TY - JOUR
AU1 - Yang, Yao Ming
AU2 - Sun, Qian
AU3 - Xiu, Jiang-Fan
AU4 - Yang, Ming
AB - Abstract During the transformation of immature aquatic dipteran insects to terrestrial adults, the prothoracic pupal respiratory organ enables pupae to cope with flood-drought alternating environments. Despite its obvious importance, the biology of the organ, including its development, is poorly understood. In this study, the developing gills of several Simulium Latreille (Diptera: Simuliidae) spp. were observed using serial histological sections and compared with data on those of other dipteran families published previously. The formation of some enigmatic features that made the Simulium gill unique is detailed. Through comparisons between taxa, we describe a common developmental pattern in which the prothoracic dorsal disc cells not only morph into the protruding respiratory organ, which is partially or entirely covered with a cuticle layer of plastron, but also invaginate to form a multipart internal chamber that in part gives rise to the anterior spiracle of adult flies. The gill disc resembles wing and leg discs and undergoes cell proliferation, axial outgrowth, and cuticle sheath formation. The overall appendage-like characteristics of the dipteran pupal respiratory organ suggest an ancestral form that gave rise to its current forms, which added more dimensions to the ways that arthropods evolved through appendage adaptation. Our observations provide important background from which further studies into the evolution of the respiratory organ across Diptera can be carried out. Simuliidae, prothorax, gill, appendage, evolution The wide morphological and functional diversity of arthropod appendages is the result of successful adaptations to different habitats (Averof and Cohen 1997, Damen et al. 2002). Understanding how appendages have evolved into their current forms helps us better appreciate the evolutionary origins of the highly diverse arthropods. For example, comparative observations on the developmental patterns of wings have revealed the origin of the Holometabola (Truman and Riddiford 1999), whereas pre-wing evolution has been accepted as a key link connecting the insects and their arthropodan relatives, which has stimulated much discussion (Averof and Cohen 1997, Damen et al. 2002, Gullan and Cranston 2010, Clark-Hachtel et al. 2013, Prokop et al. 2017, Linz and Tomoyasu 2018). Of the four major holometabolous orders, the dipteran insects are the most studied due to their use in medical (e.g., mosquitos, black flies, and biting midges) and genetic research (e.g., fruit flies, mosquitos, and house flies; Gullan and Cranston 2010). However, the origins of a key feature of aquatic dipterans, the prothoracic pupal respiratory organs (gill for breathing underwater in black flies and other aquatic dipteran insects), remain poorly understood. The gills, which are covered with a layer of chitin mesh (the plastron), allow these relatively low-mobility animals to cope with the flood-drought alternations of the environments in which they live (Hinton 1966). Several studies to date have investigated the physiology, morphology, and development of the gill (Satchell 1948; Reid 1963; Hinton 1966, 1968; Eymann 1991; Arens 1995, 1998; Armitage et al. 1995). Morphological descriptions of dipteran prothoracic pupal respiratory organs provide systematic and comprehensive documentation of the organs (McAlpine et al. 1981, 1989). Borkent (2012) unified the synonyms for prothoracic gills (e.g., trumpets of mosquitos and horns of nonbiting midges) as respiratory organs and compared their features accordingly. Here, considering the anatomy and function of the organ, we refer to it as the gill. Satchell (1948) observed the development of gill tissues in species of Psychodidae (Nematocera: Psychodomorpha) in detail using histological techniques and observed many important interspecies differences. Hinton, who broadly described the physiological and morphological traits of respiratory organs in many higher taxa within Nematocera (Tipulomorpha, Psychodomorpha, and Culicomorpha) and several groups within Brachycera, proposed multiple independent origins of the organs among the studied groups, while also briefly considering the possibility of a common origin (Hinton 1966, 1968). Arens (1995, 1998) investigated the taxonomy of Blephariceridae (Nematocera: Psychodomorpha) by comparing gill morphology and fine structures. Through comparisons of the gill morphology of Blephariceridae and Deuterophlebiidae, Arens argued for a common origin of the organ. Hinton (1966, 1968) proposed the only hypothesis to date on the evolutionary origin of the respiratory organs, which, surprisingly, has commonalities with a lesser-known hypothesis on wing origins—Bocharova-Messner’s ‘spiracular flap theory’ (Kukalova-Peck 1978). Both hypotheses suggest that the thoracic dorsal appendages originated from spiracular tissue outgrowth (Hinton 1966, 1968; Kukalova-Peck 1978). These ideas compelled us to investigate whether the origins of respiratory organs and wings are somehow connected. Similar to dipteran respiratory organs, some aquatic crustaceans bear gill-like appendages on all segments (Averof and Cohen 1997, Damen et al. 2002). Crustacean appendages combine both locomotory and aquatic breathing functions. Paleodictyopterans, the ancestral winged insects, had an additional pair of prothoracic wing pads in both nymph and adult stages (Kukalova-Peck 1978). Meanwhile, some extant hemimetabolous insects (e.g., mantids, treehoppers, and lace bugs) have morphologically diverse extended pronota, some of which are considered relics of the ancestral prothoracic wing pads (Kukalova-Peck 1978, Larivière et al. 2011, Prud’homme et al. 2011, Miko et al. 2012, Schwarz and Konopik 2014, Tomoyasu et al. 2017, Rodrigues and Svenson 2018). Within the Holometabola, protruding prothoracic dorsal structures are not restricted to dipterans. Some beetles (e.g., Geotrupidae, Scarabaeinae, Dynastinae, etc.) bear horns in the center or on each side of the pronotum, which develop from imaginal discs that are tightly controlled by the genetic pattern formation of appendages or by the axis of outgrowth (Emlen et al. 2007). Interestingly, although the Mecoptera and Siphonaptera do not have prothoracic dorsal appendages, an ancestral fossil flea (Saurophthirus longipes Ponomarenko, 1976) does show a pair of protruding prothoracic spiracular structures (Rasnitsyn and Strelnikova 2017). The fossil