Taxonomic evaluation of eleven species of Microcyclops Claus, 1893 (Copepoda, Cyclopoida) and description of Microcyclops inarmatus sp. n. from America

Abstract Description and meristic analysis of eleven species of Microcyclops recorded in America were performed based on the examination of type specimens and fresh material. Microscopic analysis of oral appendages, such as the shape and armature of the distal coxal endite of the maxilla, the ornamentation on the caudal surface of the antenna, and the intercoxal sclerites and armament of the inner basis of all swimming appendages, were characteristics that allowed the differentiation between species. Among these species, our study confirmed the synonymy of Microcyclops diversus Kiefer, 1935 with Microcyclops ceibaensis (Marsh, 1919). The results of our observations showed that Microcyclops alius (Kiefer, 1935) is a junior synonym of Microcyclops dubitabilis Kiefer, 1934; the latter being confirmed as a valid species. Also, it is proposed that the records of Microcyclops rubellus (Lilljeborg, 1901) and Microcyclops varicans (Sars, 1863) in America should be revised as there are serious doubts about their distribution in America. The analysis suggested that Microcyclops anceps pauxensis Herbst, 1962 is distinct from Microcyclops anceps var. minor Dussart, 1984 and that both are likely different from Microcyclops anceps anceps (Richard, 1897). Finally a full morphological description of adult females of Microcyclops inarmatus sp. n. is presented.

In her publication of an identification key for South American cyclopoids, Reid (1985) proposed that Microcyclops diversus is a probable synonym of M. ceibaensis (in 1986, this opinion was based on similarities in the fourth leg observed by the same author) and that M. anceps var. minor is a synonym of M. anceps pauxensis. Rocha (1998) proposed a set of morphological features that would be useful for distinguishing five species previously recorded in Brazil and suggested that M. alius is a possible synonym of M. dubitabilis. However, Reid (1990) had previously suggested that M. dubitabilis is a possible synonym of M. varicans.
Therefore, some questions on the taxonomic status of some species of Microcyclops in America remain unresolved. These taxonomic problems may be related to the lack of thorough and rigorous species descriptions. Rocha (1998), Mirabdullayev (1998Mirabdullayev ( , 2007, and Mirabdullayev and Urazova (2006) have documented morphological features that are useful for differentiating some species of the genus. For instance, they proposed the following morphological features as diagnostic: ornamentation of dorsal margins of prosomites, presence or absence of pores on second endopodite of first leg, ornamentation of the inner margin of basipodite of first leg, ornamentation of caudal ramus and caudal setae, relative lengths of caudal setae, proportions of second endopodite of fourth leg, and general ornamentation of fourth leg.
In Mexico, some species with uncertain taxonomic status have been recorded, including M. ceibaensis, M. anceps, and M. dubitabilis (Elías-Gutiérrez et al. 2008). In this paper, we propose a set of morphological features that are useful for distinguishing between these species, which have been documented by biological inventories of the country. These features include the mouth appendages, the ornamentation of intercoxal sclerites, and the inner margin of the basis of the first to fourth swimming legs.

Methods
The morphological analysis was performed following current standards for the taxonomic study of cyclopoid copepods (see Williamson and Reid 2001).
Material examined. The evaluation included analyses of holotypes, paratypes, and museum specimens deposited in different collections: the Collection of Zooplankton of ECOSUR at Chetumal (ECO-CH-Z), the collection of Copepoda of the Muséum National d'Histoire Naturelle, Paris (MNHN), the Staatliches Museum für Naturkunde, Karlsruhe (SMNK) and the National Museum of Natural History, Smithsonian Institution, Washington, DC (USNM) (Table 2, as Suppl. material 1).
The morphology of several species was examined using light microscopy: A1, A2, the mouthparts, the entire structure of all of the swimming legs, and other taxonomically relevant structures, such as the frontal or caudal ornamentation of BspA2, the ornamentation of the distal coxal endite of the maxilla, and the ornamentation of maxillular palp, were illustrated with the aid of a camera lucida.
Sources for the morphological data were the types, paratypes, and other museum specimens (Table 2, as Suppl. material 1), and original descriptions of eleven named species and two named subspecies recorded in America. Detailed descriptions based on the microscopic and morphometric analyses of the adult females of each species are presented.

