Ita korotyaevi, Meregalli & Borovec, 2011
Meregalli, Massimo & Borovec, Roman, 2011, Radiation in the halophytic coenoses of the Peri-Tethys: taxonomy and biogeography of the genus Ita (Coleoptera: Curculionidae), Journal of Natural History 45 (21 - 22), pp. 1331-1401: 1385-1398
treatment provided by
Ita korotyaevi sp. nov.
Holotype ♂: TURKEY, “ Turkey, 85.7 km N Aksaray, Lake Tuz, 18.VI.1999, Korotyaev ” ( ZIN).
A species of Ita vicariant of I. kirschi , characterized by the rostrum narrower at the base, more slender in lateral view, with sides linearly, regularly and scarcely broadened from the antennal insertion to the apex, the antennae inserted at a distance from the base slightly higher than the width of the rostrum at the antennal insertion, the elytra with elliptical scales, and the aedeagus curved, with slightly sinuate sides of the median lobe and sclerotized margins as broad as the median membranous part.
Description of the male. Base of rostrum, head, pronotum and elytra black; apex of rostrum, coxae, trochanters and antennal club dark ferruginous; femora, tarsi and antennal segments 6 and 7 ferruginous, tibiae, antennal scape and antennal segments Figure 19. Ita kirschi , Baku, ♂: (A, B) body; (C, E) rostrum; (G, H) aedeagus; (I) fore tarsus; (J) antenna; (K) scales of side of elytra and (L) dorsum and Ita kirschi , Baku, ♀: (D, F) rostrum. Scale bar: (A, B) 1 mm; (C–J) 250 µm.
1–5 yellow or light ferruginous (Figures 20A, B). Scales slender, whitish, 3.5 to 4 times as long as wide, moderately densely inserted on elytra, not completely covering integument, visible amidst scales, scales on pronotum slightly narrower on dorsum and shorter on sides; on legs smaller and less dense, alternate with thick semi-erect setae; tarsi with hair-like setae and some short narrow scales (Figures 20K, L). Rostrum medially robust, in lateral view distinctly curved and slightly gibbous above antennal insertion, sides in dorsal view strongly concave between base and antennal insertion, less wide at antennal insertion than at base at anterior margin of eyes, moderately and nearly linearly broadened from behind antennal insertion to apex; surface sparsely and minutely punctured in basal part, more shallowly punctulate towards apex. Scrobes oval, deep and fovea-shaped, margins shortly interrupted basad (Figures 20C, E). Antennal scape short, barely longer than rostrum at antennal insertion, curved forwards and sharply thickened in apical half; funicle short, segment 1 at apex twice as broad as segment 2, and twice as long as wide; segments 2–7 subquadrate, small and short; club elliptical. Pronotum with moderately and regularly curved sides, maximum width at mid-length. Elytra convex, distinctly broadened from base to slightly beyond mid-length, rounded at apex, in lateral view declivity distinctly rounded, uniformly curved from dorsum to apex. Tarsi slender, segment 1 scarcely broadened at apex, nearly four times as long as wide, segment 2 2.5 times as long as wide, very feebly conical, segment 3 with elongate, divergent lobes; onychium narrow, moderately and regularly broadened (Figure 20I). Aedeagus curved, median lobe with sclerotized margins broadened post-medially, as wide as membranous median part at point of maximum width; sides distinctly sinuate towards apical lamella, ostium elongate, elliptical; lamella broad, upward-curved (Figures 20G, H).
Description of the female and variation. Rostrum slightly longer, particularly from antennal insertion to apex, and more shallowly punctulate (Figure 20D, F); antennal funicle slightly longer, particularly segment 3 (Figure 20J). The other traits are very similar in both the sexes.
The variation is very limited.
We name this species after our good friend Boris Korotyaev ( ZIN), in acknowledgement of his friendship and his continuous support to our studies.
Boris Korotyaev (personal communications) found the specimens on Halocnemum strobilaceum (Pall.) Bieb.
