Printer Friendly

Aggression and oxygen uptake in Siamese fighting fish.

ABSTRACT: Oxygen uptake/gram/hour (V[O.sub.2]) was determined for Siamese fighting fish, Betta splendens, denied access to the surface either in mirrored or non-mirrored (control) test vessels. Aggressive responses by fish in mirrored vessels resulted in higher V[O.sub.2] across a temperature range of 20 to 36[degrees]C. V[O.sub.2 ] was greater in higher temperatures in both the mirrored and control groups, and was significantly higher ([alpha] = .01) in fish in mirrored vessels.

KEY WORDS: Betta splendens; oxygen; mirrors; aggression


The Siamese fighting fish, Betta splendens, suborder Anabantoidie, family Belontiidae, is an air-breathing fish. It has a labyrinth organ that lies behind its gill chamber. These fish take in atmospheric air and extract molecular oxygen with this structure. It also uses dissolved oxygen (D.O.) within its gill structure from its surrounding water environment. Males of this species are used in behavioral studies to examine stimuli that cause aggression. One such stimulus is the appearance of a male Betta within the area of another male of the same species. To examine this type of behavior Bettas have been placed in containers separated by a glass partition, which allows the fish to see each other, and their responses are noted. Another method to observe aggressive behavior employs a single male fish in a container with one or more mirrors placed around the container to allow the fish to see their reflection in the mirror. Using the mirror technique, Colyer and Jenkins (1976) observed that male Bettas produced an increase in ventilation (opercular movements). Metabolic costs in terms of respiration due to aggression is not reported for Bettas.

The purpose of this study was to determine if the aggressive behavioral response reported for this species results in a measurable increase of V[O.sub.2]/gm/hr.


Male Bettas were obtained from several commercial sources. Fish were maintained in the laboratory at 21[degrees]C ([+ or -] 0.5[degrees]C). Prior to use fish were acclimated for at least two days to the temperature of the test. Temperature within each period of acclimation varied less than 0.5[degrees]C. The range of test temperatures was 20 to 36[degrees]C. All tests were performed under lighted conditions. A 380 ml plastic see-through rectangular container was utilized as a static respirometer (Fig. 1 [a]). The dimensions of the respirometer are; interior height 98 mm, top width 65 mm, bottom width 57 mm. A cover was placed on the vessel for the duration of the test. The cover was vented to allow air to escape when emplaced and resulted in a minimal (one to 2 mm) space between the surface water and the cover. For those tests designed to allow the fish to express a response to another male (itself), a mirror was placed on each of two opposite sides outside the test vessel. The mirrors were held in place by rubber bands surrounding the vessel. A single fish was placed into the respirometer. The interior bottom of the respirometer contained a screen. This prevented the fish from being drawn into the drain when, at the conclusion of the test, the water in the vessel was drained to determine the D.O. A second screen, placed beneath the surface of the water, prevented the fish in both mirrored containers, and non-mirrored control containers, from reaching the surface to gulp in the minimal space. Fish were maintained in the respirometers, which were in water baths at the test temperature, for two hours. At the conclusion of the test, the tube was withdrawn from its rubber band support, and then drained into a BOD bottle (Fig. 1 [b]). D.O. was determined by the Winkler method (Standard Methods, 1965). All of the Winklers were performed by the author, and D.O. determinations made within three minutes of the termination of each test. Several fish were returned to the colony, then retested again several days later, not necessarily at the same temperature as in their initial test. An average of 10 fish were tested each day at their temperature of acclimation and included blank vessels for both before and after D.O. determinations. The following formula was used to determine the mls of oxygen per gram per hour: average D.O. of one initial blank vessel (no fish) and a final blank vessel D.O., minus the final D.O. in the test vessel (the fish respirometer). This was multiplied by 0.380 (mls of water in the respirometer chamber) minus the weight of the fish, divided by the duration of the test (two hours), divided by the weight of the fish (grams). This was multiplied by 0.7 to convert to mls of 02. Typically, oxygen consumption is reported in mls [O.sub.2] (Prosser, 1973).

