© EDP Sciences, 2020

1 Introduction

It is known that body size is highly variable among the organisms and it is closely related to biotic (e.g., genetic) and abiotic (e.g., environmental) factors for many species. Body size can show a variety of differences within the members of the same taxonomic group (e.g., species, genus, etc.). Body size has been shown to influence species function and life history strategies (Brown et al., 1993), and often plays a critical role on its survival (Atkinson, 1994). For example, sexual dimorphism among individuals of the same species (i.e., intra-specific variability) can be characterized by the body size differences where females are usually larger than the males (Cohen and Morin, 1990; Fairbairn et al., 2009).

Most recently, studies focusing on the relationship between temperature and body size revealed some evidence that increasing temperature can induce reduction in body size in different aquatic organisms (e.g., Węsławski and Hessen, 2017). Although the debate on body size and temperature relationship continues, (see discussion) and requires further study (Panov and McQueen, 1998), other influential biotic (e.g., predation) and abiotic factors (i.e., altitude) coupled with temperature changes at different altitudinal levels can be considered to play critical role on the size differences. This can be important for ectothermic species (i.e., ostracods, and other aquatic invertabrates), inhabiting a variety of water bodies located at different altitudinal ranges. Body size reduction is frequently observed with increasing altitude corresponding to decrease in air and water temperatures. At this point, one may consider ecological rules to explain the body size-altitude relationship; however, these explanations are not widely accepted for many species. For example, Eweleit and Reinhold (2014), working on two supspecies of Orthoptera Poecilimon veluchianus Ramme, 1933, did not find supportive evidence for Bergmann's rule, which states that animals in cold habitats are being larger than those inhabiting warm habitats, explain changes in body size of these two with changing altitudinal ranges. In addition, their results also violated another rule called Rensch's rule, which proposes that sexual size dimorphism increases in species where males are larger in body size than females, when sexual size dimorphism in their study organisms did not differ even though body size decreased with altitude. Similarly, Külköylüoğlu et al. (2012a) underlined that Rapoport's rule, which suggests negative correlation between species diversity and altitude, did not provide supportive evidence about the relationship between reproductive modes (implying sexual dimorphism) of ostracod species and altitude and/or any other environmental variables used therein.

Ostracods are common invertebrate animals found in almost all aquatic habitats. They are composed of soft body parts within two calcareous valves (carapace). The carapace can be easily fossilized in sediment due to its CaCO₃ formation. Well preserved fossil species are known from Early Ordovisian age (ca 485 million years old) (Williams et al., 2008). Thus, this calcified structure has been commonly used in palaeontological studies (e.g., paleolimnological, paleoclimatic) (Forester, 1991; Holmes et al., 1998; Külköylüoğlu, 1998; Mischke et al., 2005; Jiang et al., 2008). The carapace has several different types of structures such as tubercules, pores, nodes and spines, and plays key role for ostracod activity (Keyser, 2005; Danielopol et al., 2018). Since these features are evident in ostracod fossils, the carapace provides possibilities for comparative studies of past and present habitat conditions, and aids in the reconstruction of the history of the ecological conditions of particular habitats. This is mostly due to species-specific habitat preferences of individual species (Külköylüoğlu, 2003, 2005). Considering that there are about 2330 species of non-marine ostracods (Meisch et al., 2019), any knowledge about the current (i.e., live) species' ecological preferences can offer a tremendous amount of scientific information about habitat conditions in past.

Many ostracod species display variation in body size (cf. size differences between the female and male) and most specifically size differences between left and right valves (or so-called size dimorphism). Although it cannot be generalized (e.g., see exceptions in the genus Homocypris whose valves are almost equal in size), for many species the left valve is typically larger than right one, and there is no specific mechanism identified that explains this pattern. In addition to size dimorphisms, soft body parts can also exhibit dimorphism. According to Martins et al. (2017), size dimorphism seen in some of the species of the genus Cyprideis, dimorphism was also the case for some in the species Cyprideis torosa Jones 1850 where soft body parts found on the left side were generally larger than the right.