This species is thus far known only from the type locality. The lake Tuz is one of the broadest and most important halophytic habitats of western Asia, and a large number of goosefoot species have been recorded there ( Çetiner 2006). It seems probable that other sectors of the coasts host populations of Ita korotyaevi . More halophytic Figure 20. Ita korotyaevi , paratype ♂: (A, B) body; (C, E) rostrum; (G, H) aedeagus; (I) fore tarsus; (K) scales of side of elytra and (L) scales of dorsum of elytra. Ita korotyaevi , paratype ♀: (D, F) rostrum; (J) antenna. Scale bar: (A, B) 1 mm; (C–H) 250 µm; (I, J) 200 µm.
This species appears to be a geographic vicariant of I. kirschi . The latter has more flattened elytra, not regularly curved from the dorsum to the apex in lateral view (Figure 20B vs 18B, 19B), rostrum shorter and narrower in lateral view (Figures 20E, F vs 18E, F and 19E, F), with more basal antennal insertion, and aedeagus less curved and with less broadened sclerotized sides (Figures 20G, H vs 18G, H and 19G, H). The other Anatolian species, I. caldarai , is very easily distinguished by the much larger scales, nearly completely covering the integument, the antennae inserted in the basal third of the rostrum, and the shorter tarsi (Figure 15).
Ecology, phylogeny and historical biogeography
The genus Ita ranges from Macaronesia to the western part of central Asia ( Figures 21 View Figure 21 , 22 View Figure 22 ), along a narrow strip between 29 and 41 ◦ N, where the species are strictly exclusive to halophytic environments. These habitats are relatively widespread in this region, being more sparse along the coasts of the Mediterranean sea and generally frequent along those of the large closed Caspian and Dead Seas; they also appear as isolated patches of vegetation in the otherwise desert regions of northern Africa and Turkmenistan, often in basins or depressions originated by subsidence phenomena.
The halophyte-rich ecosystems in which the former Chenopodiaceae , sometimes together with the Zygophyllaceae , form the dominant plant association, are part of several ecoregions ( Olson et al. 2001), the most significant being the Saharan halophytics (PA0905), an ecoregion occurring in isolated patches throughout Egypt, Morocco, Algeria, Tunisia, Libya, Western Sahara and Mauritania, and the corresponding Caspian lowland desert (PA1308) and central Asian southern desert (PA1312) ( National Geographic Society and World Wildlife Fund 2001). Salty wetlands also appear at higher altitude, in areas with a negative precipitation/evaporation balance, such as the eastern Anatolian montane steppe (PA0805) and the central Anatolian steppe (PA0803).
Throughout these regions, species of Ita appear to be exclusively associated with chenopod shrubs, including species of Atriplex L., Halocnemum M. Bieb. , Salicornia L., Salsola L. and Suaeda Forssk. ex J.F. Gmelin , that occupy the borders of brackish marshes or colonize the remnants of old lakes, now transformed into salt pans, where these plants are often, although not always, the dominant species. Such ecosystems are very stable, and in many cases developed as a consequence of the negative hydric balance triggered by the late Miocene climate change towards warmer temperature and dry conditions ( Griffin 2002; Micheels et al. 2009), that resulted in an accumulation of salt in the soil.
The specimens of Ita can be extremely abundant and represent the dominant weevil component in an optimal environmental situation, such as a flat, warm or hot plain with salty soil, or the margin of brackish marshes ( Figure 23 View Figure 23 ). The apparent rarity of many species, known from only one or very few specimens, is apparently because these halophyte stands are erroneously regarded as almost devoid of life and are thus neglected by collectors. Indeed, these weevils seem to require very stable and unaltered halophytic coenoses, and so can be considered as bioindicators of healthy ecosystems.
The insects crawl in daytime along the stems of the host plant ( Figure 24 View Figure 24 ) and feed upon the succulent stems and leaves, or hide in the debris at the base of the shrubs; all the specimens whose collecting dates are known were found between March and June; however, this is the main collecting period in the deserts and very few entomological expeditions have ever been organized in the second half of the year, so the lifespan of the imagines cannot be estimated. It seems certain that chenopods are also the hosts of the preimaginal stages, although it was not possible to set up a laboratory experiment to elucidate details of ecology and ethology, due to difficulty in keeping the host plants, which are strictly adapted to their peculiar habitat conditions, alive.
