Tomoceridae (Collembola, Entomobryomorpha) from the southern Annamitic cordillera: redescription of Tomocerus ocreatus Denis, 1948 and description of a new species of Tomocerina Yosii, 1955

Two species of Tomoceridae were found near Dalat, southern Vietnam. Tomocerus ocreatus Denis, 1948 is redescribed based on a neotype specimen. Previous records of Tomocerus ocreatus in non-type localities are reevaluated. A new species Tomocerina annamitica sp. nov. is described. The new species is mainly characterized by its small body size, pointed tenent hair, compound dental spines and the absence of intermediate teeth on mucro.

To clarify the true status and distribution of Tomocerus ocreatus, a redescription of its type specimen is required, but the holotype and only specimen of the species has been lost in the Muséum national d'Histoire naturelle (Paris) where the Denis' collection is currently stored. In a recent sample collected from Dalat (Lam Dong Province) in 2008, we have found a specimen that matches well the original description in morphology. We designate this specimen as neotype of Tomocerus ocreatus.
A new tomocerid species was found in the same sample as Tomocerus ocreatus. We assign the new species to Tomocerina Yosii, 1955 because it has no toothlet on the outer basal mucronal tooth. To our knowledge, Tomocerina annamitica sp. nov. is so far the southernmost record of the genus.

Materials and methods
Specimens were collected with aspirators or Berlese funnels. Photographs were taken under a Jenoptik ProgRes C10+ camera mounted on a Leica MZ 16 stereomicroscope. Specimens were cleared in lactic acid and mounted in Hoyer's solution (Krantz 1978). For some specimens the head, the furca and the legs were cut off from the trunk and mounted separately for detailed observation. The slide-mounted specimens were studied using a Leica DMLB microscope or a Nikon 80i microscope.
We follow Fjellberg (2007) for maxillary lamellae numbering, Yu et al. (2014) for the pattern of cephalic dorsal chaetotaxy and Christiansen (1964) for body macrochaetotaxy. The description of the body chaetotaxy refers to one side only since in most case it is symmetric. The exact morphology of each chaeta was uncertain due to shedding. The dental spines formula follows that of Folsom (1913), in which the dental spines are arranged from basal to distal, with a slash indicating the separation between basal and medial subsegments and the Roman numerals referring to spines that are noticeably larger. If not mentioned specially, all descriptions are based on fully developed individuals.

Diagnosis
Moderate to large sized Tomocerinae, usually longer than 3 mm; body colour pale to dark, some species with distinct colour pattern; eyes at most 6+6; trochantero-femoral organ reduced to 1, 1 chaetae; dens of furca basally without outer strong chaetae or inner large differentiated scales; shape of dental spines from simple to compound among different species; mucro with two dorsal lamellae and two basal teeth, outer basal tooth with corner toothlet. Commonest group of Tomocerinae in Eurasia Continent, especially abundant in East Asia. Denis, 1948 Figs 1A, 2, 3

Diagnosis
Typical Tomocerus species with pale body colour, moderately long antennae and full set of 6+6 eyes. Head without distinct PAO; Th. II with relatively reduced number of macrochaetae; tenent hair clavate, moderately developed; unguis with 5-6 teeth; tenaculum unscaled, with numerous chaetae; manubrium without dorsal scales and blunt prominent chaetae; dental spines compound with numerous moderate sized denticles; no small dental spine between two distal large spines; mucro with 9-10 intermediate teeth.
Body densely clothed by scales and various types of chaetae. Scales of typical morphology of Tomocerinae, with continuous longitudinal ridges on surface (Lubbock 1873). Ordinary chaetae of different sizes. Microchaetae smooth and pointed. Macrochaetae and mesochaetae from slightly to strongly ciliated, some slightly ciliated mesochaetae appearing to be smooth under optical microscope. Most macrochaetae straight, rod-like and subcylindrical, some macrochaetae on posterior abdominal segments long, curved and acuminate. Mesochaetae acuminate, shorter and thinner than macrochaetae. S-chaetae subcylindrical, more hyaline than ordinary chaetae, as small as microchaetae except long ones on Abd. IV. Dorso-inner chaetae on dens modified as strong pointed spines. Pseudopores as small circular structures similar to chaetae sockets, distributed at least on Th. II to Abd. IV, coxae, and manubrium. formula) 4/5, 5, 4, the distal 4 chaetae stronger. Distal edge of labrum with four curved spine-like papillae ( Fig. 2A), and a well developed ventro-distal brush. Mandibular head asymmetrical, the left one with 4 teeth (including a basal rounded one) and the right one with 5, left molar plate distally with a tapered tooth (Fig. 2B). Basal teeth of maxillary lamella 5 slightly longer than apical ones, without beard-like appendage (Fig. 2C). Maxillary outer lobe with trifurcate palp, one basal chaeta and 4 sublobal hairs (Fig. 2D). Both dorsal and ventral sides of head scaled. Cephalic dorsal macrochaetotaxy: anterior area: 2, 2; interocular area: 2, 6, central uneven macrochaeta absent; postocular area: 2+2; posterior area: 2.
Pattern of body chaetotaxy as in Fig. 2F. Bothriotricha 2, 1/0, 0, 1, 2, 0, 0 on Th. II-Abd. VI, respectively, as typical in Tomocerinae. Macrochaetae densely arranged along anterior margin of Th. II (not shown in figure). Th. II with a row of macrochaetae behind anterior margin. Number of macrochaetae or large mesochaetae in the posterior row as 3, 3/3, 3, 4, 2, 4 (3 dorsal+1 lateral) from Th. II to Abd. V. On Th. II no macrochaeta near the pseudopore contrary to most other tomocerids; on Abd. III s-microchaeta and accompanying microchaeta posterior to lateral macrochaeta; Abd. IV with one lateral macrochaeta and numerous long s-chaetae; on Abd. V inner macrochaeta smaller than others in posterior row; Abd. VI with numerous chaetae of moderate size. Most mesochaetae laterally and posteriorly on terga.

