Defense Limitations of Single Parents in the Biparental Convict Cichlid Fish: A Field Study

Abstract

A field study on the biparental convict cichlids (Amatitlania siquia) in Lake Xiloá, Nicaragua was conducted to understand how the loss of a parent’s parental care affects the antipredator behavior of both parents and offspring during intruder events. We hypothesized that the combined efforts of two parents would result in increased intruder aggression and decreased offspring dispersion compared to single-parents of either sex, and that single-females and males would differ in their ability to deter predators and manage offspring dispersion. Both parents in a pair chased half the intruders that single-females did and the same number as single-males, suggesting that the presence of a partner deters intruders from encroaching and affords parents more time to engage in other parental care duties. Compared to single-parents, offspring accompanied by both parents were seldom left alone and showed greater shoal cohesion—both of which would presumably lower their risk of predation. Although there were sex differences between single-parents in terms of how often they left their offspring unattended and called to them using pelvic fin-flicks, neither sex was found to be more effective at managing the distribution of their offspring. This field study provides empirical evidence to support the need for biparental care in this species and gives insight into the selection pressures shaping parental investment.

1. Introduction

In contrast to most birds and mammals, where parental care is centered on the provisioning of food to young [1,2], in fish, predation risk to offspring likely drove the evolution of parental care [3,4]. Many cichlid fish species, such as the convict cichlid (Amatitlania siquia), are well known for their biparental care which developed largely to protect free-swimming offspring from predation [5,6,7,8]. In the wild, convict cichlid fry emerges each morning from their parental nest site and moves around the habitat foraging in a dense shoal while being shepherded by their parents. In effect, the two parents develop a mobile defended territory around the brood [9]. As fry move in swarms through high-risk environments, predation pressure may contribute to the low survival rates, with only 20–25% of convict cichlid pairs successfully raising their young to independence [10,11]. This means that to increase rearing success, parents must be constantly vigilant and able to deter predators while herding their wandering young.

Although a widowed or abandoned cichlid parent can take on all parental roles, it is unclear if the total effort of an individual can match the efforts of two parents [10,12,13,14]. Single-parents respond to the loss of their mate by increasing their defensive behavior to varying success [10,13,15,16], with males showing a greater ability to compensate for mate loss, at least in the short term [16]. However, the offspring are still likely to become vulnerable if a single-parent cannot continue to generate enough attacks to repel all predators long-term (e.g., [13]).

When considering only short-term defense against intruders, one parent may be as effective as two parents at protecting their offspring [16]. However, fry shoal distribution is also a major determinant of parental ability to defend offspring and the overall vulnerability of young to predation [17,18]. It is possible that parents must also help maintain the tight cohesion of the fry shoal, in part by flicking their pelvic fins. Pelvic fin-flick behavior has been shown to “call the fry” to the parent, causing the fry to swim closer together into a cohesive swarm that makes them easier to defend [19,20,21,22]. Thus, successful parental care of free-swimming offspring likely requires a combination of behaviors including intruder defense and offspring communication. By only examining these behaviors in isolation it is difficult to fully appreciate parental roles and the effects of mate loss on offspring security.

Here, we examine the effectiveness of a solitary parent in defending against intruders and managing fry shoal dispersion compared to control families with both parents. We examined two components of the parent-to-offspring communication system, the dispersion of the fry and the rate of fin-flicking by the parents. Regarding the fry’s dispersion, we predicted that single-parents, having to perform all activities of both parents, would more likely leave their fry alone for longer periods than paired parents. In this case, fry dispersion would be greater than if both parents were present. We examined the relationship between the number of parents and pelvic fin-flicking. We predicted that a parent would show a higher level of such behavior when their fry shoals are dispersed. Thus, based on our predicted dispersion of fry (see above), we also predicted that two parents together would generate the least fin-flicks compared to both sexes as single parents. However, Sutrisno et al. (2014) [23] pointed out that fin-flicking may be an alert system where parents use olfactory reception to estimate predatory threats. Thus, perhaps, increased pelvic fin-flicking is a response to increased potential predators which would cause a parent to increase its chasing behavior. In this case, we predict that the intensity of the threat, as indicated by the number of intruder chases by a parent(s), would correspond with the number of fin-flicks. To the best of our knowledge, this is the first field study in A. siquia to examine the effect of mate loss from the perspectives of both parent and offspring behavior.