The dispersal ability of the weevils is probably limited and flight seems to be a very last resort: specimens observed under stress conditions, i.e. kept in full sun on a sheet of tissue, firstly tried to walk away, and only in rare instances opened their wings and flew; also in these cases the flight was more properly a “jumping flight”, with the insects landing at a distance of a few centimetres and starting again to walk. The wings are relatively small, with scarcely sclerotized veins, so they are probably unfit for long flights, notwithstanding the tininess of the weevils and the presence of setae on their lower margin. This, obviously, does not prevent single individuals being carried away, even long distances, by strong winds. Any consideration of historical biogeography of the genus should thus evaluate events of dispersal as well as dispersion along regions of ecological continuity.
The strong degree of adaptation of these insects to their habitat is reflected in the high morphological uniformity, shared by all the species. The shape of the body is extremely constant, and it is also deeply differentiated from all other Curculionidae . This was the main reason that led specimens of several species to be identified as “ Ita crassirostris ”, and the synonymy of I. gracilis and I. chobauti with I. crassirostris to remain unquestioned.
Attempts to infer phylogenetic relationships based on morphology are presented in Figures 25–27 View Figure 25 View Figure 26 View Figure 27 . No molecular analysis was of course possible, since most species are known from just a few, often very old, specimens, and sometimes only from the type. The analyses were based on both a multi-state matrix and a derived binary matrix; the resulting topologies were very similar, but those obtained from the binary matrix were more strongly supported.
The topologies of the trees obtained with the three different methods, Maximum Parsimony (MP), Maximum Likelihood (ML) and Bayesian Inference (BI) are remarkably similar, thus intrinsically suggestive of a realistic inference. All clades are shown, including those supported by less than 50%. This helps understanding of possible radiative events, even though the support for these nodes is not statistically significant.
In all the trees, the eastern Mediterranean and Asian species (“eastern” group) clustered apart from the western Mediterranean species (“western” group), with the “eastern” group occupying a basal position, and with I. caldarai intermediate between the two groups. A strongly supported monophyletic basal clade in BI and MP included I. latirostris and I. turkmena ; the same species clustered in a basal position also in ML, even though not as a monophyletic unit. These species appear to be sister to all the remaining taxa of the genus. Another strongly supported monophyletic unit, defined in all the inferences, included I. kirschi – I. korotyaevi . These two assemblages of vicariant species were also recognised in the descriptive analysis. I. friedmani and I. marina were also part of the “eastern” complex. The former clustered in a monobasic unit, sister to all the remaining species of the genus, with the exception of the I. latirostris complex, but with only weak support in BI and ML, and a very low support in MP. The precise affinity of Ita marina could not be unequivocally determined, since its position in the trees differed depending on the method applied: it was the most derived species of the “eastern” group in BI and MP, but with a quite weak support, and it was included in the I. kirschi – I. korotyaevi clade in MP, with moderate support.
Ita caldarai was basal to the “western” group, with a good to excellent support in all the topologies.
The “western” group was strongly supported in all the inferences, and three monophyletic lineages were constantly defined, with a relatively high support: the group I. berbera – I. gratiosa – I. hispanica – I. gracilis , the sister species I. chavanoni – I. punica and the sister species I. chobauti – I. crassirostris .
Based on the present day range of the genus, the relationships among the species as suggested by the phylogenetic inference, and considerations on the geological evolution of the Tethys and Paratethys basins, some hypotheses regarding its biogeographic history can be formulated.
