1()

ETHOLOGY OF THE ASIAN TELEOST BADIS. I. LOCOMOTION, MAINTENANCE, AGGREGATION AND FRIGHT GEORGE W. BARLOW1 Max-Planck-Institut fur Verhaltensphysiologie, Seewiesen, Obb., Germany For such a popular aquarium spe- Wolfgang Wickler. Thanks are due cies, the biology of to Herman Kacher and Alice Boat- (Hamilton-Buchanan) is remarkably right for making the illustrations. poorly known. What little informa- I would also like to express my tion is available, mostly in aquarium gratitude to Professor Lorenz for journals, is to a large measure er- providing the facilities that made roneous. Even systematically the the research possible. appears to have been mis- I acknowledge with pleasure that placed. I hope to show in a subse- this investigation was supported by quent paper, primarily on behavior- a post-doctoral fellowship (MF al evidence, that Badis badis has 8244) from the National Institute been classified in the wrong family of Mental Health, United States and suborder. Public Health Service. The main objective of this paper is to introduce certain aspects of TERMINOLOGY the ethology of Badis badis. The Badis badis is topics considered could stand alone, The behavior of but of more importance, they con- first broken down into units and these are given names. No claim is stitute the foundation for a series of articles to follow. The treatment made that these units correspond to of locomotion, for instance, centers Fixed Action Patterns, "acts," or "components," (see Hinde, 1959). on descriptions of movements essen- Convenience has been the rule in tial to an understanding of social displays and parental care, and their fractionating the behavior. Most of derivation, rather than emphasizing the units, however, approximate hydrodynamical considerations. It Fixed Action Patterns, or subdivi- was found practical, however, to dis- sions of them. Where pertinent, the orientation components (Lorenz and cuss some of the topics as they were completed, rather than to defer this Tinbergen, 1938) are described. The prevailing terminology is usu- to later publications. ally employed when possible, but no attempt is made to follow priority ACKNOWLEDGMENTS of usage, nor in most cases are the For critically reading the manu- sources acknowledged. In this arti- script I am indebted to William cle terms both new and old referring Childers, Konrad Z. Lorenz, and specifically to Badis badis are capi- talized. Forselius (1957) has coined 1 Present address: Department of Zoology, University of Illinois, Urbana many terms to describe behavior pat- Illinois. ters observed in anabantid fishes. In [ 175 ] 176 Transactions Illinois Academy of Science spite of the similarity of the be- seems vibratory. The phasing of the havior of the anabantids and of waves between right and left fins Badis badis, and of their presumed appears to be alternate. close relationship, I prefer using my Swimming fast is accomplished in own terminology in the treatment the normal teleost manner through of the behavior of Badis badis. lateral undulations of the body. The head appears to initiate the waves MATERIALS AND METHODS by turning almost imperceptibly to the right and left. As the waves pass Badis badis is a small (maximum posteriorly through the body they standard length about 75 mm) increase progressively in amplitude. teleost fish occurring naturally in The propulsive thrust of this type tropical Asia. The stock was ob- of movement is derived from the tained from an aquarium shop in forward component of the force pro- Munich, and had been raised in that duced by the passage of the leading city. Most of the reported edges of the waves obliquely through on were bred in the laboratory. The the water (Gray, 1933a). Normally fish were kept in aquaria having ca- the fish employs both methods for pacities of 30, 60, and 120 liters ; swimming ahead, sculling being im- they were thickly planted with a plemented by occasional undulations. wide variety of aquatic plants, and While swimming rapidly forward, the bottoms were covered with fine the spinous portions of the dorsal gravel. Water with a hardness of and anal fins, as well as the pelvic about DH 4-6, a pH of around 6-7, fins, are usually folded against the and a temperature of approximately body. The lobes of the dorsal and 26° C to 28° C circulated through anal fins are held loosely open and a large filter reservoir and the move in unison with the caudal fin ; aquaria. Illumination was provided together they increase the surface by 25-watt fluorescent tubes. area of the posterior end of the fish. Most of the observations were This arrangement of the fins refers made directly. Strobe-light photog- to swimming movements uncompli- raphy was employed to supplement cated by interactions with other observations. fishes, predators, or prey. LOCOMOTION Backward. The pectoral fins push water forward by means of alternate Forward. Badis badis swims paddle-like beats. On the recovery slowly in comparison to many kinds stroke, the fin is feathered, the su- of fishes. At the slowest speeds, for- perior margin leading. In a well ward thrust is provided by move- developed backing, all the other fins ments of the pectoral fins which un- are folded except the lobe of the dulate, passing vertical waves down . The caudal fin frequently and to the rear along the fins ; this is compressed and held to one side type of movement will be referred partially cupped ; it often beats to to as sculling. In comparison to that side, probably having a greater the slower tempo of many other steering than propulsive effect. Also, teleosts, the sculling by Badis badis the fish may feel its way in the bur- Ethology of Badis badis 177 row with the caudal fin. The loosely slowest tempos, the fish barely main- expanded lobe of the dorsal fin is tained themselves in midwater ; at brought into use as a rudder. Rarely the fastest tempos, the fish often the entire dorsal fin, as well as the contacted some firm object, as though caudal fin, may remain loosely using it as an anchor. spread, particularly when backing The tempo of the pectoral fin up is incompletely expressed. movements while Hovering seems to Hovering. The fish stands in the indicate the state of excitation of water nearly motionless. The pec- the fish. Ripe 9 9 were observed toral fins, and often the lobe of the as they approached and withdrew dorsal fin, perform sculling move- from the 8 in his burrow. In each ments. The pectoral fins tend to be instance the slowest speeds were cupped with the concave surface di- recorded when the 9 was furthest rected anteriolaterally. Presumably from the S, and had been so re- the pectorals exert a backing thrust, moved for some time. Approaching resisting the forward thrust of the produced an acceleration of the tem- opercular jets (see Breder, 1926). po. The maximum invariably was The medium and pelvic fins may be achieved when the 9 actually con- either spread or depressed. Badis fronted the f. Fleeing was followed badis spends much of its time Hov- by a gradual deceleration. Among ering. sub-adults the fastest tempos were In subsequent articles, the tempo always seen during encounters with of Hovering will be compared to other sub-adults, the slowest when those of Fanning, Digging, and unmolested and full of food. Hence Shuddering, as the origins of these the faster the tempo, apparently the behavior patterns are considered. greater the excitation. Table I lists the extreme range, and Starting. In slow starting, the median, of the tempo of Hovering fish appears to change the pitch of seen in various Badis badis. At the the sculling pectoral fins slightly,

