Identification guide to some Diaptomid species (Crustacea, Copepoda, Calanoida, Diaptomidae) of “de la Plata” River Basin (South America)

Abstract An identification guide is presented for species of calanoid copepod family Diaptomidae from “de la Plata” River Basin (Argentina, Brazil, Bolivia, Paraguay and Uruguay). It was based on material collected during the summer and winter of 2010 from 43 sites across the eastern part and the lower stretches of this basin, the second largest in South America and the fourth in the world. The guide contains identification keys and species diagnoses for males and females, richly supported by scanning electronic micrographs and/or line drawings of 19 species. It also includes some general remarks on the taxonomy and phylogenetic relationships of these species. The key was adjusted to be useful for these species only, with separate keys for each sex, and is the first for females of South America. One species classified herein as incertae sedis was not included in the analysis. At least ten other species have previously been recorded in the basin but were not present in our samples. This is the first attempt to compile comprehensive taxonomic information on this group of copepods in this region, and it is expected to become a useful tool for biologists and young taxonomists interested in the crustacean biota of the Neotropical region.


Introduction
Copepods are typically small crustaceans that are widely distributed in virtually all aquatic habitats from hot springs to glacial meltwater pools, and from deep ocean trenches to high altitude lakes (Boxshall and Halsey 2004). In terms of abundance, copepods are the dominant group in the zooplankton community both in the oceans and in continental waters, where they share the role of mass and energy transfer between producers and higher level consumers with rotifers and other crustaceans, such as cladocerans (Margalef 1983).
In continental or inland waters, copepods are predominantly represented by three orders: Calanoida, Cyclopoida, and Harpacticoida. In general terms, members of the Calanoida tend to be planktonic forms whereas the Cyclopoida are represented by planktonic as well as littoral, epibenthic and benthic species. Harpacticoids are essentially benthic and interstitial. We use the term planktonic to include the potamoplankton, referring to organisms found in rivers.
All the calanoids collected during our study belong to the subfamily Diaptominae, the most diverse taxon within the family Diaptomidae (Boxshall and Defaye 2008). Recent phylogenetic analyses of the order Calanoida (Bradford-Grieve et al. 2010;Blanco-Bercial et al. 2011) continue to classify the Diaptomidae within a large superfamily (referred to either as the Centropagoidea or the Diaptomoidea) which is placed in a relatively basal position within the order, as in most schemes published since the widely adopted system proposed by Andronov (1974).
In South America, copepods of continental waters began to be studied from the 1890s, and several major works appeared in the 1920s and 1930s (e.g. Wright 1927, Brehm 1933. Since 1950 there has been greater continuity in studies (such as those by the Argentinean researchers Raul Ringuelet and Juan Paggi), and it is now easier to highlight the gaps in our knowledge of the regional copepod fauna. The literature on the inland water copepods of Ecuador, Peru and Bolivia is relatively limited, while in other countries, like Argentina, Brazil, Venezuela, and Chile, there are large geographical areas that remain unsurveyed and the information remains patchy for some large regions due to under-sampling. Biogeographical analyses of South and Central American calanoids can be found in Menu-Marque et al. (2000), Suárez- Morales (2003), and Suárez- Morales et al. (2005).
Although some of the studies were made many years ago, we found that the following papers, used jointly, are quite useful for identifying diaptomids in the region: Wright (1927Wright ( , 1937Wright ( , 1938, Brehm (1933), Paggi (1976Paggi ( , 2001, , and Santos-Silva et al. (1989,1999,2013). Based on these and other studies, traditional morphological characters with taxonomical relevance (i.e. the structure of the fifth leg of both sexes, the armature of the geniculate right antennule (A1R) of males, and the features of the female genital double-somite, among others) were used in this study since they enable accurate, reliable identifications. The terminology adopted for these structures was based on Huys and Boxshall (1991) and Santos-Silva et al. (1999). This study aims to provide identification keys and species diagnoses for males and females of the known species of the calanoid copepod family Diaptomidae from "de la Plata" River Basin, based on material collected during the summer and winter of 2010 from 43 sites across the eastern part and the lower stretches of this basin. As the original descriptions are generally published before 1950, and the illustrations often lacked details, the 19 species identified were redescribed, focusing mostly on the morphology of the male, and illustrated, including scanning electronic micrographs and/or line drawings. Some general remarks on the taxonomy and phylogenetic relationships of these species are added. This guide will help to complete the information on the rich diaptomid fauna present in this basin, in South America, and to update the taxonomy of the species.

Materials and methods
Net zooplankton samples were taken across "de la Plata" river basin in lotic stretches and in reservoirs (upstream to dam -intermediate zone -, and close to dam -lentic zone) ( Figure 1 and Table 1) in two periods during 2010, summer (January and February) and winter (June and July).
The samples were collected by vertical hauls with a conical plankton net with a 68 µm mesh, equipped with an anti-reflux adaptation following Heron (1965, 1967). The samples were fixed in 4% formalin solution for qualitative analysis and with 2.5% glutaraldehyde for scanning electron microscopy (SEM).
Back in the laboratory, the copepods were sorted and analysed under a stereo-microscope (Zeiss Stemi SV6 and Zeiss Discovery V20), or a binocular microscope (Zeiss Standard 20 and 25). Copepods were dissected with fine needles and semi-permanent slides were made in glycerine or in 70% lactophenol. Only adults were analysed. Table 1. List of sampling sites. For each site a code is provided based on a 3 or 4 letter abbreviation of the name of the reservoir or river, and a 1 letter suffix. The suffixes used for reservoirs are: U = upstream and D = dam, and for rivers: H = high stretch, M = middle stretch, L = low stretch). Country abbreviations are: ARG = Argentina, BOL = Bolivia, BRA = Brazil, PAR = Paraguay, and URU = Uruguay. Geographical coordinates and altitude (m.a.s.l.) are also given. Figure 1. Map of sampling sites in de la Plata river basin. Reservoirs sampled included the first and the last in each regulated river as well as dam-free lotic stretches. Symbols differentiate between sites with water retention times (WRT) greater than (○) or less than (•) 100 days.
Copepods were identified with the aid of a number of publications dealing with the taxonomy of Diaptomidae (Wright 1927, 1939, Brehm 1933, Kiefer 1933, Ringuelet and Martinez de Ferrato 1967, Brandorff 1973, Dussart 1984, Dussart and Frutos 1985, Santos-Silva et al. 1989, Santos-Silva 2000Cicchino et al. 2001, Paggi 2001, and Previattelli 2006, and the identity of problematic taxa was confirmed by direct consultations with specialists. This guide includes only those species that had been identified with certainty. All The keys presented in this guide were constructed using only differential characters that emerged as being consistently useful for the identification of genera and species. Similarly, the diagnoses were designed to focus on the basic features traditionally used to characterize diaptomid genera and species, as provided in the specialized taxonomic literature. The morphological structures used in this study are illustrated for both males ( Figure 2) and females ( Figure 3). The terminology employed for each structure and the abbreviations used in the text are given in Tables  2 and 3, which complement Figures 2 and 3. Throughout this guide these listed abbreviations have been used to make the text less repetitive and more concise. We followed the morphological nomenclature proposed by Santos-Silva et al. (1999), Previattelli (2006) and Paggi (2006). The abbreviations for sampling sites (Figure 1, Table 1) are also used throughout.
Morphological structures were illustrated using phase contrast microscopy (Zeiss Standard) with the aid of a drawing tube. The pencil drawings were inked in nankeen ink, then scanned and corrected for smudges and other imperfections in Adobe Photoshop 7.0 in order to obtain high-quality illustrations.
We remind users of the keys that many structures illustrated in Figures 2 and 3 can be well developed in some species but absent in others. For example, the dorsal rows of spinules on the distal margin of the pedigers are present in some species but absent in  (Table 2) B right fifth leg (P5R), illustrating characters B1-10 (Table 2) C Right geniculate antennule (A1R) illustrating characters C1-11 (Table 2) D Basal part of fifth legs (P5), illustrating characters D1-8 (Table 2) E habitus in dorsal view, illustrating features of body tagmata and caudal rami ( Table 2,  others. Similarly, the spinous processes that are typically present on segments 14, 15, 16 and 20 of the male A1R can be absent in some species. The scanning electron microscopy (SEM) was carried out in the "Electron Microscopy Center (CME)" of the University of the State of São Paulo-UNESP -Botucatu, Brazil. Material was prepared by packing each sample of individuals in hollow cylindrical polyethylene compartments, within which the copepods were washed, fixed and dehydrated. Washing was performed using 0.1 M phosphate buffer at pH 7.3 (3 washes, each for 5 min). After washing specimens were immediately fixed by immersion in 0.5% osmium tetroxide (in water) for 20 min. Dehydration was performed  (Table 3) C Second and third exopodite segment and terminal claw of right fifth leg (P5R), illustrating characters C1-6 (Table 3) D habitus in dorsal view, illustrating features of body tagmata and caudal rami (Table 3, characters D1-8) E Dorsal posterior part of prosome and genital double-somite, illustrating characters E1-10 (Table 3) F Posterior part of prosome and genital double-somite, illustrating characters F1-11 (Table 3).