evidence not only suggests that this particular group of ancestral fleas may have adapted to aquatic life, it also shows that prothoracic adaptations are not a feature unique to dipterans. Furthermore, a recent dipteran phylogenetic study combined with morphological descriptions revealed that nearly all primitive dipteran groups (including Deuterophlebiidae) are aquatic and gill bearing, whereas a few aquatic brachyceran groups also exhibit the same traits (McAlpine et al. 1981, 1989; Wiegmann et al. 2011). These observations may help to better understand the evolutionary origin of respiratory organs. Although previous investigations on gills have mainly relied on comparative morphology and physiology in different groups, comparisons of the development of gills in different taxa could yield critical details from the perspective of tissue origins, an approach not yet attempted. Here, as a preliminary investigation, we describe the development of gills in the Simuliidae and compare their histological traits with those of the Psychodidae (Satchell 1948) and with available data on other dipteran taxa (Hinton 1968; Pedelty and Arking 1981; Arens 1995, 1998). Materials and Methods The collection sites of black fly immatures and fixation reagents used in this study are listed in Supp Table S1 (online only). Debris and aquatic plants with attached black fly immatures were collected in buckets. Using a paint brush, black fly larvae, along with the substrates to which they were attached, were transported to Petri dishes filled with fixative reagents. Afterwards, the pupa-attached plants and the fixed larvae were transferred into 50-ml centrifuge tubes filled with fixative reagents. The black fly immatures were grouped based on the identification keys of regional black flies, particularly the filament patterns on larval discs and the filament counts in pupae (Chen et al. 2016, Yang et al. 2018). The cytochrome oxidase subunit I (COI) sequences of each species group were obtained using a primer pair (LCO1490, 5′-GGTCAACAAATCATAAAGATATTGG-3′; HCO2198, 5′-TAAACTTCAGGGTGACCAAAAAATCA-3′) and PCR procedures, as described by Folmer et al. (1994). The COI sequences were identified using www.boldsystem.org. Species groups with a substantial number of larvae were divided into instar groups according to the morphometric protocol and data of Yang et al. (2018). Sectioned and dissected samples were preserved as vouchers at the Key Laboratory of Medical Entomology. The larvae used for sectioning and staining were pierced in the abdomen to create a hole using a fine needle under a stereomicroscope (Nikon SMZ 18, Tokyo, Japan) and then placed into embedding cassettes (Citotest Scientific EM 104A, Suzhou, China). For dehydration and clearing, the larva-carrying cassettes were immersed into ethanol—at concentrations of 70, 80, 90, 95 (twice), and 100% (twice)—a solution of 67% ethanol and 33% benzyne, a solution of 33% ethanol and 67% benzyne, and 100% benzyne (twice) for 20 min each at room temperature. Finally, the cassettes were immersed in solutions of 67% benzyne and 33% wax, 33% benzyne and 67% wax, pure wax, and 90% wax and 10% beeswax for 30 min each at 55–58°C. At the end of the wax immersion, the larvae were embedded in a mixture of 90% wax and 10% beeswax. At each immersion stage, the cassettes were knocked to get rid of trapped air bubbles. Beeswax can harden the wax mixture, reducing the likelihood that embedded tissues will deform (Y.M.Y. and J.F.X., personal observation). The wax blocks containing larvae were trimmed to a desirable shape suitable for transverse and sagittal sectioning, which were carried out using a microtome (Leica RM2125RTS, Wetzlar, Germany). The sections were cut 6–7 µm thick. The sectioned wax tapes were mounted on glass slides in a flotation water bath heated at 40°C. Afterwards, the wax tapes on the slides were dried overnight at 30°C. For wax dissolution, the dried slides were immersed in 100% xylene twice for 8 min each time. To rehydrate the larval tissues, the slides were immersed in 100% xylene, followed by ethanol at concentrations of 100, 95, 80, and 70%, and then distilled water, for 5 min each. The rehydrated tissues were treated with hematoxylin reagent (Solarbio G1120, Beijing, China) for 4–5 min and then gently washed in water. To dye the tissues, the slides were immersed in 70% ethanol mixed with 0.5% 1 M HCl for 1 min, gently washed in water, and then immersed in 70% ethanol mixed with 0.5% 13.4 M ammonium solution for 1 min. Finally, the slides were dyed using eosin reagent (Solarbio G1120) for 10–15 s and gently washed in water for 5–8 min. After the dyeing, the washed slides were immersed in 95% ethanol (twice) and 100% ethanol (twice) for 1 min each and then in 100% xylene twice for 2 min each. Finally, the slides were mounted with a neutral balsam (Solarbio G8590, Beijing, China). The permanently mounted slide sets are preserved at the Key Laboratory of Medical Entomology. A microscope imaging workstation (Nikon Eclipse and NIS Element D, Tokyo, Japan) was used to observe and image areas of interest in the sectioned slides (Supp Fig. S1 [online only]). Results Samples Through investigating morphological characters and determining COI sequences, we identified the following species: Simulium xingyiense Chen and Zhang, Simulium ornatum Meigen, Simulium quinquestriatum Shiraki, Simulium bidentatum Shiraki, Simulium aureohirtum Brunetti, Simulium takahasii Rubtsov, and Simulium guiyangense Chen, Liu, and Yang (see Supp Table S2 [online only]). We determined the instar numbers of S. xingyiense and S. quinquestriatum as seven and eight, respectively, in agreement with previous interpretations (Supp Fig. S2 [online only]; Yang et al. 2018). Prior to histological sectioning, a preliminary stereomicroscope observation found five distinct phases of gill development in the final two instars (see Supp Fig. S3 [online only]). We refer to them by instars in the following order of occurrence: penultimate instar-1 (PI-1), penultimate instar-2 (PI-2), last instar-1 (LI-1), last instar-2 (LI-2), and last instar-3 (LI-3). In PI-1, six isolated discs on each side of the thorax, which are white or beige in color in contrast to the body pigmentation, become visible via stereomicroscope. During PI-2, the gill tissues appear to undergo branching. With the advent of the last instar, dramatic changes in the shape of the gill and wing discs occur (during LI-1), whereas the mesothoracic wing and leg discs grow large enough