TABLE 1.—TEMPO OF PECTORAL FIN BEATS WHILE HOVERING.'

TEMPERATURE RANGE MEDIAN ESTIMATED STANDARD LENGTH (MM.) (°C) (BEATS/SEC.) (BEATS/SEC.)

25 26.3 4.6-6.1 5.4 27.5 26.3 4.5-5.2 4.9 30 26.3 4.5-5.4 5.0

20 26.5 3.9-4.5 4.2 22.5 26.5 4.14.6 4.4 25 26.5 4.0-4.3 4.2

GRAND RANGE GRAND MEDIAN 3.9-6.1 5.0

THE UPPER THREE SETS OF DATA WERE TAKEN FROM THREE RIPE 99 IN AQUARIA WITH MATURE Cre, THE LOWER THREE FROM SUB-ADULTS IN AN AQUARIUM WITH OTHER SUB-ADULTS. FOR EACH FISH 10 SAMPLES OF 50 BEATS EACH WERE TALLIED. 178 Transactions Illinois Academy of Science directing the water current to the er control of the forward movement. rear. If a faster start is required But it is placed in a less favorable the head initiates lateral trunk un- position for the initiation of another dulations in the manner typical for beat, and also sacrifices some force. most fishes (Gray, 1933b), and the The Sickle-shape is usually used only spread pectoral fins are clapped when the fish intends to dart forward against the sides in unison. In either a short distance (at least less than case the fins take on the arrange- its own body length), but when con- ment described for Forward Swim- siderable accuracy in the course is ming. necessary. In certain situations the fish starts Sickle-shape and S-shape are two with an abrupt burst of speed. The extremes and all degrees of inter- most typical preparatory posture mediacy occur. Some of the other for a fast start in Badis badis, and postures derived from locomotory a great number of other kinds of movements are shown in Figures 1C fishes, is the sigmoid, or S-shape and 1D; these occur regularly in (Fig. 1A), an arrested swimming certain displays and will be dis- undulation. From this position the cussed in another article. fish can deliver the maximum instan- Turning. The mechanics of turn- taneous forward thrust. The head ing are difficult to observe without is in a position enabling it to move resorting to high speed cinematogra- through the greatest arc possible. phy. Turning seems to consist of Consequently the largest lateral am- merely bending the head in the di- plitude is produced in the tail-beat. rection of the turn while swimming. The posterior part of the fish has Among the fins, the pectorals appear at the outset the proper angle for to exert the most control in turning utilizing the greatest possible for- both in the horizontal and in the ward component of the beat. vertical. Unfortunately the pectoral fins are too transparent to be ob- served well. In some fishes, one ex- treme being the pomocentrids of the Dascyllus, the pelvic fin, on the side toward which the fish is turning, is lowered and serves as a pivot, or brake. In Badis badis the pelvic fins are abducted now and then, though usually together. Evi- dently this is done mostly as a par- tial stopping and/or stabilizing Fig. 1.—Characteristic body postures movement. Although Badis badis derived from locomotory movements, seen from above. A, S-shape; B, Sickle- can raise and lower the right and shape; C, L-shape; D, C-shape. left pelvic fins independently, it does not do so to any great extent in turn- The Sickle-shape (Fig. 1B) is a ing. The median fins of Badis badis specialized expression of S-shape. By take no special part in turning other keeping the anterior part of the than assisting in propulsion and body straight the fish achieves great- stabilization. Ethology of Badis badis 179

