ISSN 1994-4136 (print) ISSN 1997-3500 (online) Myrmecological News Volume 25 October 2017 Schriftleitung / editors Florian M. STEINER, Herbert ZETTEL & Birgit C. SCHLICK-STEINER Fachredakteure / subject editors Jens DAUBER, Falko P. DRIJFHOUT, Evan ECONOMO, Heike FELDHAAR, Nicholas J. GOTELLI, Heikki O. HELANTERÄ, Daniel J.C. KRONAUER, John S. LAPOLLA, Philip J. LESTER, Timothy A. LINKSVAYER, Alexander S. MIKHEYEV, Ivette PERFECTO, Christian RABELING, Bernhard RONACHER, Helge SCHLÜNS, Chris R. SMITH, Andrew V. SUAREZ Wissenschaftliche Beratung / editorial advisory board Barry BOLTON, Jacobus J. BOOMSMA, Alfred BUSCHINGER, Daniel CHERIX, Jacques H.C. DELABIE, Katsuyuki EGUCHI, Xavier ESPADALER, Bert HÖLLDOBLER, Ajay NARENDRA, Zhanna REZNIKOVA, Michael J. SAMWAYS, Bernhard SEIFERT, Philip S. WARD Eigentümer, Herausgeber, Verleger / publisher © 2017 Österreichische Gesellschaft für Entomofaunistik c/o Naturhistorisches Museum Wien, Burgring 7, 1010 Wien, Österreich (Austria) Myrmecological News 25 17-28 Vienna, October 2017 Larvae of trap-jaw ants, Odontomachus LATREILLE, 1804 (Hymenoptera: Formicidae): morphology and biological notes Eduardo G. P. FOX, Adrian A. SMITH, Joshua C. GIBSON & Daniel R. SOLIS Abstract This study aims to contribute to the neglected topic of larval biology in ants. The number of larval instars for three differ- ent species of trap-jaw ants, Odontomachus meinerti FOREL, 1905, Odontomachus bauri EMERY, 1892, and Odontomachus brunneus (PATTON, 1894) was estimated to three based on the maximum width frequency distributions of head cap- sules from worker and male larvae. The obtained number of larval instars was smaller than from a previous report with another species in the genus, indicating possible interspecific variation. Larvae of different sexes and instars among the three different species were generally identical, differing merely by relative dimensions and patterns of hair distribution. Dorsal "doorknob" protuberances were recorded for the first time in the genus, and observed being used to fix larvae onto nest walls. From observing several individuals, we suggest the ornamentation of spiracle peritremes and the types of body protuberances are useful characters for larval taxonomy within this group. Moreover, a few individuals were found pos- sessing anomalous structures which are reminiscent of characters from related taxa. Finally, some brief observations are made on an unidentified parasite found inside some mature larvae of O. bauri. Key words: Juvenile morphology, post-embryonic development, ecdysis, Odontomachus meinerti, O. bauri, O. brunneus. Myrmecol. News 25: 17-28 (online 8 May 2017) ISSN 1994-4136 (print), ISSN 1997-3500 (online) Received 30 November 2016; revision received 19 March 2017; accepted 21 March 2017 Subject Editor: Chris R. Smith Eduardo Gonçalves P. Fox (contact author), Red Imported Fire Ant Research Center, South China Agricultural Uni- versity, Guangzhou 510642, China. E-mail: [email protected] Adrian A. Smith, Research & Collections, North Carolina Museum of Natural Sciences, Raleigh, NC 27601, USA; De- partment of Biological Sciences, North Carolina State University, Raleigh, NC 27607, USA. Joshua C. Gibson, Department of Entomology, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA. Daniel R. Solis, Centro de Estudos de Insetos Sociais, Sao Paulo State University. 24A, 1515, Bela Vista 13506-900, Rio Claro, SP, Brazil. Introduction Excluding the poles and a few Pacific islands, ants can be ANTWEB 2016). Odontomachus colonies vary in worker found across all terrestrial habitats wherein they typically number from less than a hundred to several hundred, while play central ecological roles at multiple trophic levels (HÖLL- queens can be quite numerous depending on the species DOBLER & WILSON 1990). In the context of their nutri- (e.g., more than 80 reported by ITO & al. 1996). For Odon- tional adaptations and social plasticity, ant larvae usually tomachus, detailed morphological studies on larvae are avail- play a key role inside the colony as "social stomachs" as able only for the following species: Odontomachus haema- they store, interconnect, and passively distribute nutrients todus (LINNAEUS, 1758), Odontomachus simillimus SMITH, to nestmates (HÖLLDOBLER & WILSON 1990). Whilst the 1858 by WHEELER & WHEELER (1952, 1964, 1980), and outstanding diversity and ecological dominance of ants has Odontomachus clarus (ROGER, 1861) by PETRALIA & VIN- attracted a great deal of attention from scientists and the SON (1979). Some additional biological notes on Odonto- general public, the larval phases of ants are typically un- machus larvae are given by COLOMBEL (1971, 1974, 1978) known and unstudied; this is illustrated by the tiny frac- (on larval