Ichnos, 8:165-175, 2001 Copyright ? 2001 Taylor & Francis 1042-0940/01 S12.00-i-.00 f^. The Insect Trace Fossil Tonganoxichnus irom the Middle Pennsylvanian of Indiana: Paleobiologic and Paleoenvironmental Implications Maria G. M?ngano Conicet, Instituto Superior de Correlaci?n Geol?gica, 4000 San Miguel de Tucumdn, Argentina Conrad C. Labandeira Department of Paleobiology, National Museum of Natural History, Smithsonian Institution, Washington DC, 20560, USA, and Department of Entomology, University of Maryland, College Park, Maryland, 20742, USA Erik P. Kvale Indiana Geological Survey, Indiana University, Bloomington, Indiana 47405, USA Luis A. Buatois Conicet, Instituto Superior de Correlaci?n Geol?gica, 4000 San Miguel de Tucumdn, Argentina The ichnogenus Tonganoxichnus, produced by one or more monuran insect taxa, is now recorded from the Middle Pennsyl- vanian Mansf?eld Formation of Indiana. Tonganoxichnus is a rest- ing trace that has three important implications. First, it represents a recurrent behavioral pattern in Upper Carboniferous to Lower Permian marginal marine environments of North America. Sec- ond, it provides finely resolved anatomical information for axial and appendicular body structures and behaviors that are difficult to determine from body-fossil material alone. Third, integrated sedimentologic and ichnologic observations indicate that the Tonganoxichnus assemblage, inclusive of other ichnotaxa, is com- mon in tidal rhythmites that were developed under freshwater conditions, probably in the innermost part of estuarine systems, close to or at the fluvioestuarine transition. Keywords Carboniferous, Ichnotaxa, Indiana, Insecta, Monura, Pennsylvanian, Tonganoxichnus INTRODUCTION The ichnogenus Tonganoxichnus was proposed by M?ngano et al. (1997) to name superbly preserved insect sedimentary structures exhibiting unique anterior and posterior sets of mor- Address correspondence to Maria G. Mangano, Casilla de Correo 1 (CC), 4000 San Miguel de Tucum?n, Argentina. E-mail: ichnolog? infovia.com.ar phologic features, present in Upper Pennsylvanian (Virgilian) tidal rhythmites of Buildex Quarry in eastern Kansas. This trace fossil is one of the few from from the Late Carboniferous attributable to a specific insect clade, even though the ichno- fossil record from the Paleozoic is exceedingly dominated by evidence of plant-arthropod associations (Labandeira, 1998a). Other Paleozoic insect traces are probably trackways (e.g., Holub and Kozur, 1981; Walker, 1985; Hunt et al., 1993). After the end-Permian mass extinction, much of the taxonomically distinctive Paleozoic Insect Fauna was succeeded by mostly the hemipteroid and holometabolous lineages of the early Meso- zoic (Labandeira and Sepkoski, 1993; Jarzembowski and Ross, 1993). This diversification also is reflected by a significant increase in abundance and diversity of insect sedimentary structures. From the ichnofossil record, this rise is documented initially by soil-associated coleopteran nests and tracks of var- ied insects in eolian environments and culminated in the domi- ciles of social hymenopterans and isopterans in Cretaceous paleosols (calichnia sensu Genise and Bown, 1994; Ekdale and Picard, 1985; Buatois et al., 1998; Labandeira, 1998b). This increasingly better Mesozoic ichnologic record is attribut- able not only to the evolution of insect nesting behavior in sedimentary substrates, but also to an increased preservational potential from the constructional nature of the nesting struc- tures that includes their intermixing with mucilaginous 165 166 M. G. MANGANO ET AL. ex?dales, excrement, or ambient organic products (Genise and Bown, 1994). Tonganoxichnus buildexensis, by contrast, is a rare Paleo- zoic sediment trace that combines significant anatomical infor- mation of the fabricator with a narrow circumscription of its habitat of occurrence. It is characterized by an anterior region with a frontal pair of right-and-left maxillary palp impressions, a median head impression, and three pairs of bilaterally posi- tioned thoracic appendage impressions. The posterior abdomi- nal region is commonly segmented and displays numerous del- icate abdominal appendage impressions, including a typically separate, thin terminal extension. This ichnospecies, initially described from the Buildex Quarry in Kansas, was interpreted as the resting trace of a wingless insect, most likely a monuran (M?ngano et al, 1997). A similar Early Permian (Wolf- campian) resting trace was illustrated by Hunt et al. (1993) from the Robledo Mountains of south-central New Mexico, who suggested that a member of the Insecta was the possible fabricator. This material was investigated by Braddy (1999) who assigned one of the forms to Tonganoxichnus, making the New Mexican material the second occurrence of this ichno- genus. More recently, specimens of the ichnospecies T. buildex- ensis have been found in the Middle Pennsylvanian Mansfield Formation of Indiana. As noted by Norman and Pickerill (1996), it is not uncommon that many ichnotaxa are only known from their type localities, and in those cases the recur- rence of a behavioral pattern remains untested. Documentation of additional examples of Tonganoxichnus, therefore, is essen- tial to provide evidence supporting the utility of this ichno- taxon. Specimens of Tonganoxichnus buildexensis from Indi- ana exhibit additional morphological features by comparison to their Kansas occurrences, thus allowing for further investiga- tion of the paleobiology of the tracemaker and elucidation of the behavioral and environmental implications of this peculiar ichnotaxon. Given this context, the aim of this paper is (1) to document the second occurrence of Tonganoxichnus buildexen- sis, (2) to further explore the biology of the maker of Tonga- noxichnus and to provide links with modern analogs, and (3) to compare specimens and associated facies and traces from Indi- ana with those of Kansas in order to reconstruct the particular environmental and taphonomic conditions that led to the preservation of these delicate biogenic structures. GEOLOGIC AND DEPOSITIONAL SETTING Tide-dominated, marginal marine systems were ubiquitous in the Midcontinental Illinois Basin during the Pennsylvanian. Tidally influenced environments commonly developed within inland estuarine paleovalleys as well as more coastal settings, such as the inner epicontinental shelf (Kvale and Barnhill, 1994; Kvale and Mastalerz, 1998). Recent geochemical studies have shown that at least some of these tidal deposits formed in brackish to freshwater settings (Mastalerz et al., 1997, 1999; Kvale and Mastalerz, 1998). The insect traces described herein 88? 