New species and new records of Manota Williston from Colombia, Brazilian Amazonia, and Costa Rica (Diptera, Mycetophilidae)

Abstract The following five species are described as new: Manota clava sp. n. (Colombia), Manota multilobata sp. n. (Colombia), Manota perplexa sp. n. (Costa Rica), Manota setilobata sp. n. (Colombia) and Manota subaristata sp. n. (Colombia). In addition, new records for the following 11 species are presented: Manota acuminata Jaschhof & Hippa, 2005 (Costa Rica), Manota arenalensis Jaschhof & Hippa, 2005 (Costa Rica), Manota corcovado Jaschhof & Hippa, 2005 (Costa Rica), Manota costaricensis Jaschhof & Hippa, 2005 (Costa Rica), Manota diversiseta Jaschhof & Hippa, 2005 (Colombia, Brazilian Amazonia, Costa Rica), Manota minutula Hippa, Kurina & Sääksjärvi, 2017 (Brazilian Amazonia), Manota multisetosa Jaschhof & Hippa, 2005 (Costa Rica), Manota parva Jaschhof & Hippa, 2005 (Colombia, Costa Rica), Manota pisinna Hippa & Kurina, 2013 (Brazilian Amazonia), Manota spinosa Jaschhof & Hippa, 2005 (Colombia) and Manota squamulata Jaschhof & Hippa, 2005 (Costa Rica). Distribution patterns include (1) species known only locally in Costa Rica or Colombia, (2) distributions connecting Central America to west Andes lowlands, and (3) north-west Neotropical components, extending from Central America to Brazilian Amazonia. The possible biogeographical and taxonomical context of Manota species with a widespread distribution is considered.


Introduction
The monophyletic subfamily Manotinae of Mycetophilidae is represented by four extant genera in the world fauna, but only Manota Williston (type species Manota defecta Williston) has almost cosmopolitan distribution, with the highest diversity in tropical areas (e.g. Kurina and Hippa 2015 and references therein). The other three genera are restricted either only to the Oriental region (in case of Paramanota Tuomikoski and Promanota Tuomikoski) or to the Oriental and Australasian regions (in case of Eumanota Edwards) (Søli 2002, Papp 2004, Hippa and Ševčík 2010. The Oriental region could be an original area of distribution of the manotines (Jaschhof et al. 2011). Manota is considered to be the sister group of a clade including the other three genera (Hippa et al. , Ševčík et al. 2013. Members of Manota have a unique habitus, including small size, yellowish to brownish coloration, and considerably reduced wing venation, which makes them easily recognizable in samples (for characteristic general facies, see, e.g., Kurina 2012, Hippa et al. 2017).
The last 15 years revealed an explosion in the number of described Manota species. The number of species of the genus in the world increased from 28 (Bechev 2000) to 271 (Hippa et al. 2017). In the Neotropical region, this number moved from three species (Papavero 1978) to 67 species known to date, viz. 32 species from Peru (Hippa et al. 2017), 27 species from Costa Rica , 21 species from Ecuador (Hippa and Kurina 2013), 8 species from French Guyana (Hippa and Kurina 2013), four species from Mexico (Hippa and Huerta 2009), two from Brazil (Enderlein 1911, Lane 1948, and one species from each St. Vincent, Lesser Antilles (Williston 1896), Nicaragua (Hippa and Kurina 2013) and Argentina (Hippa and Kurina 2013). Most of the species are known from their type localities only. The genus was mentioned as occurring in Colombia by Oliveira and Amorim (2016); the details on the species are being dealt with in this paper. A considerable number of species from the Atlantic Forest of Brazil are to be described soon, increasing the diversity of the genus in the region.
The aim of this study is to increase the knowledge of the genus Manota in the Neotropical region by describing new species and by giving new records based on material collected in Colombia, Brazilian Amazonia, and Costa Rica.

