438 PROCEEDINGS OF THE AMERICAN ACADEMY. 



However, in nearly all the differences among the workers of this 

 species, it is necessary to observe that there is no break in the series, 

 no tendency to form separate classes. There is, rather, a continuous 

 variation from, one size to the next, forming, as Wheeler (:07^ p. 77) 

 says, a series of intermediates between the very large and the very 

 small. 



Although one would not expect the duties of classes in this species 

 to be as distinct from one another as they are in those polymorphic 

 species where there are sharp morphological differences, still it seemed 

 possible that polymorphism might be associated with some recognizable 

 division of labor. Furthermore, as already stated, it was thought well 

 to compare the activities of two forms, in one of which (as in Campono- 

 tus americcmus and C. pictus) there is a graded series ; and in the other 

 (as in certain species of Pheidole) there are dimorphic classes without 

 intermediate forms between the large-headed soldiers and the workers 

 (Emery, :02, p. 719 ; Wheeler, :10, p. 559). 



2. Pheidole. — In Pheidole there is, as a rule, no intermediate form 

 between the small workers of ordinary proportions and the soldiers 

 (Emery, :02, p. 719 ; Wheeler, :10, p. 559), though such intermediate 

 individuals are occasionally found. By " soldier " is meant, according 

 to Forel ('95, p. 143), merely a large worker which, through complete 

 dropping out of the intermediate forms and through adaptation to pre- 

 cise functions, has become differentiated from the small worker. It is 

 often so different from the worker proper as to be taken for another 

 species. In some species, as Ponei-a edouardi, there are, indeed, two 

 forms of soldier (Forel, '95, p. 145). In Pheidole, however, there is 

 only one. The differences between the heads of the soldier (Plate, 

 Figure 22) and the worker (Figure 23) of Pheidole piUfera may be 

 described as follows: 



1. A great difference in the size of the head, that of the soldier 

 being much larger, not only actually, in correspondence with its larger 

 body, but even relatively ; it is so heavy and clumsy, that it is held 

 bent downward at a much greater angle than that of the small worker. 



The thickness of the heads measured dorso-ventrally does not differ 

 materially. When the dorsal surface of the head is examined, however, 

 it is found that the proportions of its outline are quite different ; it is 

 nearly square in the small workers, and quite oblong in the soldiers. 

 The posterior margin is only slightly cordate in the small ants ; but in 

 the soldiers it is deeply indented in the centre, where ends a median 

 dorsal groove, which begins about the middle of the head. This groove 

 lies between the muscles of the mandibles, which cause a pair of longi- 



