Once upon a time in America: recognition of the species of Libitioides from USA, with comments on other American Cosmetidae (Opiliones, Laniatores)

. The American species of Cosmetidae, in spite of being clearly distinguished in the older literature, were mixed-up by Roewer. This, aggravated by groundless synonymies done by Goodnight & Goodnight, prevented all subsequent authors from properly recognizing and adequately naming those species. Herein, we de ﬁ ne and characterize the three most widespread species of Cosmetidae in the USA, explaining the misleading synonymies and misidenti ﬁ cations in the taxonomic literature. A recent phylogenetic analysis revalidated Libitioides Roewer, 1912 from the synonymy of Vonones Simon, 1879 to include three American species: Gonyleptes ornatum Say, 1821 (from southeastern USA), Cynorta sayi Simon, 1879 (from central-southern USA) and Cosmetus albolineatus Sørensen, 1884 (from eastern USA). Expanding on that, we herein aggregate another two species to Libitioides : Vonones modestus Banks, 1909 (from Cuba, herein revalidated from the synonymy of Libitioides ornata Roewer, 1912) and Libitioides scabrissima Roewer, 1912 (from Mexico, restored from the current combination with Vonones ). The following subjective synonymies are proposed: (1) Platycynorta Mello-Leitão, 1933 and Denticynorta Roewer, 1947 = Libitioides ; (2) Metacynorta denticus Walker, 1928 = Cosmetus albolineatus ; (3) Platycynorta secunda Roewer, 1947 = Vonones modestus ; (4) Libitioides ornata Roewer, 1912 and Cynorta ( Cynorta ) depressa Sørensen, 1932 = Cynorta sayi . Outside Libitioides , Cynorta bimaculata Banks, 1893, currently combined under Calicynorta Goodnight & Goodnight, 1943, and originally reported from California (due to a misinterpretation of label) has its type locality corrected as to be in Costa Rica instead and is transferred to Holovonones Roewer, 1912. Accordingly, Calicynorta is herein considered a junior subjective synonym of Holovonones Roewer, 1912. Platycynorta clavifemur Roewer, 1957 from Peru is newly combined under the genus Ambatoiella Mello-Leitão, 1943, otherwise known from Ecuador. The present analysis of the distributional data suggests that the occurrence of Cosmetidae is determined not directly by temperature, but by the absence of a dry season and the presence of a hot summer.


Introduction
The family Cosmetidae C.L. Koch, 1839 recently underwent a cladistic analysis (Medrano et al. 2021) in which several subfamilial and generic taxa were defi ned. The genus Libitioides Roewer, 1912, was then retrieved as the sister-group of the Taitoinae Medrano, Kury & Mendes, 2021. But because of the weak support, and the lack of the conspicuous synapomorphies defi ning an expanded Taitoinae, it was left unassigned to any subfamily. Libitioides currently has a few species in North America and the Caribbean. The three species of Libitioides which occur in the United States were used in that analysis. These species appear both in the literature and on internet websites of macro photographers mostly identifi ed as "Vonones ornata" or "Vonones sayi" (see below in the Examples of misidentifi cations section, within Results).
In this project we treat in detail the taxonomy of Libitioides and for that we access other species recorded from the USA which are allocated in other genera (such as Calicynorta Goodnight &Goodnight, 1943, andCynortoides Roewer, 1912).
Aiming to reach diagnoses for the American species of Libitioides and to defi ne their geographic distribution, we have examined material from several museums, some of it destroyed in the 2018 fi re of the Brazilian National Museum (MNRJ). Besides non-type specimens from the USA, we studied type material of the relevant species, whenever possible, or at least photographic images of this material.
In order to maximize the number of geographic records and to refi ne the areas of distribution of each species, additionally to the study of specimens, we also gleaned information from a large number of photographic records as published on the Internet websites of naturalists/macrophotographers (Flickr, Bugguide, iNaturalist, Instagram).

