Trachymolgus purpureus sp. n., an armored snout mite (Acari, Bdellidae) from the Ozark highlands: morphology, development, and key to Trachymolgus Berlese

Abstract Trachymolgus purpureus Fisher & Dowling sp. n. is described from the Ozark highlands of North America. A diversity of imaging techniques are used to illustrate the species including low-temperature scanning electron microscopy (LT-SEM), stereomicrography, compound light micrography, and digitally created line drawings. Developmental stages (larva, nymphs, and adult) and morphology are illustrated and discussed, and terminological corrections are suggested. Trachymolgus recki Gomelauri, 1961 is regarded as being described from tritonymphs. A key to Trachymolgus is presented.


introduction
Bdellidae Dugès, 1834 generally have a striated, unsclerotized integument.Exceptions occur in Cytinae Grandjean, 1938, which comprises three of the most distinctive bdelloid genera.Cyta Heyden, 1826 are common mites known for their stocky bodies, massive chelicerae, and unpaired fifth eye.Rigibdella ignea Tseng, 1978 from Taiwan have sclerotized, striated holodorsal shields (Tseng 1978;Lin and Zhang 2010).The only other bdellids with a sclerotized body are the subject of this study -Trachymolgus Berlese, 1923.Trachymolgus are infrequently collected mites with three described species.Instead of the typical fingerprint-like membranes of other Bdellidae, the integument of Trachymolgus is heavily sclerotized, divided into distinct sclerites, and dark (Fig. 1).The sclerites are foveolate, containing many indentions that create a characteristic reticulated patterning (Fig. 2a).These indentions have multiple small pits at the corners of the foveolae (Fig. 2b).This strange integument, combined with the rarity of collection, has rendered Trachymolgus the most enigmatic bdellid.Berlese (1923) erected Trachymolgus to accommodate the distinctive snout mite Canestrini and Fanzago (1876) named Bdella nigerrima.Trachymolgus niggerimus (Canestrini & Fanzago, 1876) was originally collected in northern Italy (Padova) and subsequently reported from Sicily and Lugano, Switzerland (Thor 1931), Crimea, Ukraine (Vainshtein et al. 1978), Caucasus (Vainshtein et al. 1978), and from rodent nests in Bulgaria (Sosnina et al. 1965).Grandjean (1938) grouped Cyta and Trachymolgus into a new subfamily (Cytinae) based on the number of subcapitular setae and leg trichobothria, presence of pseudotracheae, and normal chelicerae (reduced in Spinibdellinae).Two other species were described later, both known only from their type localities: T. recki Gomelauri, 1961 in Georgia (former U.S.S.R.: Tbilisi and Kashtak) and T. jesusi Mejia-Recamier & Palacios-Vargas, 1999 in Mexico (Jalisco and Chamela).Undetermined species have been reported from a Buddhist temple on Shikoku Island, Japan (Nakamura et al. 2006), from the St. Lawrence Islands National Park, Canada (Smith et al. 1996), and we collected a third from Columbus, Ohio.Here, we describe and illustrate Trachymolgus purpureus sp.n. from the Ozark highlands of North America.
ndjean, 1938, which comprises three of the most distinctive bdelloid genera.Cyta Heyden, 1826 are common mites known for their stocky bodies, massive chelicerae, and unpaired fifth eye.Rigibdella ignea Tseng, 1978 from Taiwan have sclerotized, striated holodorsal shields (Tseng 1978;Lin and Zhang 2010).The only other bdellids with a sclerotized body are the subject of this study -Trachymolgus Berlese, 1923.Trachymolgus are infrequently collected mites with three described species.Instead of the typical fingerprint-like membranes of other Bdellidae, the integument of Trachymolgus is heavily sclerotized, divided into distinct sclerites, and dark (Fig. 1).The sclerites are foveolate, containing many indentions that create a characteristic reticulated patterning (Fig. 2a).These indentions have multiple small pits at the corners of the foveolae (Fig. 2b).This strange integument, combined with the rarity of collection, has rendered Trachymolgus the most enigmatic bdellid.Berlese (1923) erected Trachymolgus to accommodate the distinctive snout mite Canestrini and Fanzago (1876) named Bdella nigerrima.Trachymolgus niggerimus (Canestrini & Fanzago, 1876) was originally collected in northern Italy (Padova) and subsequently reported from Sicily and Lugano, Switzerland (Thor 1931), Crimea, Ukraine (Vainshtein et al. 1978), Caucasus (Vainshtein et al. 1978), and from rodent nests in Bulgaria (Sosnina et al. 1965).Grandjean (1938) grouped Cyta and Trachymolgus into a new subfamily (Cytinae) based on the number of subcapitular setae and leg trichobothria, presence of pseudotracheae, and normal chelicerae (reduced in Spinibdellinae).Two other species were described later, both known only from their type localities: T. recki Gomelauri, 1961 in Georgia (former U.S.S.R.: Tbilisi and Kashtak) and T. jesusi Mejia-Recamier & Palacios-Vargas, 1999 in Mexico (Jalisco and Chamela).Undetermined species have been reported from a Buddhist temple on Shikoku Island, Japan (Nakamura et al. 2006), from the St. Lawrence Islands National Park, Canada (Smith et al. 1996), and we collected a third from Columbus, Ohio.Here, we describe and illustrate Trachymolgus purpureus sp.n. from the Ozark highlands of North America.


