IXALIDIIDAE, Hemp, Song & Ritchie, 2025

Hemp, Claudia, Ritchie, J. Mark, Cigliano, Maria Marta, Heller, Klaus-Gerhard, Warchalowska-Śliwa, Elżbieta, Grzywacz, Beata, Linde, Jackson, Uluar, Onur, Ngoute, Charly Oumarou & Song, Hojun, 2025, Out of sight, out of mind? Ixalidiidae, a new family of African forest grasshoppers (Orthoptera, Acridoidea) revealed by molecular phylogenetics and genital morphology, Zoosystema 47 (24), pp. 489-553 : 532-540

CONTEXTUALIZING IXALIDIIDAE HEMP, SONG & RITCHIE N. FAM. WITHIN THE ACRIDOIDEA

Having established the value, or “phylogenetic signal” of male genital morphology in reconstructing phylogeny in Acridoidea, this section examines the individual characters and compares the character states that are found across the Acridoidea, including the new family Ixalidiidae Hemp, Song & Ritchie n. fam., in a summary table ( Table 13 ) before looking in more detail at a few of these families to clarify the distinct identity of the Ixalidiidae Hemp, Song & Ritchie n. fam. The following comparison of Ixalidiidae Hemp, Song & Ritchie n. fam. with its nearest neighbours in the phylogenetic tree on the basis of genital morphology builds on Song & Mariño-Pérez (2013) with some minor areas of divergence. Table 13 presents twenty-four significant genitalic and other morphological characters with potential to shed light on the relationships of Ixalidiidae Hemp, Song & Ritchie n. fam. within the core clade of the Acridoidea, which (excepting Pamphagodidae and Pamphagidae ) all share the possession of a bridge-shaped epiphallus with lophi ( Song & Mariño-Pérez 2013; Song et al. 2015).

CHARACTERS OF THE MALE GENITALIA IN ACRIDOIDEA, INCLUDING IXALIDIIDAE HEMP, SONG & RITCHIE N. FAM.

Supra-anal plate

The form of the supra-anal plate or epiproct, though not an internal genitalic structure, is included here and in Table 13 because it is evidently under sexual selection (e.g. see Tangana ). It is divided into basal and apical portions by a transverse sulcus in the Ixalidiidae Hemp, Song & Ritchie n. fam. (see Figs 5; 7) as well as in Tristiridae (Cigliano 1989) , all three subfamilies of Ommexechidae ( Ronderos 1973, 1978; Carbonell & Mesa 1972), and many genera of Lentulidae . Among Romaleidae most Romaleinae have the supra-anal plate divided, except for Diponthus Stål, 1861 , Gurneyacris Liebermann, 1958 and Zoniopoda Stål, 1873 , while the Acrididae in general appear to lack the transverse division.

Epiphallus

Possession of a bridge-shaped epiphallus is shared by Pyrgacrididae , Tristiridae , Lentulidae , Ommexechidae , Romaleidae and Acrididae , and also by the Ixalidiidae Hemp, Song & Ritchie n. fam. The possession of epiphallic lophi ( Roberts 1941: 244) characterizes all Acridoidea apart from the Pamphagodidae and Pamphagidae . However, the shape of the lophi (pointed in Tristiridae and Ommexechidae , polymorphic in Lentulidae , or lobiform in Romaleidae and Acrididae ) is evidently homoplasious above family level ( Song & Mariño-Pérez 2013). The lophi are pointed in Ixalidiidae Hemp, Song & Ritchie n. fam. (though with a flattened tip in Tangana ).

The presence of epiphallic ancorae ( Roberts 1941: 241) is found in all families of Acridoidea apart from the basal groups, Pamphagodidae and Pyrgacrididae ( Song & Mariño-Pérez 2013, figure 5). However, when morphological characters are mapped onto the mitochondrial genome tree the character appears homoplasious, having developed independently in Pamphagidae as well as in the common ancestor of the Lentulidae , Tristiridae , Ommexechidae , Romaleidae and Acrididae . Ancorae must subsequently have been lost in the Ommexechidae ( Song & Mariño-Pérez 2013, figure 5) and they are also largely absent from the Ixalidiidae Hemp, Song & Ritchie n. fam., though possibly incipient in Mazaea ( Fig. 7C).

