A Review of the Genus Oecetis (Trichoptera: Leptoceridae) in the Northeastern Region of Brazil with the Description of 5 New Species

Within Leptoceridae, the genus Oecetis contains about 500 species around the world, including 53 in the Neotropics. In Brazil, there are 15 recorded species of Oecetis. These species were described over several decades by numerous authors with the results that descriptions are not comparable and diagnoses are incomplete. Also, the apparently unbranched M vein, in the forewing, a diagnostic character for Oecetis pointed by McLachlan, is controversial and no consensus has been reached about its homology. Additionally, the only revision for the genus was never published; thus the information and proposed taxa are not available according to the International Code of Zoological Nomenclature. We analyzed specimens collected in the Brazilian Northeast Region and compared these with described species and literature descriptions and Oecetis from other regions. We provide herein the description of five new species, additional characters for diagnosing seven of the species recorded from Brazil, new distributional records, and a dichotomous key to the Brazilian species. Additionally, we contrast the two hypotheses of forewing M vein homology and support the unbranched hypothesis. In this way, we improve the knowledge of the genus in the Neotropics, making the species descriptions comparable in a way that facilitates species identification.


Introduction
Leptoceridae is one of the three largest families of Trichoptera, with more than 2,000 described species [1]. The family is comprised of 48 extant genera, placed traditionally in two subfamilies, Triplectidinae Ulmer and Leptocerinae Leach, the former recorded from Australasian and Neotropical regions, while the latter has a cosmopolitan distribution [2]. Recently, Malm & Johanson [2] raised Grumichellini Morse, with Neotropical and Australian distribution [3], and Leptorussini Morse, with Australian distribution [3], to subfamily status, classifying the family into four subfamilies.
The majority of original descriptions of Oecetis species are not detailed, and include few diagnostic characters. The latest comprehensive treatment of the genus, including the Brazilian species, was implemented by Chen [9], in his unpublished PhD thesis. He proposed a classification of the genus into four subgenera: Pleurograpta, Pseudosetodes, Oecetis, and Quaria. The two former names were first proposed as genera by Wallengren [12] and Ulmer [13], respectively, but trated as synonyms by Fischer [14] and Flint [15], respectively. The name Quaria was proposed by Milne [16] as a subgenus of Oecetis. The subgenera recognized in Chen's unpublished work [9] are considered "species groups" here. These four species groups show very distinctive male genitalic characteristics and are easily recognizable. According to Chen [9], Pseudosetodes is the only species group that has a symmetrical phallus without phallic spines. Oecetis has only one phallic spine, and an asymmetrical phallus. Pleurograpta has a short tergum IX and a long sternum IX, and males of Quaria are easily distinguished by the presence of a pair of prominent dorsolateral processes on tergum IX.
Concentrating our efforts in the Northeast region, to minimize this disparity, we provide, herein, a review of Brazilian Oecetis species, describe five new species, and provide a synopsis of the previously described Oecetis, improving the original descriptions with new characters and illustrations. We also present a dichotomous key for all the known Brazilian species.
Additionally, it is known that some Oecetis species are difficult to diagnose [31]. In Brazil, we have records of four of these cited species: Oecetis excisa and O. inconspicua, which are stated as possible synonyms by some authors (e.g., [6]; [32]); and O. iguazu and O. punctipennis. In order to explore additional morphological characters for improving the diagnosis of these species, they were examined by using a Scanning Electron Microscope, and the micrographs and discussion of results are herein presented. In this way we improve the knowledge of caddisflies in Brazil, by revealing unknown diversity and distributional patterns, solving a long standing problem in caddisfly literature and propose a standard nomenclature and homology hypothesis for the genus wings.

