BUCKINGHAM. — DIVISION OF LABOR AMONG ANTS. 435 



the head fell between 1.00 mm. and 1.20 mm. ; the second class between 

 1.20 and 1.40 mm., and so on. The interval adopted for width (Figure 

 2) was 0. 15 mm. ; that for area 0. 6 sq. mm. ; and for ratios 4 per cent of 

 length. The ordinates show the number of individuals falling within 

 each class. None of the resulting " curves " shows any signs of two or 

 more maxima, except that of Figure 2, where the indication is too 

 slight to be of significance, so that there is no evidence from these 

 measurements that there are well marked classes. When a given 

 colony was plotted by itself, the results were substantially like those 

 here recorded for the combination of several colonies, and each curve 

 was therefore similar. Consequently it seems to me that no objection 

 can be raised to combining the results of several colonies, as I have 

 here done. 



I have also tried to find evidence of distinct types or classes based on 

 other grounds, such as the number of teeth on the mandibles and the 

 ratio of the width between the eyes to the width between the insertions 

 of the antennae, but with equally small success. 



I next turned my attention to Camponotus herculeanus pictus. 

 Here, too, although I did not make such measurements as are described 

 above for C. amerkanus, there seems to be a complete, graded series. 

 This fact is easily seen from the Plate, which was made from a series of 

 photographs taken with the aid of a microscope. All exhibit the 

 animal in the same position and under the same magnification, — about 

 9 diameters, — so that they represent accurately variations of size and 

 shape. But not relying on this single view of the head, since all the 

 parts were not visible in that position, studies were made with the mi- 

 croscope from different sides. When necessary in comparing the size 

 of the various organs measurements were made by means of the ocular 

 micrometer. 



1. The most striking difference (Plate, Figures 2-20) between the 

 large and the small worker is that of size, the heads of the large 

 workers being not only actually but also relatively larger than those of 

 the smaller ones. 



2. Not only is the head of the larger workers larger, but it is also of a 

 somewhat different shape. Thus, while the head of the small worker 

 is somewhat longer than broad, the head of the large worker is more 

 nearly equal in the two dimensions, or even a little broader than long 

 in the largest specimens, thus resembling Camponotus americanus. 

 Moreover, the posterior margin of the small heads is convex backward, 

 whereas that of the large workers is more nearly straight, or even 

 slightly concave backward. It might be more nearly correct to say of the 

 latter that it presents a backward convexity of the two ends of the pos- 