The lateral sclerites ( Roberts 1941: 245) (= oval sclerites ( Snodgrass 1935)) flanking the epiphallus are present in the eight core acridoid families included in Table 13 , but absent in Pamphagidae and of uncertain homology in Pamphagodidae ( Dirsh 1956) . They are also present in the family Ixalidiidae Hemp, Song & Ritchie n. fam., though sometimes closely articulated or partly fused with the lateral plates of the epiphallus.

Lateral evaginations of the epiphallic membrane ( poches dorso-latérales and poches latérales) have been noted in Pyrgacrididae ( Descamps 1968: 34 35; Fig. 11). Similar and possibly homologous lateral lobes have been illustrated in the epiphallic membrane of Tropidostethus Philippi, 1863 and Elysiacris Rehn, 1942 ( Tristiridae ) (Cigliano 1989: figures 172, 187). Amédégnato (1976: 7) indicated the occasional presence of latero-ventral sclerites of the epiphallic layer. These have also been shown in Paulinia acuminata (De Geer, 1773) ( Acrididae , Pauliniinae ) by Carbonell (2000: 174; Figs 18; 19). These latero-ventral sclerites of the epiphallic membrane are quite distinct from the paired sclerites of the ventral lobe (ectophallic) ( Snodgrass 1935; Roberts 1941; Dirsh 1956) found in most Acrididae and also in Ixalidiidae Hemp, Song & Ritchie n. fam. Too little is known about the occurrence of these epiphallic structures across the Acridoidea for them to be included in Table 13 . They have not been found in Ixalidiidae Hemp, Song & Ritchie n. fam.

Ectophallus

This layer is largely constituted by the cingulum ( Roberts 1941) a dorsal covering over the endophallus with differentiation of apodemes, rami, and in some families an arch, with or without a dorsal pair of aedeagal sclerites, all of which are discussed below. The degree of sclerotization of the ectophallus was said by Song & Mariño-Pérez (2013) to distinguish the Acridoidea (fully sclerotised) from all other superfamilies (partly sclerotised). However, in the Ommexechidae the ectophallus is visibly not fully sclerotized ( Song & Mariño-Pérez 2013; Fig. 3G) as earlier indicated by Dirsh (1956: 247). In Tristiridae ( Cigliano 1989b: 56) described the ectophallus in Tristirinae Rehn, 1906 as having “a lower degree of sclerification”. In Ixalidiidae Hemp, Song & Ritchie n. fam. the ectophallus is mostly sclerotized, heavily so in Rowellacris Ritchie & Hemp n. gen. but less so in Tangana . The cingulum is reduced to a sclerotized skeletal framework in Ixalidium , Mazaea , Eubocoana and Barombia .

The apodemes of cingulum may be present or absent and, when present, may be long or short. Both Tristiridae and Ixalidiidae Hemp, Song & Ritchie n. fam. have species or genera in which although the cingulum is present as a sclerotised plate, it lacks differentiation into distinct paired apodemes, while in other genera there are taxa with either short or elongated apodemes (e.g. Atacamacris Carbonell & Mesa, 1972 in Tristiridae (Cigliano 1989: 53, Fig 5) and Mazaea , Eubocoana , Barombia and Ixalidium in Ixalidiidae Hemp, Song & Ritchie n. fam. have elongate apodemes). Both families therefore are considered as polymorphic for these characters ( Table 13 ).

The presence of blade-like secondary cingular apodemes underlying and parallel to the main apodemes of cingulum has been illustrated for a few taxa, both in line drawings and in photographic images, but their presence has not previously been remarked upon. Dirsh (1966: 100, 102) figured these apodemes clearly in the male genitalia of Mazaea and Barombia ( Ixalidiidae Hemp, Song & Ritchie n. fam.) (C. H. Rowell, personal communication, June 2020). In this study they have also been found in Ixalidium s. str. ( Fig. 11B), and their presence confirmed in Mazaea ( Fig. 7F), Eubocoana tristis ( Fig. 9C) and Barombia . Similar structures have also been illustrated in Eneremius desertorum Saussure, 1888 ( Lentulidae ) ( Brown 1962: fig. 4; Otte 2014, fig. 15) and in E. pusillum (Uvarov, 1925) ( Dirsh 1961: 396, fig. 21 (5); Otte 2014: fig. 17). They do not appear to be present in other families of Acridoidea, but it is possible they may have been overlooked.