progressively via a graded series of ethanol as follows: 7.5%, 15%, 30% and 50% (two changes at each concentration for 5 min), then 70% (3 changes each for 10 min), and 90% and 100% (2 changes at each, for 5 min). Subsequently, the material was critical point dried in a BALZERS UNION CTD-020 equipment, using liquid carbon dioxide as the exchange medium. After drying, specimens were dissected when necessary to reveal diagnostic structures, and attached to stubs using adhesive tape. Sputter coating with 15 nm layer of gold was carried out in a BALZERS UNION MED-10 coater. Observations were made on a Philips SEM-515 microscope and images were edited in Photoshop 7.0 (Adobe).  Table 3. Key to female characters numbered in Figure 3. Numbers are given together with the name of the morphological structure and its abbreviation (Abbrev. Sensillae on both swellings on GS directed perpendicular to body axis; right margin of GS with small rounded lobe at posterior extremity ( Adult female, body length (excluding caudal setae) 1851 µm. Dorsal surface of Ped4 and Ped5 without spinule rows; complete suture present between Ped4 and Ped5; lateral wings slightly asymmetrical, with right wing larger than left; each wing with two sensillae close to postero-distal corner (Fig. 7A). GS asymmetrical, about 1.6 times longer than wide; anterior part weakly swollen, right side more swollen than left (  Remarks. In identifying species of Diaptomidae in general, caution must be exercised in using the shape of segment 20 of A1R for species identification because it is often variable within a population. This segment has a falciform process at the distal angle in several species (Paggi 1976) but, as  points out, in a sample of  Ultrafine-scale ornamentation characters might prove to be valuable in future comparative studies, including 1) the presence of a pore on segment 20 of male A1R (Fig. 5F), 2) the presence/absence of spinules of the basepodite of the A2B in both sexes (Figs 5H, 8C), 3), the presence of a pore on the outer surface of the syncoxa, and of spinules on the distal syncoxal endite of the maxilliped (Figs 5I-K, 8B), and 4) details of the spinulation on the terminal setae of ExpP4 (Fig. 8G). In the absence of comparable SEM-based data on other species, such data cannot be used for routine species discrimination at present. This species is widely distributed, extending from northeastern Brazil to the Itaipu Reservoir at the end of the upper Paraná River in southern Brazil (Fig. 9). Typically, few individuals are found in any one sample, so this is not a common species. However, among Argyrodiaptomus, this species is rarely misidentified, given its large body size and distinctive P5.
Remarks. The male illustrated was collected from the Salto Grande reservoir (SGRA-D), Uruguay River. The female illustrated here belongs to the collection of the   Museo Argentino de Ciencias Naturales (Buenos Aires) (MACN-In 29733), and was examined because of the scarcity of females in the samples from de la Plata river basin. This species can be readily identified because of the distinctive structure of its P5 and the presence of dorsal rows of spinules on the male pedigers, which are lacking in other species of Argyrodiaptomus.
This species is common in the lower regions of the basins of the Paraná and Uruguay rivers (Figure 14), and has been reported in several studies, particularly by Argentinian researchers. Santos-Silva (2008) recorded this species in southern Brazil, in the state of Rio Grande do Sul. Previattelli et al. (2013) included a record of this species from Bolivia, a distance of about 1,300 km north of the previous northernmost record. It is possible that the distribution of this species extends beyond this northern limit to at least the line of latitude 35°, but verification is needed.
Adult female, body length 1648 µm. Dorsal surface of pedigers lacking ornamentation of spinules ( Fig. 18A-C). Complete suture present between Ped4 and Ped5; lateral wings slightly asymmetrical (Fig. 17A); both wings with two sensillae each, one at distal corner. GS weakly asymmetrical, approximately 1.5 times longer than wide (Fig. 18B). Anterior of GS slightly swollen. P5 symmetrical, with small conical process at outer distal corner of Cx, bearing short triangular sensilla with bifid apex (Fig. 17B). BspP5 with short outer seta, less than half length of outer margin of Exp1P5. EnpP5 2-segmented but with incompletely-expressed transverse articulation and longitudinal groove (Figs 17C, 18D, E). ExpP5 3-segmented; lateral spine of Exp2P5 reaching about to midlength of external margin of Exp3P5 (Fig. 18F); external seta of Exp3P5 about 3/4 length of internal seta; internal seta about half length of terminal claw (Fig. 18G-H). Remarks. The specimens illustrated here were caught in the middle stretch (RPAR-M2) of the Paraná River, and the sample contained only a few females and two males. Paggi (2006) provided a redescription and confirmed the validity of A. falcifer, recognizing at the same time that Argyrodiaptomus argentinus (Wright, 1938) should be treated as a junior subjective synonym of this species. Paggi also summarized its geographical distribution across Argentina and Paraguay. Previattelli (2006) also illustrated this species, but he considered this taxon under the binomen Argyrodiaptomus argentinus.
This species, together with others of this genus, can be very abundant in small scale habitats, such as water pools, and is generally scarce in lotic environments. It was found at only three stations in the present study (Fig. 19).  both wings with pair of sensillae, located at distal corner. GS asymmetrical, about twice as long as wide; slightly swollen anteriorly, more on left side than on right (Fig. 22A). P5 symmetrical, with small conical process at outer distal corner of CxP5 bearing short, triangular sensilla (Fig. 22B). BspP5 with short outer seta (Fig. 22C), measuring about 70-80% of external margin of EnpP5. EndP5 2-segmented, although suture only visible from certain angles (Figs 22C, 23C-E). ExpP5 3-segmented; lateral spine of Exp2P5 short, not exceeding length of external margin of Exp3P5; external seta of Exp3P5 about ¼ (25%) length of internal seta; internal seta about 2/3 length of terminal claw (Fig. 23E).
Remarks. The figured specimens were collected from the Iguaçu River at the reservoir of Foz do Areia (FARE-D) (Figure 24). This species was originally described from material collected in the state of São Paulo and, according to Wright (1927), occurs in Brazil (south and southeastern, and possibly also in the mid-west) and in Argentina.   (2008) and Previattelli et al. (2013) cited several records of A. furcatus from south and southeastern Brazil. Together these studies confirm the occurrence of this species mainly in reservoirs of southeastern and southern Brazil, and in northern Argentina. Along with other large species of this genus, this species can co-occur with A. azevedoi (in southeastern Brazil) and with Notodiaptomus spiniger (Brian, 1925) (in southern Brazil and northern Argentina). Argyrodiaptomus furcatus tends to be closer to N. spiniger in body size, whereas A. azevedoi is larger.
Adult female, body length 1309 µm. Single row of spinules present along posterior margin of Ped3; irregular rows of spinules present, marking plane of posterior margin of Ped4; discrete transverse row of minute spinules present across middle of Ped5 (Fig. 27A). Lateral wings asymmetrical, similar in size but left wing located more anteriorly and right wing positioned and directed more posteriorly; both lateral wings with single sensilla at apex (Fig. 27A). GS nearly symmetrical, no more than 20% longer than wide; dilated anteriorly with left and right swellings of similar size, left sensilla on apex of swelling and aligned perpendicular to longitudinal axis of body; right sensilla directed postero-laterally. P5 symmetrical (Fig. 27B) with conical process at outer distal corner of CxP5 bearing strong triangular sensilla at apex. BspP5 with long outer seta, extending beyond middle of outer margin of Exp1P5. EnpP5 2-segmented, not reaching middle of inner margin of Exp1P5. ExpP5 3-segmented; lateral spine of Exp2P5 about as long as outer margin of Exp3P5; external seta of Exp3P5 less than 1/4 (25%) length of internal seta; internal seta almost reaching tip of terminal claw.
Remarks. The specimens illustrated were collected from the lower Paraná River (RPAR-L2). This species appears to be widely distributed in the south of the basin, from the Iguaçu River into more temperate conditions (Fig. 28).  Paggi (2001) provided good illustrations of this species and, in the same work, described a new species, N. dentatus Paggi, 2001, resembling and potentially confused with N. anisitsi. These two species deserve further investigation, as they may be part of a larger complex of species with very similar morphology.

Notodiaptomus carteri (Lowndes, 1934) Figs 29-31
Diaptomus carteri Lowndes, 1934 Diagnosis. Adult male, body length 1484 µm. Modified seta on segment 13 of A1R reaching distal end of segment 14 (Fig. 29A). Right BspP5 with longitudinal groove in surface, ornamented with small surface granulations (Fig. 29C). Right Exp2P5 with slightly outwardly curved lateral spine inserted on distal margin of segment, length  about 1/3 of terminal claw; insertion of lateral spine separated from base of terminal claw by gap barely greater than basal width of spine (Fig. 29B, C).
