The microelectrode technique has been employed by many investigators with the aim of localizing ERG components in the retina. However, the results and/or interpretations of these studies have not always been in agreement. A difference in technique may account for, in part, the difference in results, but in so far as the interpretation is concerned, disagreement may arise because of the different criterion employed in the localization of the ERG components. Three main types of criteria used so far are collected. 1) Point of maximal amplitude, 2) point of polarity reversal and 3) point of maximal potential gradient of the intraretinal response. BROWN and WIESEL (1961b) employed the maximal amplitude criterion since they could not observe a polarity reversal of the intraretinal response in the cat eye. But subsequent work by ARDEN and BROWN (1965) did reveal a polarity reversal of the ERG when the low-resistance shunt of the vitreous humor was abolished by replacing with a non-conductive oil. MOTOKAWA et al. (1959) and OGAWA (1961) also used maximal amplitude as a criterion in analyzing the potentials of the carp retina where no polarity reversal was apparent. TOMITA at al. (1960) employed polarity reversal point as a criterion, since in the depth recording from the freshly excised opened eye of the frog, potential reversals of the ERG components are always observed (" TOMITA pattern" designated by BRINDLEY (1960)), and this fact provides the evidence that these components are generated from radially oriented structures.The polarity reversal criterion, however, is limited in application. As has been analyzed by TOMITA at al. (1960) and BROWN and WIESEL (1961a), the vitreous humor acts as a low-resistance shunt which permits current generated in distant regions to flow into the recording site. Therefore, in the presence of the R membrane lying just back of the retina, an electrode penetrated in the retinal tissue