Opening and closing the fins. A the dorsal, then the caudal, and then definite way of folding or raising the anal fin, extinguishing in the the fins is observable when the move- reverse order. ments are done slowly. In a fish whose fins are tautly spread the an- RESPIRATION IN OXYGEN terior-most elements of the dorsal DEFICIENT WATER fin even incline slightly forward. In folding the dorsal fin against the The response to oxygen-deficient is similar to body the rays first incline somewhat water by Badis badis to the rear. This imparts a relaxed that of many teleost fishes lacking the ability to utilize air directly. appearance to the fin, termed loosely spread. In order to observe this response, The dorsal fin collapses from front twelve Badis badis, juveniles to to rear, and is erected in reverse young adults, were placed in a small order, from posterior to anterior. plastic dish containing water from Under certain circumstance, how- their aquarium. As the oxygen con- ever, the fish is able to erect the tent of the water presumably de- dorsal fin from front to rear. The creased, the fish moved upward and folding of the anal fin is similar to stood directly under the surface of that of the dorsal. Because of the the water. Their bodies were tilted short length of its base, the anal fin upward anteriorly about 45°, just folds more as a unit. When folded, enough to bring their mouths into the rays of the caudal fin may press contact with the surface. The oxy- together until the fin forms a point. gen-rich surface film was then in- The pelvic fins are simply brought haled arid passed over the gills. The in against the abdomen. opercular movements appeared to be The sequence of folding the fins the same as those seen in normal slowly can be seen best in a fish pre- aquatic respiration. At no time did paring to back up. The pelvic fins a Badis badis entrap air bubbles in, are adducted first followed by the or pass them through, the gill anal, caudal, and dorsal fins in that chamber. order. The sequence is usually the COMFORT MOVEMENTS same for swimming forward ; of course, the caudal fin is not folded. Included here are those movements The fins are abducted in reverse that are used in removing irritants, order, i.e., the pelvics are the last or seem to be stretching movements to be spread. Most of the time the (Baerends and Baerends-van Roon, folding movements overlap broadly. 1950, for a discussion of Comfort As an example, when the pelvics Movements). fold, the dorsal fin rays begin to Fin flickering. The pelvic fins incline more to the rear. rarely are opened and closed to- In opening and closing the fins gether rapidly once or twice. Some- the excitation runs along the perime- times one pelvic fin alone is moved ter of the body, so to speak. The through the full arc, and the other same holds for the onset of fanning opens only partially. The elements movements where the beat starts in of the dorsal and anal fin can be 180 Transactions Illinois Academy of Science raised and lowered rapidly in a currently with (3) the fins relax. flicker-like movement that may be In a less complete performance the repeated in bursts. The pectoral fins sequence remains the same, but the can be rubbed against the body in gaping is less, the floor of the mouth various ways ; these movements are is not so conspicuously depressed, exceedingly difficult to observe. and the fins are only partially, or Chafing. The fish slowly ap- barely, spread. Intention yawning, proaches an object such as a plant, consisting of slight gaping and par- the bottom, or an aquarium wall. tial erection of the fins, may precede With a quick movement one side of complete yawning. the fish apparently near the head, is placed against the object, and it FEEDING BEHAVIOR darts forward two to three body Badis badis is strictly carnivorous. lengths ; the rebound may be arcing, The post-larvae forage at first on or straight ahead ; the fins are folded. the microfauna in the aquarium, In cichlid fishes the median and pel- probably rotifers and large proto- vic fins, in contrast, are normally zoans. Within a few days after spread during the rebound (pers. swimming they capture and eat small observ. on Etro plus maculatus and cyclops and Daphnia; these are E. suratensis). caught in the open water. The post- Flexing. An exaggerated lateral larvae also forage along the bottom flexure is made in the body, often and over the surfaces of plant leaves as the fish reverses directions. It where doubtless they find small ben- may immediately straighten out thonic animals. As soon as the fish again, or the flexure may pass down are