A brief historical review regarding the type locality of Euptychia stigmatica Godman, 1905 was conducted, which suggests that its type locality is actually Rio de Janeiro, Brazil, rather than northeastern Argentina, as previously purported. Consequently, E. stigmatica and its senior synonym E. cyanites Butler, 1871, are regarded to be two species-group names representing a taxon in the euptychiine genus Caeruleuptychia Forster, 1964 known from the Brazilian Atlantic Forest. A lectotype is designated for E. cyanites. Additionally, two closely related species are named and described using an integrative approach with morphological and molecular evidence. Caeruleuptychia harrisi Nakahara & Freitas, sp. nov. and C. aemulatio Nakahara & Willmott, sp. nov. both occur in Amazonia and COI barcode data recovered these taxa as part of the caerulea clade of Caeruleuptychia.

Atlantic Forest, Brazil, Entre Rios, Euptychiina, Herbert Huntingdon Smith, Peru, Rio de Janeiro, taxonomy

Amazonia and the Atlantic Forest are two of the four major biogeographical regions in the Neotropical realm. Despite the existence of several definitions of the term “Amazon” (e.g., Eva et al. 2005), it hardly needs saying that this word applies to the largest contiguous region of tropical rainforests on earth, situated in South America encompassing nine countries: Bolivia, Brazil, Colombia, Ecuador, French Guiana, Guyana, Peru, Suriname, and Venezuela. The highly diverse but vanishing Atlantic Forest spans more than 3,000 km along the southeastern coast of Brazil and penetrates into the northern edge of Argentina and southeastern Paraguay (Morellato and Haddad 2000; Giraudo 2003; Huang et al. 2007; Ribeiro et al. 2009). These two biogeographical regions are currently separated from each other by a diagonal series of open vegetation formations encompassing the semi-arid Caatinga, the Cerrado savannas, and the dry Chaco (Santos et al. 2012; Bonatelli et al. 2022). Nevertheless, mounting evidence supports historical biotic connections between the Amazon and the Atlantic Forest (e.g., Brown 1987; Costa 2003; Pavan et al. 2011; Batalha-Filho et al. 2013; Rodrigues et al. 2014), and many sister species pairs are known to have their distribution restricted to either one of these two major biogeographical regions (e.g., Silva et al. 2014). In addition, faunistic and floristic similarities have been reported in certain parts of these biogeographical regions: the “tabuleiro” forests (= tableland forests) in the Brazilian states of Espírito Santo and southern Bahia show floristic resemblance with Amazonia by sharing 223 plant species (Andrade-Lima 1966), and faunistic similarity between these two biogeographical regions has also been indicated (Coelho et al. 2022). Recent discoveries of new species from the tableland forests of Espírito Santo further strengthen the affinity of this region with Amazonia (e.g., Lewis et al. 2017).

The butterfly subtribe Euptychiina is a diverse radiation in the Neotropics, and our understanding of its systematics has seen steady improvement (e.g., Espeland et al. 2023). As members of one of the more diverse euptychiine genera, Caeruleuptychia Forster, 1964 species are well represented throughout the lowlands of tropical South America east of the Andes. Ongoing work using both morphological and molecular data recognizes three clades within Caeruleuptychia: 1) a group with species exhibiting sexual dimorphism and males displaying an androconial patch on the dorsal hindwing (aegrota clade); 2) a group mainly including species with blue-lilac wings and possessing a concavity along the dorsal margin of the male genitalic valvae (caerulea clade); 3) a group including mainly species with brown wings, with males accompanied by secondary sexual structures (umbrosa clade). There exist a number of other morphological characters to support each of these three clades (e.g., Nakahara et al. 2020a), and a study is underway to revise the latter two clades. Nevertheless, the aegrota clade has recently been revised and its known species diversity may not increase (Nakahara et al. 2018b). On the other hand, our understanding of species richness in the caerulea clade and umbrosa clade is far from complete, since COI barcoding data suggest they contain a high number of cryptic species (unpublished data), which otherwise might have remained unrecognized based solely on external morphology. Given this situation, cataloging Caeruleuptychia diversity in the form of single to few species descriptions provides progress towards developing a firm species-level classification of the group. The urgency to document Caeruleutychia species richness in Amazonia and the Atlantic Forest biomes is increased by the high degrees of biodiversity loss in these regions, driven by forest loss, fragmentation, and degradation (e.g., de Lima et al. 2020; da Cruz et al. 2021).