The zygoma of cingulum was defined by Dirsh (1956: 228) as “a transverse dorsal part of the cingulum, connecting the apodemes and, in most cases, the cingulum itself with the apical valves of the penis”. It is shared by most Acridoidea and their sister group, the Pyrgomorphoidea. However, this character is an expression of a strongly sclerotized ectophallus, hence Cigliano (1989) stated that there is no zygoma in Tristiridae . However, the zygoma is present in all Ixalidiidae Hemp, Song & Ritchie n. fam., whether as a narrow transverse sclerite joining the cingular apodemes in Ixalidium , Mazaea and Barombia , or incorporated into a larger dorsal plate covering much of the endophallus in Rowellacris Ritchie & Hemp n. gen. In each case it connects ventrally to the arch of cingulum.

The rami of cingulum, which generally cover the sides of a part of the endophallus, are present in all the families in Table 13 , except for the Pamphagodidae and Pyrgacrididae if the interpretation of Descamps (1968) is correct.

The arch sclerite in Acridoidea, originally named “bridge of anterior phallotreme sclerites” ( Snodgrass 1935), was subsequently named “arch of dorsal valves” and “arch of aedeagus” by Roberts (1941: 241), who considered it “a development from the aedeagal valves or endophallic membrane rather than from the zygoma of the cingulum”. Both Dirsh (1956: 225) and Amédégnato (1976: 7-8, Plate II, Figs 12-15) subsequently described and illustrated the “arch of cingulum” as a sclerite of ectophallic origin connecting the cingulum and the endophallic sclerites. Amédégnato (op. cit.) stated that in groups with a well-developed aedeagus the arch could either fuse with the dorsal valves of endophallic origin or alternatively it could give rise to genuine “valves of cingulum”.

Eades (1962: 6-7) noted that “homologies of the arch, bridge and dorsal aedeagal sclerites have never been suggested except by speculation”, before speculating that in Dericorythinae Jacobson & Bianchi, 1905 (now Dericorythidae ), which he considered to be intermediate between Ommexechidae and Romaleidae and the Acrididae , the arch had initially developed as a pair of sclerites arising from the phallotreme membrane adjoining the “primitive” single pair of (ventral) aedeagal sclerites and extending dorsad to fuse with the ectophallic membrane at the rear of the cingulum, giving rise to a dorsal pair of aedeagal valves in some cases and subsequently becoming fused into a single structure bridging the ventral sclerites. Eades (1962: 6-7) indicated his belief that the arch in Dericorythidae was homologous with that in Acrididae , but he later ( Eades 2000:184) expressed the view that it is a pseudoarch “not homologous with the true arch found in Charilaidae [= Pamphagodidae ] and Acrididae ”. However, Song et al. (2018: 4) considered that the arch sclerite is homoplasious within the Acridoidea, having apparently evolved separately in Pamphagodidae and Acrididae .

Presence or absence of the arch sclerite is difficult to establish and requires dissection of the phallic complex ( Song et al. 2018: 13). Uvarov & Dirsh (1961: 153) argued that the arch was absent in several genera of Acrididae , “appearing sometimes only as a slightly sclerotised part of the ectophallic membrane”. The uncertainty derives from the variability in the degree of sclerotization of the structure in different genera and the subjectivity of a presence / absence decision. Song (2004) demonstrated that in Schistocerca Stål, 1873 the arch sclerite develops during the adult stage, being largely undeveloped at fledging, and only reaching its full size before sexual maturity. He suggested that this may have led to S. braziliensis Dirsh, 1974 being defined on the basis of immature material ( Dirsh 1974: 166).

Nonetheless, the presence of the arch of cingulum connecting the zygoma to the apical valves of the endophallus in most Acrididae and its absence in most Romaleidae has been considered of importance in establishing relationships ( Amédégnato 1976, 1977; Eades 2000). In the absence of any other synapomorphy of the Acrididae, Song et al. (2018: 13) concluded that “the fact that we have recovered the monophyletic Acrididae strongly suggests that this obscure genital character [the arch of cingulum] may indeed be a synapomorphy for the family”. In Melanoplus rotundipennis (Scudder, 1878) ( Acrididae ) Woller & Song (2017: 345, 351, 354) showed that the ‘arch of aedeagus’, arising from the dorsal valves and inserted into the lower region of the zygoma, provides through the zygoma a point of articulation and structural support for the aedeagus during mating.