2. Materials and Methods

2.1. Field Site Description

This experiment was conducted in Lake Xiloá, Nicaragua in December 2015 during the long dry breeding season from December to June [9]. Lake Xiloá is a natural volcanic crater lake within the Chiltepe Peninsula Natural Reserve located 25 mi NW of the capital city of Managua. The lake habitat is characterized by a gradual slope, low turbidity, underwater visibility of 3–4 m, and a substrate composed mostly of sand, occasional dense vegetation mats, and rock outcroppings of large and medium-sized rocks often covered in macroalgae. There are nine cichlid species, including convict cichlids, and six non-cichlid species of fish [11]. One of the six non-cichlids is the bigmouth sleeper goby (Gobiomorus dormitor); an ambush predator that often preys on adult and juvenile convict cichlids [11].

2.2. Identification of Convict Cichlid Families and Parental Treatment Group Set up

Convict cichlid pairs with free-swimming offspring were located by snorkeling along 900 m of the Eastern and Southeastern shores of the lake, and data were collected from families found at shallow depths ranging from 0.26 to 1.08 m (mean = 0.69 m). The location of each family’s nest site was marked with a uniquely numbered survey flag to avoid replication. Each family was assigned to one of three parental treatment groups that dictated which parent(s) remained with the offspring (both parents n = 26, female-only n = 10, male-only n = 10). A parent was removed by gently dropping a cast net over the family while swimming over them. The parent assigned for removal was then led to a side of the net, which made it easy to wrap the edge of the net underneath them so that they could be removed. The net was then lifted, taking care to pull it up slowly so as not to disperse the offspring. Captured fish were kept in buckets until data collection finished, at which point they were released near their family. To account for any disturbance caused by removing a parent, 10 of the 26 both-parent treatment families went through the netting process, but without placing any parents in holding buckets. There was no difference between both parent groups that had been netted and those that had not for all dependent variables except one, so they were combined (Intruders chased: t₂₄ = 0.42, p = 0.68; Percent times shoal left alone: t₂₄ = 0.32, p = 0.75; Number of pelvic fin-flicks: t₂₄ = −1.91, p = 0.07; Changes in shoal distribution: t₂₄ = −2.15, p = 0.042).

2.3. Data Collection

Each family replicate was observed while snorkeling from ~1 m distance for 10 min following a 2 min acclimation period. Before the experiment, three practice trials were conducted where both researchers observed the same family simultaneously to ensure that they characterized and quantified behaviors similarly. The following parent and offspring behavioral data were recorded on plastic slates:

• Intruder event: The encroachment of a conspecific or heterospecific fish within an approximate distance of 0.5 m or less of the target family’s territory boundary. Territory area was visually estimated as the area encompassing the parent(s) and fry when no intruders were present.
• Intruder chase: The number of times a conspecific or heterospecific intruder was chased away from the territory by a parent. If a pair of intruders were chased away simultaneously, it was recorded as a single intruder event.
• Times a shoal was left unattended: The percentage of intruder events (intruders within at least 0.5 m of the territory) where a parent was not within at least ~3 body lengths of the offspring shoal.
• Pelvic fin-flicks: The number of pelvic fin-flicks performed by parents during and within 5 s immediately after an intruder event. A fin-flick was characterized as the quick anterior extension of the paired pectoral fins while simultaneously exerting forward thrusts using the anal and caudal fins leading to no net movement during the flicking behavior [23].
• Changes in shoal distribution: The percent of intruder events where a change in the horizontal or vertical distribution of the offspring shoal was observed compared to just before the intruder event. A change could occur in either direction, but most changes resulted in less cohesive shoals.

