Descriptions of eighteen new species of Glemparon , a previously monotypic genus of Porricondylinae ( Diptera , Cecidomyiidae )

Abstract. Glemparon Jaschhof, 2013, a previously monotypic genus confi ned to Sweden, is shown here to be considerably richer in species, with most species found to occur in the Australasian region. Eighteen new species are described: G. tomelilla sp. nov. (from Sweden); G. aotearoa sp. nov., G. birhojohmi sp. nov., G. cervus sp. nov., G. didhami sp. nov, G. kaikoura sp. nov., G. nativitas sp. nov., G. orautahi sp. nov., G. otago sp. nov., G. pureora sp. nov., G. rakiura sp. nov., G. rotoiti sp. nov., G. rotoroa sp. nov., G. tewaipounamu sp. nov., G. waipapa sp. nov., G. waipoua sp. nov. (all from New Zealand); G. manuka sp. nov. and G. warra sp. nov. (both from Tasmania, Australia). Glemparon sagittifer Jaschhof, 2013 is redescribed. Genitalic illustrations are provided allowing for the effective identifi cation of all the species known thus far. Morphological data obtained here are used for revising the generic defi nition. Dicerura Kieffer, 1898 is hypothesized as the sister group to Glemparon. The case of Glemparon is discussed as a perfect example of the fact that our collective ignorance of porricondyline diversity in most parts of the world is a major impediment to a better understanding of the European species.


Introduction
The genus Glemparon Jaschhof, 2013 was introduced to absorb a single, unusual species of dicerurine Porricondylinae, of which two males had been captured by The Swedish Malaise Trap Project in two places in southern Sweden (Jaschhof & Jaschhof 2013).Morphological analysis revealed this species, Glemparon sagittifer Jaschhof, 2013, to differ from all other Dicerurini by the presence of setae on both the postfrons and the metepisternum.Other distinctions found concern genitalic structures: the ventral gonocoxal emargination, which is provided with a complex, largely membranous outgrowth (revised and called the posteromedial protuberance in the present paper) and the tegmen, which has two pairs of processes (Jaschhof & Jaschhof 2013: fi g. 64A-B).As both specimens of G. sagittifer available referring to particular characters (↓ 1 , ↓ 2 , ↓ 3 , etc.).Descriptive sections other than diagnoses, titled here "Other characters", make mention only of those characters not referred to before.Species are treated according to geographical distribution (Sweden, New Zealand, Tasmania) and, within this scheme, in alphabetical order.Most of the new species are named after geographic localities, with names used as nouns in apposition as to retain their euphony.Exceptions to this practice are explained in the respective etymology sections.

Diagnosis
Glemparon is known only from males; females and larvae remain unrecognized.Unprecedented within Dicerurini, the number of fl agellomeres varies among different species of Glemparon from 12 to 18, as far as is known.Basitarsal spines are either present (Fig. 1A) or absent (Fig. 1B), which is another unusual case of intrageneric variation.Genitalic structures show the following synapomorphous characters, with four or more present in a single species: (1) the gonostylar apex is laterally compressed (Fig. 2B); (2) the gonocoxites have a posteromedial protuberance, which is largely membranous (and thus hard to study by light microscopy) and usually structured in complicated ways (Fig. 2A; see remarks below), and (3) extensive membranous areas ventroposteriorly below the gonostyli (Figs 2A, 4A); (4) the tegmen is provided with a pair of posterolateral processes that are mostly serrate or tubercular (Fig. 15C) and (5) with a pair of elongate apodemes (Fig. 15C; called here the longitudinal apodemes, equal to the ventral processes by Jaschhof & Jaschhof (2013: fi g. 64A); see remarks below); (6) the ejaculatory apodeme is invariably single-pointed and mostly arrow-shaped (Fig. 3C); (7) the posterior edge of the ninth tergite, which is convex (as opposed to concave or bilobed), is mostly provided with megatrichia (Fig. 6D; see remarks below).

