Reproductive Guilds and the Ultimate Structure of Fish Taxocenes

Journal of Fish Biology (2006) 69, 622-628 doi:10.1111/j.1095-8649.2006.01120.x, available online at http://www.blackwell-synergy.com

Translocations of rocky habitat cichlid fishes to Nkhata Bay, Lake Malawi

M . J. G e n n e r * | , A . Botha} and G . F . T u r n e r *

* Department of Biological Sciences, University of Hull, Hull HU6 7RX, U.K. and }Aqua Africa, P. O. Box 209, Nkhata Bay, Malawi

(Received 11 April 2005, Accepted 22 February 2006)

Over the last 10 years, 12 rocky habitat ‘mbuna’ taxa have been translocated to Nkhata Bay on the north-western shores of Lake Malawi from other regions of the lake. Given potential for competition and hybridization with indigenous taxa, further translocations should be avoided. © 2006 The Authors Journal compilation © 2006 The Fisheries Society of the British Isles

Key words: community diversity; dispersal; invasion; long-term change.

Littoral rocky habitats of Lake Malawi are dominated by the ‘mbuna’ group of endemic cichlid fishes. These fishes do not readily traverse habitat boundaries such as sandy shorelines or deep water (Ribbink e t a l., 1983; van Oppen e t a l., 1997), and many species and phenotypically distinct races have highly localized distributions (Ribbink e t a l., 1983). Since 1996, 12 mbuna taxa have been recorded at Nkhata Bay on the north-western shores of Lake Malawi that were absent from previous surveys (Ribbink e t al., 1983; Genner e t a l., 1999, 2004). Source localities for translocated taxa were identified from published photographs (Konings, 2001; Spreinat, 2002); all appear to originate from the eastern or north-western shores of Lake Malawi and are exported as orna mental fishes. Of the 12 translocated taxa, 11 are species new to the area, while Cynotilapia afra (Gunther) from Cobue is considered conspecific with an indig enous population (Konings, 2001), but differs in male nuptial colouration (Table 1). Preliminarily observations using scuba revealed the translocated taxa to be most abundant between 2 and 6 m depth. To determine broad distributions of mbuna around Nkhata Bay, a 30 min snorkel survey was conducted over each of five stretches of shoreline during November 2004 (Fig. 1). The greatest concen tration of translocated taxa was at Chirundu Point. During this survey, and dives for other purposes, all 12 taxa (Fig. 2) have been observed there and eight appear to be restricted to the location (Table 1). Mbuna abundances at Chirundu Point

fAuthor to whom correspondence should be addressed. Tel.: +44 (0) 1482 465962; fax: +44 (0) 1482 465458; email: [email protected] 622 © 2006 The Authors Journal compilation © 2006 The Fisheries Society o f the British Isles LAKE MALAWI CICHILD TRANSLOCATIONS 623

T a b l e I. Locations of translocated taxa in Nkhata Bay and inferred source localities (see Fig. 1 for location codes)

Locations Year first during Additional observed Location 2004 locations Translocated Inferred in Nkhata first snorkel observed species/populations source site Bay seen surveys 1996-2004 Cynotilapia afra Cobue;, Mozambique 2004 2 2 — (Gunther 1893) Cynotilapia ‘mbweca’ Cobue;, Mozambique 2004 2 — — Cynotilapia ‘lion’ Lions Cove, Malawi 1997 2 2 — Cynotilapia pulpican Likoma Island, 2004 2 2 — Tawil 2002 Malawi Melanochromis Chisumulu Island, 1996 2 2, 3 interruptus Malawi Johnson 1975 Melanochromis Likoma Island, 2004 2 2 1 joanjohnsonae Malawi (Johnson 1974) Pseudotropheus Southern Tanzania 2004 2 2 — ‘daktari’ Pseudotropheus Chewere, Malawi 2004 2 —— ‘elongatus chewere’ Pseudotropheus Likoma or 2002 3 2, 3 — ‘elongatus ornatus’ Chisumulu Island Pseudotropheus Likoma Island, 1997 5 5 2 (Maylandia) aurora Malawi Burgess 1976 Pseudotropheus Likoma Island, 2004 2 (Maylandia) Malawi hajomaylandi Meyer & Schartl 1984 Pseudotropheus Cobue, 2004 2 (Maylandia) Mozambique ‘hajomaylandi Cobue’