to be adjacent to one another. During LI-2, prothoracic and mesothoracic discs also grow large enough to become adjacent to one another, whereas the branches of the gill discs appear to coil, becoming more complex. During the LI-3, before the pupal molt, the gill discs become cuticularized and pigmented. Using these descriptions and those of Crosskey (1962), we were able to accurately assign the larvae of the other species into the PI-1, PI-2, LI-1, LI-2, and LI-3 groups (see Supp Table S2 [online only]). Gills of S. xingyiense The spatial-temporal changes of the Simulium gill during development (using S. xingyiense as a model) are described according to the serial transverse and sagittal sections: The gill disc clearly starts to develop at the penultimate instar, based on comparisons between the PI-1 and PI-2 larvae (Figs. 1 and 2). The thoracic discs of larvae before the penultimate instar generally constitute thick layers of columnar cells that have not yet invaginated to form disc sacks (Supp Fig. S4 [online only]). In PI-1, although disc sacks had formed around the gill discs, no filament branches or any morphological differentiations were observed in either transverse or sagittal planes (Fig. 1A and 1B). No columnar primordial cells appear at the quadrifurcated joint that connects the ecdysial tube and the dorsal tracheae according to the transverse plane (Fig. 1B, V–IX and D, VIII–XII). Fig. 1. Open in new tabDownload slide Serial sagittal (A and C) and transvers (B and D) sections of Simulium xingyiense prothoracic gill discs during PI-1 (A and B) and PI-2 (C and D). Roman numerals show the order of the sections in each series. The tissues of interest—prothoracic gill disc, disc sack, external scar, ecdysial tube, and trachea—are denoted as gd, ds, es, et, and t, respectively. The arrows in A and B highlight the furrows formed by filament primordial folding. Budding filaments in (C) and (D) are numbered as c1–c5 (five cuticular side buds) and i1–i3 (three interior side buds). Dashed lines in (D) highlight the multipart chamber primordial bulge. Scale bar = 100 µm. Fig. 1. Open in new tabDownload slide Serial sagittal (A and C) and transvers (B and D) sections of Simulium xingyiense prothoracic gill discs during PI-1 (A and B) and PI-2 (C and D). Roman numerals show the order of the sections in each series. The tissues of interest—prothoracic gill disc, disc sack, external scar, ecdysial tube, and trachea—are denoted as gd, ds, es, et, and t, respectively. The arrows in A and B highlight the furrows formed by filament primordial folding. Budding filaments in (C) and (D) are numbered as c1–c5 (five cuticular side buds) and i1–i3 (three interior side buds). Dashed lines in (D) highlight the multipart chamber primordial bulge. Scale bar = 100 µm. Fig. 2. Open in new tabDownload slide Partial series of continuous sagittal (A) and transverse (B) sections of Simulium xingyiense prothoracic gill discs during LI-1. Roman numerals show the order of the sections in each series. The tissues of interest—external scar, ecdysial tube, and trachea—are denoted as es, et, and t, respectively. Budding filaments are numbered as c1–c5 (five cuticular side buds) and i1–i3 (three interior side buds). The structure marked by a dashed line and arrows is a trench formed by the invagination of the multipart chamber primordium. Scale bar = 100 µm. Fig. 2. Open in new tabDownload slide Partial series of continuous sagittal (A) and transverse (B) sections of Simulium xingyiense prothoracic gill discs during LI-1. Roman numerals show the order of the sections in each series. The tissues of interest—external scar, ecdysial tube, and trachea—are denoted as es, et, and t, respectively. Budding filaments are numbered as c1–c5 (five cuticular side buds) and i1–i3 (three interior side buds). The structure marked by a dashed line and arrows is a trench formed by the invagination of the multipart chamber primordium. Scale bar = 100 µm. In PI-2, the filament primordium (the columnar cell cluster of the gill disc within the disc sack) shows buds in a fixed and distinguishable number of branches. The branches come in two leaves (Fig. 1C, VIII and IX and D). One leaf of filaments (visible through the cuticle with the aid of a stereomicroscope) covers the interior leaf (Supp Fig. S3B–D [online only]). To form the two leaves of the buds, the filament primordial cells fold into four layers of cells, creating a ‘W’ arrangement in the transverse plane (Fig. 1D, I–VI, IX, and X), while furrowing multiple times in the sagittal plane, essentially dividing the two leaves into many buds (Fig. 1A, II–VII). Despite the entire gill disc undergoing invagination, the filament budding is an outward axial process as the branch lumen is connected to the body cavity (Fig. 1D, IV–X). A cluster of gill cells that circumvent the external scar, the multipart chamber primordium, has bulged up by this phase, a feature that is not apparent during PI-1 (Fig. 1D, VI–VII). No columnar primordial cells were observed at the quadrifurcated trachea (Fig. 1D, VIII–XII). In LI-1, the buds, which are 30–50 µm in length and columnar shaped during PI-2, elongate into 100- to 200-µm-long tapered filaments (Fig. 2A, III and IX). In the sagittal plane, the multipart chamber primordium has expanded and invaginated to form a 60- to 70-µm-long trench (Fig. 2A, II–V), to which one end of the stem of filament buds connects (Fig. 2A, V and VI) and by which the other end the anterior external scar is surrounded (Fig. 2A, I–III). Columnar primordial cells have developed at the quadrifurcated trachea by this phase when compared with PI-1 and PI-2 (Fig. 2B, XII–XIV). The dorsal tracheal trunk has dilated to a 7–10 µm radius (Fig. 2A, V). In LI-2, the filaments elongate to 300–400 µm. The filament wall, particularly the distal surface, becomes wrinkled owing to its primordial cells, which are closely compacted and columnar shaped during LI-1, becoming enlarged, oval shaped, and loosely arranged (Fig. 3A and B). The multipart chamber primordium exhibits a series of changes between LI-1 and LI-2. The midsection of the trench that appeared in LI-1 becomes enclosed to form a tubular chamber (Fig. 3B, II–V), in which the external scar-surrounding side becomes a blind end (Fig. 3A, I–VI and B, V–VII), whereas the other end forms an opening, resembling an entrance of an underground passage, which connects to the filament stem (Fig. 3A, V). Inside the chamber opening, there is a distinct cuticle layer present