Journal of Natural History ,?%. 
Vol. 42, Nos. 27-28, July 2008, 1841-1847 %) S??* 
Biology of a nocturnal bee, Megalopta atra (Hymenoptera: Halictidae; 
Augochlorini), from the Panamanian highlands 
Simon M. Tierneya*, Therany Gonzales-Ojedab and William T. Wcisloa 
"Smithsonian Tropical Research Institute, Balboa, Ancon, Panama;   Facultad de Ciencias 
Forestales y Medio Ambiente, Universidad Nacional de San Antonio Abad del Cusco - Sede 
Puerto Maldonado Jr. San Martin 451, Puerto Maldonado, Madre de Dios, Peru 
(Received 6 June 2007; final version received 6 April 2008) 
Bees of the genus Megalopta have gained attention as a result of their social 
nesting and nocturnal foraging. Seventeen nests of Meglaopta atra from the 
highlands of Chiriqui Province, Panama, were collected at the end of the dry 
season when brood rearing is expected to be at its peak. Most nests contained 
single females; within multifemale nests only one female possessed enlarged 
ovarioles, although some non-reproductive individuals were inseminated. In two 
of these nests reproductive individuals were clearly larger in body size than 
nestmates, but body size variation and macrocephaly were equivalent to those 
found in other Neotropical augochlorines. There was no evidence of a non- 
reproductive worker-like caste and multifemale nests did not appear to be more 
productive than solitary nests, which may represent pre-reproductive assem- 
blages. Megalopta atra appears to be isolated by altitude from co-geners common 
in Panama, this is discussed in comparison with temperate halictine bees, in which 
environmental dines separate solitary from social populations. 
Keywords: Augochlorini; Megalopta; nesting biology; nocturnal bees; social 
evolution 
Introduction 
The halictine sweat bees are viewed as one of the most useful groups for 
investigations into the origins of social behaviour, largely because of the plasticity 
in degrees of social organization enabled by facultative caste systems, but also 
because of the numbers and diversity of species and the range of habitats occupied 
by these bees (Michener 1974, 1990; Danforth and Eickwort 1997; Wcislo 1997; 
Yanega 1997). Extrinsic factors are increasingly thought to play a role in shaping the 
kinds of social behaviour expressed, and the trend for temperate social taxa to 
largely revert to solitary living at higher altitudes and latitudes is often cited in 
support of this claim (e.g., Sakagami and Munukata 1972; Packer 1990; Eickwort et 
al. 1996; Soucy 2002; but see Soucy and Danforth 2002). 
Transitions of species into novel environments are therefore of interest in 
understanding behavioural transitions (e.g., Wcislo 1989; West-Eberhard 2003). One 
other such example is movement into a nocturnal environment, and recently the 
augochlorine genus Megalopta has attracted the attention of biologists for exactly 
this reason (Arneson and Wcislo 2003; Smith et al. 2003, 2007, 2008; Wcislo et al. 
2004; Kelber et al. 2006; Wcislo and Gonzalez 2006). Very little is known of the 
^Corresponding author. Email: tierneys@si.edu 
ISSN 0022-2933 print/ISSN 1464-5262 online 
? 2008 Taylor & Francis 
DOI: 10.1080/00222930802109124 
http://www.informaworld.com 
1842    S.M. Tierney et al. 
biology of the stem-nesting Megalopta outside of Barro Colorado Island (BCI), 
Panama. Here we detail the nesting biology of Megalopta atra Engel from the 
western Panamanian highlands, as part of comparative biological and phylogenetic 
studies of the genus. 
Material and methods 
Fieldwork was conducted at Reserva Forestal Fortuna, situated in western Panama 
at the border of Chiriqui and Bocas del Toro Provinces. The area is a mountainous 
region formed by the Serrania del Tabasara and the Cordillera de Talamanca that 
continues west into Costa Rica (see Myers 1969). These east-west ridges form the 
continental divide with unbroken highlands up to the ~1200-m contour in Panama, 
and isolated peaks up to 3475 m (Volcan Baru). The mountain range slopes north to 
the Atlantic lowlands, and south through foothills to the Pacific Ocean. The 
highlands contain montane rain forest or cloud forest, except on the higher peaks, 
with premontane moist or wet forest at lower elevations, though extensive areas have 
been cleared for cattle ranching and farming. Temperatures are cool and equable. A 
dry season usually lasts from approximately February to April, although rain falls in 
all months, with about 4000-5000 mm of rain per year, and the intensity of the dry 
season varies greatly from year to year (see www.stri.org/english/research/facilities/ 
terrestrial/fortuna/research_info.php). Low clouds and light mist are common. 
