Springer-VerlagTokyohttp://www.springer.de101640289-07711439-5444Journal

of EthologyJ

EtholLifeSciences20010.1007/s10164-006-0200-z

J Ethol (2007) 25:21–27 DOI 10.1007/s10164-006-0200-z

© Japan Ethological Society and Springer-Verlag Tokyo 2006

ARTICLE

Kazunori Matsumoto • Masanori Kohda

Male foraging avoidance in female feeding territories in a harem polygynous cichlid in Lake Tanganyika

Received: May 9, 2005 / Accepted: February 9, 2006 / Published online: March 21, 2006 Japan Ethological Society and Springer-Verlag 2006

Abstract We studied foraging site partitioning between the sexes in Neolamprologus tetracanthus, a shrimp-eating Tanganyikan cichlid with harem-polygyny. Females maintained small territories against heterospecific food competitors within large territories of males, foraging exclusively at the inner side of their own territories (foraging areas). Males fed as frequently as females in their own territories, but mostly outside female foraging areas, although they frequently entered female territories and repelled food competitors from the territories. Soon after removal of the resident females, however, harem males, as well as many food competitors, invaded the vacant territories and intensively devoured prey of female foraging areas. This indicates that although female foraging areas appear to contain more food than outside the areas, harem males refrained from foraging there when the resident females were present. We suggest that harem males will attempt to keep female foraging areas in good condition, whereby they may get females to reside in male territories and/or promote female gonadal maturation. Key words Cichlid · Feeding site partitioning · Haremic polygyny · Neolamprologus tetracanthus · Reproductive tactics · Territory

Introduction In fishes with monogamous mating systems (Barlow 1984), male reproductive success largely depends on the size and number of clutches produced by females (Barlow 1991; Clutton-Brock 1991; Keenleyside 1991). Many monogaK. Matsumoto1 (*) · M. Kohda Department of Bio- and Geosciences, Graduate School of Science, Osaka City University, Sumiyoshi-ku, Osaka 558-8585, Japan 1 Present address: Mimuro 658-10, Gose, Nara 639-2306, Japan Tel. +81-745-652437; Fax +81-745-652437 e-mail: [email protected]

mous males attempt to increase the number of gametes produced by their mates (Clutton-Brock 1991). One of the ways to obtain more gametes is to improve female feeding conditions (Wootton 1998). For example, males of the monogamous reef fish Oxymonacanthus longirostris swim and forage in association with a small vulnerable mate, during which time the female feeding rate significantly increases in comparison with solitary foraging, probably due to reduced predation pressure (Kokita and Nakazono 1999). Males of mouthbrooding cichlids actively participate in parental care, resulting in partial emancipation of the pairing females from caregiving, thus allowing the females to spend a lot of time foraging (Yanagisawa 1986; Kuwamura et al. 1989; Kuwamura 1997). In the herbivorous cichlid Tropheus moorii with a permanent territory, males allow a female to forage intensively inside their quality territories after establishing the mating bond (Yanagisawa and Nishida 1991). In these cases, the increased food intake of females brought by male contribution has been thought to heighten the female potential reproductive rate, and, in turn, heighten the reproductive success of the males (Balshine-Earn 1995; Neat and Balshine-Earn 1999). In contrast, mating success of haremic males largely depends on the number of their mates, and the males devote their energy to increasing their harem size in competition with rival males (Warner 1984, 1988; Turner 1993; Andersson 1994; Sato 1994). The harems of many haremic fish species with territorial females have a long-term membership, and the number of females in a harem does not increase freely (e.g., Yanagisawa 1987; Kuwamura 1997; Ohnishi et al. 1997). This social situation implies that female fecundity, primarily affected by food intake, should be one of the factors limiting male reproductive success in such haremic species. Thus, it is expected that males of haremic fishes with territorial females would behave so in order to improve female feeding conditions as well as to increase harem size. However, little attention has been paid to such attempts of the males in haremic species. Neolamprologus tetracanthus, a substrate breeder with harem-polygyny, inhabits the shallow waters of Lake Tanganyika (Kuwamura 1986, 1997; Brichard 1989; Kohda