Remarks
Denis (1948) provided accurate description for Tomocerus ocreatus. For instance, he described clearly the dental spines as "covered with secondary spines, more similar to scales than denticles", and the figure showed that those scale-like denticles were of moderate size and covered at least basal half of each spine, on which account we rejected some records of non-type Tomocerus ocreatus (see below). But naturally, some later revealed characters were not mentioned by him. For instance, chaetotaxy had not been used for taxonomy in Tomocerinae before Yosii (1956). The limit of original description is a main reason for the misidentification of Tomocerus ocreatus in some subsequent records, and finally turned the species into a complex including a number of closely related forms from Vietnam to eastern Russia. To overcome the deep confusion about this species, a redescription of type specimen is necessary. Because the holotype and unique type specimen of Tomocerus ocreatus was lost, we propose to set up a neotype for the species from specimens of the type locality. Denis (1948) described Tomocerus ocreatus on one specimen, from "plateau du Lang Biang, 2400 m. alt., forêt tropicale". It is not the Lang Bian peak itself that reaches 2400 m but the Chu Yang Sin massif north of the Lang Bian (= Dalat) plateau, the Lang Bian peak being only 2197 m in altitude. The Chu Yang Sin massif was and still is of very difficult access, and was certainly not the place where the collector, Dawydoff, operated. There are two main patches of subtropical forests around Dalat: a small one on the Lang Bian peak, and a much larger one 25 km northeast on the Bi Doup massif, which reaches 2287 m and is less easy to access. In any case, it is most likely on the slopes of one of these massifs that Tomocerus ocreatus was collected. On this account, we propose to designate a specimen that we obtained from litter sample in the Bi Doup mountain as the neotype.
The neotype is highly identical with the original description in almost all described characters, including the body colour, the length of antennae, the structure of claws and the morphology and arrangement of dental spines. The only difference is that in the neotype specimen the numbers of ungual teeth, dental spines and mucronal teeth are slightly larger than in the original one. It could happen in Tomoceridae that within the same species larger individuals have a few more teeth and dental spines than smaller ones, so the slight difference mentioned above can be treated as intraspecific considering the neotype specimen is slightly larger than the lost holotype (3.6 mm versus 3.3 mm).
Regarding previous non-type records of Tomocerus ocreatus, Stach (1964Stach ( , 1965 redescribed two different forms of Tomocerus ocreatus from southeastern China and northern Vietnam, respectively, and also claimed that the Japanese record of "Tomocerus minor" by Uchida (1953) was actually Tomocerus ocreatus; Yosii described a Japanese species Tomocerus kawamurai (Yosii, 1954) and then synonymized it with Tomocerus ocreatus (Yosii 1956(Yosii , 1967; Chiba (1968) recognized three forms in Japanese Tomocerus ocreatus; Lee (1975) made descriptive notes on Korean specimens; Martynova (1977) recorded Tomocerus ocreatus in Sakhalin and made some descriptive notes on it. These redescriptions had subtle to considerable differences to each other, and were never identical to the original description. The Chinese "ocreatus" recorded in Hangzhou ("Hangchow"), Zhejiang Province by Stach (1964) has brownish body colour and distinctly short antenna about half the length of body; the northern Vietnamese record from Lao Cai Province by the same author (Stach 1965) bears a small dental spine between two large distal spines, and the denticles on spines are finer than those in the original description, therefore it is more similar to Tomocerus folsomi Denis, 1929 from Yunnan Province, China; the Japanese "ocreatus" described by Yosii (1967) has dorsal scales and 2+2 blunt principal chaetae on manubrium which are absent in the true Tomocerus ocreatus; other records (Chiba 1968;Lee 1975;Martynova 1977) all have dental spines with one or two whorls of crownlike denticles near the base which is not the morphology of the true ocreatus form. But these remarkable differences were treated as intraspecific variations when other characters showed high similarity. Inferred from the present morphological review and the molecular study on Chinese ocreatus complex , the aforementioned non-type "ocreatus" may each represent an independent species in the complex, and so far the true Tomocerus ocreatus is known as only endemic to southern Annamitic range in Vietnam.
Genus Tomocerina Yosii, 1955 Diagnosis Small to moderate sized Tomocerinae, usually less than 2 mm in length; body colour pale to light grey, seldom dark; antennae usually much shorter than body; eyes at most 6+6; trochantero-femoral organ with 1, 1 or more chaetae; dens of furca basally without outer strong chaetae or inner large differentiated scales; shape of dental spines from simple to compound; mucro with two dorsal lamellae and two basal teeth, outer basal tooth without corner toothlet. Widely distributed in Northern Hemisphere.

