DESCRIPTION

Size. Largest specimen (holotype) 2.6 mm.

Color. Background color greenish white or orange.

Scale Distribution. Antennal segments 1 to 3, head, body, legs to tibiotarsus, collophore scales only on posterior face, and ventral face of furcula.

Head. Antennae missing 2 apical segments. Eyes apparently 6+6, but actual number and relative size of individual eyes not clearly seen; interocular setae 3 or 4, scales absent. Dorsal chaetotaxy as in Fig. 1: row An with 7 to 10 macrosetae; rows A, M, and S with 3, 3, 6 paired macrosetae, respectively, unpaired macrosetae A0 and S0 present; anterior microsetae A5, M3p, Ps2, Ps3, and Ps5 present; macroseta Pa5 present, all other postero-medial macro- and microsetae absent. Prelabral and labral setae smooth; inner labral papilla conic, outer papilla blunt (Fig. 2). Basal seta of outer maxillary palp variable, either sharply tapered (Fig. 3) or parallel-sided (Fig. 4), macroseta-like; sublobal plate with 4 smooth appendages. Labial papilla E with 5 guard seta-like projections, lateral appendage short, not reaching tip of papilla. Labial palps with 5+5 smooth proximal setae. Labial triangle series M with 6 or 7 setae of diverse sizes, 2 innermost setae ciliate, other setae in series M smooth; R smooth and slightly shorter than E; E, L1, L2, and series A smooth. Postlabial setae smooth, consisting of 3+3 setae along cephalic groove, 2+2 companion setae, and 12 lateral setae (Fig. 5).

Body. Dorsal macrosetae on Th. 2 to Abd. 4 as 10, 9/3, 3, 4, 5 + 0 + 7 and distributed according to Figs. 6–8, 10–12. Abd. 4 (Fig. 12) macroseta A5 inserted in field anterior to pseudopore and bothriotrix T4; macroseta B6 anterior to, but displaced towards seta Ae; only 1 macroseta external to column F.

Legs. Trochanteral organ L-shaped, with up to 23 smooth, stiff, acuminate setae. Posterior face of hind tibiotarsus with 2 rows of smooth setae. Tenet hair smooth and spatulate on all legs (Fig. 9). Unguis with 3 small inner teeth (Fig. 9), basal pair aligned and subequal, slightly longer than unpaired tooth. Unguiculus lanceolate, smooth.

Furcula. Manubrium dorsally with 4+4 erect, stiff, smooth setae; 2+2 similar setae present on base of dens. Dens with 2 inner rows of short spines, one row with ≈ 9 spines, other row with 11 spines. Mucro bidentate, apical tooth longer than basal tooth; basal spine smooth, reaching basal tooth.

TYPE MATERIAL

Holotype: Nevis, 2.4 km WSW Brick Kiln, 17.16374°N, 62.5723°W; 298.4 m elevation, leaf litter; 19.XII.2014; slide mounted. Paratypes: same information as holotype, 1 on slide and 9 in alcohol; 1.2 km WSW Brick Kiln, 17.16519°N, 62.57144°W; 277.7 m elevation, leaf litter; 19.XII.2014, 1 on slide and 1 in alcohol.

DISTRIBUTION AND HABITAT

Nevis and Panama (see Remarks below), leaf litter in moist tropical forest.

REMARKS

Dicranocentrus icelosmarias sp. nov. is the only member of the D. marias group with 3 inner macrosetae on Th. 3 that also carries lateral macroseta a6i on Th. 3 (arrow in Fig. 7) and 5 inner macrosetae on Abd. 4. The new species is part of the D. marias species complex (see below) and differs from D. marias Wray from Puerto Rico in having macroseta a6i on Th. 3 (microsetae in D. marias) and in having 5 macrosetae on Abd. 4 (4 in D. marias, Fig. 13). We examined 17 individuals of D. marias collected at various locations throughout Puerto Rico, including populations from Isabela (Guajataca), Mayaguez (Miradero), Villalba (Toro Negro), Orocovis, and Rio Grande (el Yunque), and all of them lack macroseta a6i on Th. 2 and all have 4 inner macrosetae on Abd. 4 The new species also differs from D. marias in that head element A5 is always a microseta, but in small adults of D. marias A5 is also a microseta.

In his revision of the genus Dicranocentrus, Mari Mutt (1979) reported 4 forms of D. marias, each with a unique combination of macrosetae on Abd. 1 and Abd. 4. Populations of D. marias from Mexico have 2 macrosetae on Abd. 1 and 4 or 5 on Abd. 4; individuals from Puerto Rico (species type locality) and Panama carry 3 macrosetae on Abd. 1, but whereas the Puerto Rican form has 4 macrosetae on Abd. 4, the individuals from Panama carry 5 macrosetae. Given the geographic isolation of each variant and the apparent stability in the number of macrosetae, it is likely that all these forms represent distinct species. The new species from Nevis is identical in all described details to D. marias from Panama, but the condition of macroseta a6i on Th. 3 remains unreported for the Panamanian form. We are provisionally assigning the populations from Panama to D. icelosmarias sp. nov., but a final decision on their identity awaits analysis of fresh material.