Therefore, we here propose taxonomic changes in the caerulea clade of Caeruleuptychia in order to contribute towards cataloging species-level diversity of the genus in these distinctive biogeographical regions. In the current study, we discuss the taxonomic status and history of two species-group names currently placed in Caeruleuptychia, Euptychia cynanites Butler, 1871 and E. stigmatica Godman, 1905. These two species-group names were considered synonymous by Weymer (1911) and the identity of these two names has not been discussed by any subsequent authors with evidence. Here, we reinterpret the type locality of E. stigmatica and regard this nomen as a species-group name to represent a Caeruleuptychia taxon known from the Atlantic Forest. Further, we provide evidence to retain E. stigmatica as a junior subjective synonym of E. cyanites. Additionally, we name and describe two closely related undescribed Amazonian Caeruleuptychia species utilizing an integrative approach.

Application of species-group names (specific vs subspecific) to a sister pair of allopatric taxa involves some degree of subjectivity. Therefore, the species-level status of C. cyanites, C. harrisi sp. nov., and C. aemulatio sp. nov. can be challenged considering their low genetic divergence and allopatric distribution, coupled with their rather trivial differences in adult external morphology. Nevertheless, we consider our conclusions to be justifiable based on the following arguments provided herein. One principal morphological character to consider these three taxa to be accorded specific status is the size difference in the dorsal forewing androconial scent patches. The taxonomic value of this androconial scent patch can be inferred from two closely related species in the caerulea clade, C. coelestis and C. aemulatio sp. nov. Male specimens of C. coelestis lack the androconial scent patch, whereas this feature is present in C. aemulatio sp. nov. These two species are broadly sympatric in the southwestern Amazonia and their species-level status is delineated in the phylogenetic structure (Fig. 1A), with mean inter-specific pairwise distance of 4.0% based on the dataset used in the present study. Additionally, Nakahara et al. (2022) described C. thaliana Nakahara & Piovesan, 2022 on the grounds of broad sympatry with its sister species, C. umbrosa Butler, 1870, in conjunction with the presence/absence of the dorsal forewing androconial hair-pencil, with support from molecular data. Although the role of these androconial scales are not fully explored in euptychiines, a wealth of research has been conducted on another Satyrinae genus, Bicyclus Kirby, 1871 (Satyrini: Mycalesina): Condamin (1973) implied that the androconial scales on wings are informative species-level diagnostic characters in Bicyclus by providing such information in his dichotomous key, and subsequent studies revealed that the male sex pheromone component released by sympatric species pairs displays larger differences compared to compounds emitted by an allopatric pair of Bicyclus species (e.g., Bacquet et al. 2015). These studies on Bicyclus, coupled with the differences observed in sympatric Caeruleuptychia sister species pairs, suggest that these androconial scales can potentially play a key role in interspecific reproductive isolation in euptychiines as well. This conclusion reinforces our use of the size difference observed in the dorsal forewing androconial scent patch between C. cyanites, C. harrisi sp. nov. and C. aemulatio sp. nov. as a species-level diagnostic character. Furthermore, these three species are found separately in Amazonia (C. harrisi sp. nov. and C. aemulatio sp. nov.) and the Atlantic Forest (C. cyanites), two of the four major biogeographic regions in the Neotropics. As mentioned above, these two biogeographical regions are separated by a diagonal of open vegetation formations, currently with no known contact zone, and thus potentially leading to different biological traits. Indeed, C. cyanites appears to be a canopy dwelling species, which is an unusual feature for an euptychiine species, since members of this subtribe are often encountered in the understory of the forest (pers. obs., but see Singer et al. 1983 and Freitas et al. 2021 for discussions of other euptychiine species with similar canopy habits). With no equivalent trait data for C. harrisi sp. nov. and C. aemulatio sp. nov., it would be premature to develop the discussion further, but we consider the species-level classification introduced herein to be informative based on the aforementioned evidence, as well as taking into account the genetic data with a clear barcoding gap (Fig. 1A). The uniting of C. cyanites and C. harrisi sp. nov. and regarding one as a subspecies of the other might be more parsimonious, although the existence of relatively few euptychiine species known to occur in both Amazonia and the Atlantic Forest (e.g., Amiga arnaca (Fabricius, [1777])), argues against such an approach.

