Philodromidae Thorell, 1870 currently comprises 31 genera, of which 18 occur in the Neotropics (WSC 2020) and only Berlandiella Mello-Leitão, 1929 and Gephyrellula Strand, 1932 have been revised (Lise and Silva 2011, Santos and Rheims 2018). Berlandiella was proposed to allocate three species from Brazil: B. insignis Mello-Leitão, 1929 (type species), B. magna Mello-Leitão, 1929 and B. polyacantha Mello-Leitão, 1929. Lise and Silva (2011) did the taxonomic revision of the genus, re-describing and designating lectotypes for the known species and describing three new species, all recorded for Brazil, as follow: B. insignis, B. magna, B. polyacantha, B. meridionalis Lise & Silva, 2011, B. querencia Lise & Silva, 2011, and B. robertae Lise & Silva, 2011 (B. robertae is also recorded as occurring in Argentina). Most Berlandiella species are distributed in the south, southeast, and central-west regions of Brazil, with an isolated record of B. magna from the state of Piauí, Brazil and B. querencia from the state of Mato Grosso, Brazil. In this context, we present here the first record of the genus for northern Brazil, from the interior of the Amazon Basin.

According to Lise and Silva (2011), Berlandiella species are similar to Cleocnemis Simon, 1886 and Paracleocnemis Schiapelli & Gerschman, 1942, in the general body morphology; however, Berlandiella differs from these two genera by the absence of scopula in the tarsi and metatarsi and differs from all other Philodromidae genera by an unique set of characters: relatively high carapace, legs relatively short and densely covered by bristles (Fig. 1), and tarsi with dense claw tufts (Lise and Silva 2011). In addition, Lise and Silva (2011) described the Berlandiella male palp as possessing the following characters: cymbium covered by bristles, short and curved embolus, wider at the base, emerging from the apex of the tegulum, and tibia with a retrolateral tibial apophysis (RTA). According to Lise and Silva (2011), females of Berlandiella have a weakly depressed epigynum with a large spermathecae, usually elliptical, and with a glandular head.

Based on the revision provided by Lise and Silva (2011), it was possible to identify a new species of Berlandiella from Brasileira and Piracuruca, Piauí, named here as B. zabele sp. nov., composed of both sexes, and a couple of the known species B. querencia from Belém, Pará, whose male was unknown until now and is here described for the first time. In addition, we recorded the presence of scopula in the tarsi, metatarsi and palp of some species of Berlandiella, and recorded structures that may be present in the palp of males, such as a cymbial process. In this context, we present an updated diagnosis for Berlandiella.

The specimens were analyzed with a Leica MZ16 stereomicroscope and the illustrations were made with the assistance of a camera lucida attached to a stereomicroscope. Descriptions follow Lise and Silva (2011) and morphological terminologies follow Muster and Thaler (2004) and Muster (2009). Herein we record the position of the ventral macrosetae of the tibiae and metatarsi I-II according to Lise and Silva (2011), but we highlight the fact that there is no regular pattern of spination for species, since leg macrosetae I-II vary in number and position between individuals, and even between left and right legs. All measurements are in millimeters. Color images were obtained on a Leica M205 A stereomicroscope, coupled with a Leica MC170 HD digital camera. The LAS v. 4.9 software was used to generate extended focal range images. The scanning images were obtained with Mira 3 Tescan scanning electron microscope housed at the Laboratório Institucional de Microscopia Eletrônica de Varredura of the Museu Paraense Emílio Goeldi. For all scanning images, samples were submitted to an ultrasonic washer to clean surfaces. Then, the samples were gradually dehydrated in series of ethanol concentrations (80%, 90%, 96%, and 100%), subjected to critical-point drying and coated with gold. The distribution map of the species was elaborated with the program QGis 2.18 (QGIS 2019). The material examined is deposited in the Museu Paraense Emílio Goeldi (MPEG, curator A. Bonaldo) and Coleção de História Natural da Universidade Federal do Piauí (CHNUFPI, curator E.F.B. Lima). Abbreviations for the measurement of the eyes: (ALE) anterior lateral eyes, (AME) anterior median eyes, (PLE) posterior lateral eyes, (PME) posterior median eyes, (MOQ) median ocular quadrangle.