Material and methods


Specimens

Mites were collected primarily from leaf litter samples in the Ozark Mountains of Arkansas (U.S.A.)

Specimens
Mites were collected primarily from leaf litter samples in the Ozark Mountains of Arkansas (U.S.A.), specifically Buffalo National River and Devil's Den State Park, and extracted using Berlese-Tullgren funnels.Approximately half of the specimens were slide-mounted with Hoyer's medium (see Krantz and Walter 2009 for preparation), and half are stored in 95% ethanol at -80°C.Slide-mounted specimens were dissected along the frontal plane.Five paratypes are deposited in the Ohio State University Acarology Collection, Columbus, Ohio.Four paratypes each are deposited in the Field Museum of Natural History, Chicago, Ill., and the National Mite Collection, National Museum of Natural History, Smithsonian Institution, located in Beltsville, Md.All other type specimens are deposited in the Acari Collection of the University of Arkansas, Fayetteville, Ark.

tional Riv
r and Devil's Den State Park, and extracted using Berlese-Tullgren funnels.Approximately half of the specimens were slide-mounted with Hoyer's medium (see Krantz and Walter 2009 for preparation), and half are stored in 95% ethanol at -80°C.Slide-mounted specimens were dissected along the frontal plane.Five paratypes are deposited in the Ohio State University Acarology Collection, Columbus, Ohio.Four paratypes each are deposited in the Field Museum of Natural History, Chicago, Ill., and the National Mite Collection, National Museum of Natural History, Smithsonian Institution, located in Beltsville, Md.All other type specimens are deposited in the Acari Collection of the University of Arkansas, Fayetteville, Ark.


Terminology

An effort is made to implement terminology that is broadly applicable and well accepted across acariforms despite conventions

Terminology
An effort is made to implement terminology that is broadly applicable and well accepted across acariforms despite conventions used among bdelloid authors.Thus, two terms have been renamed herein.First, "hypostome" is used by many (Atyeo 1960;van der Schyff et al. 2004;Hernandes and Feres 2006;Krantz and Walter 2009) to refer to the entire subcapitulum, though it more accurately refers only to the portion anterior to the oral opening (Evans 1992;Krantz and Walter 2009).Thus, in addition to the subcapitulum itself, its setae have been renamed herein to ventral subcapitular setae (vs) and dorsal subcapitular setae (ds).

sed among bd
lloid authors.Thus, two terms have been renamed herein.First, "hypostome" is used by many (Atyeo 1960;van der Schyff et al. 2004;Hernandes and Feres 2006;Krantz and Walter 2009) to refer to the entire subcapitulum, though it more accurately refers only to the portion anterior to the oral opening (Evans 1992;Krantz and Walter 2009).Thus, in addition to the subcapitulum itself, its setae have been renamed herein to ventral subcapitular setae (vs) and dorsal subcapitular setae (ds).

Second, the major idiosomal divisions of bdelloids are regularly referred to as the "propodosoma" and "hysterosoma" (Bdellidae: Atyeo 1960, Me Second, the major idiosomal divisions of bdelloids are regularly referred to as the "propodosoma" and "hysterosoma" (Bdellidae: Atyeo 1960, Mejia-Recamier and Palacios-Vargas 1999, Hernandes et al. 2007;Cunaxidae: Meyer and Ryke 1959, Swift 1996, Den Heyer and Castro 2009).However, acariform segmentation is a debated topic with recent views favoring a radically reduced podosoma leaving the major idiosomal divisions (when viewed dorsally) the opisthosoma and aspidosoma (Grandjean 1969;Coineau 1974;Evans 1992;Krantz and Walter 2009).With regard to the posterior portion, hysterosoma is already widely used and accurately refers to the idiosoma posterior to the sejugal furrow (metapodosoma and opisthosoma).Moreover, it is independent of segmentation hypotheses.Therefore, we retain the use of hysterosoma.However, propodosoma is hypothesis dependent.The Grandjean (1969) hypothesis of body organization suggests the propodosoma is highly reduced dorsally, rendering propodosoma inaccurate when referring to the anteriodorsal portion of the idiosoma.Instead, aspidosoma accurately describes this region.Therefore, we abandon the use of propodosoma when referring to the dorsal morphology of acariforms.However, until more evidence exists for segmentation homologies, we also avoid the use of aspidosoma for Bdelloidea.Prodorsum is widely used by acarologists, but is usually poorly defined and represents the opposite extreme from aspidosoma--straying too far from morphological hypotheses.An exception to this is in Oribatida (e.g., Sanders and Norton 2004), where it is often used as synonymous with the aspidosomal shield (=aspis) and is therefore hypothesis dependent.Thus we recommend restricting prodorsum to casual use.Proterosoma refers to the body anterior to the sejugal furrow (propodosoma, aspidosoma, and gnathosoma of Grandjean [1969]), and is a segmentation independent term that is already widely used in acarology.Therefore, we adopt proterosoma here.
ia-Recamier and Palacios-Vargas 1999, Hernandes et al. 2007;Cunaxidae: Meyer and Ryke 1959, Swift 1996, Den Heyer and Castro 2009).However, acariform segmentation is a debated topic with recent views favoring a radically reduced podosoma leaving the major idiosomal divisions (when viewed dorsally) the opisthosoma and aspidosoma (Grandjean 1969;Coineau 1974;Evans 1992;Krantz and Walter 2009).With regard to the posterior portion, hysterosoma is already widely used and accurately refers to the idiosoma posterior to the sejugal furrow (metapodosoma and opisthosoma).Moreover, it is independent of segmentation hypotheses.Therefore, we retain the use of hysterosoma.However, propodosoma is hypothesis dependent.The Grandjean (1969) hypothesis of body organization suggests the propodosoma is highly reduced dorsally, rendering propodosoma inaccurate when referring to the anteriodorsal portion of the idiosoma.Instead, aspidosoma accurately describes this region.Therefore, we abandon the use of propodosoma when referring to the dorsal morphology of acariforms.However, until more evidence exists for segmentation homologies, we also avoid the use of aspidosoma for Bdelloidea.Prodorsum is widely used by acarologists, but is usually poorly defined and represents the opposite extreme from aspidosoma--straying too far from morphological hypotheses.An exception to this is in Oribatida (e.g., Sanders and Norton 2004), where it is often used as synonymous with the aspidosomal shield (=aspis) and is therefore hypothesis dependent.Thus we recommend restricting prodorsum to casual use.Proterosoma refers to the body anterior to the sejugal furrow (propodosoma, aspidosoma, and gnathosoma of Grandjean [1969]), and is a segmentation independent term that is already widely used in acarology.Therefore, we adopt proterosoma here.