Either a single arch sclerite or a pair of sclerites is present in all genera included within the family Ixalidiidae Hemp, Song & Ritchie n. fam. In the genus Rowellacris Ritchie & Hemp n. gen. the enlarged and expanded arch sclerite also appears to function as a stiffener and spacer that maintains the gap between the cingulum and the endophallus, otherwise usually connected only by membranes.

The presence in the aedeagus of sclerites of ectophallic origin or dorsal valves of the cingulum ( Amédégnato 1976), together with an arch sclerite, was regarded by Song & Mariño-Pérez (2013) as an autapomorphy of the Acrididae , although cingular valves and an arch of ectophallic origin are also present in Diponthus ( Romaleini, Pictet & Saussure, 1887 ) ( Uvarov & Dirsh 1961; Amédégnato 1976: 8; Pocco et al. 2023). Eades (1962: 6-7) stated that in Romaleidae and Ommexechidae the arch was only present when a dorsal pair of aedeagal sclerites were also present.

In the present study of the members of the Ixalidiidae Hemp, Song & Ritchie n. fam. a bifurcated horn-like posterior medial outgrowth of the cingular arch, possibly representing incipient valves of cingulum (C. H. Rowell, personal communication, June 2020), has been found appressed to the dorsal surface of the posterior section of the endophallus in Mazaea ( Fig. 18A, B). This structure was previously illustrated, but not commented on, in Barombia by Dirsh (1966, fig. 40) whose own dissected specimen has been photographed ( Fig. 18C). Posterior medial dorsal projections of the zygoma in Namatettix Brown, 1970 , Atopotettix Brown, 1970 ( Brown 1970: 495, 505) and Shelfordites Karny, 1910 ( Brown 1967: figs 1b,1c) ( Lentulidae ), that were considered by Brown as dorsal or cingular valves, may or may not be homologous with the structures found in Mazaea and Barombia .

The term “pseudoarch” was originally coined by Akbar (1966: 77) to describe the arch structure found in Pyrgomorphidae (specifically in Poekilocerus pictus Audinet-Serville, 1831 ). Akbar defined the pseudoarch as “a small transverse sclerite … developed in the distal part of the central membrane close to the base of the suprarami. It forms an inflection laterally, and carries dorsally a pair of valves of the cingulum”. Akbar’s drawings show that his pseudoarch was attached dorsally to the central membrane posterior to the zygoma rather than to the zygoma itself. In studies of the Tristiridae ( Amédégnato 1977: 49; Cigliano 1989a, b) the term pseudoarch has been used to describe a sclerite, said to be of uncertain origin, attaching dorsally to the rear edge of the cingulum and ventrally to the endophallus in the same position as occurs with the arch of cingulum in Acrididae , but in the absence of a zygoma and dorsal valves of cingulum in the aedeagus. This arch structure may be absent, reduced or prominent ( Cigliano 1989a) and is of significance in defining the subclades of a “ Tristira generic group” (= tribe Tristirini ) within the family. It is absent in those genera which lack an aedeagus (= without development of distal portion of endophallic sclerites, Cigliano 1989b), but well-developed in several genera, including Moluchacris Rehn, 1942 , Peplacris Rehn, 1942 ( Fig. 23A), Punacris Rehn, 1942 , Crites Rehn, 1942 , Paracrites Rehn, 1942 , Incacris Rehn, 1942 , and with incipient development in Bufonacris Walker, 1871 ( Fig. 23E), Tristira Brunner von Wattenwyl, 1900 and Circacris Ronderos & Cigliano, 1989 .

It is thus apparent that similar arch structures joining the cingulum to the aedeagus occur in different families of Acridomorpha, but there is no agreement as to their homologies or the appropriate terminology with which to describe them. A comparative morphological study of these arch structures would be a useful contribution to understanding the evolution of the genitalia in Acridomorpha.

