The effects of mixed-species assemblage on the behaviour and welfare of fish held in home aquaria

Abstract

While there is growing scientific research into the welfare of fish held under aquaculture conditions, there has been little research into the welfare of ornamental fish kept in home aquaria. With the ever increasing popularity of home aquaria, there is now an urgent need to address our current lack of knowledge regarding welfare of mixed-species assemblages. Here we investigated the welfare of four commonly kept species of ornamental fish held in different species combinations and group sizes. Specifically we asked two questions (1) whether the presence of angelfish (Pterophylum scalare) altered the behaviour of a mixed group of white cloud mountain minnows (Tanichthys albonubes), neon tetras (Paracheirodon innesi) and tiger barbs (Barbus tetrazona) and (2) whether group size affected behaviour and welfare in a mixed group assemblage of angelfish, white cloud mountain minnows and neon tetras. Behaviours including intra and interspecific interactions (darting and aggressive attacks), shoaling, and use of environmental enrichment were measured. Angelfish appeared to have a beneficial effect on the welfare of small shoaling species by reducing aggression but had little effect on other behaviours. For white cloud mountain minnows and neon tetras, larger group sizes resulted in increased natural behaviours (i.e. a tendency to shoal). Interestingly, the present study highlighted the problems in predicting the effects of environmental enrichment with both species and group size altering the way enrichment was used (e.g. as shelter from competitors or as a resource worth aggressively defending). Setting scientifically determined guidelines on appropriate species assemblages or stocking densities for ornamental fish is challenged by the plethora of group sizes and combinations found within home aquaria; research as in the present study serves to provide insights into welfare issues which may not become apparent within unrepresentative, single-species studies.

Introduction

The concept of animal welfare has, until recently, predominantly focused on terrestrial sentient animals, with little research exploring the welfare requirements of teleost fish (Braithwaite and Boulcott, 2008). The “five freedoms” approach to welfare identified by the Farms Animal Welfare Council (FAWC, 2009) states that animals should be free from; (a) thirst, hunger, or malnutrition; (b) discomfort (provided appropriate housing); (c) pain, injury and disease; (d) restriction of normal behaviour; and (e) fear and distress (FAWC, 2009). These guidelines define rules concerning appropriate biological functioning (with respect to thirst, hunger, shelter, injury and disease) and establish the foundational criteria for basic captive animal welfare (Volpato, 2009). Consideration of fish welfare has involved reworking of guidelines for farm animals, into a form more appropriate for fish (FSBI, 2002, Williams et al., 2009). However applying the welfare principles described for terrestrial animals to fish is a complex and challenging task as they possess several factors unique to aquatic organisms (Williams et al., 2009). Studies of fish welfare usually follow one of three approaches: “feelings based”, “functioning based” and “natural behaviour” (Duncan and Fraser, 1997). Most of the literature dedicated to the welfare of fish focuses on the “functioning based” approach where physical health is the most frequently used indicator (Volpato, 2009). However, behavioural studies have an important role in welfare research (Mench and Mason, 1997), partly in the context of understanding behavioural needs and partly because altered behaviour is an early and easily observed response to adverse conditions (Huntingford and Kadri, 2008). Although no single set of behavioural responses can be used to indicate impaired welfare, observation of behavioural changes or comparisons between different conditions can provide welfare information (Mench and Mason, 1997, Huntingford and Kadri, 2008, Saxby et al., 2010).

To date, fish welfare research has focused on species of fishes used in aquaculture (Ellis et al., 2002, Ashley, 2007) where improved fish welfare equates to both improved product quality and reflects increasing customer demand and awareness of animal welfare. More recently, there has been an increased interest in the welfare of fish species kept as companion animals. Over 350 million fishes are traded annually within the aquarium industry which is estimated to be worth 800 million to 30 billion US dollars (Bartley, 2000, Helfman et al., 2009). While ornamental fish are protected under the UK Animal Welfare Bill (2006) and retailers are required to provide adequate information to buyers to promote the welfare of the fish they sell (OATA, 2008, Walster, 2008), currently there are no scientifically determined guidelines on appropriate species assemblages or stocking densities for ornamental fish. Fish in home aquaria can, therefore, potentially be held in mixed species assemblages and densities detrimental to their welfare. Inappropriate mixed-species assemblages translate into provision of an inappropriate environment resulting in disruption of normal behaviours.

Mixed-assemblage home aquaria vary dramatically in both numbers of species and numbers of individuals making it impossible to consider all combinations when investigating welfare of home aquarium fish. However, it is possible to investigate how the welfare dynamic varies as set combinations of species are altered and behavioural changes observed. Previously, Saxby et al. (2010) detailed the welfare effects of group size in four species (neon tetras, white cloud mountain minnows, angelfish and tiger barbs) when held in single species groups. Here, we build on the work of Saxby et al. (2010) by investigating whether the behaviour and welfare of these species was altered when they were combined in four different mixed-species assemblages. Specifically, two questions were addressed: (1) whether the presence of angelfish altered the behaviour of a mixed group of white cloud mountain minnows, neon tetras and tiger barbs and (2) whether group size affected behaviour and welfare in mixed species assemblages. To address the first question, two combinations of fish were used. Combination one contained eight tiger barbs, five white cloud mountain minnows and five neon tetras. Tiger barbs are known to form shoals and can be particularly aggressive towards individuals of both the same and different species within home aquaria (www.aqua-fish.net). Neon tetras and white cloud mountain minnows are both small shoaling species of fish popular in home aquaria (Magurran and Pitcher, 1983, Chapman et al., 1998, Webster et al., 2008). Combination two contained the same numbers and species of fish with the addition of two angelfish. Angelfish are anecdotally considered to be peaceful ‘dither’ fish (Barlow, 1968, Loiselle, 1979; www.aqua-fish.net) and it was hypothesised that welfare would be improved by the presence of these fish. To address the second question, the behaviour of five white cloud mountain minnows, five neon tetras and one angelfish (combination three) was compared with 10 white cloud mountain minnows, 10 neon tetras and two angelfish (combination four). Saxby et al. (2010) found that single species groups of 10 neon tetras and white cloud mountain minnows showed improved welfare compared to smaller group sizes whereas no evidence for an effect of group size on the welfare of angelfish was found. It was unknown how interspecific interactions between these three species would impact on their behaviour but it was hypothesised that welfare would be improved for all species by increasing group size.