Non-steroid anti-inflammatory agents are used in the treatment of bovine coliform mastitis, in order to reduce the severity of the clinical signs, to improve cure and to minimize the adverse effects of the disease (Lohuis et al., 1989;Morin et al., 1998). In ewes, similar findings have not been reported and consequently, the efficacy of these drugs in the treatment of ovine mastitis is not known. The objective of this study was the evaluation of flunixin meglumine (FINADYNE ® Schering-Plough) in the supportive treatment of ovine mastitis.Sixty dairy ewes of various ages and local Greek breeds, with spontaneous acute clinical mastitis were included into the study. To account for the likely flock-effect and to improve the efficiency of the study in terms of the amount of information per subject studied, a matched case-control sampling design was followed (Rothman & Greenland, 1998). The matching factor was the flock of origin; within each flock and for every case-ewe (i.e. a ewe with acute clinical mastitis) the first ewe with acute clinical mastitis after the case-ewe was selected as control.Ewes were allocated into group A (test) or B (control) (n= 30). Ewes in group A were treated with cefuroxime (SPEC-TRAZOL ® , Schering-Plough) and with flunixin meglumine simultaneously and those in group B with cefuroxime only. Cefuroxime was administered intramammarily on three occasions, every 12 h, and the dose was 250 mg (one tube) per occasion per ewe. Flunixin meglumine was administered intramuscularly on two occasions, with a 24 h period between, and the dose was 100 mg (1.82-2.33 mg/kg body weight) per occasion per ewe. Treatment was instigated within 30-60 h after the initial observation of the signs of mastitis (abnormal secretion, swollen mammary gland, change of colour of the udder skin etc.) by the farmer.Before treatment (D0) an initial clinical examination of each ewe was carried out and mammary secretion samples were collected. The rectal temperature was recorded and the mammary glands were examined. Seven different mammary or systemic reactions were observed and scored on a 0 (no reaction) to 4 (very severe reaction) scale (Table 1). The total clinical reaction score was produced by addition of all mammary and systemic reactions scores and ranged from 0 to 28. Subsequently, clinical examination and collection of mam-mary secretion samples were carried out 1 day (D1), 2 days (D2), 4 days (D4) and 14 days (D14) after initiation of the treatment. Mammary secretion samples were examined bacteriologically by using conventional methods (Fthenakis, 1988(Fthenakis, , 1994.The clinical cure rate of each group was the proportion of ewes without clinically detectable abnormal signs. The bacteriological cure rate of each group was the proportion of ewes with no bacteria isolated from mammary secretion samples. The mean rectal temperature (MRTemp) in each group was calculated as: MRTemp= (Temp 1 + Temp 2 + …+Temp n )/n, and the mean total clinical reaction score (MTCRS) in each group was calculated as: MTCRS=(S 1 + S 2 + ...