The sequoia-loving sprite, a new genus and species of fungus gnat (Diptera, Mycetophilidae) from California

Abstract California is one of the most biologically diverse regions of the world, yet the diversity of fungus gnats (Mycetophilidae) remains largely undocumented within the state. A modest survey of these flies has led to the discovery of a new genus and species of gnat that lives alongside one of the most iconic trees in the world, the giant sequoia (Sequoiadendron giganteum). Spritella sequoiaphila gen. et sp. n. is described and illustrated and its status among other mycetophilid genera is analyzed and discussed.


Introduction
California is home to one of the most biologically diverse regions of the world (Myers et al. 2000), although it is unknown exactly how many species of Mycetophilidae (Diptera) occupy the state. Recent monographs suggest that the family remains in need of rigorous taxonomic development. For example, the genera Novakia Strobl and Azana Walker were not known to occur in the state until 2007 and 2010, respectively (representing four new species; Kerr 2007Kerr , 2010; all three species of Acomoptera Vockeroth living in California were undescribed until 2011 (Kerr 2011); both species of Californian Phthinia Winnertz were not described until 2014 (Fitzgerald and Kerr in press); and an additional seven new species (six of these California endemics) of Megophthalmidia Dziedzicki were only just recently discovered and described (Kerr 2014).
Our understanding of the diversity and distribution of Mycetophilidae and related flies in California will remain limited as long as material available for study is lacking in museums and other curated collections. In an effort to document latent diversity within the California Floristic Province, a modest collecting program has been conducted throughout the state over the last several years. Although woefully incomplete, this effort has generated over 5000 genus-level specimen identifications of fungus gnat specimens from over 170 different collecting events (nearly all Malaise traps) in 12 different California counties.
One of the most iconic symbols of the California Floristic Province is the giant sequoia groves of the Sierra Nevada Mountains. This habitat is special for its natural inhabitants as much as for the emotional reaction it inspires. The giant sequoia is the planet's largest living organism and found only in California, with some trees towering over 300 feet tall and having trunk diameters of over 55 feet. In 1853, when the 'Mother of the Forest' tree was cut down in Calaveras County for speculative commercial exploitation, it set off the first national awakening of environmentalist sentiments that called for the protections of public lands. This awareness, fostered by the popular publications of naturalist John Muir, led to the first protections of these trees in 1864 (Mariposa Grove in Yosemite Valley), and eventually led to the establishment of the United States National Park Service in 1872. Today, the magnificent 'Big Trees' bring over 5,000,000 people per year to visit parks that contain giant sequoia groves, generating tens of millions of dollars in tourist revenue for local economies every day (Thomas et al. 2014).
It was within giant sequoia forest habitat that an especially curious fungus gnat was recently discovered. Because of its unusual morphology -particularly, the unique wing venation of the male -the phylogenetic affiliation of this fly was not immediately clear. This paper describes this species, illustrates its morphology, and locates it within a systematic phylogenetic framework among currently known Mycetophilidae.