Descriptive section
Below those morphological structures which are shared by all the species examined herein are described.
Etymology. the name of the species means un-armed in Latin; it refers to the absence of ornamentation on the intercoxal sclerites, the lack of spinules at base of caudal furcal setae, the reduced number of setae on second antennal endopod, and the reduced ornamentation on antennal basis.
Diagnosis. Adult female: Dorsal margin of prosomal somites smooth; body length 565 to 615 µm in paratypes. Antennule 12-segmented, not reaching the distal margin of the first prosomal segment ( Fig. 2A). Fifth pediger nude; cylindrical free segment of P5 more than 3 times as long as wide, with tiny inner spine; genital double somite expanded proximally. Anal somite with strong spines on ventral distal margin; length to width ratio of caudal ramus less than 3; no spinules at base of lateral and outermost terminal caudal setae (Fig. 2B). Outer median terminal and inner median terminal caudal setae with heteronomous setulation (Fig. 2B). Endopodites and exopodites of P1-P4 bisegmented with setation formula as in Table 1, inner basis of P1 with long spine (long arrow in Fig. 2C), Enp2P1 with one pore on lateral margin (short arrow in Fig. 2C). Intercoxal sclererites of P1-P4 unarmed, long setules on inner margin of basipodites of P1-P4, medial spine of Enp2P4 almost as long as the segment and twice the length of the lateral spine (Fig. 2D).
Adult male: unknown. Description of female. Antennule 12-segmented; antenna with 3-segmented Enp armed with 1, 6, and 7 setae respectively (Fig. 3A, B -position of missing setae in specimens is arrowed). Antennal basis with one group of spinules on the basal-outer margin in caudal view (Fig. 3A, B); frontal surface of the antennal basis with two rows of tiny spinules (Fig. 3C). Nine teeth on mandibular gnathobase (Fig. 3D). Maxillule ( Fig. 3E) with unarmed palp; apical region of maxillular palp with two armed setae plus one smooth seta, three setae (one armed) on lateral lobe, proximal seta smooth (Fig. 3F). Distal coxal endite of the maxilla with two long setae: the proximal seta with two tiny spines at its base and bifurcated, distal seta with one row of tiny spines along one margin (Fig. 3G, H). Basipodite with one claw-like projection bearing 5-7 strong spines on the concave margin and one long, armed seta on its base; two-segmented Enp bearing 2 and 3 setae respectively (Fig. 3G). Because of the condition of the microscope slide preparatum we could not verify one basal seta on maxillar Enp1 (arrowed in Fig. 3H). Maxilliped with syncoxa (3 setae), basis (2 setae), and twosegmented Enp bearing 1 and 3 setae. Basis of maxilliped with a row of spinules on frontal and caudal surfaces (Fig. 3I).
Fifth pediger nude; P5 free segment cylindrical, 3.6 ± 0.8 times as long as wide, without inner spine. Free segment 0.4 ± 0.1 times the length of the distal seta (Figs 7I-K; 8C). Length to width ratio of caudal ramus 2.4 ± 0.2, inner margin naked. With or without spines at base of seta II (intrapopulation variation), spinules always present at base of setae III. Distal margin of anal somite bearing spinules: medial spinules are longer than lateral ones on ventral surface; spinule row can extend laterally or dorsally ( Fig. 8 A-C). Seta II inserted at 71 ± 5.7% of caudal ramus.
Dorsal margin of prosomal somites slightly serrated (Fig. 10A). Basis of P1−P3 with pilose inner margin. Intercoxal sclerites of P1-P3 with one row of short spinules ( Fig.  10B-G), in some populations the sclerite of P3 with two rows of spinules (Fig. 10H). Enp2P1 with two pores on lateral margin (Fig. 10C). Because of the condition of the specimen in slide USNM-222299, it was not possible to verify the presence of spinules on the sclerite as well as the pores on the second endopodal segment of P1 (Fig. 10B).
P1 basis with long medial spine reaching distal third of Enp2P1. Spine ornamented with long setules near base and with short spinule-like setules more distally (Fig.   10B, C). Inner margin of P4 basis with strong spinules. Intercoxal sclerite rectangular, and ornamented with two rows of spinules (Fig. 10I, K). Enp2P4 2.2 ± 0.1 times as long as wide; medial spine 1.5 ± 0.1 times as long as the lateral spine, and 0.6 ± 0.06 times as long as segment (Fig. 10J, K). Sixth leg with one long seta plus two short spines (Fig. 10L).
Fifth pediger nude (Fig. 11A). Urosomal somites with serrated hyaline fringes (Fig.  11B, C). Fifth leg with one cylindrical free segment 3.8 ± 1.4 times as long as wide; tiny spinule present on inner margin. P5 free segment 0.3 ± 0.1 times the length of the distal seta (Fig. 11C). Distal margin of anal somite with a continuous row of strong spinules on ventral and dorsal surfaces (Fig. 11B, C). Caudal ramus 3.6 ± 0.4 times longer than wide, inner margin naked. Spinules present at base of caudal setae II and III; seta II inserted at 69 ± 3.2% of the caudal ramus (Fig. 11C).