Material and Methods
Specimens were collected using light traps and pan light traps [33] with incandescent and fluorescent lamps placed next to water bodies. The collecting trips were made with support of Dr. Vitor Becker (Reserve Serra Bonita) and Aquatic Entomology Lab staff (LEAq-UFBA). In addition, Malaise traps were also used over low order streams. The specimens collected were stored in 80% alcohol or pinned. Genitalia were removed, together with 4 to 5 abdominal segments, cleared for identification using 85% lactic acid solution [34] or 10% KOH and stored in microvials containing glycerin [35,36]. Illustrations were made using a compound microscope with drawing tube attached. Improvements on illustrations were made using the software Adobe Illustrator CS 5. Morphological terminology followed Snodgrass [37] and Schmid [38], as implemented by Holzenthal [35] and Calor [39]. The lengths presented in the description sections are average lengths. The DELTA System [40][41][42] was used to produce the species descriptions and the dichotomous key.
Type material will be deposited at Museu de Zoologia, Universidade de São Paulo (MZUSP), Museu de Zoologia, Universidade Federal da Bahia (UFBA) and the University of Minnesota Insect Collection (UMSP), as indicated in material examined.
The specimens examined by Scanning Electron Microscope (SEM) had their genitalia cut off with a pair of iridectomy scissors, and were subsequently positioned on a carbon tape, metalized, and then the photos were taken in a JEOL JSM 6390LV microscope. Acronyms

Nomenclatural acts
The electronic edition of this article conforms to the requirements of the amended International Code of Zoological Nomenclature, and hence the new names contained herein are available under that Code from the electronic edition of this article. This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the ICZN. The ZooBank LSIDs (Life Science Identifiers) can be resolved and the associated information viewed through any standard web browser by appending the LSID to the prefix "http://zoobank.org/". The LSID for this publication is: urn:lsid:zoobank.org:pub:01A88E9B-558A-4469-83E7-C89406BCD5E0. The electronic edition of this work was published in a journal with an ISSN, and has been archived and is available from the following digital repositories: PubMed Central and LOCKSS.
Taxonomy OECETIS ACANTHOSTEMA SP. NOV. urn:lsid:zoobank.org:act:E6612BEA-6738-4836-AFDB-2401642 39E68 (Fig 1) Diagnosis: The new species presents a pair of prominent dorsolateral processes on tergum IX of males, as in species of the Quaria species group (sensu Chen [9]), but can be distinguished from others of the group by the following characters: presence of a sclerotized projection on the mesoventral region of the phallic apparatus, and the distal 1/3 of the dorsolateral process  Oecetis in Northeast Brazil enlarged in width and flattened dorsoventrally; this process is cylindrical in other species of the species group.
Thorax: Pterothorax yellowish brown in dorsal region and pale yellow in lateral and ventral regions. Forewings hyaline, yellowish brown, with 3 dark transversal bands (not visible in Fig  1A and 1B) over crossveins s, r-m and m-cu; wing vein pattern as in Fig 1A and 1B. Hind wings each with row of setae along posterior margin; forks I and V present (Fig 1B). Legs yellowish-brown. Mid leg each with longitudinal row of spines from distal half of femur to the first tarsal segment. Tibial spur formula 1,2,2. Apical spur of fore tibia very small.
Genitalia: Dorsolateral processes of segment IX slender, bent ventrad, tapering posteriorly and extending beyond length of phallic apparatus ( Fig 1C); 2/3 rd of process enlarged in width, flattened, dorsal view ( Fig 1C); four long setae on posterior margin of segment IX, near base of dorsolateral processes (Fig 1D). Preanal appendage long, digitate with short setae. Rod-like mesodorsal process above tergum X present, with few short setae on apex ( Fig 1C). Tergum X membranous, divided medially, forming two lobes, broad basally and acute apically, with shallow cleft between them in dorsal view ( Fig 1C). Inferior appendage 1-segmented, broad basally, covered with setae over its base and mid portion; upper portion digitate, covered with small setae on top; apex rounded in lateral view ( Fig 1D); presence of small spine-like setae on inner portion in ventral view ( Fig 1E). Phallic apparatus long, bent ventrad, with spine-like projection on ventral mid portion, sclerotized on apex ( Fig 1F); asymmetrical, bearing two small, digitate mesolateral projections in dorsal view ( Fig 1G).