As indicated in the introduction, the idiochaetotaxy (i.e., all differentiated elements of the chaetotaxy, including microsetae, macrosetae, and sensilla) of Dicranocentrus is represented as diagrammatic macrosetae maps that lack the necessary information to evaluate the homology of individual elements. Figures 1, 6–8, and 10–15 show a first attempt at describing the dorsal idiochaetotaxy within the framework developed by Soto-Adames (2008) and Szeptycki (1979). Comparison of the idiochaetotaxy of D. icelosmarias sp. nov. with that of Orchesella flavescens (Bourlet), Heteromurus nitidus (Templeton), Pseudosinella alba (Packard), Cyphoderus albinus Nicolet, and Seira dowlingi (Wray) (Szeptycki 1979; Soto-Adames 2008) indicate that:

A. The anterior chaetotaxy of the head (i.e., rows A, M, S, and Ps) is almost complete, only element M0 is absent. The nomenclature proposed for elements in row S (Fig. 1) is provisional, based on the alignment of S3 and S5 with Ps3 and Ps5. The posterior section of the head of all mounted specimens is densely covered by scales and with the exception of element Pa5 appears to lack all setae posterior to row Ps. The D. marias species group is characterized by the absence of posterior head macrosetae and it is possible that even the inner microsetae in rows Pa—Pp have been lost.

B. The idiochaetotaxy of the mesothorax (Fig. 6) includes the primary elements retained in most scaled species, except, perhaps, m1. The identity of the elements closest to the pseudopore is unclear; they may represent m1 and m2, or a duplicated m2 (i.e., m2 and m2i). The simplest interpretation is to consider these setae as m1 and m2. However, the relative position of the elements to each other and to the pseudopore suggest they represent m2 and m2i; when present, m1 is inserted well anterior to m2, whereas in D. icelosmarias sp. nov., the 2 macrosetae are very close to each other, forming a row. Following the simple assumptions proposed by Szeptycki (1979) for identifying the homology of individual elements, we propose the inner macroseta is m1 (Fig. 6). Nevertheless, this interpretation will need to be confirmed with evidence from studies of postembryonic development.

C. The metathorax (Fig. 7) retains all the primary elements, although the identity of p6 is unclear and it is not identified in Fig. 7. Most macrosetae can be identified with primary elements and only macroseta a6i is the result of a duplication.

D. The 1st abdominal segment (Fig. 8) retains all primary elements except p5. Seta p5 is also lost in adult H. nitidus and C. albinus.

E. The idiochaetotaxy of the 2nd abdominal segment (Fig. 10) is almost identical to that in H. nitidus, including the absence of element m4. Seta m4 is also absent in Seira and Cyphoderus, but whereas in Heteromurus the seta is absent in the 1st instar, in Seira and Cyphoderus the loss is secondary.

F. The 3rd abdominal segment (Fig. 11) retains all primary elements and is identical in organization to that of H. nitidus.

G. The number of inner elements on the 4th abdominal segment is reduced and their homology is difficult to ascertain. Figures 12 and 13 show putative homologies in D. icelosmarias sp. nov. and D. marias, respectively, following the pattern proposed by Szeptycki (1979). Homologies in those figures are based on the chaetotaxy of O. flavescens because in H. nitidus the number of elements is even more reduced than in Dicranocentrus and homologies are also uncertain. Figures 14–15 show alternative homology propositions for D. marias: Fig. 14 is consistent with the unique condition in H. nitidus in which only 2 columns are retained, whereas Fig. 15 is consistent with all other scaled entomobryoids so far described in retaining 3 columns. The lateral chaetotaxy of this segment (Figs. 12, 13) is well conserved and the homology of the 9 main elements (D1–D3, E1–E3, and F1–F3) present in the 1st instar of almost all entomobryoids is evident.

In summary, the chaetotaxy of Dicranocentrus shows clear affinities with that of Heteromurus. This affinity is most evident on the 2nd and 3rd abdominal segments where Dicranocentrus and Heteromurus share the absence of m4 (Abd. 2) and presence of m3e (Abd. 3). The homology of several elements remains uncertain: the identity of lateral elements in series S on the head; the element near the pseudopore on the mesothorax; whether the loss of element p5 on Abd. 1 is primary or secondary; and more crucially, the identity of inner elements on Abd. 4. Resolution of these uncertainties will require analysis of 1st instar chaetotaxy.

ETYMOLOGY

The epithet means ‘similar to marias’, referring to the similarity with D. marias.