Microcyclops echinatus Fiers, Ghenne & Suárez-Morales, 2000 Figures 12, 13
Description of female. This description is a complement to the original description of Fiers et al. (2000). Frontal surface of antennal basis with one basal, inner row of spinules arranged in an arc, and one longitudinal row of spinules near lateral margin (Fig. 12A). Antenna with three-segmented endopod bearing 1, 9, and 7 setae, respectively (Fig. 12B). Caudal surface of antennal basis with two rows of long spinules next to exopodal seta, one group of long spinules at basal position, one basal row on inner margin and another basal row on outer margin (Fig. 12B). Nine teeth on the distal margin of the labrum (Fig. 12C). Eight teeth present on gnathobase of mandible (Fig.  12D). Maxillular palp with three apical setae (one of these setae armed with long setules); lateral lobe with three setae, the longer seta armed; proximal seta nude (Fig.  12E). Maxillar basipodite with one claw-like projection bearing thin spines on concave margin and one long seta with one ( Fig. 12F) or four tiny spinules (Fiers et al. 2000); maxilla with two-segmented Enp bearing 2 and 3 setae on first and second endopodal segments, respectively (Fig. 12F). Maxilliped with syncoxa (3 setae), basis (2 setae), and two-segmented Enp bearing 1 and 3 setae. Syncoxa and basis of maxilliped with rows of spinules on caudal surface (Fig. 12G).
Two pores on lateral margin of second endopodal segment of P1, very long spinules present at insertion of apical spine of Enp2P1. Long medial spine of P1 basis with heteronomous setulation (Fig. 13A). Inner margin of P1−P3 basis with long hairlike setae (Fig. 13A, B), inner margin of P4 basis with one row of tiny spinules and one row of long setules (Fig. 13C). Intercoxal sclerites of all swimming legs ornamented on caudal surface: P1 with one row of spinules and P2 to P4 with two rows of spinules ( Fig. 13A-C). Enp2P4 2.5 ± 0.1 times as long as wide; medial spine is 2.0 ± 0.1 times as long as lateral spine, and 0.8 ± 0.1 times as long as the segment.

Microcyclops finitimus Dussart, 1984
Anal somite with continuous row of spinules along distal margin (ventrally and dorsally), but on ventral surface medial spinules are longer and stronger than lateral spinules (Fig. 14E). No spinules at base of caudal seta II, but spinules present at base of caudal seta III; length to width ratio of caudal ramus 2.7. Relative lengths of terminal caudal setae from outermost to innermost seta, 1.0 : 6.1 : 8.9 : 2.1. Caudal setae IV, and V with homonomous setulation, bearing long and fine setules (Fig. 14E). Dorsal caudal seta (VII) 0.7 times as long as caudal rami, and innermost caudal seta (VI) 1.2 times longer than caudal rami. Lateral caudal seta (II) inserted at 75.5% of caudal ramus length.