Holotype ( Distribution: Brazil (BA). Etymology: from Greek acanthus = spine; stemum = penis. The species name is in reference to the presence of a spine-like projection on the mesoventral region of the phallic apparatus.
Taxonomic remarks: Oecetis acanthostema sp. nov. is clearly a member of the Quaria species group due to the presence of two long dorsolateral processes on segment IX of the male genitalia. The new species has two characteristics that no other species in the genus presents: presence of a spine-like process on the mesoventral portion of the phallic apparatus, and distal 1/3 of the dorsolateral processes enlarged in width and flattened dorsoventrally, like the neck of a cobra. This species is similar to Oecetis fibra, because of the shared presence of a long phallic apparatus, short and digitate preanal appendage and similar shape of the inferior appendage, but O. acanthostema sp. nov. has an inferior appendage with a broad base. The shape of tergum X is digitate apically on O. fibra, while in the new species tergum X is similar to O. rafaeli [tergum X "divided into long, slender, lateral lobes, pointed apicad and overlaying phallic apparratus dorsolaterally", in original description of Flint ([24], p.74). However, O. rafaeli has short and straight dorsolateral processes and no rod-like process above tergum X.
OECETIS MARTINAE SP. NOV. urn:lsid:zoobank.org:act:B464EF98-7FBB-4FC4-AB7F-EBFDB1 E1C0A5 (Fig 2) Diagnosis: This species is similar to Oecetis chipiriri Martin, Gibon & Molina [6], Oecetis knutsoni Flint [22], and O. punctata (Navás) [43]. This new species has a quadrate inferior appendage, while O. chipiriri and O punctata have a rounded one. The inferior appendages of O. knutsoni and O punctata have small digitate projections on top bearing the thick setae, and a mesoventral process on the inferior appendages, both of which are absent in the new species. Moreover, O. knutsoni and O. punctata have a short, symmetrical phallic apparatus, with a Ushaped mesodorsal portion, while the new species has a long, asymmetrical, basally wide phallic apparatus, with a sinuous mesodorsal portion, which resembles the phallic apparatus of O. chipiriri. However, O. chipiriri has a rounded, thumb-like, inferior appendage, whereas Oecetis martinae sp. nov. has a quadrate one.
Thorax: Pterothorax yellow in dorsal and pale yellow in lateral and ventral regions. Forewings hyaline, yellow with brownish dots over end of veins, forks, and junctions; dark bands over crossveins absent; wing vein pattern as in Fig 2A and 2B; crossvein m-cu reaching Cu 1a (after ramification of Cu into Cu 1a and Cu 1b ). Hind wings with basal brush; forks I and V present ( Fig 2B). Legs yellowish-brown, each with row of spines on distal half of femur, all along tibia and first tarsal segment. Tibial spur formula 1,2,2; apical spur of fore tibia very small.
Holotype  This species was collected at dusk, attracted by light, in low order coldwater streams, with rocks and rough sand in bed. The vegetation around was dense, composed of rainforest, where the light traps were set and the adults captured. They are typical from higher elevations (above 500m). However, in some collection sites, such as Santa Teresinha and Pedra Branca (BA), they can be found at lower elevations, where the streams are slightly different. These water bodies are shallow with intercalated channels and small pools. The vegetation is composed of smaller and sparser trees than the rainforest. The immature stages and the female of this species are unknown.
OECETIS CLAVICORNIA SP. NOV. urn:lsid:zoobank.org:act:031C365E-7603-4E58-957E-C267 D8388133 (Fig 3) Diagnosis: The new species is similar to Oecetis scoparia Flint [18]. However, it has a small quadrate process on the dorsal region of the base of the inferior appendage, which is absent in O. scoparia. Moreover, the new species has a mesodorsal process above tergum X with a distinctly clavate apex, whereas O. scoparia has only a slightly clavate process, as described by Flint [18].