large enough they feed on small the body as a wave. worm-like animals. They seem to Coughing. The gill covers are prefer these "worms," such as Tubi- clamped down in a quick movement fcx, Enchytraeus, and many aquatic and the mouth is thrown open. Fur- insect larvae, to crustacea. If suffi- ther details are not available. Ap- cient "worms" are available to the parently a counter-current of water juveniles and adults, crustacea are is flushed through the buccal cavity. taken only sporadically. Adults also Yawning. In the most complete have been seen to seize and attempt expression of this movement all the to eat large leeches, but the leeches fins are fully spread, the mouth is escaped. Small leeches are probably opened maximally, and the opercular eaten. chamber is expanded (See Fig. 8 Feeding appears to be visually of Eupomotus gibbosus in Breder, mediated since only moving objects 1936). Then (1) the mouth is are taken. Even post-larvae, which snapped shut as (2) the floor of the have been swimming well for only mouth is depressed and quickly re- a few days, have been observed to covered, and immediately (3) the turn the head to one side and fix on operculae are adducted. This se- tiny moving particles. quence is rapid and passes a large The movements employed in feed- volume of water quickly from the ing are similar at all stages in the mouth out through the gills. Con- life history of Badis badis. Younger Ethology of Badis badis 181 animals feed on a different type of is commonly observed after the fish prey (nektonic) than do the larger has devoured several crustacea in fish (benthonic), so their way of quick succession. The fish simply foraging at first seems different. swims directly to the prey and takes A young, not recently fed, Badis it into its mouth. All intermediates badis catches a Daphnia in the fol- are seen in the degree of complete- lowing way. It approaches the prey ness of prey-capturing behavior. slowly by sculling forward. When The descriptions of activities con- the prey lies directly ahead about cerned with seizing worms apply to one head length away the fish stops juvenile fish capturing worms that and commences Hovering while con- are already imbedded in the bottom. tinually adjusting to maintain its Long before the fish reach maturity position relative to the prey. The they learn to catch the worms as already raised dorsal fin now be- they fall from the surface of the comes taut. Slowly the trunk forms water, or to pick up worms on a Sickle-shape (Fig. 1B). During the bottom before they dig in. The this brief period while the fish fixes adults seem to be "spoiled'' by the on the prey, its movements are slow easy life in the aquarium and com- and scarcely noticeable. Suddenly plete feeding behavior is seldom it darts forward with folded fins in manifested even when the worms a single snap of the trunk, and en- are well dug in. In the adults, social gulfs the Daphnia. The prey is interactions also frequently modify chewed a few times, apparently by the behavior. For instance, a fish the pharyngeal teeth, before swal- may fix on a worm from outside of lowing. another 's territory, then suddenly A young Badis badis that has al- dart from its hiding place and snatch ready eaten some Daphnia shows the worm, all in one circling dash. less complete feeding behavior. It Normally two slightly different approaches and fixes on the prey in techniques for grasping onto worms nearly the same manner. The body are observed ; in both cases the fish of the fish, however, remains straight. first swims to the protruding tip Furthermore, the fish may move for- of the worm. It pauses there when ward and to the side until the prey its mouth is roughly one head length is lateral to its snout at a distance from the worm. In the more com- about one-half of a head's length. plex method of grasping a worm the Then the fish snaps its head laterally fish rolls over onto one side, simul- toward the prey. The fish seems to taneously forming an S-shape ; the "inhale" the prey by suddenly en- head points down at the worm, and larging the mouth cavity, but this the dorsal fin is usually spread as cannot be reported with certainty. much as the body curvature allows It is also not certain whether this (Fig. 2, Top). After briefly fixing sideways snapping of the head ap- on the prey the Badis badis lunges pears as early in ontogeny as the and seizes the worm. Most of the forward dart. time this method is seen only when An even less complete perform- the fish has not recently caught a ance of the prey-capturing behavior worm. 182 Transactions Illinois Academy of Science

