As is well known the time courses of many physiological and electrical phenomena are quite similar. Therefore electrical and electrochemical models can be used to simulate physiological processes [Bethe; Bonhoeffer; Burns; Druckrey and Kupfmuller; Eccles (1); Freygang; Hermann; Hodgkin and Huxley; Kupfmuller (1); Lillie; Tasaki].1 In modern automation and computer technology refined apparatus becomes necessary which is able to perform such complex functions as learning and pattern recognition, which seemed hitherto to have been restricted to organisms with highly developed nervous systems (Steinbuch; Wiener). The design of such apparatus being rather difficult, communication and electronic engineers are interested in how nature has solved the problems of information processing in the nervous system. The investigation of nervous systems is one of the tasks of biologists, and engineers hope that their questions can be answered by the biologists. However two difficulties arise: firstly, many biologists are not interested in engineering problems, and secondly, biologists and engineers do not approach problems from the same angle.