| Summary: | <p>In this study of the functional aspects of a complex behavioural system I have looked at three different aspects of the feeding behaviour of the Oystercatcher. I started by analysing the skilled feeding techniques of the adults, but I soon became interested in the development of these techniques in the young, and in the parental feeding system which is surprisingly different from that of other waders. It forms the 'buffer' between the adults and the young and allows the young to acquire the specialised feeding techniques and skills of their parents.</p> <p><strong>Part I</strong></p> <p>I discuss the parental feeding system first because the analysis has focused on a slightly different aspect of the feeding behaviour, namely the way in which the amount of food brought to the young by the parents is precisely controlled by the behaviour of the young. The parental feeding system has two basic characteristics: (a) the parents bring food items to the young and present them, and the young take the food from the parent's bill; and (b) the young accompany their parents out onto the feeding grounds and are fed by them in long 'feeding trains'.</p> <p>The seasonal development of parental feeding starts with the auditory and tactile interactions between the incubating parent and the hatching eggs. Both parents and young-in-egg become more responsive to each other as hatching proceeds, and both show a marked selective responsiveness. These pre-hatch interactions lead to the transition from incubation to brooding behaviour in the parent, and finally result in the onset of parental feeding. There is the additional possibility that the young learn to respond to the call of their parents while still in the egg.</p> <p>I then discuss the factors which control the amount of food brought by the parents to the young. In the active phases of the daily feeding cycle, each parent brings food items to the young at random moments in time. Once the food has been presented, the subsequent behaviour of the parent depends upon the speed with which the young seizes and devours the food. If the young takes the food within a certain time (the Waiting-Time-Threshold) the parent will immediately search for and present more food, and it will continue to do this for as long as the young takes the food within the threshold time. In this way the long 'feeding trains' are initiated and maintained. However, the young can also initiate a feeding train by approaching a parent who is feeding. The feeding train is broken off by the parent as soon as the young fails to react within the threshold time. This self-regulating social system is based on complex and highly specialised interactions between the parents and the young.</p> <p>As the young grow older the organisation of parental feeding changes drastically. The parents initiate fewer feeding trains, they carry less food to the young and, instead of removing the flesh from prey and then presenting it to the young, they begin to leave the prey for the young to open up and clean out. The young however play a more and more important role in the initiation and maintenance of the feeding trains. This switch of control to the young eventually leads to the breakdown of parental feeding, for, as the young learn to feed themselves, they stop eliciting parental feeding. The consequence of this system is that in areas where the young merely have to acquire the simple skills of probing for worms, parental feeding stops at about 6–7 weeks. However, in areas where the young are learning the more difficult skills for dealing with molluscs and crustacea, parental feeding continues for 12–20 weeks, and in extreme cases for as long as 43 weeks.</p> <p><strong>Part II and III</strong></p> <p>I here describe the techniques employed by adult Oystercatchers in dealing with <em>Mytilus</em>, <em>Cardium</em>, <em>Scrobicularia</em>, <em>Tapes</em>, <em>Littorina</em>, <em>Patella</em> and <em>Carcinus</em>. Each of these techniques consists of a sequence of discrete motor patterns of which each has a different form and a different function. To take the feeding techniques on <em>Mytilus</em> as an example: either the <em>Stab</em> (a quick downwards lunge of the bill through the gaping valves) or <em>Hammering</em> (a series of hard blows aimed at the shell) serve to gain initial energy into the valve; <em>Levering</em> (side-to-side movements of the bill) and <em>Prising</em> (forcibly opening the mandibles) serve to force the valves wide apart; and <em>Chiseling</em> (rapid to-and-fro movements of the bill) serves to cut the flesh loose from the valves. Each motor pattern is precisely oriented for maximum effect: the Stab is aimed between the posterior valve margins so that it severs the strong posterior adductor muscle when the bill enters (this disabling the mussel); the Hammering attack (for which the mussel is first oriented so that the flat ventral surface is uppermost) is aimed at this area which I have shown to be by far the weakest part of the shell; and the Chiseling movements are aimed at the largest attachments between the flesh and the shell, the smaller ones being broken when the main mass of the flesh is shaken free.