With regard to hysterosomal setal notation, we follow the chaetotaxic system of Grandjean (1939Grandjean ( , 1947) ) that has been widely adopted (e.g., van With regard to hysterosomal setal notation, we follow the chaetotaxic system of Grandjean (1939Grandjean ( , 1947) ) that has been widely adopted (e.g., van der Hammen 1970;Lindquist 1976Lindquist , 1977;;Kethley 1990).Proterosomal setal notation in this system is currently problematic.Generally, we do not recommend the use of mixed approaches in terminology, especially given our attempts to utilize broadly applicable notations.However, recent suggestions in the proterosomal setal notation of bdelloids have rendered this system unreliable.In the Grandjean system, proterosomal setae are termed internal/external verticals (vi and ve) and internal/external scapulars (sci and sce).In Bdelloidea, sci are always external to sce, leaving the internal/external designations unintuitive.Den Heyer and Castro (2008a) noted this and proposed simply switching the terms when referring to bdelloids so that sce and sci are descriptive, which they have published since (Den Heyer and Castro 2008b, c;Den Heyer 2011).As a result, the literature now has sce and sci referring to both inner and outer scapulars.The homology of these setae with respect to other mites is not known.Therefore, until more evidence is provided for homology, we resort to a modified version of Atyeo (1960) when referring to proterosomal setae, which unambiguously relies on position: anterior and posterior trichobothria (at and pt), and lateral and median proterosomal setae (lps and mps).
er Hammen 1970;Lindquist 1976Lindquist , 1977;;Kethley 1990).Proterosomal setal notation in this system is currently problematic.Generally, we do not recommend the use of mixed approaches in terminology, especially given our attempts to utilize broadly applicable notations.However, recent suggestions in the proterosomal setal notation of bdelloids have rendered this system unreliable.In the Grandjean system, proterosomal setae are termed internal/external verticals (vi and ve) and internal/external scapulars (sci and sce).In Bdelloidea, sci are always external to sce, leaving the internal/external designations unintuitive.Den Heyer and Castro (2008a) noted this and proposed simply switching the terms when referring to bdelloids so that sce and sci are descriptive, which they have published since (Den Heyer and Castro 2008b, c;Den Heyer 2011).As a result, the literature now has sce and sci referring to both inner and outer scapulars.The homology of these setae with respect to other mites is not known.Therefore, until more evidence is provided for homology, we resort to a modified version of Atyeo (1960) when referring to proterosomal setae, which unambiguously relies on position: anterior and posterior trichobothria (at and pt), and lateral and median proterosomal setae (lps and mps).

Leg chaetotaxy follows Grandjean's system as reviewed by Norton (1977).However, leg chaetotaxy is poorly studied among Eupodina, and only distal tarsal setae are Leg chaetotaxy follows Grandjean's system as reviewed by Norton (1977).However, leg chaetotaxy is poorly studied among Eupodina, and only distal tarsal setae are denoted presently, which has been adopted by other eupodine authors (e.g., Jesionowska 2010).Nevertheless, we believe Grandjean's system can be employed with other leg setae, and will readdress this in a more detailed forthcoming study.
denoted presently, which has been adopted by other eupodine authors (e.g., Jesionowska 2010).Nevertheless, we believe Grandjean's system can be employed with other leg setae, and will readdress this in a more detailed forthcoming study.