DISTRIBUTION AND HABITAT

Nevis, Puerto Rico, US Virgin Islands (St. Thomas), Jamaica, Galapagos, Colombia, Brazil (Mari Mutt & Bellinger 1990; Soto-Adames 2002; Bellini & Zeppelini 2009; Katz et al. 2016). This species is common in leaf litter in disturbed habitats, including urban forest patches, and marginal wooded areas.

MATERIAL EXAMINED

Locality 3: 1 on slide.

DISTRIBUTION AND HABITAT

Nevis. Leaf litter in transition zone between disturbed margin of moist woods and grassy roadside.

REMARKS

The single individual collected clearly represents a new species, but we defer formal description until additional material becomes available.

MATERIAL EXAMINED

Locality 3: 1 on slide.

DESCRIPTION

Size. Largest specimen 1.2 mm; holotype 1.1 mm.

Color. Light blue, evenly distributed through head and body; most intense on eye patch.

Scales distribution. Head, body, and ventral face of furcula.

Head. Antennae 0.34 times as long as head and body. Ant. 4 without apical bulb; subapical sense organ with capitate sensillum and apically truncate guard seta. Eyes 2+2. Head dorsal chaetotaxy as in Fig. 16: row An with 8 macrosetae; inner margin of eye patch with 1 macroseta; 1 unpaired (A0) and 2 paired (A2, A3) anterior macrosetae; 1 paired posterior macroseta (Pa5) present; row S with 1 unpaired (S0) and 4 paired microsetae (S2, S3, S4, S5); microsetae Ps2, Ps3, and Ps5 present. Prelabral setae ciliated; proximal row of labral setae sparsely ciliated, setae on medial and distal rows smooth; distal margin of labrum smooth. Pleural setae ciliated, proximal seta longer than distal. Maxillary palp with basal seta coarsely ciliated, apical appendage smooth and as long as basal seta; sublobal plate with 3 smooth seta-like appendages. Labial papilla E with 3 guard appendages; lateral appendage anteriorly curved, blunt and surpassing tip of papilla. Proximal labial setae 5, all serrated (Fig. 17). Labial triangle (Fig. 18) as M1M2rEL1L2A1-5: r a conic microseta, all other setae coarsely ciliated; L1 shorter than L2. All postlabial setae ciliated, with 3+3 setae along cephalic groove (Fig. 18); field of columns C-O with 6–10 setae.

Body. Body macrosetae formula 10/0232+0+4; regular sensilla (labeled “s” in figures) formula for Th. 2 to Abd. 3 11/011; micro sensilla (labeled “ms” in figures) formula 10/101. Mesothorax (Fig. 19) with macroseta p3; anterolateral regular and micro sensilla present; all setae smooth. Metathorax as in Fig. 20, inner posterior setae longer than those on anterior and medial rows: p1 longest, a2 shortest; seta m4 absent, latero-anterior regular sensillum present. First abdominal segment (Fig. 21) with 8 setae and 1 latero-anterior micro sensillum; setae a3, a5, and a6 absent. Chaetotaxy of Abd. 2–3 as in Figs. 22–23; relative insertion of micro sensillum d2 on Abd. 3 variable. Fourth abdominal segment with 2 inner and 5 lateral macrosetae (Fig. 24); posterior setae absent. Fifth abdominal segment with 13 macrosetae, and 3 regular sensilla distributed as in Fig. 25.

Legs. Trochanteral organ V-shaped, with up to 13 setae (Fig. 26). Femoral organ with 9–12 short blunt ciliate setae (Fig. 26). Tenet hair smooth, weakly spatulate (Fig. 27). Unguis with 3 inner teeth (Fig. 27–28): basal teeth asymmetric, one longer; unpaired tooth as long as or longer than smaller of basal teeth. Dorsal ungual tooth minute, inserted at base of ungues; lateral teeth inconspicuous, ending on basal quarter. Unguiculus lanceolate, with large posterior tooth (Fig. 27).

Collophore. Posterior face with 3 rows of finely ciliate or denticulate setae, distal row with 3+3 setae (Fig. 29).

Furcula. Mucro bidentate, with basal membrane, without basal spine (Fig. 30); teeth subequal.

TYPE MATERIAL

Holotype, male: Nevis, 2.4 km WSW Brick Kiln, 17.16374°N, 62.5723°W; 298.4 m elevation, leaf litter; 19.XII.2014; slide mounted. Paratypes, 1 male and 2 females, same as holotype, on slides, 5 in alcohol; 1.2 km WSW Brick Kiln, 17.16519°N, 62.57144°W; 277.7 m elevation, leaf litter; 19.XII.2014; 3 in alcohol.

DISTRIBUTION AND HABITAT

Endemic to Nevis. Leaf litter in moist tropical forest.

