Ken YON, The Brain of the Bee. 159 



take in a larger region, besides appearing to have a different 

 mode of branching. 



The difference in size between figs. 23 and 26 is mostly a 

 difference of magnification, which in the former is 121 and in 

 the latter a 102 diameters, yet the fibers represented in fig. 23 

 are actually the coarser. 



Just where either of these two kinds, the fine and the 

 coarse fibers originate, or what other parts of the brain are con- 

 nected with the mushroom bodies by them is yet to be deter- 

 mined. Almost all that can be said is that six different tracts 

 of fibers terminate here, two from the optic lobe, one from the 

 ventral region, and three from the antennal lobes. To these 

 may probably be added a seventh of a commissural nature. 

 These tracts will be described further on. 



The Fibers Ending in the Stalk and the Roots. 



In certain places (fig. i, PI. XIV), as already pointed out, 

 fibers entering the anterior roots of the mushroom bodies may 

 be very readily seen in sections treated after the copper-haema- 

 toxylin method. In preparations by the bichromate of silver 

 method others may be distinguished. One large tract traced 

 to its cells of origin behind and between the stalks of the mush- 

 room bodies by Viallanes ( 87) in his study of the wasp spreads 

 out on the side of the anterior half of the anterior root and 

 branching sends one branch into it. The fibers of this 

 pass nearly horizontally across or obliquely downwards to the 

 opposite side giving off from their upper side several sub- 

 branches (fig. 21, PI. XXI) that after passing a greater or less 

 distance upward break up into a thickly bushy head. These 

 all occupy almost the same level so that often in specimens 

 treated with osmic acid or with haematoxylin, especially the 

 latter, the appearance is presented of several bands crossing 

 transverse sections of the roots. Figures 3 and 4 show these 

 bands excellently, and represent the two roots as seen in a 

 slightly oblique section treated with copper sulphate and a so- 

 lution of haematoxylin made according to one of Weigert's for- 

 mulae. Bands of this sort occur throughout the whole extent 