87? 86? 85? FIG. 1. Location map showing tlie Illinois Basin and the Crane trace-fossil locality. were found within a tidal succession of the upper part of the Mansfield Formation, of Middle Pennsylvanian age (Morrowan to Atokan), at the Naval Surface Warfare Center (NSWC) near Crane, Martin County, Indiana (Figure 1). At this locality the Mansfield Formation, by correlation to palynologically dated outcrops to the south, is probably Atokan in age. It consists pre- dominantly of sandstones, shales, interbedded sandstones and mudstones, thin coal beds, and well-developed paleosols that are informally subdivided into three successions: lower, mid- dle, and upper (Figure 2). The trace fossils occur toward the base of the outcrop in a well-sorted, laminated siltstone that coarsens upwards to very fine-grained sandstone with small cut-and-fill rill-like struc- tures, forming the lower part of the Crane section. The silt- stones are vertically stacked in layers, with each layer generally less than 1 cm thick. These layers are separated by very thin clay drapes, typically in the order of a fraction of a millimeter. The laminae appear to thicken and thin vertically just above the underlying coal, indicating a tidal influence similar to that de- PENNSYLVANIAN INSECT TRACES 167 NE NW NW 20-5N-4W Sand I 1 m c s V f m I I I I I I 219 , 217 215. 213 211 209 207 ::00::o::: Fluvial Interticial to Subtidal Inlertldal Upper Intertidal Supratidal Peaf Mire Covered Lower Interticial Upper Intertidal Insect Trace Fossils Supratidal Peat Mire ^^ Trough cross-beding -75N-5??Owjnr Ripple cross-lamination Parallel lamination ^^ Wavy bedding A Upright calamite S Plant fragment A A Rooting QOQ Vertical burrow ooo Rip-up clasts FIG. 2. Stratigraphie section at Crane. Insect trace fossils are present in the lower part of the section. 168 M. G. MANGANO ETAL. scribed for the slightly older Pennsylvanian Hindost?n Whet- stone beds (Morrowan), also of the Mansfield Formation (Kvale et al, 1989; Kvale and Archer, 1991). Well-preserved plant fragments are present in this unit; an example is an up- right calamite trunk that is rooted in the upper part of the un- derlying coal. This calamite, basically vertical in orientation, extends approximately 40 cm into the overlying siltstone, at which point it becomes almost horizontal. Other plant frag- ments include articulated fern leaves and stems and Lepido- dendron twigs and bark fragments. Tonganoxichnus buildexensis at Crane is associated with other invertebrate traces, such as Treptichnus bifurcas, Gordia indianaensis, and amphibian trackways. Raindrop impressions also are present. Strata containing the multitude of amphibian trackways found in this part of the section are informally known as the "Crane Tetrapod Quarry." The top of the very fine-grained sandstone unit overlying the siltstone package is bioturbated, exhibiting small (1-2 mm diameter) vertical bur- rows with laminated infill. The succession from the top of the underlying coal to the top of the sandstone with vertical bur- rows is interpreted as a transgressive lithofacies succession similar to that recognized for the tidal flat deposits in the tide- dominated German Bight (Reineck, 1972; Davis and Clifton, 1987). In this model, an initial relative rise in the sea level causes drowning of the coal-forming mire and landward mi- gration of the upper intertidal mudflat and lower intertidal sandflat. The presence of the rill-like structures, which are best explained as drainage features, and the trace fossil assemblage indicates that the lower part of the Crane succession remained within an intertidal environment. A three-meter-thick covered interval separates the lower and middle parts of the Crane sec- tion. Recurring thickness variations suggest deposition under tidal influence. The base of the middle section is represented by a thin pale- osol and a superposed thin coal bed, interpreted as a supratidal peat mire deposit. This coal is overlain by another coarsening upward succession. The base of this succession consists of a thin, wavy-bedded sandstone and shale interval that is erosively overlain by a fine-grained, flat-bedded, and flaggy sandstone with ripple- and parallel-laminated cosets having clay drapes. This fine-grained sandstone interval coarsens upward to a trough cross-stratified, medium-grained sandstone. Paleoflow within this unit is bipolar in this exposure, with clay layers sev- eral millimeters thick draping many of the bedforms. This part of the sequence is clearly transgressive and records deposition from upper intertidal to subtidal environments. The tidally influenced sandstone of the middle part of the Crane section is overlain erosively by a large-scale, trough cross-stratified, fine- to medium-grained sandstone. Goethite- cemented rip-up clasts occur along the erosive scour at the base of the unit. Large, poorly preserved plant fragments are also found within this unit. Paleoflow is unidirectional and to the west. This upper part of the succession probably records depo- sition in fluvial channels. SYSTEMATIC ICHNOLOGY Ichnogenus Tonganoxichnus M?ngano, Buatois, Maples and Lanier 1997 Tonganoxichnus buildexensis M?ngano, Buatois, Maples and Lanier 1997 Figure 3A,B; Figure 4A-D; Figure 5 Emended diagnosis. Tonganoxichnus with anterior area characterized by the anterior-most pair of imprints oriented subparallel to the median axis, anteromedian impression and three pairs of conspicuous ellipsoidal imprints perpendicular or at a high angle to the median axis. A second set of circular to slightly elongate lateral distal imprints may be present. The posterior area is composed either of numerous, delicate, oblique appendage markings or a wedge-like inflated structure. Thin, straight terminal extension commonly present (modified after M?ngano et al., 1997). Material. Four slabs containing 19 specimens housed at the Department of Geological Sciences, Indiana University, Bloom- ington (lU 21246, lU 21247) and at the Indiana State Museum, Indianapolis (INS M71.3.136, INSM 71.3.137), Indiana, USA. Description. Bilaterally symmetrical traces typically con- sisting of an anteromedian mark, three pairs of conspicuous an- terior appendage impressions and numerous, delicate V-shaped