Other characters
As our present study reveals, G. sagittifer is the only species of Glemparon whose postfrons and metepisternum are setose; in all the congeneric species those sclerites are devoid of setae.Therefore, the presence of pronotal and metepisternal setae cannot be regarded as generic characters (see Jaschhof & Jaschhof 2013).The fl agellomeral necks are shorter (Fig. 1H) to several times longer (Fig. 1F) than the nodes.Circumfi la, which are present on all but the terminal fl agellomeres, have posterior extensions only on the proximal fl agellomeres; those extensions, which have different lengths, are either closely adpressed (Fig. 1F) or free-ended (Fig. 1G).In species with generally short extensions, some of the circumfi la are simply ring-shaped or sinuous (Fig. 1H).The fl agellomeral nodes are completely covered with microtrichia, while the necks are always glabrous.The palpus of G. sagittifer was described to have only two to three segments (Jaschhof & Jaschhof 2013), whereas the conspecifi c specimen described in the present paper has four segments.The four-segmented condition is found, invariably, in all other species of Glemparon, but both the lengths and shapes of palpal segments are subject to variation, both among different species (see Fig. 1C vs Fig. 1D) and within one and the same species (see Fig. 1D vs Fig. 1E).Pronotal setae are either present or absent.The claws of Glemparon are subrectangular rather than evenly bent, with the distal sections clearly longer than the basal sections, and with one large and two to three small teeth basally.

Remarks on certain genitalic structures
The structure called here the posteromedial protuberance possibly indicates the presence of the ninth sternite, which in the basal subfamilies of Cecidomyiidae, including Porricondylinae, is thought to be merged, mostly untraceably, into the ventral gonocoxal bridge (see Jaschhof & Jaschhof 2013).Protuberances similar to that found in Glemparon are also present here and there in other Dicerurini, such as Dicerura complicata Spungis, 1987;D. adunca Borkent, 1990; and Linnaeomyia hortensis Jaschhof & Jaschhof, 2015;but  the tegmen, or both; however, their mostly membranous texture and small size makes these connections diffi cult to study using light microscopy.
The longitudinal apodemes arising from the tegmina of many Glemparon are, to our knowledge, unique within the Dicerurini.They arise, as far as one can discern, from the bases of the posterolateral processes, and either end freely or connect to the inside of the gonocoxal wall.Due to the muscles attached to them, these apodemes are quite conspicuous in our specimens, but one cannot be sure about their visibility in macerated specimens where they possibly become eliminated together with the muscle tissue.
The integumental projections referred to here as megatrichia are conspicuously enlarged microtrichia, which appear to be fl attened and bent apically.Megatrichia are unknown to us in other Porricondylinae, including Dicerurini.In Glemparon, they are situated next to ordinary, hair-like microtrichia as well as tubercle-like microtrichia (Fig. 14B), which, within Dicerurini, are found also on the ninth tergite of several Dicerura (e.g., Jaschhof & Jaschhof 2013: fi g. 54A).In Porricondylinae, as in other fungivorous Cecidomyiidae, the phenotypic variation of microtrichia is insuffi ciently studied.

Relationships to other Dicerurini
Male morphology suggests that Glemparon is the sister group to Dicerura.In both genera the gonocoxites are provided with ventroposterior lobes (referred to as gonocoxal processes by Jaschhof & Jaschhof (2013)), which we interpret here as a synapomorphy.The same kind of gonocoxal lobes are present here and there in other Dicerurini, but the few species concerned have genitalic structures that otherwise are very different from that in Glemparon and Dicerura.While there is always a single pair of gonocoxal lobes present in Glemparon, Dicerura may have either one or two pairs.Another derived trait shared by the two genera is the presence of modifi ed microtrichia on the ninth tergite.Apart from large, coarse microtrichia the size of setulae that are commonly found in both genera, many Dicerura have coniform, tubercle-shaped microtrichia, whereas many Glemparon have megatrichia.A third similarity in genitalic characters is that the gonostyli of both groups have exposed, densely microtrichose areas, which in Glemparon are situated at the gonostylar apex but in Dicerura at the gonostylar base ("mediobasal lobe", see Jaschhof & Jaschhof (2013)).Very few Glemparon have microtrichose bulges at the gonostylar bases, but then in a dorsal, not medial, position.Despite the fact that areas of dense microtrichia are exposed in different positions on the gonostylus -apical in Glemparon, basal to medial in Diceruratheir presence might be another indicator of the common ancestry of both genera.Indicative here are the short, blunt-ended bristles situated among the microtrichia in both Glemparon and Dicerura; those bristles are typically located at or near the gonostylar apex (not only in Dicerurini and other Porricondylinae but throughout the fungivorous subfamilies of Cecidomyiidae).It therefore stands to reason that the mediobasal lobe found in Dicerura is actually the gonostylar apex.If so, Glemparon and Dicerura concur in having exposed areas of dense microtrichia intermingled with a few short bristles at the gonostylar apex.To end on another interesting note, two species of Dicerura with structures similar to the gonocoxal posteromedial protuberance found in Glemparon, namely D. complicata and D. adunca, are so aberrant also in other genitalic characters that one might doubt the validity of their current generic placement.However, both species are characterized as true Dicerura by the possession of a bifurcate ejaculatory apodeme, which is the character distinguishing Dicerura from all other Dicerurini, including Glemparon.