Images

Most species descriptions include only a few image types; line drawings are most common in acarology (e.g., Mejia-Recamier and Palacios-Vargas 1999;Den He

Images
Most species descriptions include only a few image types; line drawings are most common in acarology (e.g., Mejia-Recamier and Palacios-Vargas 1999;Den Heyer and Castro 2008a, b, c;Łochyńska 2008).We believe image diversity increases accuracy, accessibility, and attractiveness of taxonomic products, and have therefore included many images and a range of imaging types.Other acarologists have also begun including image diversity in taxonomic works (e.g.Mąkol 2010;Wohltmann 2010;Pešić et al. 2011).

r and C
stro 2008a, b, c;Łochyńska 2008).We believe image diversity increases accuracy, accessibility, and attractiveness of taxonomic products, and have therefore included many images and a range of imaging types.Other acarologists have also begun including image diversity in taxonomic works (e.g.Mąkol 2010;Wohltmann 2010;Pešić et al. 2011).

Line drawings were created digitally with Adobe Illustrator CS5 and a Wacom Cintiq 21UX tablet using procedures outlined in Fisher and Dowling (2010).Compound light m Line drawings were created digitally with Adobe Illustrator CS5 and a Wacom Cintiq 21UX tablet using procedures outlined in Fisher and Dowling (2010).Compound light micrographs were taken with a Leica DFC300 FX camera and a Leica DM2500 DIC light microscope.Stereomicrographs were taken with a Leica MZ 16 stereoscope and a Leica DFC 290 camera.Photographs were enhanced using Adobe® Photoshop CS4.
crographs were taken with a Leica DFC300 FX camera and a Leica DM2500 DIC light microscope.Stereomicrographs were taken with a Leica MZ 16 stereoscope and a Leica DFC 290 camera.Photographs were enhanced using Adobe® Photoshop CS4.

Low-temperature scanning electron micrographs (LT-SEM) were made using an S-4700 field emission scanning electron microscope (Hitachi High Technologies America, Inc., Pl Low-temperature scanning electron micrographs (LT-SEM) were made using an S-4700 field emission scanning electron microscope (Hitachi High Technologies America, Inc., Pleasanton, Calif.) equipped with a Quorum CryoPrep PP2000 (Energy Bean Sciences, East Grandby, Conn.) cryotransfer system.To prepare specimens, mites were placed on 12 mm diameter ultra smooth carbon double sided adhesive tabs (Electron Microscopy Sciences, Hatfield, PA) which were adhered to flat speci-men holders consisting of 16x30mm copper plates that were tacked on the edges to the tabs with a small dot of Tissue Tek (OCT Compound, Ted Pella, Inc., Redding, Calif.), which acted as the cyro-adhesive upon freezing.The samples were frozen conductively, in a Styrofoam box, by placing the plates on the surface of a pre-cooled (-96°C) brass bar whose lower half was submerged in liquid nitrogen (LN2).After 20-30s, the holders containing the frozen samples were transferred to a LN2 Dewar for future use or cryotransferred under vacuum to the cold stage in the pre-chamber of the cryotransfer system.Removal of any surface contamination (condensed water vapor) took place in the cryotransfer system by etching the frozen specimens for 10-15 min by raising the temperature of the stage to -90°C.Following etching, the temperature was lowered below -130°C, and a magnetron sputter head equipped with a platinum target, was used to coat the specimens with a very fine layer of platinum.The specimens were transferred to a pre-cooled (-130°C) cryostage in the SEM for observation.An accelerating voltage of 5kV was used to view the specimens.Images were captured using a 4pi Analysis system (Durham, N.C.).Images were sized and placed together into figures using Adobe® Photoshop 7.0 and CS4.  2).Like T. jesusi, the integument is dark purple, whereas T. nigerrimus was described as black.Like T. jesusi and T. nigerrimus, there are two teeth on the fixed cheliceral digit.Like T. nigerrimus, T. purpureus has one tooth on the movable digit (T.jesusi have three) and a serrated edge proximal to the tooth (undescribed in other species).All stages have two pairs of eyes, unlike the larva, proto-and deutonymphs of T. jesusi, which lack eyes (tritonymphs and adults have two pairs).Trachymolgus purpureus pedipalpal basi-and telofemora are only fused dorsally.Trachymolgus jesusi pedipalp femora are completely fused, whereas T. nigerrimus are completely divided.Trachymolgus purpureus, like other Trachymolgus, have undivided femora on legs I-II (femora III-IV are divided).All other Bdellidae have divided femora on all legs.T. jesusi is the only bdellid reported to have undivided femora on legs II and III.The ontogeny of T. purpureus differs markedly from that described for T. jesusi, the only other species where ontogeny was investigated.Finally, there are many chaetotaxic differences on the appendages and venter between T. purpureus and T. jesusi.Most chaetotaxy of T. nigerrimus remain to be investigated.See Remarks for discussion of T. recki.
asanton, Calif.) equipped with a Quorum CryoPrep PP2000 (Energy Bean Sciences, East Grandby, Conn.) cryotransfer system.To prepare specimens, mites were placed on 12 mm diameter ultra smooth carbon double sided adhesive tabs (Electron Microscopy Sciences, Hatfield, PA) which were adhered to flat speci-men holders consisting of 16x30mm copper plates that were tacked on the edges to the tabs with a small dot of Tissue Tek (OCT Compound, Ted Pella, Inc., Redding, Calif.), which acted as the cyro-adhesive upon freezing.The samples were frozen conductively, in a Styrofoam box, by placing the plates on the surface of a pre-cooled (-96°C) brass bar whose lower half was submerged in liquid nitrogen (LN2).After 20-30s, the holders containing the frozen samples were transferred to a LN2 Dewar for future use or cryotransferred under vacuum to the cold stage in the pre-chamber of the cryotransfer system.Removal of any surface contamination (condensed water vapor) took place in the cryotransfer system by etching the frozen specimens for 10-15 min by raising the temperature of the stage to -90°C.Following etching, the temperature was lowered below -130°C, and a magnetron sputter head equipped with a platinum target, was used to coat the specimens with a very fine layer of platinum.The specimens were transferred to a pre-cooled (-130°C) cryostage in the SEM for observation.An accelerating voltage of 5kV was used to view the specimens.Images were captured using a 4pi Analysis system (Durham, N.C.).Images were sized and placed together into figures using Adobe® Photoshop 7.0 and CS4.  2).Like T. jesusi, the integument is dark purple, whereas T. nigerrimus was described as black.Like T. jesusi and T. nigerrimus, there are two teeth on the fixed cheliceral digit Like T. nigerrimus, T. purpureus has one tooth on the movable digit (T.jesusi have three) and a serrated edge proximal to the tooth (undescribed in other species).All stages have two pairs of eyes, unlike the larva, proto-and deutonymphs of T. jesusi, which lack eyes (tritonymphs and adults have two pairs).Trachymolgus purpureus pedipalpal basi-and telofemora are only fused dorsally.Trachymolgus jesusi pedipalp femora are completely fused, whereas T. nigerrimus are completely divided.Trachymolgus purpureus, like other Trachymolgus, have undivided femora on legs I-II (femora III-IV are divided).All other Bdellidae have divided femora on all legs.T. jesusi is the only bdellid reported to have undivided femora on legs II and III.The ontogeny of T. purpureus differs markedly from that described for T. jesusi, the only other species where ontogeny was investigated.Finally, there are many chaetotaxic differences on the appendages and venter between T. purpureus and T. jesusi.Most chaetotaxy of T. nigerrimus remain to be investigated.See Remarks for discussion of T. recki.