Ostracods can be found from about 5000 m below sea level (Benson, 1972; Fricke et al., 1989; Brandão and Yasuhara, 2013) to temporary or permanent artificial (e.g., canals, reservoirs, ponds, troughs,) and natural (e.g., lakes, streams, rivers, semiterrestrial habitats, springs and pools) freshwater habitats located up to 5000 m above sea level (Laprida et al., 2006; Mischke et al., 2007). Ostracods are probably the only invertebrate aquatic group with such wide altitudinal ranges (≥10 000 m). However, scientific argument about the relationship between altitude and ostracod distributions and/or their occurrences is unresolved (e.g., Mezquita et al., 1999; Laprida et al., 2006; Pieri et al., 2009; Külköylüoğlu et al., 2012a,b; Guo et al., 2013; Yavuzatmaca et al., 2015, 2018). Nevertheless, these studies account importance of the altitude and ostracod relationship. However, there is no study focusing on the relationship between carapace size/length of nonmarine ostracods and altitude. Thus, the aims of the present study were to (1) provide the first extensive data on empirical distribution and diversity of nonmarine ostracod species in the study area, (2) show the relationship between carapace size of nonmarine ostracods at different altitudinal ranges in Mersin province, and (3) examine other abiotic factors which may have influential effect on individual species.

2 Materials and methods

2.1 Study area

Mersin is part of the Mediterranean region of Turkey where 87% of the province (aka Mersin city) is mountainous (West and Central Taurus Mountains) and 13% is woodland (Mersin Governor, 2011). The area studied lies between 36−37° North latitudes and 33–35° East longitudes. Mersin's land border is about 608 km while the sea border is 321 km with an area of 15.853 km². The province has a semi-arid low humidity, mild winters, hot summers and very long shoreline in Mediterranean Sea. The average ambient temperature is about 19.2 °C and the total amount of rainfall per year is 591.6 mm (Mersin Governor, 2019).

2.2 Sampling and measurements

A total of 117 samples (Fig. 1) were randomly collected from 13 different aquatic habitats (i.e., lakes, creeks, troughs, dams, ponds, rivers, spring waters, canals, waterbodies (slough), pools, helocrene springs, rheocrene springs, and unnamed water bodies) between 3 and 9 October 2015. Ecological variables, including electrical conductivity (EC µS/cm), dissolved oxygen concentration (DO, mg/L), pH, water temperature (TW, ^°C), salinity (Sal, ppt), oxidation reduction potential (ORP, mV) were measured at each site before sediment sampling was conducted. Measurements of water variables were done using a multiprobe YSI-I Professional plus while geographical information (coordinates and altitude) were reported with GARMIN e-trex Vista H GPS. A Testo 410-2 anemometer was used to record air temperature (Ta ^°C). We also collected 200 ml of water to measure the major ions (anions and cations) using a sterile plastic bottle. The water samples (Appendices 1 and 2) were arranged to analyze by filtering from a syringe-type membrane filter (0.45 µm) and then standard method (no: 4110) was applied for anion (chloride (Cl^–), sulphate (SO₄⁻²) nitrite (NO₂^–) and nitrate (NO₃^–) and cation (sodium (Na²⁺), calcium (Ca²⁺), potassium (K²⁺), magnesium (Mg²⁺)) analyses using Ion Chromatography (Dionex 1100) at the Department of Engineering, Bolu Abant İzzet Baysal University.

Ostracods collected with sediments (ca. 1–5 cm of depth) from each sampling site using a hand net (150 µm mesh size) were kept in 250 ml of plastic bottle in %70 alcohol. All samples were preserved in coolers during transport to the laboratory. In the laboratory, each bottle was filtered through four standard sieves (0.5, 1.0, 1.5- and 2.0-mm mesh size). Afterwards, the samples were sorted from sediments under a stereomicroscope (Olympus ACH 1X) and prepared for taxonomic identification. We used taxonomic keys of Meisch (2000), Karanovic (2012), and other related keys whenever possible to identify specimens to the species level. Both the soft body parts and carapace were preserved in slides with solution of lactophenol while valves were kept in the micropaleontological slides. Each slide was numbered and deposited at the Limnology Laboratory of Bolu Abant İzzet Baysal University, Turkey. Scanning Electron Microscope (SEM) photographs were obtained from TÜBİTAK-MAM (Gebze, İstanbul). A light microscope (Olympus BX-51) and inverted microscope (Leica DMi1) were used to measure carapace length and height. A Leica MC170 HD camera with Leica Application Suite V4 software were also used for valve length measurements. We measured the longest distance on the valves from the posterior end to the anterior end of the valve, and the height from the highest peak of the valve in lateral view. Witdh was represented as the widest distance of the carapace in the dorsal view. Total of 3980 left and right adult valves (or 1990 complete carapaces) were measured. Intra-species and inter-species variabilities of the carapace were compared for the same species found at different altitudinal ranges.