Identifi cation of species
Males of Glemparon can be identifi ed to species using genitalic characters.Each of the 19 species known to date is distinguished by a specifi c design of either the tegmen or the gonocoxites.Even so, we usually refer to three different structures -usually gonocoxites, gonostylus and tegmen -when diagnosing species, for there is every indication that similar-looking species of Glemparon exist that might be found in the future.Illustrations can describe those structures best and scanning them is, in our estimation, the most effective way to identify these taxa.For the same reason we refrain here from presenting a key.The present study, though based on only one to three specimens per species, revealed that non-genitalic characters tend to vary intraspecifi cally, such as the number of fl agellomeres and the outline of palpal segments.Consequently, such characters should be used with caution until the extent of variation is better known.

Species of Glemparon in Sweden
Glemparon sagittifer Jaschhof, 2013 Fig. 2A-B Since the fi rst description of this species (Jaschhof & Jaschhof 2013), only two additional specimens, again males, have been found, one of which is fi gured here to show some of the genitalic structures in better detail.The gonostylus of G. sagittifer is strongly compressed apically (Fig. 2B), the outline typically found in Glemparon.The gonocoxal posteromedial protuberance possibly bears a pair of dorsally directed processes, of which only the transverse sections are visible in ventral view (↓ 1 , Fig. 2A).One of our specimens has the antennae completely preserved, including 15 fl agellomeres.The palpus, which is conspicuously short, has either three or four segments (not either two or three as previously described); pronotal setae are absent (not mentioned in the original description); and the metepisternum has one to three setae (not only one or two as previously described).

Diagnosis
The ventroposterior edge of the gonocoxites has a V-shaped emargination fl anked by subtriangular lobes of moderate size (↓ 1 , Fig. 3A); the posteromedial protuberance has a slightly sinuous edge and a pair of ovate substructures, whose outline may remind one of eyeglasses (↓ 2 , Fig. 3A).The roughly H-shaped tegmen is provided with a pair of posterolateral processes whose spike-bearing apices are bent ventrolaterally (↓ 3 , Fig. 3D).The vestiture of the ninth tergite (Fig. 3A) consists of normal setae and microtrichia; megatrichia are absent.

Etymology
The specifi c epithet refers to Tomelilla, a small town in Sweden's province Skåne, where the holotype specimen was collected.

Differential diagnosis
Glemparon tomelilla sp.nov.differs from G. sagittifer, the only congener in the Palearctic, in the outline of the posterior gonocoxal edge, the structuring of the tegmen, and the more slender gonostylus.Furthermore, G. sagittifer has metepimeral setae, which are missing in G. tomelilla sp.nov.A species with genitalic structures generally similar to that of G. tomelilla sp.nov. is G. warra sp.nov.from Tasmania (Fig. 20A-D).

Distribution and phenology
This species is known from two locations in southern Sweden, where only two specimens were obtained as a result of extensive Malaise trapping all over Sweden by both the Swedish Malaise Trap Project and ourselves for several years.A common feature of the two collecting sites is that grasses predominate in the ground cover, with one of the sites being markedly dry, the other rather swampy.Specimens were collected during the main vegetation period in May-August.

