The Neotropical tanyderid Araucoderus gloriosus (Alexander) (Diptera, Tanyderidae), with description of the egg, larva and pupa, redescription of adults, and notes on natural history

Larvae, pupae and adults of Araucoderus gloriosus (Alexander) were collected during fieldwork in Chilean Patagonia, December 2013 and January 2014. Eggs were obtained from females that oviposited in captivity. Association of all life stages is based on co-occurrence and rearing of individual larvae to adults. A diagnosis for the genus and species is provided. Descriptions of the egg, larva and pupa and redescriptions of the male and female are completed. Eggs of A. gloriosus are the first described for Tanyderidae. Natural history characteristics for this species, including microhabitat, copulatory behavior and oviposition, are discussed. supernumerary cross-veins terminalia: 90 dorsal bridge paramere pair of anterolaterally; lateromedial paramere small lobe 4–6 setiform dentate lobe set five lobe inserted proximally, set many small trichoid dorsomedial basally dorsal bridge, apically lateral ejaculatory processes aedeagus; parameral gonocoxite base reniform lateral several prominent setiform inserted point of some

Since discovery of the first extant tanyderid (Osten Sacken 1859), information pertaining to the biology or behavior of this family has been of great interest, but fragmentary. Except for studies by Exner & Craig (1976) and Lukashevich & Scherbakov (2014), most published information about tanyderid natural history includes brief mention of habitat or life history (Colless & McAlpine 1970;Krzemiński & Judd 1997;Judd 2004) usually in the context of a species description (Alexander 1930;Crampton 1930;Wood 1952;Knight 1963;Rose 1963;Podeniene & Gelhaus 2013;) or geographic range extension (Alexander 1935).
Information about the reproductive biology of primitive crane flies, particularly regarding mating behavior and oviposition, has been hypothetical at best. Knowledge of natural history is lacking partly due to sampling difficulties for adults and especially immature stages. Larvae of some species are known to burrow in soft, saturated wood (Hinton 1966;Colless & McAlpine 1970, 1991Krzemiński & Judd 1997;Lukasevich & Scherbakov 2014) while others are free living in the shallow hyporheic zone of cobble-, gravel-and sand-bottom streams (Alexander 1930;Crampton 1930;Wood 1952;Knight 1963;Rose 1963;Exner & Craig 1976;Anthon 1988;Krzemiński & Judd 1997;Judd 2004;Courtney & Merritt 2008;Marshall 2012). A single larva of Protanyderus stackelbergi Savchenko (1971) was recorded from a leaf pack (Podeniene & Gelhaus 2013), but it is unclear if this is the primary larval habitat for the species. Pupae have been found on the damp ridges above the water line of sandy banks of streams and rivers (Wood 1952;Knight 1964). Adult tanyderids are rarely collected, cryptic and believed to be short lived, which contributes further to a poor understanding of the family.
Araucoderus Alexander, 1929 is one of three monotypic tanyderid genera in the Neotropical region: Neoderus Alexander, 1928 andTanyderus Philippi, 1865. Of these, Araucoderus gloriosus (Alexander 1920) was originally placed in Tanyderus, and later Alexander (1927) transferred the species to Radinoderus Handlirsch, 1909. Subsequently, Alexander (1929 proposed Araucoderus as a genus to include A. gloriosus. Despite the species being reclassified multiple times, the immature stages remain undescribed and, aside from wing venation, little is known about adult morphology. Moreover, no information has been recorded regarding the natural history of this species.
During a recent expedition to southern Chile, we collected specimens of all life stages of A. gloriosus and obtained eggs from ovipositing captive females. Detailed observations were made of larval, pupal and adult habitat and behavior. In the current paper, we redescribe A. gloriosus adults, describe its eggs, larvae, and pupae, and provide details of the behavior, habitat and development of the species. This is the first study including complete descriptions of the morphology and natural history for all life stages of any tanyderid species.

Material and methods
Study area. This project focused Chilean Patagonia where samples were taken from streams and their riparian vegetation. Localities range from montane to lowland between latitudes 39°31'N and 48°02'N, including Araucanía (Region IX), Los Ríos (Region XIV), Los Lagos (Region X) and Aysén (Region XI) (Fig. 2).
Material. All life stages of Araucoderus were examined. Most specimens were collected between December, 2013 and January, 2014. Additional larval specimens were collected during October, 2007. Adult specimens were borrowed from the National Museum of Natural History, Smithsonian Institution, Washington, D.C. (USNM).
Voucher collections from the current study are deposited in the following: USNM; Academy of Natural Sciences, Philadelphia, PA (ANSP); Iowa State Insect Collection, Ames, IA (ISIC); Canadian National Collection of Insects, Ottawa (CNCI) and Museo Nacional de Historia Natural, Santiago, Chile (MNNC).