Euptychia cyanites was described by Butler (1871: 282–283) based on an unspecified number of male specimen(s) from Brazil in [William Wilson] Saunders’ collection. The sex is not explicitly stated in the original description, although the mention of “the curious scaly whitish patch on the interior margin of the anterior wings on the upperside” in the diagnosis clearly points to the examined specimen(s) being a male. The male syntype of E. cyanites (Fig. 2A, B) is housed at NHMUK and the label information is provided above. The forewing length of this syntype is 25 mm (the right forewing is angled away and appears distorted in Fig. 2A, B), which falls within the forewing length range of C. cyanites (see Table 1). Other notable features include a large scent patch on the dorsal forewing in cell 2A, which is larger than that of C. harrisi sp. nov. and C. aemulatio sp. nov. (see Table 1), as well as the presence of pale powder-blue modified scales visible in the middle of the ventral forewing cell Cu₂. These characteristics suggest that the species name cyanites cannot be applied to concepts represented by the two other species-group names introduced in the present work, but applies to individuals of a Caeruleuptychia species known from the Atlantic Forest. Although the type locality of E. cyanites was only stated as “Brazil” with no further information as to its provenance, evidence suggests William Saunders’ collection did hold materials from southeastern Brazil (Hewitson 1858). Additionally, for an unknown reason, Gaede (1931: 444) associated Minas Geraes [Gerais] as the locality for cyanites, thus we have some support besides external morphology to narrow down the identity of E. cyanites.

Euptychia stigmatica Godman, 1905 was described by Frederick DuCane Godman based on a single male specimen from “Entre Rios, Argentina” collected by H[erbert] H[untingdon] Smith (Godman 1905: 186). The statement of one specimen in the original description enables us to regard this male specimen as a holotype fixed by monotypy following Article 73.1.2 of the ICZN (1999). In addition to this holotype, Godman (1905) mentions the existence of specimens in the British Museum [=NHMUK] of what he considered as a conspecific pair from “Rio River”. The transcribed label data for these two specimens housed at the NHMUK are provided above. Butler (1869: 347) stated that “Rio R.” of [William Chapman] Hewitson probably refers to “Rio Real” in Bahia, Brazil. This interpretation of “Rio R.” representing a site situated in Bahia, Brazil falls within the range (i.e., Atlantic Forest) of C. cyanites and the whereabouts of Godman’s “Rio River” is unknown. Subsequently, Weymer (1911: 218) regarded E. stigmatica to be a junior subjective synonym of E. cyanites. Gaede (1931: 444) considered both E. cyanites and E. stigmatica to be synonymous with Papilio crantor, although with no justification. Gustav Weymer also did not provide rationale for his synonymy and referred to the type locality of E. stigmatica as “Entre Rios, Argentina”, in addition to stating that [Richard] Haensch collected both sexes [of E. cyanites] in Minas Gerais (Weymer 1911). The illustration accompanying the original description on pl. X, fig. 2 (Fig. 2D, E) clearly shows a large scent patch in the dorsal forewing cell 2A. This feature is also visible in the male holotype housed at the NHMUK (Fig. 2C, D; see above for transcribed label data), which readily excludes this specimen representing many other species in the caerulea clade of Caeruleuptychia, including C. harrisi sp. nov. and C. aemulatio sp. nov. (see above). The scent patch length/inner margin length of this male holotype is 27%, which is greater than that of C. harrisi sp. nov. (14–19%, mean 16.2%) C. aemulatio sp. nov. (21–27%, mean 23.7%) and in accordance with C. cyanites (25–31%, mean 28%). Despite the lack of dorsal hindwing black spot in cell Cu₁ and pale powder-blue scales on the ventral forewing in cell Cu₂ not being discernible (due to overlap with the hindwing?), the forewing length of this holotype is 25.4 mm, which supports its identity as C. cyanites, thus we continue the discussion by following the footsteps of the remarkable naturalist who collected the holotype of E. stigmatica.