According to Kadereit et al. (2006) the Chenopodiaceae tribe Salicornioideae (corresponding to the present-day Amaranthaceae subfamilies Salicornioideae and Salsoloideae) diversified in Eurasia during the Eocene–early Oligocene, along the northern margin of the Tethys. By the middle Miocene, all the major lineages had originated; also thanks to the onset of C 4 metabolism ( Pyankov et al. 2002), these halophytes and hygrohalophytes enjoyed a very successful radiation in the most arid, often salty, Mediterranean and Asian coenoses. Thus, it seems likely that it was not later than the first half of the Miocene that the association between Ita and the host plants had become fully established. This host plant association favoured the weevils’ diffusion, either as a dispersion or a dispersal, with a subsequent radiation in the regions dominated by this halophyte vegetation. The actual “mosaic” distribution, with species not closely related living in neighbouring localities, was probably determined by speciation events that occurred over a relatively long period and gave rise to independent lineages. Obviously, the apparently limited vagility of the extant species cannot be automatically attributed also to their ancestors, which may well have had greater dispersal ability. The present-day range of Ita is in agreement with a differentiation on the south-eastern coasts of the Paratethys sea, where the genus underwent a Miocene radiation that led several species to occupy halophytic coenoses, with the most basal groups inhabiting the coasts of the Caspian sea. During the Miocene in the eastern Paratethys, in the area corresponding to today’s region of Yerevan, shallow, brackish basins were connected to the Kura gulf and the Caspian Sea ( Popov et al. 2004, 2006). The Armenian plateau started uplifting towards the end of the middle Miocene ( Popov et al. 2004, 2006), and following the uplift these basins became landlocked. The understanding that the Yerevan region hosted halophytic coenoses with brackish water connected with the Caspian Sea is confirmed by the recent description of a Yerevan Miocene fossil species of Aphanius (fish, Cyprinodontidae ), a genus whose extant species live in coastal and brackish habitats along the Mediterranean Sea, and in the Caspian and Turanian region ( Vasilyan et al. 2009). Two species of Ita are nearly sympatric around Yerevan, Ita kirschi and Ita latirostris . The former is also present with morphologically undifferentiated populations along the coasts of the Caspian Sea, near Baku, whereas I. korotyaevi , the fully supported sister species of I. kirschi , is associated with the salty coenoses in central Anatolia. Ita latirostris is sister to Ita turkmena , native to the western part of the Turkmen desert. The phylogenetic analysis suggested that these last two species occupy the most basal position in the evolution of the genus: they probably derive from an ancestral upper-Miocene taxon, inhabiting the southern coasts of the South Caspian Depression, at least from the Kura gulf to the margin of the Turkmen desert (cf. Popov et al. 2006). After the Armenian uplift the populations at the extremes of the range became isolated and differentiated, originating the present day vicariant sister taxa I. latirostris and I. turkmena .
More debatable is the biogeographical analysis of the complex of I. kirschi . In this case, the presence of reciprocally undifferentiated populations in Yerevan and along the coasts of the Caspian Sea at Baku, and that of the only slightly differentiated I. korotyaevi in central Anatolia, would render such a reconstruction less likely. Unless the characters are considered to be extremely conservative and slowly evolving, it may appear more likely that the salty, high-altitude habitats of Armenia were colonized through dispersal events in more recent times, or that a certain gene flow is or was maintained between the Yerevan and Baku apparently isolated populations. Little can be said regarding the single male of I. kirschi found around Tiflis, if not that this area was also in connection with the Caspian coasts. Regarding I. korotyaevi , the halophytic vegetation in central Anatolia had a very ancient onset, at least dating back to the late Eocene, when the Tuz Gölü basin became isolated from the sea and extensive evaporites started being deposited ( Çemen et al. 1999). However, the relatively limited morphological variation between I. korotyaevi and I. kirschi would favour a more recent diversification, possibly following dispersion events from the Armenian habitats via eastern Anatolian halophytic coenoses.
Because of the uncertainty about its range and its position on the proposed phylogenies, nothing can be said about I. marina , except that this species undoubtedly belongs to the “eastern” group. Also the exact placement of I. friedmani , from the Negev desert and the coasts of the Dead Sea, is relatively uncertain, as the node from which it descends was only moderately supported. It is probably one of the most basal species in the evolution of the genus, and this is in good agreement with the antique origin of the salt deposits in the Dead Sea area, that were already widespread at least from the middle Miocene ( Neev and Hall 1979); in this epoch the ancestor of I. friedmani may have colonized this area.
In the middle Miocene, the Tethys ocean closed and the coasts of the Anatolian block came in continuity with the coasts of the Arabian block; the eastern coastal habitats united to the southern coasts of the Paratethys sea (cf. Meulenkamp 2003; Popov et al. 2004, 2006), and the genus, from its north-eastern quarters in the eastern Peri-Tethys, could reach the southern Tethys platform, where a further radiation occurred.
I. caldarai is basal to the “western” group, with very good support; if the inference is correct, the proposed reconstruction is upheld and I. caldarai can be seen as a remnant of the expansion of the range of the genus from the South Caspian Depression towards the coasts of the western Tethys.
