MECHANICS AND NEURAL RESPONSE OF RECEPTOR SYSTEMS 



407 



canals will be stimulated. Another con- 

 ceivable difficulty would be the possibility 

 that canals may interact mechanically with 

 one another. This, however, does not seem 

 likely from Lorente de No's result (cited by 

 Spiegel and Sommers, 37), that endolymph 

 flow in one canal does not produce endo- 

 lymph flow in the other two canals. 



In closing this section at least brief men- 

 tion should be made of stimulation thresh- 

 olds. We have said above that it seems 

 reasonable to take <r as the rate at which the 

 central nervous system is being stimulated 

 by the peripheral receptors. Assuming that 

 there is some criterion by which one may 

 decide that the central nervous system has 

 responded, e.g., the onset of nystagmus, one 

 may inquire as to what the smallest <r is 

 which will just elicit the response. This 

 is the rheobasic value of cr. A purely formal 

 but convenient picture which has developed 

 in physiology is that the stimulation builds 

 up in the responding system some sort of 

 excitatory state; at the same time this state 

 is subject to dissipation by independent 

 means. If in this offset the excitatory state 

 ever reaches or exceeds a certain threshold 

 value, however, it unstabilizes some hypo- 

 thetical configuration and we have the 

 response. In these terms it is easy to "ex- 

 plain" why any moderate rate of stimulation 

 requires some time (so-called "latency") be- 

 fore it is able to evoke the response. The 

 rheobasic value of cr (what is apparently 

 referred to in psychological literature as 

 the "threshold" value) has then an infinite 

 latency. Carrying these concepts over to 

 the motion-sickness problem, it is evident 

 that whether or not the a corresponding to a 

 particular ship motion exceeds the threshold 

 value for effecting sickness is a question of 

 great importance. That this is not so for 

 the cr's resulting from ship rotations seems 

 to be rather widely accepted (e.g., by Sjoberg 

 (34) and by McNally and Stuart (21); see 

 also Morton et al. (26)). These conclusions 

 should be considered with considerable 

 caution, because in at least one instance (9) 



the various angular accelerations were not 

 at all long enough to establish a true 

 threshold.^ Mowrer's (27) careful study 

 was on a different species (pigeon) and may 

 not be directly applicable to man; however, 

 he found a nystagmus threshold (.79° per 

 sec.^) much lower than the average angular 

 accelei'ations encountered on many ships.^'^ 

 In this connection it should also be noted 

 that Spiegel et al. (36) have cited important 

 evidence to show that the nystagmic thresh- 

 old may be above that of responses of vegeta- 

 tive components of the autonomic nervous 



Fig. 3. Orientation of the left semicircular 

 canals viewed from the left side. According to 

 Summers et al. (23), the Euler angles ^, 6, (p, have 

 values 0°, 30°, and 45° respectively. The con- 

 ventional axes, ^, 77, f, have been marked a, p, h, 

 to indicate that they are normal to the planes of 

 the anterior, posterior, and horizontal canals 

 respectively. 



system associated with illness. It has also 

 been shoAvn (Henry, cited in reference 36, 

 and also Spiegel, 35), that a typical partial 

 symptom of motion siclmess, the laby- 

 rinthine vasomotor reaction, may still be 



^ It should be noted parenthetically that the 

 "thresholds to constant angular velocity" deter- 

 mined by Dodge were actually thresholds to un- 

 known accelerations at the onset of the motion, for 

 as we have seen in equation (8) a turning torque 

 on the cupula acts only at the beginning of such 

 motions. 



^^ See, however, the summary tabulation of 

 threshold determinations (20). 