Etymology
The specifi c epithet, aotearoa, is the Maori name for the country of New Zealand, meaning "land of the long white cloud".TERMINALIA.Posterior edge of ninth tergite with large, hair-like microtrichia.Gonocoxites with narrow, unsetose, separate section ventrobasally (Fig. 4A).Gonostylus twice longer than broad; outside densely setose; inside with dense, fi ne pubescence, and very few fi ne setulae (Fig. 4B).Apex of ejaculatory apodeme small, membranous, arrow-shaped (Fig. 4C).

Distribution and phenology
The single specimen known of G. aotearoa sp.nov.was Malaise trapped at the height of summer in an old-growth native forest in the north of New Zealand's South Island.The same trap collected simultaneously four other species of Glemparon: G. birhojohmi sp.nov., G. cervus sp.nov., G. rotoroa sp.nov., and G. tewaipounamu sp.nov.

Diagnosis
The following combination of genitalic characters is characteristic of G. birhojohmi sp.nov.: the gonostylus, which is twice longer than broad, is slightly constricted subbasally (↓ 1 , Fig. 5A); the gonocoxal posteromedial protuberance, which is fl anked by small, subtriangular lobes, has a weakly sclerotized, T-shaped process that projects ventrally (↓ 2 , Fig. 5A); below the protuberance is an assemblage of conspicuously dense setae (↓ 3 , Fig. 5A); and the medial bridges have mostly setae of various sizes rather than microtrichia.

Etymology
In naming this new species birhojohmi, we honor Birgit Rhode and John Mitchell, of Papakura, Auckland, New Zealand.Birgit and John, both gifted with an artistic, creative disposition, have been supportive of our taxonomic work for many years.The name should be treated as a noun in apposition.

Differential diagnosis
Glemparon nativitas sp.nov., a broadly similar species found in Stewart Island, differs in broader gonostyli (Fig. 9B) and details of the tegmen (Fig. 9C).

Remark on variation
The paratype (Fig. 5B) differs slightly from the holotype (Fig. 5C) in that the tooth situated most anteriorly on the tegminal processes is conspicuously large and bent.These teeth are apparently somewhat variable in shape, size, and relative position.

Distribution and phenology
See G. aotearoa sp.nov.The paratype of G. birhojohmi sp.nov.was captured almost three months later than the holotype at the same site, indicating a long adult fl ight period in this species.

Etymology
The specifi c epithet, cervus, is the Latin word for deer, an allusion to the antler-shaped tegminal processes found in this species.The name is a noun in apposition.

Distribution and phenology
See G. aotearoa sp.nov.inconspicuous (↓ 2 , Fig. 7A); the ventral emargination is sclerotized basally (↓ 3 , Fig. 7A); and ventral setae are reduced to the lateral portions.The tegminal processes have multiple small tubercles and points, which are harder to discern than Figure 7B (↓ 4 ) suggests.The ninth tergite has a broadly rounded posterior edge, whose vestiture is of ordinary microtrichia and setae of various sizes.

Etymology
This new species is named after Raphael K. Didham, ecologist at The University of Western Australia, Perth, who collected several of the specimens studied here, including the holotype of G. didhami sp.nov.TERMINALIA.Gonocoxites: membranous areas below gonostylus small (Fig. 7A).Gonostylus twice longer than broad, with slight depression above densely microtrichose apex (Fig. 7A).Sclerotized portion of ejaculatory apodeme slightly thickened apically, covered by membranous cap (Fig. 7A).

Distribution and phenology
The only specimen known of G. didhami sp.nov.was Malaise trapped at the height of summer in the native bush of Stewart Island, the smallest and southernmost of New Zealand's main islands.The same Malaise trap collected simultaneously two other species of Glemparon: G. nativitas sp.nov.and G. rakiura sp.nov.

Distribution and phenology
The holotype of G. kaikoura sp.nov.was collected at the end of autumn in a patch of regenerating native bush on the east coast of New Zealand's South Island.

Etymology
The specifi c epithet is a Latin noun in apposition meaning Christmas, an allusion to Christmas Village Hut on Stewart Island, the type locality.

Distribution and phenology
See G. didhami sp.nov.

Etymology
The specifi c epithet, orautahi, is the Maori name for Smoky Beach, the place on Stewart Island where the holotype was collected.

Distribution and phenology
The only specimen known of G. orautahi sp.nov.was collected at the height of summer in the native bush of Stewart Island.

