The history of the families which now compose the Samooidea and Zalmoxoidea is rather complex, with many generic transfers between them. Older authors such as Sørensen and Thorell described isolated families mainly from SE Asia. Roewer (1912) considered all of them to be subfamilies of an immense and meaningless Phalangodidae or even merged some in Phalangodinae. In an unpublished dissertation Kury (1993) recognized a large superfamily Zalmoxoidea, including Biantidae, Minuidae, Podoctidae, Samoidae, Stygnommatidae and Zalmoxidae, but this suprafamilial name only appeared in press much later (Kury and Cokendolpher 2000). In Kury’s catalogue (Kury 2003), the families of Grassatores were not included in superfamilies. Only in a synoptic work by Giribet and Kury (2007), were the superfamilies of Grassatores explicitly defined for the first time. They recognized Zalmoxoidea as (Icaleptidae, Guasiniidae, Zalmoxidae, Fissiphalliidae), while the Samooidea included (Samoidae, Podoctidae, Biantidae, Minuidae, Stygnommatidae). There was a lapse with the name of this latter superfamily being called either Samooidea or Biantoidea in the same book. This happened because composition was uncertain, and in the last minute Samoidae was also included, being older than Biantidae. Giribet et al. (2010) recovered a paraphyletic Samooidea (Biantidae, Escadabiidae, Kimulidae, Samoidae, Stygnommatidae) respective to the Zalmoxoidea, while they included in Zalmoxoidea the same Fissiphalliidae, Guasiniidae, Icaleptidae, and Zalmoxidae, with a paraphyletic Samooidea as sister group. In that paper they also found that Podoctidae should be placed far from both, isolated among the basal Grassatores. Later, Sharma and Giribet (2011) recovered a reduced monophyletic Samooidea with only 3 families (Biantidae, Samoidae and Stygnommatidae) while they augmented Zalmoxoidea transfering Escadabiidae and Kimulidae to it, resulting in 6 included families. Recent molecular analyses recovered a sister-group relationship between Samooidea and Zalmoxoidea (e.g., Sharma and Giribet 2011, with BS = 85% and PP = 1.00). This hypothesis is also supported by morphology (Pérez-González and Kury 2007b). However, the support for each superfamily is unequal. On one hand we have evidence for a strong Zalmoxoidea sensu Sharma and Giribet (Sharma and Giribet 2011) (BS = 89%, PP = 1.00) whereas Samooidea it is recovered as monophyletic but with limited support (BS = 53%, PP = 0.62) (Sharma and Giribet 2011). That could be a reflection that whereas the taxon sampling for Zalmoxoidea has been greatly enhanced lately (Sharma and Giribet 2011, Sharma and Giribet 2012) the taxon sampling for Samooidea still remains unsatisfactory. More effort is needed in studying and sequencing African and Indo-Pacific lineages (Sharma and Giribet 2011), as well as enhancing the representation of Neotropical and Australasian terminals in order to test Samooidea as a natural group and improve the hypotheses of relationship inside the Samooidea/Zalmoxoidea clade.

Escadabiidae. This small Brazilian family was created by Kury & Pérez (in Kury 2003) for 4 genera of Phalangodinae from northeastern Brazil. Kury and Pérez-González (2007) also included Spaeleoleptes H. Soares, 1966, which was originally in Phalangodidae Minuinae and latter in incertae sedis as per Kury (2003). Kury et al. (2010) further included the monotypic Brotasus Roewer, 1928, then in Grassatores incertae sedis, but originally in Phalangodidae, Phalangodinae. Escadabiidae was placed in Samooidea by Giribet and Kury (2007), Kury and Pérez-González (2007) and Giribet et al. (2010), but finally it was moved to Zalmoxoidea (Sharma and Giribet 2011), a transfer which has been subsequently supported (Sharma and Giribet 2014), although monophyly remains uncertain. Sharma and Giribet (2012) did not recover Baculigerus H. Soares, 1979 as monophyletic.

Fissiphalliidae. Martens (1988) erected this micro-diverse family for three new species placed into the single genus Fissiphallius, all from the vicinity of Bogotá, Colombia. He discussed other families in Grassatores, but allied Fissiphalliidae most closely with the Podoctidae (a group now considered to be phylogenetically distant from both Zalmoxoidea and Samooidea). The almost simultaneous publication of the resurrection of the Zalmoxidae (Staręga 1989) must have prevented any comparison between each other. Kury and Pérez-González (2002) included Fissiphalliidae in a clade along with Zalmoxidae and Icaleptidae, and commented that keeping Fissiphalliidae as a distinct family could render Zalmoxidae paraphyletic. Giribet and Kury (2007) formally included Fissiphalliidae in Zalmoxoidea closest to Zalmoxidae, and molecular analyses (e.g. Sharma and Giribet 2011, 2015) corroborated this view. In the last decade four other species have been described from Brazil, and placed in the same original genus, but without any groundbreaking phylogenetic discussion.

