Sperm competition is recognized as an important selective force in reproductive biology (Birkhead & M ller 1998) and it occurs when sperm from more than one male compete for fertilization of the ova (Parker 1970). Sperm competition has led to the evolution of adaptive sperm traits (Andersson 1994; Birkhead & M ller 1998), and several studies on ¢sh species have shown a strong positive correlation between fertilization or hatching rate and sperm motility (Kime,Van Look, Mcallister, Huyskens, Rurangwa & Ollevier 2001; Kupriyanova & Havenhand 2002; Gage, Macfarlane, Yeates, Ward, Searle & Parker 2004; Schulte-Hostedde & Burness 2005; Casselman, Schulte-Hostedde & Montgomerie 2006). However, in species with sperm competition, males not only need to be able to fertilize the eggs, but in addition, do better than the competing males to achieve fertilization. Although sperm competition is widespread among ¢sh species (Stockley, Gage, Parker & M ller 1997; Taborsky1998), few studies have examined the in£uence of diierent sperm traits on relative paternity when males compete for fertilization. The Atlantic cod (Gadus morhua L.) aggregate annually at spawning areas in spring, and spawning pairs are often joined by several satellite males that shed their sperm among the newly released eggs (Brawn 1961). In addition, multiple paternities per spawning bout are found in captivity (Hutchings, Bishop & Mcgregor-Shaw 1999; Rakitin, Ferguson & Trippel 1999). Thus, it seems likely that the Atlantic cod is naturally adapted to sperm competition. Wild-caught sexually mature adult Atlantic cod were trawl captured in the basin of Balsfjord (69119 0N, 19121 0E), Northern Norway, in April 2005. Four ripe males and ¢ve ripe females were randomly selected for the breeding experiment, and their gametes were collected by gently applying pressure to their abdomen after drying the ¢sh surface to avoid contaminating the samples. As each female produces millions of small eggs that are externally fertilized, we were able to conduct controlled in vitro fertilization while concurrently measuring sperm traits. Sperm from fourmales weremixed and shed over unfertilized eggs from ¢ve diierent females in an aqueous environment to which Atlantic cod is naturally adapted, in order to examinewhat sperm traits determine relative paternity. An aliquot (o0.12 mL) of undiluted milt placed on a cooled (approximately 5 1C) microscope slide with an 80-mm-deep chamber was used for spermmotility analyses. Sperm motility was induced in a one-step procedure by adding 4.5 mL seawater before applying the coverglass. A video camera (Sony XC-ST50CE, Sony, Tokyo, Japan) mounted on a negative phasecontrast microscope (Olympus CH30, Olympus, Tokyo, Japan) ( 10 objective) was used to record sperm activity from each male. Sperm swimming activity was recorded twice for eachmale and stored on MiniDV tapes. Computer-assisted spermanalysis was used to examine sperm motility in ¢sh, and a total of 2548 sperm cell trajectories were analysed using an HTM-CEROS sperm tracker (CEROS version 12; Hamilton Thorne Research, Beverly, MA, USA). The sperm analyser was set as follows: frame rate 50Hz; number of frames 25; minimumcontrast 9; andminimum cell size 8 pixels. Four qualitatively diierent Aquaculture Research, 2008, 39, 212^216 doi:10.1111/j.1365-2109.2007.01863.x