178 BEES AND BEE-KEEPING. 



the corner (/) ? as, for reasons previously given, she 

 must, and that, too, from the very position in which 

 we have placed her, if the straight boundary lines 

 are to be modelled. The orthodox accommodation 

 cell, which is really partially double, is seen 

 (Fig. 3) above two unsealed drone larvse, and 

 in it the septem is not continued to the top. A 

 few somewhat irregular forms, in addition to such 

 a one, will enable the bees to pass completely 

 from one size to the other. But even where the 

 greatest difficulties are presented, no angles of less 

 than ioo° are found — e.g., in B, Fig. 39, made from 

 a tracing of actual comb, constructed from pieces of 

 drone and worker, placed near to each other, for the 

 industrious little insects to repair and join ; a few 

 irregular cells are made to complete the accommoda- 

 tion, the impossible angular point, of course, not trans- 

 piring. The nearest approach to an angle lies at e, 

 where the width of the bee's head determines the 

 obtuseness, while at b the cell is only a depression, 

 not extending to the midrib, because its small size 

 prevents the entrance of the worker's body. Such a 

 cell fills a gap, but is in no other way utilised. 



It will be at once remarked, that the normal cells 

 in B stand between hexagons and circles. This is 

 true, more or less, of all comb, which, if cut through 

 in the middle of its cell partitions, shows these to be 

 very nearly straight up to the angles of the hexagons, 

 although some thickening is observable at the line of 

 junction ; but the end of the walls, at the face of the 

 cell, is always loaded by a rim of wax, which converts 

 the mouth into an approximate circle. This thicken- 



