The first physiological theory of emotions was formulated at the end of XIX century. According to this theory, set by William James and C.G. Lange, emotions are perceptions of bodily changes. In late 1920s this theory was abandoned due to W.B. Cannon’s criticism. Recently the interest in James-Lange’s theory has been revived especially due to neurological researches done by Antonio Damasio – he claims that the emotion is the perception of bodily changes. Note that this is not James’ thesis, but its modifications. In the paper I show how James-Lange’s theory could have been defended in the framework of their theories alone, independently of recent neurological results. It follows that for understanding emotions it is essential to understand the nature of perception.