Imago description.Females and males similar, except for genitalia, size, and chaetotaxic differences noted in Table 1.Color dark purple; occasionally immatures and adults were collected from the Buffalo National River (Arkansas) with an internally green coloration, which rendered the normally purple mite teal; teal specimens returned to purple after a few days in 95% ethanol, and were indistinguishable from normal specimens when slide-mounted (we also collected Penthaleus, a normally black to dark blue mite, from the same habitat exhibiting green internal coloration).Integument divided into heavily armored sclerites with foveolate sculpturing (Fig. 2a).The foveolate indentions (foveolae) are bordered with pits (Fig. 2b).Measurements in Tables 2-5.table 1. Leg chaetotaxy.Female (♀), male (♂), tritonymph (3N), deutonymph (2N), protonymph (1N), larva (L), pedipalp (Pp), legs I-IV (I-IV).Numbers represent setal counts for barbulate setae (undesignated), solenidia (s), and trichobothria (tr).Male setal counts that are not different from the female are denoted with an asterisk (*).Absent characters are denoted with a dash (-).Fused segments are denoted by fused cells.Numbers in parentheses denote occurrences of two solenidia on tarsus II in some specimens.Dorsal idiosoma (Fig. 3).Idiosoma dorsally armored with two large tergites: proterosomal and hysterosomal shields (see Terminology).Dorsal membrane (between proterosomal and hysterosomal shields and between dorsal and lateral shields) striated and accompanied with raised bumps similar in size to the foveolate indentions (Figs 4,.Proterosoma ending anteriorly in a crenulated, tri-lobed shelf (crown) covering the stigmata.Two pairs of eyes present.Two pairs of minutely barbulate trichobothria: anteri Imago description.Females and males similar, except for genitalia, size, and chaetotaxic differences noted in Table 1.Color dark purple; occasionally immatures and adults were collected from the Buffalo National River (Arkansas) with an internally green coloration, which rendered the normally purple mite teal; teal specimens returned to purple after a few days in 95% ethanol, and were indistinguishable from normal specimens when slide-mounted (we also collected Penthaleus, a normally black to dark blue mite, from the same habitat exhibiting green internal coloration).Integument divided into heavily armored sclerites with foveolate sculpturing (Fig. 2a).The foveolate indentions (foveolae) are bordered with pits (Fig. 2b).Measurements in Tables 2-5.table 1. Leg chaetotaxy.Female (♀), male (♂), tritonymph (3N), deutonymph (2N), protonymph (1N), larva (L), pedipalp (Pp), legs I-IV (I-IV).Numbers represent setal counts for barbulate setae (undesignated), solenidia (s), and trichobothria (tr).Male setal counts that are not different from the female are denoted with an asterisk (*).Absent characters are denoted with a dash (-).Fused segments are denoted by fused cells.Numbers in parentheses denote occurrences of two solenidia on tarsus II in some specimens.Dorsal idiosoma (Fig. 3).Idiosoma dorsally armored with two large tergites: proterosomal and hysterosomal shields (see Terminology).Dorsal membrane (between proterosomal and hysterosomal shields and between dorsal and lateral shields) striated and accompanied with raised bumps similar in size to the foveolate indentions (Figs 4,.Proterosoma ending anteriorly in a crenulated, tri-lobed shelf (crown) covering the stigmata.Two pairs of eyes present.Two pairs of minutely barbulate trichobothria: anterior (at) and posterior trichobothria (pt).Barbules are difficult to discern with light microscopy (Fig. 5c).Two pairs of barbulate setae are present: lateral proterosomal (lps) and median proterosomal setae (mps).Setae lps are oriented dorsomedially and lay in a groove posterior to the first pair of eyes (Fig. 5d); mps are the longest barbulate setae.table 2. Body measurements.Stage (St), female (♀), male (♂), tritonymph (3N), deutonymph (2N), protonymph (1N), and larva (L), mean (M), standard deviation (S), range (R), number examined (n), idiosomal length (Idi L) and width (Idi W), and lengths of proterosomal shield (Pro), hysterosomal shield (Hys), lateral shield (Lat), subcapitulum (Sub), chelicerae (Chel), pedipalps (Ped), anal shield (Ana), genital shield (Gen), and legs I-IV (L I-IV).Absent characters are denoted with a dash (-).All measurements in micrometers.