Adult specimens were hand-collected or swept from riparian vegetation. Pupae were collected by hand from the wet root mats of marginal vegetation and damp marginal substrata, while larvae were collected in kick samples and by hand-picking them from marginal substrata. Definitive associations of all life stages are based on rearing of individual larvae to adult emergence. Species associations of non-reared adult males and females were based on observed copulation and morphological similarity of wings. Association of eggs was based on two copulations and resulting eggs from two different females.
Specimen preparation. Field-collected and reared adults were fixed primarily in 70% or 95% EtOH, whereas most larvae and pupae were killed either in 95% EtOH or by immersion in hot water and subsequent placement in 95% EtOH. Eggs were preserved in 70% EtOH at different times after oviposition to address color changes. Undissected adult specimens and remaining dissected structures were dehydrated in a graded ethanol series (95, 100, 100, 100%), followed by a graded hexamethyldisilazane (HMDS) series (25, 50, 100, 100%). Morphological studies were based on partial-or whole-animal preparations, slide mounts, and scanning electron microscope (SEM) micrographs of selected body parts. For selected specimens, the head and terminalia were macerated in a solution of cold potassium hydroxide (approximately 10%) and slide mounted permanently or temporarily using Canada balsam or glycerin, respectively.
Light microscopy. Specimens were observed using an Olympus SZX-9 and SZX-12 stereo microscopes and a Nikon Eclipse E800 compound microscope, the latter equipped with differential interference contrast and both fitted with an ocular micrometer. Drawings were rendered with the aid of a drawing tube on both systems. Photomicrographs were captured using a Nikon DS-Fi1 mounted interchangeably on both microscopes. Composite images were generated using NIS-Elements 3.2 imaging software and edited in Adobe Photoshop ® 12.1x64. Illustrations and plates were produced using Adobe Illustrator ® CS5 15.1.0 and Adobe Photoshop ® 12.1x64.
Scanning electron microscopy. The head capsule, thorax and genitalia were dissected from select adult specimens while the head capsule and anal division were dissected from select larval specimens. All material was sonicated in EtOH for 5-10 seconds. After sonication, specimens were dried either chemically with HMDS or by using a critical point drier, then sputter coated with gold-palladium. Material was examined using a JEOL 5800LV Scanning Electron Microscope.
Terminology. Larval terminology follows Courtney et al. (2000) and Wipfler et al. (2012), with head sensilla numbered from anterior to posterior and dorsal to ventral. Pupal terminology follows Borkent (2012). Descriptions of adult morphology follow Cumming & Wood (2009). Terminology of the male genital tract follows Borkent & Sinclair (2012). The parameres in Tanyderidae are complex, with multiple elements herein referred to as "dorsal bridge of paramere", "dorsomedial element of paramere", "parameral lobe at gonocoxite base" and "lateromedial element of paramere". Female post abdomen terminology follows Cumming & Wood (2009) and (Kotrba 2000). In the material examined section, "instar IV larvae" and "pupae" are abbreviated below as "L" and "P", respectively. We use the term "natural history" in reference to all observational data.   Descriptive format. Diagnoses of the genus and species are provided. Complete descriptions are given for previously unknown life stages and revised descriptions are given for adults. When applicable, sample size is provided before each description with measurements in millimeters presented as a mean followed by a range in parentheses. Measurements: Larval characters refer to instar IV; total length was measured from the anterior-most part of the outer epistomal ridge to the anal aperture. Pupa: total length was from apex of the dorsal apotomal tubercle to the apical-most point of terminal fleshy lobes; respiratory organs were measured from base to apex. Adult: head width was measured at the point of greatest width of the eyes, and head length from the anterior margin of the clypeus to the occiput; leg segments were measured along the dorsal margin; approximate wing length and width were measured at the point of greatest length and width, respectively.
Distribution. Araucoderus is restricted to Patagonia. All confirmed records lie between latitudes 38°31'S and 48°02'S, ranging in altitude from 2 to 1700masl. A survey of the central regions of Chile yielded no specimens. There is one published Araucoderus record from Argentina (Alexander, 1959).
Type material. Chile. Región XI (Aysén): Holotype ♂: Chile Rio Aysén, approximately, coll. P. Dusen, no date given; the specimen is deposited in two museums. Naturhistoriska Riksmuseet Stockholm, Sweden (NHRM) and the USNM. The latter possesses only the wing, antenna and middle leg. Only the USNM material was examined during the current study.