Diagnosis
The genitalia of this species (Fig. 11) differ from what is typically found in Glemparon.In particular, while the gonostylus is narrowed apically, it is not laterally compressed; the gonocoxites lack membranous areas below the gonostyli; and the apex of the ejaculatory apodeme is not arrow-shaped.An absolutely distinguishing structure is the tegmen, which ends in a pair of sclerotized points (↓ 1 , Fig. 11) and has a pair of lateral processes each consisting of about 10 tubercles (↓ 2 , Fig. 11).

Etymology
The specifi c epithet, otago, is of Maori origin and refers to the Otago Region in the south of New Zealand's South Island, where the holotype was collected.

Discussion
Our reasons to classify this species in the genus Glemparon are the possession of 17 fl agellomeres, i.e., more than the 14 fl agellomeres usually found in Dicerurini, and the presence of fi ve out of eight genitalic characters regarded as synapomorphous in Glemparon (see the generic diagnosis above).There is no better, alternative generic placement for G. otago sp.nov.The somewhat aberrant genitalic morphology of this species gives an idea of the diversity of structure met in Glemparon, a fact to bear in mind as other unusual, hard-to-place Dicerurini will be found in the future.Glemparon rotoiti sp.nov., which is described below, possibly lies between G. otago sp.nov.and other, more typical species of Glemparon.

Distribution and phenology
The holotype of G. otago sp.nov.was collected in summer in the high mountains of New Zealand's Southern Alps, in a patch of native bush according to the specimen label.

Etymology
The specifi c epithet, pureora, refers to the type locality of this species, Pureora Forest Park in the North Island of New Zealand, which protects one of the earth's fi nest podocarp forests.

Distribution and phenology
The only specimen known of G. pureora sp.nov.was collected in summer, in shrubland embedded in an extensive podocarp forest in New Zealand's North Island.Another species of Glemparon, G. waipapa sp.nov., was captured one month later at the same site, perhaps even with the same Malaise trap.

Diagnosis
Glemparon rakiura sp.nov. is distinguished by the gonocoxal ventroposterior lobes, which are conspicuous by their globular shape and dense cover with thick, spine-like microtrichia dorsomedially (↓ 1 , Fig. 13A).The gonostylus, which basically is rather slender, has a densely microtrichose protuberance dorsobasally (↓ 2 , Fig. 13A).The ninth tergite has a narrow posterior portion, which is separated from the much broader anterior portion by a constriction and is covered with microtrichia of various sizes; the posterior edge is provided with about 10 megatrichia (↓ 3 , Fig. 13B).

Diagnosis
The elongate gonostylus of simple shape is not compressed at the apex, which is rounded and covered with dense microtrichia (↓ 1 , Fig. 14A).The gonocoxites are peculiar for both the membranous portions ventroposteriorly, whose outline resembles a stair (↓ 2 , Fig. 14A), and the ventral emargination, which is large, somewhat rectangular and vaguely delineated at the base (↓ 3 , Fig. 14A).The ninth tergite has an almost truncate, strongly sclerotized posterior edge (↓ 4 , Fig. 14B), which is a unique feature in Glemparon, and on the inside numerous small tubercles in indistinct rows.The tegmen is provided with a pair of sclerotized, serrate points (↓ 5 , Fig. 14C) and, at the points' bases, sclerotized processes each consisting of two curved teeth (↓ 6 , Fig. 14C).

Etymology
The specifi c epithet refers to Lake Rotoiti in the north of New Zealand's South Island, where the holotype was collected.WING.Length / width ratio 2.6.Rs short, one fi fth of apicR 1 .

Etymology
The specifi c epithet refers to Lake Rotoroa in the north of New Zealand's South Island, where the holotype was collected.THORAX.Pronotal setae present.WING.Length / width ratio 2.9.Rs short, one eighth of apicR 1 .