</p> <p>I found that these motor patterns are applied to all the prey species mentioned, and that they have always the same general function. However, in each case the orientation of the motor patterns are specifically adjusted to the anatomy of the particular prey. For example, the Hammering attack is direct at the ventral valve margins of <em>Mytilus</em>; at the posterior, ventral or anterior valve margins of <em>Cardium</em> (all these valve margins are equally friable); at the lateral walls of <em>Patella</em> (which are most easily dislodged by a blow from this direction); and at the mouth of <em>Carcinus</em> (which kills the crab as well as making a hole in it.)</p> <p>I then analyse the organisation and control of the sequence and duration of the motor pattern. Firstly, each motor pattern is elicited by a different external stimulus. Second, the duration of the motor patterns is likewise under continuous external control; for example, the duration of Levering in <em>Cardium</em> depends solely upon the size of the prey, i.e. in functional terms, to the amount of Levering that is needed to force the valves apart. Thirdly, the sequence in which the motor patterns occur are seen to be simple chain responses in which the performance of one motor pattern brings about the external stimuli necessary to elicit the next. It is this <em>total</em> dependence on external control which imparts a remarkable flexibility, and thus efficiency and economy of effort to the techniques, for each motor pattern is performed only if and when necessary, and only for as long as is necessary.</p> <p>The skill, efficiency and versatility of the adult techniques are due to two basic factors: firstly, a basic repertoire of motor patterns each adapted to a very generalised function; and secondly, their dependence on external stimuli, which allows (a) the precise adjustment of the orientation of each motor pattern to the anatomy of any one prey, and (b) flexibility in the initiation, duration and sequential organisation of the motor patterns.</p> <p>Very surprisingly, there is good evidence that individual Oystercatchers specialise not only on one particular prey but even on one particular technique of catching it. For example, the breeding population of Ravenglass (in Cumberland) is made up of individuals who eat crabs. Furthermore, in each breeding pair both partners have the same specialisation. There are a number of both casual and functional reasons for this specialisation, most of which are bound up with the development of the techniques in the young Oystercatchers.</p> <p><strong>Part IV</strong></p> <p>The starting point for my study of the development of the feeding sills were the observations that the young of mussel-Stabbing parents, mussel-Hammering parents and crab-eating parents all develop the same feeding skills and techniques as their parents. I therefore cross-fostered the eggs of individually known pairs with different specialisations. In every case the young developed the specialisations of their <em>foster parents</em>. This meant that the young had the <em>potential</em> to develop a number of different techniques but that some environmental factor was channelling the developmental system, so that only one technique developed. This led to a study of (a) the basis of the flexibility in the developmental system, and (b) the ways in which the channelling was brought about so that only one technique developed, and I have tried to analyse these from both a functional and a causal viewpoint.</p> <p>I first discuss the Independent Feeding Activities (IFA) of young Oystercatchers. These are shown from a few hours after hatching, and they are easily recognised as incipient forms of the adult feeding behaviour. A full repertoire of discrete motor patterns, all similar to the motor patterns of the adults, is present in all Oystercatcher young, irrespective of the specialisations of their parents and irrespective (within the range observed) of the environment in which they are growing up. The IFA are directed initially at a wide range of objects (both edible and inedible alike), and the stimuli that elicit them are very similar to those that elicit the same motor patterns in the adults.</p> <p>During IFA the young gain experience with the prey species (in this study with mussels or crabs – depending on the specialisations of the parents). The opportunities to experience are the following: (a) the young remove scraps of flesh from the shells discarded by the parents, (b) the young open up and empty out partially prepared prey which the parents leave for them, (c) when the young accompany their parents onto the feeding grounds they constantly see prey being caught and opened in their immediate presence, and (d) there are numerous undisturbed prey on the feeding grounds which the young attempt to catch and open up for themselves.