According to the phylogenies, the species belonging to the “western” group occupy the terminal nodes of the clades. They are reciprocally less differentiated, and this fact is reflected in the low support of many of the internal nodes of this assemblage, which is however strongly supported. The three clades consistently inferred from all the analyses imply that distinct events of speciation followed one another; nowadays the ranges of the taxa derived from these events overlap in a “mosaic” distribution.
Several species of the “western” group are present in northern Africa; some of them live in marshes near the Mediterranean coasts, but Ita chobauti and I. chavanoni are localized inland in two of the broadest and more stable salty pans, in drainage plains that were formed by Miocene and Pliocene subduction events, respectively the Chott Melrhir and the Tamlelt plain. Basins and salty marshes in Tunisia, usually not far from the sea, are inhabited by I. punica . Populations of Ita must have already been widespread in this region during periods of marine incursions that, all along the central part of northern Africa, connected shallow basins with the Mediterranean sea, at least until the Late Tortonian (about 8–7 million years BP, cf. Meulenkamp et al. 2003). It is well known that the halophyte vegetation was never absent from the Sahara, although the territory occupied by this vegetation expanded and contracted several times during and after the Miocene, and up to only a few thousands years ago ( Geyh and Thiedig 2008). By the Pliocene, the Sahara desert had replaced the previously warm and humid subtropical vegetation, and patches of Salicornioideae were widespread throughout the whole of northern Africa ( Giresse 2008), providing corridors for the dispersion of the associated fauna. The currently closed depressions, the Algerian Chott Melrhir and the Tunisian Chott el Jerid, were originally connected to the Mediterranean Gulf of Gabes through the very large Gabes drainage system, that was disrupted in the Pliocene ( Griffin 2002; Goudie 2005). When the basins became landlocked and desiccated, the vegetation patches around water pools and marshes of increasing salinity were reciprocally isolated by more or less broad strips of barren desert. This scenario supports the slight, but distinct, morphological specificity observed in each of the nowadays isolated populations of I. punica . It might be expected that I. chobauti and I. punica , present in the lands derived from the Gabes drainage system, would be closely related, but the topologies of the trees, and also the analytical study of the species, indicate that they belong to different species-groups, I. punica being an eastern vicariant of I. chavanoni , a species that occupies the Tamlelt depression in south-eastern Morocco. Marine deposits in this basin date back to the Cretaceous, thus too far back in time, but Miocene marine deposits extend as far as north of Tendrara, not particularly distant from the depression. The Tamlelt Miocene deposits have a lacustrine origin, the most significant salt deposits being quite recent, from the Holocene (Du Desnay 1966; Chichagov 2008; H. Hamid, personal communication), so the presence of a species of Ita in this region should date back to this epoch. I. chavanoni and I. punica may thus derive from a common ancestor that inhabited the southern coasts of the Mediterranean Sea, and speciated in the post-Miocene halophytic coenoses.
The distribution of Ita chobauti seems to have been determined by different, perhaps more antique, events. This species is related to I. crassirostris from Sicily, and to a Tunisian taxon, only very slightly distinct from I. crassirostris . Ancestors of Ita chobauti may have reached the Chott Melrhir following patches of Salicornioideae in northern Africa, and have speciated after isolation when the basin became salty and landlocked in the Pliocene, while I. crassirostris may have reached Sicily during the Messinian salinity crisis, when part of the Mediterranean became a series of brackish lakes and desiccated tidal flats, a habitat ideal for extensive chenopod populations, about 6 million years BP ( Hsu 1983; Popov et al. 2004).
Thus, the phylogenetic analysis and the geological history of the western Mediterranean support the hypothesis that a first Miocene radiation in coastal Mediterranean coenoses may have given rise to the ancestors of each of the two lineages, I. punica – I. chavanoni and I. chobauti – I. crassirostris . Late- and post-Miocene events have triggered the speciation of the geographic vicariants in each of the two groups.