Thorax: Pterothorax yellowish brown in dorsal region and yellow in lateral and ventral regions. Forewings hyaline, yellowish brown with dark bands over cord; tuft of dark long setae present on base of M vein ( Fig 3A); wing vein pattern as in Fig 3A and 3B; crossvein m-cu Oecetis in Northeast Brazil reaching Cu 1a (after branching of Cu into Cu 1a and Cu 1b ). Hind wings with basal brush; forks I and V present ( Fig 3B). Legs yellowish brown, middle and posterior legs both with row of small spines on distal half of femur, all along tibia and tarsus. Tibial spur formula 0,2,2.
Holotype Distribution: Brazil (BA). Etymology: from Latin clavos = club, cornus = horn. The epithet is in reference to the clubshaped process over tergum X.
Taxonomic and collection remarks: this species seems to be morphologically related to Oecetis peruviana (Banks) [17], O. scoparia Flint [18] and O. traini Rueda-Martín, Gibon & Molina [6]. Tergum X and preanal appendages of all four species are similar, as are the inferior appendages, except for the small quadrate basal process in the new species. However, the phallic apparatus of each of these species is different; that of the new species is most similar to Oecetis scoparia. The clavate process with a ventral hump is an exclusive character of Oecetis clavicornia sp. nov. (O. scoparia has a clavate process, but the hump is dorsal). Additionally, Oecetis peruviana, O. scoparia and the new species share a setal brush on the base of the M vein.
This species was collected under the same conditions and in the same collecting sites as Oecetis martinae sp. nov. The only exception is for the specimens collected in Uruçuca, around the Conduru Mountains State Park, where the specimens were collected in shallow streams surrounded by a sparser rainforest than those in Wenceslau Guimarães. Like Oecetis martinae sp. nov., this species seems to be typical from higher altitudes. The immature stages and female of this species are unknown.
OECETIS FURCATA SP. NOV. urn:lsid:zoobank.org:act:F5C0B55C-BFBE-43BC-95C4-27042AE4 66B5 (Fig 4) Diagnosis: This species is a member of the Quaria species group and can be diagnosed from its congeners by having dorsolateral processes on segment IX that are divided into ventral and dorsal lobes, with the ventral one shorter than the dorsal one. Also, the inferior appendages have a singular shape among species in the genus, with a long and terete dorsal lobe and a very acute, spine like distal lobe. The new species seems to be morphologically similar to Oecetis falicia Denning & Sykora [44], although the dorsal lobe on the dorsolateral process on O. falicia is very short and the distal lobe on the inferior appendage is not acute.
Thorax: pterothorax yellowish brown in dorsal region and pale yellow in lateral and ventral regions. Forewings yellowish brown with cord crossveins slightly thickened; wing vein pattern as Fig 4A and 4B; crossvein m-cu reaching Cu 1a (after branching of Cu into Cu 1a and Cu 1b ). Hind wings with basal brush; forks I and V present; false vein near Cu 1a (Fig 4B). Legs yellowish brown, mid and hind legs both with with row of small spines on tibia and tarsus. Tibial spur formula 1,2,2, fore tibial spur very small.
Holotype Taxonomic and collection remarks: based on the presence of the dorsolateral processes on segment IX, this new species is a member of Quaria species group. There are a few described species recognized in this group, such as Oecetis falicia Denning [44], O. fibra Chen & Morse in Quinteiro & Calor [7], O. morsei Bueno-Soria [45] and O. rafaeli Flint [24]. However, Chen included additional undescribed species in his unpublished thesis. Within Quaria, only O. acanthostema sp. nov. and O. furcata sp. nov. have the process over tergum X enlarged apically; all of the other members of the species group have the process rod-like. With regard to the shape of the phallic apparatus, the new species is similar to O. falicia, but it seems that the former has a shorter phallobase compared to the latter. This shape of phallic apparatus seems to be very common in Quaria group.
This species is known only from Serra Bonita, a reserve of Atlantic Forest in the south of Bahia State. It was collected near coldwater low order streams with dense surrounding rainforest. The immature stages and female of this species are unknown.