which divided the chamber into parts resembling the letter ‘8’ in the sagittal plane (Fig. 3A, IV–VI). Fig. 3. Open in new tabDownload slide Partial series of continuous sagittal (A) and transverse (B) sections of Simulium xingyiense prothoracic gill discs during LI-2. Roman numerals show the order of the sections in each series. The tissues of interest—external scar, ecdysial tube, and trachea—are denoted as es, et, and t, respectively. Filaments are numbered as c1–c5 (five cuticular side buds) and i1–i3 (three interior side buds). The structure highlighted by the dashed lines is a trench formed by the invagination of the multipart chamber primordium. The arrows indicate the pupal spiracular apparatus. Scale bar = 100 µm. Fig. 3. Open in new tabDownload slide Partial series of continuous sagittal (A) and transverse (B) sections of Simulium xingyiense prothoracic gill discs during LI-2. Roman numerals show the order of the sections in each series. The tissues of interest—external scar, ecdysial tube, and trachea—are denoted as es, et, and t, respectively. Filaments are numbered as c1–c5 (five cuticular side buds) and i1–i3 (three interior side buds). The structure highlighted by the dashed lines is a trench formed by the invagination of the multipart chamber primordium. The arrows indicate the pupal spiracular apparatus. Scale bar = 100 µm. The observations from LI-3 forward are complementarily consistent with previous description on Simulium gills before and after pupation (Hinton 1957, 1976). In the LI-3, filament cuticular sheath—which eventually give rise to the plastron after pupation—is formed, which comprise a forest of branched struts situated vertically to the filament surface (Figs. 4B and 5A, I–II). The distal parts of the struts connect to each other forming a thin mesh layer (Figs. 4B, IV and 5A). The strut-formed sheath is thicker (20–30 µm) in the proximal region—particularly the filament stem—than the distal area on the filaments, where it is less than 10 µm thick. As shown by Fig. 5A (III–V), immediately adjacent to the opening of invaginated chamber, there is a thin layer of stained membrane formed at the larval body wall, whereas a string of stained material, presumably chitin, present at the gill lumen entrance. The lumen entrance as shown is disjointed with the larval body but left unenclosed by the disc cells to form a basal fenestra, an unusually thin cuticular oval membrane that bursts at pupation to allow ambient water to enter the gill lumen (Adler et al. 2004). This is deduced from its anatomical position, oval shape, and absence of plastron and cell tissue. Fig. 4. Open in new tabDownload slide Partial sagittal section series of Simulium xingyiense prothoracic gill discs during LI-3. The sections (A) and (B) show a continuous series. Roman numerals show the order of the sections in each series. The series shown in (C) represents a set of skewed sagittal sections with detail of the adult spiracular end of the multipart chamber. The boxed tissues of interests are magnified twofold in the upper right corner. The tissues of interest—plastron, invaginated multipart chamber, filter apparatus, fine struts, external scar, ecdysial tube, and trachea—are denoted as p, ic, fa, fs, es, et, and t, respectively. Filaments are numbered as c1–c5 (five cuticular side buds) and i1–i3 (three interior side buds). The arrows indicate the pupal spiracular apparatus. Scale bars represent 100 µm in (A), 200 µm in (B), and 50 µm in the magnified boxes. Fig. 4. Open in new tabDownload slide Partial sagittal section series of Simulium xingyiense prothoracic gill discs during LI-3. The sections (A) and (B) show a continuous series. Roman numerals show the order of the sections in each series. The series shown in (C) represents a set of skewed sagittal sections with detail of the adult spiracular end of the multipart chamber. The boxed tissues of interests are magnified twofold in the upper right corner. The tissues of interest—plastron, invaginated multipart chamber, filter apparatus, fine struts, external scar, ecdysial tube, and trachea—are denoted as p, ic, fa, fs, es, et, and t, respectively. Filaments are numbered as c1–c5 (five cuticular side buds) and i1–i3 (three interior side buds). The arrows indicate the pupal spiracular apparatus. Scale bars represent 100 µm in (A), 200 µm in (B), and 50 µm in the magnified boxes. Fig. 5. Open in new tabDownload slide Partial series of continuous transverse sections of Simulium xingyiense prothoracic gill discs during LI-3. Roman numerals show the order of the sections in each series. The series depicted in (A) and (B) are each continuous, but some slides have been omitted between each series. The tissues of interest—plastron, multipart chamber, fine struts, external scar, ecdysial tube, and trachea—are denoted as p, ic, fs, es, et, and t, respectively. The arrows indicate the pupal spiracular apparatus that conjoined with the chitin hardened opening. Scale bar = 100 µm. Fig. 5. Open in new tabDownload slide Partial series of continuous transverse sections of Simulium xingyiense prothoracic gill discs during LI-3. Roman numerals show the order of the sections in each series. The series depicted in (A) and (B) are each continuous, but some slides have been omitted between each series. The tissues of interest—plastron, multipart chamber, fine struts, external scar, ecdysial tube, and trachea—are denoted as p, ic, fs, es, et, and t, respectively. The arrows indicate the pupal spiracular apparatus that conjoined with the chitin hardened opening. Scale bar = 100 µm. The multipart chamber primordium dilated at both ends to form a dumbbell-shaped tube where the invagination slit at the midpart is completely sealed off showing subchambers: distal, spiral-ridged, and adult spiracular subchambers (Fig. 4B, I–IV). Inside the adult spiracular subchamber, the dorsal tracheal trunk connected to the ecdysial tube, dilated to approximately 20–30 µm, is bent at a perpendicular angle and exposed to the chamber interior (Fig. 5A, IV and B, V–VI). This subchamber becomes an oval-shaped orifice with a cluster of fine protruding struts that overlays the exposed trachea (Figs. 4B, I–IV and C, I–IV and 5A, I–IV). The oval orifice with protruding struts eventually gives rise to the adult anterior spiracle. At the distal subchamber, the opening is sealed by a cuticularized and hardened peritreme that is completely filled by the plastronal struts (Fig. 4B, I–IV), while the layer of cuticle that divides