Materials and methods
The Colombian material was collected by Malaise traps within the framework of a collection project leaded by Dr. M. Sharkey (National Science Foundation Grant DEB-0205982; see also Oliveira and Amorim 2016). This project resulted in a huge amount of material, including fungus gnats, of which only a few genera have been worked to date (e.g., Oliveira and Amorim 2012, Kurina and Oliveira 2015. The Costa Rican material was collected by Malaise traps or sweeping from rainforest near the Soltis Center for Research and Education, San Isidoro. The material from Brazil comes from Malaise traps at the Reserva Ducke, in Manaus, State of Amazonas, and at the State of Roraima, close to the border with Venezuela. All the material was initially stored in ethyl alcohol. In most cases, the hypopygium was detached from the specimen and macerated in warm 20% potassium hydroxide (KOH). Several specimens, especially those collected in Colombia, were faded after being more than a decade in alcohol. After macerating in KOH and washing in distilled water, the hypopygium was stained with Chlorazol Black and thereafter mounted in "Euparal" between two pieces of coverslip, which allowed a study from both sides under a compound microscope. These preparations are now attached to a normal microscope slides by two strips of adhesive tape across their edges and are easily detached when needed, together to the remainder of the body, which was not macerated, but dehydrated and mounted in "Euparal" under a coverslip.
The morphological terminology follows mainly Søli et al. (2000), while the term "parasegment" is used in accordance with Jaschhof and Hippa (2005). The terminology of the hypopygium follows Hippa and Papp (2007), but the term aedeagus is used here instead of tegmen. The terminology of hypopygium is explained in Figs 1-5. The mid tibial organ is an area of tightly placed setae basoventrally on the mid tibia (Jaschhof and Jaschhof 2010). The hind tibial organ is a similar area apicoventrally on the hind tibia (Jaschhof et al. 2011). Wing length was measured from wing base to wing tip. Description of colour was made from specimens on slides under a stereomicroscope; when available, additional specimens in ethanol were used to confirm sclerite colours. In slides, medial part of the scutum, scutellum and abdominal tergites appear somewhat darker due to their curvature on slide, while they are unicolorous in ethanol preserved specimens. Illustrations were made with the aid of a drawing tube attached to a Leitz Diaplan compound microscope. The slide mounting was done under a Leica MZ16 stereomicroscope; compound microscopes Leica DM 2500 and Leica DM 6000 B were used for final identification of species.
The material is deposited in the following collections:
Female. Unknown. Discussion. The setose laterotergite, non-setose anterior basalare, sternite 9 laterally free from gonocoxa, and gonocoxa without a remarkable posterolateral lobe group together Manota clava sp. n. with M. caribica Jaschhof & Hippa, 2005 (Costa Rica) and M. micula Hippa & Kurina, 2013 (Ecuador, Peru). All three species have also the sternite 9 posteriorly broadly concave and a similar aggregation of setae on plate-like lobe ventrally from dorsal medial margin of gonocoxa. Parastylar lobe is distinct be-tween all three species: large, apically broadened with 3-5 posterior setae in M. clava, large, subtriangular with three posterior setae in M. micula and small, stout with 2-3 setae posteroapically in M. caribica. The gonostylus of M. clava is subrectangular and posterolaterally drawn out while it is oval or almost circular in case of the two other species. Manota clava and M. micula have the juxtagonostylar megasetae complex with transverse and leaf-like expansions, while they are simple and pointed in M. caribica.
Etymology. The specific epithet is Latin, clava [club or mace], referring to the prominent club-shaped parastylar lobe, and is a noun used as in apposition. Diagnosis. Laterotergite non-setose; anterior basalare non-setose; sternite 9 posteriorly broadly and deeply concave, anterior half fused to gonocoxa; parastylar lobe transversally oblong, with ca 20 setae; gonocoxa drawn into a short and broad posterolateral lobe; gonostylus widening apically, somewhat sunken into gonocoxa; two juxtagonostylar megasetae, ventral one flame-shaped, dorsal one twisted; two and one apically setose finger-like lobes anteriorly and posteriorly from juxtagonostylar megasetae, respectively.