Fig. 1 Distribution of 117 sampling sites from 13 counties (Akdeniz, Anamur, Aydıncık, Bozyazı, Çamlıyayla, Erdemli, Gülnar, Mezitli, Mut, Silifke, Tarsus, Toroslar, Yenişehir) of Mersin, Turkey.

2.3 Statistical analyses

Microsoft Excel 97–2003 version was used to arrange actual data collected from each site. Two tailed t-tests with un/equal variances were used to compare the mean values of the length and height of the species among the altitudinal ranges. Paired-t tests along with Pearson correlation analyses were also used for the comparison of left and right valves of the same species found at each altitudinal range. During the measurements, 36 species were identified as the most common species across different elevational ranges. We used a simple regression model to test possible correlations of some environmental variables (e.g., water temperature) with carapace size. Canonical Correspondence Analysis (CCA) was used to examine correlation between environmental variables and ostracod species occurring in three or more sites. CCA was run along with Monte Carlo Permutation test (499 permutations) where rare species were downloaded. Before running CCA, Detrended Correspondence Analysis (DCA) was applied to evaluate the suitability of data for CCA. Accordingly, if the length of DCA was found to be greater than three, data were considered suitable (Ter Braak, 1987; Birks et al., 1990). Five major variables (TW, DO, Sal, pH, Altitude) were found to be suitable for CCA analysis. CANOCO software version 4.5 was used to find the relationships between environmental variables and ostracod species composition among the sampling sites. CCA also provides understanding of the most effective variable(s) on species along the altitudinal transects. We used the C2 program (Juggins, 2003) to estimate most common species' tolerance and optimum values for some environmental variables. In which, we applied a transfer function from the weighted averaging of regression models for each species and variables during the analyses of tolerance and optimum estimates.

3 Results

We found 50 ostracod taxa (36 recent and 14 sub-fossils) from 66 of 117 sites (Appendix 2). Four taxa (Bradleycypris cf. obliqua (Brady, 1868); Cypretta cf. reticulata Lowndes, 1932; Fabeaformiscandona cf. wegelini (Petkovski, 1962); Hemicypris cf. largereticulata (Rome, 1969)) were new records for the Turkish Ostracoda fauna (Appendix 3). Including these three new species, there are now 39 nonmarine ostracod species known to the Mersin province. Comparing to the other areas studied in Turkey, Mersin exhibits unequivocally high species diversity. Eleven species with occurring at three or more different sampling sites were encountered from different altitudinal ranges, while the rest of the species showed patchy distribution among the ranges. Three species (I. bradyi Sars, 1890; H. salina (Brady, 1868); P. fallax Fox, 1967) were widely distributed in almost all altitudinal ranges (Fig. 2). The numbers of species decreased with altitude. Maximum numbers of species (17 spp.) were found from 0 to 165 m of range followed by 12 and 11 species at 166–330 m, and 826–990 m, respectively (Tab. 1).

We found evidence that carapace size of a few species increased with altitude, but most others did not show such a clear trend. Testing the null hypothesis of “no significant difference between the mean length of right and left valve of the species” revealed a significant result (p < 0.05) when all species were considered. Similarly, comparing the height of the valves showed significant differences among the species. However, the length and height of the five most common species (P. fallax; I. bradyi; H. salina; H. incongruens (Ramdohr, 1808); Ps. olivaceus (Brady and Norman, 1889)) exhibited variations at different altitudinal ranges (Tab. 2). For example, valve length and height of I. bradyi both increased with increasing altitude. However, the situation was different for H. salina that its height increased with altitude, but its length did not portray such a trend. Heterocypris incongruens is a well-known cosmoecious species with a wide geographical distribution and high levels of tolerance to different environmental conditions. Interestingly, this species was only found within a limited altitudinal range from 496 m to 1155 m. Both the length and height of this species appear to increase with altitude within its limited distribution. Similar trends were also seen for P. fallax (t  =  1.77, p = 0.01) and Ps. olivaceus (t  =  2.51, p = 0.01) when correlation between the left and right valves was generally strong and significant.

The CCA diagram (Fig. 3) outlined that TW (p = 0.006, F = 3.259) and pH (p = 0.006, F  =  3.251) were the two most effective variables on species occurrences while the other three variables (DO (p =  0.272, F = 1.187), salinity (p = 0.398, F = 0.835), altitude (p = 0.280, F = 1.970)) did not show strong relationships with the 12 most common ostracods (Tab. 3). Troughs and creeks were the two habitat types with highest species numbers (Tab. 4).