OECETIS  [18], but O. froehlichi has a thumb-like dorsal lobe with spines over the surface, not found in any other species in the genus. Also, the inferior appendages are thinner than O. scoparia. The new species has an acuminate lateral projection on segment IX, as in Oecetis osteni Milne [16], but that of the new species is shorter. The new species also has a long phallic spine, as in Chen´s illustration of O. excisa [19]  Head: yellowish brown (alcohol). Antennae long, about 3 times length of forewings; scape and pedicel both short and stout. Maxillary palps yellowish brown, 5-segmented, all segments subequal in length and width, densely covered with setae. Labial palps yellow, 3-segmented.
Thorax: Pterothorax brown in dorsal region and pale yellow in lateral and ventral regions. Forewings hyaline, yellowish brown with dark bands over cord and dark spots over base of M, fork of Cu 1 and Cu 2 and at ends of M, Cu 1a , Cu 1b and Cu 2 ( Fig 5A); wing vein pattern as Fig  5A and 5B; vein M-Cu reaching Cu 1a (after branching of Cu into Cu 1a and Cu 1b ). Hind wings with basal brush; forks I and V present; fork I very short; false vein near Cu 1a (Fig 5B). Legs yellow, mid legs each with row of small spines on tibia and tarsus. Tibial spur formula 0,2,2.
Holotype Taxonomic and collection remarks: this new species shares a few characters with some congeners. Its tergum X is similar to O. scoparia, O. traini and O. clavicornia sp. nov., as is the shape of the inferior appendage. However, the shape of the phallic apparatus and phallic spine are unique. In addition, the bilobate endotheca is not common in Oecetis. This more complex endotheca seems to be present more often in members of Quaria species group. However, the posterolateral processes in O. froehlichi does not seem to be homologous to those dorsolateral processes that define the Quaria group. In this way, it seems that this species could not be in the Quaria group. Even so, the presence of a phallic spine, and complex endotheca may indicate an affinity with the Quaria species group. Additionally, an m-shaped mark can be seen after the clearing process on the base of the mesodorsal process above tergum X. This mark was not found in any other species of Oecetis.
This species is known only from Estação Ecológica Wenceslau Guimarães and the adults were collected with light traps. The environmental conditions were the same as for Oecetis martinae sp. nov., Oecetis clavicornia sp. nov. and Oecetis furcata sp. nov. The immature stages and female of this species are unknown.

Additional species records and synopsis of Brazillian species
In an attempt of facilitate Oecetis identification, and to provide a foundation for subsequent works on morphology, we present redescriptions and new illustrations of species of Oecetis previously recorded from Brazil, including new characters and a standardized terminology. We also provide new distribution records. The primary distributions were based on Flint et al. Head: color yellowish brown [17]. Antennae long, about 3 times length of forewings; scape and pedicel both short and stout, about same length as segments of flagellum. Maxillary palps brown, 5-segmented, all segments subequal in length and width, densely covered with setae. Labial palps yellow, 3-segmented, first segment very small, covered with brown setae.
Thorax: pterothorax yellowish brown [17]. Forewings brown [17,23]; dark bands over s, r-m and m-cu; area around cord hyalinized; dark spots absent; m-cu crossvein reaching fork V ( Fig  6A). Hind wings with forks I and V [17]; basal brush present (Fig 6B). Legs yellowish brown; mid legs each with row of spines on distal half of femur and on tibia and tarsus; hind legs each with row of spines on tarsus. Tibial spur formula 0,2,2.