markings or more commonly, a wedge-like inflated structure at the posterior part (Figures 3A,B; 4A-D). Some specimens dis- play apparent transverse annulations at the posterior part of the trace (Figure 4B,D). In addition, two elongate frontal impres- sions parallel to subparallel to the median axis are present in most specimens (Figures 3A; 4A,B). The frontal anteromedian impression, crescentic or subtriangular in shape, is a prominent feature in some specimens (Figures 3A; 4A,B ), but it may be poorly preserved or absent in others (Figure 4D). The three pairs of ellipsoidal lateral impressions are oriented perpendicu- lar to oblique to the median axis (Figure 3A,B). In addition, a second set of circular or elongate lateral markings, more or less equidistant to the median axis, is present in several specimens (Figures 3A, 4B). The posterior part of the trace is quite vari- able in morphology. In the Mansfield material, wedge-like, high relief posterior elements represent the dominant morphol- ogy (Figures 3A, 4C). However, more shallow preservational variants composed of three to six nested V-shaped markings and/or transverse segment-like annulations also are present (Figures 3B; 4B,D). Upon close examination, each V-shaped mark can be seen to be composed of two discrete elongate impressions, obliquely inclined to the median axis. In well- preserved specimens, each segment-like impression is coupled with a pair of oblique impressions (Figure 4A,B ). The angle subtended by each V-shaped mark typically decreases posteri- orly and varies from 87? to 20?. Two distinctive "half body" preservational variants are present in the Mansfield material. In a few specimens, the anterior area of the trace is virtually ab- sent, and the last pair of prominent anterior impressions is suc- ceeded by the posterior part of the trace (Figure 4A, lower left). PENNSYLVANIAN INSECT TRACES 169 \ ?? i.V. ?V ? i---- ? i,.i l\ -fr i.. i?( A .*; *'^ ?-"'?: fe. ?-?.?. ?-..... *? =?*-*#,*. ? -* ?*"? FIG. 3. Typical morphology of Tonganoxichnus buildexensis. Scale bars = 0.5 cm. (A) Negative epirelief of the anterior-most paired palp impressions, promi- nent head, three pairs of thoracic appendage impressions at the anterior part of the trace, and inflated wedge-like posterior part. Note the second set of circular anterior impressions representing the digitigrade tarsus of each thoracic appendage. The straight marks (lower part) are interpreted as terminal filament drag marks produced by the Tonganoxichnus tracemaker. (B) Positive hyporelief of a pooriy preserved subtriangular head impression, three pairs of better preserved and con- spicuous thoracic appendage impressions at the anterior part, and V-shaped shallow impressions at the posterior part. Note the presence of circular to elongate im- pressions at a distance of approximately 0.75 that of the body width, which most likely represent the walking trackway of the digitigrade tracemaker responsible for the resting structure. Other indistinct circular imprints (lower right) may be undertracks of previously produced trackways. The straight mark in the central portion of the photograph represents a drag mark of a terminal filament produced prior to the resting trace. In other specimens, the frontal mark and two or three pairs of lateral impressions are preserved, and the posterior part of the trace is absent (Figure 4B, lower right). A thin, short, cylindri- cal axial projection has rarely been observed. Tracks of disor- ganized circular imprints (Figures 3B, 4B), and slightly curved or straight marks (Figure 3A,B ) are commonly associated with distinct resting traces. Irregular traces formed of multiple seg- ment-like impressions are common within this trace fossil as- semblage (Figure 4B). The traces are predominantly preserved as positive hyporeliefs on soles of very fine-grained silty sand- stones or more rarely as negative epireliefs. Remarks. The ichnogenus Tonganoxichnus was proposed by M?ngano et al. (1997) for monuran traces preserved in Upper Pennsylvanian tidal rhythmites of eastern Kansas. Two ich- nospecies, T. buildexensis for resting traces and T. ottawensis for feeding traces, were proposed. M?ngano et al. (1997) also commented on two resting traces from the Early Permian of the Robledo Mountains in New Mexico, illustrated by Hunt et al. (1993), suggesting similarity to T. buildexensis. Recent re- search by Braddy (1998, 1999) on the ichnofauna of the Rob- ledo Mountains provided additional information on these and other resting traces. Braddy (1999, figure 1) illustrated four resting traces. He regarded two of these as new resting traces and described the other two informally as new ichnospecies of Tonganoxichnus and Rotteroidichnium. Superficially, the nested V-shaped markings of Tonganoxichnus may be confused with some preservational variants of chevronate locomotion traces of bivalves included in Protovirgularia M'Coy 1850 and its junior 170 M. G. MANGANO ET AL. m -??/Mr m ..-.'*?' _?' , ?:???. ii^'t?^*t**^?? V'/---Vi. '''??" J?'?'l'' j'.rj- .-!i-jr? ??7? i-: v~ -.;???? h. ^f *, ?? !.' ' ? ? ..?4. .-*,? ;? "'?'? \^M(?t?^??A:.:^^ J^ ^/?i^. :}i?^:Mi?^M:^^Scim .'S?J FIG. 4. Preservational variants of Tonganoxichnus buildexensis in positive hyporelief. Scale bars = 0.5 cm. (A) Upper specimen exliibits typical anterior and posterior features of T. buildexensis; at lower left a specimen displays a "half body" position that shows the posterior, or abdominal, aspect of the original trace- maker. (B) A cluster of three specimens. An exquisitely preserved specimen (upper left) with five nested, delicate, V-shaped marks at the posterior part and a sec- ond set of three pairs of associated circular impressions revealing tarsal contact points. A specimen displaying annulated morphology (upper center) interpreted as three partly superimposed "half body" imprints, each representing seven evident abdominal segments without associated tarsal impressions. A specimen show- ing the half body anterior-part preservation (lower right). Note the presence of circular imprints (lower) most likely representing partially preserved undertracks. (C) A cluster of two specimens showing an inflated posterior part with poorly preserved, posterior morphological features. (D) Closeup of a specimen showing very shallow, almost indistinct, head impressions; three pairs of thoracic appendage impressions; an annulated, segment-like posterior part; and probable circular tarsal