the distal and spiral-ridged subchambers becomes patently hardened forming an apart of the pupal spiracle apparatus, an apodeme that is a part of a ‘pinch-cock’ device. A cuticle layer with spiral ridges, similar to the tracheal wall but thicker, has formed in the mid-subchamber (Fig. 4A, V–VII). Detailed first by Hinton (1957), muscle tissues are involved in constituting the pupal and adult spiracular apparatuses that constrict and divide the invaginated chamber into distal, spiral-ridged, and adult spiracular subchambers. These accounts are in agreement with our observation here. A schematic representation of the protruding gill and cuticular plastron and their relationships with the invaginated multipart chamber is illustrated in Fig. 6. Fig. 6. Open in new tabDownload slide A schematic representation of a developing Simulium protruded gill and invaginated multipart chamber. Fig. 6. Open in new tabDownload slide A schematic representation of a developing Simulium protruded gill and invaginated multipart chamber. Gills of Other Simulium spp. The early phases of gill development in the other Simulium spp. appear to show the same patterns of buddings between PI-1 and PI-2 (Supp Figs. S3 and S5 [online only]). The bulging primordium of multipart chamber is not present in PI-1 (Supp Fig. S5A and B) but is present in the PI-2 gills of most Simulium spp. (Supp Fig. S5, C’, D’ and E’ [online only]). In LI-1, although S. takahasii and S. xingyiense (from the Wilhelmia subgenus) grow shorter and stockier filaments (Fig. 3A and Supp Fig. S5C, I–II [online only]), all other species grow longer and slimmer filaments in a clockwise inward direction that become more convoluted (Supp Fig. S6A–B and D–F [online only]). In LI-2, although a wrinkled filament surface was present in all the species we investigated, those with slimmer and longer branches showed less furrowing (Fig. 7 and Supp Fig. S7 [online only]). Finally, no apparent differences were observed regarding the morphogenesis of the invaginated multipart chamber, while basal fenestra formation or the disjointing of gill and body occurs at this time. However, the thickness of the plastron that seals the multipart chamber entrance does appear to vary between species, with larger clusters of filament stems yielding thicker proximal plastrons (Fig. 7). Fig. 7. Open in new tabDownload slide Transverse sections of Simulium gill discs during LI-2 and LI-3. The larval tissues in (A)–(G) and (A’)–(G’) are sectioned from LI-2 and LI-3 larvae of S. xingyiense, S. takahasii, S. ornatum, S. bidentatum, S. aureohirtum, S. quinquestriatum, and S. guiyangense, respectively. The tissues of interest—furrowed multipart chambers and segmented leg discs—are denoted as fr and ld, respectively. Scale bar = 250 µm. Fig. 7. Open in new tabDownload slide Transverse sections of Simulium gill discs during LI-2 and LI-3. The larval tissues in (A)–(G) and (A’)–(G’) are sectioned from LI-2 and LI-3 larvae of S. xingyiense, S. takahasii, S. ornatum, S. bidentatum, S. aureohirtum, S. quinquestriatum, and S. guiyangense, respectively. The tissues of interest—furrowed multipart chambers and segmented leg discs—are denoted as fr and ld, respectively. Scale bar = 250 µm. Wing Discs of Simulium spp. At the PI-1 instar, wing discs undergo invagination showing disc sacks with no clear morphogenesis observed (Supp Fig. S8A and C [online only]). In PI-2, the wing discs exhibit single-fold axial outgrowth as filament budding occurs (Supp Fig. S8B and E [online only]). With the advent of the last instar, the developing leg discs become segmented, whereas wing discs clearly show tracheal veins (Supp Fig. S8D–E [online only]). At the LI-2, primary wing parts (tracheal veins and axillary muscle primordia) have appeared (Supp Fig. S9A–F [online only]). Finally, in LI-3, the developing wings have formed cuticle sheaths and exhibit wrinkles and furrows similar to the filament wrinkles (Supp Fig. S9A–F [online only]). After pupation, the wrinkles and furrows of the gills and wings stretch and flatten, and the cuticle hardens significantly, expanding the surfaces that interact with the ambient environment. Discussion Our descriptions of the gill development of Simulium spp. using serial sections provide a novel holistic view on the development of the organ. These include the formation of enigmatic features described by previous writers, e.g. Hinton (1957, 1964, 1976) and Adler et al. (2004), such as the basal fenestra, the pupal spiracular apparatus, the multipart chamber that in part give rise to the anterior adult spiracle. Furthermore, the Simulium spp. we identified included the widely distributed species S. quinquestriatum, S. ornatum, and S. bidentatum, among others, in which the branch numbers are even and ranged from 4 to 10, representing the majority of Simuliidae species (72.8% among North American species), even though the plesiomorphic Parasimulium shows three main branches (Rubtsov 1995, Adler et al. 2004, Coscaron and Coscaron-Arias 2007, Chen et al. 2016). The two-leaf axial budding pattern explains, to some extent, why the branches of most Simuliidae species bud in even numbers. Assuming the proportion of energy allocated for gill development during metamorphosis is fixed in some way, the number of cells required to form a gill must be fixed within a small range. At the same time, only limited space is available for the developing gill to occupy because wing and leg discs are also taking up space inside the larval cuticle. Because number of cells and space for arranging a gill are limited or even fixed, an S-curve relationship between the gill surface area and the branch number can be reasonably deduced. On the one hand, the surface area variation along the branch number when branch number is low (e.g., 20) is going to significantly affect the individual’s survival. When branch number exceeds dozens, on the other hand, the branches are inevitably small; thus, one or two branches added or removed does not affect the survival in a significant way. Notwithstanding the plesiomorphic gill is trifurcated, axial symmetrical branch budding implies tightly controlled and repeating patterning, inherently shows selective advantage. Hence, even numbered gill gradually becomes the predominate gill type in the extending Simuliidae species. Meanwhile, some species with plesiomorphic and odd number of branches and exceptionally flourishing number of branches become the minority gill types. The developing gills exhibit many