lying on one side the fish is able to use its body as a lever (Fig. 2, Bot- tom). The head is raised, pulling the worm free, while a counteract- ing force is exerted downward by swimming movements of the tail. Swimming forward or backward with the worm usually follows fixing in an upright position. Side lever- age normally follows fixing from the on-the-side S-shape. All combina- tions, however, are seen. Fig. 2. Top—fixing on a worm in the The completeness of the observed bottom. Bottom—pulling on the worm. behavior seems to be influenced by exogenous, as well as endogenous, In the less elaborate technique of factors. This can be documented by worm catching the body remains in observations on the activities in- the normal upright position, though volved in pulling a worm free of its long axis becomes inclined down- the substrate after grasping onto ward anteriorly at an angle of about it. It often happens that the fish 450; the already raised fins become backs up with the worm, the pulling tautly spread. While thus fixing on movement that commonly seems the the worm the fish forms a slight last to drop out, but finds the worm Sickle-shape, and then strikes for- will not yield. This may evoke dart- ward grasping the free tip of the ing forward. And if the worm still worm. Hence this maneuver is the remains fastened, the fish may resort same as that used in capturing free- to lying on the side, or to making swimming prey, except it has a dif- stronger lunges forward. The en- ferent orientation. The upright ap- vironmental feed-back seems to play proach is often used in catching the a smaller role in the movements used first worm. However, if both the in stalking the prey. upright and on-the-side approaches Once a juvenile Badis badis suc- occur in the same feeding bout the ceeds in pulling a worm free from on-the-side technique more often ap- the bottom, it swims a distance of pears first and subsequently is re- several body lengths up into the placed by the upright method. water with the worm dangling from The techniques used to pull worms its mouth. But a fully grown Badis free of the substrate can be grouped badis with a cluster of worms pro- into three characteristic types, swim- truding from its mouth usually backs ming backward, swimming forward away slowly ; a single worm simply (including circling), and lying on disappears into the mouth. Backing one side. Swimming backward and up apparently serves to keep the forward are largely self-explana- free ends of the worms from touch- tory ; the variation lies in the amount ing the face. of force exerted, and the number of The most significant characteristic times the fish pulls on the worm. By of hunting behavior in Badis badis Ethology of Badis badis 183 is that catching and eating each prey length) they can be observed ap- produces steadily increasing parsi- proaching and interacting with other mony of behavior in capturing the Badis badis juveniles. Other than next, if this is done at once. The between juveniles, and perhaps there most complete expressions of this as well, aggregating appears to re- behavior contain the most move- sult only from special circumstances. ments and positions deviating most Immature fish. The congregating from the norm. In short, they re- of juveniles is a weakly developed quire the greatest expenditures of type of aggregating. The fish come energy. Hence increasing energy re- together, usually by one approach- quirements appear to be positively ing the other, fight briefly, and sepa- correlated with progressively higher rate. One often appears to be driv- thresholds of the motor patterns of ing the other way, or the two merely hunting behavior (see also van Iersel withdraw from each other. The and Bol, 1958). spacing out is not so great as to Prey-capture behavior may per- keep the fish from continually in- sist in an that has fed to teracting with one another. Dis- repletion (Lorenz, 1937, 1939 ; persal seems to be counteracted by RAber, 1950; Thorpe, 1956). Badis some degree of attraction between badis was never observed to hunt them. This conclusion, however, is after it stopped eating, even if no highly tentative. If the fish actively hunting had been involved. Al- disaggregated as a result of fighting though the motor patterns of hunt- one would expect them to become ing drop out quickly, Badis badis uniformly spread out, and the fight- may continue to eat if it can pick ing to be reduced thereby. Such up loose worms. Thus the threshold was not the case even though food of hunting seems to rise faster than was not confined to one place, and that of eating. In nature the animal the fish were not crowded. Aggre- probably never, or seldom, has