Research was based out of the Fortuna field station of the Smithsonian Tropical 
Research Institute (STRI; 8?43'N, 82? 14'W; ~ 1235m elevation). Collections 
were made towards the end of the dry season - 24 April to 1 May 2007. Nests were 
collected and sealed in the field, and were later opened in the laboratory. Adults were 
preserved in absolute ethanol and brood cells were removed intact and housed in cell 
culture trays to rear as many adults as possible. Flying bees were collected using 
mercury vapour light traps situated at the STRI field station and at three locations of 
varying altitude down the Caribbean slope of the divide (~1160m, 1010m and 
664 m). 
Reproductive activity of adult females was inferred by dissection following 
Schwarz (1986). Ovarian development was measured as the sum length of the three 
largest terminal oocytes and mating status was identified by the presence or absence 
of sperm in the spermatheca. Body size was determined by intertegular distance, 
following Tierney et al. (2008). The number of nicks and tears along the distal 
margin of this wing was taken as a measure of wing wear. Voucher specimens are 
deposited in the Dry Reference Collection of the Smithsonian Tropical Research 
Institute, Balboa, Ancon, Panama. 
Results 
Seventeen M. atra nests were collected (Table 1). Nest and brood cell architecture did 
not differ from that described for other Panamanian species of Megalopta (Wcislo et 
al. 2004). Nest lengths ranged from 47 to 372 mm and nest tunnel width varied from 
8 to 14 mm. While there was no evidence to suggest that nests with multiple females 
were longer than the nests of single females, the largest nest in this study, containing 
eight females (nest 13), did have a larger tunnel width and was in a stem of much 
broader diameter that contained two parallel sets of tunnels. 
Journal of Natural History    1843 
Table 1. Megalopta atra nest census data. 
Nest number Nest length Tunnel n adult n sealed n cells being n total 
(mm) width 
(mm) 
females cells 
(brood) 
provisioned 
with pollen 
brood cells 
1 47 1 2 1 3 
2 271 11 1 0 1 1 
3 372 9 1 2 3 
4 96 8 0 3 3 
6* 275 11 2 0 1 11 
7 68 9 1 2 2 
8 - - 1 2 1 3 
9 237 9 1 5 5 
10 190 11 1 dead 0 3 
11 132 12 0 1 6 
12* 135 10 2 1 3 
13* 242 14 8 7 8 
14 119 8 1 0 1 1 
15 195 11 1 1 1 3 
16 156 11 1 0 2 
17 230 9 1 3 1 4 
18 223 9 0 0 1 
Mean 186.8 10.1 1.4 1.7 0.4 3.6 
+ SE 21.5 0.4 0.4 0.5 0.1 0.6 
Data show physical nest properties and the numbers of adults and brood. 
*Multifemale nest. 
Three nests were abandoned but contained brood cells. Eleven nests contained 
single females (one deceased) and three were multifemale; two nests contained two 
females each and one nest had eight females. All nests contained brood cells: the 
nests of single females contained one to five cells and the nests with multiple females 
contained three to eight cells. A better indication of colony productivity at the time 
of nest collection is the number of cells sealed or being provisioned: for singletons 
this ranged from zero to five cells and for multifemale nests it ranged from one to 
seven cells. Based on relatively small sample sizes, it does not appear that group 
nesting results in increased productivity on a per capita basis. 
There is considerable intraspecific variation in female body size (range of 
intertegular width=2.56-3.76mm), including enlarged processes of the head and 
genal spines, as known for some other Neotropical, social augochlorine bees 
(reviewed in Sakagami and Moure 1965; Tierney et al. 2008), and large body size is 
thought to be associated with social dominance (Arneson and Wcislo 2003; Smith 
et al. 2008). An ovarian index was measured (sum of three largest oocytes/ 
intertegular width) to account for the scaling effects of body size on ovary size. In 
multifemale nests (Table 2) only one female possessed enlarged ovarioles and in nests 
12 and 13 these individuals also displayed more wing wear than their nestmates. In 
nests 13 (eight females) and 6 (two females) the reproductive individuals were 
macrocephalic (per Sakagami and Moure 1965), but in nest 12 both females were of 
similar body size, the slightly smaller one being the reproductive individual. These 
1844    S.M. Tierney et al. 
Table 2. Reproductive status within multifemale nests of Megalopta atra. 