Herbert Huntingdon (or Huntington) Smith (1851–1919) (Fig. 7B) was an American naturalist, as well as being known as a conchologist later on in his life, who traveled throughout tropical America collecting faunistic and floristic materials (Holland 1919; Clapp 1920). Smith was first introduced to the delights of the Neotropical fauna and flora in 1870, when he traveled to the Brazilian Amazon as a member of the Morgan expeditions with Charles Frederick Hartt (Smith 1879a; Kunzler et al. 2011). In 1874, Smith returned to Brazil and spent two years around Santarém (Pará, Brazil) (Smith 1879a), followed by a year exploring the northern tributaries of the Amazon basin and the Tapajós, and returned to the United States after residing for four months in Rio de Janeiro. He was then able to make additional trips to Brazil supported by the publishing firm Messrs Scribner & Co., and Smith wrote six articles for their Scribner’s Monthly journal, based on his revisit to Brazil in return (Kunzler et al. 2011). One of these articles was entitled “Coffee culture in Brazil” (Smith 1879b), which also appeared in chapter 18 (The Story of Coffee) in Smith’s (1879a) “Brazil the Amazon and the coast”, where he reported his studies on the coffee industry and referred to the locality he visited as “Entre Rios”. In his introduction of this site, Smith (1879a, b) stated that “the story begins on the hills around Entre Rios, away back of [from] the Organ mountains”. The mention of Organ mountains [= Serra dos Órgãos] implies that the place named “Entre Rios” in Smith (1879a, b) is located in the vicinity of Rio de Janeiro, Brazil, already contradicting the type locality of E. stigmatica stated by Godman (1905) as “Entre Rios, Argentina”. Smith (1879a: 530) further refers to his Entre Rios as “Entre Rios is on a branch of the Dom Pedro Segundo Railroad [Estrada de Ferro D. Pedro II], where the latter meets the Uniao e Industria [Estrada de Rodagem União e Indústria]”. Estrada de Ferro D. Pedro II departed from Côrte [= Rio de Janeiro?] on March 29, 1858 and reached Entre Rios, a municipality in Rio de Janeiro currently known as Três Rios, on October 13, 1867 (Fig. 7A; Picanço 1884: 172). The inauguration of Estrada de Rodagem União e Indústria took place in 1861 (Bartholomeu 2019), and connected Petrópolis (Rio de Janeiro) and Juiz de Fora (Minas Gerais). Bartholomeu (2019: fig. 2) shows passing of Estrada União e Indústria through Três Rios [= Entre Rios], which corresponds to Smith’s statement “Entre Rios is on a branch of the Dom Pedro Segundo Railroad, where the latter meets the Uniao e Industria”, since this is the only place represented by the name Entre Rios reached by both the União e Indústria and Estrada de Ferro D. Pedro II. Smith (1879a: 511) started his “The Story of Coffee” by stating “In 1878, Brazil exported more than five hundred million pounds of coffee.” Thus, assumption can be made that Smith was studying the coffee industry in Entre Rios at some point in 1878. In the chapter entitled “Ceará and the Drought” in his “Brazil…coast”, Smith (1879a: 421) wrote “I reached Fortaleza [Ceará, Brazil] on 19^th of that month [December 1878]”, and based on his (seemingly) chronologically written preface of “Brazil…coast”, he likely studied the coffee industry prior to his travel to Ceará. De Abreu (1922) also stated that Smith passed through Ceará and Pernambuco on his way back to the USA, therefore, he likely traveled to eastern Brazil in 1878, towards the end of his journey. If this holds true, Smith was in Entre Rios to study the coffee culture prior to December 1878. This assumption is in accordance with the printed label “Sept. [year not indicated]” associated with the holotype of E. stigmatica. It is worth noting that a species of a fly, Microdon inermis Williston, 1888 (Diptera: Syrphidae), was also described based on a single specimen collected by H. H. Smith in “Entre Rios” in September [year not indicated] (Williston 1888: 258). Despite the referral of its type locality to Argentina in subsequent catalogs (e.g., Thompson et al. 1976), the labels associated with the type specimen (i.e., holotype, housed at AMNH) of M. inermis bear no label indicating the country of origin, except for a handwritten label “Entre Rios Sept.”, in accordance with the original description (Fig. 7C). This observation raises the likelihood that the holotype of M. inermis was collected by Smith in September of an unknown year when he also sampled the holotype of E. stigmatica in Entre Rios, currently known as Três Rios (Rio de Janeiro, Brazil), perhaps in 1878.