Stage
sally, and longitudinal striation distally (Fig. 11c); with two dorsal barbulate setae.Fixed digit ending in a hook, and with two teeth (one small and one large and triangular); movable digit with one small tooth and a serrated edge proximal to the tooth (Fig. 10d).Tables 1-6.Developmental stages are illustrated in Figures 13-16.Like other mites, developmental stages can be easily recognized by leg number (larvae have six legs) and genital development (Fig. 17).Chaetotaxic differences and femoral divisions are also helpful (Tables 1, 6).All immature stages appear soft bodied (despite dorsal sclerites) and vary in color from light green or purple to yellowish-white (Fig. 12).

Immatures description. Measurements and chaetotaxy of immatures are given in
Due to the unique armored morphology of Trachymolgus, other interesting developmental changes are present.These are discussed below.
Dorsal sclerites and setae.None of the immature stages of T. purpureus have complete dorsal shields as seen in the adult.This is unlike T. jesusi, which was described as having an armored tritonymph and unsclerotized proto-and deutonymphs.In T. purpureus, all stages have dorsal sclerites.Shield sculpturing is underdeveloped in the larva with foveolate indentions absent but pits present (Fig. 13); protonymphs also lack foveolate indentions, but the pits are more organized, reminiscent of the indentions (Fig. 14); deutonymphs begin to develop foveolate indentions (Fig. 15), which are nearly complete in the tritonymph (Fig. 16).The proterosomal shield of the larva does not encompass the posterior pair of eyes, and the anterior crown is not developed, leaving the gnathosomal membrane appearing as a collar.The protonymph has a welldeveloped proterosomal shield that encompasses all eyes and has a complete crown.Hysterosomal shield of the larva only encompasses c 1 and d 1 ; nymphal stages also encompass e 1 .Small sclerotized regions containing pits, but not foveolate indentions, are present around c 2 in the deuto-and tritonymphs.A posterior shield encompassing h 1 , contiguous with the hysterosomal shield in adults, is present in nymphal stages, but not larvae.Larvae completely lack f 2 .

Shield
sculpturing is underdeveloped in the larva with foveolate indentions absent but pits present (Fig. 13); protonymphs also lack foveolat indentions, but the pits are more organized, reminiscent of the indentions (Fig. 14); deutonymphs begin to develop foveolate indentions (Fig. 15), which are nearly complete in the tritonymph (Fig. 16).The proterosomal shield of the larva does not encompass the posterior pair of eyes, and the anterior crown is not developed, leaving the gnathosomal membrane appearing as a collar.The protonymph has a welldeveloped proterosomal shield that encompasses all eyes and has a complete crown.Hysterosomal shield of the larva only encompasses c 1 and d 1 ; nymphal stages also encompass e 1 .Small sclerotized regions containing pits, but not foveolate indentions, are present around c 2 in the deuto-and tritonymphs.A posterior shield encompassing h 1 , contiguous with the hysterosomal shield in adults, is present in nymphal stages, but not larvae.Larvae completely lack f 2 .

Lateral shields.Lateral shields are present in deuto-and tritonymphs (Figs 15, 16), but do not encompass h 2 or ih, as in adults (Fig. 3).Furthermore, in addition to lateral shields, larvae lack h 2 .Lyrifissure ih was not found in any immature stage.

Pseudotracheae.As described for T. jesusi, pseudotracheae are lacking in the larva and protonymph, but are well-developed in the deutonymph (Fig. 17).

Membranes.As discussed above, adult Lateral shields.Lateral shields are present in deuto-and tritonymphs (Figs 15, 16), but do not encompass h 2 or ih, as in adults (Fig. 3).Furthermore, in addition to lateral shields, larvae lack h 2 .Lyrifissure ih was not found in any immature stage.
Pseudotracheae.As described for T. jesusi, pseudotracheae are lacking in the larva and protonymph, but are well-developed in the deutonymph (Fig. 17).
Membranes.As discussed above, adult T. purpureus striations are accompanied with bumps (Fig. 5b), unlike other bdellid membranes that exhibit fingerprint-like striations.However, larvae and protonymphs lack bumps and have typical fingerprintlike striations (Figs 13,14).Membrane bumps begin to develop on the deutonymphal dorsum (Fig. 15), and are well developed in the tritonymph (Fig. 16).All stages have normal, fingerprint-like striations on the venter.
T. purpureus striations are accompanied with bumps (Fig. 5b), unlike other bdellid membranes that exhibit fingerprint-like striations.However, larvae and protonymphs lack bumps and have typical fingerprintlike striations (Figs 13,14).Membrane bumps begin o develop on the deutonymphal dorsum (Fig. 15), and are well developed in the tritonymph (Fig. 16).All stages have normal, fingerprint-like striations o the venter.