Material and methods
We have looked for phenotypic characters with potential to allow distinction among the species, and easily verifi able by the study of photographs only. Five characters were then selected and mapped against individual records onto a political map of the eastern USA. We then tabulated the combinations of those fi ve characters against the distribution to fi nd unique combinations of characters that defi ned discrete species.
Because of the complex taxonomic story of the taxa involved, not only protonyms and aponyms (sensu Dubois 2000) are used in the logonymies of genera and species herein, rather a complete logonymy is given including chresonyms.
in several reports and may end up as part of the popular culture. On Table 1, a few examples illustrate the wide use of misidentifi ed "Vonones" spp.

Taxonomic history of Libitioides
[1] The fi rst cosmetid (ornatus) There were already 54 species of harvestmen described in the world, when William Kirby (1819) described the fi rst three Laniatores Thorell, 1876 (two from Brazil, one from Uruguay), placing them in the new genus Gonyleptes Kirby, 1819. In the very next paper in the taxonomic story of Opiliones Sundevall, 1833, Thomas Say (1821) described the fi rst laniator from the USA, Gonyleptes ornatum Say, 1821. This was also the fi rst species described of what we now know as Cosmetidae. Say was, therefore, the fi rst to conceive what would later become the Laniatores (as we consider today), by recognizing the intimate relationship between what are now gonyleptids and cosmetids. Soon later, other researchers (Perty 1833;Hope 1836) proposed a different hypothesis (today refuted), by joining the cosmetids with what are now the Eupnoi Hansen & Sørensen, 1904.
Say described the new species Gonyleptes ornatum Say, 1821 (incidentally getting the grammatical infl ection incorrectly, in what should have been ornatus, masculine instead of neuter gender) from Cumberland Island, Georgia and East Florida. Say's description is detailed and accurate enough to allow this species to be recognized without mistake (Fig. 2a). The name 'ornatus' ('decorated'), as applied to a cosmetid, is adequate enough, but it has exceedingly low discriminatory value, and indeed, it was later used to name several unrelated species. This species was fi rst combined with Cosmetus Perty, 1833 only 50 years later by Butler (1873).
[2] The Texan caramel cosmetid (sayi) Almost half a century after the description of Gonyleptes ornatum, Horatio Wood (1868) recorded this species from Texas. Wood's redescription is clear enough to make evident it was an entirely different species, unarmed, "dark ferruginous" or light coffee colored with a pair of large diffuse "nearly black spots" on the posterior part of abdominal scutum (Fig. 2b). Table 1. Examples of misidentifi cation in reports of "Vonones" spp.