After the data collection period, the total lengths (TL) of parents and the average TL of fry within a shoal were visually estimated. To calibrate visual measurements, approximately 20% of parents and 30% of shoals (2–5 fry from each shoal) were captured and measured. On average, visual estimates of adults and offspring were 0.15 cm and 0.21 cm less than actual TL measurements, respectively. After adjusting all estimated measurement data based on these differences, male TL was 6.80 ± 0.17 cm, female TL was 4.81 ± 0.13 cm, and offspring TL was 0.95 ± 0.06 cm on average. Offspring shoals were estimated to contain 34 ± 2.2 fry on average. It was not possible to verify shoal size estimates.

2.4. Statistical Analysis

A one-way ANOVA was used to analyze each behavioral data set collected. Parent treatment served as the between-subject factor for all analyses, and had three levels (both parents, female only, male only). If the assumption of homogeneity of variance was violated, then Welch corrected F statistics were reported (i.e., pelvic fin-flicks, changes in shoal distribution). Tukey HSD post hoc testing was performed when the assumption of homogeneity of variance was upheld, and Bonferroni post hoc testing was used when homogeneity of variance was violated. Effect sizes are reported as partial eta² (η_p²). All data were analyzed using SPSS ver. 21.

3. Results

3.1. Single-Females Chase More Intruders than Single-Males or Both Parents Combined

A common parental response to encroaching intruders is to chase them away from the area. The number of intruders chased varied based on which parents remained with the offspring (Figure 1; F_2,43 = 5.04, p = 0.011, η_p² = 0.190, 90% CI: 0.03, 0.33), with single-females displaying the most intruder aggression. On average, the sum of intruders chased by both parents combined was nearly half that chased by single-females (Figure 1; p = 0.009). The same near-significant trend was observed between single-males and single-females, with females chasing nearly double the number of intruders (Figure 1; p = 0.059). The chases performed in concert by both parents did not differ from those of single-males (Figure 1; p = 0.956).

3.2. Offspring Shoals Were Unattended More Often in Single Versus Both-Parent Treatments

Since chasing an intruder away requires the parent(s) to momentarily leave the territory area, we asked how often this action would leave offspring unattended. The prediction was that the offspring of single-parent treatments would be left alone significantly more than those with both parents present. The percent of times that an offspring shoal was unattended by a parent during an intruder event depended on the parental treatment group (Figure 2; F_2,43 = 114.58, p < 0.001, η_p² = 0.842, 90% CI: 0.75, 0.88). As predicted, shoals with both parents were left alone significantly less (Figure 2). Offspring were unattended for only 9.1% of all intruder events in the both-parent treatment compared to 84.2% in single-female (p < 0.001) and 66.4% in single-male treatments (p < 0.001). However, single-parent treatments also differed from one another, with males remaining near offspring during a greater number of intruder events than females (Figure 2; p = 0.034). This fits with the observation that there was an increased number of intruder chases in single-female treatments (Figure 1), which would have kept them away from offspring more often.

3.3. Fin-Flicks Following Intruder Events

Pelvic fin-flicks are a communication tool parents use to alert and encourage fry to undertake antipredator behavior, including enhanced shoal cohesion and orientation towards the parent [19,20,21,22]. The number of pelvic fin-flicks observed during and immediately following an intruder event differed between parental treatment groups (Figure 3; F_2,12.41 = 4.55, p = 0.033, η_p² = 0.313, 90% CI: 0.02, 0.60). Our prediction that single-parents would perform more fin-flicks than both parents was only partially supported. Single-females performed more fin-flicks during intruder events than the sum of both parents combined (p < 0.001) and single-males (p = 0.031), which did not differ from each other (p = 0.703) (Figure 3).

3.4. Changes in Offspring Shoal Distribution

Shoal cohesion during intruder events was examined by observing if the horizontal or vertical distribution of the fry shoal changed noticeably in any direction following an intruder event. There was a main effect of the parental treatment group on the percent of intruder events that were followed by an observed change in the distribution of shoaling offspring (Figure 4; F_2,13.5 = 10.50, p = 0.002, η_p² = 0.386, 90% CI 0.22, 0.73). As predicted, shoals with both parents present were relatively stable with only 7.2% of intruder events followed by a change in distribution, which was significantly different in comparison to single-parents whose shoals were more in flux (Female only: p < 0.001; Male only: p = 0.023) (Figure 4). Shoals belonging to single-females did not differ from those belonging to single-males (p = 0.225).