Figure 7. 

A table from Picanço (1884: 172) showing Estrada de Ferro D. Pedro II reaching Entre Rios (= Três Rios) in 1867 (image reproduced from Harvard Business School (HBS) Archives, Baker Library, HBS at Harvard University) B portrait of Hebert H. Smith from volume 30 of The Nautilus, where Clapp’s (1920) article appeared C labels associated with the holotype of Microdon inermis (Diptera) (AMNH) D labels associated with the paratype of Amblystira nyctalis (Hemiptera) (USNM).

Nevertheless, Smith explored the Brazilian Amazon again, by signing a contract with Ladislau de Souza Mello e Netto (then director of the MNRJ) at the end of 1881 (Doc. MN 237, folder 20, of 12/23/1881). In particular, Smith stayed in Mato Grosso for four years and collected extensively in “Chapada” [= Chapada dos Guimarães, Mato Grosso, Brazil] during this period (de Abreu 1922), until his return to the United States in 1886. His entomological samples of an estimated 10,000 insect species collected in Chapada yielded many species then unknown to science.

Smith’s (1922) “Do Rio de Janeiro a Cuyabá”, accompanied with a preface by Capistrano de Abreu, provides various records of Smith’s trips in Brazil and neighboring countries between 1881 and 1886. It is worth highlighting that Smith refers to a trip to Buenos Aires in chapter 19 of his “Do Rio de Janeiro a Cuyabá” (Smith 1922). Smith apparently traveled from Corumbá (in Mato Grosso) to Buenos Aires, via the Paraná River, consequently passing through Entre Rios province in Argentina. The details of his boat trip to Buenos Aires are not clear from the passage, but he did mention a visit to the Natural History Museum in Buenos Aires where he examined the extensive collection of Argentinean insects. Despite no mention of Entre Rios throughout over 350 pages in Smith (1922), it is therefore theoretically possible for him to have collected in Entre Rios province in Argentina on his way to Buenos Aires. This possibility is supported by the fact that Williston (1888: 249) examined dipteran specimens from “Corumbá” and “Rio Parana, near Bella Vista [Corrientes province, Argentina]” collected by Smith, likely during his trip to Buenos Aires via the Paraná River.