Manota multilobata
Description. Male. Colour. Head brown, face somewhat paler. Antenna light brown, including scape and pedicel. Clypeus and mouthparts yellowish. Thorax light brown. Legs yellowish, basal third of femur 3 infuscated. Wing with brownish tinge because of microtrichia; halter yellow with blackish knob. Abdomen with tergites brownish, sternites somewhat lighter. All vestiture pale, yellowish or brownish, thicker setae and trichia seeming darker than finer ones. Head. Antennal flagellomere 4 ca. 2.1 times as long as wide. Palpomere 3 of maxillary palpus with apicomesial thumblike extension, its curved sensilla not discernible; palpomere 4 with parasegment; palpomere 5 not measurable on holotype. Number of strong postocular setae 9. Thorax. Anepisternum with 26 setae; anterior basalare, preepisternum 2 and laterotergite nonsetose; metepisternum with 8 setae. Legs. Mid and hind tibial organs absent. Wing. R 1 meeting C within basal half of costal margin; sclerotized part of M 2 extending to level of tip of R 1 ; wing length, 2.5 mm. Hypopygium ( Fig. 2A-C). Sternite 9 broad, extending to the middle of gonocoxa, anterior half fused to gonocoxa, posterior half free, posterior margin broadly and deeply concave, anterior margin shallowly and angularly incised, laterally with bare narrow area, medially setose with setae similar to the adjacent ventral setae of gonocoxa. Ventral medial margin of gonocoxa simple. Posterior margin with two long setae having prominent sockets. Parastylar lobe transversally oblong, well exposed in ventral view, bearing ca 20 setae. No paraapodemal lobe observable. Posterolateral part of gonocoxa not drawn into a remarkable lobe. Dorsal medial margin of gonocoxa simple, bulging medially, contiguous with the dorsal posterior margin. Two juxtagonostylar megasetae arising from separate basal bodies, dorsal megaseta somewhat twisted, with a basal body ca. one fourth of seta's length, ventral megaseta flame-shaped, with a basal body slightly less than seta's length. Two finger-like lobes anteriorly from juxtagonostylar megasetae: more anterior lobe subequal to basal body of dorsal juxtagonostylar megaseta with one seta apically, more posterior lobe subequal to basal body of ventral juxtagonostylar seta with three setae apically. Posteriorly from the juxtagonostylar megasetae, a lobe, subequal to basal body of ventral juxtagonostylar seta, bearing one strong apical seta and one weak subapical seta. Gonostylus somewhat sunken into gonocoxa, apically widening, with 5 strong and long setae at posterior and posterolateral margins, other setosity similar to that on gonocoxa ventrally. Aedeagus elongate, narrowly subtriangular, the lateral sides slightly concave, apex curved ventrally. Hypoproct extending posteriorly to level of apex of gonostyli, each side with 6 setae on apical third ventrally. Cerci broad, medially separated.
Discussion. Manota multilobata sp. n. groups together with M. setilobata sp. n. by having the non-setose anterior basalare, non-setose laterotergite, indistinct or short posterolateral lobes of the gonocoxa, and the megasetae and aggregations of setae at the dorsal medial margin of the gonocoxa all placed far posteriorly. Both species have the obovate gonostylus, which is somewhat sunken into the gonocoxa, and have 4-5 strong apical and subapical setae deviating from other setae, similar arrangement of small setose lobes around juxtagonostylar setae, and sternite 9 basally fused with the gonocoxa. The species differ as follows: 1) in M. multilobata there are two finger-like lobes close together anteriorly from the juxtagonostylar megaseta, the more anterior one with one, the more posterior one with three strong setae (in M. setilobata there is a plate-like lobe anteriorly bearing one seta widely separated from a posterior group of several setae), 2) in M. multilobata posteriorly from the juxtagonostylar megasetae there is a finger-like lobe with one strong and one weak seta (in M. setilobata a flat lobe with numerous fine setae), 3) in M. multilobata the gonocoxa is drawn into a short and broad posterolateral lobe (in M. setilobata it is drawn into a short and narrow lobe), and 4) in M. multilobata sternite 9 has the posterior margin broadly v-shaped incised (in M. setilobata there is narrow and deep medial cleft).