Adult female, body length 1770 µm. Ped4 and Ped5 separated by incomplete suture; process present on mid-dorsal surface of Ped4 (Fig. 30A, C); lateral wings slightly asymmetrical, with two pairs of sensillae on each side; internal sensilla of right side thin, setule-like; left sensilla slightly larger than right. GS asymmetrical, about 1.5 times longer than wide, swollen anteriorly (Fig. 30C), swelling on left side greater than that on right margin; left side swelling hemispherical, bearing single posterolaterally-directed sensilla; swelling on right side with projecting semi-circular lobe with posteriorly-directed sensilla. Right distal margin of GS longer than left, each side with rounded process at posterior end. P5 symmetrical ( Fig. 30B) with small conical process at distal corner of Cx bearing short and triangular sensilla, about as long as wide. BspP5 with outer seta barely reaching middle of outer margin of Exp1P5. EnpP5 onesegmented, reaching 2/3 distance along inner margin of Exp1P5. ExpP5 3-segmented; lateral spine of Exp2P5 short, about as long as outer margin of Exp3P5; external seta of Exp3 about half length of internal seta; internal seta reaching middle of terminal claw.
Remarks. The illustrated specimens were collected from the lower Paraná River (RPAR-L2). The body length is slightly larger than the known range for males, 1315 to 1439 µm, given by Ringuelet and Martínez de Ferrato (1967). These authors published the first record from Argentina, near Santa Fe (equivalent to site RPAR-M3 in this study, Fig. 1), and they associated this species with the presence of floating aquatic macrophytes. The original description of this species (Lowndes 1934) was based on material from Paraguay, collected in flooded regions and wetlands, typically where there are macrophytes. In the present study the records of this species (Fig. 31) were about 500 km downstream from the locality sampled by Ringuelet and Martínez de Ferrato (1967). Santos-Silva (2008) provided records of this species from Paraguay, Argentina and southern Brazil, in the Patos Lagoon. Ringuelet and Martínez de Ferrato (1967) provided physical and chemical data for the water in the ponds where they found this species. Given the range of sites reported in this study, we infer that this species can occur in a wide range of habitat types.
Females of N. carteri can be readily distinguished from congeners by the asymmetrical swellings of the genital double-somite.
Adult female, body length 1485 µm. Ped4 and Ped5 fused, without trace of suture; lateral wings symmetrical, both with two pairs of sensillae (Fig. 33A), one at distal tip, the other close to inner margin of wing. GS almost symmetrical, 1.3 times longer than wide; slightly dilated anteriorly, with swelling on right side slightly more marked than on left, both swellings with apical sensilla; left sensilla aligned perpendicular to longitudinal axis of body, right sensilla directed slightly anteriorly. P5 symmetrical (Fig. 33B), with short conical process at distal corner of Cx bearing triangular sensilla, longer than wide; BspP5 with long outer seta, almost reaching distal margin of Exp1P5; EnpP5 2-segmented, about as long as internal margin of Exp1P5. Exp 3-segmented; lateral spine of Exp2P5 similar in length to external margin of Exp3P5; external seta of Exp3 1/3 as long as internal seta; internal seta almost reaching apex of terminal claw.
Remarks. The illustrated specimens were collected from the Parnaíba River at Emborcação Reservoir (Fig. 34), where this species co-occurred with at least two other diaptomids: A. azevedoi and N. iheringi . This species has been widely reported from across the north and northeast of Brazil and in Venezuela. In de la Plata river basin, Tundisi and Matsumura-Tundisi (1994) found this species in Barra Bonita Reservoir. This species can be confused with N. iheringi, but differs in the setal formula of some segments of the A1, in the absence of a row of spinules on the first segment of A1R, and in having a smaller body than N. iheringi.

Notodiaptomus conifer (Sars, 1901) Figs 35-38
Diaptomus conifer Sars, 1901 Diagnosis. Adult male, body length 1548 µm. Segment 1 of A1R with spinule row (Fig. 36F, G); segment 15 typically with well-developed spinous process reach-  ing to end of segment (Figs 35A, 36A, B, E); process sometimes absent; segment 20 of A1R lacking distal projection. Enp1 of A2 ornamented with spinule row and single pore (Figs 35G, 36J). Right BspP5 with longitudinal fissure ornamented with surface granulations (Fig. 36H, I); Right Exp2P5 twice as long as wide (Figs 35B, 36C, D); lateral spine positioned close to outer distal angle of segment and directed posteriorly relative to longitudinal axis of body, length of lateral spine less than width of segment; terminal claw long and slightly curved (Fig. 35C-F), more slender than in N. cearensis. Adult female, body length 1734 µm. Ped4 and Ped5 separated by complete suture; lateral wings slightly asymmetrical, each wing with sensilla at apex (Fig. 37A). GS asymmetrical, about 1.5 times longer than wide; slightly dilated anteriorly, with swelling on left side larger than on right, left swelling hemispherical with sensilla directed slightly posteriorly; right swelling more pronounced, sensilla located on dorsal surface, not on lateral margin; right lateral margin of GS with small projection about at 2/3 length and with small notch at posterior border (Fig. 37A). P5 symmetrical (Fig.  37B), with small expansion at outer distal corner of Cx, bearing large robust, triangular sensilla, approximately 1.5 times longer than wide. BspP5 with long outer seta, almost reaching distal margin of Exp1P5. EnpP5 with incomplete suture, similar in length to inner margin of Exp1P5. Exp 3-segmented; lateral spine of Exp2P5 similar in length to external margin of Exp3P5; external seta of Exp3P5 about 1/5 (20%) length of internal seta; internal seta attaining 3/4 length of terminal claw.
Remarks. The illustrated specimens were collected from the upper Tiete River, at the Barra Bonita Reservoir. In addition to the well-developed spinous process on segment 15 of A1R of the male, N. conifer can be distinguished from other congeners like N. iheringi and N. cearensis, by its larger body size. These species also differ in their relative lengths of the lateral spines on the right Exp3P5 of the male.
Earlier studies (Nogueira 2001, Nogueira et al. 2008Matsumura-Tundisi and Tundisi 2003), which sampled several rivers in São Paulo State (Brazil) between 1970 and 2005 noted changes in the abundance of N. conifer. Matsumura- Tundisi and Tundisi (2003) suggested that such changes might be in response to variations in the ionic concentrations resulting from decreasing water quality and increasing conductivity over the surveyed period. Notodiaptomus conifer was found in the current study (based on samples taken in 2010) in only two reservoirs (JUR and BBON) (Fig. 38), and in both of these Matsumura- Tundisi and Tundisi (2003) had indicated that the species had disappeared. Studies on long time series (e.g. Polli and Simona 1992) have demonstrated cycles of 25 years for some diaptomid species, over which they dramatically decrease in population density but subsequently recover to become dominant again. It is necessary to sample extensively before putative disappearances can be confirmed.
Notodiaptomus conifer has been reported from Argentina to the northeast of Brazil, thus suggesting a widespread but scattered occurrence across South America.
Adult female, body length 1411 µm. Suture between Ped4 and Ped5 incomplete; conical dorsal process present on Ped4 (Figs 41A, 42A, B); lateral wings symmetrical, with two pairs of sensillae, larger pair on apex of each wing (Fig. 42B), each about twice as long as wide; inner pair of sensillae located dorsally near internal margin of each wing, each about as long as wide. GS asymmetrical, approximately 1.7/1.8 times longer than wide; anterior part slightly dilated, each anterior swelling with apical sensilla; left sensilla distinctly curved posteriorly, right sensilla aligned perpendicular to longitudinal body axis (Fig. 41A). P5 symmetrical with small conical process at outer distal corner of Cx bearing short, robust and triangular sensilla (Fig. 41B). BspP5 with external seta almost reaching middle of external margin of Exp1P5. EnpP5 2-segmented, almost reaching end of inner margin of Exp1P5; EnpP5 with 2 strong unequal apical spines. ExpP5 3-segmented; lateral spine of Exp2P5 almost as long as external margin of Exp3P5; internal seta on Exp3P5 about 3.5 times longer than external seta; internal seta reaching middle of terminal claw.
Remarks. The specimens illustrated here were collected in the upper Paraguay River. This is the only calanoid species that was found throughout the Paraguay basin (Fig. 43). Frutos et al. (2006) previously reported this species from along the Paraguay River. We did not find it in the basin of the Paraná River or the Uruguay River but it was previously observed in the middle section of the Paraná River by Dussart and Frutos (1985). Santos-Silva (2008) mentioned a record from the Itaipu Reservoir in the upper Paraná River basin, and other reports indicate its presence in the lower Paraná River, near the delta (e.g. Ringuelet 1958). The type locality of this species is in the Amazon (Wright 1927).