22

et al. 1996). Both sexes of this fish feed on shrimps on the lake bottom (M. Hori, 1990, personal communication). Inside large territories of males, females maintain small feeding territories against many food competitors (Matsumoto and Kohda 1998, 2004). During field observations, we noticed that haremic males appeared to avoid foraging inside female territories, although they frequently entered the territories. If males avoid foraging in female territories although they can potentially forage there, it could be thought of as an attempt by the males not to reduce food resources within female territories. The present study reports whether N. tetracanthus males do indeed refrain from foraging inside female territories. We tested this by conducting a female removal experiment, in which male foraging sites were compared between, before, and after female removal.

Materials and methods Subject fish Neolamprologus tetracanthus is a small fish, up to 14 cm in total length (TL), belonging to the tribe Lamprologini. Many males have large territories enclosing multiple female territories. Large males tend to have a large harem size, up to 14 females at a time (Matsumoto and Kohda 1998). Female territories include sandy substrate for foraging sites and several stones for spawning sites. Females ready to spawn dig a hole of ca. 20 cm deep along the side of a stone partly buried in the sand, exposing the vertical surface of the stone on which ca. 50–100 eggs are deposited (Matsumoto and Kohda 1998; K. Matsumoto, personal observation). Females in this population guard the brood from brood predators exclusively for ca. 1.5 months (Matsumoto and Kohda 1998). This fish feeds on small shrimps of 3–5 mm in size by pecking on the sandy substrate. Field observations Field studies were carried out at Wonzye Point near Mpulung in Zambia, at the southern end of Lake Tanganyika, from the end of August to the middle of November 1992. Water temperature during the study period ranged from 24 to 28 °C. A total of 63 cichlid species and 10 non-cichlid fish species are recorded at this location (Kuwamura and Karino 1991). A 10 m × 14-m study area with 2 m × 2 m grids was laid out on the sandy bottom with rocks and stones at a water depth of 5–7 m, where many fish inhabited. Underwater observations were conducted during 09:00–15:00 h with the aid of SCUBA. Fish behavior was recorded on topographic maps of the lake bottom drawn at a scale of 1:20. Neolamprologus tetracanthus could be individually identified by the pattern of white spots on the body. The TL was estimated underwater to the nearest 0.5 cm using a scale bar placed near the fish. The sexes were identified based on

courtship interactions. In the study area, 54 females were recorded at the beginning of the study period (see Matsumoto and Kohda 1998). Of these, 45 females had territories enclosed by those of 7 harem males, and 11 females in the harems disappeared within 1 month. We conducted 10-min observations on each individual 4– 30 times (mean ± SD = 12.8 ± 5.6) for the 34 females and 2– 13 times (6.3 ± 5.1) for the 7 harem males. We recorded individual positions every 10 s, location of feeding pecks and intra- and interspecific aggression. The swimming range of each fish was demarcated by connecting the outermost points of the fish positions. N. tetracanthus defended their swimming ranges against conspecifics of the same sex and also against various heterospecific food competitors in the case of the females, and the defense sites nearly coincided with the outline of the ranges (Matsumoto and Kohda 1998, 2004). Thus, the outline of the swimming ranges was defined as the boundary of a territory. During the study period, the territories of the 7 males and 34 females were stably maintained. Harem males occasionally visited the female territories. The visits of the males (entering a female territory or an approach to within 10 cm from the territory boundaries) were also recorded while observing females. Of the 34 females, 7 females spawned eggs, and 5 of them guarded fry. The behavioral data on breeding females were omitted from analyses. To examine the foraging site utilization of N. tetracanthus in detail, 10 cm × 10 cm grid sections over the study area were drawn on the topographic maps. ‘Female foraging areas’ were defined as sets of the sections that contained female feeding pecks. For the six females used in the removal experiment (see below), mean feeding frequency (pecking number per min) of each grid section was calculated. The sections were categorized into six grades, from 0 to ε, according to the feeding frequencies (0 pecks per min < α ≤ 0.03 < β ≤ 0.06 < γ ≤ 0.09 < δ ≤ 0.12 < ε). When comparing male feeding frequencies between, inside, and outside female foraging areas before female removal experiment, we calculated the frequencies per grid section for each male so as to remove the effect of the size difference between, inside, and outside the foraging areas in male territories. In the justification, we used neither the number of feeding pecks and grid sections outside male territories nor the number of grid sections containing only rocky substrate unsuitable for feeding.