Original ID Our revised ID Locality Reference
Vonones sayi Pseudopucrolia discrepans (Roewer, 1943) [Gonyleptidae] Alagoas, northeastern Brazil danpostou ofi cial (2022) (Sørensen, 1884) Eleven years later, Simon (1879) noticed that this "Gonyleptes ornatum" of Wood was a misidentifi cation of a yet undescribed species. He strangely enough named this new species Cynorta sayi, in homage to Say (when the standard practice would have been to call it "woodi" after the person who did the misidentifi cation). Simon also transferred both American cosmetids to the genus Cynorta C.L. Koch, 1839, erected 40 years earlier for Brazilian species.
[3] The northernmost cosmetid (albolineatus) Only fi ve years after Simon had named Cynorta sayi, Sørensen (1884)  Latin description is enough to allow recognition of this species (Fig. 2d). By then, the recognition of three USA species was clear enough, so that Weed (1893) also transferred albolineatus to Cynorta and easily summarized the distinction of the three American species of Cynorta. Weed offered redescriptions of those Cynorta, providing some additional records -in the case of C. sayi he also mentioned a distinct morph (see [7]) with a "yellow marking on dorso-meson behind eye eminence, and a transverse line near posterior margin of abdominal scutum." In the case of C. albolineata, he also provided illustrations which match precisely Sørensen's description and also considerably expanded the known distribution of the species. As for C. ornata, Weed described the variation of the "canary-yellow" markings and noted it as "exceedingly abundant in the pine woods under logs." [4] The humble Cuban cousin (modestus) Banks (1909: 171) described the new species Vonones modestus Banks, 1909 from Cuba. This species was wrongly synonymized (by Roewer 1912, see [5]) with Libitioides ornata, a name used by Roewer (see [5]) which included all American cosmetids, and subsequently forgotten.
Four decades later Roewer (1947, see [7]) described a similar species from Cuba, placing it in Platycynorta Mello-Leitão, 1933, as P. secunda Roewer, 1947. In spite of Banks' type of V. modestus being lost, his description matches in all details the male holotype of P. secunda (Fig. 3).
This Cuban species is extremely similar to L. albolineata and only a specifi c focused study will be able to establish its status. This species does not closely resemble any other Cuban cosmetid, and the hypothesis of introduction should also be considered.
[7] The striped Texan morph (Platycynorta depressa) Sørensen (in Henriksen 1932) described the new species Cynorta depressa Sørensen, 1932 from "America centralis" based on males and females in the Dresden Museum. The arachnid collection in Dresden has been destroyed during World War II, but a subsample survived in ZMUC, which has been examined by ABK (Fig. 4). Henriksen (the editor of Sørensen's notes posthumously published) placed this species in the subgenus Cynorta, into which he synonymized a great number of generic nomina by  (Banks, 1909) Roewer, including Libitioides. He commented: "This species is nearly [sic] related to C. ornata Wood, described above, and is thus to be referred to the genus Libitioides in the Roewerian system in spite of the different armature of the third and fourth abdominal somites." Mello-Leitão (1933) made this Cynorta depressa the type of his new genus Platycynorta (Fig. 2c). The rationale for the separation of both genera was given as: "Area III tuberculis rotundis areae primae similibus munita" [Platycynorta] X "Area III processis conicis duobus armata" [Libitioides]. (Simon, 1879), morph C, syntype, ♂ (ZMUC). a. Habitus, dorsal view. b. Same, sinistrolateral view. c. Same, ventral view. d. Same, detail of carapace, dorsal view. e. Labels. Roewer (1947: 8) described Platycynorta secunda from Cuba. Our colleague Abel Pérez-González kindly shared his images of the holotype, and to our astonishment, we could not fi nd any difference with Sørensen's species C. albolineata from USA. We do not know if under more thorough examination this species proves different from L. albolineata (or even if the locality was mislabeled), therefore we simply associate this species with the Cuban Vonones modestus and make the transfer to Libitioides leaving it otherwise undisturbed. Roewer (1952: 42) described the new species Platycynorta transversalis Roewer, 1952 from Seagoville near Dallas, Texas. Roewer's illustration is surprisingly faithful (showing the sinuous omega-stripe) and his text clearly mentions the typical granulated retrolateral patch on femur IV of male. Cokendolpher & Jones (1991: 86) commented about the unsatisfactory state of the taxonomy of American cosmetids and synonymized Platycynorta transversalis with Vonones sayi.

Fig. 4. Libitioides sayi
Finally, Roewer (1957: 89) described yet another Platycynorta, P. clavifemur Roewer, 1957 from Peru. It does not have anything to do with any of the species treated here and is referred below to the genus Ambatoiella Mello-Leitão, 1943, to which we herein give an extended diagnosis.
[9] The Goodnights blender Goodnight & Goodnight (1953a) performed an important setback, clustering 64 genera of Cosmetidae into three genera, thus inexplicitly synonymizing Metacynorta with Vonones, only mentioning the combination Vonones ornata (instead of ornatus, see Kury (2003)) fi ve years later (Goodnight 1958). Goodnight overlooked the fact that Vonones is masculine, so the specifi c epithet should be infl ected to match it as Vonones ornatus. This error propagated for decades until Kury (2003) noticed and corrected it.
[10] What we are proposing herein We propose herein the recognition of three species of American Libitioides based on our interpretation of exclusive sets of morphological features. After discarding the spurious report of occurrence of Cosmetidae in California, we fi nd that the members of Libitioides occur only in the eastern portion of the USA, with clear climatic constraints. L. ornata occurs in the southeastern tip of this area (mostly the Florida Peninsula). The other two species are mostly allopatric, L. sayi occurs almost exclusively in Kansas, Oklahoma and Texas while L. albolineata occurs to the east of this area, up to the Atlantic coast.