This species has a wide distribution in rivers and associated systems like marginal ponds, but it is not typically recorded in reservoirs. This species is easily distinguishable by the position of the lateral spine on a lobe on the coxa of P5 and by the length of spinous processes on segments 13 and 15 of the male A1R. Some studies have reported this species under the name Notodiaptomus coniferoide (sic.) (see ), but the name presented in the original description by Wright (1927) is N. coniferoides.  made by previous authors revealed some distinctive features. However, we consider that the specimens from Venezuela identified as N. coniferoides by Dussart (1984) are in fact N. simmilimus, based on the differences in the proportions of BspP5 and the size of the modified seta on segment 13 of A1R. Dussart (1984) and Cicchino et al. (2001) did not refer to differences in the size of the modified seta on this segment, which tends to be smaller in N. simmilimus than in N. coniferoides. In the Amazonian specimens of N. coniferoides, the modified seta reaches the end of segment 14, according to the original description (Wright 1927), whereas in N. simmilimus it only extends to the middle of this segment.
We recommend a thorough comparative analysis of specimens of N. coniferoides found in the south of Brazil and in the lower Parana River. Comparison of our N. coniferoides with Amazonian specimens indicates that the material described in the present study was relatively smaller and we consider it is necessary to confirm the identity and status of N. coniferoides in de la Plata River Basin.
Adult female, body length 1335 µm. Incomplete suture present between Ped4 and Ped5, with spinule row marking plane of fusion (Fig. 45A); row of spinules present dorsally along posterior margin of Ped4 (Fig. 45A); lateral wings of similar size  but slightly asymmetrical; two sensillae, one large and one small, present on each wing; apical sensilla on left side located on small projection and with apex directed posteriorly; right sensilla located at apex of projection directed perpendicular to body axis. GS asymmetrical (Fig. 45A), approximately 1.9 times longer than wide. Anterior part slightly dilated, left swelling larger than right one; left side hemispherical, with strong sensilla at apex of swelling, about 1.5 times longer than wide; right side ornamented with sensilla at apex of triangular projection, sensilla about twice as long as wide; posterior to sensillate swelling, external margin sinuous in dorsal view, ending in welldeveloped lobe at posterior end of right side. P5 symmetrical (Fig. 45B), with small conical process at outer distal corner of CxP5 bearing triangular sensilla, about 1.4 times longer than wide. BspP5 with long outer seta long, reaching almost to end of outer margin of Exp1P5. EnpP5 one-segmented, length exceeding middle of inner margin of Exp1P5. Exp 3-segmented; lateral spine of Exp2P5 just reaching end of external margin of Exp3P5; external seta of Exp3P5 about 1/4 (25%) length of internal seta; internal seta extending almost to tip of terminal claw.
Remarks. The illustrated specimens were collected from the middle stretch of the Paraná River (Fig. 46, RPAR-M3). The body length of the sampled specimens is slightly less than that reported by Paggi (2001) but falls within the known range for this species. Paggi (2001) highlighted two important characteristics of this species: 1) the presence of sclerotized processes on the surface of the right Exp2P5 of the male, and 2) the presence of a small tooth-like denticle on the outer margin of external caudal seta, near its base. This tooth-like process provides the basis for the etymology of the species. Diagnosis. Adult male, body length 1123 µm. Dorsal suture between Ped4 and Ped5 incomplete (Fig. 48B); surface of Ped3, Ped4, and Ped5 ornamented posterodorsally and laterally with fine covering of slender spinules (Fig. 48C-D). Lateral wings small, slightly asymmetrical, directed posteriorly; left side larger than right; both lateral wings with short, posteriorly-directed sensilla at distal corner, left sensilla better developed than right (Fig. 48B). GS slightly asymmetrical. Rostrum asymmetrical  ( Fig. 48H). Segment 1 of A1R ornamented with spinule row (Fig. 48E, F); modified seta on segment 13 reaching to end of segment 14 (Figs 47A, 48A, I). Bsp of A2 ornamented with spinules (Fig. 48G, J); Enp1 of A2 with spinule row (Fig. 47E). Cx and Bsp of P1 ornamented with long setules laterally (Fig. 49A). Right CxP5 with conical process bearing sensilla at apex (Figs 47H, J, 49G). Inner margin of right BspP5 with sclerotized knob, covered with tiny granulations (Figs 47C, D, 49C, E, F); right BspP5 with oblique fissure ornamented with small surface granulations (Fig. 47C); external seta of right BspP5 inserted subdistally, short, typically shorter than outer margin of Exp1P5. Right Exp1P5 bearing triangular process distally; right Exp2P5 with internal margin concave proximally, lateral spine slightly curved (Figs 47K, 49D) and inserted sub-terminally on external margin, less than 1/3 length of terminal claw (Fig. 47G-K). CxP5L with small conical process at outer distal corner bearing short apical sensilla; BspP5L with short external seta inserted distally; internal margin of BspP5L slightly concave, bearing two small sclerotized knobs proximally (Figs 47B-D, 49B, G), each ornamented with tiny granulations (Fig. 49C, E, F). EndP5 one-segmented, with row of spinules on inner distal margin (Fig. 47B, F). Exp2P5L with digitiform and sclerotized terminal process, lacking ornamentation. Adult female, body length 1275 µm. Complete suture present between Ped4 and Ped5 (Fig. 50A); lateral surfaces of Ped3, Ped4 and Ped5 ornamented with covering of fine spinules (Fig. 51C); lateral wings more or less symmetrical (Fig. 51A); both wings with two sensillae, larger sensilla located at apex of wing and about 1.5 times longer than wide, smaller sensilla located near posterior margin of wing (Fig. 50A). GS asymmetrical, about 1.9 to 2 times longer than wide; slightly dilated anteriorly, with swelling of similar size on each side; each swelling with sensilla at apex; sensilla on right side of GS located at apex of hemispherical swelling; small lobe located at posterior end of right margin of GS (Figs 50A, 51A). Rostrum symmetrical (Fig. 51B). Fine setules present on segments 1 and 3 to 5 of A1 (Fig. 51E, F). P5 symmetrical (Fig. 50B) with small conical process at outer distal angle of CxP5, with short and strong sensilla, barely longer than wide. BspP5 with long external seta, extending beyond middle of external margin of Exp1P5. EnpP5 extending to middle of internal margin of Exp1P5, with incomplete suture (Fig. 51D). Exp 3-segmented; external seta of Exp2P5 exceeding length of external margin of Exp3P5 and similar in length to external seta of Exp3P5; external seta of Exp3P5 about 40% length of internal seta. Internal seta about 2/3 length of terminal claw.
Remarks. The specimens depicted here were collected in Furnas Reservoir on the Grande River (Fig. 52, FUR-U).  commented on the possible confusion between N. amazonicus  and N. henseni. However, comparisons between the specimens of N. henseni found in this work and N. amazonicus collected in Balbina Reservoir (Uatumã River, Amazonas) showed clear differences in the structure of P5 (Fig. 53), the dorsal spinular ornamentation of the prosomal somites, and in body length. In the present study, specimens of N. amazonicus similar to those obtained in the Amazon region were not found, but N. henseni was found to be widely distributed in the Upper Paraná River. There are significant differences in size range between N. henseni and N. amazonicus (Fig. 54). Notodiaptomus amazonicus males had a body length of 1608 µm and a body width of 397 µm, both significantly larger than N. henseni, whose body length does not exceed 1300 µm. It is possible that some records of N. amazonicus from de la Plata basin are erroneous, reflecting the taxonomic confusion between this species and N. henseni. For example, Ringuelet and Martinez de Ferrato (1967) reported a body length of about 1270 µm for a diaptomid identified as N. amazonicus in Argentina. However, this length is much less than would be expected for this species and falls within the typical size range of N. henseni.
In addition, Ringuelet and Martinez de Ferrato (1967) mentioned that their specimens of N. amazonicus that carried a process on segment 20 of the male A1R were larger (about 1375 µm), a condition which is closer to the expected size of N. amazonicus from the Amazon region, although still relatively small. More comparisons are needed between specimens of N. amazonicus from the Amazon basin and those reported by other researchers from de la Plata River Basin.