Manota perplexa
Description. Male. Colour. Head brown, face somewhat paler. Antenna light brown, scape, pedicel and two basal flagellomeres slightly paler. Clypeus and mouthparts yellowish. Thorax light brown. Legs yellowish. Wing with light brownish tinge because of microtrichia; haltere yellow with brown knob. Abdomen with tergites dark brown to blackish, sternites yellowish. All vestiture pale, yellowish or brownish, thicker setae and trichia seeming darker than the finer ones. Head. Antennal flagellomere 4 ca. 1.8 times as long as wide. Palpomere 3 of maxillary palpus with apicomesial thumb-like extension, with three apically curved sensilla; palpomere 4 with parasegment; palpomere 5 ca. as long as palpomere 4. Nine strong postocular setae. Thorax. Anepisternum with 46 setae; anterior basalare, preepisternum 2 and laterotergite nonsetose; metepisternum with 15 setae on anterior part. Legs. Mid and hind tibial organs absent. Wing. R 1 meeting C within basal half of costal margin; sclerotized part of M 2 extending to level of tip of R 1 ; wing length 1.6 mm. Hypopygium (Fig. 3A-D). Sternite 9 laterally entirely fused to gonocoxa, posterior margin free with protruding posterolateral corners; setae similar to adjacent ventral setae of gonocoxa. Ventral medial margin of gonocoxa simple, posteromedial corner drawn into a lobe, posterolateral part of gonocoxa not drawn into a lobe. Parastylar lobe indistinct, apparently represented by one seta by ventral medial margin of gonocoxa. No paraapodemal lobe observable. Dorsal medial margin of gonocoxa with a transverse shallow incision medially. A large plate-like lobe bearing four simple megasetae medioventrally from dorsal medial margin and anteriorly from the juxtagonostylar setae, anteriormost with its own basal body ca. 1/3 longer than others. Two juxtagonostylar megasetae present, more anterior one pointed, slightly curved simple megaseta, arising from a basal body which is shorter than the megaseta, more posterior one bifurcate, one of the branches whip-like, the other flat and dilated, arising subapically from a very prominent basal body which is longer than the megaseta itself and ca. as long as gonostylus. Ventrally from the more anterior juxtagonostylar megaseta there is an apically twisted megaseta. Dorsally at the posterior margin of gonocoxa a long finger-like lobe apically bearing a seta (in holotype the seta is broken on both sides). Gonostylus subtriangular, with prominent lateral angle, with one very strong and 2-3 weaker setae at posteromedial corner, and with one curved strong seta at posterior margin. Aedeagus narrowly subtriangular, the apex curved ventrally, otherwise the details not visible in the mount. Hypoproct posteriorly extending to the base of gonostyli, with ca. 35 ventral setae on each side. Cerci medially separated, with their apical parts narrowed.
Female. Unknown. Discussion. In the key to Costa Rican species by Jaschhof and Hippa (2005) Diagnosis. Laterotergite non-setose; anterior basalare non-setose; sternite 9 posteriorly and anteriorly deeply incised, posterior third laterally free; parastylar lobe indistinct; posterolateral part of gonocoxa drawn into a narrow lobe; dorsomedial margin of gonocoxa with a large plate-like lobe bearing one strong seta at posteromedial corner; gonostylus elongated subquadrangular, slightly sunken into gonocoxa; two juxtagonostylar megasetae, both twisted, the more dorsal one apically flattened and dilated; posteriorly from the juxtagonostylar megasetae a narrow flat apically setose lobe.