Distribution and climate
The cosmetids of the genus Libitioides in the USA are distributed exclusively in the Köppen-Geiger climate regions Cfa and Dfa (Fig. 5). The letters C and D (at the beginning of the acronyms) stand respectively for mesothermal and microthermal climates. This strongly suggests that the factors determining the distribution of these species are not directly temperature, but the absence of a dry season (second letter "f") and the presence of a hot summer (third letter "a"). However, the Cosmetidae are conspicuously absent from the northern half of Dfa.

Recognition of morphs in Libitioides
We can recognize fi ve morphs for the American species treated herein (Fig. 6), two pairs of them (B, C and D, E) merge into each other. Their marked features are listed below: Morph A: Libitioides ornata (Fig. 6a). The strongly spined and profusely marked Floridan species. It seems endemic to the WWF ecoregion NA0529 (Southeastern conifer forests).
Morph E: "yellow Libitioides albolineata" (Fig. 6e). It occurs in the contact zone of L. sayi and L. albolineata, restricted to a small area mainly in the Gulf Coastal Plain of Texas. WWF ecoregions: NA0523 (Piney Woods forests) and NA0701 (Western Gulf coastal grasslands).
We are also marginally dealing with other species associated with Libitioides and its synonyms as well as the monotypic Calicynorta, mistakenly reported as being from the USA.
Platycynorta clavifemur does not share any special similarity with any Libitioides. Its distribution in NW Peru already hints at the alienness of it compared to Libitioides. We made an effort to relate this species to the described Andean genera, and found special similarity with Ambatoiella Mello-Leitão, 1943, so this species is below formally combined under that genus. Other genera from the Peruvian Andes such as Moselabius Roewer, 1956 andSocotabius Roewer, 1957 are surely closely related to Ambatoiella, and at least members of the Libitiinae, but in the absence of detailed information of those genera and of a phylogenetic analysis to test those hypotheses, we refrain from proposing a formal transfer. Calicynorta bimaculata is very similar to Holovonones compressus, a common south Mexican species. As it is known by a female only, we do not know if there are signifi cant contrasts between Costa Rican and Mexican/Guatemaltecan populations. There seems to be small differences in the scutal armature and the two longitudinal stripes. Therefore, we include C. bimaculata in Holovonones as a distinct species and assume that the Costa Rican H. compressus must be H. bimaculatus instead.

[11] Diagnostic characters within Libitioides
Because the easy distinction of L. ornata from the other Libitioides, we have tackled here only the issue of distinction of the western species (morphs B and C: L. sayi) from the eastern species (morphs D and E: L. albolineata) by listing the main features which can be used for that purpose.
1) Color background of dorsal scutum and legs (Fig. 7). This character sharply distinguishes two almost allopatric species: (a) western (morph B, C, but also E): yellow to rusty-orange background, with two fuzzy dark spots on scutal area III; trochanters I-IV yellow, gradually getting darker (olive) distally and (b) eastern (morph D): body and legs uniform dark chestnut brown, with only a faint hint of darker spots at area III. 2) Chevron (Fig. 8). We may distinguish three states of this character: [0] chevron entirely absent, [1] chevron reduced to a small median arch, [2] chevron complete, reaching laterals of carapace. The western species (morphs B, C) mainly lacks chevron or has it reduced (0 >> 1), while the eastern species (morphs D, E) mainly has well-developed chevron, sometimes reduced (2 >> 1).
3) Backbone (Fig. 9). We may distinguish four states of this character: of area III, sometimes broken in two, [4] a pair or arches almost formed, but each one severed in more than one place, [5] a pair of entire arches, but clearly separated in the middle, [6] omega stripe complete, left and right halves joined in the middle. The western species mostly has no omega stripe and even when present it is rarely complete (states 1 to 4, the more dissociated shapes, correspond to the L. depressa morph C) while the eastern species has only the two more complete stages (states 5 and 6). 5) Ribs (Fig. 11). We may distinguish three states of this character: [1] ribs absent, [2] ribs poorly defi ned, formed by a few dots on groove I, [3] ribs more well-formed by dashed lines on grooves I and II. The western species (morphs B, C) entirely lacks ribs while the eastern species (morphs D, E) has all degrees of dissociation from entire ribs to none.
6) Markings on lateral areas. This character is exclusively present in the eastern species (morphs D, E) and it is very variable, from absent in both sides to present in a single side (asymmetric).   Goodnight & Goodnight (1953a: 60); synonymy rejected by Kury (2003: 37)]. Type-species by original designation: Ambatoiella vigilans Mello-Leitão, 1943.