Microcyclops anceps anceps (Richard, 1897) Figures 15-17
Description of female. Dorsal posterior margin of second prosomal somite with crenulated hyaline fringe (Fig. 15A, B), posterior margin of fourth prosomal somite wrinkled (Fig. 15B). Caudal surface of antennal basis with three oblique rows of tiny spinules near inner margin and two basal (proximal) rows of long spinules near outer margin (Fig. 15C, D, F). Antenna with three-segmented endopod bearing 1, 9, and 7 setae, respectively. Frontal surface of antennal basis with three rows of spinules: one proximal oblique, one near lateral (outer) margin in middle of segment, and one next to exopod seta (Fig. 15E). Mandible with nine teeth on gnathobase (Fig. 15G). Maxillular palp with two armed and one naked setae apically; one armed seta plus two nude setae on lateral lobe, proximal seta with tiny spinules (Fig. 15H). Distal coxal endite of maxilla with two long setae: proximal seta with two long, basal setules and bifurcated    apically; distal seta armed with a continuous row of tiny spinules along one (inner) margin (Fig. 15I). Basipodite with claw-like projection bearing two stout teeth followed by a row of tiny spinules, and one long smooth seta on its base; two-segmented Enp bearing 2 and 3 setae, respectively (Fig. 15I). Maxilliped with a row of spinules in syncoxa, Bsp, and Enp1, on frontal view (Fig. 15J).
Inner margin of basipodite with long and fine hairs in P1−P3 (Fig. 16A, B). One pore present on lateral margin of Enp2P1. Spine absent on inner margin of BspP1 (Fig. 16A). Inner margin of BspP4 with long. Inner margin of BspP4 with long spinules (Fig. 16D). Intercoxal sclerites naked in P1 and P2 (Fig. 16A, B). Usually one distal row or sometimes two rows of spinules present on intercoxal sclerite of P3 (Fig.  16C). P4 sclerite with two rows of spinules (Fig. 16D): spinules in distal row larger and stronger than those in proximal row (Fig. 16D). Medial apical spine of Enp2P4 1.3 ± 0.1 times as long as lateral apical spine, and 0.7 ± 0.03 times as long as segment; length to width ratio of segment 2.5 ± 0.1.
Strong spinules present (Figs 16E, 17D) or absent (Fig. 17A) on fifth pediger near base of lateral seta of P5. This character shows both inter-and intrapopulation variation; in one population, the females do not have spinules next to lateral seta (MNHN-Cp7296, unfigured here), while the males of the same population do (Fig. 17E).
Free segment of fifth leg 2.5 ± 0.2 times longer than wide, with relatively large spinule in distal position (Figs 16E, 17A); free segment 0.4 ± 0.08 times as long as apical seta. Distal margin of anal somite with continuous row of strong spinules on ventral and dorsal surfaces (Figs 17B, F). Caudal ramus 3.7 ± 0.3 times longer than wide, inner margin naked. Spinules present at base of caudal seta III. Caudal seta II inserted at 71.1 ± 1.15% of caudal ramus length (Fig. 17B).

Discussion
Microcyclops anceps anceps showed the least variation in the qualitative and morphometric characters even though specimens were examined from a wide latitudinal range (Venezuela, Mexico, Guyana, Brazil, Guatemala, and Uruguay).
Microcyclops echinatus (from southeastern Mexico) and M. ceibaensis (from Honduras and southeastern Mexico) appeared morphologically similar. Similarities between these species are in: the length and width ratio of Enp2P4; the P4 sclerite with two rows of spines; the length ratio of the terminal caudal setae III and IV; the presence of spines at the insertion of setae II, and III; and the heteronomous ornamentation of the spine on the inner margin of Bsp P1. But the features that separate the specimens of these species were the insertion of the caudal seta II (69.5% in M. ceibaensis vs. 73.2% in M. echinatus); the length and width ratio of the caudal ramus is 3.6 in M. ceibaensis The type specimens labelled as M. dubitabilis (from Trou Caiman, Haiti) and Microcyclops alius (from Barra Sta. Luzia, Uruguay) were morphologically similar to: 1) specimens identified as M. rubellus [including the specimen analysed by Reid (1992)], 2) some specimens from Southeastern Mexico, 3) M. alius from Brazil described by Rocha (1998), and 4) another specimen labelled as M. dubitabilis from Guadeloupe. The micro-structural analysis showed that the following features are common in all of the above mentioned specimens: the number of setae on the endopodal segments of the antenna and the number of rows of spines on the caudal surface of antennal basis; the ornamentation of the setae of the maxillular palp; the shape of the maxilla, and in particular, the structure of the distal coxal endite, the basipodite, and the seta on the claw-like projection; the shape and length of the spine on the inner margin of P1 basis; the shape and ornamentation of the intercoxal sclerites and the inner basis of P1, and P4; the meristic characters of P4, all traits of the caudal rami setae; the structure of P5, the anal somite, and the caudal rami. Therefore, M. alius is considered here as a junior synonym of M. dubitabilis, as it was suggested by Rocha (1998). Also, several specimens recorded under the name M. rubellus in the Americas likely refer to M. dubitabilis.
Microcyclops dubitabilis clearly differs from M. varicans s. str. at least in the next features: the spines at the insertion of caudal seta III are present in M. dubitabilis, but absent in M. varicans; caudal ramus is 3.5-4 times as long as wide in M. varicans and shorter in M. dubitabilis. Medial spine of Enp2P4 is around 0.8 times as long as the segment in M. dubitabilis, whereas that in M. varicans is shorter (around 0.5); and the seta inserted at base of claw-like projection in the maxilla is armed only with strong teeth at its base in M. varicans, but this armament is more complex in M. dubitabilis. Therefore M. dubitabilis is not a synonym of M. varicans.
Microcyclops rubellus and M. varicans have been recorded in several regions of the world and were thought to be likely cosmopolitan (Reid 1992) and highly variable in morphology. This is especially the case for M. varicans (Franke 1989, Alekseev 2002. The type material of M. rubellus and M. varicans is probably lost and both species were originally described from North Europe (Sars 1863, Lilljeborg 1901). Our review of the descriptions and drawings of M. rubellus from some European localities however revealed differences between the European and American specimens here examined in the medial surface ornamentation of the basis of the fourth swimming leg, in the length proportion of the medial apical spine and the Enp2P4, and the ornamentation of the distal margin of the anal somite (see Einsle 1993). Hence, M. rubellus s. str. probably is not distributed in America.