Four species (P. fontinalis (Wolf, 1920); I. bradyi; H. salina; P. olivaceus) had the highest tolerance values for altitude while four others (P. fallax, I. bradyi, H. incongruens, and P. olivaceus) had high optimumum estimates (Tab. 5). Heterocypris salina was the only species high values of both tolerance and optimum estimates for electrical conductivity (referring to salinity). A few species had different levels of tolerance and optimum values for some of those major ions; for instance, P. fallax had maxiumum tolerance to sodium ion while its optimum value was very small for water temperature (Appendix 2). On the contrary, there was no significant relationship between the size and/or occurrences of the species and those of ions used in the present study.

Fig. 2 Length (L) and height (H) measurments of left (L) and right (R) valves of the species (A) H. salina and (B) I. bradyi at different altitudinal ranges.

Fig. 3 CCA diagrams of (A) five variables with 12 taxa, and (B) with sampling sites shown with empty circles. Abbreviations: Candona sp (Csp), Cypridopsis vidua (Cv), Heterocypris incongruens (Hi), H. reptans (Hre), H. salina (Hs), Ilyocypris bradyi (Ib), I. inermis (li), Limnocythere inopinata (Li), Potamocypris fallax (Pof), Prionocypris zenkeri (Pz), Psychrodromus fontinalis (Pf), P. olivaceus (Po), Sal (salinity), Alt (altitude), DO (dissolved oxygen), TW (water temperature).

4 Discussion

Until this study, five ostracod species (H. incongruens; Eucypris virens (Jurine, 1820); E. mareotica (Fischer, 1855); Tonnacypris lutaria (Koch, 1838); H. salina) were known from Mersin (Hartmann, 1964; Gülen, 1988). Of which, H. salina and H. incongruens were also common in our study. Now, total numbers of ostracods are reached upto 39 species in the area. Hence, comparing to the other areas with similar habitat types (e.g., cf. Akdemir and Külköylüoğlu, 2014; Külköylüoğlu et al., 2012b, 2019; Yavuzatmaca et al., 2015, 2018; Yavuzatmaca et al., 2017; Akdemir et al., 2016), this number indicates relatively high ostracod diversity for this area. However, we believe that this number is most likely underestimated because of at least two possible factors. First, many ostracod species have different seasonal occurrences, but our sampling was done once from each site in the fall. Second, although we sampled 117 aquatic sites with 13 different habitat types, microhabitat preferences of individual species can vary depending on altitudes (cf. Appendix 1). Accordingly, aiming to complete faunal richness and diversity of ostracods may not be achieved without considering these factors. Nevertheless, our study is the first study on nonmarine ostracods in the area within a wide altitudinal sampling from −3 m to 1630 m a.s.l. and provides information for future studies.

Finding a few species (e.g., I. bradyi) with relatively high positive correlation to increasing altitude where there was a significant difference between the size of left and right valves of all species considered is not clearly explained by any ecological rules stated above. The situation seems to be more complex among common species (H. salina, H. incongruens, P. fallax, P. olivaceus) than others at different altitudinal ranges due to their scarce and/or patchy distributions. For example, height of H. salina increased with altitude but its length decreased, and correlation was not significant between the lengths of both valves. Implication of these results suggests the fact that we cannot make a general and effective conclusion about the size-altitude relationship for this species. On the other hand, we have a good support that the correlation between the valves was strong when left valve was being larger than the right valve. Unfortunately, our study fails to provide a clear answer about why such dimorphism exists in the carapace size of this (if not many) species at different altitudes.

The influence of altitude may be coupled with some other factors such as temperature and salinity changes in water bodies. For example, temperature-controlled size difference in a brackish-marine ostracod (Loxoconcha matagordensis (Swain 1955)) was found to be related to seasonal differences in water temperature where the species was larger in cooler waters (Kamiya, 1988). In contrast, studies on salinity-controlled experiments revealed that there may not be a linear relationship between carapace size of a brackish water species (Cyprideis torosa (Jones, 1850)) and salinity in laboratory experiments (e.g., Boomer et al., 2017). These authors reported larger valves below the 8.5–9.2‰ salinity range while the relationship between salinity and carapace size was found to be little or none above this range. In this study, we did not find a significant correlation between salinity and carapace size. Besides, salinity was not found as an effective factor among other factors which were not included in this study. Rather, as seen in CCA diagram (Fig. 3, Tab. 3), there was a significant correlation between water temperature and species occurrence, but this did not explain temperature-size relationship. In the light of these studies, we may conclude that altitude alone was not a determinant factor on carapace size during our study. On the other hand, we could also interpret the combination of relationship among altitude, temperature and salinity to conclude that temperature is generally declines with increasing altitude, where cold water is usually expected to be less saline than warmer waters. Even if this relationship correctly explains species occurrence, it still fails to explain the size-temperature correlation in our study. This is because the size of many (but see exceptions) ostracod species used here does not increase with decreasing water temperature and altitude. As underlined above, change in carapace size, except a few species with weak correlations, was not truly explained with any of the environmental factors measured in our study.

