The scarab genus Termitotrox Reichensperger, 1915 is composed of ten blind, flightless species from the Ethiopian region (eight species) and the Indian subregion of the Oriental region (two species) (Krikken 2008). Wasmann (1918) originally established the tribe Termitotrogini in the subfamily Coprinae of Scarabaeidae to include this genus, but the rank of the taxon was treated differently by different authors from a distinct family of Scarabaeoidea (Termitotrogidae: Tangelder and Krikken 1982; Dellacasa 1988), a subfamily of Scarabaeidae (Termitotroginae: Smith 2006; Bouchard et al.2011), to a tribe of Aphodiinae (Termitotrogini: Scholtz and Grebennikov 2006). Currently, the subfamily status or the tribal status is accepted, but it remains unstable until a precise phylogenetic position of Termitotrox within Scarabaeidae is investigated. Most species have been found in fungus gardens of the fungus-growing termite genera Odontotermes Holmgren, 1912 or Protermes Holmgren, 1910 (Isoptera, Termitidae, Macrotermitinae). Recently, I collected an extraordinary, undescribed species of Termitotrox in Cambodia which is far from the known localities of the genus in the Oriental region (India) and is the first discovery of the genus in the Indo-Chinese subregion. It was collected from fungus garden cells of Hypotermes Holmgren, 1917 (also Macrotermitinae), a previously unknown host. This paper describes it as a new species.

When Arrow (1920) described Termitotrox minutus (originally in the genus Aphodiocopris Arrow, 1920) from India, he stated “probably the smallest Lamellicorn beetle [= scarab] hitherto discovered”. Termitotrox minutus is approximately 1.5 mm in length. Subsequently Pittino (2007) described Microtermitodius atomus, ranging from 1.4 to 1.7 mm, which reduced the size of the smallest known scarab still further. However, the new species is 1.2 mm long, and is therefore likely the world’s smallest scarab found thus far.

In August 2012, I visited Angkor Wat, Siem Reap, Cambodia, a typical tropical monsoon forest, for five days and worked a total of 15 hours collecting termitophiles associated with fungus-growing termites. I examined fungus gardens of the genera Macrotermes Holmgren, 1910 (four species), Odontotermes (three species), Microtermes Wasmann, 1902 (one species) and Hypotermes (one species). I found 25 Termitotrox beetles from fungus garden cells of Hypotermes makhamensis Ahmad, 1965 by examining more than 80 fungus gardens. The beetles were put in a killing tube (35 ml) with tissue paper and ethyl acetate; two hours later they were removed from the tube and kept in 80% ethanol. Twenty two specimens were dried and mounted for morphological observation, and remaining three are kept in 99% ethanol for future DNA extraction. Specimen photographs were taken using a Canon EOS 60D with a Canon MP-E 65 mm 1-5× macro lens and mounted using a software CombineZM. Images of living beetle were taken by the same camera set in the field by Dr. Takashi Komatsu. Terminology of the species description follows Krikken (2008). The type series is deposited in the author’s collection in the Kyushu University Museum, Fukuoka (KUM); Canadian Museum of Nature, Ottawa (CMN); The Natural History Museum, London (NHM); and the University of Nebraska State Museum (UNSM).

Termitotrox cupido specimens were found only on the walls of the fungus garden cells of Hypotermes makhamensis (Figs 7, 8), unlike many other termitophiles associated with fungus-growing termites, which are usually found inside the fungus gardens themselves. In the same habitat (on the cell wall), undescribed species of Odontoxenia Schmitz, 1915, Clitelloxenia Kemner, 1932, Ridiculiphora Disney, 1997 (Diptera, Phoridae) and Discoxenus Wasmann, 1904 (Coleoptera, Staphylinidae) were found, but they were also found inside the fungus gardens.

When the fungus garden was removed from the cell, the beetles walked slowly on the cell wall to escape to a tunnel connected to the other fungus garden cells, by following the termites. No direct contact with the beetle by the termites was observed even though the beetle walked among highly dense columns of the termites. Since Termitotrox cupido is so small (≈ 1.2 mm) and the termites were rushing to escape the disturbance, further observations of its termite association were not possible.

No behavioural information of Termitotrox species is available, other than that reported here for Termitotrox cupido. In two species of Corythoderini, the other termitophilous scarab tribe associated with fungus growing termites, beetles were being carried by worker termites as they do for their own nymphs (Kistner 1982; Maruyama 2012): the worker termite grabbed a handgrip-like structure which is formed by the median lobe of the pronotal base (medially projected posterior margin) and a median projection of the elytral base of the beetles. Though Termitotrox species do not possess an elytral median projection, they share the pronotal median lobe which is well developed and is handgrip-shaped, hence it could be hypothesized that Termitotrox beetles are likewise carried by the worker termites.

Termitotrox cupido is characterized by the large trichomes on the elytral costae, and these trichomes are not known among other Termitotrox species. Among inquilinous Aphodiinae trichomes on body surfaces have been observed in members of the tribes Corythoderini, Termitoderini, Stereomerini, and Rhyparini. Wasmann (1903) observed secretory tissue in Chaetopisthes assmuthi Wasmann 1911 (Corythoderini) under the posterior part of the pronotum and apices of the elytra, where the trichomes originate. This kind of tissue could exist in the trichomes of other species, and it probably produces chemical signals that target the host termites. Krikken (2008) speculated that the body surface convexities (including elytral costae) in the other Termitotrox species that do not possess trichomes might be filled with glandular tissue. The trichomes on the elytral costae in Termitotrox cupido indicate the presence of glandular tissue in the convexities of the elytral costae, suggesting glands might by present in this body region in the other species.

In the Oriental region the known distribution of Termitotrox species was restricted to the Indian subregion before the present finding of Termitotrox cupido in Cambodia in the Indo-Chinese subregion. Krikken (2008) mentioned that the Indo-Afrotropical distribution of Termitotrox is analogous to that of the Corythoderini. Recently, I collected a new genus and species of Corythoderini in the same place as Termitotrox cupido in Cambodia (Maruyama 2012). Aanen & Eggleton (2005) suggested that fungus-growing termites originated in the African rainforest, and their distributional range subsequently extended from tropical Africa to Southeast Asia. Corythoderini and Termitotrox might also have originated in Africa and have extended their distributions along with their host termites.