Sixteen water ponds were analyzed from Aguascalientes, México: 1. El Niagara 2. Tanque de los Jiménez 3. El Tepetate de Abajo, Mal Paso 4. Sierra Fría, Bordo 1: 5. Sierra Fría, Bordo 4; 6. Sierra Fría, Bordo 5: 7. Boca de Túnel; 8. Bordo Siglo XXI; 9. Tapias Viejas 1; 10. El Cedazo Park: 11. Rodolfo Landeros Park: 12. Pulgas Pandas; 13. UAA; 14. Los Gavilanes: 15. El Ocote: 16. Villa Hidalgo. We provide additional details of the study sites in Suppl. material 1: appendix A. We examined ephippia of 23 taxa (Table 1).

Using an acrylic tube (2 m × 7.5 cm), we randomly collected sediment samples (cores) from three different points at each study site. We used only the upper 3 cm of cores to extract potentially vital ephippia (Pérez-Martínez et al. 2013). Samples were stored in the dark in a plastic bag at 4 °C. To extract ephippia from sediments we used the sugar flotation method of Onbé (1978) modified by Marcus (1990). For some species, ephippia were collected from the littoral zone and in other cases from algal mats. Additionally, ephippia of Simocephalus mixtus (Sars, 1903) were obtained from a laboratory culture (see details in Suppl. material 1: appendix B).

We recorded a photomicrograph of the ephippia using a Nikon Eclipse light microscope (LM) with a digital camera DS-Fi2 under 4×, 10×, or 20× magnification. To initiate hatching, individual ephippia were placed in wells of a 96-well, polyethylene microplate (CELLTREAT® Scientific Products, 20 Mill St Ste 130, Pepperell, MA 01463) and incubated in a bioclimatic chamber in commercial water (Ciel®, Coca Cola®) under the conditions of 16:8 light/dark period, 20 °C temperature, and white light with an intensity of 345.50 ± 20.54 µmol s^-1 m^-2). Once hatched, the females were cultured until they matured, at which time they were identified using the key of Elías-Gutiérrez et al. (2008).

To analyze their ornamentation, ephippia were isolated and fixed in 4% formalin. For SEM study, specimens were dehydrated using a graded ethanol series (60, 70, 80, 90, 96%), after which a critical point drying was performed. Ephippia were attached to a SEM stub (1 cm high × 1.2 cm in diameter) and sputter coated with gold. All samples were observed under a SEM JEOL 5900 LV®, photomicrographs were taken to document ephippial characteristics.

Characterization of ephippia. To identify ephippia to species level we used six characteristics that we could see using LM and SEM: size, shape, color (including transparency), number of resting eggs, presence of ornamentation and/or accessory structure, and type of ornamentation or accessory structures.

Here we provide data of six morphological characteristics on ephippial morphology on 23 taxa (Figs 2–7). Most of our results provide evidence of taxon-specificity in ephippial morphology among the studied species, but there were several challenges. For example, identification of Ceriodaphnia species was problematic. When using light microscopy, Vanderkerkhove et al. (2004) also reported difficulty in identifying Ceriodaphnia laticaudata P.E. Müller, 1867, Ceriodaphnia pulchella (Sars, 1862), Ceriodaphnia quadrangula O.F. Müller, 1785, and Ceriodaphnia reticulata (Jurine, 1820). On the other hand, Kotov et al. (2018) were able to differentiate among four species. Berner (1985) in a study on the Ceriodaphnia cornuta complex, found high variability between species and proposed the separation of C. cornuta and C. rigaudi by comparing different details of the ephippia. In SEM studies, we also found that C. cornuta differed from other species in the genus. However, the smooth surface of ephippia in C. dubia, C. laticaudata, and C. reticulata made taxonomical distinctions difficult. Another example is seen in ephippia of S. mixtus: those from cultures showed clear ornamentation, while those from sediments did not. We also note that in some Daphnia species ephippia from sediment samples did not always retain their posterior spinulae. Although differences between some external ornamentations in cladocerans are linked to plasticity, the latter ephippial structure may degrade while in the sediment. Thus, ornamentations or appendages can be present when ephippia are formed in cultures, but may be absent in specimens collected from sediments, as seen in many of our Daphnia samples.