Etymology
The name of this species, tewaipounamu, is the original Maori name for the South Island of New Zealand, where the holotype was collected.TERMINALIA.Posterior edge of ninth tergite slightly three-lobed, central lobe rounded, with about 15 megatrichia, fl anked by angular-shaped, densely microtrichose side lobes.Gonocoxites sparsely setose ventrally; membranous areas below gonostyli large; ventral emargination U-shaped; ventroposterior lobes large, subtriangular (Fig. 16A).Gonostylus twice longer than broad; outside densely setose; inside with dense, fi ne pubescence, a very few fi ne setulae (Fig. 16B).Apex of ejaculatory apodeme largely unmodifi ed (Fig. 16C).Tegmen with distinct pattern of sclerotization in center (Fig. 16C).

Distribution and phenology
See G. aotearoa sp.nov.

Etymology
The specifi c epithet, waipapa, highlights the Waipapa Ecological Area of the Pureora Forest Park in New Zealand's North Island, where the holotype of this species was collected.

Other characters
BODY SIZE.1.6 mm.HEAD.Eye bridge 1-2 ommatidia long dorsally.Nine fl agellomeres retained; fourth fl agellomere with neck and node equally long.Palpus clearly shorter than head height, fi rst segment very small, unsetose, second to forth segments with ordinary setae; apical segment longest of all.THORAX.Pronotal setae present.WING.Length / width ratio 3.2.Rs short, one fi fth of apicR 1 .

Distribution and phenology
See G. pureora sp.nov.

Etymology
The specifi c epithet, waipoua, refers to Waipoua Forest, a sanctuary for the fi nest kauri forest preserved in New Zealand, which is the type locality of this new species.TERMINALIA.Posterior edge of ninth tergite with about 20 megatrichia medially fl anked by inconspicuous microtrichose lobes (Fig. 18B).Gonocoxites with narrow, unsetose portion ventrobasally; ventral setae reduced to a transversal stripe in center; membranous areas below gonostyli large; ventroposterior lobes small, rounded; posteromedial protuberance unmodifi ed except for a distinct longitudinal fold; medial bridges with short, densely microtrichose sections (Fig. 18C).Gonostylus more than 3.0 times longer than broad (Fig. 18A).Ejaculatory apodeme with elongate, membranous apex (Fig. 18E).

Distribution and phenology
This species, known only from the holotype, was collected in the subtropical winter of New Zealand's Northland, in a forest predominated by kauri trees.

Notes on a possibly closely related, unnamed species
Among the Glemparon studied here is a male (NZAC, no.CEC1413) from the Westland Region in New Zealand's South Island, which largely fi ts the description of G. waipoua sp.nov.given above, with the exception of the genitalic structures that resemble those shown in Figure 18D.In this specimen, the gonocoxal ventroposterior lobes are smaller; the ventral emargination has a sclerotized basal edge (no microtrichose bulge as found in G. waipoua sp.nov.); and, most importantly, the tooth-bearing processes of the tegmen protrude beyond the gonocoxal wall, which is something we have noted in no other Glemparon.We refrain from describing this specimen as a new species, because it resembles G. waipoua sp.nov. in all other respects and we cannot be absolutely sure whether the peculiarities observed are just due to preparation artifacts.

Diagnosis
The complex genitalic structures found in G. manuka sp.nov.make this the most unusual species we have seen of Glemparon.The tegmen and ejaculatory apodeme form a structural unit, which is also linked to the gonocoxal ventral wall; for simplicity we refer to this entire structure as the tegmen.The abundance of serrate and tooth-bearing lobes on the tegmen, which is unique, is shown in Figure 19A.
The gonocoxal posteromedial protuberance is missing as a distinct structure, but we suppose it is merged into the tegmen.

Etymology
The specifi c epithet, manuka, refers to the Manuka Road in the Warra Long Term Ecological Research site, Tasmania, where the holotype specimen of this species was collected.Manuka (Leptospermum scoparium J.R. Forst.& G. Forst.), a small tree of the Myrtaceae family, is a common native of southeast Australia and New Zealand.

Differential diagnosis
Genitalic morphology indicates that G. manuka sp.nov. is most closely related to G. rotoroa sp.nov.from New Zealand (Fig. 15).The gonostyli, gonocoxites and ninth tergites of both species are broadly similar, while the tegmina are completely different.HEAD.Eye bridge 0-1 ommatidium long dorsally.Nine fl agellomeres retained; neck of fourth fl agellomere 2.2 times longer than node (Fig. 1F).Palpus about as long as head height, 4 setae-bearing segments; apical segment longest of all.

