28 THE PHYSIOLOGY OF INSECT SENSES 



with mechanoreceptive hairs is not always clear (Fig. 16). It is reported 

 as being inserted on that part of the hair shaft that extends below the 

 socket, on the socket itself, or at some point part way up the shaft of the 

 hair. It is enclosed in a scolopoid sheath, but may or may not be 

 capped with an apical body. The sheath may be ribbed and possess 

 hickened zones. Just below the cap may be seen with the light micro- 

 scope some darkly staining spots, the sense rods (Snodgrass, 1926) or 

 points sensoriels (Hsu, 1938). Further observation with the electron- 

 microscope may reveal more complicated structural relationships 

 similar to those seen in tympanal chordotonal sensilla (Gray, 1960). 



Richard (1952) showed clearly in the case of sensilla trichodea of 

 termites that at moulting the sheath and the neuron process extending 



.«''sw>'»«*».^n«<JujW 



1.0 

 MV 



1 1 



MMMiffMMI 



Fig. 17. Response to repetitive mechanical stimulation of a chemosensory 

 hair on the wing ofSarcophaga. The arrow indicates the onset of a rapid 

 one-directional displacement, and the direction of the displacement 

 with reference to the preceding stimulus. Positive potential at recording 

 electrode is down. Time marks recur at 0-2-second intervals. (Courtesy 

 of M. L. Wolbarsht.) 



through and beyond it are lost. A new sheath and filament are regener- 

 ated to supply the new hair. During the period when the new cuticle is 

 forming the distal fibre remains attached to the old hair to provide 

 tactile sensitivity until the last moment. Shortly before the moult the 

 distal fibre ruptures, and at this time there is a loss in tactile sensitivity 

 (Richard, 1952; Wiggles worth, 1953). 



From the geometry of these sensilla it is likely that bending of the 

 hair in its socket deforms the terminal region of the distal process of 

 the neuron. The response to angular deflexion varies with direction 

 relative to the long axis of the joint (Pumphrey, 1936; Pringle, 1938 a). 

 Stimulation is followed by an electrical response that can be divided 

 into two components. The work of Wolbarsht (1960) has shown that 

 there is a graded slow potential, the receptor potential, that can be 

 recorded from the distal process. It occurs prior to any impulses, varies 