Female removal experiment To examine whether males refrain from foraging inside female territories, a female removal experiment was conducted in November. Six females (a–f), randomly chosen in four harems (A–D) (Fig. 1), were removed from their territories using a gill net. After female removal, a 30-min observation was conducted for each vacant territory, and we recorded intrusions by con- and heterospecific fishes into the vacant territories, interspecific aggression, and feeding pecks of conspecifics. After each observation, the removed females were returned to each territory.

23 Fig. 1. Territory arrangement, the positions of male feeding pecks and female foraging areas for Neolamprologus tetracanthus in the study area. Each territory is represented by a polygon: thick lines for harem males (A– G) and thin lines for females. Thick dotted lines show territories of small males without females. Six females (a–f) were used in the removal experiment. Dots and shaded areas indicate male feeding pecks and female foraging areas (sets of 10 cm × 10 cm grid sections with female feeding pecks), respectively

G D C

f

F e c b A a 2m

Results

E

d

B

males did not respond to these competitors passing near female territories.

Territories In the study area, 7 N. tetracanthus males encompassed 2– 12 female territories (Fig. 1). In the harems, males were larger than females (mean ± SD = 12.4 ± 0.9 cm in TL, n = 7 for males, 9.5 ± 0.6, n = 34 for females, Mann–Whitney Utest, z = 4.21, P < 0.0001). Males attacked conspecific males near the territory boundaries with a mean frequency of 0.02 times per min (±0.02 SD, n = 7), repelling them from their territories. Females maintained territories partially surrounded by some rocks on the sandy substrate, attacking and repelling various other fish species (total 35 species in 24 genera) with a mean frequency of 0.64 times per min (±0.3 SD, n = 34). Female territories did not tightly adjoin each other; agonistic behavior among neighboring females was rarely observed (mean ± SD = 0.01 ± 0.02 times per min, n = 34). Males visited females one after another. Females were visited by the harem male with a mean frequency of 0.19 times per min (±0.11 SD, n = 34). Although females approached the visiting males, they did not show any agonistic behavior toward the males, which stayed still for a short duration (usually <30 s) in or near the territories. The visiting males attacked other fish species (total five species in five genera) intruding into female territories or foraging there; the attack frequency was 0.14 times per min (±0.13 SD, n = 7). The attacked species were mainly shrimp eaters (69.1% of the total 55 aggressive encounters) such as Lamprologus callipterus and Lobochilotes labiatus, whereas

Foraging sites Feeding frequency (pecking number per min) did not differ between males (mean ± SD = 0.74 ± 0.44 times, n = 7) and females (0.99 ± 0.33, n = 34, Mann–Whitney U-test, z = 1.13, P > 0.2). Pecking points of males were scattered all over the territories (Fig. 1). The detailed examination of male foraging sites indicated that 62.0% of the total pecks (±21.9 SD) was performed outside female territories, 21.6% (±11.5) inside female territories but outside female foraging areas and 16.4% (±14.2) inside the foraging areas. Males pecked more frequently outside female foraging areas than inside the areas (Wilcoxon signed-ranks test, z = 2.37, P < 0.05, Table 1). Male feeding frequencies per grid section were also higher outside female foraging areas than inside the areas except for male G (Wilcoxon signed-ranks test, z = 1.18, P > 0.2 as for all males, but z = 2.20, P < 0.05 as for males A–F, Table 2). Male G had a small territory with many rocks, and the number of grid sections containing sandy substrate suitable for feeding was smaller than in the other male territories. Removal experiment Within a few minutes after female removal, heterospecific fishes began to intrude into the vacant territories. A total of