Diagnosis
Scutum outline alpha-type, with much elongated coda, lateral border in posterior view following the general body curvature. Protoglyph guards are very robust, triangular. Mesotergal grooves are shallow, straight. All scutal areas unarmed, III-IV however each with a transverse row of minute tubercles. Areas I-IV each with a pair of faintly colored rounded fl ecks. Cheliceral hand is not swollen, but basichelicerite of male is slightly stouter, with coarsely tuberculate posterior rim. Legs very short (all femora shorter than scutum). Coxa IV oblique, not surpassing coda in situ and projected laterally in dorsal view, with reduced apical apophysis and with groin warts. Femur IV sexually dimorphic, either with strongly clavate prolateral crest on proximal half (in A. clavifemur) or with a row of strong proventral spines increasing in size posteriorly (in A. vigilans). Basitarsus I thickened in male. Tarsal counts: 5(3)/7-10(3)/6/6. Male genitalia (only known from A. vigilans): VP (ventral plate) latero-distal corners wide, projected. Wattle is complete. MS (macrosetae) D1 is clearly smaller than C. Two long lateral patches of microsetae T4 without midfi eld.

Comment
The species does not match the diagnosis of Libitioides (which in principle the former species of Platycynorta are expected to be assigned to), and in turn it matches the diagnosis of Ambatoiella. Emended diagnoses for both genera are given here so they can be compared. Mello-Leitão, 1943 Fig. 12a

Comment
Ambatoiella vigilans and V. sexpunctata are nearly sympatric, both coming from Tungurahua State in Ecuador. The description and illustrations by Roewer are detailed enough to allow recognition of both being the same species.

Diagnosis
Scutum outline beta-type, scutum in lateral view growing strongly convex backwards. Protoglyph guards are small, blunt triangular. Scutal areas I, III and IV each with a pair of acuminate tubercles, growing larger backwards, so that those of area IV are clearly larger. Cheliceral hand is not swollen, but basichelicerite of male is clearly stouter, with a coarsely tuberculate posterior rim. Coxa IV with oblique orientation. Femur IV clavate and keeled only distally, pro-ventrally armed with a comb of tubercles somewhat fused to each other in the base. Basitarsus I thickened in male. Tarsal counts: 6(3)/14-17(3)/7-8/8-11. Male genitalia: VP subrectangular. Wattle is complete. MS D1 is almost as large as C. Two long lateral patches of microsetae T4 without midfi eld.

Combined distribution
Bahamas, SE tip of Florida, Greater Antilles, except Puerto Rico.

Comment
Besides the species of Libitioides, there are unpublished records which should probably belong to C. quadrispinosa from the southeastern tip of Florida, around Miami City. The more typical species of Cynortoides hail from Cuba, while there are at least four others in Jamaica and Hispaniola which resemble more closely this Floridan cosmetid. Those Jamaican/Haitian/Dominican species are poorly studied, some only known from the female holotypes, therefore, it is not possible for now to use masculine sexually dimorphic features to look for identifi cation clues. Due to the gap in distribution between these species, it is likely that the USA specimens are introduced from the Greater Antilles.

