Crematogaster rochai Forel , 1903:255. Syntype worker, queen, male: Brazil , Ceara ( Diaz da Rocha ) [ MHNG ] (examined). Emery, 1922:134: combination in C. (Orthocrema) . Forel, 1912:213; Gallardo, 1934:21: race/subspecies of brevispinosa .
Range
Widespread in Neotropics from southern Mexico to Argentina.
Description of worker (Costa Rica)
Differing from crinosa and torosa by the following combination of characters: dorsal and posterior face of propodeum in nearly the same plane, such that the propodeal suture appears very shallow with no posterior wall, the propodeum forming a single declivity from the propodeal suture to the petiolar insertion (very large workers have a short dorsal face that drops to propodeal suture); promesonotum strongly arched, not flattened; anteroventral petiolar tooth long, sharply acute, triangular to spine-like; fourth abdominal tergite completely lacking erect setae.
Measurements
HL 0.851, 0.519, 1.139; HW 0.915, 0.535, 1.238; HC 0.905, 0.509, 1.207; SL 0.559, 0.375, 0.681; EL 0.198, 0.114, 0.257; A11L 0.247; A11W 0.123; A10L 0.110; A10W 0.104; A09L 0.062; A09W 0.078; A08L 0.045; A08W 0.060; WL 0.913, 0.500, 1.259; SPL 0.098, 0.069, 0.158; PTH 0.200, 0.121, 0.265; PTL 0.271, 0.158, 0.407; PTW 0.279, 0.166, 0.343; PPL 0.232, 0.128, 0.286; PPW 0.262, 0.167, 0.352; CI 108, 103, 109; OI 23, 22, 23; SI 66, 72, 60; PTHI 74, 77, 65; PTWI 103, 105, 84; PPI 113, 130, 123; SPI 11, 14, 13; ACI 0.57.
Queen (Costa Rica)
A normal queen (dorsal face of propodeum drops steeply from postscutellum and much of propodeum appears ventral to scutellum and postscutellum, Fig. 1) with general shape, sculpture, and pilosity characters of the worker; size characters as in Figures 4 and 5.
Biology
Crematogaster rochai has a biology very similar to crinosa and torosa . It occurs primarily in open, seasonally dry areas, highly disturbed areas, pasture edges, and beach margins. It occasionally occurs in mangroves, although crinosa is the more common mangrove inhabitant. I have never collected it in rainforest areas.
Nests are large, polydomous, and distributed in a wide variety of plant cavities. Dead branches and knots in living trees are most often used. In Guanacaste Province in Costa Rica colonies may occupy ant acacias and may invade acacias occupied by Pseudomyrmex . I have seen workers distributed in small chambers scattered in the corky bark of Tabebuia trees (Bignoniaceae) and Erythrina trees (Fabaceae). Workers often construct small carton baffles to restrict nest entrances and small carton pavilions that shelter Homoptera on surrounding vegetation.
Foraging is primarily diurnal. Workers are generalized scavengers and they frequently visit extrafloral nectaries. Often columns of workers move between nests.
I often find cockroach egg cases scattered in the nest chambers of C. rochai , at a much higher density than in the environment generally. The nature of the relationship between cockroaches and the crinosa group would be worth investigation.
Comments
This is a member of the crinosa complex and may not always be distinguishable from crinosa and torosa . See under crinosa for further discussion. In Costa Rica rochai always has the fourth abdominal tergite completely devoid of erect setae, and the anteroventral petiolar process is long and sharp. Costa Rican material also lacks a differentiated dorsal face of the propodeum, but material from central and southern South America develops a stronger propodeal suture, thus approaching the condition in other crinosa group material. Also, southern material often has one to five erect setae on the anterolateral portions of the fourth abdominal tergite.
Alto Paraguay, Canindeyú , Central, Concepción , Pte. Hayes (ALWC, IFML, INBP, JTLC, LACM, NHMB). Literature records: “Paraguay” (s. loc.) (Kempf 1972 [as “ malevolens ”]).
Crematogaster rochai Forel 1903: 255. [w syntypes examined, MHNG ; Ceará , Brazil , ( Rocha )]
Crematogaster brevispinosa r. rochai Forel . Forel 1912c: 213.
Crematogaster (Orthocrema) brevispinosa st. malevolens Santschi 1919: 41. NEW SYNONYMY . [w syntypes examined, NHMB ; Nueva Pompeya , Chaco , Argentina ( Joergensen )] .
Crematogaster (Orthocrema) brevispinosa subsp. rochai Forel . Emery 1924a: 134.
Crematogaster (Orthocrema) brevispinosa rochai Forel . Kempf 1972: 86.
Crematogaster (Crematogaster) rochai Forel . Longino 2003: 102.
Crematogaster (Crematogaster) malevolens Santschi . Longino 2003: 131. Raised to species.
Crematogaster rochai can be diagnosed by the following combination of characters: antennal scapes short, less than.8x head length, in repose failing to reach posterior margin of head in full-face view; dorsal surface of head largely devoid of erect setae; promesonotum slightly arched in profile; propodeal suture present but not deep; postero-dorsal face of propodeum forming a single concave slope, not broken into distinct dorsal and posterior faces; propodeal spines short and upturned; first gastric tergite (= abd. tergite 4) with <6 standing setae exclusive of posterior row.
Mature colonies in the Crematogaster crinosa complex, including C. rochai , are polymorphic in the worker caste and occasionally produce rather large workers. Santschi’s C. malevolens seems to be little other than a large C. rochai . I can find no consistent characters to separate C. malevolens from the C. rochai type or from Paraguayan material that Jack Longino has identified as C. rochai . Longino (2003) elevated C. malevolens to species with the intent to clear trinomials from the C. crinosa complex but did not provide additional reasoning. If C. malevolans is indeed conspecific with what I call C. rochai in Paraguay, an alternate resolution would be to split these southern populations off from C. rochai under the name C. malevolens . Specimens from southern South America are more pilose and bear a stronger propodeal suture than more northerly material Longino (2003), character states that apply equally to C. malevolens . Considering the allopatric nature of the variation, however, I prefer to retain a single species.