24 Table 1. Feeding frequency of Neolamprologus tetracanthus males inside and outside female foraging areas before female removal Males

Observation time (min)

A B C D E F G Average ± SD a

120 20 120 100 30 20 20

Number of feeding pecks per minute Inside female foraging areasa

Outside female foraging areas

0.18 0.30 0.18 0.02 0 0.05 0.15 0.12 ± 0.11

0.88 1.00 0.72 0.21 0.30 1.00 0.20 0.61 ± 0.37

Sets of 10 cm × 10 cm grid sections with female feeding pecks

Table 2. Feeding frequency of Neolamprologus tetracanthus males per grid section inside and outside female foraging areas in male territories before female removal Males

Observation time (min)

A B C D E F G Average ± SD a

120 20 120 100 30 20 20

Number of feeding pecks per sectiona per minute Inside female foraging areasb

Outside female foraging areas

0.37 × 10−3 2.59 × 10−3 0.44 × 10−3 0.17 × 10−3 0 0 4.05 × 10−3 (1.09 ± 1.59) × 10−3

0.61 × 10−3 5.36 × 10−3 0.46 × 10−3 0.33 × 10−3 0.21 × 10−3 1.33 × 10−3 0.89 × 10−3 (1.31 ± 1.82) × 10−3

10 cm × 10 cm grid sections containing sandy substrate Sets of the grid sections with female feeding pecks

b

Table 3. Feeding frequency of Neolamprologus tetracanthus males inside and outside female foraging areas after female removal Removed females

a b c d e f Average ± SDb a

Males

A A A B C D

Observation time (min)

30 30 20 30 30 30

Number of feeding pecks per minute Inside female foraging areasa

Outside female foraging areas

1.07 2.03 1.20 0.90 0.77 1.20 1.08 ± 0.31

0.13 0.13 0.40 0 0 0.27 0.12 ± 0.14

Sets of 10 cm × 10 cm grid sections with female feeding pecks n=4

b

22 species in 15 genera invaded the territories with a mean frequency of 3.10 times per min (±0.98 SD, n = 6), devouring benthic animals on the sandy substrate with apparently a quite high feeding frequency in relation to their usual foraging, although unfortunately we did not record the frequency. These fishes included many individuals of three species of exclusive shrimp eaters, L. callipterus, L. labiatus and Xenotilapia sima, and the intrusion frequency of the shrimp eaters into female foraging areas was more than nine times higher after female removal than before (see Matsumoto and Kohda 2004). The harem males also intruded into the vacant territories within 10 min after the removal, and spent almost all their time in foraging during

the observation. All males pecked much more frequently inside female foraging areas than outside the areas (Table 3), although the sample size was small for the analysis via paired-sample test. Moreover, males intensively pecked in the sections where females had frequently pecked and males had only rarely done so before the removal (Fig. 2). The relative number of male feeding pecks at each categorized section differed between, before, and after female removal (Kolmogorov–Smirnov test, male A: χ2=26.2, P < 0.0001; male B: χ2 = 10.2, P < 0.05; male C: χ2 = 10.6, P < 0.01) except for male D (χ2 = 6.8, P > 0.06). Intrusion into a vacant territory by an unidentified female was observed after 10 min of the removal of female f.