the gating in this fashion might not oc- opportunity to capture several cur in nature. And if it did, it worms in quick succession. The would be difficult to rule out selec- elimination of the more elaborate tion of some common environmental elements of the hunting behavior factors, which is also the case in would be adaptive, allowing the ani- the aquarium. mal to capture more worms in a That juveniles probably are not given amount of time, and to con- really solitary is indicated by one serve energy as well. feature of their color pattern. The region embraced by the first few ele- AGGREGATING BEHAVIOR ments of the dorsal fin is dark ; this is followed by a white bar, then dark Badis badis is throughout its life again. As the juveniles, otherwise an essentially non-aggregating spe- cryptically colored, move about this cies. The post-larvae seem to be en- white patch with dark borders tirely solitary. At about the time flashes on and off with the opening they begin developing a barred color and closing of the fin. There prob- pattern (ca. 8-10 mm standard ably is a reason for the elaboration 184 Transactions Illinois Academy of Science of such a striking feature, and the From the increased disposition of reason may lie in the intraspecific the fish to show fright behavior (see signal value of the structure. below), the appearance of the school- Sexually motivated females. Ag- ing behavior itself, and a considera- gregating between two gravid 9' 9 tion of the stimuli provoking it, it is clear cut. The 9 9 , though, must seems reasonable to assume fright be deprived of contact with a S. is the predominant motivation of Such 9' will then spend most of schooling in Badis badis. their time together, only separating occasionally. Infrequently they may FRIGHT BEHAVIOR fight, but this almost never proceeds Flight. In forward flight the beyond initial displays. Two fish darts away at top speed with not sexually motivated, on the other folded fins. The course taken is often hand, will enter into well developed toward the cover of plants, burrows, fights. Thus the aggregating of two or aquarium corners. When shelter gravid, S -deprived 9 probably re- is too far away, or when the fish sults from substituting one another is temporarily too disoriented to find for a f, and is made possible through a suitable hiding place, an adult the suppression of fighting behavior. Badis badis may dart down to the Frightened fish. The last type of bottom and sit there motionless (see aggregating behavior could be called below). Juvenile Badis badis in the schooling behavior. It has only been same situation are more apt to swim observed in and sub-adults. upward. If badly harassed by an- Males might be capable of schooling other Badis badis, an adult will swim if properly manipulated. The fol- upward as a last resort. Then it lowing description applies to Y usually stands in a corner (head When placed in a small, relatively down or up) or among plants (often bare aquarium (but provided with head up). When a juvenile swims hiding places), the fish at first be- upward or forward in flight, the come motionless on the bottom, a body normally swings down pos- typical fright response. After sev- teriorly at the end of the flight if eral minutes one usually makes a the fish is not in contact with the short dart toward another. In this substrate. The body angle, with re- way they gradually assemble in a spect to the bottom, ranges between loose aggregation resting on the bot- 45° and 80°. tom. Then, little by little, they move Flight backwards is performed as in one direction, each swimming in described for normal backward short hops. Eventually they travel swimming. The movement is done several times around the aquarium slowly and the distance travelled is together in this saltatory fashion. seldom more than one body length. The behavior is infectious. If one There are two flight situations in or two residents are present they which backing up is commonly ob- soon behave as the introduced fish. served. The less common occasion As the fish seem to become accus- is when one fish avoids another ; tomed to their new surroundings the this is sometimes observed when a schooling behavior gradually disap- juvenile Badis badis is confronted pears. by a larger one. The behavior most Ethology of Badis badis 185 often observed is the fish backing but sometimes are folded. Rarely into some shelter. This frequently just the anterior-most elements of goes over smoothly into a type of the dorsal fin are erected. If Hover- behavior called Nestling (see below). ing in the water, the transparent Forward and backward flight ap- pectoral fins continue to scull. pear to have different