Guasiniidae. González-Sponga (1997) proposed this new micro-diverse family for two Venezuelan species, with one genus from Isla Guasina at sea level, another from a tepui (tepuis are table-top mountains found in the Guiana Highlands of South America, especially in Venezuela and western Guyana), 1350 m high. He compared it chiefly with Oncopodidae, without proposing any serious phylogenetic discussion, although the Oncopodidae (now Sandokanidae) are currently considered as an evolutionarily distant and archaic lineage. Pinto-da-Rocha and Kury (2003) added a third species from Brazilian Amazonia, and related Guasiniidae with the Zalmoxidae, Fissiphalliidae and Icaleptidae, a clade established shortly before (Kury and Pérez-González 2002). Giribet and Kury (2007) formalized its inclusion in Zalmoxoidea, a placement subsequently supported by molecular and mophological studies.

Icaleptidae. Another micro-diverse family, with only two described Andean species in two genera. In the original description, Kury and Pérez-González (2002) stated: “Among Grassatores, Icaleptidae is most closely related to Zalmoxidae and Fissiphalliidae”. Although they did not cite any superfamilial assignment. Icaleptidae was assigned to Zalmoxoidea very early (Giribet and Kury 2007) and this placement has never been challenged. Icaleptidae was found sister to one unidentified species of the genus Costabrimma from Costa Rica, and of uncertain affinities by Giribet et al. (2010). Many unidentified "Icaleptes" sp. are nested with two Costabrimma in Sharma and Giribet (2012). However, the flea like IV leg used to identify is not anymore a diagnostic character for this family, and it is present in several other Zalmoxoidea taxa, therefore the accurate familial identification of specimens in those later studies needs to be confirmed using genital characters together with other morphological features. Costabrimma may represent a lineage of Zalmoxoidea, not yet studied under a revisionary morphological viewpoint. This genus has not been used in the most recent analysis of Sharma and Giribet (2011), where Icaleptidae appeared as sister group to Zalmoxidae + Fissiphalliidae

Kimulidae. Sørensen in Henriksen (1932) created this family as Minuidae with 7 new genera, mostly from Venezuela and only one from southern Brazil. Mello-Leitão (Mello-Leitão 1933, Mello-Leitão 1938) subdivided some of Sørensen’s genera and removed Microminua and Minuides from Minuinae. Roewer never cited Minuinae either as a separate family or as a subfamily of Phalangodidae. H. Soares (1966) expanded the Brazilian representation of Minuinae, describing two new genera from southeastern Brazil. Šilhavý (1978) and González-Sponga (1987) considered both subfamilies Minuinae and Minuidinae as synonyms of Phalangodinae. Kury (1995) started to shrink the Minuidae by removing the original southern Brazilian genus Phera to the Gonyleptidae. Kury (2003) reinstated Minuidae as a separate family, expanding it with the inclusion of some Caribbean and Venezuelan genera of Phalangodinae, and both genera of Minuidinae, while removing 3 other genera, including the two remnant Brazilian genera. In 2007 Pérez-González, Kury, and Alonso-Zarazaga (in Pérez-González and Kury 2007a) detected that the type genus was invalid due to homonymy (replaced by Minuella) and were forced to change the family name to Kimulidae. Pérez-González and Kury (2007a) newly included Tegipiolus from NE Brazil in Kimulidae. They also transferred Minuides to the Zalmoxidae, automatically carrying the synonymy of the Minuidinae. Kimula has been chosen as type genus, leading to the new name of this small family, now comprised of nine genera.