Does form follow function (Gould 1971) in cladoceran ephippia or is their morphology simply the residue of their formation while part of the female? This is a difficult question to answer. For example, ornamentation in the ephippium of Chydorus sphaericus complex is the same as the shell ornamentation in the female adult (see Fig. 2(8A–8C)). Despite the difficulty in coming to specific conclusions about the diversity of Chydorus sphaericus group (Kotov et al. 2016), ephippial eggs were still useful to identify the species group. In Daphnia species, spinules in the resting eggs are also present in the female (see Fig. 3(9–12A, B, C)). On the other hand, Moina micrura, Ilyocryptus agilis, Macrothrix mexicanus, and Simocephalus vetulus are clear examples of different ornamentations in diapausing eggs in comparison to the female. These examples could suggest that these structures possess a specific function during dormancy. However, we must remember that selection pressures do not operate on a single trait, they work on the entire animal within its environment: this is referred to as the Pareto optimality (Tendler et al. 2015). Thus, any research on ephippia must consider the importance of ephippial characteristics in the ephippial female, as well as while it resides in the sediment. For instance, the peduncle presents in Daphnia pulicaria Forbes, 1893 from Sierra Nevada (Spain) might be used for dispersion or to remain in the surface of the lake, as was observed by Guerrero-Jiménez et al. (2020). In S. vetulus, the margin of the ephippia is ornamented and the chamber in which the DE resides is orange eggs were mostly floating (see Suppl. material 1: appendix B). Different levels of carotenoids produce the orange pigmentation present in ephippia of cladoceran species permitting greater light absorption, which could improve hatching (Stross, 1966). On the other hand, spinules on ephippia of Daphnia spp. are probably a residual of the structures that protect adult females from predators.

In Mexico, approximately 150 cladoceran species have been reported (Elías-Gutiérrez et al. 2008), so our study has only covered a small proportion (15%) of the total diversity. This number is likely to increase simply because only 1% of watersheds in the area have been explored (Elías-Gutiérrez et al. (1999). While our work analyzed 23 morphotypes this represents the beginning of the studies in ephippia structures from subtropical areas. We urge that more work should be done on this subject, including paleolimnological research, such as an adaptation to a (sub)tropical context of the ‘Atlas of Subfossil Cladocera from Central and Northern Europe’ by Szeroczyfiska and Sarmaja-Korjonen (2007). Additionally, by combining light microscopy and SEM techniques, useful results could be obtained. An example of this is the remarkable work of Kotov (2013), who demonstrated that even in subfossil samples, good ultrastructure of the ornamentation could be obtained, leading to positive identifications at detailed levels.

In this work, we could identify some useful diagnostic traits for different morphotypes, even for some where high plasticity is well known, such as the Chydorus sphaericus complex, Daphnia, and Alona. However, we keep in mind that this condition will surely change when research of species increases. We recommend studying the ornamentations of ephippial structures to detect the boundaries for each taxon or complex of species, and ultimately, contribute to understand the evolution and biodiversity of water fleas in contemporary and past lakes. Our efforts augment current information which could shed light on cryptic speciation in cladocerans as several complex groups have been reported by Petrusek et al. (2003), Elías-Gutiérrez and Valdez-Moreno (2008), Quiroz-Vázquez and Elías-Gutiérrez (2009), and Bekker et al. (2016). While identification of cladoceran ephippia by itself has proven difficult (Murat and Sevil 2014), improving our ability to identify them to species level, will extend our knowledge of cladoceran distribution, especially to habitats that cannot be sampled throughout the year: i.e., those that are too remote and ephemeral ponds.