We present an updated diagnosis of Berlandiella based on the revision carried out by Lise and Silva (2011). Scopula was not recognized in Berlandiella specimens in the description of the type species B. insignis by Mello-Leitão (1929) and by Lise and Silva (2011). Until now the absence of this structure was considered as on one of the diagnostic features of the genus. However, we found scopula in the anterior legs of both males and females of B. zabele, males of B. querencia, B. magna (MCTP-31691), and B. meridionalis (MCTP-24716, MCTP-24718), although the scopula of the latter two species is less dense. Apparently some male and female specimens of B. meridionalis (MCTP-18928, MCTP-35602) lack scopula. In addition, figure 71 in Lise and Silva (2011) shows scopular setae in a section of one leg of B. robertae, indicating that other species of the genus also have scopula. The type-species of the genus, B. insignis, was not examined.

Scopula seem to be present only in sexually mature individuals, since we analyzed a juvenile of B. querencia without this structure. The scopular setae appeared in the anterior legs after ecdysis, when the specimen reached sexual maturity, becoming an adult male (MPEG.ARA 35228). Although we have not examined all species of the genus, it is clear that scopula should not be used to diagnose Berlandiella because the presence/absence of this structure varies among specimens of the same species and with degree of maturity.

We also removed from the diagnosis any mention about the height of the carapace, as well as the presence of dense claw tufts. In fact, some females of Berlandiella may have a relatively high carapace, and even males have convex carapace, although it is lower than females. As the understanding about the height of the carapace is subjective, we decided to remove this character from the diagnosis. The use of claw tufts in the genus diagnosis is not informative, since these structures are present in many others genera of Philodromidae.

Since we documented the presence of the scopula in Berlandiella specimens, the absence of this character used to separated Berlandiella from Cleocnemis become invalid and was removed from the diagnosis. As consequence we could not establish a clear limit between Berlandiella and Cleocnemis. The original description of the type species, C. heteropoda Simon, 1886, as well as the redescription by Mello-Leitão (1929), present some somatic characters shared with Berlandiella representatives, such as a strongly recurved posterior eye row, with the PME smaller and more separated than in PLE; short legs with similar length, covered with many bristles, elongated bulbous male palp and short embolus. In addition, Mello-Leitão (1929) describes low carapace for some Cleocnemis species, including C. heteropoda, and high carapace for Berlandiella. However, some Berlandiella have relatively low carapace, which show us that this character is not informative, as discussed above.

The limits between Berlandiella and Cleocnemis should be established based on the detailed analysis of the Cleocnemis type species, as well as a phylogenetic analysis to understand the relationship between the two genera. Both are beyond of the scope of this paper, although we have the concern of not describing a new taxa which can become a synonym of a species already described and allocated in Cleocnemis. Then, to make sure that B. zabele has not been already described under Cleocnemis we checked the type localities, original descriptions and illustrations for all 14 known species of this genus as follow summarized: C. bryantae (Gertsch, 1933), C. lanceolata Mello-Leitão, 1929, C. nigra Mello-Leitão, 1943, C. paraguensis (Gertsch, 1933), C. spinosa Mello-Leitão 1947, C. taquarae (Keyserling, 1891) and C. xenotypa Mello-Leitão, 1929 differ from B. zabele based on the genitalia. Cleocnemis moschata Mello-Leitão, 1943 and C. rosea Mello-Leitão, 1944 has several ventral macrosetae in the legs I-II and B. zabele has only 2–2 ventral macrosetae in this legs. Cleocnemis mutilata (Mello-Leitão, 1917), C. punctalata (Taczanowski, 1872), C. rudolphi Mello-Leitão, 1943 and C. serrana Mello-Leitão, 1929 has several somatic characters different from B. zabele, especially legs with unequal length. Cleocnemis nigra and C. rudolphi were registered from the state of Paraíba, relatively close to the type locality of B. zabele, however, differs from this species by the previously mentioned characters. The description of C. heteropoda is similar to B. zabele based on somatic characters, however, C. heteropoda occurs in Rio de Janeiro, relatively far from the type locality of B. zabele. Additionally, the heteropoda epithet refers to unequal legs, as indicated by Mello-Leitão (1929) in the redescription of the species. On the other hand, B. zabele has all legs similar in color and length.