Zalmoxidae. Sørensen (1886) created the new family Zalmoxioidae only for the new genus Zalmoxis, with two species from Fiji. Soon afterwards, Thorell (1889) synonymized this with family Epedanoidae and three years later, Loman (1902) included Zalmoxis in an expanded Epedanidae with many genera. Roewer (1912) included Zalmoxis as a genus of his Phalangodinae within the immense family Phalangodidae. During eight decades, Zalmoxis and closely related genera remained buried in Phalangodinae while other poor arrangements were made regarding the small Grassatores, and while a legion of would-be Zalmoxidae were described under Phalangodinae. Mello-Leitão (1933) erected the subfamily Minuidinae in Phalangodidae, but later (Mello-Leitão 1938), he wrote: “Minuidinae (n. subfam.)”. This taxon (Mello-Leitão 1938) was formed by two genera sorted from Sørensen’s Minuidae — Minuides and Pseudominua. H. Soares (1972) created Stygnoleptinae as a new subfamily of Gonyleptidae containing 3 entirely distant genera – Glysteroides Roewer, 1943 (currently in Gonyleptidae), Saramacia Roewer, 1913 (currently in Manaosbiidae), and Stygnoleptes Banks, 1914 (currently in Zalmoxidae). Staręga (1989) finally resurrected Zalmoxidae to include 5 Paleotropical genera, remarking that a great number of genera had been sunken into Zalmoxis by Goodnight and Goodnight (1957). Staręga (1992) included 3 more Afrotropical genera from Seychelles and Madagascar. Kury (1997) dismantled the Stygnoleptinae, synonymizing this name with Zalmoxidae, and proceeded in subsequent studies (Kury and Cokendolpher 2000, Kury 2003) to promote a mass exodus from the Phalangodidae into Zalmoxidae, to expand into what is now the largest family of Zalmoxoidea. Pérez-González and Kury (2007a) presented evidence that Minuides is a Zalmoxidae, therefore Minuidinae is a synonym of this family. Sharma and Giribet (2012) demonstrated that the family Zalmoxidae, “similar to the Pacific iguanas, constitutes the unusual case of a lineage of Neotropical origin that colonized the Indo-Pacific, likely by ancient transoceanic dispersal during the Late Cretaceous.” Accordingly, Sharma and collaborators (Sharma et al. 2011a, Sharma 2012, Sharma et al. 2012) while describing many new Paleotropical species, sought to merge all Old World genera of Zalmoxidae to better accommodate this single origin within a wider familial clade,which is much more diverse in the Neotropics.

Biantidae. Karsch (1880) described the genus Hinzuanius from the Comoros Islands in the family Gonyleptoidae. Pavesi (1884) described a second species from Ethiopia. Simon (1885) described the new genus Biantes in the Phalangodidae, but compared it to Hinzuanius and Stygnus, both then in Gonyleptidae. Biantes originally included one species from India, and another from Madagascar. Thorell (1889), in a work on Burmese harvestmen, created the family Biantoidae to include only Biantes (known previously from India, Madagascar and from then on also Burma) and Hinzuanius (this one treated only briefly because it was outside the area of study, being an African genus). Soon afterwards, he (Thorell 1891b, Thorell 1891a) repeated the composition of Biantoidea restricted to Biantes alone. Sørensen (1896) confirmed the presence of the family in continental Africa,adding the new genus Lacurbs (from Cameroon), and including most of his EpedanoidaeSørensen 1886 (specifically, the Australasian Ibalonius and Mesoceras, plus the African Hinzuanius). Loman (1898) expanded the known distribution of Biantes in the Afrotropics, describing two new species from South Africa. Loman (1901) made a summary of the distribution of the family, mentioning the disjunctive areas in Africa (Cameroon, Abyssinia, many places in southern and eastern Africa and Madagascar) and SE Asia from Sri Lanka to Sumatra. Loman (1902) placed Ibalonius and Mesoceras elsewhere, created the new genus Acudorsum from the Seychelles and synonymized Biantes with Hinzuanius, thus recognizing only three genera Acudorsum, Hinzuanius and Lacurbs as Biantidae. Pocock (1902) followed the generic composition by Loman, but mistakenly thought that the family should take its name from the oldest genus, thus replacing Biantidae with Hinzuanidae. Roewer (1912) downgraded Biantidae to a subfamily of Phalangodidae, correctly using Biantinae instead of Hinzuaninae for the subfamily name. He revalidated Biantes from the synonymy of Hinzuanius and created the new genus Heterolacurbs from “Togo” as well as a few species in the other genera. From then on, during decades the Biantinae were gradually expanded with descriptions of new genera and species (e.g, Roewer 1923; Lawrence 1965; Martens 1978) without changes in the concept of the subfamily. Henriksen (1932) basically kept Roewer’s arrangement, only raising subfamilies and families one rank, so his Phalangodoidea included a family Biantidae, which was equivalent to Biantinae and not meant as a rupture with Roewer’s system. For many years the Biantidae were the same as today’s Biantinae, until Mello-Leitão (1938) included in Biantidae three of Roewer’s (Roewer 1923) subfamilies of Phalangodidae — the Biantinae, Stygnommatinae and Dibuninae. This proposal went universally ignored in the 1940s-60s, and those continued to be treated as subfamilies of Phalangodidae (e.g, Roewer 1949 for Stygnommatinae; Kauri 1961 for Biantinae). Lawrence (1959) created the new subfamily Lacurbsinae without detailed explanations (i.e. did not treat it in his paper, because it was only concerned with fauna of Madagascar). This subfamily went largely ignored until being resurrected by Kury (2003). Šilhavý (1973) was the first to observe Mello-Leitão’s proposal, recognizing for the first time in 60 years the Biantidae as a separated family, including the Biantinae, Stygnommatinae and Dibuninae (as in Mello-Leitão 1938) plus the new Caribbiantinae Šilhavý 1973for Caribbean species. Martens (1978) significantly treated the Biantidae as a separate family, solidifying the standard from then on. Also, following Šilhavý (1973), Suzuki (1977) and Kauri (1985) accepted Dibuninae as a biantid. Kury (2003) was the first (reaffirmed in Kury 2007) to assign Dibuninae to the Epedanidae, removing them from the Biantidae. Kauri (1985) added the monotypic subfamily Zairebiantinae from Central Africa. Pinto-da-Rocha (1995) discovered the Caribbiantinae also had representatives in South America, having been previously described by Roewer as Stenostygninae, hitherto regarded as a subfamily of Stygnidae (or Gonyleptidae), which had priority over Šilhavý’s family group name.