Smith returned to the United States in 1886 and his insect samples, including over 30,000 Lepidoptera specimens, were partly acquired by F. D. Godman in London (UK) and the Carnegie Museum of Natural History (USA) (Kunzler et al. 2011). Subsequently, numerous taxa were named and described based on these materials brought back by Smith: Godman (1905) described Pierella chalybaea Godman, 1905 (Nymphalidae: Satyrinae), simultaneously with E. stigmatica, based on three specimens collected by Smith in Chapada; prior to Godman (1905), Godman also described eight riodinid species based on specimens collected by Smith in Chapada (Godman 1903); Williston (1888) also incorporated at least 400 specimens representing more than 80 dipteran species sampled by Smith in Chapada when he described M. inermis in his “Diptera Braziliana”. Williston’s (1888) introduction begins with “More than a year ago Mr. Herbert H. Smith, who is well known to zoologists for his writings on Brazil, placed in my hand for study a collection of Diptera made by him during the past few years in Southern Brazil”. Given the mention of “more than a year ago,” as well as having the work published in 1888, implies that Williston received materials subsequent to Smith’s various excursions to Chapada and other sites he presumably visited between 1881 and 1886, perhaps in late 1886 upon his return from Brazil or at some point in 1887. The statement “made by him [H.H. Smith] during the past few years” refers to specimens originating from Smith’s entomological expeditions to Chapada and other places in Brazil which took place between 1881 and 1886, not samples obtained (if any) by Smith in the 1870’s – this is supported by the fact that a number of dipteran specimens examined and incorporated in Williston (1888) were collected in Chapada. Both Samuel W. Williston and F. D. Godman worked on Biologia Centrali-Americana: zoology, botany and archaeology, a ground-breaking work on the Central American fauna and flora, to which Smith contributed with materials from his trip to Mexico in 1888–1889 (e.g., Osten Sacken 1887). Although Godman’s (1903, 1905) aforementioned works appeared after Smith’s travel to Mexico, Williston (1888) already had Smith’s entomological samples from Brazil prior to 1888, thus it is possible that Godman also had Smith’s Brazilian Lepidoptera specimens in his hands before 1888. In order to develop this discussion further, it is important to take into account Drake’s (1922) work on the hemipteran family “Tingitidae” [= Tingidae] based predominantly on specimens housed at the Carnegie Museum of Natural History. Drake (1922: 361) referred to a single specimen as a paratype in the author’s collection from “Entre Rios, Brazil” in his description of Amblystira nyctalis Drake, 1922 (Hemiptera: Tingitidae). Like E. stigmatica and M. inermis, this paratype of A. nyctalis housed at USNM also bears a label stating that it was collected in September of an unknown year (Fig. 7D). The holotype of A. nyctalis is a male from Chapada. Drake (1922: 361) stated that these two specimens were “collected by Mr. and Mrs. H. H. Smith”. Herbert H. Smith married Miss Amelia Woolworth Smith on October 5, 1880 and she participated in Smith’s trip to Brazil between 1881 and 1886 (Holland 1919). Thus, the specimen from “Entre Rios, Brazil” used in Drake’s (1922) description of A. nyctalis is based on materials collected while Smith and his wife made their headquarters in Chapada between 1882 and 1886. Among nine species of Amblystira known to occur in South America, two have been recorded from Argentina, both in the province of Misiones (Montemayor 2010). Furthermore, by consulting with an expert, the locality of A. nyctalis is more likely to be Brazil, not in northeastern Argentina (Sara Montemayor, pers. comm., November 2022). From a similar biogeographical point of view, we can call into question the type locality of E. stigmatica being a site in northeastern Argentina, since records for Caeruleuptychia from this region are not known to us to date.