thresholds, When the animal is resting on the that for backward being lower than bottom the pectorals continue to that for forward, particularly when scull with a fast tempo if the cause the fish is in its burrow. Bringing of the fright is relatively weak (far a strange object progressively nearer away, smaller, etc.). As the strength to the shelter releases first backing, of the stimulus increases the pectoral then swimming forward to the en- fins cease moving and the amplitude trance and fleeing forward. Some- of the opercular beats diminishes times the threshold of forward flight perceptibly. The already abducted is extremely high ; touching, enter- fins become even more tautly spread. ing, or even lifting away the burrow For convenience, the situations may not produce immediate flight. evoking suppression of movements Nestling. The performance of may be considered under five groups : Nestling, though species-typical, is (1) predator near (or any novel variable because it depends upon the object), (2) open spaces, (3) attack shape and light conditions of the from a conspecific, (4) mating be- shelter, as well as the disposition havior, (5) stalking food object. of the fish. The fish backs up while Only numbers one and two will be seemingly probing for a crevice with taken up at this point ; number five its tail. Then it presses its body, has already been dealt with, and especially the posterior part, against numbers three and four will be the rear-most region of the shelter treated in later articles. where the light intensity is low. The The protection afforded through median and pelvic fins are folded inhibition of movements in the pres- until the fish assumes a "satisfac- ence of a potential predator, and tory position. Then these fins are when in the open where Badis badis usually spread, and the fish becomes is vulnerable, seems clear. Move- almost motionless (Fig. 3). ments both attract the attention, and elicit the feeding behavior of preda- tors. Two things deserve to be men- tioned about the inhibition of move- ments in exposed places when no predator is actually present. First, the animal must already have a low fright threshold, as when in new Fig. 3.—A male nestling in the rear of a flower-pot. surroundings. Second, the inhibi- tion is often incomplete. The fish Movement suppression. The fish moves extremely slowly across open is almost motionless, and the median spaces, most of the time just over and pelvic fins most often are spread, the bottom. When cover is finally 186 Transactions Illinois Academy of Science at hand, the fish darts into it. Once is not manifested, but is only de- inside, it may turn around, probe duced. When Flight is inferred the bottom, and so forth. The pres- from non-Flight behavior, the term ence of cover (= fish out of preda- is no longer used objectively. The tor's view) removes the suppression second difficulty in using Flight of movement. arises when one attempts to unravel Defense. Fleeing could be con- the complex motivational causes of sidered as a type of defense in the social behavior, particularly dis- negative sense, as could also hiding. plays. Flight in itself lies further Positive defense consists of spread- down in the hierarchical chain than ing the fins. This may be accom- do Sexual or Aggressive behavior, panied by rotating the body slight- against which it is usually compared. ly, presenting the erect dorsal fin A more balanced comparison of mo- spines to the object eliciting the tivations lies in the application of behavior. the higher level concept of Fright, Interactions. Andrew (1957) or Fear. uses the term Fear as, " . . . the Fright in Badis badis finds expres- tendency to give fleeing and related sion in three more or less conflicting responses." Figure 4 should make kinds of behavior, Avoiding, Con- clear what is meant by Fright in cealing, and Aggregating (Fig. 4) ; the present article ; it is roughly defending is compatible with all equivalent to Andrew 's usage of three expressions. The most extrewe Fear. type of Avoiding is panicky flight ; There has been an inclination in the behavior has negative valence investigators to use the terms Escape since the only taxis is away from or Flight, or Flight (Escape) Tend- the stimulus, if the stimulus is lo- ency, in preference to Fright or cated. The extreme manifestation of Fear. For one thing Flight Tend- Concealing is complete absence of ency is considered more objective movements. Goal directed flight al- because it is founded on overt be- ready contains an element of Con- havior. Yet the notion of Flight is cealing, since its goal is usually a used freely in situations when Flight hiding place. Backward flight may