marks at left. PENNSYLVANIAN INSECT TRACES 171 synonym Biformites Linck 1949 (see, for example, figure 2H of Miller and Knox, 1985). However, overall morphology and dis- tinct anatomical details of Tonganoxichnus clearly distinguish this ichnogenus from chevronate bivalve trails. Morphologic elements of T. buildexensis are illustrated in Figure 5. Based on the anterior-most elongate pair of impres- sions (interpreted as maxillary palp resting sites), the frontal anteromedian impression (representing the head), the three pairs of conspicuous anterior markings (that undoubtedly are thoracic appendage marks), and delicate abdominal structures, we conclude that the Crane material warrants inclusion within Tonganoxichnus buildexensis (cf. M?ngano et al., 1997). In ad- dition, the size range is similar to that of T. buildexensis. The second set of distal circular lateral impressions commonly as- sociated with the thoracic area most likely represent digitigrade oo Al LI FIG. 5. Morphologic components of Tonganoxichnus (modified from M?ngano et al, 1997). AI = Anteriormost pair of appendage imprints, A2-A4 = Second to fourtli pairs of anterior appendage imprints, FI = Frontal impres- sions, PI -P8 = First to eight pairs of posterior appendage imprints, CP - Cylin- drical projections, TI = Transverse segment-like body impressions, LI = Lateral impressions. tarsal impressions; the more proximal, ellipsoidal thoracic markings adjacent to the axis are interpreted as impressions of the coxae. Interestingly, although not described in the original material, these impressions seem to be partially preserved in the Tonganoxichnus buildexensis holotype (cf. M?ngano et al., 1997; figure 3 A) and other specimens of the Tonganoxie Sand- stone Member The fact that some morphologic features are lacking in some specimens favors an undertrace origin for many of the preservational variants of Tonganoxichnus buildex- ensis. As opposed to a surface trace, the undertrace represents the duplicate imprint on lower surfaces (see Goldring and Seilacher, 1971). The digitigrade tarsal impressions, palp seg- ment, and last-segment abdominal impressions are more shal- low than other impressions, such as the coxae, and may not be preserved on relatively deep subsuperficial laminae. Although the terminal filament mark is not in physical continuity with the abdominal impressions, isolated linear marks and anterior palp impressions coupled with posterior linear marks indicate that a tail-like feature was present in the tracemaker. A similar mode of occurrence was documented by M?ngano et al. (1997; see figure 3B), and interpreted as a result of sliding-on-the-water behavior with the sensorial maxillary palps probing the sedi- ment surface and the tail-like projection passively touching the substrate. Interestingly, the preservational variant characterized by the third pair of thoracic appendages and abdominal im- pressions (Figure 4A, lower left) is similar to the insect trace il- lustrated by Hunt et al. (1993, figure 10, left). This "half body" undertrace, with the anterior features virtually absent, suggests the center of gravity was located on the third thoracic ap- pendage and may represent a prejumping position (cf Evans, 1975). Undertracks of circular or elongate imprints are too in- complete to deserve formal description, but they most likely represent a fast-walking gait performed by thoracic appendages (see next section). Irregular, multiple, segment-like impressions are difficult to interpret in terms of ethology; they are variable in morphology and record the impressions of different parts of the body. Some may well represent landing structures of bod- ies that are upside down or on their side (Figure 4B, center; see next section). M?ngano et al. (1997) interpreted differences in ichnofossil morphology, principally in the relative proportions of the head, thorax, and abdomen that occur between large and small spec- imens. These differences reflect different developmental, or in- star, stages. The size range and morphological characters of the specimens from the Mansfield Formation are compatible with those documented for adult monurans (cf Sharov, 1961; Dur- den, 1978; Kukalov?-Peck, 1987, 1991). THE PALEOBIOLOGY OF TONGANOXICHNUS The Lower Pennsylvanian Mansfield Formation of Indiana is the third locality in North America to reveal Tonganoxichnus traces and, together with the Kansas material, preserves deli- cate hyporelief and rare epirelief impressions consistent with fossil monuran and modern archaeognathan ventral anatomy. 172 M. G. MANGANO ET AL. The morphological similarities between Pennsylvanian to Per- mian Monura and modern Archaeognatha are striking, as both groups have been considered to be closely related clades (Lauterbach, 1972; Kukalov?-Peck, 1991) or subgroups of the same clade (Carpenter, 1992; Rasnitsyn, 1999; Goldbach, 2000), even though these two lineages are probably derived from an earlier ectognathous, monocondylic lineage present during the Early and Middle Devonian (Shear et al., 1984; La- bandeira et al., 1988). Archaeognathans are some of the earli- est known detritivores in terrestrial ecosystems (Shear and Kukalov?-Peck, 1990; see also Watson, 1989) and have a spo- radic and sparse record extending from the Early Devonian to the present (Kukalov?-Peck, 1987; Sturm and Bach de Roca, 1993; Bitsch and Nel, 1999). The Monura, however, are only known from the Middle Pennsylvanian to early Late Permian (Sharov, 1957, 1966; Durden, 1978; Kukalov?-Peck, 1987; Rowland, 1997). Machilid Archaeognatha are not only mor- phologically similar to Monura (Kukalov?