About the new characters described for Berlandiella here, males of B. zabele sp. nov. and B. querencia present a cymbial process, which differs from all other males of the genus. A similar structure occurs in several species of Philodromus Walckenaer, 1826 and has been called the paracymbial lamella (Braun 1965), tutaculum (Levy 1977), cymbial process (Muster and Thaler 2004), and dorsal cymbium bulge (Muster 2009). The paracymbial lamella refers to a thin structure projected from the cymbium, which is different from that found in taxa described here. The tutaculum, used by Levy (1977), is a specific term to define a type of cymbial process that houses the embolus tip, which functions like a conductor in some genera of Thomisidae (Comstock 1940, Schick 1965, Benjamin 2011) and is different from the structure described here. In relation to the cymbium bulges, analyzing the illustrations of Muster (2009), we concluded that such structures are irregular projections on the dorsum of the cymbium, also different from the structure described here. Finally, Muster and Thaler (2004) described cymbial process as a small projection at the retrolateral proximal ventral border of cymbium, similar to structure found in males of B. zabele and B. querencia. Therefore, we decided to adopt this terminology for this structure.

Structures that emerge from the margin of the cymbium have been defined several times in the literature as paracymbium (see Griswold et al. 2005), however, this term is widely used to define cymbial projections present in Araneoidea palps, often being indicated as a group synapomorphy (Griswold et al. 1998, Griswold et al. 2005). In order to avoid misunderstandings, we decided not to use the term paracymbium to designate the structure present at the base of the cymbium of B. zabele and B. querencia; rather, we instead use the term cymbial process.

Finally, despite the difficulty in distinguishing Cleocnemis from Berlandiella, this latter is a revised genus and presents uniformity in the diagnostic characters present in the type-species and in the others (see diagnosis above). For this reason, and due to the similarity between B. zabele and other Berlandiella species, we decided to allocate the species in this genus.

We would like to thank the Museu Paraense Emílio Goeldi (MPEG) by providing the laboratory structure to carry out this work, especially Alexandre Bonaldo, responsible for the Laboratório de Aracnologia of the MPEG. We also thank Hilton Tulio Costi and Laura Miglio for providing the use of the SEM lab of MPEG and for all support given to the authors. We are very grateful to the reviewers and the editor for their careful reading of the text, for all valuable comments and suggestions, which were very helpful in improving the paper: Cristina Anne Rheims, Martín Ramírez, Ricardo Pinto-da-Rocha and anonymous. Finally, we greatly thank André Prado for our discussions about Cleocnemis taxonomy. The authors were supported by the following grants: PP – Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, 130702/2019-6) and Programa de Pós-graduação em Zoologia – UFPA/MPEG; MD – Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES, 88882.442371/2019-01) and Programa de Pós-graduação em Biodiversidade e Evolução – MPEG; RS – PandD ANEEL/CELPA # PD-0371-0017/2013. The laboratory facilities at Museu Paraense Emílio Goeldi were provided by Project #3362 CELPA/FADESP Monitoramento (REF.: 061/2013), also coded at Agência Nacional de Energia Elétrica (ANEEL) as PD-0371-0017/2013, entitled “Monitoramento dos possíveis impactos da linha de transmissão do Marajó sobre a fauna.” The last version of this paper was written when RS was supported by the Project #1199/FSADU (Fundação Sousândrade de Apoio ao Desenvolvimento da UFMA), entitled “Desenvolvimento experimental de metodologia para a detecção e redução de fatores de vulnerabilidade da vegetação que causam interrupções no fornecimento de energia”, funded by Companhia Energética do Maranhão-CEMAR.