Moreover, it appears that temperature coupled with several different physico chemical reactions in waters may be a good candidate to explain changes in carapace size. Altitude, however, may probably play a secondary (or indirect) role on the size, occurrence, and numbers of species. Kobayashi et al. (2018), using regression models, showed that relationships between altitude (and latitude) and occurrence of calanoid copepods, a close taxonomic group to ostracods, was negatively related in 94 of 353 edge samples collected from 321 sites of lotic waters located within 2–1834 m of altitude in New South Wales, Australia. The authors argued that finding wide confidence intervals at higher altitudes was probably associated with the limited sample sizes at higher altitudes where they had limited numbers of edge habitats. However, their study failed to include environmental factors (e.g., temperature, salinity, etc.), and did not highlight correlational patterns of these factors on the calanoid species. In this study, we reported reduction in the numbers of species with altitude, but the reduction was inconsistent among altitudinal ranges. It is, however, important to point out that the most common species had almost continous occurrence patterns through out the ranges. One may also consider the fact that these common species have relatively wide tolerance ranges for many environmental variables. The advantage of having such tolerance ranges for the species likely increases their chances of survival (Akdemir et al., 2016) and the probability of occurrence within broad distributional patterns from low to high altitudes (Külköylüoğlu et al., 2019). Thereby, species with high tolerances may not be widely affected by changes in environmental factors and may not show sudden reactions to those changes within their tolerance ranges. In the much broader perspective, this also applies to their size. It seems that altitude is considered as an important geographical barrier to species distribution; however, we did not measure a direct relationship with body size of many ostracods in our study. Some other factors warrant consideration. For example, a recent study in some ostracod species of the Order Podocopida showed that body size can have strong and significant (p < 0.05) interaction with habitat type, which has direct affect on genome size (haploid nuclear DNA) (Jeffery et al., 2017). Since all of our ostracods are freshwater species in the same order, the findings of Jeffery et al. (2017) may help to produce additional explanations for the size-altitude or size-environmental relationships. A Genome-body size correlation was out of the scope of the present study; therefore, we cannot make detailed comments about these relationships with the species we studied. However, two species (C. vidua (O.F. Müller, 1776) and H. incongruens) listed in the study of Jeffery et al. (2017) were also found in the present study. This may bring out some insights for future studies to show probabilities of size differences with genome size.

The present study sought to document the size-altitude relationship in non marine ostracod species. Our study is unique, in that it is one of the first documenting as many species over such larger habitat gradients. This study can also be important as a reference point for future studies because body size is accepted as a fundamentally important biological unit in/directly associated with ecological properties and processes, as much as geographic distribution and genetic characteristics. Węsławski and Hessen (2017) argued that size-temperature correlations were predicted as the third universal response to global warming. Under these conditions we would predict that increasing temperature will cause reductions in body size. However, the broader application of this relationship will likely only be expressed in species that can respond quickly to such changes. In this sense, we believe that ostracods can be used as good indicator species for these large climatological patterns. The effects of global warming will be especially intense, and its impacts will be seen faster on species inhabiting cooler aquatic habitats. Cooler habitats are likely to warm more quickly, and species inhabiting these habitats will either adapt to increasing temperatures or continue to persist, be forced to disperse to other suitable habitats, or become locally extirpated. Gilabert et al., (2019) revealed that body size along with the mobility of taxa were the best predictors of 15 different centrality indices in the Gulf of California food web. However, they also underlined that only the mobility exhibited a significant positive relationship to the generalized linear models (GLMs) characterizing the trophic levels of the taxa. In our case, species with active or passive dispersion abilities will likely be the only ones able to move up to higher altitudes where there may be cooler microclimates. Thus, species persistence will likely be seen more often in cosmoecius species that have wide tolerance ranges to temperature. In our study, the five most common species are indeed widely distributed in different water bodies at different altitudes. These species do not show significant patterns of correlation to water chemicals or ions as far as the present study shows. This interesting finding opens a new window for future works indicating necessities of further studies in much broader geographical ranges (i.e., altitude, latitude) and perspectives.