-Peck, 1991; Rasnit- syn, 1999), but also they occur in coastal, humid environments (Delany, 1954; Benedetti, 1973; Sturm and Bach de Roca, 1993), thus allowing useful comparisons regarding the mor- phology of the possible tracemakers and the overall behavior and feeding habits represented by Tonganoxichnus. Evidence regarding the habitat of the Tonganoxichnus fabri- cator indicates that it inhabited coarse silty substrates on tidal flats that were periodically inundated with water and influxes of sediment. Placement of these trace fossils along a seaward- to-landward gradient suggests typically freshwater conditions that extended marginally into shore sediments that were de- posited by marine tidal currents (Buatois et al., 1997; M?ngano et al., 1997). The size of the Tonganoxichnus buildexensis cul- prit fits comfortably into the adult range of modern machilid Archaeognatha, which ranges from 8 to 14 mm for adult fe- males and males from several measured taxa (Reilly, 1915; Wygodzinsky, 1944; Sturm, 1952, 1955; Delany, 1957, 1959, 1960). Many modern machilid archaeognathan species are re- stricted to coastal localities, including Dilta littoralis Womers- ley occurring along coastal regions of western Europe but sometimes extending somewhat inland (Delany, 1954), and Petrobius brevistylis Carpenter and P. maritimus Leach inhabit- ing the rocky coasts of northern North America and northwest- ern Europe, respectively (Delany, 1959; Wygodzinsky and Schmidt, 1980). Whereas modern archaeognathans are funda- mentally detritivores (Adis and Sturm, 1987; Sturm and Bach de Roca, 1993), their dietary strategy in these coastal habitats is better described as opportunistic and includes bacteria, sur- face-film algae, lichens, decaying fruit, dead leaves, fungi, and unidentified solid organic debris (Willem, 1924; Wygodzinsky, 1944; Delany, 1954, 1959; Smith, 1970; Sturm, 1984). Some species consume spores and pollen (Kaplin, 1978; Benedetti, 1973), and mineral matter also has been recovered from gut contents (Wygodzinsky, 1944). During feeding, machilid ar- chaeognathans employ long, leglike, seven-segmented maxil- lary palps to probe and scratch the substrate and also to assist in food handling (Willem, 1924; Sturm, 1955); a similar activ- ity occurs during copulation, in which there is "drumming" by the maxillary palps (Goldbach, 2000). These palps are typically held in a curved, sigmoidal position at rest with the dorsal sur- face of the distal segments positioned on the substrate surface and aligned in a subparallel fashion, a behavior also inferred from Tonganoxichnus buildexensis (Figures 3A, 4A,B). These maxillary-palp impressions range from relatively short ellip- soids representing one segment each to slightly asymmetrical structures indicating a more oblique position (Figure 3A,B). Modern archaeognathans use sensillae-rich abdominal styli for tactile communication with the substrate and to prop their abdomen when slowly walking or at rest (Willem, 1924; Smith, 1970). Only under rapid locomotion do abdominal styli not contact the ground. Styli apparently are not internally muscu- lated and thus are minimally movable structures that leave an en echelon series of a few to several V-shaped impressions on soft substrates (Sturm, 1955; Smith, 1970). Such chevron- shaped impressions are deployed segmentally and imbricately in extant machilid Archaeognatha (Delany, 1959; Machida, 1981; Sturm and Messner, 1992), a feature that is clearly recorded in several Tonganoxichnus buildexensis specimens from Indiana (Figures 3B, 4A, and especially Figure 4B) where there are up to five well-preserved sets of styli indentations. Unlike acuminate styli, thoracic legs bear a double claw that contacts the substrate in a digitigrade stance and produce foot- print tracks that are placed from the lateral body margin at a distance of approximately 0.75 times that of the body width (Willem, 1924, Figure 2), corresponding closely to the body- margin to circular distal footprint distances in Tonganoxichnus buildexensis (Figures 3A; 4B). By contrast, the terminal fila- ment, typically elevated above or barely in contact with the substrate surface (Smith, 1970; Kukalov?-Peck, 1987), is used as a jumping device that can launch the insect up to 20 cm in the air (Evans, 1975; Sturm and Bach de Roca, 1993). This is done in conjunction with dorsoventral abdominal flexion and movement of paired maxillary palps. Subsequent landing is somewhat haphazard and can occur in various orientations, in- cluding touchdowns on all or some of the tarsi, on lateral body exposures, or other positions (Sturm, 1955). Linear impres- sions of the terminal filament, which can approximate or ex- ceed the body length in modern bristletails (B?r, 1912), are commonly associated with some specimens of Tonganoxichnus buildexensis (Figure 3A,B), although these are interpreted as shallow drag marks. Also, there is no evidence of conjugate jumping and landing structures, such as those documented in Tonganoxichnus ottawensis from eastern Kansas (M?ngano et al., 1997). Although speculative, some irregular multiple seg- ment impressions (Figure 4B, center) may record random land- ing structures. A considerably wider, postero-medial structure extending from the posterior-most abdomen may be a female ovipositor (Figure 4A,B), representing a stouter, thicker struc- ture when compared to a gently-tapering median filament (Agrell, 1945; Sturm and Bach de Roca, 1992). PENNSYLVANIAN INSECT TRACES 173 In a recent paper by Rasnitsyn (1999), a new monuran species was described, Dasyleptus noli, from the Late Pennsyl- vanian Commentary Basin in France. Also, three additional congeneric species were redescribed: D. lucasi Brongniart (1885), D. brongniarti Sharov (1957), and D. sharovi Durden (1978). Also mentioned was an interpretation of Tonganoxich- nus buildexensis and T. ottawensis from Kansas (M?ngano et al., 1997), in which it was claimed that this trace fossil was un- likely to be attributable to the Monura. The principal reason given was that Tonganoxichnus did not preserve a relatively deep, postabdominal furrow that would be produced by a ter- minal filament (median cercus) as it was suddenly depressed into the sediment for an aerial jump (Smith, 1970; Evans, 1975). However, an essential factor that was not discussed was that most of the Tonganoxichnus traces illustrated by M?ngano et al. (1997) are landing or resting traces, with the terminal fil- ament interpreted as either erect in the air or more rarely lying on the substratum. This anatomical position of the terminal fil- ament was illustrated by Evans (1975), demonstrating that the posterior-most abdominal segments and terminal filament were kept in an upward position, clear of the sediment surface (M?ngano et al., 1997; figure 7). Thus, with the possible ex- ception of Figure 5A,B of M?ngano et al. (1997), the Kansas Tonganoxichnus specimens were not take-off traces and thus are not characterized by much deeper postabdominal troughs that would have been consistent with the sudden depression of the terminal filament into the sediment for springing a jump. The Tonganoxichnus traces from Indiana apparently exhibit ev- idence only for landing or resting. In addition, although cited by Rasnitsyn (1999) as an unexplained