Materials and methods
Terminology for thoracic, wing, and genitalic morphology is consistent with Kerr 2011, which follows Söli (1997, Vockeroth (1981), and Matile (1990). The terms "genitalia" and "terminalia" are used interchangeably. Genitalia were macerated in 10% KOH at approx. 95 °C for 15-20 minutes to remove soft tissue, then rinsed in distilled water and dilute glacial acetic acid, and dissected in water. All genitalia preparations were placed in a small genitalia vial containing glycerol, and pinned beneath the specimen. Figures were made using Adobe Illustrator and Adobe Photoshop Creative Suite software, with digital images taken using a Nikon DS-Fi1 scope-mounted digital camera. Habitus images were taken with the same digital camera (or the Nikon DS-Fi2), using an LED dome lighting system (Kerr et al. 2008). Material examined is deposited in the California State Collection of Arthropods, Sacramento, CA (CSCA), as indicated in square brackets after the transcribed specimen label data.
All measurements are made in millimeters. Ranges are given for body length, wing length, and the mean for each of these values is provided. Measurements of the holotype are given in square brackets. The number of individuals measured is noted in parentheses. All measurements are of critical-point dried specimens.  Diagnosis. Three ocelli, antennae with 14 cylindrical flagellomeres, maxillary palpus 4-segmented, scutum raised above level of head, upper half of anepisternum with setae, ventro-posterior area of katepisternum with microsetae, tibial spurs 1:2:2. Wing membrane with macrotrichia; costa produced beyond tip of R 5 ; subcosta long, ending at C, approximately at midpoint of wing; sc-r present, arising beyond origin of Rs; r-m missing because Rs and M 1 touching or r-m present, short; M 2 arising from discal cell basad of origin of Rs or at base of M 1 ; cubital vein unforked, A 1 well developed, reaching beyond origin of Rs. Male gonostylus without basal appendages.
Spritella gen. n. resembles Acnemia Winnertz by its lack of a posterior fork and foreshortened medial stem. However, the new genus is readily separated from Acnemia by the presence of setae on the anepisternum and katepisternum; sc-r arising well beyond origin of Rs; and by having male gonostylus without a basal process. Other sciophiline genera that also lack the cubital fork include Afrocnemia Matile, Cluzobra Edwards, Monoclona Mik, and Parvicellula Marshall. In the new genus, crossvein sc-r is clearly present unlike in Afrocnemia and Cluzobra; R 4 is absent unlike in Parvicellula; and the macrotrichia of wing membrane are decumbent, directed toward wing apex unlike in Monoclona. The long subcostal vein of Spritella gen. n. (relative to wing length) and position of sc-r relative to Rs is also distinctively different from these genera.
Description. Head shape in anterior view subequal, approximately as long as wide; medial eye margins farther apart dorsally than ventrally; antennal eye notch present, at least two ommatidia deep; interommatidial setulae present between all ommatidia; ocelli three, nearly linear; lateral ocellus between 1× and 1.5× its own diameter from eye margin, between 2.5× and 3× its own distance from median ocellus; all ocelli dorsad of eye margin; occipital suture from median ocellus to occiput absent; frontal suture between median ocellus and ventral margin of frons complete, suture between lateral ocelli and eyes also present; frons with setae; face approximately 2× longer than wide, parallel-sided along most of length, bearing setulae throughout; face and clypeus separated by complete suture; clypeus ovate, approximately one-half length of face, covered with short setae. Antennal scape and pedicel subequal in size; scape with setae approximately 2× scape length; pedicel setae approximate length of pedicel; antennal flagellomeres 14, cylindrical, approximately 3× longer than wide, approximately the same length but thinner distally, densely covered with short setae. Palp with 4 visible segments, none with apparent sensory pit.
Thorax (Fig. 2) raised, scutum dorsad of head position; short setae distributed throughout scutum, acrostichal setae present, bristles present along lateral margins of scutum; postalar wall and callus separated by carina; scutellum clearly wider than long, narrower than scutum; antepronotum and proepisternum with bristles; anepisternum with setae dorsally; anterior basalare bare; anapleural suture incomplete; katepisternum with setae ventro-posteriorly; anepimeron bare; anepisternum with few inconspicuous setulae; laterotergite raised ventrally, with bristles and shorter setae; metepisternum bare; mediotergite with three bands of bristles ventrally and shorter setae that extend along dorsoventral length, medially. Wing membrane covered with microtrichia and macrotrichia that are arranged irregularly; C ending beyond R 5 ; dorsal surface of humeral vein without setae, ventral surface with setae; subcostal vein setose on both sides, ending in C, approximately at midpoint of wing; sc-r present, arising distad of origin of Rs; R 1 setose on dorsal and ventral surfaces, although bare basad of Rs vein ventrally; vestigial M vein within discal cell present or absent; R 4 not present, r-m present or absent (R 5 joining M 1 at junction with Rs); M 1 setose above, bare below; M 2 setose above, bare below, either arising from bM, from junction of M 1 and Rs, or from base of M 1 ; cubital vein unforked, setose above, bare below, ending at wing margin; CuP strong at base, extending apically as weak fold; anal vein strong, setose on both sides (less so ventrally), extending beyond origin of Rs. Legs elongate; coxae with dark, erect setae and lighter, shorter decumbent setae; femora with short, appressed setae and microtrichia; mid tibial organ absent; tibial spur formula 1:2:2; tibiae with short, appressed setulae and short, erect setae that are no longer than half widest width of tibia; tarsal claws small; empodium developed.
Diagnosis. This species may be distinguished by the characters of the genus and by the male genitalia; particularly the gonostylus which has two apical lobes, the ventral one being distinctly sclerotized and darkened.
Coloration (Figs 1, 2). Head brown; face and clypeus brown; palpomeres light brown to brown. Antennal scape and pedicel yellowish light brown, base of first flagellomere yellowish light brown, otherwise brown; all remaining flagellomeres brown. Thorax variously yellow, yellowish to orangish brown to brown; scutum yellowish light brown dorsally, with faint brown band postero-medially, orangish brown to brown laterally; scutellum light brown to brown; antepronotum, proepisternum, anepisternum, and katepisternum brown; anepimeron light brown; laterotergite and anepisternum brown; metepimeron and metakatepisternum brown to dark brown; halteres yellowish light brown; mediotergite yellowish light brown. Wing membrane lightly brown infuscated. Coxae dark brown; femora yellowish light brown; tibiae slightly darker yellowish light brown; tarsi brown. Abdominal segments light brown to brown, lighter in color near anterior margin on segments 1-6. Terminalia brown.
Thorax and Abdomen. Wing as Fig. 3; Rs distinctly elongate; vestigial M vein present within discal cell; M 1 arises from R 4+5 at origin of Rs so that r-m not present; M 2 weak at base. Tergite 8 reduced (Fig. 4), with setae on posterolateral margin, and a     Male Genitalia (Figs 6-9). Epandrium deeply emarginate both anteriorly and posteriorly, with pair of submedial apical fins (Fig. 6) and pair of short lateral processes (Fig.  8). Gonostyli with pair of setose apical lobes, the ventral one with dark sclerotization.
Coloration. Similar to male. Thorax. Wing as Fig. 3; Rs not as long as in male; vestigial M vein within discal cell absent; r-m present; M 2 arises at or near base of M 1 .
Etymology. The species epithet is an adjective, referring to its affiliation with giant sequoia groves ("sequoia-loving").