Samoidae. Samoidae was described in a paper where Sørensen (1886) created many other familial subdivisions in the Laniatores. It initially included Badessa, Feretrius and Samoa, all from Pacific islands. Roewer (1912) proposed a retrograde classification which endured for decades, recognizing a huge meaningless Phalangodidae including many of Sørensen’s families. So Samoidae became Phalangodidae Samoinae. Roewer also added to Samoinae the genus Mitraceras previously described from the Seychelles by Loman (1902) in Assamiidae. Roewer (1933) described the 5th genus of the subfamily, Psyctrapus from Costa Rica, as the first acknowledged Neotropical Samoinae. Roewer (1949) expanded considerably the Samoinae, with addition of Microconomma Roewer, 1915 (from Cameroon, originally placed in Phalangodinae) and the new genera Badessania (from the Australian continent), Sawaiellus (from Samoa), Waigeucola (from Indonesian Papua) and Maracaynatum (from Venezuela), the latter as the first South American Samoinae. Goodnight and Goodnight (1957) then described Parasamoa from Micronesia, while Lawrence (1959) described three new genera from Madagascar: Anaceros, Hovanoceros and Malgaceros, and Roewer (1963) added the Australasian Fijicolana. Šilhavý (1977) described two new genera – Arganotus and Akdalima – from Mexican caves (Arganotus known also from epigean milieu). Soon afterwards, Šilhavý (1979) described four new genera from the Caribbean (Hummelinckiolus, Orsa, Reventula and Vlachiolus), mentioning the other existing genera, but overlooking the ones from Madagascar. Goodnight and Goodnight (1983) transferred Pellobunus Banks, 1905 from Costa Rica to Samoinae, synonymizing Psyctrapus with it. Rambla (1984) studied material from the Seychelles, added a new Mitraceras, the first Seychellan Samoa and the new genus Benoitinus (with the first anophthalm species). Staręga (1989) placed Microconomma in an undescribed family, which only much later was formally described as Pyramidopidae (Sharma et al. 2011b). Critically, Staręga (1992) restored Samoidae as a family and considered it to be closest to Biantidae than to Phalangodidae. He also removed Anaceros to the Biantidae and transferred Tetebius Roewer, 1949 from the Phalangodidae to the Samoidae. Microconomma reappeared in his list of Samoidae without comment. Kury (2003) followed Staręga and considered Samoidae as a family. He also transferred to Samoidae Cornigera González-Sponga, 1987 and Neocynortina Goodnight & Goodnight, 1983 from Phalangodinae and Kalominua Sørensen, 1932 from Minuidae. Pérez-González and Kury (2007b) noted that typical samoids are restricted to Polynesia, Melanesia, Australia, Mexico, the West Indies, and Venezuela and cast doubt upon the samoid kinship of the Indonesian and African species. They also transferred the Australian Zalmoxista Roewer, 1949 from Phalangodinae to Samoidae. Currently this family is not subdivided into subfamilies.