absence, the distal tarsi of all thoracic legs do contact the sediment in the Indiana ma- terial, now indicated by three pairs of lateral point-indentations into the sediment located alongside the main body axis (Figures 3; 4B, D), which perhaps remained unnoticed in the Kansas material (M?ngano et al., 1997; but see figure 4B). Interest- ingly, the lower-most Permian New Mexico locality from which Rowland (1998) described Dasyleptus sp., confirmed by Rasnitsyn (1999), is somewhat older than the Lower Permian site containing Tonganoxichnus mentioned by Braddy (1998, 1999) approximately 180 km to the south. DISCUSSION AND CONCLUSIONS A comparison among all three occurrences of Tonganoxich- nus buildexensis reveals remarkable similarities. The age of the youngest occurrence of Tonganoxichnus is Early Permian, whereas the ages of the older trace fossil-bearing units is Mid- dle to Late Pennsylvanian, with the Tonganoxichnus-hearing Mansfield Formation at Crane being slightly older (Atokan) than the Tonganoxie Sandstone Member at Buildex (Virgilian). These occurrences are consistent with the Middle Pennsylvan- ian to early Upper Permian stratigraphie range of the Monura. Although the trace fossil context of the Indiana Mansfield For- mation deserves further investigation, the insect trace fossil as- semblage is similar to that of the Tonganoxie Sandstone Mem- ber of Kansas. In both units, T. buildexensis is associated with probable insect feeding [Treptichnus bifurcus (Miller)] and grazing [Gordia indianaensis (Miller)] traces, as well as am- phibian tracks and plant traces. The Buildex ichnofauna was analyzed systematically, and other forms, particularly arthro- pod trackways, were also documented (Buatois etal., 1998), al- though comprehensive trace-fossil sampling and study of the Crane rhythmites has not been undertaken. Bioturbation, namely vertical disturbance of primary structures, is negligible in both the Tonganoxichnus-hearing portions of the Buildex and the Crane deposits. The presence on some bedding planes of crenulated curvi- linear disruptions of laminae indicates the possible feeding on an algal surface mat. Although other surface-dwelling arthro- pods may be responsible for these distinctive microlaminar irregularities, it is highly likely that they are the result of palpal probes and mandible manipulation of a thin algal mat by monu- rans. Primary sedimentary fabric and surface traces are exquis- itely preserved at the Indiana site. Interestingly, in both locali- ties T buildexensis is preserved in similar lithofacies, namely rhythmically bedded laminated siltstones with clay drapes, and structures indicative of subaerial exposure, such as rill marks and raindrop impressions, are present on bedding planes. Well- preserved plant fragments and in situ upright trunks are com- mon in both units. Strata are stacked, forming thickening and thinning packages, suggesting tidal deposition. Delicate preser- vation of surface traces indicates absence of erosion. Estimations at Buildex, based on an analysis of tidal rhythmite cychcity, sug- gest an elevated sedimentation rate of 3.8 m/yr (Lanier et al., 1993). Comparable calculations for the Crane site are not avail- able, but field observations indicate similar depositional rates. Marine indicators, either as trace fossils or as body fossils, are absent in both Buildex and Crane. Buatois et al. (1997) in- terpreted the Buildex ichnofauna as recording the activity of a typical freshwater to terrestrial benthos. These authors sug- gested that the existence of this mixed freshwater to terrestrial ichnofauna in tidal rhythmites was indicative of tidal flats in the most proximal area of the inner estuary under freshwater con- ditions, in a zone between the maximum limit of landward tidal currents and the salinity limit further towards the sea. The dom- inance of insect traces and the absence of a marine burrowing infauna in Crane also suggest freshwater conditions. Rapid bur- ial, absence of significant erosive events, and absence of infau- nal brackish water organisms were most likely the factors in- volved in the preservation of these delicate, superficial insect traces. Modern studies in tidal flats reveal a complex interplay between sedimentation and microbial growth (Noffke et al., 1996). High sedimentation rate and intense physical re- working commonly prevent significant biomass accumulation and development of microbially induced sedimentary struc- tures. At Buildex and Crane, however, high sedimentation rates are interpreted as related to short-term depositional events, most likely seasonal flooding of fluvial systems (cf. Lanier et al. 174 M. G. MANGANO ET AL. 1993). At Buildex some sedimentary structures, such as wrin- kle marks and pinch-out lamination, may be related to incipient tissue-like mats (cf. Noffke et al., 1996). The sharpness of mor- phologically delicate features of Tonganoxichnus also is sug- gestive of a relatively firm substrate that acted as a plastic casting medium. Cohesive behavior has been interpreted as a useful indicator of microbial mat colonization in siliciclastic sediments (Schieber, 1999). The recording of the ichnogenus Tonganoxichnus in the Mansfield Formation of Indiana has three important implica- tions. First, it confirms that this ichnotaxon documents a recur- rent behavioral pattern in Upper Carboniferous coastal ecosys- tems. Second, it provides detailed ichnologic documentation for some body structures and behaviors that are difficult to infer from body-fossil material alone. Third, sedimentologic and ich- nologic observations indicate that the Tonganoxichnus assem- blage seems to be common in tidal flat rhythmites developed under freshwater conditions, probably in the innermost part of estuarine systems, adjacent to the fluvio-estuarine transition. We predict that further occurrences of Tonganoxichnus in late Paleozoic, fine-grained tidal rhythmites will be recorded in the near future. ACKNOWLEDGMENTS Support for research was provided to MGM through Grants- in-Aid-of-Research by Sigma Delta Epsilon (Graduate Women in Science) and the Antorchas Foundation, and to CCL from the Walcott Fund of the Paleobiology Department at the National Museum of Natural History. EK thanks the Naval Surface War- fare Center, Crane, Indiana. Ichnos reviewers Franz F?rsich and Michael Schlirf provided useful comments. Jim Hunsicker and Tom Brent of the Department of the Navy helped facilitate the collecting. Doug