Unlike Smith’s study of the coffee industry at Entre Rios, Brazil, in the late 1870s, documented in Smith (1879a, b), we were unable to find any statement in the existing literature which can be accepted without ambiguity to support that Smith actually visited Entre Rios (= Três Rios) in Brazil between the years of 1881 and 1886. Nevertheless, some inference can be made. For example, in chapter 24 of his “Do…Cuyabá”, Smith (1922: 247) wrote “É de notar que nossas colecções paraguayas [de insectos] differiam muito das que fizeramos na região costeira do Brasil” [It is to be noted that our Paraguayan [insect] collections differed greatly from those we did in the coastal region of Brazil], which implies Smith did collect insects in the coastal region of Brazil (i.e., the Atlantic Forest). It must be noted that Chapada, located in Mato Grosso is unlikely to be considered as a site lying in the coastal region of Brazil, thus leaving the possibility of Smith collecting insects elsewhere between 1881 and 1886. De Abreu (1922) stated that in addition to staying for four years in Chapada, Smith spent a few months in Pará, ten days in Pernambuco, six months in Rio de Janeiro, and six months in Rio Grande do Sul, starting from May 1881. Thus, it is possible that Smith visited Entre Rios (= Três Rios) while residing in Rio de Janeiro for six months, especially since he would have been familiar with the region due to his prior visit in the 1870s. This possibility can also be supported based on the botanical materials collected by Smith and currently housed at the New York Botanical Garden’s C. V. Starr Virtual Herbarium (https://sweetgum.nybg.org/science/vh/). Among 17 botanical samples in this Herbarium collected by Smith, 16 samples are from Mato Grosso, whilst there exists a single specimen from Minas Gerais. Considering the fact that the whereabouts of the majority of the botanical samples amassed by Smith remain unknown (Kunzler et al. 2011), it is not clear how these materials ended up in New York. Notwithstanding this situation, the presence of an individual from Minas Gerais among samples from Mato Grosso, can be seen as evidence to support the supposition that Smith visited Entre Rios at some point between 1881 and 1886. Entre Rios (= Três Rios) is situated near the border of the states of Rio de Janeiro and Minas Gerais, and the Estrada de Ferro D. Pedro II railway would have passed through Entre Rios on its way from Rio de Janeiro to various stations in Minas Gerais. William J. Holland, a then director of the Carnegie Museum of Natural History, estimated that approximately 30,000 species and 200,000 insect specimens entered the collection of the Carnegie Museum of Natural History (Holland 1919). These figures illustrate a discrepancy between the number of insect species collected by Smith at Chapada (10,000 species) (de Abreu 1922), suggesting that he conducted collecting trips in various other places besides Chapada, including the areas around Rio de Janeiro. Indeed, Williston (1888) did cite more than 130 specimens representing 36 dipteran species collected by Smith in “Rio de Janeiro”, reinforcing Smith’s entomological survey in the Atlantic Forest between 1881 and 1886.

In the present study, we call into question the purported type locality of E. stigmatica representing a site in northeastern Argentina. Based on the discussion developed above, the holotype of E. stigmatica was more likely to have been collected in Entre Rios in the Brazilian state of Rio de Janeiro, a place currently known as Três Rios. Herbert H. Smith likely collected the holotype of E. stigmatica in September 1881. The argumentation supporting the conclusion regarding the type locality of E. stigmatica in the present study can be summarized as follows:

1. Smith visited Entre Rios (= Três Rios) in the state of Rio de Janeiro in the 1870s.
2. Given the proximity, he was able to revisit Entre Rios (= Três Rios) while residing in Rio de Janeiro for six months in 1881.
3. Simultaneous descriptions of taxa based on specimens from both Entre Rios and Chapada by Godman (1905), Williston (1888), and Drake (1922) suggested specimens from Entre Rios were also part of the entomological samples brought back to the United States by Smith in 1886.
4. Entre Rios was considered to be in Argentina by Godman, but Williston (1888) did not associate the country of origin in his description (also reflected in the type label), and Drake (1922) linked Entre Rios with Brazil, not Argentina.
5. Type materials of E. stigmatica, M. inermis, and A. nyctalis all bear labels indicating that they were collected in September (see Figs 2, 7C, D).
6. Evidence suggests Smith did collect insects apart from Chapada, between the years of 1881 and 1886 (see immediately preceding paragraph).
7. The paratype of A. nyctalis was collected after Smith’s marriage in 1880.
8. Current understanding of Amblystira and Caeruleuptychia distribution suggests Rio de Janeiro to be a more plausible locality for A. nyctalis and E. stigmatica (no records exist for these two genera from Entre Rios province in Argentina).