REMARKS

Following Mari Mutt & Gruia (1983), the presence of a femoral organ places the new species in the subgenus Sulcuncus. Metasinella (Sulcuncus) radkei sp. nov. is the only member of the genus with bidentate mucro, 2 eyes, and 3 paired (A2, A3, Pa5) dorsal head macrosetae; all other described Metasinella have 5 (A2, A3, M2, S3, Pa5) or 6 (A2, A3, A5, M2, S3, Pa5) paired macrosetae. The new species keys out to M. (S.) millsi Mari Mutt & Gruia in Mari Mutt & Gruia (1983), but M. millsi carries 5 paired dorsal head macrosetae whereas M. radkei sp. nov. has 3 macrosetae; in addition, M. millsi has 4 inner ungual teeth and the dorsal tooth is inserted at the level of the basal inner teeth, whereas the new species carries 3 inner teeth and the dorsal tooth is inserted near the base of the unguis.

The genus Metasinella comprises 2 well characterized subgenera (Mari Mutt & Gruia 1983): Metasinella s. str. with long antennae, 2 mesothoracic macrosetae, trochanteral organ with long flexible setae, dental spines, and lacking femoral organ; and Sulcuncus with short antennae, 1 mesothoracic macroseta, trochanteral organ with short, stiff, spine-like setae, femoral organ present, and without dental spines. The separation of the genus Metasinella into 2 subgenera is taxonomically sound, as each taxon is diagnosable by non-overlapping discreet characters, but they clearly are part of the same lineage. Most species in the subgenus Sulcuncus are known only from surface collections and show few adaptations to subterranean life. Some of the distinguishing characters of M. (M.) acrobates Denis, on the other hand, likely represent adaptations to living in caves. For example, evolution of large body size and long antennae is common in troglomorphic species (Christiansen 1961), and the dental spines may provide traction while jumping in wet, slippery surfaces. Similar adaptations are present in Pseudosinella spinosa (Delamare-Debouteville), the largest Pseudosinella in North American, and one of the most morphologically adapted cave forms (Delamare-Debouteville 1949). Thus, it seems appropriate to retain Sulcuncus as a subgenus of Metasinella.

As pointed out by Soto-Adames (2002), the dorsal chaetotaxy of the metathorax and 1st abdominal segments in Metasinella is reduced with respect to most Lepidocyrtus and Pseudosinella (cf. Figs. 20–21 and 31–32). The original interpretation of the metathoracic chaetotaxy by Soto-Adames (2002) indicated the absence of seta m2, a seta considered by Szeptycki (1979) to be diagnostic of Lepidocyrtini. The interpretation of Soto-Adames (2002) is here amended (Fig. 20) based on the clear observation of the complete chaetotaxy of the segment; the small size of the inner element closest to the pseudopore suggests it is m2, whereas the long element external to m2 is interpreted as p2. Provided this new interpretation, the only element missing in the metathorax of Metasinella is m4. The nomenclature of Abd. 1 (Fig. 21) is also amended with respect to that proposed by Soto-Adames (2002). Based on their relative size and insertion, the elements external to a2 are most likely homologous to m3 and m4. The new species also carries element m5, which is absent in the species from the Virgin Islands.

Minor reductions in idiochaetotaxy are relatively common in Lepidocyrtini. Some species of Lepidocyrtus lack Abd. 1 seta a6 (e.g., L. lanuginosus [Gmelin]), whereas in P. hirsuta (Delamare-Debouteville) and L. nigrosetosus species groups Abd. 2 lacks seta m3e (Christiansen & Bellinger 1998; Bernard et al. 2015). A more extensive reduction in the chaetotaxy of Th. 3 and Abd. 1 is seen in P. violeta Mari Mutt (Figs. 33–34). Pseudosinella violeta is also an Antillean species and it might represent part of the lineage from which Metasinella was derived.

ETYMOLOGY

The epithet honors Garett Radke, who provided accommodations in Nevis that made the collecting trip possible.

DISTRIBUTION AND HABITAT

This species is found in disturbed, grassy road-sides as well as secondary forests in Brazil, Bolivia, Puerto Rico, the Virgin Islands (Mari Mutt & Bellinger 1990; Mari Mutt et al. 1997–2016), the Dominican Republic, and the state of Florida (the last 2 localities are unpublished records).

MATERIAL EXAMINED

Locality 3: 1 in alcohol.

DISTRIBUTION AND HABITAT

This species is known from Puerto Rico and the US Virgin Islands where it is frequent in leaf litter in secondary forests (Soto-Adames 2002).

MATERIAL EXAMINED

Locality 3: 1 on slide, 6 in alcohol.

MATERIAL EXAMINED

Locality 1: 1 on slide, 2 in alcohol. Locality 2: 2 in alcohol.

REMARKS

This appears to be a new species, closely related to T. sauron Soto-Adames, Jordana & Baquero but all individuals are small juveniles.