82 PROCEEDINGS OF THE AMERICAN ACADEMY 



cell of the bee has not the strict conformity to geometrical accuracy so 

 often claimed for it, but, as the best observers have maintained, is liable 

 to marked variations, chief among which the following may be men- 

 tioned. 



1st. The diameters of worker cells may so vary, that ten of them 

 may have an aggregate deviation from the normal quantity, equal to 

 the diameter of a cell. The average variation is a little less than one 

 half that amount, viz. nearly 0.10 inch, in the same number of cells. 



2d. The width of the sides varies, and this generally involves a va- 

 riation of the angles which adjoining sides make with each other, since 

 the sides vary not only in length, but in direction. 



3d. The variation in the diameters does not depend upon accidental 

 distortion, but upon the manner in which the cell was built. 



4th. The relative size of the rhombic faces of the pyramidal base is 

 liable to frequent variation, and this where the cells are not transitional 

 from one kind to another. 



5 th. When a fourth side exists in the basal pyramid, it may be in 

 consequence of irregularity in the size of the cells, or of incorrect 

 alignment of them on the two sides of the comb. 



6th/ Ordinarily, the error of alignment does not amount to more 

 than one or two diameters of a cell. But occasionally the rows of 

 cells on one side of the comb may deviate from their true direction 

 with regard to those on the other, to the extent of 30°. In conse- 

 quence of this deviation and the continual crossing of the rows on op- 

 posite sides, the pyramidal base is not made, and the cell is thereby 

 shortened. 



7th. "When a piece of comb is strongly curved, or two portions 

 form an angle with each other, there is no constant way in which the 

 tendency to the distortion of the cells is met ; consequently the cells 

 found at the curves or angles have a variety of patterns. 



8th. Deformed worker and drone cells are used for rearing the 

 young. 



All of these variations are found in the three different kinds of cells, 

 but are much more frequent and marked in the honey than in either 

 worker or drone cells. In view of the frequency of such, however 

 near the bee may come to a typical cell in the construction of its comb, 

 it may be reasonably doubted whether a type cell is ever made. Here, 

 as is so often the case elsewhere in nature, the type-form is an ideal 

 one, and, with this, real forms seldom or never coincide. Even in crys- 