Straight-line linear regression (Sigmaplot 2002 for Windows Version 8.02) for V[O.sub.2] uptake/gram/hour was performed for both control and for mirrored fish.

Data was then grouped into three temperature ranges for both groups. The temperature groups were low, 20 to 25[degrees]C; medium, 26 to 30[degrees]C; and high, 31 to 36[degrees]C. Initially a t-test was performed comparing control and mirrored fish, then t-tests were performed for each of the three temperature groups to determine if the V[O.sub.2] differed significantly between the controls and the mirrored fish.

T-tests were performed within the three temperature groups to determine if the weights of the fish, chosen at random for test placement, differed.


A total of 173 male fish were tested, including 96 controls (no access to surface waters, and no mirrors), and 77 fish with no surface access and mirrors (Figure 2). Three observations of fish with no surface access and mirrors were not included in the analyses. Their V[O.sub.2]/gm/hr values exceeded the mean values for their test temperatures by a standard deviation greater than 2.5.

Oxygen consumption was significantly higher in fish exposed to mirrors ([alpha]= 0.01). By temperatures grouped low, medium, and high, males that could see their reflections had increased V[O.sub.2]/gm/hr of 13, seven, and seven percent respectively when compared with controls of these groups (Table I). Differences in oxygen consumption in the low temperature group was significant ([alpha]= 0.01), but was not significant in both the medium and high temperature groups ([alpha] = 0.01).

Weight (grams) differences for controls and for mirrors did not vary more than seven percent (Table I). Weight differences for all groups did not differ significantly ([alpha] = 0.01). Weights for control and for mirrored fish are shown in Table II.

Levels of oxygen saturation dropped to 35 percent in several vessels at the conclusion of these tests. This occurred in some of the vessels at temperatures of 31, 32, and 33[degrees]C. The fish in these vessels showed no outward signs of stress (i.e. increased activity).


An increase in respiration and a concomitant increase in V[O.sub.2] consumption per gram of fish per hour occurs in Betta splendens as a result of seeing their mirrored reflection. The fish that sees its own mirror image initiates threat displays (TDs) that are elicited by its own mirror image (Lissman, 1932). TDs include erection of fins and opening of gill membranes to their maximum for only a few seconds, and only when looking directly at its image (Fantino et al, 1972). While swimming back and forth the gills are only partially extended (Fantino et al, 1972). TDs wane during prolonged exposure when the fish is prevented from fighting (Figler, 1972). Meliska and Meliska (1976) indicate that Betta's TDs least wane when threats are identically reproduced by the opponent (i.e. mirrored fish). TD intensity varies based upon aggressiveness. Across the temperature range in this study there was an approximate 20 percent increase in V[Osub.2]. However, the difference in V[O.sub.2] consumption declines between the control and the mirrored fish as the temperature increases. This becomes apparent in the high temperature range (31-36[degrees]C).

Fish were selected for control and mirrors at random. For each of the temperature groups the average weight of the controls was less than the average weight of mirrored fish, but was not significant ([alpha] = 0.01). Prosser (1973) reported that when large and small adults of a species, or when different sized species of the same general type of animal are compared, it is found that the total metabolism of the larger animals is higher, but that the metabolic rate of the small exceeds that of the large. For example, Summer and Urless (1942) reported that for Crenichthyes baileyi, small fish have a higher V[O.sub.2] (0.307) than larger fish (0.215) at temperatures of 36[degrees]C. Dorfman (unpublished data) found that for Betta splendens weighing between 0.55 and 2.70 grams, in comparable temperatures, fish weighing less than one gram had higher V[O.sub.2] than those weighing more than one gram, with differences ranging from 40 to 90 percent. In this study small fish, at comparable test temperatures, tended to have a higher V[O.sub.2].