Wing. R 1 meeting C within basal half of costal margin; sclerotized part of M 2 extending to level of tip of R 1 ; wing length, 2.4 mm. Hypopygium (Fig. 4A-F). Sternite 9 ca. 2/3 as long as gonocoxa, anterior 2/3 laterally fused to gonocoxa, posterior 1/3 free, posterior and anterior margins with deep incisions which separate the sclerite almost into two halves, covered with setae similar to adjacent ventral setae of gonocoxa. Ventral medial margin of gonocoxa simple. Parastylar lobe not identifiable with certainty, possible fused with gonocoxa and in Fig. 4A comprising the part visible between the posterior margin of sternite 9 and the gonostylus. No paraapodemal lobe observable. Posterolateral part of gonocoxa drawn into a narrow lobe. Dorsomedial margin of gonocoxa simple. In a more ventral level, a large plate-like lobe bearing one strong seta at posteromedial corner. Two juxtagonostylar megasetae present, both twisted, the more dorsal one apically flattened and dilated, both arising from basally fused basal bodies which are as long as the megasetae. Dorsally from juxtagonostylar megasetae, a flat apically setose lobe, connected with a thin, one seta bearing plate-like lobe anteriorly from it. Posteriorly from the juxtagonostylar megasetae a narrow flat apically setose lobe. Gonostylus elongated subquadrangular, slightly sunken into gonocoxa, with 4-5 strong and long setae at posterior margins, other setosity similar to that on gonocoxa ventrally, dorsal side non-setose. Aedeagus subtriangular, lateral sides slightly concave, apex curved ventrally. Hypoproct extending posteriorly over apex of gonostyli, each side with 4-5 strong setae apically and ca. 30 fine setae on ventral surface. Cerci medially separated.
Female. Unknown. Discussion. Manota setilobata sp. n. resembles M. multilobata sp. n. For a more detailed discussion on distinguishing characters, see above.
Etymology. The specific epithet is Latin, setilobata [with seta-bearing lobes], referring to the apically setose lobes dorsally on the gonocoxa (adjective).  Diagnosis. Laterotergite non-setose; anterior basalare non-setose; sternite 9 laterally fused to gonocoxa except for posterior fifth; parastylar lobe indistinct; posterolateral part of gonocoxa drawn into a lobe; dorsomedial margin of gonocoxa with a plate-like lobe bearing one anterior and two posterior simple megasetae; gonostylus in dorsal and ventral view narrow, crescent-shaped; two juxtagonostylar megasetae, more dorsal one subbasally geniculate and apically bifurcate, more ventral one simple, slightly flattened whip-like; posteriorly from juxtagonostylar megasetae a short fingerlike lobe with 3-4 strong setae.
Female. Unknown. Discussion. Manota subaristata sp. n. is similar to M. aristata Hippa & Kurina, 2013 in having the dorsal juxtagonostylar megaseta with a long whip-like branch. Manota subaristata, however, has the megaseta subbasally geniculate, arising from a separate basal body, while it is basally straight and arising from apical half of the common basal body with the ventral juxtagonostylar seta in M. aristata. Manota subaristata has 3-4 strong setae on a finger-like setose lobe posteriorly from the juxtagonostylar megaseta, which are absent in M. aristata. By the latter character, the species resembles M. acutistylus , but the megasetae at the dorsal medial margin of gonocoxa in M. subaristata are longer and there are two of them in the posterior group, not three as in M. acutistylus (see also the discussion for M. aristata in Hippa and Kurina 2013: 109). In M. subaristata, the juxtagonostylar megasetae are subequal in length, while M. acutistylus has the dorsal juxtagonostylar megaseta remarkably shorter, ca. half of the length of ventral one. All these three species have the gonostylus apically tapering in dorsal and ventral view, and sternite 9 largely fused to the gonocoxa, and may compose together a small clade within the Neotropical diversity of the genus.
Etymology. The specific epithet is Latin, formed from the specific epithet of M. aristata by the prefix sub-[somewhat], as a reference to the similarity of the two species (adjective).
Remarks. The species was earlier known from Costa Rica , Ecuador (Hippa and Kurina 2013) and Peru (Hippa et al. 2017), hence widespread at the north-western corner of South America and Central America.
Remarks. Manota arenalensis was earlier known only from Costa Rica .
Remarks. The terminalia of the specimen studied here slightly differs from those figured by Jaschhof and Hippa (2005: fig. 16): three internal megasetae on the gonocoxa subapically (= position IV by Jaschhof and Hippa 2005) are more smoothly outlined, the gonostylus is slightly wider and the apical re-curved seta on gonostylus is pointed instead of being blunt. These differences are here considered to be within intraspecific variation or are differently exposed due to different position at the slide-mounting. The large posterolateral lobes of the gonocoxa, the number and arrangement of megasetae at ventromedial margin of the gonocoxa, the shape of juxtagonostylar megasetae and sternite 9 are identical to the specimens of the original description of M. corcovado. The species was earlier known only from Costa Rica . Remarks. The species is known only from Costa Rica .  , the species has subsequently been recorded from Ecuador, French Guyana (Hippa and Kurina 2013) and Peru (Hippa et al. 2017).