Remarks.In the early 1980s, Trachymolgus was collected by Cal Welbourn on a rocky bluff in the Buffalo National River (Arkansas).John Kethley recollected three specimens from the same bluff a few years later.Another specimen (one female) was collected by Evert Lindquist in the St. Lawrence Islands National Park, Canada (Smith et al. 1996), but the specimen has since been lost.Since then, we have col-lected this seemingly restricted, rare mite fr Remarks.In the early 1980s, Trachymolgus was collected by Cal Welbourn on a rocky bluff in the Buffalo National River (Arkansas).John Kethley recollected three specimens from the same bluff a few years later.Another specimen (one female) was collected by Evert Lindquist in the St. Lawrence Islands National Park, Canada (Smith et al. 1996), but the specimen has since been lost.Since then, we have col-lected this seemingly restricted, rare mite from a wide variety of microhabitats including litter, talus, rock outcrops and bluffs, moss, cedar stands, hardwood stands, and in both wet and dry conditions.Furthermore, the range is potentially not restricted to the Ozark highlands.Amusingly, we collected one tritonymph and one adult from leaf litter less than 200ft from where the Ohio State University Summer Acarology Course is taught (downtown Columbus, Ohio).Morphologically, these specimens cannot yet be distinguished from T. purpureus, potentially extending the range to eastern U.S. We were able to extract DNA from one of these specimens and will publish our findings later.
m a wide variety of microhabitats including litter, talus, rock outcrops and bluffs, moss, cedar stands, hardwood stands, and in both wet and dry conditions.Furthermore, the range is potentially not restricted to the Ozark highlands.Amusingly, we collected one tritonymph and one adult from leaf litter less than 200ft from where the Ohio State University Summer Acarology Course is taught (downtown Columbus, Ohio).Morphologically, these specimens cannot yet be distinguished from T. purpureus, potentially extending the range to eastern U.S. We were able to extract DNA from one of these specimens and will publish our findings later.

Biogeography.The known distribution of North American Trachymolgus is Mexico (T.jesusi), Ozark highlands (T.purpureus), central Ohio (undet.species), and the northern Appalachian mountains (undet.species).Other groups have a similar distribution, and the biogeographic affinity between the Ozark and Appalachian mountains, and between Mexico and the eastern U.S. has been well documented.Examples include mosses (Crum 1952;Redfearn 1986), higher plants (Br Biogeography.The known distribution of North American Trachymolgus is Mexico (T.jesusi), Ozark highlands (T.purpureus), central Ohio (undet.species), and the northern Appalachian mountains (undet.species).Other groups have a similar distribution, and the biogeographic affinity between the Ozark and Appalachian mountains, and between Mexico and the eastern U.S. has been well documented.Examples include mosses (Crum 1952;Redfearn 1986), higher plants (Braun 1955;Dressler 1954;Miranda and Sharp 1950;Watson 1891), fungi (Miranda and Sharp 1950;Sharp 1948), snakes, flying squirrels, and plethodontid salamanders (see Martin and Harrel 1957).Recently, a mite was implicated as sharing this affinity (Skvarla et al. 2011).It is tempting consider T. purpureus as representative of these biogeographic events, but much more sampling is necessary before this conclusion can be justified.
un 1955;Dressler 1954;Miranda and Sharp 1950;Watson 1891), fungi (Miranda and Sharp 1950;Sharp 1948), snakes, flying squirrels, and plethodontid salamanders (see Martin and Harrel 1957).Recently, a mite was implicated as sharing this affinity (Skvarla et al. 2011).It is tempting consider T. purpureus as representative of these biogeographic events, but much more sampling is necessary before this conclusion can be justified.

Temperature tolerance.T. purpureus seems to have extremely high temperature tolerances.They were found crawling on rock surfaces in direct sunlight during a drought in the hottest and driest time of year (August), and were collected near the surface during the winter.When preparing live specimens for LT-SEM, mites are set atop a metal bar that is subjected to liquid nitrogen fumes which freezes them midstride for imaging.When T. purpureus was subjected to liqui Temperature tolerance.T. purpureus seems to have extremely high temperature tolerances.They were found crawling on rock surfaces in direct sunlight during a drought in the hottest and driest time of year (August), and were collected near the surface during the winter.When preparing live specimens for LT-SEM, mites are set atop a metal bar that is subjected to liquid nitrogen fumes which freezes them midstride for imaging.When T. purpureus was subjected to liquid nitrogen temperatures however, they would simply run, curl their legs, and roll off the plate (see Fig. 18a).This made imaging live specimens very difficult.
nitrogen temperatures however, they would simply run, curl their legs, and roll off the plate (see Fig. 18a).This made imaging live specimens very difficult.

