36 THE BUILDING OF THE CITY 



a very large area had been thus worked upon that 

 the cells began to assume the hexagonal form. 

 The fact is that there is a certain thickness, or 

 rather, thinness, beyond which the bees will not 

 go. As soon as they reach a point when, working 

 as they do from opposite sides at the same time, 

 the slightest further amount of work would break 

 through the partition, they stop. It seems clear 

 to me that, starting with circular cells, and 

 working in this manner, the hexagonal form is 

 inevitable. The hexagonal shape makes it 

 necessary that the cells of one row should be 

 situated not exactly under those above, but in 

 the intervals between them. This result corre- 

 sponds with what one would naturally expect 

 when the bees are working together as closely 

 as is the case, for the most economical method 

 of working is the quincunx system, which is 

 precisely the same as the disposition of the cells. 

 Very much has been said about the perfection 

 and beauty of the combs, and I yield to no one 

 in my admiration of the excellent work, but it is 

 quite overdoing it to say they are always mathe- 

 matically exact in size and shape. Quite apart 

 from what are known as " transition " cells, which 

 occur when a change is made from constructing 

 worker cells, in order to begin building the larger 

 drone cells and also the queer-shaped cells which 

 occur at the points of attachment and in odd 

 corners where it would be impossible to fill the 