Recently, Matsumura-Tundisi et al. (2010) described a new species, Notodiaptomus oliveirai Matsumura-Tundisi, Espindola, Tundisi, Souza-Soares & Degani, 2010, which shows a close similarity to N. henseni, and stated that many records of N. henseni from the state of São Paulo (Brazil) would represent records of N. oliveirai. However, their description is supported by illustrations lacking many important details (e.g. description of A1 formula, dorsal spinules) and the diagnostic difference between these two species as stated by these authors was based only on the curvature of the outer margin of the second segment of the exopod of right P5 (Exp2P5R). We consider this evidence insufficient to justify the establishment of a new species. The form of the P5 appears to vary according to the angle of observation (e.g. Fig. 47G to 47K), and this can be misleading. The degree of curvature alone does not provide adequate evidence upon which to base the establishment of a new species. The typical sclerotized

Notodiaptomus iheringi (Wright, 1935) Figs 55-60
Diaptomus iheringi  Diagnosis. Adult male, body length 922 µm. Posterior margin of Ped3 ornamented with rows of spinules; Ped4 and Ped5 ornamented with spinule rows along posterior margins and on lateral surfaces (Fig. 56B-F). Rostral filaments asymmetrical (Fig. 56A). Single row of spinules present on first segment of A1R (Figs 55D, 56G, H); modified seta on segment 13 of A1R strong, with minutely bifid apex, reaching beyond level of insertion of proximal seta on segment 14 (Fig. 55A); spinous process of segment 15  longer than spinous process of segment 16) (Fig. 57A, B). Enp1 of A2 ornamented with pore and spinule row (Figs 55G, 56I). Cx of P1 with setules on outer surface (Fig. 57E). Right and left internal margins of P5 without sclerotized processes (Figs 55B, C, 57G). Right BspP5 with irregular oblique fissure on middle of surface; distal part of fissure ornamented with small surface granulations; external seta inserted distally on lateral margin. Lateral spine on right Exp2P5 inserted in distal third of external margin, length about 1/6 (16%) of length of terminal claw (Figs 55E, F, 57C, D, F, G). Adult female, body length 1093 µm. Incomplete suture present between Ped4 and Ped5 with plane of fusion marked by transverse row of strong spinules, with double row in middle section of dorsal surface (Figs 58A, 59D); lateral surfaces of posterior prosomal somites ornamented with spinules (Fig. 59F). Lateral wings bearing two unequal sensillae; large sensilla located at apex about 1.5 times longer than wide. GS asymmetrical, about 1.3 to 1.4 times longer than wide; dilated anteriorly, swellings of similar size; left swelling hemispherical; both swellings bearing sensilla approximately 2.5 times longer than wide, right sensilla inserted on dorso-lateral surface and not on apex of swelling (Figs 58A, 59A). P5 symmetrical (Fig. 58B) with small conical process at outer distal corner of Cx bearing short, robust triangular sensilla, about 1.1 times longer than wide. BspP5 with long external seta, extending beyond distal end of external margin of Exp1P5 (Fig. 58B). EnpP5 one-segmented (Fig. 59C, G), about 3/4 length of internal margin of Exp1P5. Exp 3-segmented; lateral spine of Exp2P5 not reaching end of external margin of Exp3P5; external seta of Exp3P5 approximately 3.5 times shorter than internal seta; internal seta reaching just beyond middle of terminal claw (Fig. 59D).
Remarks. Our specimens were taken in the Grande River at Furnas Reservoir. In the present study this species was found in southeastern and southern Brazil and in the upper part of the Paraná River basin, with its southerly distribution boundary represented by the Iguaçu River (Fig. 60). Other studies (Santos-Silva 2008) indicate a widespread distribution in Brazil, and the type locality is in Paraiba State in northeastern Brazil, but it also occurs in some parts of northern Argentina. This species can be confused with its congeners N. cearensis and N. isabelae , but details of the male A1R and P5 are useful to distinguish N. iheringi from these other two species. This species also resembles N. conifer in possessing a small lateral spine on the P5, but N. conifer has a well-developed spinous process on segment 15 of A1R (Fig. 35A), which is much larger than that of N. iheringi (Fig. 55A). In the present study this species occurred in a variety of habitats with different trophic states, from oligotrophic, such as the Emborcação and Furnas reservoirs, to the eutrophic Barra Bonita and Foz do Areia reservoirs. Its presence or absence, therefore, is not indicative of trophic status and this species is not suitable for biomonitoring purposes.
Adult female, body length 1310 µm. Complete suture present between Ped4 and Ped5; transverse row of short spinules present along posterior margin of Ped4 (Fig. 63A); row of long setules, present dorsally on Ped5 (Fig. 64B); lateral wings asymmetrical (Fig. 64A, B), left wing narrower than right, each with two unequal sensillae, large sensilla at apex of wing. Sensilla about 1.5 times longer than wide.  GS asymmetrical, approximately 1.5 times longer than wide: slightly dilated anteriorly, with swelling on right side larger than left; sensilla present at apex of each swelling, slightly on dorso-lateral surface on right side (Fig. 63A). P5 symmetrical (Figs 63B, 64C) with small conical process at outer distal corner of CxP5 bearing long blunt sensilla, approximately 2.6 times longer than wide. BspP5 with long external seta, equal in length to external margin of Exp1P5. EnpP5 approximately 3/4 of length of internal margin of Exp1P5. Exp 3-segmented; lateral spine of Exp2P5 as long as external margin of Exp3P5; internal seta of Exp3 about 2.3 times longer than external seta; internal seta reaching almost to middle of terminal claw (Fig. 64C).
Remarks. This species was collected in the Machadinho Reservoir (MAC-U) on the Uruguay River and it can be easily identified by the unusual form of the caudal setae. It is distributed across the southern part of de la Plata river basin, including southern Brazil, south of the Iguaçu River (Fig. 65), and it also occurs widely in Argentina and Uruguay. In the present study, N. incompositus was among the dominant species and it tended to occur abundantly in reservoirs and river stretches irrespective of their trophic status (eutrophic, mesotrophic, and oligotrophic), as observed for N. henseni. In general, N. henseni was the dominant species in the northern part of de la Plata river basin, while N. incompositus was dominant in the southern sector.

Notodiaptomus isabelae (Wright, 1936) Figs 66-68
Diaptomus isabelae  Diagnosis. Adult male, body length 919 µm. Transverse row of strong spinules present along posterior margin of Ped4 (Fig. 66B, C). Modified seta on segment 13 of A1R extending beyond middle of segment 14, but not reaching end of segment; spinous process present on segments 15 and 16 of A1R (Fig. 66G). Two small expansions present proximally on internal margin of right BspP5 (arrowed in Fig. 66A); right Exp1P5 with wide distal expansion on margin (Fig. 66D, F); right Exp2P5 broad, with width about ¾ length (Fig. 66A, E); lateral spine short, inserted distally, close to insertion of terminal claw. Adult female, body length 1056 µm. Complete suture present between Ped4 and Ped5, ornamented with row of strong spinules dorsally (Fig. 67A); lateral wings asymmetrical, left wing smaller and located anterior to right wing; two pairs of sensillae present on each wing, large sensilla on left positioned on small dorsal projection; large sensilla on right positioned at apex of wing. GS asymmetrical, approximately 2.5 times longer than wide; anterior part slightly dilated, with swelling on right side larger than that on left; left swelling rounded with strong sensilla at apex, approximately 1.5 times long than wide; anterior swelling on right side with strong sensilla at apex, approximately twice as long as wide; anterior swelling extending back along two thirds of lateral margin to acute indentation; margin straight posterior to indentation (Fig.  67A). P5 symmetrical (Fig. 67B) with small conical process at outer distal corner of CxP5, bearing sensilla with rounded apex, barely longer than wide. BspP5 with external seta reaching middle of external margin of Exp1P5. EnpP5 one-segmented, about as long as inner margin of Exp1P5. ExpP5 three-segmented; lateral spine of Exp2P5 almost reaching distal tip of external margin of Exp3P5; external seta of Exp3P5 about one third length of internal seta; internal seta about two thirds length of terminal claw.
Remarks. The specimens were collected in the middle section of the Paraná River at Yaciretá Reservoir (Fig. 68, YACI-D). The type locality of this species is in the state of Pernambuco (northeastern Brazil) but this species is distributed across much of Brazil and Argentina, with its southernmost boundary in the middle region of the Paraná River. It does not seem to occur in the Iguaçu and Uruguay rivers, and the low winter temperatures in this region may be a factor limiting its distribution. Dussart and Frutos (1986) recorded this species in the middle section of the Paraná River.
This species can be easily distinguished from its congeners by the proximal processes on the internal margin of the male right BspP5 and by the shape of the wings and GS of females. A comparison with the description by Paggi (1976) revealed minor differences in the shape of segment 20 of the male A1R. No falciform process was found on this segment in our specimens, but this absence can occur in some populations, and a similar range of variability of this process has been noted for other species (Ringuelet andMartínez de Ferrato 1967, Paggi 1976).