characteristics of imaginal discs and show little resemblance to the developing trachea. Taking Drosophila melanogaster and Manduca sexta as examples, tracheal morphogenesis does not involve cell proliferation during embryonic development, whereas, at the larval stage, the tracheal system sprouts new branches periodically around each molt, but arrests in development otherwise (Ghabrial et al. 2003, Callier and Nijhout 2011, Nijhout and Callier 2015). In comparison, gill budding and chamber invagination are continuous processes that begin at the larval stage for both Simuliidae (described in the present study) and Psychodidae (described previously; Satchell 1948). Additionally, branch sprouting in the tracheal system, guided by Branchless FGF signaling in hypoxic tissues, is different from the branch budding of Simuliidae and Psychodidae where a disc-like axial branching pattern is observed (Satchell 1948). Furthermore, although trachea-like spiral ridges form in the interior of the multipart chamber, fine struts present in the walls of distal and adult spiracular subchambers. The gill formation of Simuliidae, Psychodidae, and many other dipteran insects requires muscle tissue, to which the multipart chamber is attached, which creates a pinch valve that allows adjustment of the air exchange volume (Satchell 1948, Hinton 1968). The development of a multipart chamber with interior strut ornaments and pinch valve do not occur during tracheal development, as shown by reports on D. melanogaster and M. sexta (Callier and Nijhout 2011, Ghabrial et al. 2003, Nijhout and Callier 2015). Thus, its pattern of development shows that it is unlikely that the gill originates from a tracheal extension, instead fulfilling many of the criteria of appendage development (e.g., axial outgrowth of discs, involvement of muscle tissue, and cuticle sheath formation). The descriptions of Simuliidae and Psychodidae gill morphogenesis combined with those of respiratory organ development in other families collectively suggest that the lower Diptera share the same developmental pattern. The key feature explaining discrepancies between these taxa is the way the protruding gill mechanically connects and to the multipart chamber. Although forming a monolayered outgrowth, the developing gills of Psychodidae gradually furrow to form a columnar double-layered tube or cone, called felt chamber, in which the proximal end circumvents and encloses the opening of the multipart chamber underneath (Satchell 1948). The lumen of the felt chamber and the invaginated chamber are integrated into one large vertical columnar lumen (after pupation) as the wrapping slit of the double-layered tube is sealed up presumably leaving small pores along the slit (Satchell 1948). And only the most primitive Simuliidae, Parasimulium, possess this structure, whereas, contrast to the Psychodidae and Parasimulium, Simuliinae gills, formed by monolayered cells, only have the proximal thick plastron to seal up the opening of the multipart chamber (Adler et al. 2004). Understandably, the felt chamber type of gill structures could be interpreted as originating from a tracheal extension when only the morphology of the pupal gill is taken into account. To date, our knowledge of the respiratory organs of many other dipteran insects have been limited to morphological or histological descriptions at the pupal stage; our work, combined with the descriptions of developing Psychodidae gills, collectively provides a framework for understanding dipteran prothoracic gill development that can allow further investigation. Of the plastron-bearing dipteran insects, Deuterophlebiidae (the most ancestral dipteran family), Blephariceridae (Psychodomorpha), and Simuliidae (Culicomorpha) pupate on substrates in running streams and are generally immobile. Interestingly, in the Deuterophlebiidae and Blephariceridae gills, the protruding tube (the felt chamber) is not present similar to the Simuliinae gill structure, whereas the Deuterophlebiidae gill is monolayered and Blephariceridae gill section is unavailable (Hinton, 1968; Arens 1995, 1998). Meanwhile, Tipuloidea (Tipulomorpha), Psychodidae (Psychodomorpha), Chironomidae, and Culicidae (Culicomorpha), and some higher dipteran insects (Canaceidae and Dolichopodidae) share the Psychodidae gill structure in which felt chamber connects to the invaginated chamber, as their pupae are more mobile and generally live freely in lagoons, lakes, or ponds (Satchell 1948; Reid 1963; Hinton 1966, 1968; Arens 1995, 1998; Armitage et al. 1995; Nishiura 2002; Ha et al. 2017). As shown by sections in Hinton (1968), many of the Tipulidae gill must share the Psychodidae gill morphogenesis because they also possess double-layered gill wall with a slit on one side. The contrast between the two types of gill development sheds some light on pupal behavioral and habitational adaptations, which could establish grounds for the systemic classification of dipteran respiratory organs. In higher Diptera, the prothoracic dorsal (humeral) discs of D. melanogaster start to develop in the last instar. From a pair of protruding cuticularized anterior spiracles, the surrounding donut-shaped discs invaginate along the spiracular tracheae toward the dorsal tracheal trunks, forming a chamber that eventually surrounds the joint of the spiracular trachea and the dorsal trachea (Madhavan and Schneiderman 1977, Pedelty and Arking 1981). Presumably, this structure becomes the adult thoracic spiracle after it is filled with gas due to results shown in a recent study (Wang et al. 2018). The humeral disc morphogenesis is the same as that of, and resulting in similar structural configurations as the multipart chamber of lower dipteran insects. As the flies have diverged from a single lineage of aquatic dipteran insect (Wiegmann et al. 2011), it makes sense for the flies to gradually lose the budding gills with only the multipart chamber being retained. Most notably, this disc has long been shown to have potential of forming wing structures when Sex comb reduced (Scr) expression is lowered (Rogers et al. 1997). In summary, comparisons of respiratory organ development show that dipteran prothoracic gills share essentially the same morphogenesis pattern with less obvious discrepancies (such as formation of basal fenestra, gill disc wrapping, and enclosing process) that made seemingly vastly different organs. This indicates that an ancestral prothoracic dorsal appendage gave rise to all dipteran respiratory