Genitalia: Segment IX annular and short [18]; two small acrotergites present dorsolaterally. Preanal appendages fused with tergum X, forming a hoodlike structure [18], dorsal view ( Fig  7C). Rod-like process above tergum X absent. Inferior appendages slightly enlarged basally [18]; dorsal lobe absent; ventral lobe absent; distal portion narrow [18], tapering posteriorly with apex acuminate (Fig 7D); small spine-like setae absent (Fig 7E). Phallic apparatus curved downward, short, comma shaped, with elongate ventral tip [18] (Fig 7F); phallic spine absent [9]; phallotremal sclerite horseshoe shaped (Fig 7G). Taxonomic remarks: according to Chen [9], Oecetis connata, O. punctipennis and O. iguazu form a small group within Oecetis with the comma shaped phallic apparatus as a grouping character. Oecetis dominguezi can also be included in this group due to its shared comma shaped phallic apparatus. Oecetis connata is easily separated from the other three species by its preanal appendages fused to one another, and also fused to tergum X, forming, according to Flint [18], a single, hood-like structure. No other described species in the Neotropics has this characteristic. Also, O. connata is the only species in the Neotropical region that does not have fork V in the hind wings (Fig 7B). The position of the m-cu crossvein was variable in the examined material.
Genitalia: Segment IX annular and short; one small acrotergite present mesodorsally. Preanal appendages short and rounded, setose. Rod-like process above tergum X absent. Tergum X membranous, undivided in dorsal view [6]; composed of a single elongate lobe, broad basally, digitate apically, abruptly narrowed in dorsal view (Fig 8C). Inferior appendages broad basally, covered with setae [19]; dorsal lobe acute, narrow basally; ventral lobe absent; distal portion narrow, tapering posteriorly, with apex acuminate; forming with dorsal lobe a deep C-shaped incision ( Fig 8D); small spine-like setae absent (Fig 8E). Phallic apparatus curved downward, large, rounded and inflated, extending the length of segment IX ( Fig 8F); phallic spine present, curved helically, counter clockwise (Fig 8G).  Taxonomic remarks: Ulmer's description matches the specimens examined, except for one character: the tibial spur formula on the specimens observed was 0,2,2. No fore tibial spur was present, even in small specimens. Oecetis excisa and O. inconspicua are easily recognized by their large, rounded and inflated phallic apparatus bearing a curved phallic spine. However, the distinction between them is difficult because of the morphological similarities between their genitalia. Flint [23] and Rueda-Martín et al. [6] stated that it is possible that these species may be synonymous. However, in comparing identified material from UMSP (see material examined section) it was possible to find some small differences between the two species. They have a similar tergum X, preanal appendages and phallic apparatus, but the inferior appendages are different.
Oecetis excisa has a narrower dorsal lobe than O. inconspicua. Also, the apex of the inferior appendage seems to be longer in O. excisa, with the result that, in ventral view, it appears to project further beyond the C-shaped basomesal invagination of the appendage. These character differences were easily recognizable when both species were compared using Scanning Electron Microscopy (Fig 9A and 9B). The morphological differences pointed out by Rueda-Martín et al. [6] concerning the shape of tergum X (digitate in O. excisa and trapezoidal in O. inconspicua) and hind wing venation (O. excisa with fork I and O. inconspicua without) were not observed. Both species had fork I and the specimens examined had digitate as well as trapezoidal tergum X.
Thorax: pterothorax yellowish brown dorsally [20]; yellow in lateral and ventral views. Forewings yellowish brown, hyaline [20]; dark bands over cord absent; dark spots on forks, junctions and end of veins [20]; m-cu crossvein reaching Cu or fork V (Fig 10A). Hind wings with forks I and V [20]; basal brush present (Fig 10B). Legs yellow; mid legs each with row of spines on tibia and tarsus; hind legs each with row of spines on tarsus. Tibial spur formula 1,2,2, fore tibial spur very small.
As mentioned above, Oecetis iguazu, O. connata and O. punctipennis form a small group within Oecetis sharing the comma shaped phallic apparatus as one of the most conspicuous diagnostic characters [9]. Because the differences between Oecetis iguazu and Oecetis punctipennis are very subtle, the diagnosis of these two species is frequently difficult.
Both of them have identical preanal appendages, in dorsal and lateral views. Tergum X is also identical in dorsal view. However, in lateral view, O. iguazu has the apex of tergum X slightly acute, while O. punctipennis has a rounded one. However, these differences are seen only when both species are compared side-by-side. The phallic apparatus is comma shaped and, as already mentioned, identical in these three species.