Notodiaptomus santafesinus (Ringuelet & Martínez de Ferrato, 1967) Figs 69-73
Diaptomus santafesinus Ringuelet & Martínez de Ferrato, 1967 Diagnosis. Adult male, body length 967 µm. Dorsal and lateral surfaces of Ped3, Ped4 and Ped5 ornamented with scattered setules (Fig. 70A). Segment 1 of A1R with setule row (Fig. 70H); segment 13 with modified seta reaching beyond middle of segment 14; segments 15 and 16 each with small spinous processes (Figs 69I, 70D); distal margin of segment 20 of A1R terminating in small bifid process (Fig. 69J, K). Tips of rami of P4 ornamented with minute spinule combs (Fig. 70C, G). Left and right CxP5 each with conical process bearing sensilla at apex (Figs 69A, G, 70B, E). BspP5L with smooth inner surface, lacking spinulation or processes (Figs 69A, 70E). Right BspP5 with smooth inner surface, lacking spinulation or processes (Fig. 69A), outer seta short, less than half length of external margin Exp1P5; lateral spine on right Exp2P5 inserted close to insertion of terminal claw, strong and slightly outward curved, approximately four times longer than wide (Figs 69B-H, 70F); terminal claw with main curvature in proximal 1/3. Adult female, body length 1271 µm. Incomplete suture present between Ped4 and Ped5; lacking rows of spinules on posterior margin of pedigers (Fig. 71A). Dorsal process present on midline of Ped4 (Fig. 71A, C); lateral wings slightly asymmetrical; both projections with pair of sensillae, one large and one small. GS slightly asymmetrical, about 1.8 to 1.9 times longer than wide; anterior part slightly dilated, with sensilla at apex of each swelling; left sensilla positioned slightly anterior to right; left sensilla directed slightly posteriorly, right sensilla directed slightly anteriorly. P5 symmetrical (Figs 71D, 72A) with small conical process at outer distal corner of CxP5, bearing sensilla approximately 1.5 times longer than wide; BspP5 with external seta of medium length, not exceeding length of internal margin of Exp3P5. EnpP5 unsegmented, reaching middle of inner margin of Exp1P5. Exp   3-segmented; lateral spine of Exp2P5 shorter than external margin of Exp3; external seta of Exp3 about half length of internal seta; internal seta almost reaching middle of terminal claw (Fig. 72B).
Remarks. The specimens examined were collected in the lower stretch of the Paraguay River (RPAG-B). This species is found in Argentina in the middle and lower stretches of the Paraná River and it can be considered to be a common species in the zooplankton community of this region (Fig. 73). However, Ringuelet and Martínez de Ferrato (1967) found that this species was uncommon during the period of their study, occurring in lotic or adjacent systems only between February and April. It was not found in reservoirs. This species can be readily distinguished from its congeners because of the position of the robust lateral spine on the right Exp2P5, close to the insertion of the terminal claw.
Adult female, body length 1688 µm. Complete suture present between Ped4 and Ped5; with several irregular rows of spinules present dorsally along posterior margin of Ped4 (Figs 76A, 77A-C); lateral wings slightly asymmetrical, left wing larger than right; both wings bearing pair of sensillae, one large and one small; large sensilla on left side located on hemispherical projection, right sensilla located on apex of wing (Fig. 77C). GS asymmetrical, about 1.5 times longer than wide; anterior part slightly dilated, with swelling on left side larger than on right, sensilla present on apex of each, both about twice as long as wide; sensilla on right side directed perpendicular to long axis of body (Fig. 76A). Right margin of GS longer than left. Cx of P1 with spinules on antero-lateral surface (Fig. 77G). P5 symmetrical (Fig. 76B) with small conical process at distal corner of CxP5, bearing triangular sensilla approximately 1.8 times longer than wide. BspP5 with outer seta of medium length, reaching middle of external margin of Exp1P5 (Fig. 77D). EnpP5 one-segmented, slightly longer than internal margin of Exp1P5. Exp 3-segmented; lateral spine of Exp2P5 long, reaching middle of external seta of Exp3P5 (Fig. 77E); internal seta of Exp3P5 about 3.5 times longer than external seta of Exp3P5; internal seta of Exp3P5 extending beyond middle of terminal claw.
Remarks. The illustrated specimens were caught in the upper reaches of the Uruguay River, in the Machadinho Reservoir. The northernmost boundary of its distribution may lie in the northern sector of Paraná State (Brazil), its northernmost record is from the floodplain of the upper Paraná River (Fig. 78).
This species has been the subject of much taxonomic confusion. Ringuelet and Martínez de Ferrato (1967) argued that misinterpretation of the intraspecific variability of the spinous processes on segment 15 and of the falciform process on segment 20  of the male A1R is responsible for the establishment of "Diaptomus" toldti Pesta, 1927, a junior subjective synonym of N. spiniger. They added that individuals with welldeveloped spinous processes typically had a larger body size. Brehm (1933) included this species in the genus Argyrodiaptomus, but Kiefer (1936) doubted its inclusion in the genus because of its possession of a mammiform process on the internal margin of the right BspP5 and thus suggested an affinity with Notodiaptomus. Some subsequent authors (e.g. Ringuelet and Martínez de Ferrato 1967) have placed this species in "Diaptomus" as D. spiniger, while Dussart and Defaye (2002) treated it as incertae sedis within Argyrodiaptomus.
There are two other synonyms: "Diaptomus" birabeni Brehm, 1957 andNotodiaptomus orellanai Dussart, 1979 as proposed by Paggi in his MS Thesis in 1994. In each case the new species was established on the basis of morphological characters (Brehm 1957, Dussart 1979, but both synonymies are now widely accepted.
Further study of this species is necessary, including study of museum collections, in order to verify these synonymies and assess the evidence supporting the placement of this species in the genus Notodiaptomus, in part because of the presence of a mammiform process on the internal margin of the right BspP5, a character which conflicts with the current diagnosis of the genus.
Remarks. The specimens were collected in the low stretch of the Uruguay River at Salto Grande Reservoir (Fig. 83, SGRA-D). Only three individuals were found and a full re-description was published by Perbiche-Neves et al. (2012). This species is on the red list of endangered species (IUCN-Red List of Threatened Species, 2010-Reid 1996, due to lack of recent records. The only previous confirmed record of this species was its original description (Brehm 1933). There is a subsequent record of this species from pools in Venezuela, but it is doubtful and requires confirmation (Perbiche-Neves et al. 2012). Ringuelet (1958) did not find this species in his work on Argentinian co-   pepods, nor did Paggi and José de Paggi (1990). This was only the second confirmed record of this rare species in 77 years.
The genus Odontodiaptomus comprises three species (equivalent to the thomseni group of , and there is a great interest in the genus because of its apparent isolation from other Neotropical Diaptomidae (Santos-Silva 2008). A second species of this genus, Odontodiaptomus paulistanus  is relatively common in Brazil, but was not found in the present study. The third species, O. michaelseni (Mrázek, 1901), was not found in our study. G), length similar to width of Ur4. CR asymmetrical, left ramus larger than right, internal seta of CR with narrow section close to base (Fig. 84F). Segment 11 of A1R with modified seta on anterior margin longer than modified seta on segment 13; spinous processes present on segments 15 and 16 (Figs 84C, 85A, C); segment 20 with small, curved, distal process (Fig. 85E). Two rows of spinules present on surface of Enp1 of A2 (Fig. 84D). Small mammiform process present proximally . Lateral seta of CxP5L as long as width of segment; BspP5L as wide as long (Fig. 84B). Adult female, body length 1120 µm. Incomplete suture present between Ped4 and Ped5; dorsal surface smooth, without spinules (Fig. 86A); lateral wings asymmetrical, left wing larger than right. GS slender, asymmetrical, about 1.6 times longer than wide; anterior part slightly dilated, with swelling on left side larger than right. P5 symmetrical (Fig. 86B) with small conical process at distal corner of CxP5, bearing short, triangular sensilla. BspP5 with long external seta, extending beyond base of terminal claw. EnpP5 one-segmented. ExpP5 3-segmented; lateral spine of Exp2 longer than external margin of Exp3P5; external seta of Exp3P5 about two thirds length of internal seta; internal seta reaching mid-length of terminal claw. Remarks. Several individuals of this species were found in the Yaciretá Reservoir (Fig. 87, YACI-B) in the middle Paraná River. This species had been found by S. M. Frutos (pers. comm.) in Yaciretá Reservoir, and J. C. Paggi (pers. comm.) found three males in the middle section of the Paraná River in 1971 and in 1993. This species is distinct from all other Diaptomidae found in South America, especially with regard to the P5 and the dorsal process on the Ur4. A preliminary phylogenetic analysis based on morphological characters was inconclusive with regard to which genus this species should be attributed to, thus suggesting the possibility that it represents a new genus. However, Perbiche-Neves et al. (2013) decided to maintain the status as "Diaptomus" sensu lato until a comprehensive phylogenetic analysis of Neotropical diaptomids could be completed. Diagnosis. Adult male, body length 1292 µm. Irregular row of spinules present dorsally and laterally along posterior margin of Ped4 (Fig. 88A). Modified seta on segments 11 and 13 well developed, that on segment 13 longest, extending to end of segment 14 (Fig. 89G, H); spinous process well developed on segments 15 and 16 of A1R; spinous process on segment 15 longer than that on segment 16; falciform process on segment 20 of A1R in all specimens analysed (N=16), reaching beyond middle of apical segment, proximal surface of process rugose (Fig. 88G, 89I, J). Right CxP5 with well-developed distal process bearing short sensilla (Fig. 89E). Right Exp1P5 short, about 1.5 times wide than long; right Exp2P5 flattened, subtriangular in shape (Fig. 89C-D), wider than long, with swollen outer margin proximal to insertion of outer seta (Fig. 89A, C, 88B); outer seta slightly curved (Fig. 88B, 89C). Terminal claw strongly curved near tip H,89F). Strong seta present on distal margin of CxP5L. Adult female, body length 1346 µm. Incomplete suture present between Ped4 and Ped5; rows of strong spinules marking position of posterior margin of Ped4 (Figs 90A, 91B, C); lateral wings slightly asymmetrical, similar in size; both wings with sensilla at distal tip (Fig. 91A). GS symmetrical, approximately 1.8 times longer than wide; anterior part dilated, with swelling on left side larger than on right; sensilla on apex of each, about 1.8 times longer than wide; swelling on left side hemispherical, swelling on right more conical (Figs 90A, 91F). P5 symmetrical (Fig. 90B), with small conical process at outer distal corner of CxP5, bearing short, strong, triangular sensilla, about 1.5 times longer than wide. BspP5 with long outer seta, about twice length of EnpP5. EnpP5 one-segmented. Exp 3-segmented; lateral spine of Exp2 as long as external margin of Exp3 (Fig. 91D); external seta of Exp3 reaching one third length of   internal seta; internal seta almost reaching mid-length of terminal claw; terminal claw with comb of robust spinules in mid-section (Fig. 91E).