instars, artificial rearing, and the production of tion (0.4%, ca. 64) of total ant species which have had gynes), WHEELER (1918) (briefly on larval feeding), HART their larval stages described in any detail (see a tentative & TSCHINKEL (2012) (on seasonal brood production), and list in SOLIS & al. 2010a; estimated total number of ant BOTTCHER & OLIVEIRA (2014) (briefly on larval nutrition). species based on ANTWEB 2016). Therefore, the present study is part of a recent effort to Ants in the genus Odontomachus LATREILLE, 1804 (67 expand available information on ant larvae. Larval mor- described species) are relatively large and common ants phology of three species – Odontomachus bauri (EMERY, from the tropics and subtropics that possess elongate spring- 1892), Odontomachus brunneus (PATTON, 1894), and Odon- loaded mandibles used to capture prey or initiate escape tomachus meinerti (FOREL, 1905) – was evaluated using jumps (DEYRUP & al. 1985, LARABEE & SUAREZ 2014, light and scanning electron microscopy. Intraspecific varia- tions are reported from comparing a large number of spe- above listed Florida populations and established in the la- cimens (> 20 for each species), including unprecedented boratory without brood. Worker-laid male larvae were ob- aberrations. Observations of larvae inside laboratory cul- tained from these colonies which were kept under con- tures of live colonies are also presented and discussed, trolled lab conditions of 27 ºC and relative humidity 40 - given the paucity of published information for this genus. 50%, held in plastic Petri dish nest chambers with mois- tened dental plaster substrate. Obtained specimens were Material and methods all stored directly in 70% ethanol. Sample collection: Worker larvae. Nest fragments of Odon- Larval instars: The number of larval instars was pri- tomachus meinerti and O. bauri were obtained from the marily estimated following the principle of Dyar's rule as urban municipality of Rio de Janeiro, Rio de Janeiro, Bra- discussed in PARRA & HADDAD (1989) and adapted by zil, with further specimens of O. bauri collected from a SOLIS & al. (2010a), as follows: "The maximum head widths single nest in the campus of Joint Research Unit Ecology of the larvae were measured […] and plotted in a fre- of Guianan Forests (EcoFoG) in Kourou, French Guiana. quency distribution graph, wherein every distinct peak was Obtained specimens were fixed and stored in 70% ethanol. considered to correspond to a different larval instar; the A total of nine whole queenright colonies of O. brunneus obtained number of larval instars was then tested against were collected from the following Florida, United States Dyar's rule." populations: MacArthur Agro-ecology Research Centre For Odontomachus meinerti workers we had sufficient in Lake Placid, Apalachicola National Forest in Tallahas- samples to allow for bracketing growth limits, as described see, and Pine Jog Environmental Education Centre in West by SOLIS & al. (2010a): "The first larval instar and the Palm Beach. These colonies were queenright and out of last larval instar can be explicitly identified and used as the reproductive phase given the total absence of alates, reference to bracket others. First instar larvae are equiva- either as pupae or adults. Queenright ant colonies out of lent to the mature embryo, which can be measured inside the reproductive phase are expected to produce almost ex- the egg through transparency, and last instar larvae have clusively worker brood (HÖLLDOBLER & WILSON 1990). the developing pupa showing from within (also termed Moreover, according with COLOMBEL (1971, 1978) there "prepupae")." Prepupae can be removed from cocoons to is strong inhibition upon the production of gynes in the confirm measurements from last larval instars. Because of presence of the queen, where "the aptitude of larvae to reduced sample size, the same limits could not be reli- become winged females is totally inhibited (...)" (COLOM- ably established for O. bauri, for which the few embryos BEL 1978). Colombel remarked obtaining gynes was quite obtained were used for light microscopy (i.e., head width hard, reporting ca. 1% success in artificial rearing and a is altered) and the few cocoons carried only fully-formed feeble 15% under ideal conditions with orphaned colo- pupae. For these species all measurements were taken from nies. On top of that, the presence of reproductive brood is fixed larvae as in SOLIS & al. (2010a). deemed obvious as the larva mature, given reproductive For Odontomachus brunneus, given the samples origin- prepupae and pupae acquire a characteristic morphology. ated from > 20 small colonies, the individuals were meas- Therefore, as our small colonies contained only even-sized ured with special care to