organs. Evolutionary studies on the epipods/gills of Branchiopoda (ancestral arthropods) suggest that these features may be the ancestral form of insect wings and abdominal book gills of aquatic chelicerates as well as the book lung, tubular trachea, and spinnerets of spiders (Averof and Cohen 1997, Damen et al. 2002). Recently, a series of studies on a ground plan gene (scr) have suggested the enlarged pronotum of some hemimetabolous insects and the prothoracic horn of horned beetles are homologous to the wings (Chesebro et al. 2009, Medved et al. 2015, Elias-Neto and Belles 2016, Hu et al. 2019). These studies have further supported our suspicion of the origins of the respiratory organs has some relations to wing evolution. The Simuliidae and the other taxa for which data on gill development are available are not model organisms, and thus, there is currently a lack of sufficient genetic data to validate our key hypothesis; however, our study establishes a base from which further investigations into respiratory organ evolution using ground plan genes (scr and other Hox genes) in model organisms (mosquitos and fruit flies) can be carried out. Acknowledgments We thank equipment and assistances from the Key Laboratory of Medical Entomology and the Research Center of Basic Medical College, Guizhou Medical University, China. This work was funded by the National Natural Science Foundation of China (grant no. 31860324) and the Foundation of Science and Technology of Guizhou Medical University (grant no. YJ2016-27), awarded to M.Y. References Cited Adler , P H , D C Currie, and D M Wood. 2004 . The black flies (Simuliidae) of North America . Cornell University Press , New York . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Arens , W . 1995 . Structure and evolution of spiracular gills in pupae of net-winged midges (Nematocera; Blephariceridae). Part I. Paulianina and Edwardsina (subfamily Edwardsininae) . Can. J. Zool . 73 : 2318 – 2342 . Google Scholar Crossref Search ADS WorldCat Arens , W . 1998 . Structure and evolution of spiracular gills in pupae of net-winged midges (Nematocera; Blephariceridae): Part III: gill diversity in Paulianina (subfamily Edwardsininae) . Ann. Natal Mus . 39 : 83 – 114 . Google Scholar OpenURL Placeholder Text WorldCat Armitage , P D , P S Cranston, and L C V Pinder. 1995 . The Chironomidae: biology and ecology of non-biting midges . Chapman and Hall , London, United Kingdom . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Averof , M , and S M Cohen. 1997 . Evolutionary origin of insect wings from ancestral gills . Nature 385 : 627 – 630 . Google Scholar Crossref Search ADS PubMed WorldCat Borkent , A . 2012 . The pupae of Culicomorpha – morphology and a new phylogenetic tree . Zootaxa 3396 : 1 – 98 . Google Scholar Crossref Search ADS WorldCat Callier , V , and H F Nijhout. 2011 . Control of body size by oxygen supply reveals size-dependent and size-independent mechanisms of molting and metamorphosis . Proc. Natl. Acad. Sci. USA 108 : 14664 – 14669 . Google Scholar Crossref Search ADS WorldCat Chen , H B , C L Zhang, M Yang, and X J Wen. 2016 . Chinese blackflies (Diptera: Simuliidae) . Guizhou Science and Technology Publishing House , Guiyang, China . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Chesebro , J , S Hrycaj, N Mahfooz, and A Popadić. 2009 . Diverging functions of Scr between embryonic and post-embryonic development in a hemimetabolous insect, Oncopeltus fasciatus . Dev. Biol . 329 : 142 – 151 . Google Scholar Crossref Search ADS PubMed WorldCat Clark-Hachtel , C M , D M Linz, and Y Tomoyasu. 2013 . Insights into insect wing origin provided by functional analysis of vestigial in the red flour beetle, Tribolium castaneum . Proc. Natl. Acad. Sci. USA 110 : 16951 – 16956 . Google Scholar Crossref Search ADS WorldCat Coscaron , S , and C L Coscaron-Arias. 2007 . Neotropical Simuliidae (Diptera: Insecta) . Pensoft , Moscow, Russia . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Crosskey , R W . 1962 . The identification of the larvae of African Simulium . Bull. World Health Organ . 27 : 483 – 489 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Damen , W G , T Saridaki, and M Averof. 2002 . Diverse adaptations of an ancestral gill: a common evolutionary origin for wings, breathing organs, and spinnerets . Curr. Biol . 12 : 1711 – 1716 . Google Scholar Crossref Search ADS PubMed WorldCat Elias-Neto , M , and X Belles. 2016 . Tergal and pleural structures contribute to the formation of ectopic prothoracic wings in cockroaches . R. Soc. Open Sci . 3 : 160347 . Google Scholar Crossref Search ADS PubMed WorldCat Emlen , D J , L Corley Lavine, and B Ewen-Campen. 2007 . On the origin and evolutionary diversification of beetle horns . Proc. Natl. Acad. Sci. USA 104 ( Suppl. 1 ): 8661 – 8668 . Google Scholar Crossref Search ADS WorldCat Eymann , M . 1991 . Flow patterns around cocoons and pupae of black flies of the genus Simulium (Diptera: Simuliidae) . Hydrobiologia 215 : 223 – 229 . Google Scholar Crossref Search ADS WorldCat Folmer , O , M Black, W Hoeh, R Lutz, and R Vrijenhoek. 1994 . DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates . Mol. Mar. Biol. Biotechnol . 3 : 294 – 299 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Ghabrial , A , S Luschnig, M M Metzstein, and M A Krasnow. 2003 . Branching morphogenesis of the Drosophila tracheal system . Annu. Rev. Cell Dev. Biol . 19 : 623 – 647 . Google Scholar Crossref Search ADS PubMed WorldCat Gullan , P J , and P S Cranston. 2010 . The insects: an outline of entomology , 3rd ed. Wiley-Blackwell , Oxford, United Kingdom . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Ha , Y R , E Yeom, J Ryu, and S J Lee. 2017 . Three-dimensional structures of the tracheal systems of Anopheles sinensis and Aedes togoi pupae . Sci. Rep . 7 : 44490 . Google Scholar Crossref Search ADS PubMed WorldCat Hinton , H E . 1957 . Some little known respiratory adaptations . Sci. Prog . 45 : 692 – 700 . Google Scholar OpenURL Placeholder Text WorldCat Hinton , H E . 1964 . The respiratory efficiency of the spiracular gill of Simulium . J. Insect Physiol . 10 : 73 – 80 . Google Scholar Crossref Search ADS WorldCat Hinton , H E . 1966 . Plastron respiration in marine insects . Nature 209 : 220 – 221 . Google Scholar Crossref Search ADS WorldCat Hinton , H E . 1968 . Spiracular gills . Adv. Insect Physiol . 5 : 65 – 162 . Google Scholar Crossref Search ADS WorldCat Hinton , H E . 1976 . The fine structure of the pupal plastron of Simuliid files . J. Insect Physiol . 