25 Fig. 2. Female foraging areas and the positions of male feeding pecks in female territories before and after female removal in Neolamprologus tetracanthus. Each polygon shows female territory. Shaded-in squares (10 cm × 10 cm) indicate female feeding frequencies (pecking number per minute): 0 < α ≤ 0.03 < β ≤ 0.06 < γ ≤ 0.09 < δ ≤ 0.12 < ε. Female feeding pecks were not observed in nonshaded areas in the territories. Solid and open circles indicate male feeding pecks before and after female removal, respectively. Open circles with figures in the territory of female e show the points with plural feeding pecks. Crosses in the territory of female f show the feeding pecks of an unidentified intruder female after the removal

a

b

c

d

e

f

α β γ δ ε

5 7

1m Harem male D never attacked the newcomer female and foraged in the marginal area of her foraging site (Fig. 2). At all of the six vacant territories, males performed agonistic behavior against foraging heterospecifics (total of ten species in seven genera) with a mean frequency of 0.32 times per min (±0.56 SD, n = 4).

Discussion The results of the present study clearly indicate that foraging-site partitioning occurred between harem males and females in N. tetracanthus. Although harem males entered female territories without receiving any agonistic behavior from the territory owners, they did not forage actively in female foraging areas when females were present. However, after female removal, the males intensively foraged in female foraging areas of the vacant territories, indicating that the presence of females discouraged males from feeding in the foraging areas. Thus, it is evident that foraging site partitioning is achieved by male refrainment. Aggression intensity of N. tetracanthus females against food competitors is positively related to the degree of dietary overlap with the latter, and females attack and repel shrimp eaters more intensively than other fishes with different food habits so as to defend their food resources (Matsumoto and Kohda 2004). Shrimp eaters of other species foraged outside female territories with feeding frequencies similar to N. tetracanthus (Hori 1987; K. Matsumoto,

personal observation). After female removal, however, they invaded female territories much more frequently than before the removal (Matsumoto and Kohda 2004), and their feeding frequencies inside the vacant territories were apparently much higher than that outside the territories. The devouring of many fish of heterospecific shrimp eaters strongly suggests that female territories were rich in food resources of benthic prey in comparison with areas outside the territories, since other ecological factors (e.g., predation pressure and feeding site structure) affecting fish feeding activities (Gerking 1994) were likely to be almost uniform over the study area. Females allowed harem males to enter female territories. Then, why do the males refrain from foraging in female foraging areas? We did not experimentally examine the effect of male food consumption on female foraging areas. Shrimps comprise one of the lowest biomass levels within the zoobenthos community within the shoreline of this lake (Yuma and Kondo 1997; Yuma et al. 1998). Both males and females of N. tetracanthus visually aim at a shrimp on the sandy bottom in pecking (Matsumoto and Kohda 1998), and the feeding frequency did not differ between the sexes. Thus, if males target prey inside female foraging areas, food consumption by the males will reduce available food resources in female foraging areas. Male foraging avoidance will probably contribute to provision of good quality foraging sites ensuring sufficient food intake for their mates. Other ecological factors that prohibit males from foraging in female foraging areas just in the presence of the territory owners were not evident in the study area, and we cannot find other