Parker, Al Cartwright, Jim Watson, Char- lie Sinn, and Todd Thompson assisted EK in the field. Pho- tographs of the slabs were taken by Barbara Hill and John Day (Indiana Geological Survey). This is contribution 77 of the Evolution of Terrestrial Ecosystems Consortium at the National Museum of Natural History, Washington, DC, USA. REFERENCES Adis, J. and Sturm, H. 1987. On the natural history and ecology of Meinertel- lidae (Archaeognatha, Insecta) from dryland and inundation forests of Cen- tral Amazonia. Amazoniana, 10:197-218. Agrell, I. 1945. Die Arthropodenfauna von Madeira nach den Ergebnissen der Reise von Prof. Dr. O. Lundblad Juli-August 1935. Archiv f?r Zoologi, 35B:l-5. B?r, H. 1912. Beitr?ge zur Kenntnis der Thysanuren. Zoologische Zeitschrift, 48:1-92, pis. 1-6. Benedetti, R. 1973. Notes on the biology o? Neomachilis halophila on a Cali- fornia sandy beach (Thysanura: Machilidae). Pan-Pacific Entomologist, 49:246-249. Bitsch, J. and Nel, A. 1999. Morphology and classification of the extinct Ar- chaeognatha and related taxa (Hexapoda). Annales de la Soci?t? Ento- mologique de France, 35:17-29. Braddy, S. J. 1998. An overview of the invertebrate ichnotaxa from the Robledo Mountains ichnofauna (Lower Permian), southern New Mexico. New Mex- ico Museum of Natural History and Science Bulletin, 12:93-98. Braddy, S. J. 1999. Terrestrial trace fossils from the Robledo Mountains ichno- fauna (Lower Permian) of southern New Mexico. Geoscientist, 9:5-6. Brongniart, C. 1885. Les insects fossils des terrains primaries. Bulletin de la Soci?t? des Amis des Sciences Naturelles de Rouen, 1885: 50-68. Buatois, L. A., M?ngano, M. G., Maples, C. G., and Lanier, W. P. 1997. The paradox of nonmarine ichnofaunas in tidal rhythmites: integrating sedimen- tologic and ichnologic data from the Late Carboniferous of eastern Kansas, USA. Palaios, 12:467^81. Buatois, L. A., M?ngano, M. G., Maples, C. G., and Lanier, W. P 1998. Ich- nology of an Upper Carboniferous fluvio-estuarine paleovalley: The Tonga- noxie Sandstone Member, Buildex Quarry, eastern Kansas, USA. Journal of Paleontology, 71:152-180. Carpenter, F. M. 1992. Volume 3. Superclass Hexapoda. In Moore, R. C, Kaesler, R. L., Brosius, E., Kiem, J., and Priesner, J. (eds.). Treatise on In- vertebrate Paleontology, Part R, Arthropoda 4, 3: xxi + 655 p. Geological Society of America: Boulder, Colorado, and University of Kansas Press, Lawrence, Kansas, USA. Davis, R. A., Jr., and Clifton, H. E. 1987. Sea-level change and the preservation potential of wave-dominated and tide-dominated coastal sequences. In Nummedal, D., Pilkey, D. H., and Howard, J. D. (eds.), Sea-level Fluctua- tions and Coastal Evolution. Society of Economic Paleontologists and Min- eralogists Special Publication, 4:167-178. Delany, M. J. 1954. Studies on the life history and ecology of Dilta littomlis (Womersley, 1930) (Thysanura, Machilidae). Transactions of the Royal En- tomological Society of London, 105:31-63. Delany, J. 1957. Life histories in the Thysanura. Acta Zool?gica Cracoviensia, 2:61-90. Delany, J. 1959. The life histories and ecology of two species of Petwbius Leach, P. brevistylis and P. maritimus. Transactions of the Royal Society of Edinburgh, 63:501-533. Delany, M. J. 1960. A study on the post-embryonic development of Machiloides delanyi Wygodzinsky (Thysanura: Meinertellidae). Proceed- ings of the Royal Entomological Society of London (A), 36:81-87. Durden, C. J. 1978. A dasyleptid from the Permian of Kansas, Lepidodasypus sharovi, n. gen., n. sp. (Insecta: Thysanura: Monura). Pearce-Sellards Series, 30:1-9. Evans, M. E. G. 1975. The jump of Petrobius (Thysanura: Machilidae). Jour- nal of Zoology, 176:49-65. Genise, J. F. and Bown, T. M. 1994. New Miocene scarabeid and hymenopter- ous nests and Early Miocene (Santacrucian) paleoenvironments, Patagonian Argentina. Ichnos, 3:107-117. Goldbach, B. 2000. The evolutionary changes in the reproductive biology of the Archaeognatha (Insecta). Zoologischer Anzeiger 239:215-229. Goldring, R. and Seilacher, A. 1971. Limulid undertracks and their sedimento- logical implications. Neues Jahrbuch f?r Geologie und Pal?ontologie, Ab- handlungen, 137:422-442. Holub, V. and Kozur, H. 1981. Arthropodenf?hrten aus dem Rotliegend der CSSR. Geologie und Pal?ontologie Mitteilungen, Innsbruck, 11:95-148. Hunt, A. R, Lockley, M. G., Lucas, S. G., Macdonald, J. P, Hotton, N. III, and Kramer, J. 1993. Early Permian tracksites in the Robledo Mountains, South Central New Mexico. New Mexico Museum of Natural History and Science, Bulletin, 2:23-31. Jarzembowski, S. A. and Ross, A. J. 1993. Time flies: the geological record of insects. Geology Today, 9:218-223. Kaplin, V. G. 1978. A comparative ecological and morphological description of bristle-tails of the family Lepismatidae (Thysanura) in the eastern Karakum sands. Entomological Review, 56:72-84, Kukalov?-Peck, J. 1987. New Carboniferous Diplura, Monura, and Thysanura, the hexapod ground plan, and the role of thoracic side lobes in the origin of wings (Insecta). Canadian Journal of Zoology, 65:2327-2345. Kukalov?-Peck, J. 1991. Fossil history and the evolution of hexapod structures. In Naumann, I. D., Came, P. B., Lawrence, J. F., Nielsen, E. S., Spradbery, J. P, Taylor, R. W., Whitten, M. J., and Littlejohn, M. J., (eds.), The Insects of Australia, 1:141-179. Cornell University Press, Ithaca, NY, USA, 2nd edn. Kvale, E. P. and Archer, A. W. 1991. Characteristics of two, Pennsylvanian-age, semidiurnal tidal deposits in the Illinois Basin, USA. In Smith, D. G., Rein- PENNSYLVANIAN INSECT TRACES 175 son, G. E., Zaitlin, B. A., and Rahmani, R. A. (eds.), Clastic Tidal Sedimen- tology. Canadian Society of Petroleum Geologists Memoir, 16:179-188. Kvale, E. P. and Barnhill, M. L. 1994. Evolution of Lower Pennsylvanian estu- arine faci?s within two adjacent paleovalleys, Illinois Basin, Indiana. In Dal- rymple, R. W., Boyd, R., and Zaitlin, B. A. (eds.), Incised-valley Systems: Origin and Sedimentary Sequences. Society of Economic Paleontologists and Mineralogists Special Publication, 51:191-207. Kvale, E.P. and Mastalerz, M. 1998. Evidence of ancient freshwater tidal de- posits. In Alexander, C, Davis, R., and Henry, V. (eds.), Tidalites: Processes and Products. Society of Economic Paleontologists and Mineralogists Spe- cial Publication, 61:95-107. Kvale, E. P., Archer, A. W., and Johnson, H. R. 1989. Daily, monthly, and yearly tidal cycles within laminated siltstone of the Mansfield Formation (Pennsylvanian) of Indiana. Geology, 17:365-368. Labandeira, C. C. 1998a. Early history of arthropod and vascular plant associ- ations. Annual Review of Earth and Planetary Sciences, 26:329-377. Labandeira, C. C. 1998b. The role of insects in Late Jurassic to Middle Creta- ceous ecosystems. In Lucas, S. G., Kirkland, J. I. and Estep, J. W. (eds.). Lower and Middle Cretaceous Terrestrial Ecosystems. New Mexico Mu- seum of Natural History and Science Bulletin, 14:105-124. Labandeira, C. C. and Sepkoski, J. J. 1993. Insect diversity in the fossil record. Science, 261:310-315. Labandeira, C. C, Beall, B., and Hueber, F. 1988. Early insect diversification: evidence from a lower Devonian bristletail from Quebec. Science, 242:913-916. Lanier, W. P., Feldman, H. R., and Archer, A. W. 1993: Tidal sedimentation from a fluvial to estuarine transition, Douglas Group, Missourian-Virgilian, Kansas. Journal of Sedimentary Petrology, 63:860-873. Lauterbach, K.