Preliminary results from an inexpensive motion analyzer for free - swimming zooplankton

Mirrors are used to reflect two orthogonal views of a small aquarium into the lens of a single video camera, thus creating distortionfree, three-dimensional images. Zooplankton motion is quantified with computer software that categorizes different zooplankton behaviors and measures parameters such as distance, duration, angle of movement, and overall swimming speed. Using this system, we have found a strong behavioral response of Diaptomus minutus to the presence of conspecifics; Diaptomus greatly increases swimming speed and frequency of jumps under moderately crowded conditions. Animals often modify their movement patterns to optimize food gathering, mate location, and predator and competitor avoidance. Therefore, many studies in behavioral ecology require detailed information on movements and activity levels of various animals. In the zooplankton, behavioral observations are impeded by the small size of study animals and their ability to make rapid movements. We now rely on technologies such as high-speed microcinematography (Strickler 1977, 1982) to partly overcome these difficulties. This technique has made accessible details of food encounter, capture, and ingestion in tethered zooplankton (Koehl and Strickler 198 1). Techniques used to study the movements of zooplankton that are freely swimming in a three-dimensional environment are not as well developed, however, and our understanding of these behaviors is limited as a result. Most researchers restrict zooplankton to Acknowledgments This research was financially supported by Natural Sciences and Engineering Research Council of Canada operating and Special Equipment grants to Gary Sprules. We thank Peter Hiscocks for his advice and assistance with image analysis equipment and techniques, and Dewey Meyers for the idea of the three-dimensional viewing system. We also thank Milos Legner for providing the continuous culture of Chlamydomonas, and Howie Riessen, Michael Arts, Andrea Locke, and Robert Baker for critical review of this manuscript. swimming in two-dimensional, narrow or shallow tanks in order to simplify both direct and videotaped observation (e.g. Stearns 1975; Wong et al. 1986). Videotaping is generally used to quantify motion through “frame-by-frame” analysis-an arduous and time-consuming process of stopping the playback at regular, brief time intervals and measuring an animal’s position (Wong et al. 1986; Young and Getty 1987). Automated motion analyzers based on minicomputers have been used effectively (Buskey 1984; Koltes 1985) but are quite expensive. With cheaper, microcomputer-assisted techniques, operators must still manually locate the animals on either a video screen or digitizing tablet. Here, using only one video camera and an automated method for motion quantification that works on a microcomputer, we describe an inexpensive technique for reconstructing three-dimensional movements of freely swimming animals. We create a three-dimensional image that shows both X-Y and Z-Y planes on either half of one video image by using a system of four first-surface mirrors (Fig. 1). The mirrors simultaneously reflect two orthogonal views of an aquarium (10 X 10 X 12 cm high) into the field of one video camera (Panasonic WV1850 with a Canon 16100 mm, f1.9 zoom lens). Both the camera lens and mirrors are adjusted to make the angle between the two different lines of sight (angle 0) as close to 90” as possible in order to minimize perspective distortions and thereby simplify calculations of distance and position. During videotaping, the tanks are illuminated from above with a collimated infrared (minimum wavelength, 720 nm) light source that makes zooplankton appear as bright dots against an even, dark background. Using infrared light reduces behavioral responses of zooplankters to the strong directional lighting. Since zooplankton are the only bright parts of the image, our com-