Remarks about the new species
The specimens from southeastern Mexico assigned to the new species M. inarmatus were morphologically similar to that from Laguna Rincon, Haiti identified as M. dubitabilis (SMNK-2391(SMNK- , 2392 and to the specimen examined by Reid (1992) and classified as M. varicans. The shared morphology of the antenna, maxilla, P1 to P5, and urosome is obvious in all of these specimens (figured and described in the descriptive section before).
Microcyclops inarmatus sp. n. can be distinguished from M. varicans by the following characters (see also Rylov 1948, Einsle 1993: M. varicans has a more elongated caudal rami (3.5-4 times as long as wide), the basipodite of P4 bears short spinules on inner margin, there are more setae on the second endopodite of A2, on the distal margin of anal somite the spinules are present ventral, lateral and dorsally; and the medial spine of the second endopod of fourth leg has around the half length of the segment. All these features clearly differ to M. inarmatus sp. n.
The analysis that included specimens from a wide latitudinal range showed a highly conserved morphology primarily in the inner region of each swimming leg and oral appendages. Thus, we may speculate that some reports of M. varicans and M. rubellus recorded in the Americas are in fact M. inarmatus sp. n. and M. dubitabilis, respectively.
Microcyclops inarmatus sp. n. has some similarities also to M. dubitabilis, but the following features differentiate these two species: setal formula of the antennal endopod (1, 9, 7 in M. dubitabilis vs. 1, 6, 7 in inarmatus); the ornamentation on the caudal surface of the antennal basis is less complex in M. inarmatus than in M. dubitabilis; setae on maxillular palp are more armed in M. inarmatus than in M. dubitabilis; and the basal seta inserted at base of claw-like projection in the maxilla is more simple in M. inarmatus, whereas M. dubitabilis has two opposite rows of different spines.
The inner margin of the basis of the first swimming leg has a long spine with heteronomous ornamentation in M. inarmatus, in comparison to the short, homonomously setulated spine on this site in M. dubitabilis. In addition, the inner margin of the basis of the fourth swimming leg bears long hair-like setae and the fourth sclerite is almost as long as wide in M. inarmatus, whereas in M. dubitabilis, this inner margin bears short setae and the fourth sclerite is wider than long. The free segment of the fifth leg has a tiny spine on the medial margin in M. inarmatus (not described in Reid (1992), but clearly observed in the slide USNM-251321), and this spine is absent in M. dubitabilis. Finally, the lateral caudal seta is located near the middle of the caudal ramus in M. inarmatus, whereas in M. dubitabilis, this seta is located in the distal third. In all of the material analysed, no spines were observed at insertion of outermost terminal caudal seta in M. inarmatus, whereas in M. dubitabilis, these spines were observed in every specimen.
Other species of Microcyclops which has 12-segmented antenna, caudal rami with innermost terminal caudal setae longer than outermost terminal caudal setae, spines present only ventrally on the distal margin of the anal somite, caudal rami short (no more than 3 times as long as wide), one spine on inner basis of P1, and the intercoxal sclerite of P4 quadrangular and naked, are M. davidi (Chappuis, 1922) and M. richardi (Lindberg, 1942). Microcyclops inarmatus sp. n. differs from these species in the surface ornamentation of P4 basipodite: long hair-like setules vs. short spine-like setules in M. davidi (sensu Mirabdullayev et al. 2002) and M. richardi (see Lindberg 1942). The genital double-somite in M. inarmatus is short around 0.8 times as long as wide -similar in M. davidi -, but it is elongated in M. richardi, around 1.4 times longer than wide; and the second endopodite of A2 bears 9 setae in M. davidi, but only 6 setae in M. inarmatus.
Additionally, in M. richardi the free segment of P5 has no spine on medial margin, and the medial spine of Enp2P4 is short (0.6 times the length of the segment) in comparison with the new species. Finally, the seta on the base of the claw-like projection of the maxillar basipodite, is armed with tiny spinules in M. inarmatus, whereas in M. davidi it bears strong teeth on its base. Herbst, 1962 and M. anceps var. minor Dussart, 1984 Microcyclops anceps pauxensis and M. anceps var. minor, described from the Amazonian region and Venezuela respectively, are similar in the number of segments of A1 (12-segmented); the length ratio of Enp2P4 (2.35 vs. 2.46); the intercoxal sclerite of P4 with two rows of spines; the length ratio of the lateral and medial apical spines of Enp2P4 (0.64 vs. 0.51); the length ratio of the medial apical spine of En2P4 and the segment (0.7 in both species), the insertion of the caudal seta II (lateral) is at 68% of the caudal rami length in M. a. pauxensis, and 70% in M. a. minor, and the continuous row of spines along the ventral and dorsal margins of the anal somite.