26

scenarios that explain male foraging avoidance. Moreover, harem males defended female territories from food competitors, although they usually foraged outside female territories. This male defense is also understandable in the context of keeping female foraging sites in good condition. Reproductive success of females is limited mainly by available energy for investment in gamete production, and the courting males of many birds and insects offer nutrition to females, resulting in strengthening the pair bond and enhancing female fecundity (Andersson 1994). The foraging refrainment by N. tetracanthus males may result in improvement of the female energy gain essential for gonadal maturation, although the degree of the effect is unknown. Foraging site partitioning between the sexes has been observed in the territorial and herbivorous cichlid Tropheus duboisi, where dominant males do not forage inside female territories that partially overlap with male territories (Yanagisawa 1993). The author suggests that the males yield good feeding sites to their mates so as to promote female gonadal maturation, which is quite similar to the case of N. tetracanthus. The provision of quality foraging sites by males for their mates will have another function, that is, facilitation of the residence of females. In general, females of haremic species (resource defense polygyny) prefer male territories containing quality and/or abundant resources (Orians 1969; Emlen and Oring 1977; Andersson 1994). If N. tetracanthus males target prey in female territories, the reduced territory value may not well compensate females for the cost of territorial defense, thereby inducing females to abandon their territories. In one case of the female removal experiment, a harem male seemed to encourage a newcomer female to reside by giving the best foraging site to her. If males can maintain a quality foraging area with spawning stones, such an area will attract females to reside, so therefore males should maintain it. However, males did not exhibit such attempts regarding the vacant territories after female removal. Harem males visited females one after another, and engaged in the defense of their own large territories against conspecific males. Thus, even if males have a quality foraging area for future mates, they need to leave the area occasionally, thereby exposing it to many other food competitors. Males do not seem to be able to maintain a quality foraging area effectively, and they would scramble to get plenty of food resources in the vacant territories in competition with heterospecific food competitors. In haremic fishes, males adopt some tactics to gain high reproductive success. The major tactic is to increase harem size through obtaining a quality and/or large territory in competition with rival males (Krebs and Davies 1987; Barlow 1991; Clutton-Brock 1991; Andersson 1994). N. tetracanthus males surely adopt this tactic. In some fishes, males attempt to prevent females from leaving the harem (e.g., Kuwamura 1984; Sakai 1997) or arbitrate fights between females (Walter and Trillmich 1994; Kokita and Nakazono 2001) in order to maintain a large harem size. The provision of good foraging sites for females observed in N. tetracanthus is another male reproductive tactic in haremic fishes.

Acknowledgments We thank the staff of Lake Tanganyika Research Unit of the Department of Fisheries, Zambia for their support. This work was financially supported by Grant-in-Aid for Scientific Research (Nos. 040441781, 04044088 and 08640811) from the Japanese Ministry of Education, Science, Sports and Culture and by the Studentship from Osaka City University.

References Andersson M (1994) Sexual selection. Princeton University Press, Princeton Balshine-Earn S (1995) The costs of parental care in Galilee St Peter’s fish, Sarotherodon galilaeus. Anim Behav 50:1–7 Barlow GW (1984) Patterns of monogamy among teleost fishes. Arch FischWiss 35:75–123 Barlow GW (1991) Mating systems among cichlid fishes. In: Keenleyside MHA (ed) Cichlid fishes: behaviour, ecology and evolution. Chapman & Hall, London, pp 173–190 Brichard P (1989) Cichlids and all the other fishes of Lake Tanganyika. TFH, Neptune City Clutton-Brock TH (1991) The evolution of parental care. Princeton University Press, Princeton Emlen ST, Oring LW (1977) Ecology, sexual selection, and the evolution of mating systems. Science 197:215–223 Gerking SD (1994) Feeding ecology of fish. Academic, San Diego Hori M (1987) Mutualism and commensalism in a fish community in Lake Tanganyika. In: Kawano S, Connell JH, Hidaka T (eds) Evolution and coadaptation in biotic communities. University of Tokyo Press, Tokyo, pp 219–239 Keenleyside MHA (1991) Parental care. In: Keenleyside MHA (ed) Cichlid fishes. Chapman & Hall, London, pp 191–208 Kohda M, Yanagisawa Y, Sato T, Nakaya K, Niimura Y, Matsumoto K, Ochi H (1996) Geographical colour variation in cichlid fishes at the southern end of Lake Tanganyika. Environ Biol Fishes 45:237–248 Kokita T, Nakazono A (1999) Pair territoriality in the longnose filefish, Oxymonacanthus longirostris. Ichthyol Res 46:297–302 Kokita T, Nakazono A (2001) Sexual conflict over the mating system: the case of a pair-territorial filefish without parental care. Anim Behav 62:147–155 Krebs JR, Davies NB (1987) An introduction to behavioral ecology, 2nd edn. Blackwell, Oxford Kuwamura T (1984) Social structure of the protogynous fish Labroides dimidiatus. Publ Seto Mar Biol Lab 29:117–177 Kuwamura T (1986) Parental care and mating systems of cichlid fishes of Lake Tanganyika: a preliminary field survey. J Ethol 4:129–146 Kuwamura T (1997) The evolution of parental care and mating systems among Tanganyikan cichlids. In: Kawanabe H, Hori M, Nagoshi M (eds) Fish communities in Lake Tanganyika. Kyoto University Press, Kyoto, pp 57–86 Kuwamura T, Karino K (1991) Littoral fish fauna along the southern coast of Lake Tanganyika, Zambia, with special reference to the family Cichlidae. Ecol Limnol Study Lake Tanganyika Adjacent Reg 7:14–17 Kuwamura T, Nagoshi M, Sato T (1989) Female–male shift of mouthbrooding in a cichlid fish, Tanganicodus irsacae, with notes on breeding habits of two related species in Lake Tanganyika. Environ Biol Fishes 24:187–198 Matsumoto K, Kohda M (1998) Inter-population variation in the mating system of a substrate-breeding cichlid in Lake Tanganyika. J Ethol 16:123–127 Matsumoto K, Kohda M (2004) Territorial defense against various food competitors in the Tanganyikan benthophagous cichlid Neolamprologus tetracanthus. Ichthyol Res 51:354–359 Neat FC, Balshine-Earn S (1999) A field survey of the breeding habits of Eretmodus cyanostictus, a biparental mouthbrooding cichlid in Lake Tanganyika. Environ Biol Fishes 55:333–338 Ohnishi N, Yanagisawa Y, Kohda M (1997) Sneaking by harem masters of the sandperch, Parapercis snyderi. Environ Biol Fishes 50:217–223 Orians GH (1969) On the evolution of mating systems in birds and mammals. Am Nat 103:589–603 Sakai Y (1997) Alternative spawning tactics of female angelfish according to two different contexts of sex change. Behav Ecol 8:372–377