Nest Ovarian index Intertegular width 
(mm) 
Inseminated Wing wear 
6 2.94 3.48 + 0 
026 2.88 - 0 
12 2.44 2.72 + 4 
0.42 2.88 + 0 
13 2.61 3.76 + 7 
0.66 308 + 0 
0.51 3.00 - 0 
049 3.04 + 1 
0.47 246 - 0 
0.41 2.56 + 0 
0.38 3.04 + 0 
033 2.68 - 0 
Single female 1.52 2.99 8+, 2- 1.5 
(mean values H=10) 
Individuals listed in descending order of ovarian index, and the reproductive female in each 
nest is indicated in bold type. 
nests showed no direct evidence of worker roles in non-reproductive bees and it may 
be that these colonies are only pre-reproductive assemblages, with some females 
possibly waiting to inherit the maternal nest or disperse. This interpretation is 
supported by the almost complete lack of wing wear in non-reproductive individuals 
(Table 2). A worker caste should be characterized by reduced reproductivity and 
evident wing wear. 
Twenty-two M. atra (21 females, one male) were collected at light traps between 
05.15 and 06.13 h at the STRI station (?1235 m) on the continental divide and at the 
Caribbean slope site a (?1160m). Traps set at slope sites b and c (~1010m and 
664 m, respectively) attracted the two other species of nocturnal Megalopta 
commonly found in Panama (M. genalis and M. ecuadoria - see Moure and Hurd 
1987; Wcislo et al. 2004), but these lower elevation traps did not attract M. atra, 
which has been collected only at higher elevations (see Engel 2006). 
Discussion 
Environmental conditions may lessen the probability for eusociality to occur, as the 
result of a shorter foraging season, constraining life cycles such that there is 
insufficient time for adults reared in the first brood to rear through a second brood 
of potential reproductives. Studies comparing populations of temperate halictine 
bees have shown that populations at higher altitudes (e.g., 2850 m, Eickwort et al. 
1996) tend to be solitary, while populations at lower elevations are social (Sakagami 
and Munukata 1972; Soucy 2002). The same preclusion of sociality occurs in a 
halictine bee with regard to higher latitudes (Packer 1990), as expected based on 
biogeographic parallels between increasing altitude and increasing latitude in 
temperate zones (Merriam 1895). However, a subsequent phylogeographic study 
Journal of Natural History    1845 
utilizing mitochondria! sequence data (Soucy and Danforth 2002) proposed that 
solitary and social populations of North American Halictus rubicundus are 
genetically distinct lineages, and that this may be evidence for some degree of 
genetic determination of social behaviour. 
Geographic seasonal restrictions mentioned for these halictines are for temperate 
taxa. A tropical social spider Anelosimus eximius shows an increase in the proportion 
of solitary webs at altitudes between 700 m and 1200 m (Purcell and Aviles 2007), 
although social colonies are not excluded at these upper altitudes. In other tropical 
halictines, social species (e.g., Lasioglossum aeneiventre) and species that are largely 
solitary (e.g., Lasioglossum figueresi) occur at similar altitudes > 1200 m (Wcislo et 
al. 1993). Given that temperatures are equable year-round at midlevel altitudes in the 
tropics, seasonal constraints associated with temperature may be less important than 
phenological flowering patterns, which are lacking for the Fortuna region. 
Temperature during brood rearing has also been proposed to account for the 
variation in body size among temperate halictines, both positively (Yanega 1989; 
Richards and Packer 1996; Soucy 2002) and negatively (Kamm 1974; Plateaux- 
Quenu 1993). Yanega (1989) and Soucy (2002) suggest that the actual length of the 
brood rearing season, and so greater variations in temperature, may be more 
important in defining this variation in size within a population. However, this 
reasoning cannot account for the highly variable cephalic and body sizes found 
among Neotropical augochlorines, where seasonal temperatures are less variable and 
polymorphisms are more extreme (Sakagami and Moure 1965; Tierney et al. 2008). 
For Megalopta, body size appears to be an important factor determining social 
reproductive status (Arneson and Wcislo 2003), although whether size differences 
are environmentally determined by floral availability, or by the reduction of pollen 
provisions by brood cell parasites (Smith et al. 2008), requires additional data and 
broader assessment. 
Megalopta atra was not found at altitudes lower than ~ 1160m and other species 
of Megalopta common to Panama were found up to ?1010 m. We are unaware of 
the upper altitudinal range of M. atra, but the species extends at least into Costa 
Rica (Engel 2006). Although this study does not involve an intraspecific comparison 
of populations at different altitudes, M. atra does appear to be isolated by altitude, 
and therefore provides a valuable comparison with lowland congeners, M. genalis 
and M. ecuadoria (e.g., Smith et al. 2003, 2007; Wcislo et al. 2004), and a 
comparative phylogeographic study of these species should be informative (cf. Soucy 
and Danforth 2002). These lowland studies detail much higher rates (percentages) of 
multifemale social nesting and it may be that the largest colony of M. atra described 
here is only a reflection of the larger size of the nesting substrate. It may also be 
presumed that the cooler mean temperatures in Fortuna would both shorten 
foraging times and slow the developmental time of immature M. atra (approximate 
average minimum temperatures: BCI, 22-25?C; Fortuna, 14-17?C). However, initial 
predawn flight times (both the coldest period of the day and the major foraging 
period) recorded for M. atra do not differ greatly from those recorded for other 
Megalopta species on BCI (?05.00-05.lOh) based on activity at nests (Wcislo et al. 