Endophallus

Amédégnato (1976, 1977) considered the sclerotised endophallus to be constituted in three sections, anterior, middle and posterior. The anterior part consists of the paired endophallic apodemes; the middle part is a pair of sclerites, in Acrididae called the lateral plates of Roberts (1941), which strengthen and support the walls of the ejaculatory and spermatophore sacs, while the posterior part (which may be present or absent) participates in the formation of an aedeagus, where this is present. Between the middle part and the posterior part of the endophallus there may be a fracture, creating a division. Possession of a divided endophallus groups the Ixalidiidae Hemp, Song & Ritchie n. fam. with the Pamphagodidae and Pamphagidae together with the Pyrgacrididae , Lithidiidae , some Tristiridae ( Fig. 6) and some Lentulidae (see below). Other genera of Lentulidae ( Lentula Stål, 1878 , Eremidium Karsch, 1896 ) have an undivided endophallus ( Dirsh 1956: 244), while in many Acrididae the endophallus is flexured, but without a break. Song & Mariño-Pérez (2013: table 3) considered the flexure as equivalent to a division, but that interpretation is not followed here, as explained below.

The presence of a divided endophallus with basal and apical pairs of sclerites articulated rather than completely disconnected, was considered by Song & Mariño-Pérez (2013: 253) to group the Tristiridae , Lentulidae , Pamphagidae , Ommexechidae , Romaleidae and Acrididae . However their character state “endophallus articulated” actually combined two distinct types of linkage between basal and apical sclerites of the endophallus. Firstly there may be a visible disjunction or fracture ( Amédégnato 1976), forming a “hinge”, as noted by Song & Mariño-Pérez ( op. cit.) or, alternatively, there may be a spring-like thinning of the endophallus, the sigmoid flexure ( Roberts 1941; Dirsh 1956, 1961), which may achieve the same function without creating a disjunction. This is the condition in many Romaleidae and Acrididae , including Ommatolampidinae , e.g. Eujivarus Bruner, 1911 and Eugenacris Descamps & Amedegnato, 1972 ( Amédégnato 1976: figs 26, 28), Oedipodinae Walker, 1871 , e.g. Oedaleus Fieber, 1853 and Gastrimargus Saussure, 1884 ( Ritchie 1981, 1982), Acridinae ( Popov et al. 2019) and Catantopinae (e.g. Rowell et al. 2018). Conflating these two forms of endophallic linkage in the character matrix may potentially obscure significant differences. In most members of the acridid subfamily Hemiacridinae , the endophallus is completely separated into basal and apical parts ( Dirsh 1956: 155).

In Ixalidiidae Hemp, Song & Ritchie n. fam. the divided endophallus is hinged, with separate basal and apical parts, distinct but closely appressed anterior to the arch sclerite(s). The medial sclerites of the endophallus are dorso-ventrally flattened anteriorly and attenuated in Mazaea ( Fig. 7H), Barombia , Eubocoana and Ixalidium continuous with the basal sclerites ( Fig. 11D). In Ixalidium the conjoined medial sclerites are evidently flexible enough to allow the apical section of the endophallus to be folded upwards by as much as 30°, compressing the spermatophore sac. However, in Rowellacris Ritchie & Hemp n. gen. ( Fig. 15H) and Tangana ( Fig. 17T) the medial sclerites have been reduced almost completely, so that the hinge occurs between the anterior basal section (the endophallic apodemes) and the apical sclerites which, together with an ectophallic sheath, form the aedeagus.

The ejaculatory sac is normally ventral to the basal valves of the endophallus in the families of Acridoidea listed in Table 13 , but the sac is apparently vestigial in Rowellacris Ritchie & Hemp n. gen., in which the ejaculatory duct is only minimally widened before becoming internalized within the fused basal valves of the endophallus. In Tangana , the ejaculatory duct also passes caudad within the fused basal valves of the endophallus to the dorsally-positioned spermatophore sac. However, there appears to be a small sac, appended to the ejaculatory duct well forward of its point of entry into the basal valves of the endophallus, which is not always preserved during dissection of the genitalia. This may represent the reduced ejaculatory sac.

