Notes on the biology of Pheidole lamia (Hymenoptera: Formicidae) at its type locality (Austin, Texas).

Abstract

The rediscovery of Pheidole lamia at its type lo cality is reported, and several aspects of its biology are clarified: 1) P. lamia is not confined to undisturbed woodlands, but is also found in recently disturbed wooded habitat. 2) P. lamia is not rare in Austin, although its inconspicuous nesting sites and foraging habits make it appear so. 3) It is suggested that the size of Austin colonies is probably near 500-1000 minor workers, the same order of magnitude as colonies found in Florida. In 1901, and more formally in 1908, William Morton Wheeler described Pheidole lamia from a small number of specimens collected in Austin, Tex as. The major workers of this species are strikingly different from the major workers of all other North American members of Pheidole. The heads of P. lamia major workers are elongate, subcylindrical in cross section and obliquely truncated in front, closely resembling the heads of Camponotus (Colobopsis) major workers. Wheeler believed colonies of P. lamia to be small, hypogaeic and, at least in the Austin area, very patchily distributed. Furthermore, he suggested that the major workers may serve a dual function. First, like Camponotus (Colobopsis) major workers, they may use their phragmotic heads to block the entrances to nest galleries. Second, due to the large size of their gasters and their small numbers (1-4) in a colony, the major workers may also function as the queen caste. Until recently the validity of Wheeler's statements on P. lamia remained unchecked. P. lamia was not collected again until 1931 in Fayette, Missis sippi (Smith, 1931) and for only a third time in 1954 in Decatur County, Georgia (Gregg, 1956). Creighton repeatedly searched for P. lamia in the Austin area without success (Gregg, 1956). These geographically disjunct collecting records led Gregg (1956) to suggest that P. lamia is extremely rare and patchily distributed throughout the southeastern United States. More recently, Buren et al. (1977) collected numerous colonies of this species in Leon County, Florida. Unlike the collections reported by Wheeler (1901, 1908), Smith (1931) and Gregg (1956), these colonies were not small 1 A contribution from The University of Texas Brackenridge Field Laboratory, Austin, Texas. 2 Present address, Department of Zoology, University of Iowa, Iowa City, Iowa 52242. Received for publication 4 June 1980. This content downloaded from 207.46.13.167 on Thu, 21 Jul 2016 06:08:28 UTC All use subject to http://about.jstor.org/terms 270 JOURNAL OF THE KANSAS ENTOMOLOGICAL SOCIETY and contained over 200 major workers and 1000 minor workers. Further more, observations by Buren et al. (1977) suggested that the heads of P. lamia major workers are not phragmotic, but, instead, are specialized for subterranean warfare against marauding thief ants, Solenopsis (Diplorhop trum) spp. Thus Wheeler appears to be mistaken on most of his conjectures regarding P. lamia. Buren et al. (1977), following Gregg (1956), suggested that Austin, Texas is a geographical area for P. lamia, and that this fact may account for its apparent rarity and small colony size at the type locality. In this paper I report the rediscovery of Pheidole lamia at its type locality and present evidence that suggests this species is more common than previously thought and may maintain large colonies in this geographical fringe area. Study Area and Methods I made all observations and collections at The University of Texas Brack enridge Field Laboratory, Austin, Texas. This area, adjacent to the Colo rado River, is a 32 hectare tract of semi-deciduous woodland in various stages of secondary succession. Residential structures stood on some of this land as late as 1960. The woodland is chiefly composed of oak (Quercus fusiformis and Q. texana), hackberry (Celtis reticulata and C. laevigata), cedar elm (Ulmus crassifolia) and post cedar (Juniperus ashei). In general the litter in this woodland is largely composed of U. crassifolia and Q. fusiformis leaves. To determine the composition of the litter-foraging ant fauna I established two 100 m transects and four 10 m x 10 m quadrats in 5 separate areas of the woodland. Transect 1 passed through Quadrat 2. All of these areas were in the same stage of secondary succession. Both transects contained 10 sites 10 m apart with 5 sampling stations per site. The 5 sampling stations were at the corners and center of squares 1.5 m on a side. Each of the 10 m x 10 m quadrats contained 81 sampling stations 1 m apart. In August 1975 I collected one 25 cm x 25 cm litter sample at each sampling station along both transects and extracted the arthropods using Berlese funnels. All litter was then returned to the appropriate stations. In 1976 and 1977, from mid April until late October, I monitored spatial distributions and activity pat terns of ant species along these transects by tunafish baiting at the same stations where litter samples were collected in 1975. In summer 1979 I mon itored diel activity patterns by baiting for 24 hour periods in each of the 4 quadrats. I always placed tuna baits on the surface of the litter.