EXCITATORY AND INHIBITORY PROCESSES 291 



short, brush-like, endings which might undergo less well sustained deforma- 

 tion during stretch, thus accounting for the more rapid adaptation of this 

 receptor (Fig. 3).* In this case a mechanical effect might be largely responsible 

 for the striking differences in adaptation rate. However, one cannot assume 

 that differences in adaptation are produced only by mechanical factors. No 

 comparative studies concerned with the role of applied currents on the 

 discharges of the slowly and the rapidly adapting receptors have yet appeared. 

 Terzuolo and Bullock (1956), and Hagiwara et al. (1960) studied the effects of 

 appUed current on the slowly adapting receptor only. 



Processes Leading to the Production of Conducted Axonal Impulses 



By appropriate placement of external leads or by insertion of a micro- 

 electrode into the soma, it is possible to record a fairly well maintained 

 depolarization which occurs whenever the preparation is stretched. Applying 

 a weak tension one is able to record this type of potential only, or, by using 

 small doses of novocaine one can block the propagated nerve impulses and 

 study only this slow response under greater degrees of stretch (Fig. 4). 

 Similar potentials have been recorded from other receptor organs such as the 

 mammahan Pacinian corpuscle (Alvarez-Buylla and Ramirez de Arellano, 

 1953; Gray and Sato, 1953; cf. also, Loewenstein, 1959), the frog muscle 

 spindle (Katz, 1950), the retina of some invertebrates (cf. Harthne et al., 1952) 

 and from olfactory receptors (Ottoson, 1956). In line with a suggestion by 

 Granit (1947) this potential has been called the generator potential. In the 

 crustacean stretch receptor it is produced at the end of the dendrites; it is 

 non-propagated, graded and spreads electrotonically over the rest of the 

 system. In normal preparations, propagated potentials are set up if the 

 generator potential reaches a critical level of about 10 mV in slowly adapting 

 cells (Fig. 5) and of about 20 mV in fast-adapting neurons. The actual 

 magnitude of the generator potential produced at the dendritic terminals is 

 unknown since it has been recorded only at some distance from its site of 

 origin. Consequently, it must have been attenuated considerably, although it 

 is difficult to estimate the degree of attenuation since the space constant of 

 dendrites is unknown. 



In slowly adapting cells the generator potential seems to be better main- 

 tained than in rapidly adapting neurons where a rather conspicuous decay 

 occurs during constant stretch. This phenomenon might explain the fact that 

 rapidly adapting neurons discharge only for a very short period of time since 

 the generator rapidly declines to values below threshold at the region respon- 

 sible for setting up propagated impulses. 



* The term pre-potential as used in the illustration and subsequently refers to a slowly 

 increasing depolarization that precedes the spike. Consequently, the pre-potential is 

 smaller immediately after the previous orthodromic discharge and increases by several 

 millivolts until the next orthodromic discharge appears (cf. Eyzaguirre and Kuffler, 1955a). 