Diagnosis
Scutum outline alpha-type, scutum fl attened in lateral view, lateral border in posterior view following the general body curvature. Protoglyph guards are triangular. Mesotergal grooves are shallow, straight.
Scutal areas I-IV are each armed with a pair of small fl at tubercles which are white, sharply contrasting with the brown background, area V with a transverse row of a few such tubercles. Cheliceral hand is somewhat swollen in male, and the basichelicerite of male is clearly stouter, with a coarsely tuberculate posterior rim. Coxa IV with oblique orientation. Femur IV sexually dimorphic, clavate in males, in female thinner and curved proximally. Basitarsus I thickened in male. Tarsal counts: 5(3)/8-9(3)/6-7/7. Male genitalia: VP subrectangular. Wattle is complete. MS D1 is clearly smaller than C. Two long lateral patches of microsetae T4 without midfi eld.

Combined distribution
SE Mexico, Guatemala and Belize, with an isolated occurrence in Costa Rica.

Comment
Holovonones biangulatus is a species from southern Guatemala and adjacent part of the Mexican state of Chiapas. It is recognizable by the L-shaped stripes on the laterals of the abdominal scutum, contrasting with the thicker arches of H. bimaculatus and H. compressus. No male has been reported so far. (Banks, 1893)

Comments on type data
Specimen originally reported by Banks as from "California, San Diego". On MCZ website there are images of the holotype and the labels. One of them says clearly "San Jose", and the specimen sheet reads "COSTA RICA, San José". Neither San Diego nor California are mentioned anywhere in the labels or the associate data. Also in the same website it is said [incorrectly]: "there is no locality label with specimen". Therefore, the ascription of "Costa Rica" to this locality "San José" is doubtful. There is no record of anything similar from Costa Rica. San José must be one of the commonest toponyms in Latin America. Given the great similarity of the images of the female holotype with H. compressus, and if this "San José" refers to a toponym in Mexico or Guatemala, this species could well be a senior synonym of Holovonones compressus. However, with such fragmentary evidence, we refrain from doing this synonymy here, especially because Holovonones compressus is a much more widely used name. (Pickard-Cambridge, 1904) Fig. 12d

Diagnosis
Scutum outline lambda-type, scutum fl attened in lateral view, lateral border in posterior view clearly marked by a deep groove and raised to break the outline of scutum. Protoglyph guards are blunt triangular. Mesotergal grooves are obsolete. Scutal areas I, III and IV each with a pair of blunt tubercles (L. ornata: area III with robust spines tilted backwards and IV unarmed). Cheliceral hand and basichelicerite are sexually monomorphic. Legs moderately elongate (femora II and IV barely longer than scutum). Coxa and trochanter IV of male almost entirely parallel to the main body axis, so that femur IV lies straight, while in females the coxa/trochanter are inserted more obliquely and the femur is slanted. Coxa IV surpasses coda in situ only by its apical region, with well developed proapical apophysis and without groin warts. Femur IV sexually dimorphic, in male more thickly granulous and straight, in female thinner and curved proximally. Basitarsus I thickened in male. Tarsal counts: 5(3)/8-9(3)/6/6. Male genitalia: VP subrectangular. Wattle extends from the middle to the tip of the stylus. MS D1 is almost as large as C. Two long lateral patches of microsetae T4 without midfi eld.

Spurious included species
Because the relevant genera were heterogeneous to begin with, there was a legacy of species of uncertain taxonomic position and which are all treated here: Platycynorta secunda Roewer, 1947 (Cuba), Platycynorta clavifemur Roewer, 1957 (Peru), Libitioides riveti Roewer, 1914 (Ecuador).

Combined distribution
Cuba, NE tip of Mexico, eastern/southeastern USA (map in Fig. 13).