22 : 1061 – 1070 . Google Scholar Crossref Search ADS WorldCat Hu , Y , D M Linz, and A P Moczek. 2019 . Beetle horns evolved from wing serial homologs . Science 366 : 1004 – 1007 . Google Scholar Crossref Search ADS PubMed WorldCat Kukalova-Peck , J . 1978 . Origin and evolution of insect wings and their relation to metamorphosis, as documented by the fossil record . J. Morphol . 156 : 53 – 125 . Google Scholar Crossref Search ADS PubMed WorldCat Larivière , M C , D Burckhardt, and A Larochelle. 2011 . Peloridiidae (Insecta: Hemiptera: Coleorrhyncha) . Fauna NZ 67 : 1 – 78 . Google Scholar OpenURL Placeholder Text WorldCat Linz , D M , and Y Tomoyasu. 2018 . Dual evolutionary origin of insect wings supported by an investigation of the abdominal wing serial homologs in Tribolium . Proc. Natl. Acad. Sci. USA 115 : E658 – E667 . Google Scholar Crossref Search ADS WorldCat Madhavan , M M , and H A Schneiderman. 1977 . Histological analysis of the dynamics of growth of imaginal discs and histoblast nests during the larval development of Drosophila melanogaster . Roux. Arch. Dev. Biol . 183 : 269 – 305 . Google Scholar Crossref Search ADS WorldCat McAlpine , J F , B V Peterson, G E Shewell, H J Teskey, J R Vockeroth, and D M Wood. 1981 . Manual of Nearctic Diptera , Vol. 1 . Canadian Government Publishing Centre , Hull, Canada . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC McAlpine , J F , B V Peterson, G E Shewell, H J Teskey, J R Vockeroth, and D M Wood. 1989 . Manual of Nearctic Diptera , Vol. 2 . Canadian Government Publishing Centre , Hull, Canada . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Medved , V , J H Marden, H W Fescemyer, J P Der, J Liu, N Mahfooz, and A Popadić. 2015 . Origin and diversification of wings: insights from a neopteran insect . Proc. Natl. Acad. Sci. USA 112 : 15946 – 15951 . Google Scholar Crossref Search ADS WorldCat Miko , I , F Friedrich, M J Yoder, H M Hines, L L Deitz, M A Bertone, K C Seltmann, M S Wallace, and A R Deans. 2012 . On dorsal prothoracic appendages in treehoppers (Hemiptera: Membracidae) and the nature of morphological evidence . PLoS One 7 : e30137 . Google Scholar Crossref Search ADS PubMed WorldCat Nijhout , H F , and V Callier. 2015 . Developmental mechanisms of body size and wing-body scaling in insects . Annu. Rev. Entomol . 60 : 141 – 156 . Google Scholar Crossref Search ADS PubMed WorldCat Nishiura , J T . 2002 . Coordinated morphological changes in midgut, imaginal discs, and respiratory trumpets during metamorphosis of Aedes aegypti (Diptera: Culicidae) . Ann. Entomol. Soc. Am . 94 : 498 – 594 . Google Scholar OpenURL Placeholder Text WorldCat Pedelty , J , and R Arking. 1981 . The metamorphosis of the humeral disc of Drosophila melanogaster . J. Exp. Zool . 216 : 141 – 148 . Google Scholar Crossref Search ADS WorldCat Prokop , J , M Pecharová, A Nel, T Hörnschemeyer, E Krzemińska, W Krzemiński, and M S Engel. 2017 . Paleozoic nymphal wing pads support dual model of insect wing origins . Curr. Biol . 27 : 263 – 269 . Google Scholar Crossref Search ADS PubMed WorldCat Prud’homme , B , C Minervino, M Hocine, J D Cande, A Aouane, H D Dufour, V A Kassner, and N Gompel. 2011 . Body plan innovation in treehoppers through the evolution of an extra wing-like appendage . Nature 473 : 83 – 86 . Google Scholar Crossref Search ADS PubMed WorldCat Rasnitsyn , A P , and O D Strelnikova 2017 . Tracheal system and biology of the early cretaceous Saurophthirus longipes Ponomarenko, 1976 (Insecta, ?Aphaniptera, Saurophthiroidea stat. nov.) . Paleontol. J . 51 : 171 – 182 . Google Scholar Crossref Search ADS WorldCat Reid , J A . 1963 . A note on the structure of the pupal trumpet in some mosquitoes . Proc. R. Entomol. Soc. A . 38 : 32 – 38 . Google Scholar OpenURL Placeholder Text WorldCat Rodrigues , H M , and G J Svenson. 2018 . Epaphroditidae sensu novo, an endemic caribbean family of morphologically divergent praying mantises (Insecta, Mantodea) . Neotrop. Entomol . 47 : 502 – 507 . Google Scholar Crossref Search ADS PubMed WorldCat Rogers , B T , M D Peterson, and T C Kaufman. 1997 . Evolution of the insect body plan as revealed by the Sex combs reduced expression pattern . Development 124 : 149 – 157 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Rubtsov , I A . 1995 . Blackflies (Simuliidae) , 2nd ed. Brill , Leiden, The Netherlands . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Satchell , G H . 1948 . The respiratory horns of Psychoda pupae . Parasitology 39 : 43 – 52 . Google Scholar Crossref Search ADS PubMed WorldCat Schwarz , C J , and O Konopik. 2014 . An annotated checklist of the praying mantises (Mantodea) of Borneo, including the results of the 2008 scientific expedition to Lanjak Entimau Wildlife Sanctuary, Sarawak . Zootaxa 3797 : 130 – 168 . Google Scholar Crossref Search ADS WorldCat Tomoyasu , Y , T Ohde, and C Clark-Hachtel. 2017 . What serial homologs can tell us about the origin of insect wings . F1000Res 6 : 268 . Google Scholar Crossref Search ADS PubMed WorldCat Truman , J W , and L M Riddiford. 1999 . The origins of insect metamorphosis . Nature 401 : 447 – 452 . Google Scholar Crossref Search ADS PubMed WorldCat Wang , Y , J Berger, and B Moussian. 2018 . Trynity models a tube valve in the Drosophila larval airway system . Dev. Biol . 437 : 75 – 83 . Google Scholar Crossref Search ADS PubMed WorldCat Wiegmann , B M , M D Trautwein, I S Winkler, N B Barr, J W Kim, C Lambkin, M A Bertone, B K Cassel, K M Bayless, A M Heimberg, et al. . 2011 . Episodic radiations in the fly tree of life . Proc. Natl. Acad. Sci. USA 108 : 5690 – 5695 . Google Scholar Crossref Search ADS WorldCat Yang , Y M , R Jia, H Xun, J Yang, Q Chen, X G Zeng, and M Yang. 2018 . Determining the number of instars in Simulium quinquestriatum (Diptera: Simuliidae) using k-means clustering via the Canberra distance . J. Med. Entomol . 55 : 808 – 816 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2020. Published by Oxford University Press on behalf of Entomological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Comparisons of Respiratory Pupal Gill Development in Black Flies (Diptera: Simuliidae) Shed Light on the Origin of Dipteran Prothoracic Dorsal Appendages
JF - Journal of Medical Entomology
DO - 10.1093/jme/tjaa208
DA - 2021-03-12
UR - https://www.deepdyve.com/lp/oxford-university-press/comparisons-of-respiratory-pupal-gill-development-in-black-flies-dAoHcSAqyh
SP - 588
EP - 598
VL - 58
IS - 2
DP - DeepDyve
ER -