Mbuna indigenous to Nkhata Bay: Cyathochromis obliquidens Trewavas 1935, Cynotilapia afra (Gunther 1893), Cynotilapia axelrodi Burgess 1976, Cynotilapia ‘mbamba’, Genyochromis mento Trewavas 1935, Gephyrochromis lawsi Fryer 1957, Labeotropheus fuelleborni Ahl 1926, Labeotropheus trewavasae Fryer 1956, Labidochromis caeruleus Fryer 1956, Labidochromis maculicauda Lewis 1982, Pseudotropheus crabro (Ribbink & Lewis 1982), Melanochromis parallelus (Burgess & Axelrod 1976), Melanochromis robustus Johnson 1986, Petrotilapia genalutea Marsh 1983, Petrotilapia ‘small blue’, Petrotilapia tridentiger Trewavas 1935, Pseudotropheus elongatus Fryer 1956, Pseudotropheus ‘elongatus ruarwe’, Pseudotropheus fuscus Trewavas 1935, Pseudotropheus (Maylandia) livingstonii (Boulenger 1899), Pseudotropheus (Maylandia) callainos Stauffer & Hert 1992, Pseudotropheus (Maylandia) zebra (Boulenger 1899), Pseudotropheus (Maylandia) ‘zebra gold’, Pseudotropheus minutus Fryer 1956, Pseudotropheus (Tropheops) ‘band’, Pseudotropheus (Tropheops) ‘black’, Pseudotropheus (Tropheops) ‘deep’, Pseudotropheus (Tropheops) lucerna Trewavas 1935, Pseudotropheus (Tropheops) ‘mauve’, Pseudotropheus (Tropheops) ‘olive’, Pseudotropheus (Tropheops) ‘rust’, Pseudotropheus tursiops Burgess & Axelrod 1975, Pseudotropheus williamsi (Gunther 1894) and Pseudotropheus ‘zebra dwarf’.

F ig. 1. The study area: (a) Lake Malawi and (b) Nkhata Bay. Locations of the five snorkel surveys labelled 1 to 5. During surveys at: site 1, only 23 indigenous taxa were present; site 2, 22 indigenous taxa and seven translocated taxa were present; site 3, 25 indigenous taxa and two translocated taxa were present; site 4, only 22 indigenous taxa were present; site 5, 17 indigenous taxa and one translocated taxon were present. In total 34 indigenous mbuna species have been observed in Nkhata Bay between 1996 and 2004. were determined within twelve 6 x 6 m transects laid at 500 m intervals and 3 m depth on 2nd November 2004 using scuba. Mean abundance of translo cated taxa was 0 03 m~2 (range: 0-011 m~2) while that of indigenous taxa was 2 03 m~2 (range: 1 08-2-78 m-2). On average, translocated taxa comprised 5-4% of all taxa (range: 0-16-7%) and 1-5% of all individuals (range: 0-4-4%). The high diversity of translocated taxa at Chirundu Point indicates it is the main introduction site. Exporters regularly hold fishes caught from other loca tions in mesh-covered plastic barrels suspended in the lake on the northern side of Chirundu Point before transportation to export stations. Only Pseudotropheus (Maylandia) aurora Burgess 1976 had a disjunct distribution, the two Nkhata Bay populations were separated by long stretches of rocky habitat, implying two separate introductions have taken place. Pseudotropheus ‘elongatus ornatus’, Melanochromis interruptus Johnson 1975 and Melanochromis joanjohnsonae (Johnson 1974) were also observed at more than one site, but these were all directly adjacent suggesting dispersal across deep water or sandy habitat. Despite evidence of low dispersal rates from molecular analyses (van Oppen e t a l., 1997), and restricted distributions of most mbuna species (Konings, 2001), mbuna do have a limited capability