The presence of a gonopore, defined as a constriction between the ejaculatory sac and the spermatophore sac ( Snodgrass 1935), was regarded by Song & Mariño-Pérez (2013) as distinguishing the Acridoidea from all other superfamilies. However, in Ixalidiidae Hemp, Song & Ritchie n. fam., while the ejaculatory sac is constricted at its junction with the ejaculatory duct in Ixalidium , Mazaea and Barombia , in both Rowellacris Ritchie & Hemp n. gen. and Tangana , the presence of a gonopore is currently inferred rather than observed, due to the vestigial condition of the ejaculatory sac described above.

The gonopore processes are pointed postero-ventral protrusions of the endophallic apodemes (basal valves of the endophallus), constricting the gonopore in Acridoidea. Though initially regarded as absent in Pyrgomorphidae ( Kevan et al. 1969: 185, 231), they were subsequently identified with the endophallic sclerites ( Eades & Kevan 1974: 250). They were scored by Song & Mariño-Pérez (2013, table 3) in their character matrix as present only in the Acrididae and Romaleidae and absent from the Tristiridae and Ommexechidae , so that when morphological characters were superimposed onto their mitochondrial genome tree ( op. cit., fig. 5B) the presence of gonopore processes was shown as a synapomorphy of the Acrididae , Ommexechidae and Romaleidae , that had subsequently been secondarily lost in the Ommexechidae . However, Eades (1961: 162) had previously illustrated the presence of gonopore processes in the Ommexechinae , in contradiction to Dirsh (1956: 247), confirming the synapomorphy across all three families. According to the current paradigm in which the ventral endophallic sclerites of Pyrgomorphidae , Lentulidae and Tristiridae are considered to be homologous with the gonopore processes of Acrididae , gonopore processes must, by definition, be present in all those families, making this a synapomorphy that also includes Lentulidae and Tristiridae .

Prior to their synonymy with Lentulidae , the medial sclerites of the endophallus in Lithidiidae were also identified as enlarged gonopore processes ( Eades 2000: 194). Table 13 therefore reflects this probable state. Given the apparent sister status of Ixalidiidae Hemp, Song & Ritchie n. fam. to Tristiridae in the phylogenetic tree derived from the mitochondrial genome ( Fig. 1), it is likely that the medial sclerites of the endophallus in Ixalidiidae Hemp, Song & Ritchie n. fam. are also derived from enlarged gonopore processes and thus homologous with the ventral branch of the endophallus in Tristiridae .

Eades (2000: 185) describes additional small sclerites appended to the gonopore processes in Ommexechidae and some other acridoids which he termed antero-ventral flanges of the endophallic sclerite. In Tristiridae Cigliano (1989b: 56) noted that “a projection (gonopore process?) arises ventrally from the anterior region, the development of which is variable. This projection is barely hinted at in Elasmoderini and Atacamacridinae . In Tropidostethini it presents a greater development, joining the dorsal endophallic sclerite through a zone of lesser sclerification. In Tristirinae it is prominent”. If the original gonopore processes have become the ventral endophallic sclerites in Tristiridae , as proposed by Eades (1962), then perhaps these ventral projections of the gonopore processes ( Fig. 23B) represent the antero-ventral flanges of Eades (2000). In Ixalidiidae Hemp, Song & Ritchie n. fam. these antero-ventral flanges of the endophallus are either absent or incipient (in a medial position in Mazaea and Ixalidium ( Figs 7G; 10C).

The spermatophore sac is positioned distal to the ejaculatory sac in Acridoidea. It is placed dorsally in relation to the endophallus in Pamphagodidae , Pamphagidae , Pyrgacrididae , Lentulidae and in Ixalidiidae Hemp, Song & Ritchie n. fam. However, it is situated ventrally, below the endophallus, in Ommexechidae and Romaleidae . In Tristiridae the spermatophore sac is situated between the dorsal and ventral branches of the endophallic sclerites and therefore in a dorsal position relative to the ventral endophallic sclerites. Song & Mariño-Pérez (2013: 250) scored the spermatophore sac in Acrididae , uniquely, as “transverse” rather than ventral since part of the sac is situated above the flexure of the endophallus in Acrididae . If it were regarded as ventral, this character state would be an uncontroverted synapomorphy of this terminal clade of the Acridoidea.