For this research, using four-sided test vessels, two mirrors were placed on opposite sides of the vessels. Additional studies with a single mirror, or with four mirrors, might yield lower or higher V[O.sub.2]/gm/hr because of fewer TDs or almost constant TDs respectively. The test duration of two hours allows adequate uptake to obtain a meaningful D.O. reading, and to limit possible effects of habituation.


The author would like to thank K. Chapman, W. Elliot, and T. Tsoutsas of Monmouth University, for their statistical and computational assistance. This study was partially funded by the Monmouth University Grant-In-Aid for Creativity.


COLYER, S.W., and C. JENKINS. 1976. Pheromonal control of aggressive display in Siamese Fighting fish (Betta splendens). Perceptual and Motor Skills. 42(1):47-54.

DORFMAN, D. 1999. Oxygen uptake in Betta splendens. MACUB Nassau CC. Abstract in Program.

FANTINO, E., S. WEIGELE, and D. LANCY. 1972. Aggressive display in the Siamese Fighting Fish (Betta splendens) Learning and Motivation. 3:457-458.

FIGLER, M.H. 1972. The relation between eliciting stimulus strength and habituation of the Threat display in male Siamese fighting fish. Behavior. 42:63-96.

INNES, W.T. 1935. Exotic Aquarium Fishes a Work of General Reference. Innes Publishing Company. Phi., Pa.; 9th Ed.: p. 511.

LISSMAN, H.W. 1932. Die umwelt des kampffishes (Betta splendens). Zeitschrift Fuer Vergleichende Physiologic. 18:65-111.

MELISKA, J.A., and C.J. MELISKA. 1976. Effects of habituation on threat display and Dominance establishment in the Siamese fighting fish, Betta splendens. Animal Learning and Behavior. 4:161-171.

PROSSER, C.L. 1973. Comparative Animal Physiology. Third Ed. W.B. Saunders Co. Pa.: 966 p. and Appendix.

Standard Methods for the Examination of Water and Wastewater. 12th Ed.: Publ. Office Am. Public Health Assoc., Inc., N.Y: p. 769.

SUMNER, F.B., and U.N. d. 1942. Studies of the respiratory metabolism of warm and cool spring fishes. The Biological Bulletin. Vol. 82;313-327.


Table I. Average uptake of VOZ ml/gm/hr by Betta splendens.

GROUP MIRRORS V[O.sub.2] ([alpha] = 0.01)

All groups Control (96) * 0.1616 S
20-36[degrees]C Mirrors (74) 0.1866
20-25[degrees]C Control (55) 0.1304 S
 Mirrors (35) 0.1496
 Control (19) 0.1801
26-30[degrees]C Mirrors (18) 0.1940 NS
31-36[degrees]C Control (22) 0.2237 NS
 Mirrors (21) 0.2418

* N = Sample size

Table II. Weight values by temperature group.


All groups Control 1.59 2.39
20-36[degrees]C Mirror 1.66 2.42
20-25[degrees]C Control 1.58 2.39
 Mirror 1.69 2.42
26-30[degrees]C Control 1.55 1.75
 Mirror 1.66 2.04
31-36[degrees]C Control 1.62 2.03
 Mirror 1.62 2.07

GROUPS WEIGHT N * ([alpha] = 0.01)

All groups 1.06 96 NS
20-36[degrees]C 1.15 74
20-25[degrees]C 1.06 55 NS
 1.26 35
26-30[degrees]C 1.25 19 NS
 1.15 18
31-36[degrees]C 1.42 22 NS
 1.18 21

N * = Sample size
COPYRIGHT 2005 New Jersey Academy of Science
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2005 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Dorfman, Donald
Publication:Bulletin of the New Jersey Academy of Science
Date:Sep 22, 2005
Previous Article:Analysis and prediction of long period precipitation in New Jersey using statistical methods.
Next Article:Medieval, copper-based green pigment.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters