in populations of cichlid fishes from rocky habitats in Lake Tanganyika.  Pp. 97-111 in: Kocher, T. D. and Stepien, C. A. (eds.) Molecular Systematics of Fishes. Academic Press, San Diego, CA, xi + 314 pp.

Takahashi, K., Terai, Y., Nishida, M. and Okada, N. (2001) Phylogenetic relationships and ancient incomplete lineage sorting among cichlid fishes in Lake Tanganyika as revealed by analysis of the insertion of retroposons. Molecular Biology and Evolution, 18 (11): 2057–2066.

Turner, G. F., Seehausen, O., Knight, M. E., Allender, C. J. and Robinson, R. L. (2001) How many species of cichlid fishes are there in African lakes? Molecular Ecology, 10 (3): 793–806.

Yanagisawa, Y. and Nishida, M. (1991) The social and mating system of the maternal mouthbrooder Tropheus moorii (Cichlidae) in Lake Tanganyika. Japanese Journal of Ichthyology, 38 (3): 271–282.

Zadenius, M. (1981) Tropheus. Ciklidbladet, 14 (6): 5-35. Nordiska Ciklidsällskapet.

Zadenius, M. (1991) Importnytt - Nyheter från Åldes Akvarium. Ciklidbladet, 24 (9): 42-45. Nordiska Ciklidsällskapet.

Zadenius, M. (1992a) Tanganyikasjöns skönheter - Tropheus-släktet, del 2. Ciklidbladet, 25 (1): 14-22. Nordiska Ciklidsällskapet.

Zadenius, M. (1992b) Nya fakta om Tropheus-släktet. Akvariet, 66 (3): 69-71