F ig. 2. Taxa translocated to Nkhata Bay, all photographs are of adult males taken in Nkhata Bay: (a) Cynotilapia afra (from Cobuei), (b) Cynotilapia ‘lion’, (c) Cynotilapia ‘mbweca’, (d) Cynotilapia pulpican, (e) Pseudotropheus (Maylandia) hajomaylandi, (f) Pseudotropheus (Maylandia) ‘hajomaylandi Cobue’, (g) Pseudotropheus (Maylandia) aurora, (h) Pseudotropheus ‘daktari’, (i) Melano- chromis joanjohnsonae, (j) Melanochromis interruptus, (k) Pseudotropheus ‘elongatus chewere’ and (l) Pseudotropheus ‘elongatus ornatus’.

© 2006 The Authors Journal compilation © 2006 The Fisheries Society of the British Isles, Journal of Fish Biology 2006, 69, 622-628 626 M. J. GENNER E T AL. to migrate across sandy habitats. When artificial reefs were placed on sand 1 km from the nearest rocky habitat at Otter Point in the south of the lake, within a year five mbuna species had colonized, and within 5 years 10 species had established themselves (McKaye & Gray, 1984). During the late 1960s or 1970s, at least 13 species were translocated from natural locations to Thumbi West Island (14°01-260 S; 34°49-250 E; 275 km south of Nkhata Bay) as a result of the ornamental fish trade (Ribbink e t a l., 1983; Genner & Turner, 2005). To forecast changes in the structure of the Nkhata Bay mbuna community, long-term abundance and diversity changes following the translocations at Thumbi West were examined using two published data sets. Survey dates from the first study (Ribbink e t a l., 1983) were not reported, so it was assumed that they were conducted in the year of publication. Repeat surveys were conducted between December 1998 and May 1999 (Duponchelle e t a l., 2000). Two sites corresponded in both surveys, labelled in Ribbink e t a l. (1983) as Thumbi West sites B and C, and Duponchelle e t a l. (2000) as T8 and T13, respectively. Both studies surveyed territorial male fishes within 25 x 2 m transects at depths of 2, 6 and 10 m. Mean abundance of each species at each site was derived from summary figures in Ribbink e t al. (1983) and from tabulated data of five sampling events in Duponchelle e t al. (2000). Two-way ANOVA tests of the Thumbi West data revealed that with increas ing depth there were significant declines in total number of species [P < 0 01; Fig. 3(f)], and number of indigenous species [P < 0 05; Fig. 3(d)], but not in the number of translocated species [P > 0 05; Fig. 3(b)]. They also revealed sig nificant increases in the total number of species between 1983 and 1998-1999 [P < 0 01; Fig. 3(f)] and the number of translocated species [P < 0 001; Fig. 3(b)], but no change in the number of indigenous species [P > 0 05; Fig. 3(d)]. With increasing depth there were significant declines in total number of individual territorial males [P < 0 01; Fig. 3(e)] and number of males of indigenous spe cies [P < 0 01; Fig. 3(c)], but no change in num ber of males of translocated species [P > 0 05; Fig. 3(a)]. There were also significant increases between 1983 and 1998-1999 in total number of territorial males of translocated species [P < 0 01; Fig. 3(a)], but no change in the number of males of indigenous spe cies [P > 0 05; Fig. 3(c)] or in total num bers of males of all species combined [P > 0-05; Fig. 3(e)]. Assuming that abundance trends of the translocated species at Thumbi West can be used an indicator of the future at Nkhata Bay, it is likely that abun dance of many of the translocated taxa will increase considerably. From a con servation perspective, a major concern is that competitive interactions between translocated and indigenous taxa may result in competitive exclusion. The lack of extirpations or significant declines in abundance of indigenous territorial males at Thumbi West over a 16 year period, despite increases in number and abundance of translocated species, implies that males of invading species are able to occupy available space at little or no cost to indigenous males. Since no published data assessing female abundance are available for comparison, it is unclear if total adult population sizes of indigenous taxa have been affected. An additional conservation concern is introgression between translocated and indigenous taxa (Streelman e t a l., 2004). Laboratory first generation hybrid progeny tend to be intermediate between parental forms in anatomy (Albertson

e 150 25

120 & 20

90 '3 15

60 o 10

30 5

0 S3 0

^ 160 25

20 120 15 80 10 40 5

0 0 10 2 Transect depth (m)