The third lineage includes the remaining “western” Ita species. The range of Ita hispanica can be interpreted either as a dispersal from northern African populations to the Iberian peninsula, or as a dispersion along the brackish lakes of the western Mediterranean basin during the Messinian salinity crisis. The relatively good differentiation of Ita hispanica with respect to the north African entities, and its well supported affinity with I. gracilis from Algeria, may strengthen this second hypothesis, which is furthermore reinforced by the concomitant presence of various wingless taxa sharing northern African affinities in the halophytic environments of south-eastern Spain (personal observations, and Osella and Meregalli 1986). The localization of I. hispanica in northern Spain, around the brackish basins of the Monegros area, was permitted by the peculiar geological evolution of the region. According to Garcia-Castellanos et al. (2003), the Ebro foreland basin started forming during the Palaeocene, becoming landlocked in the late Eocene, so that it was filled with alluvial, fluvial and lacustrine sediments. It remained closed at least until the late Miocene, when the endorheic fluvial system opened to the Mediterranean through the present Ebro River. In the same epoch, salt deposits started to accumulate, indicating that the basin water was brackish; it was during these times that the typical halophytic associations became fully established ( Pedrocchi-Renault and Sanz-Sanz 1991). The penetration of Ita hispanica in this area cannot precede this epoch, but it is probably more recent: the insects may have reached this apparently isolated region following patches of Salicornioideae associations along the Mediterranean coast and up the Ebro valley; this hypothetical scenario is supported by the morphological similarity between the Monegros and the Alicante populations, regardless of the relatively consistent distance and the apparent reciprocal geographical isolation.
The presence of a species of Ita in Macaronesia is quite surprising, also in consideration of the apparent absence of species of the genus from western Morocco (but data on the salty coenoses are so incomplete and scattered, as demonstrated by the high number of new taxa described here for a relatively well known region, that such an absence cannot be taken for granted). The island of Lanzarote, to which La Graciosa is closely associated, originated about 14.5 million years BP ( Carracedo et al. 2002; Mangas-Viñuela 2007); it is the nearest island to the African continent and is characterized by a maritime semi-arid climate, with very low precipitation ( von Suchodoletz et al. 2009). Up to about a decade ago, land connections between Lanzarote (and the other eastern islands) and the African continent had been postulated, to explain the origin of a part of the Macaronesian faunistic component (see, for example, Franz-Sauer and Rothe 1972), but in recent times this possibility has been ruled out ( Carracedo et al. 2002). If Lanzarote has never been part of the African continent, then Ita gratiosa must have colonized La Graciosa island thanks to a dispersal event from a northern African population. Anemochoric transport may have happened, and it has been demonstrated that, especially from the late Quaternary, wind-transported Saharan sand accumulated in Lanzarote valleys ( von Suchodoletz et al. 2009). Marine transportation on a raft of debris and vegetation, brought by a flood to the ocean from an African population, is an alternative possibility. This species seems to be related to the complex including also I. berbera , I. gracilis and I. hispanica , a group whose ancestors may have extensively moved and radiated in the western basin of the Mediterranean during the salinity crisis.
In conclusion, the hypothetical scenario that can be proposed from comparison and examination of the present-day distribution of the genus, the inference of the phylogenetic relationships among the species, and the palaeogeographical and palaeoecological reconstruction of the Peri-Tethys territories, would suggest that Ita underwent a first radiation along the south-eastern coasts of the Paratethys in relatively ancient times, at least as early as the upper Miocene, after adaptation to the Salicornioideae as host plants. Further events of dispersion, dispersal and, also favoured by the reduced vagility, speciation by vicariance, occurred all along the southern and south-eastern coasts of the basin, and in some inland favourable environments. These radiative events were determined or favoured by the tectonic movements that involved the Tethys and Paratethys basins and the Peri-Tethys lands, and by the frequency of the halophytic coenoses, consequent to the late Tertiary palaeoclimatic changes towards a rapid increase in the dryness and desertification of the region. The late Tertiary and Quaternary radiation, still following its course, is well-documented in the present-day pattern of several fragmented, isolated, relict species, characterizing this most typical weevil–host plant association.
Russian Academy of Sciences, Zoological Institute, Zoological Museum
Universidad de Los Andes
No known copyright restrictions apply. See Agosti, D., Egloff, W., 2009. Taxonomic information exchange and copyright: the Plazi approach. BMC Research Notes 2009, 2:53 for further explanation.