Genus
Remarks. The specimens were found in the middle part of the Paraná River (Fig.  92, RPAR-M2) and in the Yaciretá Reservoir (YACI-D). The species was relatively abundant in the samples and occurs in reservoirs and other water bodies near these locations, according to S. M. Frutos (pers. comm.). In a previous phylogenetic analysis, this species was placed close to the genus Scolodiaptomus Reid, 1987, which is monospecific [Scolodiaptomus corderoi ], because they share important features such as the subtriangular shape of Exp2P5R and the presence of modified setae or spinous processes on segments 13, 15 and 16 of the male A1R. However, it is very unlikely that "D." curvatus belongs to the genus Scolodiaptomus, which is characterised by an unornamented hyaline lamella on segment 20 of A1R and by the presence of a dorsal cylindrical process on the Ped3. Again, the affinities of this species need to be tested as part of a comprehensive phylogenetic analysis of Neotropical diaptomids. It is likely that this species could be assignable to a new genus.

Discussion
Nineteen species of diaptomid copepods were found during the present survey, but this figure probably is markedly lower than the estimated total for de la Plata river basin. The reason for this discrepancy may be related to sample representativeness due to limitations in terms of seasonality and kind of sampled freshwater habitats (only large rivers and reservoirs). Historical records suggest that there are at least another 15 species to be found in the basin. Boxshall and Defaye (2008) summarized published reports for the entire Neotropics and calculated a total of 82 species of Diaptomidae from the region. The 19 species found in this survey represent about 24% of that total. Despite this limitation, our sampling effort generated sufficient data to significantly improve our understanding of the spatial distribution of diaptomids throughout de la Plata river basin. Together with the existing literature, we can now begin to understand the large scale distribution patterns of most diaptomid species in the Neotropics. Andean diaptomids like the genus Tumeodiaptomus Dussart, 1979 were not found. Only three species, N. coniferoides, N. conifer and N. isabelae were distributed throughout the whole basin. However, detailed taxonomic studies are still required to investigate whether these morphological species represent complexes of cryptic species with limited distributional ranges, as is possibly the case at least for N. coniferoides. Notodiaptomus coniferoides occurred only in lotic stretches and it is likely that it does not occur frequently in reservoirs. Notodiaptomus simmilimus is another species that requires further detailed analysis.
There was no gradual decrease in the number of species towards the south of the continent, as might be inferred from Jablonski et al. (2006), for example, at least as far as the southern boundary of de la Plata basin. The central valley of the Paraná River serves as a route for the dispersal of several species of copepods. In its middle section it is possible to find species both from the northern group (e.g. from the nordestinus complex) and from the southern group (Argentinian species). There is evidence of the cooccurrence of members of the northern and southern groups together with other species which exhibit restricted distributions and a high degree of endemism. At sample site RPAR-L2 in the mid Paraná River and in Yaciretá Reservoir, for example, we observed the largest number of calanoids per sample anywhere in the basin. Some previous studies have mentioned the large number of species found in the Paraná River (Paggi and José de Paggi 1990, Dussart and Frutos 1985, Lansac-Toha et al. 2004. With the exception of the most widespread species, N. coniferoides, N. conifer and N. isabelae, the other diaptomids could be roughly divided into three groups according to their geographical distribution patterns, and their relationship to limnological variables and climatic factors (Perbiche-Neves et al. 2014). Calanoids in the northern basin (especially the nordestinus complex, see Santos-Silva 2000) had a southern limit at the high/middle stretch of the central channel of the Paraná River, in the Iguaçu River, and in the upper Uruguay River. South of this region, there were species typical of the Argentinian fauna in the middle and lower basin, but in the middle section it is possible to find species from the northern and southern parts of the basin, as well as other species restricted to this particular range.
As emphasized in the diagnoses of species given above, the detailed analysis of segment 20 of the male A1R revealed substantial variation in the shape and state of development of the distal process of this segment: a falciform process is usually present but other shapes may occur, and many species lack a process on this segment. It is necessary to be cautious in interpreting the wide variability in this process exhibited by species of some diaptomid genera. Based on our observations, it should not be used alone or as a strict diagnostic character for the identity of a particular taxon. Paggi (1976) highlighted this morphological variation in diaptomids of the middle Paraná River region in Argentina. Similar patterns of variation have also been observed in species of Argyrodiaptomus, for example, A. azevedoi and A. denticulatus, as well as in the Notodiaptomus species, N. spiniger, N. henseni, and N. isabelae. Species of the genus Argyrodiaptomus typically have a falciform process. Brehm (1933) created the genus Argyrodiaptomus and designated A. granulosus Brehm, 1933 as the type species. However, it is noteworthy that Wright (1927) had already identified what he called the "bergi" group, which would later be incorporated into the genus Argyrodiaptomus. Subsequently, in 1938, Wright reviewed the "bergi" group in South America, disagreeing with Brehm (1933) in certain respects, such as the inclusion of particular species and the use of the name of Argyrodiaptomus. However, the genus persists and is widely used today. The "bergi" group was centred on Argyrodiaptomus bergi (Richard, 1897). It is worth noting that both Argyrodiaptomus bergi and A. granulosus are relatively rare, and the latter has not been recorded since its original description (Brehm 1933).
Notodiaptomus spiniger has been considered by some researchers to belong to the genus Argyrodiaptomus (see Dussart and Defaye 2002), but it is clear that it does not fully conform to the diagnosis of either Notodiaptomus or Argyrodiaptomus. As currently constituted, it appears that Notodiaptomus is not a monophyletic taxon. It is possible that the generic concept should be restricted to fewer species, such as the nordestinus complex and perhaps a few other species distributed across the Brazilian shield and in the Amazon and de la Plata basins.
In the genus Notodiaptomus characters such as the presence or absence of spinular ornamentation on the dorsal surface of the pedigerous somites and its pattern and shape proved to be highly informative diagnostic features, thus facilitating the discrimination between certain pairs of species or closely related groups of species. The combination of this character with others (for example the state -complete or incomplete -of the suture between Ped4 and Ped5) can provide additional, robust criteria to achieve accurate identifications in this genus.
The fifth leg (P5) of adult diaptomids provides the main set of differential characteristics for most species. However, for some species of the nordestinus complex, for instance, the differences may be minimal, even with reference to the finest scale details.
Body length is useful for the identification of some species of Notodiaptomus. It can, for example, help to separate N. conifer from other species that share the possession of a small lateral spine on Exp2P5R, such as N. iheringi and N. cearensis. Some studies, such as Ringuelet and Martínez de Ferrato (1967), have highlighted the link between body size and the presence or absence of a falciform process on segment 20 of the maleA1R. According to their results, males with a well-developed process on the distal margin of segment 20 A1R, tend to be larger than those which lack this process. Males with well-developed process on segment 20, also tended to have a larger spinous process on segment 15. Laboratory experiments under various environmental conditions would be useful to clarify the significance of any possible linkage between the states of these two different characters. Such studies can also provide new insights into potential synonymies between species, as demonstrated by Ringuelet and Martínez de Ferrato (1967) who provided evidence to establish the synonymy between "Diaptomus" toldti and N. spiniger (as "Diaptomus" (Notodiaptomus) spiniger).
Notodiaptomus incompositus showed a distribution restricted to the middle and southern parts of the basin. This species shares several morphological characteristics with N. deitersi (Poppe, 1891), the type species of Notodiaptomus, which is included in the nordestinus complex. Notodiaptomus spiniger is not treated as part of this complex, even though it is also restricted to the south of the basin, as are N. dentatus and N. carteri, among others. Although N. isabelae occurred only in the middle and lower basin, there are several published records from the upper part of the Paraná River basin and from other smaller basins nearby within the Brazilian shield, for example, in the State of Minas Gerais, Brazil (Maia- Barbosa et al. 2008).