Remarks on M. anceps pauxensis
However, according to Herbst (1962) and Dussart (1984), the inner basis of P1 is naked in M. a. pauxensis, and hairy in M. a. minor; the inner basis of P4 bears small setules in M. a. pauxensis but this is naked in M. a. minor. The ratio between the lengths of the caudal setae VI and III is lower in M. a. pauxensis than in M. a. minor (1.81 vs. 2.52); the ratio between the lengths of caudal seta VI and caudal rami is higher in M. a. pauxensis (1.44) than in M. a. minor (1.07); the length ratio between the free segment of P5 and distal seta of P5 is 0.18 in M. a. pauxensis, and 0.34 in M. a. minor. Other differences between both species were observed in the length ratio between the dorsal caudal seta and caudal ramus (1.5 in M. a. pauxensis vs. 0.57 in M. a. minor), and the spines on the base of the outer caudal seta (spinules present in M. a. pauxensis vs. absent in M. a. minor). All of these differences suggest that M. a. minor is distinct from M. a. pauxensis.
Therefore, these taxa may constitute different species. The evaluation of mouthparts and the ornamentation of the inner region of each swimming leg in the type material may facilitate species delimitation. Unfortunately, the type material of both "subspecies" was not available to us.

Conservative characters among species
Based upon morphological and morphometric features, eleven species and two subspecies of Microcyclops recorded in America were recognized. The following set of characters distinguishes between species: the ornamentation of the caudal surface of the antennal basis; the ornamentation of the setae of the maxillular palp; the shape and armature of the distal coxal endite of maxilla; and the basal seta in front of the claw-like projection of the maxillar basis. Previously, similar structures have been useful for differentiating other Cyclopinae species, such as Mesocyclops (Van de Velde 1984a, 1984b, Hołyńska 2000.
Among the specimens examined, the organization of the spine pattern on the antennal basipodite is similar to that proposed by Van de Velde (1984b) for Mesocyclops which is more complex on the caudal side than on the frontal side. Additionally, the caudal surface ornamentation of the antennal basis in Microcyclops here examined is similar to that in most New World Mesocyclops: the simple ornamentation pattern found in Neotropical Mesocyclops was considered by Hołyńska (2000) and Wyngaard et al. (2010) as an ancestral state. The pattern observed in Microcyclops is much less complex in comparison to those reported for some eucyclopinae species from the genus Macrocyclops (Karanovic and Krajicek 2012), Paracyclops (Karaytug 1999) and Eucyclops (Alekseev et al. 2006, Mercado-Salas et al. 2015. The micro-structures of the swimming legs as diagnostic characters have been explored in Mesocyclops. In Mesocyclops, the coxal and basis armament of the first and fourth trunk limbs are important (Van de Velde 1984a, 1984b. In Eucyclops, the coxal seta of P4 or the intercoxal sclerites of all trunk limbs are informative (Alekseev et al. 2006). Our results show that features such as the medial surface ornamentation of basis of all four legs, the shape and ornamentation of the sclerites of P1 to P4, the presence/ absence or length and armature of the spine on the inner basis of P1, and the shape or armature of the free segment of P5 were useful for differentiating between species.
