OLFACTORY SENSE OF THE HONEY BEE 319 



parent spot, which may be round, oblong, or sHthke. On the 

 wings these transparent spots are always round or oblong (fig. 

 5, PorAp) and never slitlike as on the legs (fig. 12, A, PorAp). 

 By slowly focusing down on this transparent area it appears to 

 diminish in size and at the lowest level is a transparent spot, 

 which is perfectly round on the wings and legs. This spot is an 

 opening in the chitin, the pore aperture (PorAp). If the speci- 

 men examined is light, the dark pigmented pore border is more 

 noticeable. On the wings this border is narrow (fig. 5, PorB) 

 while on the legs, where the pores are not so close together, it is 

 wider and when the pores are near each other the borders (fig. 

 12, A, PorB) merge and appear more conspicuous. 



The chitin between the pores on the legs is marked, as else- 

 where, with light colored lines (fig. 12, A, ChM). These mark- 

 ings on the chitin are never present between the pores on the 

 wings but occur everywhere outside the groups. These mark- 

 ings on the chitin have been omitted in all the drawings except 

 figure 12, A. 



In comparing a hair socket with an olfactory pore the following 

 differences are noticed; the largest pore is never as large as the 

 largest socket, and the boundary of the hair socket (fig. 12, A, 

 PorWHr) is always much darker and heavier, its aperture is 

 never slitlike, and even if round or oblong it is ragged, showing 

 where the hair has torn the edges when pulled out. As a rule 

 the border of the hair socket (PorBHr) and even the chitin inside 

 the boundary are much darker than are comparable parts of the 

 olfactory pore. The borders of the hair sockets are usually much 

 narrower than are those of the pores. All these differences are 

 fairly constant, but the sockets of the smallest of the spinelike 

 hairs (fig. 15, SpHr^i) on the legs are almost identical in appear- 

 ance with the smallest pores and they are distinguished only by 

 the ragged condition of the sockets. In figure 15, on the tro- 

 chanter, the various kinds of hairs and the pores which belong 

 to groups 7, 8, and 9 are shown. The pores are shown with 

 slitlike openings while the hair sockets have irregular rings round 

 the sockets. If these pores were drawn with dark borders, as 

 they really appear when seen through the microscope and as the 