2004; Kelber et al. 2006), and this suggests that M, atra bees do not leave the nest in 
the morning until astronomical twilight begins. Our data on foraging times were 
inferred from captures at light traps. Flight times of M. genalis and M. ecuadoria on 
BCI  that  were  inferred  from  light-trap  data  (Roulston   1997)  approximately 
1846    S.M. Tierney et al. 
correspond to patterns of observed foraging behaviour from nests (Wcislo et al. 
2004; Kelber et al. 2006), suggesting that our data provide a reasonable estimate of 
foraging times. Whether climate in this context effectively reduces social nesting of 
M. Mr a in this highland habitat remains equivocal. Longer-term studies to assess 
seasonal variation are required to test these assumptions. 
Acknowledgements 
We would like to thank STRI staff Orelis Arosemana, Carlos Espinosa, Marcela Paz and Don 
Windsor for logistical support, and Emmet Gowin for practical assistance with light traps. We 
are indebted to Charles Michener and Adam Smith for constructive advice on earlier versions 
of this manuscript. This work was supported by an Earl S. Tupper Postdoctoral Fellowship 
from the Smithsonian Tropical Research Institute to S.M.T. and by a National Geographic 
Society Research and Exploration Grant awarded to W.T.W. and S.M.T. Collections were 
made under ANAM Permit no. DAPVS-01-2007. 
References 
Arneson L, Wcislo WT. 2003. Dominant-subordinate relationships in a facultatively social, 
nocturnal bee, Megalopta genalis (Hymenoptera: Halictidae). J Kansas Entomol Soc. 
76:183-193. 
Danforth BN, Eickwort GC. 1997. Evolution of social behaviour in the augochlorine sweat 
bees (Hymenoptera: Halictidae) based on a phylogenetic analysis of the genera. In: Choe 
JC, Crespi BJ, editors. The evolution of social behaviour in insects and arachnids. 
Cambridge (UK): Cambridge University Press, p. 270-292. 
Eickwort GC, Eickwort JM, Gordon J, Eickwort MA. 1996. Solitary behaviour in a high 
altitude   population   of  the   social   sweat   bee   Halictus   rubicundus   (Hymenoptera: 
Halictidae). Behav Ecol Sociobiol. 38:227-233. 
Engel MS. 2006. A new nocturnal bee of the genus Megalopta, with notes on other Central 
American species. Mitt Int Entom Vereins. 31:37-49. 
Kamm DR. 1974. Effects of temperature, day length and number of adults on the sizes of cells 
and offspring in a primitively social bee (Hymenoptera: Halictidae). J Kansas Entomol 
Soc. 47:8-18. 
Kelber A, Warrant EJ, Pfaff M, Wallen R, Theobald JC, Wcislo WT, Raguso RA. 2006. Light 
intensity limits foraging activity in nocturnal and crepuscular bees. Behav Ecol. 17:63-72. 
Merriam CH. 1895. The geographic distribution of animals and plants in North America. 
Yearbook US Dept Agriculture. 1894(1895):203-214. 
Michener CD. 1974. The social behaviour of the bees. Cambridge (MA): Harvard University 
Press, 404 p. 
Michener CD. 1990. Reproduction and castes in social halictine bees. In: Engels W, editor. 
Social insects, an evolutionary approach to castes and reproduction. Berlin: Springer 
Verlag. p. 77-121. 
Moure JS, Hurd PD. 1987. An annotated catalogue of the Halictid bees of the western 
hemisphere. Washington, DC: Smithsonian Institution Press. 405 p. 
Myers CW. 1969. The ecological geography of cloud forest in Panama. Am Mus Novit. 
2396:1-52. 
Packer  L.   1990.   Solitary  and eusocial nests  in  a population  of Augochlorella striata 
(Provancher) (Hymenopter, Halictidae) at the northern edge of its range. Behav Ecol 
Sociobiol. 27:339-344. 
Plateaux-Quenu C. 1993. Modalites de la socialisation chez les Halictinae (Hymenoptera, 
Halictidae) I. Biologic des Halictinae. Ann Biol. 32:183-204. 