Finally, we note that the aforementioned conclusion of the type locality reinforces the current taxonomic status of E. stigmatica being a junior subjective synonym of E. cyanites. The type locality of E. stigmatica likely being Entre Rios (= Três Rios) situated in the state of Rio de Janeiro, coupled with six historical specimens from Rio de Janeiro and São Paulo housed at DZUP and MfN (see above), suggests that C. cyanites occurs in the southern part of the Atlantic Forest. Nevertheless, the three females from “Araras, São Paulo” housed at DZUP may be mislabeled since the senior author (AVLF) and colleagues have sampled extensively in the region (e.g., Brown and Freitas 2000) without documenting this species. Despite the lack of recent materials from this region, as well as provenance of three specimens currently unknown, two males from São Paulo housed at MfN do fall into the concept of C. cyanites discussed above based on the forewing length and scent patch size as shown in Fig. 1B. Also, two females from Rio de Janeiro (in DZUP and MfN) do phenotypically agree with the sequenced female from Espírito Santo (BC-DZ-007). As supported by a number of butterfly taxa (e.g., Scada reckia (Hübner, 1808); Heliconius nattereri C. Felder & R. Felder, 1865; Pierella hyalinus (Gmelin, [1790]) (all Nymphalidae)), as well as literature investigating inter-specific genetic divergence (e.g., Pirani et al. 2022), Espírito Santo, northeastern Minas Gerais and the coastal region of Bahia can be characterized as faunistically different from the southern region of the Atlantic Forest, including the state of Rio de Janeiro. Apparently, no records of Caeruleuptychia exist in inventories compiled for butterflies from Rio de Janeiro (e.g., Capronnier 1881; Soares et al. 2011), suggesting its rarity and perhaps disjunct distribution. In the absence of genetic data from the holotype of E. stigmatica, one may argue that the available evidence is insufficient to draw conclusion as to its identity. On the other hand, it is apparent that there is only a single species of Caeruleuptychia with blue phenotype known from the Atlantic Forest. If we accept the type locality of E. stigmatica is in the state of Rio de Janeiro, coupled with the existence of two known male Caeruleuptychia from this region with the same habitus (both in MfN), it seems reasonable to apply the same species-group name stigmatica to these historical specimens in DZUP and MfN and regard them as conspecific with the holotype of E. stigmatica. Consequently, we parsimoniously apply the same specific epithet to specimens from Espírito Santo, northeastern Minas Gerais, and the coastal region of Bahia. We provided evidence from external morphology and literature supporting that E. cyanites is likely from southeastern Brazil as well. Thus, it seems reasonable to retain the synonymy between E. stigmatica and E. cyanites, although we admit that this taxonomic hypothesis is subject to revision if additional evidence surfaces. We thus settle the matter here by designating a lectotype for E. cyanites based on the male syntype discussed above and consider this species-group name to represent a species in Caeruleuptychia known to occur in the Atlantic Forest as a senior synonym of E. stigmatica (lectotype designation).

We are extremely grateful to Brian Harris, Robert K. Robbins, Neil Rosser, Blanca Huertas, Melissa Murphy, Heather Oswald, Bruno Ferreira, Gerardo Lamas, Vitor O. Becker, John V. Calhoun, Sara Montemayor, David Grimaldi, David Boufford, and Agnieszka Pierwola for assisting us in various ways throughout the course of preparing this manuscript; Gerardo Lamas and Maryzender Rodríguez-Melgarejo are acknowledged for reviewing the manuscript post-submission by providing helpful comments and suggestions; SN acknowledges Edward O. Wilson Biodiversity Postdoctoral Fellowship at Harvard University; EPB thanks Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) (2016/15873-8 and 2018/21432-0) and the Brazilian Research Council (CNPq) (162673/2020-5) for postdoc fellowships. AVLF acknowledges support from FAPESP (BIOTA-FAPESP grants 2013/50297-0, 2021/03868-8), from the Brazilian Research Council–CNPq (303834/2015-3, 421248/2017-3 and 304291/2020-0) and from the National Science Foundation (DEB-1256742). This publication is part of the RedeLep “Rede Nacional de Pesquisa e Conservação de Lepidópteros” – SISBIOTABrazil/CNPq (563332/2010-7). The present study is registered in the Brazilian Sistema Nacional de Gestão do Patrimônio Genético e do Conhecimento Tradicional Associado (National System for the Management of Genetic Heritage and Associated Traditional Knowledge) SISGEN (A9676AF).