Distribution and phenology
This species is known from a single specimen collected in autumn, in a stretch of old-growth wet eucalypt forest in lowland Tasmania.

Diagnosis
The tegmen consists of a massive, subtrapezoid portion basally and a biramous portion apically, both separated by a constriction; the rami end in leaf-shaped processes whose bases are covered with numerous tiny spikes (↓ 1 , Fig. 20C).The arrow-shaped apex of the ejaculatory apodeme is sclerotized (↓ 2 , Fig. 20D), not membranous as in most other species of Glemparon.The gonostylus is slightly broadened towards the apex (↓ 3 , Fig. 20B).

Etymology
The specifi c epithet, warra, refers to the type locality of this species.The Warra Long Term Ecological Research (LTER) site in southern Tasmania is a prime example of a temperate broadleaf (eucalypt) wet forest.concave portions medially and the angular-shaped, densely microtrichose apex (Fig. 20B).Ejaculatory apodeme conspicuously thick (Fig. 20A).

Distribution and phenology
The only specimen known of G. warra sp.nov.was Malaise trapped at the end of summer in the same habitat as the previous species.

Discussion
Taxonomic descriptions based on single specimens carry the risks of misinterpreting preparation artifacts as taxonomic characters and underestimating intraspecifi c variation (e.g., Jaschhof & Jaschhof 2009).In the present study we used only perfectly prepared (slide-mounted) specimens, whose structures were not distorted or otherwise affected during the preparation process, thereby minimizing the artifact problem.Furthermore, genitalic structures, which provide the characters of diagnostic merit in Glemparon, rarely vary within a species.Even so, whenever we felt that limitation of our material might pose a problem, we interpreted our observations with special caution, such as in the case of G. waipoua sp.nov.
The genus Glemparon can serve as an example for the fact that monotypic genera are diffi cult to interpret (as there is little substance available to compare with), while the discovery of a single additional congener (here G. tomelilla sp.nov.) may function as a catalyzer for the cognitive process.This, we believe, proves the potency of the comparative morphological method for resolving questions of character evolution, phylogeny and classifi cation.For example, the true nature of the tegminal longitudinal apodemes, previously thought to be just another kind of tegminal processes (see Jaschhof & Jaschhof 2013), could be understood only by comparing its various different manifestations within a wide range of species.Another example is the structure called here the gonocoxal posteromedial protuberance; its complex structuring and relevance as a synapomorphy of Glemparon was only realized by comparative examination throughout the genus.Also, the generic defi nition of Glemparon changed fundamentally through the input of data from a number of newly discovered species.Last but not least, intimately knowing as many species of a genus as possible is crucial for realizing the systematic affi liation of species whose morphology is somehow aberrant (see, e.g., G. manuka sp.nov.), for which almost all speciose genera provide examples.In other words, describing new species matters, for they provide the raw data for all kinds of subsequent analyses -a causal relationship whose relevance for Cecidomyiidae and other "open-ended taxa" is sometimes questioned (see Bickel 2009).
We do not think that the disjunct, bipolar distribution of Glemparon as revealed by our study is a matter of real facts; rather it is the consequence of our collective ignorance of Porricondylinae diversity in most parts of the globe.Enough is known about Porricondylinae in the West Palearctic as to establish that Glemparon are uncommon and poor in species there; however, we are much more badly informed about the situation in the East Palearctic, not to speak about the Oriental region.At the same time there is no doubt that Glemparon is an enormously speciose genus in the Australasian region (and possibly even beyond), which was previously unknown.Considering that our small amount of study material was gathered more or less randomly, one wonders what a systematic search for Glemparon in trap samples from all over New Zealand and Australia would reveal.Our observation of fi ve different species cooccurring at a single site also raises the question of how these species are adapted to different ecological niches in the smallest spaces.Thus far, our study has generated more question marks than it has removed -a typical situation with such basic taxonomic studies.Systematic research into Glemparon in the future should consider using morphology along with molecular genetic evidence for characterizing species, even though this means additional expenditures and adequately preserved specimens.As the case of G. waipoua sp.nov.shows, not all alpha-taxonomic problems in Glemparon can be solved using alone male adult characters.