Diagnosis
Tomocerina species with typical body size and pigmentation; antennae about 0.8 times as long as body, relatively long for Tomocerina; Th. II with only one bothriotrichum; tibiotarsi with multiple strong chaetae; tenet hair small and pointed; unguis with 1-2 inner teeth; dental spines compound with denticles; only one distal spine distinctly larger; mucro without intermediate tooth.

Etymology
Specific name derived from its type locality: the southern Annamitic range.

Description
Body length 1.6-1.8 mm. Body colour light grey, antennae light purple, eye patches black, scales brown (Fig. 1B). Clothing of Tomocerinae type, similar to that of Tomocerus ocreatus.
Ventral tube anteriorly with a few small scales, each side with 20-25 chaetae; posteriorly unscaled, with about 40 chaetae; lateral flap unscaled, each side with about 25 chaetae. Tenaculum with 4+4 teeth, unscaled, anterior face with 1 chaeta about as long as rami (Fig. 5D). Ratio manubrium: dens: mucro= 2.0-2.2:2.9-3.2:1.0. Manubrium ventrally with only scales, laterally with scales and 8 chaetae, proximal 2 chaetae small and slender, distal 6 chaetae distinctly stronger; dorsal scales absent; each dorsal chaetal stripe with about 70 chaetae in different sizes, without blunt chaetae; pseudopores 5-6 on each side (Fig. 5E); external corner chaeta as large as small chaetae in chaetal stripe (Fig. 5F). Dental spines formula 4/3, I, a small pointed scale present ventrally to basal spines; all spines with several large denticles near base, superficial texture different between dorsal and ventral sides: dorsally with only fine longitudinal ribs, ventrally with longitudinal ribs and distal tiny serrations (Fig. 5G). The largest spine about 0.1 times as long as dens. Dens dorsally with feather-like chaetae between ordinary chaetae. Stripe of feather-like chaetae starting from centre of middle subsegment of dens, ending a short distance to the apex of dens. Ventral side of dens covered by scales, several chaetae present apically. Mucro elongated and multisetaceous, with two dorsal lamellae; both basal teeth with proximal lamellae, outer tooth without toothlet; apical tooth elongated, stronger than subapical tooth; intermediate teeth absent (Fig. 5H).

Remarks
Among Tomocerina, the new species is similar to those of the minuta group in its small body size, grey body colour and shape of mucro, but is different from them mainly in the chaetotaxy and the shape of dental spines. It is similar to Tomocerina purpurithora Liu, Hou & Li, 1999 from Sichuan Province, China in the shape of dental spines, but is different from the latter mainly in the cephalic chaetotaxy, the number of tibiotarsal strong inner chaetae, the number of chaetae on tenaculum, the dental spines formula and the number of teeth on mucro (Table 1). Tomocerina annamitica sp. nov. is the southernmost distributed species of Tomocerina, a genus mostly diversified in northern temperate to cold temperate regions. However, so far Tomocerina is poorly defined by only a minute character: the absence of corner toothlet on outer basal tooth of mucro. In this respect, the presence of only 1+1 bothriotricha on Th. II of Tomocerina annamitica sp. nov. might provide an interesting character, while most other species of Tomocerinae have 2+2. Investigation on this character in different groups of Tomocerinae is in progress to assess its phylogenetic value.

Discussion
Tomoceridae are mostly distributed in the northern temperate areas, with a diversity peak in eastern Palearctic: Japan (Yosii 1967), Korea (Lee 1975), Russian Far-East (Martynova 1977) and China (Ma 2004). The diversity rapidly decreases towards southern regions, but this family is still present in the caves or at high altitude of the subtropics (southern China and Vietnam, Denis 1929;Yu et al. 2014;Yu & Deharveng 2015;Nepal, Yosii 1966), or even at high altitude in tropical regions (Sumatra, Oudemans 1891;Vietnam, Denis 1948). Before this study, all three species of Tomoceridae known from Vietnam belong to Tomocerus. The fourth species added here, Tomocerina annamitica sp. nov., belongs to another genus which had so far a single species distributed south of 30°N (Tomocerina simplex Yosii, 1966 from Nepal). Unexpectedly, the new species was frequent in litter and soil as low as 1000 m of elevation. Besides Tomocerus ocreatus and Tomocerina annamitica sp. nov., at least one other species is probably present in our collections from southern Annamitic Cordillera, confirming its status of southern extension for Palaearctic fauna of springtails (Deharveng & Bedos 2000).