Phylogenetic analysis
The phylogenetic analysis was carried out by adding Spritella seqoiaphila gen. et sp. n. to the scored character matrix created by Borkent and Wheeler (2013), with minor modifications (Suppl. material 1). The following changes were made to avoid problems of non-independence: Character # 43, wing macrotrichia orientation (decumbent or reflexed), was changed to "?" for taxa whose wing macrotrichia was scored as absent (41: 1). For taxa whose ventral surface of subcostal vein was scored as bare (54: 0), character # 55 was not applicable and changed to "?" since the presence or absence of setae on the ventral base of subcostal vein was not free to vary (it was already scored as bare). The same type of non-independence was found in character # 57 which treats the dorsal side; the subcostal vein (bare or setose), was changed to "?" for taxa that had already been scored as bare in the previous character (56: 0). Character # 68, position of anterior fork origin, was changed to "?" for taxa whose anterior fork (M1 + M2) was scored as absent (67: 1). For taxa whose posterior fork was scored as absent (73:  Character #30, anapleural suture (single, double, or absent), was re-scored as "single" (30: 0) for Acomoptera plexipus (Garrett) and A. vockerothi Kerr since this character state could be verified with specimens in the CSCA collection.
All characters were scored for S. sequoiaphila except the following: 47, humeral vein (oblique or curved); 84, arrangement of vestiture of tibia (irregular, apical portion with parallel lines, or all in parallel lines); and 87, vestiture arrangement on tarsomeres (irregular or in parallel lines). Due to problems of interpretation and reproducibility, these characters were scored as "?" in the final matrix; I could not score these characters confidently for any taxa, including those used in the original Borkent and Wheeler (2013) matrix. For wing characters, the female (Fig. 3) was used to facilitate homology recognition. The modified matrix was analyzed using parsimony; 1000 heuristic search replicates and 500 bootstrap replicates were performed using PAUP* 4.0b10 (Swofford 2001), with random-taxon-addition, tree bisection reconnection (TBR) branch swapping, steepest decent and 'MulTrees' options in effect. All characters were treated unordered and assigned equal weights. Mesquite (Maddison and Maddison 2011) was used to analyze character change and support in the phylogenetic tree.

Phylogenetic results and discussion
Parsimony heuristic searches found 61 most parsimonious trees (Fig. 14). The strict consensus of these trees differs from the results provided by Borkent and Wheeler (2013), although the underlying data matrix is largely the same. Bootstrap values above 50% support the monophyly of most genera while support for intergeneric relationships is largely lacking. A monophyletic Sciophilinae includes Syntemna hungarica Lundström and its sister lineage, although excludes species of Aneura Marshall, a genus normally thought to be included in the subfamily. Inside this group are two major lineages that contain the remaining taxa, sister to Taxicnemis marshalli Matile. Using implied weighting, Borkent and Wheeler (2013) may have found a similar tree, with Syntemna Winnertz as the sister to the rest of the Sciophilinae and Phthinia Winnertz and Polylepta Winnertz united as sister taxa. Figure 15. Strict consensus tree of 61 equally parsimonious trees from re-analysis of Borkent and Wheeler (2013), with the addition of Spritella sequoiaphila sp. n. Bootstrap values provided above branches where support ≥ 50% (500 reps).
The concept of the "Azana group" (Matile 1998) is recovered with the inclusion of Morganiella fusca Tonnoir and Paramorganiella adventurosa Tonnoir. This result is consistent with the composition of this group as suggested by Amorim and Oliveira (2008), although both Morganiella Tonnoir & Edwards and Paramorganiella Tonnoir have a complete posterior fork (character 73:0) unlike the other members of this group. Sciophila Meigen, which forms a trichotomy and may or may not be included in this group, also has a posterior fork that is complete (or interrupted basally; as in S. interrupta, S. cincticornis, and S. fractinervis). Within this clade, the new genus is included.
Spritella sequoiaphila is recovered sister to Afrocnemia whitfieldae Matile. The two genera are united by the ocellar arrangement being linear (6: 2) and the first flagellomere being slightly offset (20: 1). Along with most species of Acnemia, Afrocnemia whitfieldae and Spritella sequoiaphila have a distinctively short anterior wing vein fork. This character is difficult to score discretely for phylogenetic analysis (e.g., at what point is it "short" or "long"?) and was not scored by Borkent and Wheeler (2013) but appears to have phylogenetic signal that should be considered for future analyses.
Acnemia is recovered as a paraphyletic group, which is not especially surprising, since members of this genus can vary widely and in a way that suggests that this group is in need of systematic revision. Borkent and Wheeler (2013) did find that Acnemia are united by a unique synapomorphy, however, in that the gonostylar lobe bears one to three thin processes. These processes are lacking in Spritella sequoiaphila.