Because exposure to seminal plasma enhances the detrimental effects of cold shock on boar sperma- tozoa (Lasley & Bogart, 1944; Pursel, Johnson & Rampacek, 1972), many procedures for preserving semen by deep freezing employ sperm-rich ejaculate fractions (Pursel & Johnson, 1975; Larrson & Einarsson, 1975), thereby excluding much of the seminal plasma. Our previous studies showed that a basic protein fraction of seminal vesicle origin binds irreversibly to boar spermatozoa at ejaculation (Moore & Hibbitt, 1976) and during cooling promotes sperm membrane disruption (Moore, Hall & Hibbitt, 1976). We therefore tested the possibility that boar spermatozoa could be frozen more effectively in the absence of seminal vesicular fluid. Spermatozoa from intact boars and from boars without seminal vesicles (Davies, Hall, Hibbitt & Moore, 1975) were frozen in various diluents and comparisons were made of the sperm quality after thawing and the conception rates of gilts. Whole ejaculates were collected from three intact boars and four boars without seminal vesicles (Large White breed), and spermatozoa were frozen to —196°C by a method based on that of Pursel & Johnson (1975). Previous tests with fresh semen inseminations had indicated that the fertility of all the boars was good. Semen samples exhibiting good progressive motility of spermatozoa at 37°C were transferred in a polythene bag to a 28°C water bath. Antibiotics (Streptopen: Glaxo Ltd; 1000 i.u./ml) were added and the semen was cooled to 22°C over 2 h. After centrifugation at 700 g for 15 min, the spermatozoa were resuspended at a concentration of 1 ·2 109 cells/ml in a buffer solution, pH 7-5 (40 mM-Tes, 17 mM-tris and 180 mM-glucose) containing one of the following: (a) 20% (v/v) egg yolk; (b) 2% (w/v) casein; or (c) 2% (w/v) casein, 2-5 rriM-phosphatidyl serine and 10 min- cholesterol. Aliquots (5 ml) were cooled to 5°C at 6°C/h and then diluted with an equal volume of the same diluent to which 2% (v/v) glycerol had been added. These spermatozoa were immediately pellet frozen (in 0-2 ml volumes) on solid C02 (Nagase & Niwa, 1964) and transferred to liquid nitrogen for storage (1-3 weeks). Fifty semen pellets (10 ml) were thawed by first placing them in a styrofoam box (6 20 cm) for 3 min and they were then transferred to 50 ml of IVT diluent (du Mesnil du Buisson & Jondet, 1961) at 50°C. Each sample was assessed for post-thaw motility at 37°C. The percentage of eosinophilic (live/dead) spermatozoa was estimated from nigrosin-eosin smears examined the same day. Random samples were centrifuged and the supernatants assayed for L-aspartate 2-oxoglutarate aminotrans- ferase (GOT) activity as described by Moore et al (1976). Enzyme release was measured as a percent¬ age of the total solubleenzyme activity of thecell. The latter was determined by homogenizing sperma¬ tozoa in IVT diluent, centrifuging at 10,000 g for 30 min and then assaying enzyme activity in the supernatant. Gilts in oestrus were cervically inseminated with semen thawed within the previous 30 min. Each