Thus, the best character to distinguish Oecetis iguazu and Oecetis punctipennis is the shape of the inferior appendages, in lateral view (Figs 10D and 13D). Oecetis iguazu has the inferior appendages without lobes or with a smooth hump dorsally. In O. punctipennis, a pronounced dorsal expansion on the inferior appendages is noticeable and the apex is more acute than O. iguazu. These differences are clearly seen with a Scanning Electron Microscope (Fig 9C and  9D).
Oecetis inconspicua (  Taxonomic remarks: Diagnostic characters for this species are discussed in the section of O. excisa. Ross [10] stated that the position of the crossvein m-cu in the forewing of this species is variable. The examined material showed two degrees of this variation: m-cu reaching fork V and m-cu reaching Cu 1a (after branching of Cu 1 into Cu 1a and Cu 1b ) and that could be evidence that Oecetis inconspicua is a complex of species. Zhou et al. [63] and Floyd [64], based on molecular and larval data respectively, stated that the inconspicua group needs revision and there may be more than one species under the same name. Oecetis inconspicua has distribution records from south of Canada to south of Brazil, which is very unusual for caddisflies. This may be another evidence for a complex of species. A thorough examination on the inconspicua group like that made by Blahnik and Holzenthal on the avara group [5] is necessary.
Thorax: pterothorax brown in dorsal view and yellow brownish in lateral and ventral views. Forewings with golden-brown setae [32]; dark bands over cord absent; dark spots on forks, junctions and end of veins [32]; m-cu crossvein reaching Cu 1a (after branching of Cu into Cu 1a and Cu 1b ) (Fig 12A). Hind wings with forks I and V [32]; basal brush present (Fig 12B). Legs yellowish brown, mid and hind legs each with inner row of small spines on tarsus. Tibial spur formula 1,2,2, fore tibial spur very small.
Genitalia: Segment IX annular and short [32]; one acrotergite present, dorsomesally. Preanal appendages short and rounded [32], with setae. Tergum X membranous, deeply divided mesally, forming two lobes, apex with small setae in dorsal view [9,32] (Fig 12C). Inferior appendages broad basally, covered with setae; dorsal lobe broad and rounded [32]; ventral lobe broader than dorsal lobe and rounded; distal portion short and narrow [32], apex rounded ( Fig  12D); small spine-like setae absent (Fig 12E). Phallic apparatus, in lateral view, slightly curved downward, short, straight with ventral elongated tip [32], constricted in middle line ( Fig 12F); phallic spine present [32], straight; phallotremal sclerite U-shaped and a small membranous lobe capped by a cluster of spicules on tip [32] (Fig 12G). Taxonomic remarks: the description provided by Flint matches the specimens examined, although the colors have small differences when comparing dry and alcohol preserved specimens. This is especially notable in the forewing setae. They tend to fall off in alcohol, so it was not possible to see the golden brown setae and the dark setae over the forks, junctions and ends of veins.

Forewing Venation Discussion
It is traditionally accepted that the M vein in Trichoptera branches into M 1+2 and M 3+4 similarly to the Cu 1 vein (presence of fork V) [38]. In this way, the forewing venation in caddisflies is considered to be fairly simple and the variations differing from the cited pattern always constitute simplifications [38]. However, it is possible to observe that other more distal branching events can occur and the order is filled with examples of different degrees of variation in M and Cu 1 branching. But all of them follow the above rule of branching in which M branches into M 1+2 and M 3+4 and sometimes they branch even more.
In some genera, such as Molanna Curtis [67] or Rhyacophila Pictet [68], M 1+2 and M 3+4 also branch into M 1 , M 2 , M 3 and M 4 (presence of forks III and IV). Also, there are intermediary forms, such as taxa in which M 1+2 branches and M 3+4 does not (presence of fork III only), for example, species of the genus Ithytrichia Eaton [69].
