An illustrated guide to the identification of the known species of Diatraea Guilding (Lepidoptera, Crambidae, Crambinae) based on genitalia

Abstract The genus Diatraea Guilding is one of the most economically important groups of moths in the Western Hemisphere. The larvae are stem borers that feed on species of Poaceae, or grasses, such as sugarcane, corn, rice, and sorghum, as well as many other native grasses. Interest in this group has risen considerably since sugarcane and other grasses have been utilized and/or investigated as biofuels. This is the first modern study to treat all 41 valid described species. Most type specimens were examined and we provide a checklist with 19 new synonyms. We provide keys for the identification of most species in this genus based on morphology of the male and female genitalia and modern illustrations of male and female genitalia. We also provide an updated table of species distribution by country.


Introduction
The genus Diatraea Guilding is composed of externally similar species, i.e. species cannot be identified using external characters only, and occur in the Western Hemisphere. The type species is Diatraea saccharalis (Fabricius, 1794) (Fig. 1), a major pest of sugarcane. The literature is abundant with studies on the biology of this and other closely related species that are economically important beginning with Guilding (1828). In this paper we consider 41 distinct taxa represented by 41 valid names and 46 synonyms. Some synonyms (e.g., D. busckella Dyar & Heinrich, 1927) at one time were considered valid species or subspecies based on insignificant amounts of variation and/ or locally disparate distributions. Fortunately, the morphology of the genitalia has provided excellent characters for identification for most species. This study treats the entire genus as it is currently circumscribed throughout the Western Hemisphere. We provide a table of the species distributions as is currently known (Suppl. material 1) compiled from Box (1931) and the USNM collection (National Museum of Natural History, Smithsonian Institution, Washington, DC). Absence of a country from this table may not indicate that it does not occur there; it indicates that we have not seen material of that species from that country. During the course of this study we discovered more new species and the potential for cryptic species (e.g. Joyce et al. 2014, Solis et al. 2015, but we decided to publish keys to the identification of described species due to the number of identification requests and workshop requests that were being submitted to MAS (e.g. Solis 2004, Vargas et al. 2013. We provide a key to adults of the Crambinae as modified from Munroe and Solis (1998). The Diatraea diagnosis is from an excellent study of the North American Crambinae by Landry (1995).
Over eighty species names have been associated with Diatraea or related genera since Fabricius described the type species as Phalaena saccharalis in 1794. Early studies by Dyar and Heinrich (1927) and Box (1931) listed 39 and 48 species, respectively. More modern checklists list more, various numbers of species. Bleszynski (1967) lists 55 species, Munroe et al. (1995) lists 57 species, and Nuss et al. (2015) list 58 species. The first overview of Diatraea and related genera was in the 1927 treatment by Dyar & Heinrich. They treated the 22 known species and described 9 new species; they created entirely new species concepts. Several species names were unrecognized, that is, their only reference was a species description and not specimens. They were the first to comparatively use and illustrate the male and female genitalia using pen and ink. The last major overview of Diatraea and related genera was Box (1931). He recognized 48 species, including 10 new species. He illustrated the genitalia with black and white photos when he deemed the pen and ink illustrations deficient. Unfortunately, his photographs were often with insufficient magnification. For the first time, he provided a distribution chart by country for the 48 species. He provided a key to external characters primarily using the frons, forewing color, and venation, although he suggested that genitalia dissections be done whenever possible.