Captive reared larvae became increasingly active a few days prior to pupation, possibly in search of a suitable pupation site. During morning field collections, larvae were sometimes found crawling on the surface of the marginal substrate. Captive individuals became increasingly active during the morning from 0900-1100hrs. In the field, mature larvae were often found beside a rock that was much larger than the larva, sometimes partially buried in the contiguous sandy substrata (Fig. 52).
Resting behavior of adults. In both the field and laboratory, we observed no A. gloriosus adults resting on horizontal surfaces. The few specimens found on sloped surfaces used their tibial spurs as the main point of support. In the field, most adults were found on the underside of broad leaves in the riparian zone or among exposed roots of undercut river banks in the same general location as larvae , sometimes more than 10 m away from the stream. Among the most common resting sites were the undersides of leaves of Gunnera sp., Chusquea sp., or Fuschia sp. A few individuals of both sexes were found under bridges, between the gabions armoring bridge abutments or on vertical surfaces of rocks and logs. A morphological study of the tarsal segments in A. gloriosus revealed similarities to those of Mischoderus Handlirsch, 1909(Friedemann et al. 2014) and suggests that specific tarsal attachment devices are not required for the observed resting behavior in Araucoderus and Mischoderus.  Mating behavior (Fig. 54). On December 13 th 2013, a male and female of A. gloriosus were captured and kept alive in a clear container for further observation. Within 30 minutes of capture, the flying male collided several times with the vertically resting female, until coming to rest on the vertical wall of the enclosure side by side. Soon after, the male flexed the abdomen laterally, clasped the female abdomen at mid-length and guided its posterior end toward that of the female using an alternate constricting motion of the gonopods until their terminalia were aligned somewhat perpendicular. The male subsequently turned 180°, displaying a tail to tail copulation position. The copulation event was brief (10-20 seconds), followed by the male dislodging from the female. Neither individual showed further interest toward the other. The same female was observed copulating again several days after the first event. During the second copulatory event, three males were kept in the same enclosure.
During observed copulations, the apical tarsal segment of the foreleg was not used to hold the female, which disagrees with the general mechanism proposed by Hennig (1968). In fact, our observations suggest the female is restrained entirely by the gonopods. The only previous account of tanyderid mating position (Borkent et al. 2008) suggested a generalized orientation in which the mating position after landing was tail to tail while both individuals are standing up. This agrees with our observations except that Araucoderus was unable to stand freely on a horizontal surface.
Our observations of mating in Araucoderus show some similarities with the copulatory behavior of Sycorax silacea Curtis, 1839 (Psychodidae) described by Jung (Jung 1956 p.119) "… The male stands on the side of the female, with the body directed in the same direction as her. Suddenly he turns rapidly backwards with open claspers onto the female's abdomen. If the copulation succeeds, both animals stand for about 5 minutes calmly, forming a straight line with their bodies." This observation may be a coincidence, but could be additional evidence supporting a close relationship between Tanyderidae and Psychodidae. Furthermore, Jung's (1956) observations for S. silacea may also characterize the mating behavior of the Neotropical psychodid Aposycorax chilensis Tonnoir, 1929, which exhibits similar behavior and morphology such as short flight pattern, resting in a hanging position and 90° rotation of the male genitalia (Curler et al. 2015).
Oviposition. Clusters of eggs ( Fig. 4-6) were laid on water droplets on the enclosure floor. Night collections. Previous nocturnal collections of some tanyderids suggest a correlation of flight activity with sunset and evening hours (Judd 2004). Furthermore, black-or incandescent lights can attract some species of Protanyderus Handlirsch (1909), Protoplasa Osten Sacken, 1859 and Mischoderus (Krzemiński & Judd 1997;Judd 2004;Courtney unpublished observations). During our 2013-2014 expedition, black-light sampling was conducted at different times throughout the night, in areas where immature and adult Araucoderus were present during the afternoon. However, these efforts yielded no adult specimens, suggesting that Araucoderus is not attracted to black lights. On one occasion, with no A. gloriosus adults approaching the light and no flight activity observed during sunset hours, a subsequent search of the marginal vegetation revealed nearly two dozen resting A. gloriosus adults. Based on our observations, Araucoderus adults may be most active during the morning hours.
Predation. Mid-air predation of adult Araucoderus was observed by a swallow (Passeriformes: Hirundinidae) and a dragonfly (Odonata: Anisoptera). Although no predation of larvae or pupae was observed, we found a dead mature pupa parasitized by over two dozen larval Phoridae.

Discussion
A lack of information about the habitat, behavior, and life history of tanyderids has hampered our understanding of the family and presumably contributed to the scarcity of collection records. In addition to new information on morphology and a redescription of Araucoderus gloriosus, this study has provided valuable new information on the natural history of a tanyderid. It is our hope that the data and observations presented here will ultimately lead to additional collections, information, and insights on the biology of these unusual flies.