2o6 



PSYCHE 



[Oct.— Dec. 



while remaining very quiet as if absorbed in the dehght of feeding. 'Fhe mandi- 

 bles seemed to be too feeble to cut or puncture even so thin a chitinous invest- 

 ment as that of the termites. 



The antennae are undoubtedly the most important and remarkable organs of 

 Cerapachys. This is shown by their great thickness (whence the name of the genus), 

 the differentiation of the glandiform terminal joint of the fiagellum, and their sin- 

 gular freedom of movement. This freedom is permitted by the clypeus which is 

 much compressed laterally m the form of a vertical crest, and does not overlap the 

 scapes on the sides. The antennae are, in fact, kept in continual vibration in a 

 plane perpendicular to the head with their elbows uppermost and not directed side- 

 wise as in other ants, while the glandiform joint of the flagellum plays over the 

 surfaces with which it comes in contact. The insects must be guided almost exclu- 

 sively by the contact-odor sense in these organs since they have no trace of eyes. 

 Pronounced negative heliotaxis, evidently depending on some photodermatic sense, 

 was apparent when the ants were exposed to the sun, though they did not. respond 

 very readily to the lower intensities of diffuse day-light. Strangely enough, this neg- 

 ative heliotaxis was not associated with a high degree 

 of positive thigmotaxis, as it is in many other insects, 

 since the animals showed no decided tendency to con- 

 ceal themselves with their bodies applied to the earth 

 or glass. Often the whole colony would lie exposed 

 for hours on the surface of the soil, merely agglom- 

 erated in a mass about their eggs. 



As the number of eggs visibly increased during 

 the confinement of the colony, it is clear that some of 

 these must have been laid by the workers, as there was 

 no female in the nest. No ants could be more careful 

 of their eggs. They enveloped them with their bodies 

 so that the packet could not be seen except by disturb- 

 ing the whole colony. When the packet was broken 

 apart the eggs were eagerly sought and deftly brought 

 together again. This brooding over the eggs is quite 

 unlike anything I have seen in other ants except Eciton, 

 the smaller species of which {^E. opacithorax, schmitti^ 

 si/mic/irasti, etc.) have a very similar habit. Besides 

 affording protection to the eggs, it may, perhaps, serve 

 to hasten the embryonic development. The eggs are very slender (Figure, a), 

 being fully four times as long as broad. They are not kept in several packets 

 with the long axis of the component eggs parallel with one another, as in some 



