Ulmer (1939) established the genus Rhithrogeniella to accommodate the species Rhithrogeniella ornata Ulmer, 1939 known from imagos and subimagos from Java and Sumatra. According to Ulmer (1939), the genus was characterized by genitalia resembling those of Rhithrogena, with two simple lobes lacking spines or titillators. On the other hand, the new genus differed from Rhithrogena in the tarsal proportions on fore- and hind legs. Interestingly, in the key to the genera (Ulmer 1939), Rhithrogeniella is keyed with the genus Afronurus Lestage, 1924 from which it differs by the tarsal composition of the foreleg. During the following years, nothing substantial was added to the knowledge of the genus, and keys which included Rhithrogeniella were based on Ulmer’s description and drawings (Tomka and Zurwerra 1985; Tshernova 1974).

Major advancement was made by Soldán and Braasch (1986) who described a new species from Vietnam (Rh. tonkinensis) based on subimagos, together with the first description of the nymphal stage. This nymph exhibits typical Ecdyonurinae morphology, with scattered setae on the maxillae. The genus can be distinguished from all relatives by the peculiar structure of the cerci and terminal filament which possess a row of stout setae in the proximal part and bunches of long and thin setae in the medial and distal parts. Later, Rhithrogeniella tonkinensis was reported from Thailand and the male imago described (Braasch 1990). According to figures of Braasch (1990), the genitalia bear median titillators, the penis lobes are much more rounded than in Rhithrogeniella ornata, and the styliger fig is of an unusual shape with two triangular sublateral processes. In their revision of worldwide Heptageniidae, Wang and McCafferty (2004) proposed several nomenclatorial changes; in particular, they combined Rhithrogeniella ornata with the genus Rhithrogena, hence placing Rhithrogeniella in synonymy with Rhithrogena, and they assigned Rhithrogeniella tonkinensis to the genus Ecdyonurus (E. tonkinensis) based on subimaginal, larval and egg morphology. This account was never discussed later on and Braasch and Boonsoong (2010) mentioned the presence of Rhithrogeniella ornata in West Malaysia based on male subimagos, and Boonsoong and Braasch (2013) listed Rhithrogeniella tonkinensis in the Heptageniidae fauna of Thailand.

Two questions need to be resolved. Are Wang and McCafferty (2004) correct in synonymizing Rhithrogeniella with Rhithrogena (subfamily Rhithrogeninae), and assigning Rhithrogeniella tonkinensis to the genus Ecdyonurus (subfamily Ecdyonurinae)? Is the association between unreared nymphs and subimagos of Rhithrogeniella tonkinensis accurate or not?

The type material of Rhithrogeniella ornata, deposited in the collection of the Zoological Museum of Hamburg University, Germany (ZMH) has been reinvestigated together with new material from Sumatra. It is now possible to provide the first description of the nymph of Rhithrogeniella ornata.

Material studied here is deposited in the following institutions:

Zoologisches Museum und Biozentrum Grindel, Hamburg, Germany [ZMH]

Musée cantonal de zoologie, Lausanne, Switzerland [MZL]

Lembaga Ilmu Pengetahuan Indonesia (Indonesian Institute of Sciences), Museum of Zoology, Bogor, Indonesia [LIPI] (Bogor was formerly known as Buitenzorg)

Drawings were made with the help of a camera lucida taken from stereomicroscope Leica DM 750 and pictures from microscope Zeiss Axioscop 2 or Visionary Digital Passport II. Final digital drawings were performed on Adobe Illustrator CS6. For scanning electronic microscope (SEM) pictures, the eggs were dehydrated, carbon coated, and observed under a LEO 1525 at 5.00kV; maxillae were dehydrated, critical point dried, and then platinum coated, and observed under a FEI Quanta 250 at 5.00kV. Final figs were assembled in Adobe Photoshop CS6.

Nymphs and adults were associated with the help of the egg structure (Fig. 5).

The genitalia of the male imago differ slightly from those described by Ulmer (1939, page 577, Fig. 169), being wider and less cylindrical than illustrated. The presence of a transverse suture on the mesonotum together with the shape of the depression of the furcasternum (not narrowed anteriorly) indicates that Rhithrogeniella ornata cannot be a member of Rhithrogeninae; thus, it is not a synonym of Rhithrogena as suggested by Wang and McCafferty (2004). Moreover, the presence of scattered setae on the ventral side of the maxilla is a character only found among members of Ecdyonurinae, as suggested already by Soldán and Braasch (1986). When using the key of Webb and McCafferty (2008), Rhithrogeniella will key to the genus Afronurus Lestage, 1924 for the adults and to the genus Nixe Flowers, 1980 for the nymphs. Nearctic workers consider the genus Nixe as valid (Flowers 1986; McCafferty 2004; Wang and McCafferty 2004; Webb and McCafferty 2008), but European authors think that its concept is similar to Paracinygmula Bajkova, 1975 (Jacob et al. 1996). The Holarctic species joernensis (Bengtsson, 1909) is therefore treated either as Paracinygmula joernensis (Bauernfeind and Soldán 2012) or as Nixe joernensis (Kjaerstad et al. 2012). Nymphs of Rhithrogeniella share some characters with this concept, including the presence of swimming setae on cerci and terminal filament, and the chorionic structures of the eggs (Flowers 1980). Nymphs of Nixe/Paracinygmula however present gills with a weakly developed fibrillose part, either absent or reduced to a single filament in gill VI, which is not the case in Rhithrogeniella (Fig. 22 and Soldán and Braasch, 1986, fig. 4). Contrary to Nixe/Paracinygmula, the male genitalia have a very different shape and lack well developed median titillators as well as basal sclerite spines.

