The influence of piglet sensory systems and the role of odors from the sow's ventrum on piglet nipple attachment were investigated. In Experiment 1, four sows had their ventrum washed with organic solvents and four were unwashed control sows. Piglets were given one of four treatments shortly after birth: nares that were mechanically blocked (B), a lidocaine flush of the olfactory system (LFO), a saline flush of the olfactory system (SFO) or nontreated controls (C). Time to attach to a nipple, up to a maximum of 600 sec, was recorded. On unwashed sows, nipple attachment was affected (p less than 0.01) by piglet treatment. Piglets have B or LFO treatments took longer to attach to a nipple than SFO or C piglets. Piglets of all treatments took longer to attach to washed sows than to unwashed controls. Seven unwashed sows and their litters were used in Experiment 2. Piglets were tested after birth with one of the following treatments: an olfactory system flushed with lidocaine, lidocaine on nose (LN), lidocaine applied to the tongue (LT) or nontreated controls. All lidocaine-treated piglets took more time to attach than C piglets. Nipple attachment latencies were intermediate for LN or LT piglets. In Experiment 3, three unwashed sows and their litters were used. Treatments were: lidocaine applied to both the nose and the tongue (LNT), an olfactory system flushed with lidocaine or nontreated controls. LFO piglets had an increased latency to attachment compared to controls while LNT piglets were intermediate in latency to attach. Piglet olfaction, gustation and tactile sensory modalities as well as odors on the sow's ventrum influenced nipple attachment.(ABSTRACT TRUNCATED AT 250 WORDS)