27 Sato T (1994) Active accumulation of spawning substrate: a determinant of extreme polygyny in a shell-brooding cichlid fish. Anim Behav 48:669–678 Turner GF (1993) Teleost mating behaviour. In: Pitcher JJ (ed) Behaviour of teleost fishes, 2nd edn. Chapman & Hall, London, pp 307–331 Walter B, Trillmich F (1994) Female aggression and male peacekeeping in a cichlid fish harem: conflict between and within the sexes in Lamprologus ocellatus. Behav Ecol Sociobiol 34:105–112 Warner RR (1984) Mating behavior and hermaphroditism in coral reef fishes. Am Sci 72:128–136 Warner RR (1988) Sex change in fishes: hypotheses, evidence, and objections. Environ Biol Fishes 22:81–90 Wootton RJ (1998) Ecology of teleost fishes, 2nd edn. Kluwer, Dordrecht Yanagisawa Y (1986) Parental care in a monogamous cichlid Xenotilapia flavipinnis in Lake Tanganyika. Jpn J Ichthyol 33:249–261

Yanagisawa Y (1987) Social organization of a polygynous cichlid Lamprologus furcifer in Lake Tanganyika. Jpn J Ichthyol 34:82– 90 Yanagisawa Y (1993) Long-term territory maintenance by female Tropheus duboisi (Cichlidae) involving foraging during the mouthbrooding period. Ecol Freshw Fish 2:1–7 Yanagisawa Y, Nishida M (1991) The social and mating system of the maternal mouthbrooder Tropheus moorii (Cichlidae) in Lake Tanganyika. Jpn J Ichthyol 38:271–282 Yuma M, Kondo T (1997) Interspecific relationships and habitat utilization among benthivorous cichlids. In: Kawanabe H, Hori M, Nagoshi M (eds) Fish communities in Lake Tanganyika. Kyoto University Press, Kyoto, pp 87–103 Yuma M, Narita T, Hori M, Kondo T (1998) Food resources of shrimpeating cichlid fishes in Lake Tanganyika. Environ Biol Fishes 52:371– 373