-E. 1972. Die morphologischen Grundlagen f?r die Entstehung der Entognathie bei den apterygoten Insekten in phylogenetischer Sicht. Zo- ologische Beitr?ge, N.F., 18:25-69. Linck, O. 1949. Lebens-Spuren aus dem Schilfsandstein (mittl. Keuper km2) NW-W?rttembergs und ihre Bedeutung f?r die Bildungsgeschichte der Stufe. Verein f?r vaterl?ndische Naturkunde in W?rttemberg Jahreshefte, 1949:97-101. M?ngano, M. G., Buatois, L. A., Maples, C. G., and Lanier, W. P 1997. Tonganoxichnus, a new insect trace from the Upper Carboniferous of east- ern Kansas. Lethaia, 30:113-125. Machida, R. 1981. External features of embryonic development of a jumping bristletail, PedetonUis unimaculatus Machida (Insecta, Thysanura, Machili- dae). Journal of Morphology, 168:339-355. Mastalerz, M., Kvale, E. P., Stankiewicz, A. B., and Portle, K. 1997. Organic geochemistry in Pennsylvanian tidally-influenced sediments from SW Indi- ana. Organic Geochemistry, 30:57-73. Mastalerz, M., Salmon, G., Stankiewicz, A. B., Kvale, E. P., and Millard, C. 1999. Environments following peat deposition: examples from the Mans- field Formation (Lower Pennsylvanian) in Indiana. International Journal of Coal Geology, 33:275-299. M'Coy, F. 1850. On some genera and species of Silurian Radiata in the collec- tion of the University of Cambridge. Annals and Magazine of Natural His- tory, ser. 2, 6:270-290. Miller, M. F. and Knox, L. W. 1985. Biogenic structures and depositional envi- ronments of a Lower Pennsylvanian coal-bearing sequence, northern Cum- berland Plateau, Tennessee, U.S.A. In Curran, H. A. (ed.), Biogenic Struc- tures: Their use in interpreting depositional environments. Society of Economic Paleontologists and Mineralogists, Special Publication, 35:67-97. Nofflce, N., Gerdes, G., Klenke, T, and Krumbein, W. E. 1996. Microbially in- duced sedimentary structures?examples from modern sediments of silici- clastic tidal flats. Zentralblatt f?r Geologie und Pal?ontologie Teil I, 1996:307-316. Norman, Y. J.-C. and Pickerill, R. K. 1996. Lobichnus variabilis Kemper 1968 from the Devonian of Maine, northeastern U.S.A. Ichnos, 4:233-237. Rasnitsyn, A. P. 1999. Taxonomy and morphology of Oasyleptus Brongniart, 1885, with description of a new species (Insecta: Machilida: Dasyleptidae). Russian Entomological Journal, 8:145-154. Reilly, A. J. 1915. Notes on the British Machilidae with descriptions of two new species. Annals and Magazine of Natural History, 16:10-15. Reineck, H.-E. 1972. Tidal flats. In Rigby, J.. K. and Hamblin, W. K. (eds.). Recognition of Ancient Sedimentary Environments. Society of Eco- nomic Paleontologists and Mineralogists. Special Publication, 16:146- 159. Rowland, J. M. 1997. The Late Paleozoic insect assemblage at Carrizo Arroyo, New Mexico. Bulletin of the New Mexico Museum of Natural History and Science, 11:1-7. Schieber, J. 1999. Microbial mats in terrigenous elastics: the challenge of iden- tification in the rock record. Palaios, 14:3-12. Sharov, A. G. 1957. Peculiar Paleozoic wingless insects of the new order Monura (Insecta, Apterygota). Doklady Akademii Nauk SSSR, 122: 733-736 (in Russian). Sharov, A. G. 1961. Paleozoic insects of the Kuznetsk Basin. Trudy Paleonto- logicheskogo Institu?a, 85:65-68. (In Russian). Sharov, A. G. 1966. Basic Arthropodan Stock. Pergamon: Glasgow, 271 p. Shear, W. A. and Kukalov?-Peck, J. 1990. The ecology of Paleozoic terrestrial arthropods: the fossil evidence. Canadian Journal of Zoology, 68:1807- 1834. Shear, W. A., Bonamo, R M., Grierson, J. D., Rolfe, W. D. 1., Smith, E.-L., and Norton, R. A. 1984. Early land animals in North America: evidence from Devonian age arthropods from Gilboa, New York. Science, 244:492- 494. Smith, E.-L. 1970. Biology and structure of some California bristletails and sil- verfish (Apterygota: Microcoryphia, Thysanura). Pan-Pacific Entomologist, 46:212-225. Sturm, H. 1952. Die Paarung bei Machilis (Felsenspringer). Naturwis- senschaften, 39:308. Sturm, H. 1955. Beitr?ge zur Ethologie einiger mitteldeutscher Machiliden. Zeitschrift f?r Tierpsychologie, 12:337-363. Sturm, H. 1984. Zur Systematik, Biogeographie und Evolution der s?- damerikanischen Meinertellidae (Machiloidea, Archaeognatha, Insecta). Zeitschrift f?r zoologische Systematik und Evolutions-Forschung, 22:27-44. Sturm, H. and Bach de Roca, C. 1992. New American Meinertellidae (Ar- chaeognatha, Machiloidea). Pan-Pacific Entomologist, 68:174-191. Sturm, H. and Bach de Roca, C. 1993. On the systematics of the Archaeognatha (Insecta). Entomolog?a Generaiis, 18:55-90. Sturm, H. and Messner, B. 1992. Zwei neue Typen von Sinnesorganen bei M?nnchen von drei Mesomachilis-hntn (Machilidae, Archaeognatha, In- secta). Zoologische Beitr?ge N.E, 34:375-394. Walker, E. F. 1985. Arthropod ichnofauna of the Old Red Sandstone at Dunure and Montrose Scotland. Transactions of the Royal Society of Edinburgh. Earth Sciences, 76:287-297. Watson, R. T. 1989. Niche separation in Namib Desert dune Lepismatidae (Thysanura, Insecta): detritivores in an allochthonous detritus ecosystem. Journal of Arid Environments, 17:37-48. Willem, V. 1924. Observations sur "Machilis mar?tima" (Premi?re note). Bul- letin Biologique, 58:306-320. Wygodzinsky, P. 1944. Contribui??o conhecimento da familia Machilidae (Thysanura) do M?xico, do Brasil e da Argentina. Revista de Entomolog?a, 15:54-96. Wygodzinsky, P. and Schmidt, K. 1980. Survey of the Microcoryphia (Insecta) of the northeastern United States and adjacent provinces of Canada. Ameri- can Museum Novitates, 2701:1-17.