Important diagnostic morphometric features for Microcyclops were the relative position of the lateral seta on the caudal ramus; the relative length of the outermost terminal caudal seta (III) and the outer median terminal caudal seta (IV); the relative length of caudal seta III and the inner median terminal caudal seta (V); and the length: width ratio of caudal ramus. Traditionally, the length ratio of the second endopod and its apical spines of the fourth trunk limb have been used as features to separate species of Microcyclops; however, as in another genus such as Eucyclops or Mesocyclops, the surface micro-structures together with the integumental armature and the meristic characters of the caudal rami are more informative.
This study is the first attempt to clarify the taxonomy of the species of Microcyclops recorded in America using detailed morphological analysis.

Conclusion
The microscopic analysis of oral and thoracic appendages facilitated better delineation of Microcyclops species recorded in America. The characters that better distinguish between species are the ornamentation of antennal basipodite, the armature of the coxal endite and basipodite of the maxilla, the surface ornamentation of the inner basis of P1, the structure of intercoxal sclerites of the trunk limbs, the length: width ratio of caudal ramus, the length proportion of the caudal setae, and the relative position of the lateral seta on the caudal ramus.
The analysis performed here show that M. alius is a junior synonym of M. dubitabilis, and support the opinion about the synonymy of M. ceibaensis and M. diversus.
Microcyclops inarmatus sp. n. can be distinguished from other known species of the genus by the unique combination of several characters such as: morphometric characters of the second endopodite of fourth trunk limb and caudal ramus, presence of 6 setae on the second endopodal segment of antenna, antennal basipodite with just one group of spinules on caudal surface, lack of ornaments on the intercoxal sclerites of all swimming appendages, absence of spinules at base of lateral caudal and outermost terminal caudal setae, and basipodites of first to fourth swimming legs with long hairlike setules on inner margin.

Key to the American species of Microcyclops (females)
The key is mainly based on the analysis performed in the descriptive section of this manuscript. Original descriptions were consulted in those species in which no microscopic observations could be made [Microcyclops anceps pauxensis (Herbst 1962); M. anceps var. minor (Dussart 1984); M. mediasetosus (Dussart and Frutos 1985); M. pumilis (Pennak and Ward 1985); and M. medius (Dussart and Frutos 1986)]. 1 Cylindrical free segment of P5 smooth, without inner spine (Fig. 7I-K)  Inner basis of P1 with hair-like setae, medial spine absent (Fig. 14B); inner basis of P4 hairy (Fig. 14D) Anal somite with a row of spines on ventral margin; no spines at the bases of the caudal setae II and III (Figs 5A, C-E); intercoxal sclerites of P1 to P4 unarmed ( Fig. 4A-D); basipodite of P4 with long hair-like setae on inner margin (Fig. 2D); one group of spines on caudal view of antennal basis ( Fifth pediger with spines on ventral and lateral surfaces (Fig. 13D); caudal ramus is 5.9 ± 0.4 times longer than wide (Fig. 13E); inner basis of P4 with heteronomous ornamentation: short spine-like plus long hair-like setae (Fig.  13C); caudal surface of antennal basis with two rows of long spines next to exopodal seta (Fig. 12B)  Fifth pediger nude ventrally and laterally; caudal ramus is 3.6 ± 0.4 times longer than wide (Fig. 11C); inner basis of P4 with homonomous ornamentation: strong spine-like setae (Figs 10I, K); caudal surface of antennal basis without spines next to exopodal seta (Fig. 9F)  W ratio of caudal ramus is 3.7 ± 0.3 (Fig. 17B); inner basis of P1 hairy (Fig. 16A); inner basis of P4 with long spine-like setae (Fig. 16D)