Stygnommatidae. Roewer (1923) created Stygnommatinae as a subfamily of Phalangodidae, to include only the monotypic genus Stygnomma Roewer, 1912, previously in Phalangodinae. Later, Roewer (1927) added another monotypic genus, Stygnomimus, from the Riau Archipelago. Roewer (1928) erected the new genus Stygnommatiplus for two species previously described from Costa Rica and Panama and also Zygobunus, originally in Gonyleptidae, resulting in 4 genera. Roewer (1933) added a 5th genus, Poascola, from Costa Rica. Meanwhile, both attemps to elevate Stygnommatinae to family (Henriksen 1932 and Mello-Leitão 1949) were ignored. Goodnight & Goodnight added Antagona from Puerto Rico (Goodnight and Goodnight 1942) and Flaccus from Mexico (Goodnight and Goodnight 1947). Goodnight and Goodnight (1951) then perpetrated a big step backwards, by synonymizing all genera into Stygnomma, including that in Phalangodinae, and merging five different species under the nameStygnomma fuhrmanni. In the 1970s and 1980s, several new species of “Stygnomma” were described just adding diversity, but without any change in the composition of the group and keeping it in Phalangodinae (e.g. Rambla 1976; Soares and Avram 1981). González-Sponga (1987) finally restored Stygnommatinae as a subfamily of Phalangodidae, from the synonymy of Phalangodinae. In his Ph.D. thesis, Kury (1993) restored Stygnommatidae to family, although this change officialy took years to appear in press (Kury and Cokendolpher 2000). Pérez-González (2006) produced a Ph.D. thesis focusing on this family, and proposing many important changes, although haven't yet appeared in print through other articles.

As part of the present project we made an enormous effort in order to solve the huge amount of taxa incertae sedis that are currently included in the Grassatores. A total of 32 taxonomic changes were proposed in the companion paper to this contribution (Kury and Pérez-González 2015), but a considerable amount of taxa still remain without any suprafamiliar allocation. A total of 59 genera and 81 species are not currently clearly assigned to any one grassatorean family. From this total, only the genus Phalangodella with seven species is considered as a Zalmoxoidea incertae sedis, while the rest remain for the moment simply as Grassatores incertae sedis. The major reason to retain that amount of taxa as uncertain is because we do not yet have feasible evidence to support either their familial or superfamilial assignment. The most common reason responsible for this picture is the absence of enough detailed, accurately illustrated descriptions for both sexes (e.g. omitted in Roewer 1923, Roewer 1949) and further need of detailed study of the male genitalia. In modern papers, the male genitalia became a mandatory set of characters to be included in taxonomical descriptions of Opiliones, and their importance for familial allocation have been continuously proved and reinforced (e.g. Pérez-González 2011; Pinto-da-Rocha et al. 2012, Kury 2014). But in some cases, even addition of the morphological information from the male genitalia is not enough to decide the familial placement (e.g. Phalangodella spp.). This is an indication that our current knowledge about Opiliones diversity still is unsatisfactory and evidence of this is the great amount of new families proposed in the last decades (e.g.Kury 2014, Kury and Villarreal M. 2015). New discoveries mainly arise from the study of the tiny, cryptic and litter-dwelling harvestmen fauna. Other familial groups of Opiliones likely still remain to be described or reconsidered, and an important aspect of this will be the study of the taxa here still considered as incertae sedis. Molecular works have been playing an important role in pointing out some research targets and systematic steering. Some interesting insights about several remaining incertae sedis have been pointed by the most recent molecular study of many lineages discussed here (e.g. Sharma and Giribet 2012) such as Costabrimma sp. nested inside Icaleptes spp. (therefore rendering Icaleptes as paraphyletic), Turquinia cf. montana recovered as part of Zalmoxoidea, as well as Parascotolemon, Phalangodella and Urachiche nested outside of Zalmoxidae. Undoubtedly the future points to the need for much further work to improve the systematics of Samooidea/Zalmoxoidea. We need to make the taxon sampling denser, explicitly include the name-bearing taxa, unveil the taxonomical identity of the terminals used in the previous molecular studies, and all of those together alongside detailed morphological descriptive work.We expect synergic action and feedback between molecular and morphological based research will increase our knowledge of harvestmen systematics in a near future, and help reduce the amount of remaining incertae sedis in the group.