Acknowledgments

We thank Kenneth Nussear (University of Nevada, Reno) for his comments and suggestions on the manuscript. Also, thanks are given to Ozan Yılmaz, Umutcan Gürer, Meriç Tanyeri, Çağatay Çapraz and Filiz Batmaz for their excellent assistance both in the field and laboratory works. Dr. Nusret Karakaya is also thanked for his help during ion analyses. This work was supported by the project (No: 2130172) from The Scientific and Technological Research Council of Turkey (TÜBİTAK).

Appendix 1

Appendix 2

Appendix 3

Appendix 3A A, Candona neglecta; B, C. weltneri; C, Cypridopsis vidua;D, Cypris pubera; E, Darwinula stevensoni; F, G (postero ventral margin of left valve), H (anterior margin of right valve), Hemicypris cf. largereticulata; I, Herpetocypris helenae; J, H. intermedia;K, H. reptans; L, M, Heterocypris incongruens; N, H. reptans; O, H. rotundata; P, H. salina; R, Ilyocypris bradyi (female); S, I. bradyi (male); T, I. gibba; U, I. inermis; V, I. monstrifica; Y, Isocypris beauchampi; Z (female), W (male), Plesiocypridopsis newtoni.

Appendix 3B A, Limnocythere inopinata; B, Potamocypris arcuata; C, P. fallax; D, P. smaragdina; E, P. variegata; F, P. villosa; G, Prionocypris zenkeri; H (Left valve external view), I (Right valve internal view) of Pseudocandona albicans; J, Psychrodromus fontinalis; K (Right valve external view), L (Left valve external view) of P. olivaceus; M, N, Pseudostrandesia n. sp.; O (posterior margin), Candona weltneri; P (valve surface), Ilyocypris gibba.

References

All Tables

All Figures

	Fig. 1 Distribution of 117 sampling sites from 13 counties (Akdeniz, Anamur, Aydıncık, Bozyazı, Çamlıyayla, Erdemli, Gülnar, Mezitli, Mut, Silifke, Tarsus, Toroslar, Yenişehir) of Mersin, Turkey.
In the text

	Fig. 2 Length (L) and height (H) measurments of left (L) and right (R) valves of the species (A) H. salina and (B) I. bradyi at different altitudinal ranges.
In the text

Fig. 3 CCA diagrams of (A) five variables with 12 taxa, and (B) with sampling sites shown with empty circles. Abbreviations: Candona sp (Csp), Cypridopsis vidua (Cv), Heterocypris incongruens (Hi), H. reptans (Hre), H. salina (Hs), Ilyocypris bradyi (Ib), I. inermis (li), Limnocythere inopinata (Li), Potamocypris fallax (Pof), Prionocypris zenkeri (Pz), Psychrodromus fontinalis (Pf), P. olivaceus (Po), Sal (salinity), Alt (altitude), DO (dissolved oxygen), TW (water temperature).

In the text

Appendix 3A A, Candona neglecta; B, C. weltneri; C, Cypridopsis vidua;D, Cypris pubera; E, Darwinula stevensoni; F, G (postero ventral margin of left valve), H (anterior margin of right valve), Hemicypris cf. largereticulata; I, Herpetocypris helenae; J, H. intermedia;K, H. reptans; L, M, Heterocypris incongruens; N, H. reptans; O, H. rotundata; P, H. salina; R, Ilyocypris bradyi (female); S, I. bradyi (male); T, I. gibba; U, I. inermis; V, I. monstrifica; Y, Isocypris beauchampi; Z (female), W (male), Plesiocypridopsis newtoni.

In the text

Appendix 3B A, Limnocythere inopinata; B, Potamocypris arcuata; C, P. fallax; D, P. smaragdina; E, P. variegata; F, P. villosa; G, Prionocypris zenkeri; H (Left valve external view), I (Right valve internal view) of Pseudocandona albicans; J, Psychrodromus fontinalis; K (Right valve external view), L (Left valve external view) of P. olivaceus; M, N, Pseudostrandesia n. sp.; O (posterior margin), Candona weltneri; P (valve surface), Ilyocypris gibba.

In the text