Methods
Type specimens of Diatraea species were studied to confirm identity of species. Approximately 50% of the types are located at the Natural History Museum (BMNH) in London, United Kingdom, and most of the others are located at the National Museum of Natural History (USNM) in Washington, DC, USA. Almost all of the type specimens at the USNM had been previously dissected for the study by Dyar and Heinrich (1927). Dissections of material from the Carnegie Museum of Natural History (CMNH), Pittsburgh, PA, USA, were labeled CMNH and given sequential numbers.
Genitalia preparation for identification (Clarke 1941, Robinson 1976): the abdomen is removed by pushing the abdomen up with forceps. If the metathorax is still attached to the abdomen it should be separated from the abdomen. The male or female abdomen is then placed in a vial with 10% + KOH. The vial is then placed in a beaker with boiling water. The abdomen in KOH is boiled until air bubbles can be seen in the abdomen. Alternatively, the abdomen can be left in cold 10% KOH overnight. The abdomen is then removed from the KOH and placed in water. Then a brush is used to clean the scales from the abdomen, particularly the anal area if it is a male. For males the genitalia can be removed by holding the valvae and uncus with forceps, and then pulling posteriorly at the same time that the abdomen is being held anteriorly with either the brush or the forceps. The male genitalia of Diatraea can be very sclerotized so often staining is not required to see the structures. At this point, when the genitalia is in water, the male can be identified using the key to males below. A pair of forceps can force the valvae apart or a small piece of glass can be placed on the genitalia to flatten it out to be able to see certain structures. For structure recognition in the male genitalia, two views, lateral ( Fig. 1) and flattened (Fig. 2), are given and labeled.
To remove the female genitalia, the abdomen should be cut laterally the entire length and then around the abdomen between segments VI and VII. The female may  have more tissues surrounding the genitalia and must be cleaned with a brush carefully. The corpus bursae varies in length and width, so it is better to examine it when it is still turgid. If slide mounted, care should be taken to not fold, tear or collapse the corpus bursae. The female apparatus is usually membranous, but may not need staining. If staining is required to see structures use a saturated solution of chlorazol black for only a few seconds. The female can then be identified using the key to females below. The genitalia can then be stored in genitalia vials with glycerin if available, or in 70% alcohol that may harden, but will also preserve, the genitalia as vouchers. Two videos are also available to view dissection techniques in great detail (Brown et al. 2009(Brown et al. , 2011. Terminology within the keys and the major structures of the male genitalia are as follows (Figs 1, 2): valva (e) (harpe of Dyar and Heinrich 1927), basal lobe from the costa of the valvae, juxta (anellus of Dyar and Heinrich 1927), gnathos, tegumen, lateral lobe of the tegumen (but see Landry 1995, p. 69, "a pair of extensions posterad from base of ventral margin"), vinculum, and the phallus (aedoeagus of Dyar and Heinrich 1927) that includes a vesica with a cornutus (or cornuti).
The female genitalia (Figs 27,29) consists of the papillae anales or ovipositor, anterior and posterior apophyses, an ostium bursae (genital opening of Dyar and Heinrich 1927), ductus bursae, and corpus bursae; bursae copulatrix is the term used for the ductus bursae + corpus bursae. Associated with the corpus bursae in a few Diatraea species is a sclerotized signum or many signa that may take various forms and may be diagnostic of species. Associated with the ostium bursae are: sterigma (= ostiolar sclerites (Gaskin 1971)), sclerotized structures, sometimes very complex, surrounding the ostium bursae; lamella antevaginalis, the anterior, often the ventral, side, of the sterigma; lamella postvaginalis, the posterior, often dorsal side, of the sterigma. The section of the ductus bursae near the ostium bursae is called the antrum, and a sclerotized structure just below it, or anterior to it, if present, is called a colliculum.

Results
The adapted key below using external and tympanal characters from Munroe and Solis (1998) can aid in identifying a species as Crambinae.

Diagnosis of Diatraea
In the Crambinae Diatraea is morphologically defined by a combination of derived characters including a lack of ocelli on the head (absent or reduced in the externally similar Donacoscaptes and Xubida (B. Landry, pers. comm.)), the presence of pockets with specialized scales on the male second abdominal segment, hair tufts on the male hind tibia, in the male genitalia basal extensions of the tegumen in some or most species (Landry 1995). Landry (1995) also suggested that the shape of the female sterigma with shallow sclerotized, often spinose, depressions on each side of the ostium bursae, may be unique to Diatraea (in contrast to the externally similar Donacoscaptes and Xubida where "the setation of the female segment VIII is concentrated apico-dorsally" and "the female sterigma and segment VIII are sometimes linked by a narrow sclerotized bridge which may be single or double" (Landry 1995)). Another potentially derived structure in the male genitalia could be the lack of muscle attachments in the lateral lobes of the tegumen (Solis and Metz 2011). Note: Described from a series of 7 specimens from Venezuela and Brazil. Box (1931: Plate III) figured two female genitalia dissections (BMNH #141 and #142) that we studied. Figured in this paper is Figure 26d that is BMNH #142 (= Box #3). Box considered the dissected female, BMNH #141, to be the last remaining syntype of Walker's original series. The locality is unknown, but presumed by Box (1931:41-42) to be Venezuela: "The only specimen which we can to-day assert to have been included among the above, is the female type in the British Museum from Venezuela."

Checklist of Diatraea
Note: This appears to be a variable species based on the number of specimens available and barring numerous dissections. Despite differences in size and coloration, the male genitalia are consistent throughout with the same morphology for the uncus and gnathos, lateral process of the tegumen, and costal processes. The females, however, are not consistent. In the impersonatella form, the lamella postvaginalis has the transverse ridges at an angle so that medially they are farther from the ostium bursae than the lateral ends and the membranous area in the middle is wide and widens past the ridges forming an hourglass shape. In the moorella form it is either like the impersonatella form or has the transverse ridges completely absent and the lamella postvaginalis a large roughened patch without a wide membranous area in the middle. The pallidostricta and flavipennella (and some moorella) form females have a much rounder lamella postvaginalis that is often glabrous and the transverse ridges are arcuate. Nomenclaturally, the synonomy for this group of names is also confounded by the lack of single typification for the syntype series and that some of these species are represented by female holotypes among what seems to be variable female genitalia. Dyar, 1911: 205. Type locality: Castro, Parana, Brazil. Syn. n. Diatraea angustellus Dyar, 1911. Schaus, 1922 Mexico.