FRIGHT

AVOIDING CONCEALING AGGREGATING

FLIGHT FORWARD MOVEMENT SUPPRESSION

FLIGHT BACKWARD SCHOOLING HORIZONTAL UP DOWN HORIZONTAL Fig. 4.—A hypothetical construct of the hierarchy of fright induced behavior. Ethology of Badis badis 187 be simple Avoiding, but often it is Fright behavior in Badis badis influenced by a need for Concealing. is usually an integrated perform- Actions still more intermediate be- ance involving both Avoiding tween Avoiding and Concealing are (flight) and Concealing (movement Nestling (see above) and perhaps suppression), or Aggregating Digging (M.S.). Aggregating could (schooling). be considered as flight behavior, but with positive valence because the SUMMARY transport is toward the fellow Badis badis. Concealing also enters in ; Badis badis spends much of its the individuals in the aggregation time hovering by means of pectoral show considerable suppression of fin undulations. This sculling action movement. Aggregating, however, of the pectorals also propels the fish should be recognized as a factor on backwards, and forwards. Lateral a par with Concealing and Avoiding trunk undulations may appear when once schooling behavior emerges swimming forward. In general, the fully. median and pelvic fins are folded The change in valence of fright while swimming, and spread while orientation in social behavior is par- stopping. ticularly relevant to motivational Comfort movements recorded in- analysis. Movement away from an- clude fin flickering and rubbing, other individual is often accepted chafing, flexing the body, coughing, as an expression of flight tendency. and yawning. As just pointed out, however, fright Badis badis eats only live, moving behavior may have a positive valence, prey. The fish approaches the prey under proper conditions, bringing slowly, pauses, flexes the body, then the individuals closer together. This darts forward, grasping the prey does not cause much confusion in with its mouth. If the prey is im- analyzing the displays of Badis bedded in the substrate the fish may badis, but it becomes an important lay on its side and use leverage to consideration when studying the be- pull it loose. havior of more complex fishes, such Usually Badis badis is solitary. Some attraction between juveniles, as the Cichlidae. . It is possible to demonstrate the and between ripe C , is apparent. conflict between fleeing, backward Strong fright sometimes induces or forward, and suppression of move- schooling behavior. ment in a Badis badis by placing it Fright is indicated by three types in a small aquarium with inadequate of behavior : flight (avoiding), move- shelter. Tapping on the glass with ment suppression (concealing), and one's finger, varying the strength schooling (aggregating). and distance of the tapping, causes the fish to lie still, suddenly jump LITERATURE CITED forward and immediately back, and ANDREW, R. J. 1957. The aggressive and so forth. Thus the behavior switches courtship behaviour of certain em- berizines. Behaviour, 10: 255-308. back and forth rapidly between the I3AERENDS, G. P., and J. M. BAERENDS-VAN possible outlets of Fright. Romv. 1950. An introduction to the 188 Transactions Illinois Academy of Science

ethology of cichlid fishes. Behaviour of a Science, Study 1, Vol. II. S. Koch, Suppl., 1: 1-242. ed. McGraw-Hill, New York, pp. 561- BREDER, C. M., JR. 1926. The locomo- 610. tion of fishes. Zoologica, 4: 159-297. IERSEL, J. J. A. VAN, and A. C. A. BoL. BREDE:R, C. M., JR. 1936. The reproduc- 1958. Preening of two tern species. tive habits of the North American A study on displacement activities. sunfishes (family Centrarchidae). Behaviour, 13: 1-88. Zoologica, 21: 1-48. LORENZ, K. Z. 1937. tber die Bildung FORSELIUS, S. 1957. Studies of anabantid des Instinktbegriffs. Die Naturwiss., fishes. I. A qualitative description 25: 289-300, 307-318, 324-331. of the reproductive behaviour in ter- LORENZ, K. Z. 1939. Vergleichende Ver- ritorial species investigated under haltensforschung. Zool. Anz. Suppl., laboratory conditions with special re- 12: 69-102. gard to genus Colisa. Zool. Bid. Fran LORENZ, K. Z., and N .TINBERGEN. 1938. Uppsala, 32: 93-302. Taxis und Instinkthandlung in der GRAY, J. 1933a. Studies in animal loco- Eirollbewegung der Gr augans. I. motion. II. The relationship between Z. Tierpsychol., 2: 1-29. waves of muscular contraction and RARER, H. 1950. Das Verhalten gefan- the propulsive mechanism of the eel. gener Waldohreulen ( Asio otus otus) J. Exptl. Biol., 10: 386-390. und Waldkauze ( Strip aluco aluco) GRAY, J. 1933b. Studies in animal loco- zur Beute. Behaviour, 2: 1-95. motion. III. The propulsive mecha- THORPE, W. H. 1956. Learning and in- nism of the whiting ( Gadus merlan- stinct in animals. London, Methuen, gus). J. Exptl. Biol., 10: 391-400. 493 pp. HINDE, R. A. 1959. Some recent trends in ethology. In Psychology: a Study Manuscript received July 7, 1961.

Reprinted from the Transactions of the Illinois State Academy of Science, Vol. 54, Nos. 3 and 4, 1961