Fig. 3. Changes of the Thumbi West Island mbuna community between 1983 (■) and 1998-1999 (Q). Mean number ± s.d. of: (a) territorial males belonging to translocated species, (b) translocated species, (c) territorial males belonging to indigenous species, (d) indigenous species, (e) territorial males belonging to all species and (f) all species. Data from Ribbink et al. (1983) and Duponchelle et al. (2000).

& Kocher, 2005) and breeding colours (pers. obs.), and using these criteria there is no evidence of hybridization between translocated and indigenous taxa at Nkhata Bay. Given the high degree of uncertainty concerning the effects of translocated taxa on indigenous fishes, care should be taken to avoid further introductions.

We thank the Fisheries Department of the Government of Malawi, especially S. Chimatiro and M. Banda. D. Botha, S. De Pooter and P. Nichols helped with field work. The manuscript was improved with comments from two reviewers.

References Albertson, R. C. & Kocher, T. D. (2005). Genetic architecture sets limits on transgressive segregation in hybrid cichlid fishes. Evolution 59, 686-690. Duponchelle, F., Ribbink, A. J., Msukwa, A., Mafuka, J. & Mandere, D. (2000). The potential influence of fluvial sediments on rock-dwelling fish communities. In Fish Ecology Report. Lake Malawi/Nyasa/Niassa Biodiversity Conservation Project (Duponchelle, F. & Ribbink, A. J., eds), pp. 227-255. Washington, DC: South African Development Community, Gabarone, Botswana and Global Environment Facility. Genner, M. J. & Turner, G. F. (2005). The mbuna cichlids of Lake Malawi: a model for rapid speciation and adaptive radiation. Fish and Fisheries 6, 1-34. Genner, M. J., Turner, G. F. & Hawkins, S. J. (1999). Resource control by territorial male cichlid fish in Lake Malawi. Journal of Animal Ecology 68, 522-529. Genner, M. J., Taylor, M. I., Cleary, D. F. R., Hawkins, S. J., Knight, M. E. & Turner, G. F. (2004). Beta diversity of rock-restricted cichlid fishes in Lake Malawi: importance of environmental and spatial factors. Ecography 27, 601-610. Konings, A. (2001). Malawi Cichlids in their Natural Habitat, 3rd edn. El Paso, TX: Cichlid Press. McKaye, K. R. & Gray, W. N. (1984). Extrinsic barriers to gene flow in rock-dwelling cichlids of Lake Malawi: macrohabitat heterogeneity and reef colonisation. In Evolution of Species Flocks (Echelle, A. A. & Kornfield, I., eds), pp. 169-181. Maine, ME: University of Orono Press. van Oppen, M. J. H., Turner, G. F., Rico, C., Deutsch, J. C., Ibrahim, K. M., Robinson, R. L. & Hewitt, G. M. (1997). Unusually fine-scale genetic structuring found in rapidly speciating Malawi cichlid fishes. Proceedings of the Royal Society of London B 264, 1803-1812. Ribbink, A. J., Marsh, B. J., Marsh, A. C., Ribbink, A. C. & Sharp, B. J. (1983). A preliminary survey of the cichlid fishes of the rocky habitats of Lake Malawi. South African Journal of Zoology 18, 155-309. Spreinat, A. (2002). Aqualex-catalog Cichlids from Lake Malawi, 2nd edn. Ettlingen: Dahne Verlag. Streelman, J. T., Gmyrek, S. L., Kidd, M. R., Kidd, C., Robinson, R. L., Hert, E., Ambali, A. & Kocher, T. D. (2004). Hybridization and contemporary evolution in an introduced cichlid fish from Lake Malawi National Park. Molecular Ecology 13, 2471-2479.