256 
Psyche 
[Vol. 88 
From recent investigations we know that the pygidial gland is 
quite common in the Myrmicinae (Kugler 1978; Holldobler and 
Engel 1978). In at least two species its secretions serve as an alarm 
pheromone (Holldobler et al. 1976; Kugler 1979). On the other 
hand, in several ponerine species the pygidial gland secretions have 
been demonstrated to function as a recruitment pheromone during 
tandem running (Holldobler and Traniello 1980a) or trail communi- 
cation (Maschwitz and Schonegge 1977; Holldobler and Traniello 
1980b). From our findings in O. versicolor it appears now that this 
ponerine trait might have been preserved in the Dacetini, whose 
origin presumably dates back to early Tertiary times (Brown and 
Wilson 1959). If this speculation is right, we should expect that the 
most primitive dacetine species, Daceton armigerum (Brown and 
Wilson 1959; Wilson 1962), has a well developed pygidial gland, 
resembling closely that found in many ponerine ants, and its 
secretions presumably serve as an alarm-recruitment pheromone. In 
fact, Wilson (1962) observed that workers of D. armigerum often 
moved to areas of excitement and when a worker just had dis- 
covered prey it moved in “excited broken running patterns” by 
which other ants in the vicinity might be attracted. Wilson (1962, 
1971) hypothesized that this running pattern might serve as a 
communicative signal of the kind of “Stager’s kinopsis”, i.e. the 
large-eyed Daceton workers might respond to the visual stimuli 
produced by the excitedly moving nestmate. We have now to 
consider the possibility that a Daceton huntress which pursues a 
prey, discharges a short-range recruitment pheromone from the 
pygidial gland, and that consequently the attraction of other 
huntresses in the close vicinity is caused by this chemical signal. 
Acknowledgments 
I would like to thank E. O. Wilson for reading and commenting 
on the manuscript. Hiltrud Engel and E. Seling (SEM-Lab of the 
MCZ, Harvard) provided valuable technical assistance. Many 
thanks to R. W. Taylor and the Division of Entomology of CSIRO, 
Canberra (Australia) for their support and hospitality. R. W. Taylor 
identified the ants; voucher specimens were deposited in the Austral- 
ian National Insect Collection, Canberra. This work was supported 
by grants of the National Geographic Society, National Science 