Usually, genera in Trichoptera have a well defined hypothesis of vein homology not requiring a discussion. However, two different interpretations of forewing venation have been used in Oecetis with no consensus reached. The two hypothesis are: 1. "M vein branches" (supported by [45,59,70], for example); and 2. "M vein does not branch" (supported by [4,71,72], for example). The first approach postulates that the M vein in the forewing branches into M 1+2 and M 3+4 , while the second postulates that the M vein remains unbranched until the end of the wing.
The unbranched M vein hypothesis was first proposed by McLachlan [4] who used this character as diagnostic for the genus Oecetis. No direct reference to a different hypothesis was made until Betten [59] thoroughly discussed venation hypothesis for many species of caddisflies, including Oecetis. Betten [59] pointed out that the unbranched condition of the M vein in Oecetis was only a misinterpretation, and Ulmer [71] and McLachlan [4] were mistaken in describing the M vein as unbranched.
Both hypotheses have been followed by different authors and no further discussion about their validity has been raised lately. This makes the wings hard to compare due the absence of consensual homology hypothesis among the veins. Therefore, this condition could lead to mistakes when describing new species or proposing phylogenetic hypothesis, for example. For this reason, we argue in favor of a standardized Oecetis wing nomenclature, based on a consistent homology hypothesis. We present evidence that an unbranched M vein constitutes a better homology interpretation.
Observing the positions of the m-cu crossvein in some Oecetis species, such as O. arcada Mosely [73], O. laustra Mosely [73], O. parallela Wells [74] and O. pseudoamazonica Rueda-Martín, Gibon and Molina [6], it is noticeable that the crossvein varies in position along the Cu vein. Sometimes it reaches Cu vein anterior to the fork, other it intersects posterior to the fork, and, in a third way, it intersects the Cu vein at the fork. The last pattern is the most common in Oecetis, as in O. excisa Ulmer [19], and may lead the false impression that M vein branches and Cu does not. This variation in the position of the m-cu crossvein may be seen even within a single species, as pointed by Ross ([10], p. 243) for O. inconspicua: "Position of crossveins forming the cord extremely variable, ranging from a condition in which the three crossveins form an almost straight line to one in which they are far removed and steplike." Comparing the M and Cu 1 branching patterns in Leptoceridae (under the most recent phylogenetic hypothesis [2]), it is noticeable that in all of them the fork V is present, i.e. the Cu 1 vein branches in all cases in Cu 1a and Cu 1b and that applies also to Oecetis. If the "M vein branches" hypothesis is accepted, we would have to assume that the Cu 1 does not branch in Oecetis, what seems very unlikely to happen observing the wing patterns in leptocerids. Yet, it would be possible to assume that the M vein branches into M 1+2 and M 3+4 and the Cu 1 branches into Cu 1a and Cu 1b . If so, we would have to assume that the Cu 2 would be fused to the anal veins (as Betten [59] does) which seems to be very unlikely due to the often clear separation of the anal area in most of the caddisfly families.
At last, it can be considered that if we assume the false vein to be interpreted as Cu 2 , we could use a hypothesis in which Cu 1 branches and the M vein can be interpreted as also branched. However, we consider that the false vein is an artifact of the convoluted membrane, as described by Morse [70]. When observing the specimens, it is possible to notice that the false vein has a distinct texture and it is usually much thinner than any other vein on the wings. It leads us to believe that the false vein is not a "true" vein and that is the reason we exclude this vein from our discussion and, as a consequence, do not considerate this hypothesis with both Cu 1 and M branching.
Considering the evidences listed above, it is possible to indentify which one is the m-cu crossvein in Oecetis. With that in mind, we assume that the hypothesis in which M does not branch, seems to be the most reasonable one.
Standardizing a forewing nomenclature will make all homology hypotheses comparable to each other, as well as to provide a single framework for phylogenetic hypothesis within Leptoceridae and Oecetis.
It is understandable that many other researchers will still support the branched M vein hypothesis. In this way, more accurate studies, such as developmental biology, remain useful in this area to provide stronger evidence in this subject.