Water salinity influences fish development and growth. Better growth is often observed when fish are reared at intermediate salinities, but the underlying mechanisms are a matter of debate (reviewed by BÝuf & Payan 2001). Growth rate and food conversion efficiency of cod (Gadus morhua L.) were proposed to correlate with trypsin activity (Lemieux, Blier & Dutil 1999), and Atlantic salmon (Salmo salar L.) with trypsin isozyme TRP-2 Ã 92 had better digestion, absorption, protein utilization and growth than fish with other isoenzymes (Torrissen & Shearer 1992; Torrissen, Lied & Espe 1994). Trypsin and chymotrypsin are the major intestinal proteases (Lemieux et al. 1999 and references cited therein), and any changes in their activity may influence digestion and absorption of the food. Exposure of fish to water of different salinities results in changes in drinking rates (Usher, Talbot & Eddy 1988), and it is possible that digestive enzyme activity could be affected by the salinity of the gut contents. The objective of the present study was to examine the effects of salinity on the activity of major digestive proteases in a euryaline sparid, the gilthead sea bream (Sparus aurata L.). Forty juvenile gilthead sea bream, obtained from a local fish farm (ICHTHYO SA, Vathikoilo Pelasgias, Greece), were divided into two groups of 20 fish and stocked in 150-L glass aquaria. Fish were subjected to either 33%(High salinity: Hs) or 20%(Low salinity: Ls) and during the 8-week experiment they were offered a commercial pelleted diet (Ecolife 15, 4.5 mm, Biomar Hellenic, Volos, Greece), supplied at a ration of 3.5% b. wt dayÀ 1. The diet composition according to the manufacturer’s specification was: 47% protein, 20% lipids, 8% ash and 1.7% cellulose. The entire ration was always consumed, and food supply was adjusted weekly after weighing the fish. During the experiment, photoperiod decreased from 13L: 11D to 10L: 14D, water temperature ranged between 21.6 1C and 24.8 1C and no mortalities occurred. At the end of the experiment, all fish were killed by an overdose of anaesthetic, blotted dry, weighed, dissected and the digestive tract removed. Digestive tracts were cut longitudinally, contents removed, stomachs were separated from the intestines and the tissues were then frozen individually at À 20 1C until use. Tissues were thawed by stirring them in distilled water. Stomach samples were homogenized in distilled water (1: 10 w/v), whereas intestine samples were homogenized (100mgmLÀ 1) in cold 50mM Tris-HCl buffer, pH 7.5 (Alarco¤ n, D|¤ az, Moyano & AbellaŁn 1998). Supernatants obtained after centrifugation (16000 g for 30min at 4 1C) were stored at À 20 1C until enzyme analysis. Acid protease activity was evaluated according to Anson (1938) using 0.5% haemoglobin in 0.1 mM glycine/HCl, pH 2.0. The reaction mixtures were incubated at 25 1C for 30 min and the reaction was terminated by adding 0.5 mL of 20% TCA. Alkaline protease activity was measured as described by Alarco¤ n et al.(1998) using 0.5% casein as a substrate in 50 mM Tris-HCl buffer, pH 9.0. One unit of enzyme activity was defined as 1 μg of tyrosine released per min. Trypsin activity was measured at 25 1C using BAPNA (N-benzoyl-L-arginine-p-nitroanilide) substrate in 50 mM Tris-HCl buffer, pH 8.2, containing 10 mM CaCl2 (Erlanger, Kokowsky & Cohen 1961).