4. Discussion

As expected from other field studies on the biparental convict cichlid [24,25,26,27] and other species of cichlids [28,29], we illustrated that with both parents present, one typically remained with the offspring while the other chased intruders. With one parent always being with the offspring, we predicted it would give the other parent more freedom to chase more intruders than a single-parent (e.g., [13]). However, this was not the case; paired parents attacked intruders less than single-females and about the same amount as single-male parents. Other studies have reported that single-females attack more intruders than single-males, although this could be dependent on parental experience, intruder type, and endocrine state [12,26,30,31,32]. Future studies are needed to dissect the environmental and physiological underpinnings of the behavioral differences reported in this study, including how parent-offspring dynamics may be affected by different intruder types.

We examined how offspring dispersion, and perhaps vulnerability to predation, may change depending on whether one or both parents were present. The prediction that increased chasing and/or spending more time away from the offspring would lead to increased dispersion behavior by the offspring shoal was not supported. Besides chasing more intruders, single-females also seemed to chase them for greater distances (Al-Shaer pers obs.) than single-males, which may also account for their fry being left alone for longer compared to males and could explain why single-females performed significantly more pelvic fin-flicks. One hypothesis is that single-females employed increased pelvic fin-flick behavior to mitigate shoal dispersion while offspring were left unattended during increased intruder chases. Further, additional studies are needed to ascertain if the heightened intruder chases by single-females translate to greater offspring survival. Despite differences in intruder chases, offspring dispersion was not significantly different between the two single-parent groups. Also, dispersion was significantly higher with single-males compared to paired parents, despite both groups exhibiting similar levels of chasing. When viewed across all three treatment groups, the amount of offspring dispersion seems related to the number of parents with the offspring and not the number of intruders chased, or the time spent away from the offspring. However, future studies should also consider how shoal size may also affect dispersion dynamics.

Given the apparent dangers of fry dispersing away from the protection of the parents, we presumed that parents subjected to increased offspring dispersion would show higher levels of pelvic fin-flicking. This was not supported, as both single-male and single-female parents had offspring with similar degrees of dispersion, but males fin-flicked at a much lower level and were similar to the level exhibited by paired parents. However, the high rate of pelvic fin-flicking by single-females may have reduced their fry’s dispersion so that it resembled that of offspring with the more sedentary single-males. Paired parents rarely called the fry using pelvic fin-flicks, but it is possible that the fry may have been attracted to the parents by some other means, perhaps using odors produced by the parents [33]. Further work regarding the attraction of fry to their parents is warranted. Alternatively, the lack of calling by the parents may have been the result of reduced predatory threats because both parents were present. Similarly, the larger size of the male parent may have repelled intruders even when serving as a single-parent by creating a ‘zone of inhibition’ [7], which could explain why the smaller single-females chased more intruders.

To summarize, previous studies on biparental care in cichlid fish focused on the need for both parents to protect their offspring by chasing away the myriad predators [7,26]. Here, we illustrate that the removal of one parent of either sex results in the dispersion of the fry shoal, which may make them more vulnerable to predators. We found that the tactics employed by the single-female (i.e., attacking more intruders while increasing pelvic fin-flicks) or the single-male (remaining largely sedentary near the fry with low levels of pelvic fin-flicks) did not reduce the fry dispersion compared to when both parents were present. We suggest that the behavior of the fry in relation to biparental care could be important in understanding the evolution of biparental care in cichlid fishes.

To the best of our knowledge, this is the first study to test the effect of intruders from both the perspectives of offspring and parents while also considering how single and paired parents impact behavior. By demonstrating that parent and offspring antipredator behavior shifted when a parent was lost, we provide further insight into why biparental care evolved in this species. Our study highlights the need to consider not only parents but also offspring as key evolutionary drivers of biparental care in this, and perhaps other species.