BUCKINGHAM. — DIVISION OF LABOR AMONG ANTS. 431 



series from the smallest to the largest workers (see also Emery, '96, 

 p. 397). 



Occasionally the fertile winged queens (Emery, '96, p. 399) are of 

 different sizes, as, for example, in Camponotus abdom'malis F. and C. 

 dorylus subsp. confuscus Emery ; and the differences here are parallel 

 to those of the workers. Again (Wheeler and McClendon, : 03 ; Wheeler, 

 :03^), some species may have two distinct forms of queen without in- 

 termediates, and Wheeler (:03^ p. 650) is inclined to believe this due 

 to mutation. Moreover, intermediates, though usually abnormal, are 

 found between the queen and the largest workers (see also Forel, :04% 

 p. 574; Wasmann, '90, pp. 301 et seq.; '95; :02 ; :09% pp. 46, 52, 60, 

 etseq. ; Viehmeyer, :04; Muckermann, :04; and Wheeler, :01''; :10, 

 pp. 406-408). Hence, taking ants as a whole, there is evidence of a 

 long series of gradations from the smallest workers to the largest 

 queens. 



When there are modifications of workers of the same species, they 

 are (Emery, '96, p. 398), first, modifications of size with but slight 

 differences of form of body and mandibles ; but on approaching the 

 maximum size the form of the mandibles changes rapidly, though by a 

 series of gradations. Individuals showing such gradations are difficult 

 to get together because of the scarcity of soldiers, and especially of 

 forms intermediate between them and the worker-major. There are 

 also differences of form and sculpture of the head and other parts of 

 the body. According to Forel ('95, p. 143), the large worker may be 

 distinguished by its enormous increase in size or by the peculiar form 

 of its head or its mandibles, which are suited to breaking seeds, ob- 

 structing the nest opening, or fighting, etc. If the head has no such 

 adaptation, the ant shows no peculiarities. 



Wasmann ('90, p. 300) says that with differences of size are correlated 

 differences of sculpture of separate parts of the body, but especially of 

 the back and the head. Emery (:0P; ••01^ p. 54, foot-note, and :04, 

 pp. 588-589) states that the smallest forms of driver ants, as well as 

 of other species of Dorylus, differ in the structure of the head and its 

 appendages, even in the number of antennal joints, from the medium- 

 sized and small ants, and that these differ from the largest specimens 

 in having smaller heads and toothed mandibles. 



It is an interesting fact, as Emery ('94, p. 55) points out, that in the 

 Ponerinae, which are considered morphologically as the stem form of 

 ants, there is no striking polymorphism of the worker classes (Emery 

 knows it only in the case of Melissotarsus). But polymorphism of the 

 worker does appear in many genera of all the other groups, and it 

 therefore seems that polymorphism among ants has originated polyphy- 