We did not find N. deitersi (Poppe, 1891) in this survey. The species was originally described from the region of Cuiabá (Mato Grosso State, Brazil) and it is the type species of Notodiaptomus (Santos-Silva et al. 1999). Among the adult diaptomids reported in this work, none showed the morphological characteristics of N. dietersi as presented by Santos-Silva et al. (1999) and a targeted search was carried out as we considered it highly likely that the species would occur in the basin. The morphotype of Matsumura- Tundisi ( , 2008 is different from that redescribed by Santos-Silva et al. (1999), and was absent from our samples. There is the possibility that the sampling site visited during our survey may not have provided conditions conducive to the capture of this species.
We decided to treat Notodiaptomus cf. spinuliferus as incertae sedis and not include it in this work because the vouchers deposited in MZUSP (6971) (Fig. 93), identified as N. cf. spinuliferus, are not sufficiently similar to the species descriptions provided by Dussart (1985), Dussart and Matsumura-Tundisi (1986), Paggi (2001) and Matsumura- Tundisi (2008). The specimens found in this study were more similar to those in Dussart (1985) and Paggi (2001). This species requires further taxonomic investigation in order to resolve its true identity.
As a result of previous faunistic surveys, some researchers have inferred that particular calanoid species have disappeared from the region (e.g. Matsumura- Tundisi 2003, Sartori et al. 2009). We consider that such inferences should be made with caution. In most cases there is little information on the population dynamics of the species and we found species such as O. thomseni and A. bergi (Perbiche-Neves et al. 2011, 2012 which had not been seen for several decades. Our understanding of the relationships between diaptomids and the various limnological parameters and trophic status of water bodies is not very robust and is made less secure given that these species inhabit artificial reservoirs but were originally inhabitants of lakes, ponds, and rivers. Due to its large body size and susceptibility to being preyed upon by small fish, the abundance of species of Argyrodiaptomus is generally low in open water bodies when compared to that of other calanoids. During the counting of samples for ecological studies, if only low numbers of individuals (up to 200 per sample) are identified this probably decreases the chances of finding such rare taxa. In small, more closed habitats and in the absence of visual predators, the abundance of Argyrodiaptomus can be very high (Perbiche-Neves et al. 2011). Argyrodiaptomus furcatus and A. azevedoi occurred in reservoirs that were oligotrophic as well as in eutrophic habitats. They may be limited by turbidity rather than by trophic status due to their particular feeding habits. In addition to the transient negative effect on the water column of a turbidity event, after sedimentation of the suspended material, the surface of the substrate can be covered and may result in an adverse effect on the replacement rate of planktonic taxa by preventing hatching of deposited eggs. In such cases the recolonization after conditions have become favorable again may be by individuals whose origins are in associated lagoons or unaffected upstream river stretches.
Notodiaptomus conifer was reported as missing from Jurumirim Reservoir (Paranapanema River) by Matsumura- Tundisi and Tundisi (2003) and Sartori et al. (2009), but we found it in both the Jurumirim and Barra Bonita reservoirs. This species is distributed from northeast Brazil to Argentina, but may possibly occur at low densities, thus making the records of its presence unreliable for ecological studies, in which a standard volume of water is sampled from the surface or water column.
Taxonomically, the identification of N. conifer can be problematic. We recommend comparing the spinous process on segment 15 of the male A1R as well as considering body size. Notodiaptomus conifer has a greater body length than N. iheringi and other species, such as N. henseni, which could be confused with it. The presence of a row of spinules on the first segment of the A1R in N. iheringi allows this species to be distinguished from N. conifer. In N. henseni, the presence of sclerotized processes on the basipodites of both right and left P5 also avoids any confusion with N. conifer. In some N. henseni the presence of a well-developed spinous process on segment 15 of the A1R may cause confusion with N. conifer if this feature is considered alone, so it is always advisable to try to use more than one distinguishing feature, while also taking into account the known geographical distributions.
All these data indicate that care must be taken when reporting the disappearance of a particular species from a habitat; this conclusion should be drawn only after failure to find the species after continued sampling over extended periods of time. The probability of a species being found again seems high when routine sampling is used (e.g. Perbiche Neves et al. 2012). Another important issue is that adult diaptomids are scarce in lotic habitats, compared to lentic habitats. The macro-scale view obtained in this large scale survey supports such generalizations.
Studies carried out in Europe and Asia over long time periods have demonstrated that particular species may be dominant at rare intervals over a 20 year time series. Polli and Simona (1992) noted the complete disappearance of calanoids for nearly 30 years in a small lake, and then charted their re-appearance as the system returned to an oligotrophic state. If we analyze the historical data from some dam sites and other habitats in Brazil, for example, such as flood plains, we can observe that over 40 years, many species occur and then are absent for a period of time. During these intervals, the physical and chemical characteristics of the water and weather conditions may have changed and exerted a strong influence on the fauna. Straile and Geller (1998) studied the changes in the composition of microcrustaceans in Lake Constance between 1920 and 1995. During this period the lake was initially oligotrophic, became eutrophic, and then returned to oligotrophic status. Although the species richness found in a lake in temperate latitudes is lower than that found in a tropical lake or reservoir, this study illustrated important changes in species composition. Straile and Geller (1998) reported a major change after only a decade, in which the cyclopoid species Cyclops vicinus Uljanin, 1875, hitherto nonexistent in the habitat, emerged as a common species. They attributed its success to the competitive advantages of this cyclopoid over calanoid copepodites (of the genus Eudiaptomus Kiefer, 1932), especially at high food concentrations. Predation by Cyclops on young stages of a predatory calanoid of the genus Heterocope G. O. Sars, 1863, led to the local extinction of Heterocope in the lake. Straile and Geller (1998) pointed out that Heterocope had also disappeared from other alpine lakes since 1950, but in many cases its disappearance was not associated with the presence of Cyclops vicinus. Finally, Straile and Geller (1998) compared the different sampling strategies in different years, concluding that the precise positioning of sampling locations (onshore versus offshore, for example) may have been an important factor influencing the analysis of species composition in the samples. The interactions within the zooplankton communities are complex and this is compounded by the greater species richness present in tropical regions. Melo et al. (2006) indicated that the number of known species in Brazil can be mapped by their proximity to research centers. Our survey supports this conclusion, since many studies have been undertaken in southeastern and southern Brazil, primarily focusing on the more accessible reservoirs of the upper Paraná River and its floodplain. In Argentina, the same phenomenon was observed with studies being concentrated in the middle section of the river, mostly in marginal lakes and lotic stretches. So available data are still patchy as yet with little or no information on the copepods of particular sub-regions.
Within de la Plata river basin, the results obtained in this study combined with our review of published data suggest the following recommendations for future research: 1. Continue studying sites where historical collections have been made; 2. Study large and productive reservoirs for which little information is available and where new species records are likely; 3. Conduct studies in places that are unexplored and represent gaps in information, including reservoirs, ponds and wetlands adjacent to tributaries of the Upper Paraguay River, Upper Parnaíba River, Upper Uruguay River and west of the state of Rio Grande do Sul, the region of "Esteros de Iberá" (in Argentina); and 4. Study the transition areas between the northern and southern faunas, in the rivers Iguaçu, Paraná (Yaciretá Reservoir) and middle/upper Uruguay, focusing on flood periods and correlations with bird migration routes.
There are several taxonomic problems still to be resolved within the Neotropical diaptomids and it would be useful to test whether Argyrodiaptomus and Odontodiaptomus are monophyletic. Many species currently placed in Notodiaptomus and "Diaptomus" must eventually be relocated. Diaptomus, for example, continues to serve as a temporary repository for species whose affinities are as yet unresolved, such as the unrelated species "Diaptomus" frutosae and "Diaptomus" curvatus. We also consider that Notodiaptomus spiniger, N. anisitsi and N. coniferoides do not belong in Notodiaptomus. One solution might be the creation of additional new genera to accommodate some of these distinctive Neotropical diaptomids, however, new genera should only be erected after thorough phylogenetic analysis. Phylogenetic analyses based on morphological and molecular data are currently under way by the present authors and will be published elsewhere when completed.
The following genera and species should be reviewed as their taxonomic status is currently equivocal: N. spiniger (synonymy needs verification and its placement in Notodiaptomus requires testing); N. spinuliferus (the material deposited in MZUSP should be redescribed and compared with material found elsewhere and identified as N. spinuliferus); N. amazonicus from Amazonia should be compared with N. amazonicus from de la Plata river basin, including molecular analysis; the morphology of N. coniferoides should be checked by comparison with the holotype from the Amazon; this will provide a more robust basis for determining the true distribution of this species across the continent.
The rarity and seasonallity of some species in the basin, such as N. anisitsi and A. falcifer, and the absence of N. deitersi, O. paulistanus, O. michaelseni, and S. corderoi, suggests that carrying out only two sampling programmes over a year may be inadequate in terms of recording sound data on the seasonal variability of these species. It is necessary to be very careful when using terms like "disappearing species", since their absence might reflect an inadequate sampling regime.