Silk production.LT-SEM imaging illuminated another behavioral characteristic of T. purpureus.Though other bdellids have been known to orally produce silk table 6. Recognizing life stages.Female (♀), male (♂), tritonymph (3N), deutonymph (2N), protonymph (1N), and larva (L).Numbers represent setal counts; those in parentheses denote counts when extra setae are present.Absent characters are denoted with a dash (-). to tether prey (Alberti 1973;Krantz and Walter 2009), Silk production.LT-SEM imaging illuminated another behavioral characteristic of T. purpureus.Though other bdellids have been known to orally produce silk table 6. Recognizing life stages.Female (♀), male (♂), tritonymph (3N), deutonymph (2N), protonymph (1N), and larva (L).Numbers represent setal counts; those in parentheses denote counts when extra setae are present.Absent characters are denoted with a dash (-). to tether prey (Alberti 1973;Krantz and Walter 2009), silk production in Trachymolgus was not previously known.When subjected to liquid nitrogen, T. purpureus would charge its gnathosoma with silk, making investigation of chelae impossible on living specimens (Fig. 11d).One specimen tethered itself to the plate before rolling off (Fig. 18), potentially using silk as a dragline as has been described in many spiders.Feeding behavior.We observed a tritonymph of T. purpureus feeding on a small mite approximately 200-250m long.Unfortunately, the prey could not be retrieved for identification.The tritonymph fed with prey elevated from the ground.There seemed to be a droplet surrounding the bite site, interpreted here as silk seen in Figures 11d.We silk production in Trachymolgus was not previously known.When subjected to liquid nitrogen, T. purpureus would charge its gnathosoma with silk, making investigation of chelae impossible on living specimens (Fig. 11d).One specimen tethered itself to the plate before rolling off (Fig. 18), potentially using silk as a dragline as has been described in many spiders.Feeding behavior.We observed a tritonymph of T. purpureus feeding on a small mite approximately 200-250m long.Unfortunately, the prey could not be retrieved for identification.The tritonymph fed with prey elevated from the ground.There seemed to be a droplet surrounding the bite site, interpreted here as silk seen in Figures 11d.We


Stage

taxonomy

Trachymolgus purpureusFisher & Dowling sp.n. urn:lsid:zoobank.org:act:E0FAE922-2B81-4FB5-8D50-CE0F93517CD2http://species-id.net/wiki/Trachymolgus_purpureusDiagnosis.Trachymolgus purpureus sp.n. is heavily armored with distinctive integument characteristic of Trachymolgus (Figs 1-


FiguresFigure 1 .
1
Figures

Figure 2 .
Figure 2 .
2
Figure 2. Integument of Trachymolgus purpureus sp.n.LT-SEM.A Lateral view of e Figure 2. Integument of Trachymolgus purpureus sp.n.LT-SEM.A Lateral view of eyes and pt showing foveolate indentions B Magnified view of foveolae and pits.

Figure 3 .
Figure 3. Dorsum of Trachymolgus purpureus sp.n.Lateral plates removed and displayed laterally.See text for abbreviations.

Figure 5 .
Figure 5. Morphological aspects of Trachymolgus purpureus sp.n.A LT-SEM of lateral view B enlargement of lateral membrane showing striations accompanied with bumps C Base of pt showing minute barbules D Left lateral view of lps in a groove above anterior eye, pt removed e Compound light micrograph of proterosomal shield with apodemes in focus, appearing as four dark spots F Line drawing of proterosomal shield showing apodemes.
sp.n.A LT-SEM of lateral view B enlargement of lateral membrane showing striations accompanied with bumps C Base of pt showing minute barbules D Left lateral view of lps in a groove above anterior eye, pt removed e Compound light micrograph of proterosomal shield with apodemes in focus, appearing as four dark spots F Line drawing of proterosomal shield showing apodemes.


Figure 6 .
6
Figure 6.Ventral aspects of Trachymolgus purpureus sp.n.A Venter showing pseudotracheae, legs removed B Stereomicrograph showing extruded ovipositor.


Figure 7 .
7
Figure 7. Legs of Trachymolgus purpureus sp.n.Laterodorsal view of distal podomeres.Fastigials (ft), iterals (it), prorals (p), tectals (tc), unguinals (u), and famulus (ε).Stippling denotes unstriated membrane.Asterisk (*) denotes solenidion found in only a few specimens.


Figure 8 .
8
Figure 8. Coxal fields of Trachymolgus purpureus sp.n.A Compound light micrograph of venter showing apodemes on coxae II & III B L

Figure 8 .
Figure 8. Coxal fields of Trachymolgus purpureus sp.n.A Compound light micrograph of venter showing apodemes on coxae II & III B Line drawing with emphasis on apodemes.

Figure 11 .
Figure 11.Gnathosoma of Trachymolgus purpureus sp.n.LT-SEM.A Ventral view of gnathosoma showing subcapitular sculpturing B Dorsal view of subcapitulum showing position of oral opening (o) C Lateral view of gnathosoma showing cheliceral sculpturing D Magnified view of distal gnathosoma showing lateral lips and silk charge e Dorsolateral view of removed pedipalp showing striations F Ventrodistal view of right pedipalp showing papillated striations, finely barbulate ves, and solenidion.

Figure 18 .
Figure 18.LT-SEM of silk production in Trachymolgus purpureus sp.n.A Lateral habitus showing frozen mite with legs curled, attached to LT-SEM plate with silk tether B Enlargement of anterior gnathosoma and silk tether.