Three species of Nixe known only from the nymphal stage are reported from Taiwan (Kang and Yang 1994). Bauernfeind and Soldán (2012) transferred them to the genus Paracinygmula without new data, because they considered Nixe as a subjective junior synonym of Paracinygmula. Examination of paratypes of these species, deposited in the collections of MZL, revealed that they perfectly match the concept of Rhithrogeniella developed here, and therefore the following new combinations are proposed:

Rhithrogeniella littoralis (Kang and Yang 1994) comb. n. (= Nixe (Nixe) littoralis Kang and Yang 1994 =Paracinygmula littoralis Bauernfeind and Soldán 2012;

Rhithrogeniella mitifica (Kang and Yang 1994) comb. n. (= Nixe (Nixe) mitificus Kang and Yang 1994 =Paracinygmula mitifica Bauernfeind and Soldán 2012;

Rhithrogeniella obscura (Kang and Yang 1994) comb. n. (= Nixe (Nixe) obscurus Kang and Yang 1994 =Paracinygmula obscura Bauernfeind and Soldán 2012.

Nixe/Paracinygmula is therefore restricted to the Holarctic Realm, whereas Rhithrogeniella is Oriental, reported from Taiwan, continental Southeast Asia and from Java and Sumatra in the Sunda Islands. The genus is presently recorded neither from Borneo (Braasch 2011; Sartori et al. 2003) nor from the Philippines (Braasch 2011).

Based on the Bayesian majority-rule consensus tree reconstructed from the combined data set in Vuataz et al. (2013), Rhithrogeniella appears more related to the tribe Compsoneuriini sensu Sartori (2014) (Compsoneuria, Compsoneuriella and Notonurus), than to other Ecdyonurinae (Thalerosphyrus, Asionurus, Atopopus, Afronurus), although low posterior probability and bootstrap support does not allow to determine its exact relationships. It is possible that further studies may show that a new tribe should be established to accommodate this genus.

One remaining question concerns the presence or absence of titillators on Rhithrogeniella male genitalia. These structures are mentioned by Soldán and Braasch (1986) in the male subimago of Rhithrogeniella tonkinensis as well as in the subimago of Rhithrogeniella ornata (Soldán and Braasch 1986, page 204). Although we have not dissected the holotype (the only male imago of Rhithrogeniella known at the moment), we feel confident that this specimen lacks median titillators. The structures of the subimago male genitalia, illustrated in Fig. 4, are not “well-developed, cylindrical medial titillators with sclerotized apices” (Soldán and Braasch 1986), because they are only cuticular processes, weakly sclerotized except at the apex which is spine-like. In all Ecdyonurinae subimagos which do possess true titillators, these structures are deeply sclerotized, profoundly rooted inside the penis lobes, and are present in the imaginal stage after the subimaginal molt. The cuticular processes mentioned in Rhithrogeniella are thus likely to disappear with the subimaginal molt. We conclude therefore that, to our present knowledge, Rhithrogeniella lacks true titillators. The supposed male imago of Rhithrogeniella tonkinensis briefly described by Braasch (1990) possesses median titillators as well as a very curious styliger fig, with two large triangular processes. These processes should already be present in the subimago and easily visible; but because they are present neither in the male subimago of Rhithrogeniella ornata nor Rhithrogeniella tonkinensis, we can conclude that the male of Braasch (1990) is misassociated and possibly belongs to a species of Afronurus.

Rhithrogeniella ornata appears to be closely related to Rhithrogeniella tonkinensis, known from Vietnam and Thailand. It differs from the latter mainly by the ornamentation of the crown of the galea-lacinia, with 13–14 comb-shape setae, median ones with 10–11 teeth, whereas Rhithrogeniella tonkinensis bears only 10–11 comb-shape setae, median ones with 6–8 teeth. Additional nymphal characters, and egg chorionic structure are also very similar. Differences between subimagos of both species proposed by Soldán and Braasch (1986) are tenuous, and rely mainly on the tarsal composition of the hind leg (1=2=5>3≥4 in Rhithrogeniella ornata compared to 1=5>2=3>4 in Rhithrogeniella tonkinensis). Tibia of foreleg is distinctly shorter than the femur in Rhithrogeniella ornata, whereas it is reported as subequal to the femur in Rhithrogeniella tonkinensis. Subimaginal male genitalia are rather similar, although penis lobes appear more rounded in Rhithrogeniella ornata than in Rhithrogeniella tonkinensis.

Compared to the Taiwan species, Rhithrogeniella ornata can be easily separated from Rhithrogeniella littoralis and Rhithrogeniella obscura by the shape of the mandibles with inner and outer incisors subequal in length (inner incisor much shorter in Rhithrogeniella littoralis and Rhithrogeniella obscura), from Rhithrogeniella mitifica and Rhithrogeniella obscura, by the higher number of teeth on the comb-shape setae of the galea-lacinia (4–5 teeth only in Rhithrogeniella mitifica and Rhithrogeniella obscura vs 10–11 in Rhithrogeniella ornata), from Rhithrogeniella mitifica by the shape of the spines on the posterior margin of the tergites (pointed in Rhithrogeniella ornata vs tabular in Rhithrogeniella mitifica), and from Rhithrogeniella littoralis by the much more elongated gill VII.