Comment on relationships
Libitioides resolved as the sister-group of the Taitoinae in the phylogenetic analysis of Medrano et al. (2021). The authors, however, then refrained from including it in this subfamily because of the low resampling support, choosing instead to present a tighter Taitoinae (better supported by synapomorphies) as opposed to the "quasi-taitoine" with Libitioides. We here note that Libitioides seems to be also very closely related to the genera Paravonones Pickard-Cambridge, 1904 and Boneta Goodnight & Goodnight, 1944, which currently include six nominal species from Mexico, El Salvador and Guatemala and were not included in that analysis. Both genera share the scutal outline and armature, the vaulted laterals of dorsal scutum, monomorphic chelicerae, the sexually dimorphic insertion of coxa and trochanter IV. The species of Paravonones contrast with Libitioides by possessing more sinuous leg IV on males, with stronger armature. The type species of Boneta is very similar to Libitioides, contrasting with it by having remarkably granulated coxae.

Diagnosis
Libitioides albolineata differs from the other species of Libitioides by the following characters: short legs (femur IV shorter than or equal to DS length), small rounded tubercles in area III (instead of spines), abdomen dorsally round without coda and yellow spots of DS with backbone and ribs (frequently fragmented) and may have lateral clouds. General color of body and appendages uniform brown or rarely light yellowish brown.

Diagnosis
Libitioides modesta is strongly similar to L. albolineata, only differing from it by the two paramedian granules in area IV of abdomen (a row of granules in L albolineata).

Diagnosis
Libitioides ornata differs from the other species of Libitioides by the following characters: only moderately elongated legs (femur IV longer than DS length), high spines in area III slanted backwards (instead of rounded tubercles) and small coda. The yellow spots of DS with complete chevron and omega stripe, and backbone and ribs frequently fragmented. Lateral clouds are always present. General color of body and appendages uniformly brown.

Type data
Types of Gonyleptes ornatum USA • Georgia, Cumberland Island and Florida, Eastern portion; Cabinet of the Academy of Natural Sciences of Philadelphia (not examined).

Comment on the combination Vonones ornata
Whenever this specifi c epithet is combined under Vonones as Vonones ornata, it is incorrect from the nomenclatural viewpoint, because Vonones being a masculine noun, the specifi c epithet should be infl ected as ornatus (mandatory infl ection of adjective to agree in gender with the neuter generic name; ICZN Code Art. 34.2).

Distribution
Species known from the USA Atlantic coastal plain, including the coast of the Gulf of Mexico. WWF Terrestrial Ecoregions: predominant occurrence in the Southeastern conifer forests (NA0529), but absent from the Everglades fl ooded grasslands (NT0904).

Diagnosis
Libitioides sayi differs from the other species of Libitioides by the following characters: short legs (femur IV shorter than or equal than DS length), small rounded tubercles in area III (instead of spines), small coda. Frequently without any spots in DS, chevron and omega stripe may be present (frequently fragmented). Ribs, backbone and lateral clouds never present. General color of body and appendages mustard yellow with a pair of much darker spots on scutal area III.

Type data
Types of Cynorta sayi USA • Texas, without further locality data; whereabouts unknown (not examined).

Diagnosis
May be separated from its congeners by the coarsely granulated integument, area I with paramedian rounded granules, area III with slanted spines and unarmed area IV. Tibia and metatarsus with retrolateral row of small spines.

Distribution
Only known from the type locality. WWF Terrestrial Ecoregion: Trans-Mexican Volcanic Belt pine-oak forests (NT0310).