Remarks. The species was earlier known only from the Iquitos area in Peru (Hippa et al. 2017). We have not been able to see much material from Manaus and this is the only species identified from the state of Amazonas, a species shared with other areas of the Amazon Basin. Remarks. Manota parva was earlier known only from Costa Rica  and Ecuador (Hippa and Kurina 2013). The additional records make it one of the widespread species at the north-west corner of South America and Central America.  (Hippa and Kurina 2013), the species has subsequently been recorded from Peru (Hippa et al. 2017). Without the Peruvian record, the species would represent a typical Guyana Shield distribution. However, its presence in Iquitos makes it probably another widespread species at least in north-west South America. Remarks. The species was earlier known from Costa Rica  and Peru (Hippa et al. 2017). This distribution is the same of that of M. parva, M. acuminata, M. diversiseta, and probably M. squamulata.
Remarks. Having been described from Costa Rica , the species has subsequently been recorded from Ecuador (Hippa and Kurina 2013).

Discussion
A distribution map of the species described in this paper is depicted in Fig. 6, while maps in Figs 7-8 sum up the distribution of other recorded species including earlier data from Jaschhof and Hippa (2005), Hippa and Kurina (2013), and Hippa et al. (2017). Among the new species, only M. subaristata, sp. n. is known from more than one locality, all of them in Colombia. The other four new species are known only from the type locality. The species previously described accumulate more records that suggest some distribution patterns.
There are general patterns known for the Neotropical region (Amorim andPires 1996, Amorim 2009) in which the fauna of north-western South America, including Amazon basin elements, connects to that of Central America. For some groups, as e.g. monkeys and some sciarids, as Rhynchosciara (Amorim and Pires 1996), the patterns refer to species restricted to smaller areas in lowlands on both sides of the Andes, Central America and Mexico, as well as in the Brazilian Amazon. The observed Manota distribution patterns show individual species with a considerably wide distribution, which fit in this larger pattern-named as North-west Neotropical (Amorim 2009). In some cases, the species distribution is slightly more restricted and connects populations of lowlands in the Chocó region of Colombia, west to the Andes, to populations in Central America. This is a quite well-known pattern and in the genus Manota it is the case of M. multisetosa and M. parva. Future collections may show that this is either a real pattern or that these species actually have wider distributions and they just were still not found in other parts of South America. In other cases, species as M. acuminata, M. diversiseta, M. spinosa, and M. squamulata, present in Costa Rica, are also found east of the Andes, including Iquitos, at the west of the Amazon basin. This fits into an important biogeographical component, a triangular area delimited by the Andes, the Solimões river at the north, and Madeira-Mamoré rivers at the southeast. Some of the above-mentioned patterns can be discerned also in other groups like phorids of the genera Apocephalus Coquillett and Dohrniphora Dahl (Brown 2002, Brown andKung 2007).
Although the patterns sometimes are obvious, explanations can be more complex. Nominal species distributions that enclose areas of different endemism may correspond to: (1) secondary expansion of younger species with prior local distribution; (2) lack of response of older species to barriers that affected the younger groups; or (3) clades with undetected, cryptic species. This cannot be answered for the Manota species in question. A phylogeographic study would be useful to verify whether populations of these widespread species at the extremes of their distribution are beyond the threshold of genetic differentiation, often used to recognized separate species. This kind of problem of insect species is hard to distinguish using only morphological features, as they can involve also "hidden" molecular divergence. This aspect has been addressed in a number of recent papers (e.g. Laamanen et al. 2003, Meier et al. 2006, Bickford et al. 2007, Tan et al. 2010, Rohner et al. 2014. If these taxonomical entities in Manota presently called species actually correspond to clades of more local species, the genus would show even more intensely its condition of an open-ended taxon (Bickel 2009). This would put Manota even closer to Megaselia as one of the most diverse and taxonomically complex genera in flies.