A revised character table for the Acridoidea

The basal families of the Acridoidea clade are the Pamphagidae and Pamphagodidae which are consistently recovered as sister clades ( Leavitt et al. 2013; Song et al. 2015, 2020 and Fig. 1), Both families lack the bridge-shaped epiphallus bearing lophi that is shared by the Pyrgacrididae and the remaining eight families (including Ixalidiidae Hemp, Song & Ritchie n. fam.) of the core clade of Acridoidea. Accordingly, although they have been included in Table 13 , they are not considered in depth in this study.

Table 13 uses a traffic light approach to indicate the congruence of character states between Ixalidiidae Hemp, Song & Ritchie n. fam. and the other acridoid families, with green for full congruence, amber for partial congruence and red for incongruence. Most of the characters defined by Song & Mariño-Pérez (2013) have been used, with some modification and augmentation, including the addition of characters of the male supra-anal plate (epiproct) and the female genitalia and the omission of a few characters which have identical character states for all families, or which appear to be inapplicable, unclear, or overly subjective. An extensive survey of relevant literature indicates that some of the male genitalic characters found to be monomorphic in the exemplar taxa studied by Song & Mariño-Pérez (2013) are in reality polymorphic at family level. In Table 13 , out of a total of 24 genital characters analyzed, just 14 characters are found to be unambiguously monomorphic for the Ixalidiidae Hemp, Song & Ritchie n. fam. Among those, the largest number of monomorphic character states shared with another family is 10 with the Lentulidae and Pyrgacrididae , followed by nine with Pamphagodidae , Pamphagidae , Tristiridae and Acrididae , eight with Ommexechidae , and six with Romaleidae .

The interpretation of individual characters and their significance and application in some of the core families of Acridoidea are further examined in the Discussion section.

BIOACOUSTICS

Up to now, in species of Ixalidiidae Hemp, Song & Ritchie n. fam., no sound producing organs or specialized modified structures have been found. Nevertheless, when kept in captivity, males of three species of Rowellacris Ritchie & Hemp n. gen. ( R. obscuripes n. comb., R. usambarica n. comb., R. sp. (Lutindi W Usambara)) and two putative Tangana species ( T. asymmetrica , Tangana sp. (coastal Tanzania and Kenya, Zanzibar)), as well as females (documented as Tangana sp. only), displayed the ability to generate relatively loud rhythmic sounds through drumming/tapping with their hind knees on the substrate. The observed echemes, consisting of 9-27 impacts with varying rates (12-16 Hz for Rowellacris Ritchie & Hemp n. gen. spp., Tangana asymmetrica , and 35-38 Hz for Tangana sp. (T=23-27°C; Fig. 20), were produced at irregular intervals. Both legs were moved largely in phase. Females were observed to either respond to male signals or spontaneously produce similar signals.

CYTOTAXONOMY

All three examined genera, namely Ixalidium , Rowellacris Ritchie & Hemp n. gen., and Tangana ( Fig. 23 A-J), exhibited a diploid chromosome number of 2n = 25 in males and 2n= 26 in females, with a sex chromosome system of X0 in males and XX in females. The autosomes displayed a gradual reduction in size, while the sex chromosome (X) was acrocentric. During male spermatogonial mitosis and meiosis, C-positive blocks were consistently observed in the paracentromeric region of all chromosomes, with interstitial C heterochromatin bands present in the sex chromosome of Tangana asymmetrica ( Fig. 21A, D, H).

Silver staining revealed the presence of two active nucleolar organizer regions (NORs) per haploid genome in I. sjostedti and one in R. usambarica n. comb. and T. asymmetrica . These NORs were situated in the paracentromeric region of two or one medium-sized bivalent, respectively ( Fig. 21B, E, I). In addition, a substantial cluster of 18S rDNA was detected during mitotic metaphase or within bivalents from diakinesis to metaphase I, coinciding with the active NORs identified by Ag-NOR staining ( Fig. 21C, F, G, J).

To further probe the chromosomal structure, fluorescence in situ hybridization (FISH) using the (TTAGG)n probe (tDNA-FISH) was performed on spermatogonial mitoses and/or spermatocyte nuclei during meiosis, specifically at diakinesis and metaphase I. In all analyzed taxa, signals were consistently detected at the distal ends of each chromosome. Notably, the tDNA-FISH signals on chromosomes of T. asymmetrica appeared notably stronger compared to those observed in the other species of Ixalidium and Rowellacris Ritchie & Hemp n. gen. ( Fig. 21J).