Journal of Natural History    1847 
Purcell J, Aviles L. 2007. Smaller colonies and more solitary living mark higher elevation 
populations of a social spider. J Anim Ecol. 76:590-597. 
Richards MH, Packer L. 1996. The socioecology of body size variation in the primitively 
eusocial sweat bee, Halictus ligatus (Hymenoptera: Halictidae). Oikos. 77:68-76. 
Roulston TH. 1997. Hourly capture of two species of Megalopta at black lights in Panama 
with notes on nocturnal foraging by bees. J Kansas Entomol Soc. 70:189-196. 
Sakagami SF, Moure JS. 1965. Cephalic polymorphism in some neotropical halictine bees. An 
Acad Brasil Ciencias. 37:303-313. 
Sakagami SF, Munukata M. 1972. Distribution and bionomics of a transpalaearctic eusocial 
halictine bee, Lasioglossum (Evylaeus) calceatum, in northern Japan, with reference to its 
solitary life cycle at high altitude. J Fac Sci, Hokkaido Univ Ser VI, Zool. 18:411^439. 
Schwarz MP. 1986. Persistent multi-female nests in an Australian allodapine bee, Exoneura 
bicolor (Hymenoptera, Anthophoridae). Insectes Soc. 33:258-277. 
Smith AR, Wcislo WT, O'Donnell S. 2003. Assured fitness returns favor sociality in a mass- 
provisioning sweat bee, Megalopta genalis. Behav Ecol Sociobiol. 54:14-21. 
Smith AR, Wcislo WT, O'Donnell S. 2007. Survival and productivity in the sweat bee, 
Megalopta genalis (Hymenoptera: Halictidae). Behav Ecol Sociobiol. 61:1111-1120. 
Smith AR,  Wcislo  WT,  O'Donnell  S.  2008.  Body  size  shapes  caste expression,  and 
cleptoparasitism  reduces   body   size   in   the   facultatively   eusocial   bees   Megalopta 
(Halictidae). J. Insect. Behav. DOI 10.1007/sl0905-008-9136-l. 
Soucy SL. 2002. Nesting biology and socially polymorphic behaviour of the sweat bee Halictus 
rubicundus (Hymenoptera: Halictidae). Ann Entomol Soc Am. 95:57-65. 
Soucy SL, Danforth BN. 2002. Phylogeography of the socially polymorphic sweat bee 
Halictus rubicundus (Hymenoptera: Halictidae). Evolution. 56:330-341. 
Tierney SM, Gonzales Ojeda T, Wcislo WT. 2008. Nesting biology and social behaviour of 
two Xenochlora bees (Hymenoptera: Halictidae: Augochlorini) from Peru. J Kansas 
Entomol Soc. 81:61-72. 
Wcislo WT.  1989. Behavioural environments and evolutionary change. Annu Rev Ecol 
System. 20:137-169. 
Wcislo WT.   1997.  behavioural environments of sweat bees (Halictinae) in relation to 
variability. In: Choe JC, Crespi BJ, editors. The evolution of social behaviour in insects 
and arachnids. Cambridge (UK): Cambridge University Press, p. 316-332. 
Wcislo WT, Wille A, Orozco E. 1993. Nesting biology of tropical solitary and social sweat 
bees,  Lasioglossum   (Dialictus) figueresi Wcislo   and  L.   (D.)   aeneiventre  (Friese) 
(Hymenoptera: Halictidae). Insectes Sociaux. 40:21^10. 
Wcislo WT, Arneson L, Roesch K, Gonzalez VH, Smith AR, Fernandez-Marin H. 2004. The 
evolution of nocturnal behaviour in sweat bees, Megalopta genalis and M. ecuadoria: an 
escapefrom competitors and enemies? Biol J Linn Soc. 83:377-387. 
Wcislo WT, Gonzalez VH. 2006. Social and ecological contexts of trophallaxis in facultatively 
social sweat bees, Megalopta genalis and M.  ecuadoria (Hymenoptera: Halictidae). 
Insectes Soc. 53:220-225. 
West-Eberhard MJ.  2003.  Developmental plasticity and evolution.  New York:  Oxford 
University Press. 794 p. 
Yanega D. 1989. Caste determination and differential diapause within the first brood of 
Halictus rubicundus in New York (Hymenoptera: Halictidae). Behav Ecol Sociobiol. 
24:97-107. 
Yanega D. 1997. Demography and sociality in halictine bees (Hymenoptera: Halictidae). In: 
Choe JC, Crespi BJ, editors. The evolution of social behaviour in insects and arachnids. 
Cambridge (UK): Cambridge University Press, p. 293-315.