Diagnosis
See Medrano et al. (2022: 206). Roewer, 1914 Rhaucoides Riveti Roewer, 1914: 125, pl Roewer (1914) studied the harvestmen collected by French ethnologist Paul Rivet (1876-1958 in two campaigns in the northern páramos of Ecuador (1901)(1902)(1903). The material was offi cially split: one part was to be deposited in the MNHN, Paris and subsamples were taken to the Roewer Collection #1. As it happened, it seems that not all material was duly returned to Paris, but rather deposited in Roewer's collection, from where it ultimately ended in SMF, because it was not located by occasion of ABK's visit to Paris in 2017. Roewer mentioned only 1 male for R. riveti, but the type series (SMF RI 477) consists of two males. Maybe because of that indication, SMF curators called one of them "holotype" and the other "non-type". As for L. riveti, the paper mentions "1 male, 1 juvenile", of which, presumably the male would be in Paris. The juvenile is in SMF (SMF RI 313).

Remarks
Both external and genital morphology are conserved and well delimited in Rhaucoides members and L. riveti matches the genus diagnosis at least in external morphology. Decoloration is common in preserved opilions, and more common as in the ancient material studied by Roewer. Despite that, the juvenile syntype in SMF shows a pattern somewhat similar to R. riveti, a species described by Roewer with material collected by Rivet himself. Here, we propose the synonymy of Libitioides riveti with Rhaucoides riveti based on the morphological similarity and the geographic proximity (both type localities are in mountain tops 20 kilometers apart). Likewise, that decision avoids the creation of a secondary homonymy within Rhaucoides and the necessity of a replacement name.

Discussion
The combination of museum specimens and published photographs was essential for accurately identifying the different species of Libitioides and determine their distributions. By carefully selecting easily recognizable characters that could be evaluated from online photographs and mapping these characters individually, we were able to establish a combination of diagnostic characters for each species, particularly L. sayi and L. albolineata. This approach proved effective in accurately distinguishing these species. This method is likely only applicable to Cosmetidae due to the distinctive white markings on the dorsal scutum, which can be easily observed in photographs. It may not be effective for identifying other families of Laniatores.
We believe that American species of Libitioides are now adequately represented in our study, with neglected older names considered and synonymy clarifi ed. Further investigation is needed for L. scabrissima, as it is poorly studied and has limited sampling. It is unlikely to also occur in the USA, specifi cally southern Texas, unless it is locally sympatric with the other two known species in that region. The American Libitioides serve as a prime example of a subject that could benefi t from genetic analysis to better understand our fi ndings, as the plastic phenotypic characters can obscure clear distinctions.

Libitioides ornata Roewer, 1912 vs Cynorta sayi Simon, 1879
Wood (1868) misidentifi ed an as yet undescribed species as "Gonyleptes ornatum", and this misidentifi cation was already recognized by Simon (1879), who proposed a name for that species: "L'espèce rapportée par M.H. Wood au Gonyleptes ornatus Say, paraît en différer complètement et nous proposons de l'appeler C. Sayi." Roewer (1912) corroborated Simon's hypothesis that Wood's material did not refer to Gonyleptes ornatus Say 1821, and correctly considered that it deserved a nomenclatural status of its own. However, Roewer proposed yet another new name -Libitioides ornata -keeping the same specifi c epithet used by Wood for a species in another genus (a procedure that he would adopt several times, often creating confusion, as shown in Kury 2020).
Both ICZN Art. 11.10 (Deliberate employment of misidentifi cations) and Art. 50.1 (Identity of authors) indicate that Roewer, by deliberately using Wood's misidentifi cation of Gonyleptes ornatus for the type species of the new nominal genus-group taxon Libitioides Roewer, 1912 denoted a new nominal species Libitioides ornata Roewer, 1912 (in spite of Roewer's wrong assumption that Wood was to be the author). This nomen competes for synonymy where it is superseded by Cynorta sayi Simon, 1879. The male holotype of L. scabrissima does not entirely match the general features of the remnant species of Libitioides, especially by the coarsely granular body, the coxae IV not parallel to the main body axis and the short, and the arched male femur IV.

The type locality of Cynorta depressa
In spite of the original data placing this species in Central America, we believe that it originated from a mislabeling. As demonstrated above, this specifi c Morph C possesses a narrow distribution in eastern Texas.