</p> <p>The actual learning processes that take place are shown to be of at least three different types:</p> <ol> <li>the first interactions with the prey are seen when the small young (2–3 weeks of age) Chisel scraps of flesh from the discarded prey remains; subsequently, the young spend more and more time in investigating the prey remains. Experiments with hand-raised young (which were given experience of mussel shells containing flesh) showed that neither finding food in mussels by itself, nor Chiseling in mussels without a food reward, nor even performing any other motor pattern than Chiseling – with or without a food reward – leads to conditioning to the mussel. The conditioning <em>only</em> occurs when the young have to Chisel <em>and</em> receive food as a <em>result</em> of Chiseling. This pre-requisite for conditioning has two important functional aspects: firstly, it ensures that the young will become conditioned only to potential prey rather than to any other object lying about, and secondly, the young will become conditioned to whatever prey the parent is feeding to it. We see here the first example of both the flexibility and the channelling of the developmental system.</li> <li>This experience with the discarded shells also leads to the initial development of Chiseling, and <em>practice</em> with the partly opened prey left by the parents leads to the further development of efficient Chiseling, as well as to the functional organisation of Levering, Prising and Chiseling, i.e. to the correct orientation and sequence of these motor patterns. Much of this change in behavioural organisation can be accounted for by trial-and-error learning, but I also suggest that other factors may lend directionality to such learning processes by biasing what is learned.</li> <li>The development of the hunting techniques (i.e., locating and Stabbing or Hammering mussels, and locating and opening crabs) seem to develop as a direct result of <em>parental example</em>. At 4–5 weeks of age (before the young have started to successfully catch and open prey for themselves) there are already marked differences in the hunting behaviour of the young of mussel-Stabbing parents, mussel-Hammering parents and crab-eating parents. Both the 'Stabbing' and 'Hammering' young show elements of both techniques, but the most frequent behaviours shown in each case are those of their parents' techniques. The 'crab' young at this age completely ignore the mussels and only hunt for and attempt to open crabs. Once the young start to successfully catch and open prey (at 6–7 weeks of age) they employ only the relevant hunting behaviour.</li> </ol> <p>I then discuss the developmental system from the viewpoint of a learning machine with pre-set programmed instructions. The Oystercatcher's 'programming' shows specific adaptations to (a) develop a skilled feeding technique yet (b) retain flexibility so that a number of different feeding skills may potentially be developed. The instructions are therefore of a very general nature when compared with the instructions of developmental systems where the environment is more 'predictable'.</p> <p>I also discuss the coadaptation between the parental feeding system and the requirements of the developmental system. Firstly, the parent is instrumental in channelling the developmental system of the young; secondly, there are changes in the organisation of parental feeding as the young grow older which result in the parent leaving the partially opened prey which the young require for practice; and thirdly, I discuss the evolution of parental feeding as a necessary coadaptation to the evolution of a highly specialised feeding behaviour which the young have to acquire at an early age.</p> <p>Finally, I consider some broader ecological aspects of the development of the feeding skills. The first important point is that the young take at least three years to become fully efficient at eating mussels, and I argue that other feeding techniques will probably take as long to become fully efficient. There are a number of ecological consequences to this protracted development. Firstly, Oystercatchers do not breed until they are three, and usually four years of age, and I suggest that this is an evolutionary adaptation to the prolonged development. Secondly, during the prolonged development it would be disadvantageous to attempt to acquire a new technique, for the feeding behaviour would become even more inefficient. By the time that the one technique has become fully efficient, the bird may have to be so fixated on one particular prey and one particular technique that it will not respond to the relevant stimuli from other prey. This may be the basis of the persistent specialisations shown by adult Oystercatchers. However, some individuals (about 20%) do possess more than one skilled technique, and perhaps they can switch through essentially the same processes that young Oystercatchers employ when learning from their parents. It may, therefore, be of advantage even for adults to retain a certain degree of flexibility.</p>
|
|---|